{
    "0208/astro-ph0208530_arXiv.txt": {
        "abstract": "We observed the Seyfert~1 galaxy Mrk~509 for $\\sim 59$~ks with the {\\it Chandra} High-Energy Transmission Gratings, simultaneously with {\\it HST}/STIS and {\\it RXTE}. Here we present a detailed analysis of the soft X-ray spectrum observed with {\\it Chandra}. We measure strong absorption lines from He-like Ne and Mg, and from H-like N, O, and Ne. Weaker absorption lines may also be present. The lines are unresolved except for \\nela and \\nenine \\resonetwo ($\\lambda 13.447\\AA$), which appear to be marginally resolved. The profiles are blueshifted with respect to the systemic velocity of Mrk~509, indicating an outflow of $\\sim -200$~$\\rm km \\ s^{-1}$. There is also a hint that the profiles may have a velocity component near systemic. The soft X-ray spectrum can be described in remarkable detail with a simple, single-zone photoionized absorber having an equivalent neutral Hydrogen column density of $2.06^{+0.39}_{-0.45} \\rm \\ \\times 10^{21} \\ \\rm cm^{-2}$ and an ionization parameter of $\\log{\\xi} = 1.76^{+0.13}_{-0.14}$ (or $\\log{U}=0.27$). Although the photoionized gas almost certainly is comprised of matter in more than one ionization state and may consist of several kinematic components, data with better spectral resolution and signal-to-noise would be required to justify a more complex model. The UV data, on the other hand, have a velocity resolution of $\\sim 10$~$\\rm km \\ s^{-1}$ and can easily detect eight kinematic components, covering roughly the same velocities as the X-ray absorption profiles. Even though the X-ray and UV absorbers share the same velocity space, the UV absorbers have a much smaller column density and ionization state. We show that models of the X-ray data do not predict significant UV absorption and are therefore consistent with the UV data. Finally, we do not detect any soft X-ray emission lines. ",
        "introduction": "\\label{sec:intro}  X-ray and UV absorption and emission by photoionized circumnuclear gas in type~1 Seyfert galaxies is a key observational diagnostic. While it has been possible to study the UV absorption with a velocity resolution of $\\sim 10$~$\\rm km \\ s^{-1}$, the kinematics of the X-ray absorber observed with CCDs (such as those aboard \\asca) could only be studied with a velocity resolution $>10,000$~$\\rm km \\ s^{-1}$. Inadequate spectral resolution, combined with a lack of simultaneity between X-ray and UV observations has resulted in major uncertainties in the dynamics, physical state, location, and geometry of the X-ray and UV absorbers, as well as the relation between the two. The launch of \\chandra and \\xmm began a new era in the study of X-ray photoionized circumnuclear gas. The energy resolution of the \\chandra transmission gratings is currently the best available in the 0.5--10 keV band, and is as high as $\\sim 280$~$\\rm km \\ s^{-1}$ at 0.5 keV. High resolution X-ray spectroscopy with \\chandra now allows the gas kinematics to be studied seriously for the first time, and the detection of individual absorption and emission lines can now place very strong constraints on the ionization structure of the gas.  The existence of warm, or partially ionized, X-ray absorbing gas in type~1 active galactic nuclei (AGNs) was first suggested by \\einstein observations of QSO MR~2251$+$178 (\\cite{halpern84}). Subsequently,  \\rosat and \\asca showed this to be a common phenomenon, present in roughly $\\sim 50-60\\%$ of type~1 AGNs, and studies focused on the measurements of the \\oxyseven and \\oxyeight absorption edges which appeared to be the most prominent features in these low to moderate resolution spectra (\\cite{nandra92}; \\cite{fabian94}; \\cite{reynolds97}; \\cite{george98}). High-resolution grating observations with \\chandra have since been used to study the X-ray warm absorbers in several Seyfert~1 galaxies (\\cite{coll01}; \\cite{lee01}; \\cite{sako01a}; \\cite{pounds01}; \\cite{kaastra02}; \\cite{kaspi02}; \\cite{yaqoob02a}). Discrete, narrow absorption features (FWHM less than $\\sim 2000 \\ \\rm km \\ s^{-1}$), often unresolved, are found to be typically blueshifted, with outflow velocities ranging from a couple of hundred to a couple of thousand  $\\rm km \\ s^{-1}$ relative to systemic. Moreover, in some cases multiple velocity components have been identified (\\cite{coll01}; \\cite{kaspi02}; \\cite{kaastra02}). There are typically fewer features in emission than absorption (e.g. \\cite{kaastra02}; \\cite{kaspi02}). UV absorbing gas in Seyfert 1s was first observed with \\iue (\\cite{ulrich88}). Multiple discrete kinematic components to the UV absorbers have been  observed and it has been suggested that the X-ray absorber is associated with with one or more, but not necessarily all,  of the UV components (e.g. \\cite{mathur95}; \\cite{mathur99}). Observations with the improved sensitivity and resolution of \\hst and \\fuse have since shown that $\\sim 60\\%$ of Seyfert 1s exhibit intrinsic UV absorption and that multiple kinematic components are common (\\cite{crenshaw99}).   At present, models of the X-ray absorbers span a wide range in distance from the central ionizing source, from winds originating at the accretion disk (\\cite{elvis00}), out to the putative (parsec-scale) molecular torus (e.g. the multi-temperature wind model of Krolik \\& Kriss (2001), and beyond, to the NLR (e.g. Ogle \\etal 2001). In addition, for two particular Seyfert~1 galaxies observed by \\xmm (\\mcg and Mrk~766) it has been proposed that relativistically broadened soft X-ray lines from an accretion disk can account for some of the spectral features traditionally attributed to a warm absorber (Branduardi-Raymont \\etal 2001; Sako \\etal 2001a). However, Lee \\etal (2001) have argued that the \\chandra grating data, for \\mcg at least, can be modeled with a dusty warm absorber without the relativistic emission lines. One thing is clear however: all of the models must stand up to the scrutiny of an increasingly large body of results as the results of new observational campaigns become available. The purpose of the present paper is to present the results of one such new campaign, namely simultaneous \\chandra \\hetg, \\hst, and \\rxte observations of the luminous ($L_{2-10 \\rm \\ keV}$ typically $1.3-2.6 \\times 10^{44} \\ \\rm ergs \\ s^{-1}$ \\footnote{We use $H_{0}=70 \\ \\rm km \\ s^{-1} \\ Mpc^{-1}$ and $q_{0}=0$ throughout this paper, unless otherwise stated.}, \\cite{weav2001}) Seyfert~1 galaxy, \\mk ($z=0.0344$, Fisher \\etal 1995). \\src has been studied extensively in the UV and by every major X-ray astronomy mission since {\\it HEAO-1~A2}. Being so bright and exhibiting interesting absorption structure in the UV (e.g. \\cite{kriss00}) and X-ray bands (e.g. Pounds \\etal 1994; \\cite{reynolds97}; \\cite{george98}; \\cite{pounds01}; \\cite{perola00}) \\src makes an excellent candidate for this kind of study. The driving principles behind our campaign were to measure, with the highest spectral resolution available, the X-ray and UV absorption features {\\it simultaneously} in order to eliminate uncertainty due to variability, and to measure the hard X-ray continuum {\\it simultaneously} with the highest throughput available (i.e. with \\rxte) in order to compensate for the poor efficiency of the \\chandra gratings at energies above $\\sim 2$ keV. In the present paper we focus on the soft X-ray spectroscopy results; detailed results from the UV data are presented in a companion paper (Kraemer \\etal 2002, hereafter Paper~II). Our campaign was also designed to measure the narrow and broad components of the Fe-K line and associated Compton-reflection continuum, but these results are reported elsewhere (\\cite{yaqoob02b}).  The paper is organized as follows. In \\S\\ref{sec:obs} we present the data and describe the analysis techniques. In \\S\\ref{sec:overall} we discuss gross features of the X-ray spectrum, including the intrinsic continuum form and interpret the data in the context of historical, lower spectral resolution CCD data. In \\S\\ref{sec:features} we qualitatively discuss the discrete X-ray spectral features before describing detailed spectral modeling. In \\S\\ref{sec:modelling} we describe in detail the modeling of the X-ray spectrum using the photoionization code, XSTAR. In \\S\\ref{sec:compare} we compare our results with those from a previous \\xmm grating observation by Pounds \\etal (2001). In \\S\\ref{sec:uv} we discuss the relationship between the X-ray and UV absorbers. Finally, in \\S\\ref{sec:conclusions} we summarize our conclusions. ",
        "conclusions": "\\label{sec:conclusions}  We observed \\mk for $\\sim 59 \\ {\\rm ks}$ with the \\chandra \\hetg on 2001 April 13--14, simultaneously with \\rxte and \\hst. The complex Fe-K line and Compton-reflection continuum are discussed in Yaqoob \\etal (2002b). Details of the UV data and results from the \\hst observation are given in Paper~II. Here we summarize the main results from the \\chandra \\hetg soft X-ray spectrum, in view of relevant constraints from the \\rxte and \\hst observations.  \\begin{enumerate} \\item{Combined \\chandra MEG, HEG and \\rxte data show that the hard X-ray spectrum of \\mk is well described by a simple power law, with little Compton reflection continuum required, and a photon index of $\\Gamma = 1.67$ above 2 keV. At lower energies, below $\\sim 1$ keV, the spectrum steepens, the soft excess attaining a magnitude of about 60\\% higher than the extrapolated hard power law. When the intrinsic X-ray continuum is modeled with a broken power-law, self-consistently modeling the X-ray absorption with a photoionization model (see below) yields a break energy $E_{B}=0.95^{+0.15}_{-0.09} \\ {\\rm keV}$ and a soft power law index of $\\Gamma=2.36^{+0.21}_{-0.20}$. During the observing campaign the broadband source flux was $5.1 \\times 10^{-11} \\rm \\ ergs \\ cm^{-2} \\ s^{-1}$, typical of historical values (the corresponding 2--10 keV luminosity was $1.3 \\times 10^{44} \\rm \\ ergs \\ s^{-1}$).}  \\item{Below $\\sim 2$ keV, the {\\it observed} spectrum shows considerable deviations from a smooth continuum, due mainly to bound-free absorption opacity, and complexes of discrete absorption features. Particularly noteworthy are the \\oxyseven and \\oxyeight edges at $0.732^{+0.004}_{-0.019} \\ {\\rm keV}$ and $0.871^{+0.014}_{-0.020} \\ {\\rm keV}$ respectively. Within the errors, both values are consistent (\\oxyseven marginally so) with their respective rest energies, and from the best-fitting photoionization models the optical depths at threshold are $0.10^{+0.07}_{-0.05}$ for \\oxyseven and $0.09^{+0.01}_{-0.02}$ for O~{\\sc viii}. Compared with a previous \\asca observation (\\cite{reynolds97}; \\cite{george98}), the \\oxyseven edge is consistent, whilst the \\oxyeight edge is somewhat larger in the MEG data. However, we note that the edge depths may be variable in some AGN (e.g. Otani \\etal 1996; Guainazzi \\etal 1996). Estimates of the \\oxyeight edge depth are also complicated by nearby absorption features due, for example, to Ne and Fe. The spectral region between the \\oxyseven and \\oxyeight edges is very complex and the most difficult to model with a simple photoionized absorber. This part of the spectrum is also where one would expect Fe~L edges (e.g. due to dust) and Fe~M-shell unresolved transition arrays (UTA), neither of which are detected. The data, are, however consistent with a UTA with model parameters measured during an \\xmm observation (Pounds \\etal 2001).}  \\item{On the next level of detail, we detect absorption lines from H-like ions of N, O, Ne, He-like ions of O, Ne, Mg, as well as absorption features due to highly ionized Fe. The ionization state of the absorber is high, but not so high that absorption due to H-like Mg or Si is detected. Only the \\nela and \\nenine \\resonetwo lines appear to be marginally resolved, and the rest are unresolved (FWHM$<300$~$\\rm km \\ s^{-1}$ for N, going up to FWHM$<1330$~$\\rm km \\ s^{-1}$ for Mg). The absorption lines are consistent with an outflow of $-200$~$\\rm km \\ s^{-1}$ relative to systemic. However, the detailed velocity profiles of the absorption features are not all the same for all ionic species. The complexity and differences in profiles could be due to the contribution from different, unresolved kinematic components and/or blending and contamination from absorption/emission due to other atomic transitions.}   \\item{We have modeled the \\chandra \\hetg spectra using the photoionization code XSTAR 2.1.d, with a spectral energy distribution (SED) constructed using our UV and X-ray data. The best-fitting ionization parameter and neutral equivalent Hydrogen column density of the absorbing gas are $\\log \\xi=1.76^{+0.13}_{-0.14}$ ergs cm ${\\rm s^{-1}}$ (or $\\log{U}=0.27$), and $N_{H}=2.06^{+0.39}_{-0.45}\\times 10^{21} {\\rm cm^{-2}}$ respectively. This best-fitting model gives an excellent fit to the overall {\\it observed} spectrum. It also is able to model all the principal local absorption features to a degree ranging from fair to excellent, except for the region around 0.78--0.80 keV. The poor fit here may be, in part, due to uncertainties in the instrument response function. A curve-of-growth analysis indicates that a velocity width of $b=100 \\ \\rm km \\ s^{-1}$ gives a good match between the measured absorption-line equivalent widths. Our results are insensitive to reasonable uncertainties in the unobserved EUV part of the SED.}  \\item{We do not detect any clearly identifiable emission lines (apart from the Fe-K line), in contrast with an earlier \\xmm observation in which Pounds \\etal (2001) detected He-like \\oxyseven triplet emission, which constrained the electron density to be $>10^{11} \\rm \\ cm^{-3}$. However, this lower limit is subject to uncertainties in the line-ratio measurements. Although the broadband X-ray flux during that observation was nearly a factor of three lower than it was during our campaign, the signal-to-noise of our data is worse, and insufficient to constrain the distance of the emitter from the ionizing source. Since we cannot assume that the emitter and absorber have the same ionization state and/or column density, time-resolved spectroscopy with better signal-to-noise is required to address the question of the location of the absorber and emitter. Deducing the mass outflow of the X-ray absorber must also await better data because the density and global covering factor are unknown.}   \\item{Simultaneous \\hst observations of \\mk indicate that there are eight kinetic components comprising the UV absorbers, with velocities ranging from $-422$ to $+124$~$\\rm km \\ s^{-1}$. The X-ray absorber is compatible with sharing the same velocity space as the UV absorbers. It is possible that the velocity profiles of the X-ray absorbers are also made up of several kinematic components, possibly the same as the UV ones, but unresolved because of the factor of $>30$ worse velocity resolution of the X-ray  data compared to the UV data. The X-ray profiles do however hint at complex structure, possibly two kinematic components, or groups of components. The UV components certainly appear to cluster into two groups. There are also differences in the X-ray velocity profiles between some of the ionic species. This may either be due to genuine differences in gas dynamics but may also be due to unresolved and unmodeled line emission. On the other hand, other Seyfert~1 galaxies also show two kinematic components (or groups of components) in their absorption-line profiles, but with different offsets relative to systemic (NGC~5548, \\cite{kaastra02}; NGC~4051, \\cite{coll01}; NGC~3783, \\cite{kaspi02}).}  \\item{Although the X-ray and UV absorbers may share the same velocity space, Paper~II shows that the ionization state and column densities of the UV absorbers are too low to produce the observed X-ray absorption. Conversely, the ionization and column density of the X-ray absorber is high; some UV absorption is predicted by our best-fitting photoionization model to the \\chandra data but the relevant ionic column densities can be hidden in the UV absorbers.}  \\end{enumerate}  The authors gratefully acknowledge support from NASA grants NCC-5447 (T.Y.), NAG5-10769 (T.Y.), NAG5-7385 (T.J.T), NAG5-4103 (S.B.K.), and CXO grant GO1-2101X (T.Y., B.M.). This research made use of the HEASARC online data archive services, supported by NASA/GSFC and also of the NASA/IPAC Extragalactic Database (NED) which is operated by the Jet Propulsion Laboratory, California Institute of Technology, under contract with NASA. The authors are grateful to the \\hst, \\chandra and \\rxte instrument and operations teams for making these observations possible, and to Tim Kallman for much advice on XSTAR. The authors also thank Julian Krolik for useful discussions, and an anonymous referee for doing a very thorough job on the manuscript. TY would like to dedicate this paper to his mother, Zubaida Begum Yaqoob, who passed away in March 2001, after sacrificing so much in order that her children could get an education and pursue their goals.   \\newpage \\begin{deluxetable}{lrrrr} \\tablecaption{Absorption Lines in the Chandra HETGS Spectrum of \\mk} \\tablecolumns{5} \\tablewidth{0pt} \\tablehead{ \\colhead{Line $^{a}$} & \\colhead{EW (Data)}$^{b}$ & \\colhead{EW (Model) $^{c}$} & \\colhead{Velocity $^{d}$ } & \\colhead{FWHM} \\nl & \\colhead{(eV)} & \\colhead{(eV)} & \\colhead{($\\rm km \\ s^{-1}$)} & \\colhead{($\\rm km \\ s^{-1}$)} \\nl} \\startdata  \\nila & $0.55^{+0 e}_{-0.15}$ & 0.33 & $-290^{+130}_{-90}$ & $<310$ \\nl  \\nilb & $0.64^{+0e}_{-0.26}$ & 0.67 & $-230^{+100}_{-100}$ & $<320$ \\nl  \\oxla & $0.72^{+0e}_{-0.13}$ & 0.94 & $-85^{+200}_{-135}$ & $<450$ \\nl  \\neniner \\resonetwo ($\\lambda 13.447\\AA$) & $1.33^{+0e}_{-0.30}$ & 1.08 & $-220^{+260}_{-165}$ & $440^{+590}_{-305}$ \\nl  \\neniner  \\resonethree ($\\lambda 11.547\\AA$) & $1.12^{+0.20}_{-0.68}$ & 0.87 & $-180^{+280}_{-250}$ & $<890$ \\nl  \\mgelevenr \\resonetwo ($\\lambda 9.169\\AA$) & $0.68^{+0.48}_{-0.50}$ & 1.05 & $-620^{+330}_{-180}$ & $<1040$ \\nl  \\nela $^{f}$ & $1.89^{+0.74}_{-1.27}$ & 0.96 & $-210^{+155}_{-205}$ & $875^{+460}_{-550}$ \\nl  \\nela $^{f}$ & $1.08^{+0.29}_{-0.39}$ & 0.96 & $-355^{+205}_{-60}$ & 1 (fixed) \\nl \\enddata \\tablecomments{\\small Absorption-line parameters measured from the MEG spectrum, using simple Gaussians (see \\S\\ref{sec:features}). All measured quantities refer to intrinsic parameters, already corrected for the instrument response. Errors are 90\\% confidence for one interesting parameter ($\\Delta C = 2.706$). All velocities have been rounded to the nearest 5~$\\rm km \\ s^{-1}$. $^{a}$ Laboratory-frame wavelengths. $^{b}$ Measured equivalent widths in the \\mk frame. $^{c}$ Predicted equivalent widths using $b=100 \\ \\rm km \\ s^{-1}$ and XSTAR columns (\\S\\ref{sec:comparison}). $^{d}$ Velocity offset (\\mk frame) of Gaussian centroid relative to systemic. Negative values are blueshifts. $^{e}$ No meaningful upper limits due to poor statistics and line saturation. $^{f}$ Different centroids were obtained for \\nela depending on  whether the intrinsic width of the Gaussian was fixed at 1~$\\rm km \\ s^{-1}$ or a free parameter, indicating a complex profile (see \\S\\ref{sec:features} and \\figvelprofone).} \\end{deluxetable}  \\newpage  \\begin{deluxetable}{lr} \\tablecaption{Element Abundances used in XSTAR and CLOUDY models} \\tablecolumns{2} \\tablewidth{0pt} \\tablehead{ \\colhead{Element} & \\colhead{Abundance} } \\startdata H  & $1.00 \\times 10^{0}$ \\nl He & $1.00 \\times 10^{-1}$ \\nl C  & $3.54 \\times 10^{-4}$ \\nl N  & $9.33 \\times 10^{-5}$ \\nl O  & $7.41 \\times 10^{-4}$ \\nl Ne & $1.20 \\times 10^{-4}$ \\nl Mg & $3.80 \\times 10^{-5}$ \\nl Si & $3.55 \\times 10^{-5}$ \\nl S  & $2.14 \\times 10^{-5}$ \\nl Ar & $3.31 \\times 10^{-6}$ \\nl Ca & $2.29 \\times 10^{-6}$ \\nl Fe & $3.16 \\times 10^{-5}$ \\nl Ni & $1.78 \\times 10^{-6}$ \\nl \\enddata \\end{deluxetable}  \\newpage  \\begin{deluxetable}{lrrrrr} \\tablecaption{ Ionic Column Densities ($10^{14} {\\rm \\ cm^{-2}}$)} \\tablecolumns{6} \\tablewidth{0pt} \\tablehead{ \\colhead{Ion } & \\colhead{X-ray  $^{a}$} & \\colhead{UV  $^{b}$} & \\colhead{UV  $^{b}$ } & \\colhead{Ion } & \\colhead{X-ray  $^{a}$} \\nl & \\colhead{(Predicted)} & \\colhead{(Predicted)} & \\colhead{(Measured)} &  & \\colhead{(Predicted)} } \\startdata  H~{\\sc i} & $1.4 \\times 10^{1}$ & $1.6 \\times 10^{1}$ $^{c}$ & $4.7 \\times 10^{1}$ & Mg~{\\sc xi} & $3.0 \\times 10^{2}$ \\nl C~{\\sc ii} &$3.0 \\times 10^{-7}$ & - & $<3.5 \\times 10^{-1}$ & Mg~{\\sc xii} & $4.9 \\times 10^{1}$ \\nl C~{\\sc iii} & $6.7 \\times 10^{-4}$ & $5.5 \\times 10^{-1}$ & $<5.1 \\times 10^{-1}$ & Si~{\\sc xiv} & $3.8 \\times 10^{0}$ \\nl C~{\\sc iv} & $1.2 \\times 10^{-1}$ & 7.3 & 7.3 & Ar~{\\sc xviii} & $1.6 \\times 10^{-3}$  \\nl N~{\\sc v} & $8.5 \\times 10^{-1}$ & 7.1 & 7.0 & \\fetwelve & $8.4 \\times 10^{0}$ \\nl O~{\\sc vi} & $7.1 \\times 10^{1}$ & $1.2 \\times 10^{2}$ & $0.82 \\times 10^{2}$ & \\fefourteen & $1.0 \\times 10^{2}$ \\nl O~{\\sc vii} & $3.5 \\times 10^{3}$ & - & - & \\fefifteen & $1.8 \\times 10^{2}$ \\nl O~{\\sc viii} & $8.1 \\times 10^{3}$ & - & - & \\feseventeen & $1.6 \\times 10^{2}$ \\nl Ne~{\\sc ix} & $1.4 \\times 10^{3}$ & - & - & \\feeighteen & $5.1 \\times 10^{1}$ \\nl Ne~{\\sc x} & $7.5 \\times 10^{2}$ & - & - & \\fenineteen & $7.6 \\times 10^{0}$ \\nl \\tablecomments{\\small $^{a}$ X-ray column densities (in units of $10^{14} \\rm \\ cm^{-2}$) refer to those predicted by the best-fitting XSTAR model to the \\chandra data (see \\S\\ref{sec:modelling}). The predicted column densities for Si~{\\sc iii} and Si~{\\sc iv} from the XSTAR models were negligible. The upper limits on Si~{\\sc iii} and Si~{\\sc iv} from summing over UV kinematic components were $0.21$ and $0.07 \\times 10^{14} \\ {\\rm \\ cm^{-2}}$ respectively. $^{b}$ Predicted and measured UV column densities are summed over available measurements from all UV kinematic components except the low ionization component of 4 (see Paper~II). H~{\\sc i}, C~{\\sc iii} and O~{\\sc vi} measurements are from Kriss \\etal (2000) and {\\it not} the present observations. Predicted UV column densities are from CLOUDY photoionization models of the UV data. $^{c}$ Column from all kinetic components except 4 (low) and $4^{\\prime}$ (see Paper~II).} \\enddata \\end{deluxetable}  \\newpage"
    },
    "0208/astro-ph0208195_arXiv.txt": {
        "abstract": "It was proposed by Blitz et al. (1999) that High Velocity Clouds (HVCs) are remnants of Local Group formation and the average distance of HVCs is 1 Mpc, which is the result of a simple dynamical calculation leading to match the observed HVCs distributions. However, in this paper, we clearly show that fitting the observed HVCs distributions by a dynamical calculation {\\it cannot} provide any constraints on the average distance of HVCs. With our choices of initial conditions, the observational results in Wakker \\& van Woerden (1991) are produced in our simulations for the models of both Galactic and extragalactic origins. Moreover, because Zwaan \\& Briggs (2000) reported that they failed to locate any extragalactic counterparts of the Local Group HVCs in a blind HI 21-cm survey of extragalactic groups, we  propose to use ``remnants of galactic disc formation'' as the modification for the picture of ``remnants of galaxy group formation'' in Blitz et al. (1999) and thus reduce the possible average distances of HVCs to be about or less than a few hundred kpc. ",
        "introduction": "High Velocity Clouds (HVCs) are HI clouds at velocities incompatible with a simple model of Galactic rotation.  Most HI clouds are the tracers of the galactic discs for both the Galaxy and the extra-galaxies. The reason why HVCs are interesting is that they do not belong to the majority of the HI clouds in the Milky Way and therefore their nature and origins need to be understood.  The well-known proposed HVCs' possible origins usually fall into two categories: (a) The Local Group Formation -- the remnants of galaxy group formation, (b) Galactic Fountain -- the material injected from the galactic disc \"after\" the disc was formed. HVCs' distances are the most important key parameters to understand their possible origins.   Blitz et al.(1999) argued that HVCs should be extragalactic because it is difficult to understand why HVCs would not be metal rich or how the vertical velocities could be so high in the Galactic Fountain context. Incidently, their simulation, which seems to produce the observed HVCs' distribution on the sky, suggested the mean HVC distance is about 1 Mpc.   On the other hand, Zwann $\\&$ Briggs(2000) reported the result different form the hypothesis in Blitz et al. (1999). In a blind HI 21-cm survey of extragalactic group, they failed to locate any extragalactic counterparts of the Local Group HVCs.  \\begin{figure}[t] \\epsfysize 400 pt \\plotone{fig1_J3.ps} \\caption{The  $l-b$ plots for all models, where circles are observational data from Wakker \\& van Woerden (1991) and cross-points are our simulational results: (a) The Model of Galactic Origins, (b) The Model of Extragalactic Origins, (c) The Mixed Model, (d) The Model of HVC Complexes. } \\end{figure}  \\begin{figure}[t] \\epsfysize 400 pt \\plotone{fig2_J3.ps} \\caption{The  $l-v_{\\rm los}$ plots for all models, where circles are observational data from Wakker \\& van Woerden (1991) and cross-points are our simulational results: (a) The Model of Galactic Origins, (b) The Model of Extragalactic Origins, (c) The Mixed Model, (d) The Model of HVC Complexes. Please note that those cross-points with $|v_{\\rm los}| < 50$ km/s should be ignored since they do not satisfy the usual definition of HVCs. } \\end{figure} ",
        "conclusions": "We found that most HVCs' observational distribution on the sky ($l-b$ plane) and also $l-v_{\\rm los}$ plane can be produced for all our different models, including both Galactic origin and extragalactic origin models. Therefore, we conclude that fitting the observed HVCs distributions by a dynamical calculation {\\it cannot} provide any constraints on the average distance of HVCs.   The mechanisms to produce HVCs of Galactic origins can be the galactic fountain model, stream of satellite galaxies (as the possible stream from Sagittarius dwarf galaxy in Figure 6 of Jiang \\& Binney 2000) and also remnants of galaxy group formation.  On the other hand, the mechanisms to produce HVCs of extragalactic origins can be only remnants of galaxy group formation unless these HVCs are all close to M31 and made by the mechanisms of Galactic origin but within M31.  It is easy to see that there will be no constraint on the distance at all if HVCs are the remnants of galaxy group formation and thus this suggestion seems to be able to explain the origins of most HVCs. Especially, if these HVCs have low-metallicity, other models of Galactic origins would be less attractive. Therefore, although one should keep in mind that it is always a correct concept that HVCs are multiple origins, the  ``remnants of galaxy group formation'' is probably a good term to describe most HVCs when one has to make a choice.  However,  Zwaan \\& Briggs (2000) make a point that there should not be so many HVCs in the intergalactic space as predicted by Blitz et al. (1999) and give the picture of ``remnants of galaxy group formation'' a difficult time.   Therefore, in order to resolve the contradiction between Blitz et al. (1999) and Zwaan \\& Briggs (2000), we propose that HVCs are ``remnants of galactic disc formation'' and reduce the possible average distances of HVCs to be about or less than a few hundred kpc.  This conclusion implied from our simulations completely agrees  with recent observational results by Tufte et al. (2002), in which they observed $H_{\\alpha}$ lines of HVCs which are candidates for being at larger than average distance and found these clouds are in the Galactic halo and not distributed throughout the Local Group."
    },
    "0208/astro-ph0208476_arXiv.txt": {
        "abstract": "AE~Aquarii is a propeller system. It has the shortest spin period among cataclysmic variables, and this is increasing on a $10^7$~yr timescale. Its UV spectrum shows very strong carbon depletion vs nitrogen and its secondary mass indicates a star far from the zero--age main sequence. We show that these properties strongly suggest that AE~Aqr has descended from a supersoft X--ray binary. We calculate the evolution of systems descending through this channel, and show that many of them end as AM~CVn systems. The short spindown timescale of AE~Aqr requires a high birthrate for such systems, implying that a substantial fraction of cataclysmic variables must have formed in this way. A simple estimate suggests that this fraction could be of order one--third of current CVs. We emphasize the importance of measurements of the C/N abundance ratio in CVs, particularly via the {\\sc Civ} 1550/{\\sc Nv} 1238 ratio, in determining how large the observed fraction is. ",
        "introduction": "AE~Aqr is one of the most distinctive cataclysmic variables (CVs). It has a long orbital period of 9.88~hr \\citep{Welsh_etal:637}, and the shortest coherent pulse period (33~s), increasing on a timescale $\\sim 10^7$~yr \\citep{Jager_etal:1018}. Doppler tomography reveals the apparent absence of an accretion disc. This has led to its interpretation as a `propeller' system \\citep{Wynn_etal:624}: the white dwarf is apparently expelling centrifugally the matter transferred from the companion. To arrive at such a highly non--equilibrium spin rate, the mass transfer rate in the recent ($10^7$~yr) past must have been much higher than its current value $\\dot{M}_2 \\sim$ few $\\times \\, 10^{-9} \\, \\msun {\\rm yr}^{-1}$, and decreased on a still shorter timescale. IUE spectra of AE~Aqr \\citep{Jameson_etal:1019} reveal the most extreme {\\sc Civ} to {\\sc Nv} ratio of all CVs \\citep{Mauche_etal:661}, probably indicating strong carbon depletion and thus CNO cycling. The short spin period makes the system effectively a double--lined spectroscopic binary, with masses $M_1 \\simeq 0.89 \\pm 0.23 \\, \\msun$ for the white dwarf (WD) and $M_2 \\simeq 0.57 \\pm 0.15 \\, \\msun$ for the donor star \\citep{Welsh_etal:625,Casares_etal:623}. The donor star is of spectral type K5V \\citep{Welsh:668} which is too late for the given orbital period and too early for the measured $M_2$ if it is on the main sequence.  We show here that all of these properties are consistent with the idea that AE~Aqr descends from a supersoft X--ray binary, cf.\\ \\citet{Schenker+King:892}. The thermal--timescale mass transfer characterizing such sources ends once the secondary/primary mass ratio $q = M_2/M_1$ decreases sufficiently, leading to a rapid transition to normal CV evolution driven by angular momentum losses, cf.\\ \\citet{King:19}. If the white dwarf is magnetic, as in AE~Aqr, the high mass transfer rate in the thermal--timescale phase will have spun it up to a short spin period. This must lengthen rapidly in order to reach equilibrium with the lower transfer rate in the CV state, and this spindown is what drives its propeller action. The very short lifetime of the propeller phase implies a high birthrate for such systems, comparable to those of known CVs. This in turn suggests that systems descending from evolutions like AE~Aqr must constitute a large fraction of all CVs.   The simple idea that all CVs form with essentially unevolved, low--mass main sequence (MS) donors has been repeatedly challenged for more than a decade \\citep{Pylyser+Savonije:587,Pylyser+Savonije:588, Baraffe+Kolb:613,Schenker:724,King+Schenker:891,Schenker+King:892}. If mass transfer starts with mass ratio $q \\ga 1$ the secondary's Roche lobe $R_{\\rm L}$ tends to shrink with respect to its thermal--equilibrium radius $R_{\\rm TE}$. Mass is therefore transferred on a thermal timescale ($\\sim$ few $\\times 10^7$~yr), at rates high enough to sustain steady nuclear burning on the material accreted by the white dwarf, and thus explain (short-period) supersoft X-ray sources \\citep{Heuvel_etal:583,King_etal:640}.  After this relatively brief phase, $q$ becomes small enough that $R_{\\rm L}$ shrinks more slowly than $R_{\\rm TE}$, and thermal--timescale mass transfer ends. The systems either evolve off to longer orbital periods, driven by nuclear evolution, or switch to stable, angular-momentum--loss-driven mass transfer like ordinary CVs. We will show that AE~Aqr is at the end of this transition phase to a CV, accounting for its strange properties. AE~Aqr is passing through this transition very rapidly; its whole evolution up to its current state is short compared to the stable CV phase afterwards. In this sense AE~Aqr is not unique at all: a significant fraction of CVs must descend from similar evolutions.   In the next section we derive some rough estimates for the potential progenitor system of AE~Aqr, using a simplified analytic description of the orbital evolution during rapid mass transfer. Sect.~3 studies the behaviour of single star models under fixed mass loss of the order expected during such an evolution. This allows us to identify the influence of varying stellar parameters, and establish whether the observed properties of AE~Aqr are compatible with this scenario. We present a set of full binary evolution calculations in Sect.~4 and discuss them in Section~5. Section~6 is the Conclusion. ",
        "conclusions": "We have shown that the assumption of \\TTMT\\ in the recent past of AE~Aqr provides plausible explanations for all of its current observational properties. The system parameters for the best-fitting progenitor model presented in this paper (B6) are $M_{2i} = 1.6 \\, \\msun$ (fairly far evolved on the MS) and $P_i = 18.6 \\, {\\rm hr}$ with a primary of $M_{1i} = 0.6 \\, \\msun$ which manages to accrete about 30\\% of the total transferred mass upon reaching the current phase of AE~Aqr. This indicates a relatively large initial mass ratio, and some fraction of mass accretion during the \\TTMT\\ phase. Although the precise values may change, the idea is very robust, as the differential results from Sect.~3 \\& 4 clearly indicate that it will be always possibly to end up with an excellent model for AE~Aqr.  The short duration of the \\TTMT\\ and the transition stage (where AE~Aqr currently is) implies a large birthrate, and thus suggests a large number of systems passing through similar evolutions. We expect descendants from systems similar to AE~Aqr, and thus from supersoft binaries, to form a substantial fraction of the currently known CVs. The contortions of the $-\\dot M_2 - P$ curves of Fig.\\ \\ref{fig:types} suggest that non--magnetic descendant systems may change between recurrent novae, novalike and dwarf nova behaviour during and after the transition from \\TTMT\\ to normal CV evolution. Descendants with significant WD magnetic fields will also appear in various guises during these phases. We suggest that the long--period AM~Her system V1309~Ori \\citep{Schmidt+Stockman:713,Staude_etal:1020} is a descendant of a supersoft binary, cf.\\ \\citet{King_etal:896}. This system is apparently able to synchronise at its unusually long period of 8~hr because of the drop in mass transfer rate at the end of the \\TTMT\\ phase. The pulsing supersoft source XMMU~J004319.4+411758 found by XMM in M31 may be an example of a progenitor still in the \\TTMT\\ phase \\citep{King_etal:896}. Note that V1309~Ori, the slightly nonsynchronous polar BY~Cam \\citep{Bonnet-Bidaud+Mouchet:10001} and the intermediate polar TX Col \\citep{Mouchet_etal:10000} all show high {\\sc Nv}/{\\sc Civ} ratios similar to AE~Aqr. Abundance anomalies possibly related to stripping of a partially evolved companion were suggested by \\citet{Mouchet_etal:10000} in the latter two cases.  We conclude that in spite of its apparent uniqueness, AE~Aqr is only the first confirmed member of a much larger population of post--supersoft binaries, constituting a significant fraction of all CVs. A way of checking for this population is to determine the C/N ratio by measuring {\\sc Civ} 1550 versus {\\sc Nv} 1238. A low value here will be strongly suggestive of descent from a supersoft binary."
    },
    "0208/astro-ph0208126_arXiv.txt": {
        "abstract": "We use the Zel'dovich approximation to analyse the amplification of magnetic fields in gravitational collapse of cold dark matter during the mildly nonlinear regime, and identify two key features. First, the anisotropy of collapse effectively eliminates one of the magnetic components, confining the field in the pancake plane. Second, in agreement with recent numerical simulations, we find that the shear anisotropy can amplify the magnetic field well beyond the isotropic case. Our results suggest that the magnetic strengths observed in spiral and disk galaxies today might have originated from seeds considerably weaker than previous estimates. ",
        "introduction": "Large scale magnetic fields, with strengths $10^{-7}$--$10^{-5}$~G, have been observed in spiral and disc galaxies, galaxy clusters and high redshift condensations [see~\\cite{K,HW} and references therein]. The most promising explanation for the large scale galactic fields has been the dynamo mechanism, with the required seeds coming either from local astrophysical processes, such as battery effects, or from primordial magnetogenesis [for recent reviews, see~\\cite{GR,Wid}]. The linear evolution of large scale magnetic fields and their implications for structure formation has been studied by several authors [see, e.g.,~\\cite{RR,W,PE,EF,TB,BS,TM}]. Certain aspects of the mildly nonlinear clustering can be analysed in spherical symmetry, but this approximation inevitably breaks down as the collapse proceeds and any initially small anisotropies take over. When magnetic fields are involved, the need to incorporate anisotropic effects is particularly important, as the fields are themselves generically anisotropic sources. Numerical simulations of magnetic field evolution in galaxy clusters suggest that anisotropies lead to additional amplification of the field. Tidal effects during mergers, for example, increase the magnitude of the magnetic field~\\cite{RSB}. Shear flows in galaxy clusters can also amplify the field beyond the limits of spherical compression~\\cite{DBL1,DBL2}. The latter simulations also suggest that the final magnetic configuration is effectively independent of the field's initial set up or of the presence of a cosmological constant.  Here we use the Zel'dovich approximation to look analytically into the effect of gravitational anisotropy on seed magnetic fields in the mildly nonlinear regime. The anisotropic collapse is driven by cold dark matter (CDM). A previous analysis~\\cite{ZRS}, where the matter is purely baryonic, concluded that the seeds must be negligibly small in order to avoid the growth of magnetic fields to levels which prevent pancake formation. By contrast, this strict constraint is avoided when CDM dominates, since the magnetic field couples only gravitationally with the CDM. The field is frozen into the baryon fluid, and baryons are dragged by the CDM gravitational field. The baryons are in a different state of motion to the CDM and, unlike the CDM, they feel the magnetic backreaction. As a first approximation, however, we will ignore the magnetic backreaction and the relative motion of the baryons, effectively considering a single fluid. This approximation also maintains the acceleration-free and irrotational nature of the motion, which are key ingredients of the Zel'dovich approximation. Our approach may be seen as a qualitative starting point for a more detailed analysis.  We begin in Section 2 with a dynamical system description of the Zel'dovich approximation, which directly shows that, in a generic collapse, pancakes are the (local) attractors~\\cite{B}. In Section 3 we consider the dynamics of the magnetic field as it collapses with the matter. The magnetized dynamical system is five-dimensional. Pancakes are still the attractors, with the magnetic field squeezed in the pancake plane. Note that, as the galaxy is formed, tidal forces are generally expected to change the orientation of the galactic plane relative to the pancake. Nevertheless, the confinement of the field in the pancake plane that we find here, is qualitatively consistent with magnetic field observations in numerous spiral and disc galaxies. We also provide some quantitative results by relating the growth of the field to that of the matter density contrast. When comparing the shearing to the isotropic collapse, we find that anisotropy can lead to an appreciable increase in the amplification of the magnetic field. These analytical results are in qualitative agreement with those of the earlier mentioned numerical simulations. We interpret them as an indication that the magnetic strengths observed in numerous galaxies and galaxy clusters today could have resulted from seeds considerably weaker than previous estimates. ",
        "conclusions": "Protogalactic clouds do not collapse isotropically: in any realistic situation one expects small anisotropies in the initial velocity distribution, which will be amplified as CDM collapse progresses. We used the Zel'dovich approximation to investigate the effects of such gravitational anisotropy on the evolution of a seed magnetic field frozen into the baryon fluid. We considered the mildly nonlinear regime and ignored the magnetic backreaction on the baryons as well as the relative velocity between baryons and CDM. The CDM dominates the collapse and the gravitational anisotropy that amplifies the magnetic field. Our qualitative analysis shows that, as a generic result of the anisotropy of the collapse, the field effectively ``loses'' one of its components and is confined in the plane of the pancake. Although tidal forces are generally expected to change the orientation of the galactic plane relative to the pancake, our qualitative picture is consistent with magnetic field observations in numerous spiral and disk galaxies. More quantitatively, our results show a much more efficient magnetic amplification in a shearing rather than in a shear-free collapse. This analytical result agrees with numerical simulations of shear and tidal effects on the evolution of magnetic fields in galaxies and galaxy clusters. Taken together, these results indicate that, when the anisotropic effects of CDM collapse are taken into account, the magnetic strengths observed in galaxies today could have originated from seeds considerably weaker than previous estimates."
    },
    "0208/astro-ph0208310_arXiv.txt": {
        "abstract": "The infrared search for substellar companions to nearby white dwarfs has been going for a little more than a decade.  The most recent phase has been a wide field proper motion search carried out primarily at Steward Observatory, where we are complete down to $J=18$. Earlier phases included near field searches at the IRTF and Keck Observatory.  In the last year we have discovered ten previously unrecognized faint proper motion companions.  Of the recent discoveries, most are white dwarfs and a few M dwarfs.  GD165B, discovered in 1988 as part of our program, is still the only known companion to a white dwarf with spectral type later than M. ",
        "introduction": "We are conducting a proper motion survey at $1.25\\mu$m for low mass stellar and substellar companions to nearby white dwarfs.  The advantage of such a search is that white dwarfs are faint in the infrared, enabling discovery of brown dwarf companions even relatively close to the primary (Zuckerman \\& Becklin 1987 \\& 1992; Becklin \\& Zuckerman 1988).  The current wide field survey has been conducted mainly at the Steward Observatory beginning in 1991.  There on the 2.3m Bok telescope we have used a NICMOS array and a camera developed by Rieke \\& Rieke (Rieke et al. 1993).  With a 3 square arcminute field of view, we can detect faint companions $2-90''$ from the primary white dwarf.  We acquire five dithered 90 second images at $J$ band which enables detection of objects as faint as $J=19$ in the best observing conditions.  Generally we acquire two epochs for each field surrounding the primary white dwarf. The typical white dwarf in our sample has a small known proper motion around $0.1-0.2''$/yr. Since the mid 1990's, the white dwarfs will have moved $\\sim0.5-1.0''$ relative to background stars and galaxies.  We have found that when there are more than 5 objects which appear in both epoch fields, the scatter in their measured positions is approximately $0.2''$.  Thus, we can easily detect a $0.5''$  displacement of the primary relative to background objects in the field.  Hence with a baseline $\\geq5$ yrs between epochs, companions that move with the primary will stand out well against background objects.  The white dwarfs chosen for our survey come from the catalog of McCook \\& Sion (1999). Our targets consist mainly of white dwarfs with modest known proper motions selected from the Lowell \\& Luyten surveys along with some objects from the Palomar-Green survey. The reason for choosing white dwarfs with smaller proper motions is that stars with smaller U,V space velocities are statistically more likely to be members of the young disk (Eggen 1996) and brown dwarfs are brightest when they are young.  White dwarfs are evolved but not necessarily old; a $3M_{\\odot}$ A0 star can evolve into a white dwarf in only 0.6 Gyr.  Hence by selecting slow movers there is a greater likelihood that we are looking around younger white dwarfs.  The distance to a typical white dwarf in our sample of $\\sim240$ targets is around 50 pc. ",
        "conclusions": "The search for brown dwarfs around white dwarfs is of continuing interest for several reasons.  One is that do not yet know which temperature / spectral class is the lower cut off for a minimum mass star.  Until we start to discover and study more companions which straddle and cross the stellar / substellar boundary, the model predicted minimum temperature value is the only way we can judge whether an object is a brown dwarf or a low mass star if an accurate age is not known.  Second, by searching for M, L, and T-type companions to white dwarfs, we are probing ages when brown dwarfs have cooled significantly and low mass stars of course have not.  When our search or future searches begin to find low temperature (L-type) companions around old stars such as some white dwarfs, we can then better constrain stellar / substellar cooling model temperatures, ages and masses.  Also by conducting this kind of search we want to shed light on the formation of companions to intermediate mass stars, the progenitors of the white dwarfs we see today."
    },
    "0208/astro-ph0208583_arXiv.txt": {
        "abstract": "We constructed hydrodynamical model atmospheres for mid M-type main-, as well as pre-main-sequence objects. Despite the complex chemistry encountered in such cool atmospheres a reasonably accurate representation of the radiative transfer is possible. The detailed treatment of the interplay between radiation and convection in the hydrodynamical models allows to study processes usually not accessible within the framework conventional model atmospheres. In particular, we determined the efficiency of the convective energy transport, and the efficiency of mixing by convective overshoot. The convective transport efficiency expressed in terms of an equivalent mixing-length parameter amounts to values around $\\approx 2$ in the optically thick, and $\\approx 2.8$ in the optically thin regime. The thermal structure of the formally convectively stable layers is little affected by convective overshoot and wave heating, i.e. stays close to radiative equilibrium. Mixing by convective overshoot shows an exponential decline with geometrical distance from the Schwarzschild stability boundary. The scale height of the decline varies with gravitational acceleration roughly as $g^{-\\frac{1}{2}}$, with 0.5 pressure scale heights at \\mbox{$\\log \\mathrm{g}$}=5.0. ",
        "introduction": "The increasing number of late M-type stars, brown dwarfs, and extrasolar planets discovered by infrared surveys and radial velocity searches has spawned a great deal of interest in the atmospheric physics of these objects. Their atmospheres are substantially cooler than e.g. the solar atmosphere, allowing the formation of molecules, or even liquid and solid condensates. Convection is a ubiquitous process in these atmospheres shaping the thermal structure and the distribution of material.  Hydrodynamical simulations of solar and stellar granulation including a realistic description of radiative transfer have become an increasingly powerful and handy instrument for studying the influence of convective flows on the the structure of late-type stellar atmospheres as well as on the formation of their spectra. Here we report on efforts to construct hydrodynamical models for mid M-type atmospheres. We consider this as an intermediate step on the way to model brown dwarf and planetary atmospheres. The motivation was twofold: Firstly, pre-main-sequence evolutionary models of M-dwarfs and brown dwarfs based on mixing-length theory (MLT) depend sensitively on the mixing-length parameter~\\mbox{$\\alpha_{\\mathrm{MLT}}$} (Baraffe et al. 2002). Secondly, the distribution of dust clouds in brown dwarfs depends on the efficiency of mixing by convective overshoot. Conventionally, convection is described within the simplistic picture of MLT dependent on free parameters. Our hydrodynamical models provide a description essentially from first principles free of the uncertainties of MLT, putting stellar models on a firmer footing. ",
        "conclusions": ""
    },
    "0208/astro-ph0208060_arXiv.txt": {
        "abstract": "In this paper, I use the extension of the excursion set model of Sheth \\& Tormen (2002) and the barrier shape obtained in Del Popolo \\& Gambera (1998) to calculate the unconditional halo mass function, and the conditional mass function in several cosmological models. I show that the barrier obtained in Del Popolo \\& Gambera (1998), which takes account of tidal interaction between proto-haloes, is a better description of the mass functions than the spherical collapse and is in good agreement with numerical simulations (Tozzi \\& Governato 1998, and Governato et al. 1999). The results are also in good agreement with those obtained by Sheth \\& Tormen (2002), only slight differences are observed expecially at the low mass end. I moreover calculate, and compare with simulations, the temperature function obtained by means of the mass functions previously calculated and also using an improved version of the M-T relation, which accounts for the fact that massive clusters accrete matter quasi-continuously, and finally taking account of the tidal interaction with neighboring clusters. Even in this case the discrepancy between the Press-Schecter predictions and simulations is considerably reduced. ",
        "introduction": "In the most promising cosmological scenarios, structure formation is traced back to the hierarchical growth of primordial Gaussian density fluctuations originated from quantum fluctuations (Guth \\& Pi 1982; Hawking 1982; Starobinsky 1982; Bardeen et al. 1986 -  hereafter BBKS). Starting from these fluctuations, collapsed, virialized dark matter haloes condensed out. Within these haloes gas cools and stars form (White \\& Reese 1977; White \\& Frenk 1991; Kauffmann et al. 1999). So, the structure of dark matter haloes is of fundamental importance in the study of the formation and evolution of galaxies and clusters of galaxies. From the theoretical point of view, the structure of dark matter haloes can be studied both analytically and numerically. An analytical model that has achieved a wide popularity is the Press-Schecter (1974) (hereafter PS) formula, which allows one to compute good approximations to the mass function. PS and Bond et al. (1991) gave a detailed description of the PS statistics together with an understanding of its dynamical basis. In the PS and Bond et al (1991) papers, the authors described how the statistical properties of the initial density field, assumed to be Gaussian, together with the spherical collapse model, could be used to derive an estimate of the number density of collapsed dark matter haloes at later times, the so called universal ``unconditional\" mass function. Lacey and Cole (1993), showed how the model could be extended to estimate the merging rate of small objects to form larger ones, thus leading to the possibility of estimating the ``conditional\" mass function of sub-haloes within parent haloes. Mo \\& White (1996) applied the model to compute an approximation to the spatial clustering of dark haloes.  Although the analytical framework of the PS model has been greatly refined and extended (as testified by the previous cited papers), it is well known that the PS mass function, while qualitatively correct, disagrees with the results of N-body simulations. In particular, the PS formula overestimates the abundance of haloes near the characteristic mass $M_{\\ast}$ and underestimates the abundance in the high mass tail (Efstathiou et al. 1988; White, Efstathiou \\& Frenk 1993; Lacey \\& Cole 1994; Tozzi \\& Governato 1998; Gross et al. 1998; Governato et al. 1999). The quoted discrepancy is not surprising since the PS model, as any other analytical model, should make several assumptions to get simple analytical predictions. As previously reported, the main assumptions that the PS model combines are the simple physics of the spherical collapse model with the assumption that the initial fluctuations were Gaussian and small. On average, initially denser regions collapse before less dense ones, which means that, at any given epoch, there is a critical density, $\\delta_c(z)$, which must be exceeded if collapse is to occur. In the spherical collapse model, this critical density does not depend on the mass of the collapsed object. Taking account of the effects of asphericity and tidal interaction with neighbors, Del Popolo \\& Gambera (1998) and Sheth, Mo \\& Tormen (2001) (hereafter SMT), using a parametrization of the ellipsoidal collapse, showed that the threshold is mass dependent, and in particular that of the set of objects that collapse at the same time, the less massive ones must initially have been denser than the more massive, since the less massive ones would have had to hold themselves together against stronger tidal forces. In the second hand, the Gaussian nature of the fluctuation field means that a good approximation to the number density of bound objects that have mass $m$ at time $z$ is given by considering the barrier crossing statistics of many independent and uncorrelated random walks, where the barrier shape $B(m,z)$, is connected to the collapse threshold. Simply changing the barrier shape, SMT showed that it is possible to incorporate the ``quoted effects\" \\footnote{Namely that in the case of objects collapsing at the same time, the less massive regions must initially have been denser than the more massive ones.} in the excursion set approach. Moreover, using the shape of the modified barrier in the excursion set approach, it is possible to obtain a good fit to the universal halo mass function. \\footnote{Note that at present there is no good numerical test of analytic predictions for the low mass tail of the mass function.} As previously reported, the excursion set approach allows one to calculate good approximations to several important quantities, such as the ``unconditional\" and ``conditional\" mass functions. Sheth \\& Tormen (2002) (hereafter ST) provided formulas to calculate these last quantities starting from the shape of the barrier. They also showed that while the ``unconditional\" and ``conditional\" mass function is in good agreement with results from numerical simulations, neither the constant nor the moving barrier models (barrier obtained from non-spherical collapse) were able to describe the simulations results at small lookback times, in the case of the rescaled (in terms of $\\nu$) ``conditional\" mass function. The reason for this discrepancy is probably due to the excursion set approach's neglect of correlations between scales (Peacock \\& Heavens 1990; Bond et al. 1991; ST) or to the too simple parametrization of the ellipsoidal collapse outlined in SMT.  In the present paper, I'll use the barrier shape obtained in Del Popolo \\& Gambera (1998), obtained from the parametrization of the nonlinear collapse discussed in that paper, together with the results of ST in order to study the ``unconditional\" and ``conditional\" mass function. Finally, I'll calculate the temperature function for a CDM model by means of the mass functions previously obtained and using an improved version of the M-T relation obtained by Voit (2000), which accounts for the fact that massive clusters accrete matter quasi-continuously, and consequently that the M-T relation evolves, with time, more modestly than what expected in previous models (top-hat model) and taking account of the tidal interaction with neighboring clusters.  The reasons that motivates this study are several:\\\\ a) to study the effects of a barrier different from that used by ST on both ``unconditional\" and ``conditional\" mass function, as proposed even by ST ; \\\\ b) to study how well ST formulas really do work for several barrier shapes;\\\\ c) to test if the discrepancies between the temperature function and simulations, observed in several papers (e.g. Governato et al. 1999) are reduced using the mass function and the M-T relation obtained.  The paper is organized as follows: in Sect. ~2, I calculate the ``unconditional\" and ``conditional\" mass functions. In Sect. ~3, I introduce a model for the mass-temperature (M-T) relation and calculate the temperature function. Sect. ~4 and 5 are devoted to results and to conclusions, respectively. ",
        "conclusions": "In this paper, I calculated the unconditional, conditional mass function and temperature function by using the extension of the excursion set model of ST and the barrier shape obtained in Del Popolo \\& Gambera (1988). I showed that the barrier obtained in Del Popolo \\& Gambera (1998), which takes account of asphericity and tidal interaction between proto-haloes, is a better description of the mass functions and temperature function than the spherical collapse and is in good agreement with numerical simulations. The results are in good agreement with those obtained by ST, only some differences are observed expecially at the low mass end.  The main results of the paper can be summarized as follows: \\\\ 1) the non-constant barrier obtained from the non-spherical collapse in Del Popolo \\& Gambera (1998), taking account of the tidal interaction of proto-clusters with neighboring ones, combined with the ST model gives ``unconditional\" and ``conditional\" mass functions, is in reasonably good agreement with results from numerical cosmological simulations. \\\\ 2) The ``unconditional\" and ``conditional\" mass functions obtained with the Del Popolo \\& Gambera (1998) barrier are slightly different from those obtained by SMT and ST: smaller values at small and large $\\nu$, with respect to SMT and ST predictions, and larger values for $0.1 \\leq \\nu \\leq 4$. In the case of the rescaled ``conditional\" mass function the discrepancy observed by ST for small lookback time is smaller in the model of this paper at small $\\nu$ ($0.01 \\leq \\nu \\leq 1$). The discrepancy with simulations is connected to SMT parametrization of the collapse and not only to the neglect of correlations between scales.\\\\ 3) The mass function in SCDM \\footnote{Similar conclusion is valid for the OCDM.} is in good agreement with Governato et al. (1999) and Tozzi \\& Governato (1998) simulations.\\\\ 4)  The temperature function calculated by means of the mass functions obtained in the present paper together with an improved version of the M-T relation, \\footnote{As described in the text, the new M-T relation accounts for the fact that massive clusters accrete matter quasi-continuously, and consequently that the M-T relation evolves, with time, more modestly than what expected in previous models (top-hat model)} and taking account of the tidal interaction with neighboring clusters is in good agreement with the simulations of SCDM universes of Tozzi \\& Governato (1998) for different redshifts. \\\\ 5) The ST formulae really do work  for different barrier shapes (at least that used in this paper, that introduced in SMT and that of Monaco (1997a,b)).\\\\ 6) The behavior of the ``unconditional\" mass function at small masses is similar to that of Sheth \\& Tormen (1999), ST, and very different from that proposed by Jenkins et al. (2001) (see ST Fig. 13).  The above considerations show that it is possible to get accurate predictions for a number of statistical quantities associated with the formation and clustering of dark matter haloes by incorporating a non-spherical collapse in the excursion set approach. The improvement is probably connected also to the fact that incorporating the non-spherical collapse with increasing barrier in the excursion set approach results in a model in which fragmentation and mergers may occur, effects important in structure formation."
    },
    "0208/astro-ph0208256_arXiv.txt": {
        "abstract": "{Cosmological weak lensing gives rise to correlations in the ellipticities of faint galaxies. This cosmic shear signal depends upon the matter power spectrum, thus providing a means to constrain cosmological parameters. It has recently been proposed that {\\em intrinsic} alignments arising at the epoch of galaxy formation can also contribute significantly to the observed correlations, the amplitude increasing with decreasing survey depth. Here we consider the two-point shear correlation function, and demonstrate that photometric redshift information can be used to suppress the intrinsic signal; at the same time Poisson noise is increased, due to a decrease in the effective number of galaxy pairs.  The choice to apply such a redshift-depending weighting will depend on the characteristics of the survey in question. In surveys with $\\ave z\\sim 1$, although the lensing signal dominates, the measurement error bars may soon become smaller than the intrinsic alignment signal; hence, in order not to be dominated by systematics, redshift information in cosmic shear statistics will become a necessity. We discuss various aspects of this. ",
        "introduction": "The distortion of distant galaxies by the tidal gravitational field of intervening matter inhomogeneities has become known as ``cosmic shear''. Since this lensing signal depends upon the matter power spectrum, it is an important cosmological tool, as was proposed in the early 1990s by Blandford et al. (1991), Miralda-Escud\\'e (1991) and Kaiser (1992). This set the scene for further analytic and numerical work (eg. Kaiser 1998; Schneider et al. 1998; White \\& Hu 2000), taking into account the non-linear evolution of the power spectrum which results in increased power on small scales (Hamilton et al. 1991; Peacock \\& Dodds 1996).  During 2000, four teams announced the first observational detections of cosmic shear (Bacon et al. 2000; Kaiser et al. 2000; van Waerbeke et al. 2000; Wittman et al. 2000; Maoli et al. 2001), demonstrating the feasibility of its study. Various statistics are used to quantify cosmic shear, and to compare the observations with predictions for cosmological models. Here we focus on the shear correlation functions; other measures include the aperture mass statistic (Schneider et al. 1998) and shear variance (e.g. Kaiser 1992).  Besides its dependence on the large-scale structure in the Universe, the magnitude of the effect is also sensitive to the redshift distribution of the galaxies used in the analysis. So far, direct redshift estimates have not been obtained for the samples concerned. Typically the mean redshift has been estimated from surveys of similar depth, and a corresponding redshift probability distribution has been assumed (e.g. van Waerbeke et al. 2001; Hoekstra et al. 2002a).  Motivated by the recent interest in obtaining photometric redshifts for cosmic shear surveys, we consider how to make use of redshift information, and its impact on the constraints permitted by the two-point statistic under consideration.  In weak lensing analyses, it is assumed that the background galaxies are randomly oriented so that their mean intrinsic ellipticity $\\ave{\\epsilon^{\\rm s}}=0$ and any correlation in observed ellipticities arises from gravitational lensing. The magnitude of any {\\em intrinsic} alignment of the background source population has recently been the subject of many numerical, analytic and observational studies (e.g. Croft \\& Metzler 2000; Heavens et al. 2000; Crittenden et al. 2001 (Cr01); Catelan et al. 2001; Mackey et al. 2002; Brown et al. 2002, Hui \\& Zhang 2002). Due to differences in the assumed origin of intrinsic alignments and the type of galaxies considered, the numerical and analytic estimates span a couple of orders of magnitude in amplitude. For example, the analytic model of Cr01 assumes that any intrinsic signal arises from correlations between the angular momenta of galaxies, whereas that of Catelan et al. (2001) relies on tidal shear correlations. Nonetheless, all studies conclude that intrinsic alignments are more important for shallower surveys (becoming comparable to the lensing signal for a mean survey redshift ${\\ave z}\\sim 0.5$) and that the correlation falls off quite rapidly with source separation.  The structure of this paper is as follows. In the next section we outline the relationships between the matter power spectrum and the lensing correlation functions. The dependence on the redshift distribution and on cosmology is highlighted.  Correlation functions are derived from observational data, and in ${\\rm Sect.\\,3}$ we describe how practical estimators are used for this purpose.  In ${\\rm Sect.\\,4}$ a toy model to account for intrinsic alignments is developed; the amplitude of the correlation is normalised to that of Cr01.  The introduction of a weighting factor to minimise any contribution of intrinsic alignments to the shear correlation function is considered in ${\\rm Sect.\\,5}$. In ${\\rm Sect.\\,6}$ we present the results of applying such a weighting factor to surveys of different mean redshift. We finish in ${\\rm Sect.\\,7}$ with a discussion of the results and a future perspective of using redshift information in cosmic shear analysis.  For a review of cosmological weak lensing and the relevant aspects of cosmology, see Mellier (1999) and Bartelmann \\& Schneider (2001; hereafter BS01). In addition, the present status and outlook for cosmic shear studies is summarised by van Waerbeke et al. (2002). ",
        "conclusions": "Recent studies of intrinsic ellipticity correlations have shown that the intrinsic signal can be comparable to (or exceed) the lensing signal in the case of shallow cosmic shear surveys (Jing 2002). We investigated to what extent the contribution of galaxy pairs which are likely to have larger intrinsic correlations can be suppressed, using photometric redshift information and a simple redshift dependent weighting factor.  Such a process also results in a reduction in the effective number of pairs and hence an increase in the noise, which scales roughly as $1/{\\sqrt{N_{\\rm p}}}$. The width of the weighting factor, parameterised by $\\sigma_{\\cal Z}$, controls the fraction of pairs remaining. The photometric accuracy determines the precision with which galaxies at similar redshifts can be identified, hence the residual contribution from intrinsic alignments.  The reduction of the effective number of pairs is not the only relevant consideration for the noise in the determination of the two-point correlation function. Depending on the survey geometry, cosmic variance can become the dominant contribution to the covariance of the correlation function for larger angular separations. In particular, for a compact survey geometry, cosmic variance of the shear field dominates over shot noise for angular scales larger than a few arcminutes (e.g., Schneider et al.\\ 2002b). Hence, for those scales the reduction of the effective number of pairs becomes of little relevance.  Intrinsic alignments have also been invoked as a possible source of the so-called B-mode contributions seen in cosmic shear surveys (van Waerbeke et al.\\ 2001, 2002; Hoekstra et al.\\ 2002b). Such B-modes are not expected from lensing effects and thus are usually interpreted as a remaining systematics; whereas galaxy correlations can in principle generate a B-mode contribution (Schneider et al.\\ 2002a), its amplitude is very small. If the B-mode is due to an intrinsic alignment of galaxies, then the use of the redshift filter as discussed here should strongly suppress its relative contribution.  Choosing to apply such a weighting factor will be governed by the details of the survey. Such considerations include that the impact of intrinsic alignments is more dramatic for surveys of lower mean redshift or depth.  Another important factor is the accuracy of photometric redshift estimates, which depends upon the combination and characteristics of filters used for the survey - see e.g. Wolf et al. (2001) for a discussion of photometric redshift estimate performance in the context of filter sets. Further, the size of the error bars associated with the practical determination of the two-point correlation function will also play a part in the decision. For deep surveys, with our assumed photometric errors, in order to decrease the fractional contribution of $\\xi_{+}^{\\rm int}$ by a few percent, a reduction of several tens of percent in the number of effective pairs may result. However, removal of the intrinsic alignment systematic becomes increasingly important as the experimental error bars in surveys become smaller.  In shallower surveys, the contamination from intrinsic alignments is much more pronounced, and in this case down-weighting close galaxy pairs is more effective. Given real data, one might consider a more elaborate weighting scheme that is a function of not only the difference in photometric redshifts, but also giving more weight to higher redshift pairs. Furthermore, the weight factor ${\\cal Z}_{i,j}$ need not be chosen as a function of the estimated photometric redshift only, but can be constructed such that it depends on the full redshift probability distribution as estimated by photometric redshift techniques. For example, a natural choice would be to have ${\\cal Z}_{i,j}$ depending on the probability that the two galaxies $i,j$ lie within a narrow redshift interval $\\Delta z$, as estimated from their individual $z$-distributions. The calculation of the expectation value of the corresponding shear estimator is then slightly more complicated, but can be done for each survey at hand.  Throughout we have assumed that photometric redshift estimates will be available for each of the source galaxies, and that the dispersion in these estimates is independent of magnitude or redshift. In practice, photometric redshift estimates are reliable down to a certain limiting magnitude, depending on the survey characteristics and the relative depths of the filters used. Particularly in a deep survey, this magnitude limit may exclude many of the fainter and smaller galaxies that are likely to be in the high-redshift tail of the redshift distribution, and to make a substantial contribution to the lensing signal. Requiring photometric redshift estimates could in fact lead to an increase in the Poisson noise of such a survey by reducing the usable number of pairs beyond the effect shown in Fig.\\,4, and in addition the reduction in mean redshift could increase the relative contribution from intrinsic alignments. One possibility to counteract such affects might be to include fainter galaxies where no photometric redshift estimates are available, assuming that these are at higher redshifts where intrinsic alignment is less of an issue.  The use of cosmic shear surveys as a tool to place constraints on the 3-D power spectrum, so-called power spectrum tomography, would rely on the availability of photometric redshift estimates for the galaxies involved in the analysis (e.g. Hu 1999; 2002).  Croft \\& Metzler (2000) demonstrated that the intrinsic alignment signal is more severe when correlations within a narrow slice in source galaxy redshift space are considered. The application of a redshift dependent filter, similar to the one we have described here, would greatly suppress the intrinsic alignment systematic.  Instead of slicing galaxies in redshift, and calculating the correlation function for galaxies in the same redshift bin -- which both increases the relative importance of intrinsic alignments and reduces the total number of pairs -- one should correlate galaxies from different redshift bins $i,j$, where the bin width would be chosen to be of the same order as the redshift uncertainty. This would yield a set $\\xi_{i,j}(\\theta)$ of correlation functions for which the intrinsic signal is strongly suppressed for $i\\ne j$. When comparing these correlation functions with predictions from cosmological models, their covariance must be taken into account, which may turn out to be fairly complicated (see Schneider et al.\\ 2002b for the case with no redshift information). This redshift slicing is most straightforwardly done with the correlation function, although for other estimators of the two-point cosmic shear statistics, such as the shear dispersion or the aperture mass, similar pair redshift-dependent estimators are easily constructed.  Most of what has been said above also applies to higher-order cosmic shear statistics. It is well known that the three-point statistics, i.e. the skewness, contains very useful cosmological information (e.g. van Waerbeke et al.\\ 1999). As true for the 2-point statistics, the three-point correlation function is the quantity which is most easily derived from a cosmic shear survey. Bernardeau et al.\\ (2002a) have constructed a statistics based on the measured three-point correlation function (see Schneider \\& Lombardi 2002 for the classification of three-point shear correlators), and Bernardeau et al.\\ (2002b) obtained a significant detection in the VIRMOS-DESCART survey data. It is unclear whether, and by how much, intrinsic ellipticity correlations affect these measurements. Furthermore, up to now no B-mode estimator in the three-point function has been devised which may indicate the presence of intrinsic alignment effects. Redshift slicing can of course also be done for the three-point function which may be the only way to measure these lensing statistics without the potential influence of intrinsic effects; one needs to suppress those triplets of galaxies where the probability of all three being at the same redshift is not negligibly small (i.e., the estimator is unaffected by intrinsic effects if two of the three are at the same redshift).  Of course, the intrinsic alignment signal that we seek to suppress when performing a lensing analysis is also interesting in its own right, since it places constraints on the formation and evolution of galaxies. Suitable catalogues of galaxies to quantify this signal, and its evolution with redshift, will be a natural by-product of large cosmic shear surveys with photometric redshift information. For example, using a filter for the determination of the ellipticity correlation as in (\\ref{est}) with ${\\cal Z}'_{i,j}=1-{\\cal Z}_{i,j}$ will make this estimate dominated by the intrinsic alignments, and could thus be used as a first estimate of it. More sophisticated techniques would include the consideration of the resulting correlation function in dependence of the width of the weighting function, and extracting the lensing and intrinsic signal from this functional form.  If the intrinsic alignment of galaxies indeed occurs, and in particular if the intrinsic effect is as strong as suggested by some models (e.g. Jing 2002), a deep multi-colour wide-field survey with a broad range of filters will be necessary, and rewarding, to remove this systematic from cosmic shear measurements. Most likely, the near-IR imaging will present the bottleneck, limiting the magnitude - and thus the effective number density - of galaxies that can be used for cosmic shear. A wide-field near-IR camera in space, such as the PRIME satellite mission, would be an ideal supplement to the planned extensive ground-based optical cosmic shear surveys.  After we had completed this paper, we became aware of work by Heymans \\& Heavens (2002), also addressing how redshift information can be used to reduce the contamination from intrinsic alignments. They apply their technique to estimate the reduction of the intrinsic signal in several surveys, including the SDSS photometric and spectroscopic samples."
    },
    "0208/astro-ph0208226.txt": {
        "abstract": "{We have obtained high quality FORS1/VLT optical spectra of 85 disk \\hii\\ regions in the nearby spiral galaxies NGC 3351, NGC 3521, NGC 4254, NGC 4303, and NGC 4321. % Our sample of metal-rich \\hii\\ regions with metallicities close to solar and higher reveal the presence of Wolf-Rayet (WR) stars in 27 objects from the blue WR bump ($\\sim$ 4680 \\AA) and 15 additional candidate WR regions. This provides for the first time a large set of metal-rich WR regions. \\\\ Approximately half (14) of the WR regions also show broad \\Civ\\ emission attributed to WR stars of the WC subtype. The simultaneous detection of \\Ciii\\ emission in 8 of them allows us to determine an average late WC subtype compatible with expectations for high metallicities. Combined with literature data, the metallicity trends of WR features and the WC/WN number ratio are discussed. \\\\ The WR regions show quite clear trends between their observed WR features and the \\hb\\ emission line. Detailed synthesis models are presented to understand/interpret these observations. In contrast with earlier studies of low metallicity WR galaxies, both \\wwr\\ and \\iwr\\ are here found to be smaller than ``standard'' predictions from appropriate evolutionary synthesis models at corresponding metallicities. % Various possibilities which could explain this discrepancy are discussed. The most likely solution is found with an improved prescription to predict the line emission from WN stars in synthesis models. \\\\ The availability of a fairly large sample of metal-rich WR regions allows us to improve existing estimates of the upper mass cut-off of the IMF in a robust way and independently of detailed modeling: from the observed maximum \\hb\\ equivalent width of the WR regions we derive a {\\bf lower limit for \\mup\\ of 60--90 \\msun} in the case of a Salpeter slope and larger values for steeper IMF slopes. This constitutes a lower limit on \\mup\\ as all observational effects known to affect potentially the \\hb\\ equivalent width can only reduce the observed \\whb. \\\\ From our direct probe of the massive star content we conclude that there is at present no evidence for systematic variations of the upper mass cut-off of the IMF in metal-rich environments, in contrast to some claims based on indirect nebular diagnostics. % % ",
        "introduction": "\\label{s_intro}  Wolf-Rayet stars (WR) are the descendants of the most massive stars. Although they live during a short time (Maeder \\& Conti 1994) these stars have been detected in young stellar systems, such as extragalactic HII regions (Kunth \\& Schild 1986) and the so-called WR galaxies (Conti 1991, Schaerer \\etal\\ 1999b). They are recognized by the presence of broad stellar emission lines at optical wavelengths, mainly at 4680 \\AA\\ (known as the blue WR bump) and at 5808 \\AA\\ (red WR bump). The blue bump is a blend of N~{\\sc v} $\\lambda\\lambda$4604,4620, N~{\\sc iii} $\\lambda\\lambda$4634,4641, C~{\\sc iii/iv} $\\lambda\\lambda$4650,4658 and \\Heii\\ lines, that are produced in WR stars of the nitrogen (WN) and carbon (WC) sequences. In contrast, the red bump is formed only by \\Civ\\ and it is mainly produced by WC stars. The detection of these features in the integrated spectrum of a stellar system provides a powerful tool to date the onset of the burst, and it constitutes the best direct measure of the upper end of the initial mass function (IMF). Thus, if WR features are found in the spectra of star forming systems, stars more massive than $M_{\\rm WR}$, where $M_{\\rm WR} \\sim$ 25 \\msun\\ for solar metallicity, must be formed in the burst.  The IMF is one of the fundamental ingredients for studies of stellar populations, which has an important bearing on many astrophysical studies ranging from cosmology to the understanding of the local Universe. In particular the value of the IMF slope and the upper mass cut-off (\\mup) strongly influences the mechanical, radiative, and chemical feedback from massive stars to the ISM such as the UV light, the ionizing radiation field, and the production of heavy elements.  A picture of a universal IMF has emerged from numerous works performed in the last few years (e.g.\\ Gilmore \\& Howell 1998 and references therein). Indeed, these studies derive a slope of the IMF close to the Salpeter value for a mass range between 5 and 60 \\msun. This result seems to hold for a variety of objects and metallicities from very metal poor  up to the solar metallicity, with the possible exception of a steeper field IMF (Massey \\etal\\ 1995, Tremonti \\etal\\ 2002). However, the IMF in high metallicity (12+log (O/H) $\\ga$ (O/H)$_\\odot \\approx$ 8.92) systems is much less well constrained. Different indirect methods to derive the slope and \\mup\\ give contradictory results.  The detection of strong wind resonance UV lines in the integrated spectrum of high metallicity nuclear starbursts clearly indicate the formation of massive stars (Leitherer 1998; Schaerer 2000; Gonz\\'alez Delgado 2001). In contrast, the analysis of the nebular optical and infrared lines of IR-luminous galaxies and high metallicity \\hii\\ regions indicates a softness of the ionizing radiation field that has beeninterpreted as due to the lack of stars more massive than $\\sim$ 30 \\msun\\ (Goldader \\etal\\ 1997; Bresolin \\etal\\ 1999; Thornley \\etal\\ 2000; Coziol \\etal\\ 2001). However, the interpretation of these indirect probes relies strongly on a combination of models for stellar atmospheres and interiors, evolutionary synthesis, and photoionisation, each with several potential shortcomings/difficulties (cf.\\ Garc\\'\\i a-Vargas 1996, Schaerer 2000, Stasi\\'nska 2002). For example, recently Gonz\\'alez Delgado \\etal\\ (2002) have shown that the above conclusion could be an artifact of the failure of WR stellar atmospheres models to correctly predict the ionizing radiation field of high metallicity starbursts (see also Castellanos 2001, Castellanos \\etal\\ 2002b).  A more direct investigation of the stellar content of metal-rich nuclear starbursts has been performed by Schaerer \\etal\\ (2000, hereafter SGIT00), using the detection of WR features to constrain \\mup. They found that the observational data are compatible with a Salpeter IMF extending to masses \\mup\\ $\\ga$ 40 \\msun. Most recently, a similar conclusion has been obtained by Bresolin \\& Kennicutt (2002, hereafter BK02) from observations of high-metallicity HII regions in M83, NGC 3351 and NGC 6384.  Here, we present a direct attempt to determine \\mup\\ based on the detection of WR features in metal-rich \\hii\\ regions of a sample of spiral galaxies. To obtain statistically significant conclusions about \\mup\\ and the slope of the IMF, a large sample of \\hii\\ regions needs to be observed. For coeval star formation with a Salpeter IMF and \\mup=120 \\msun\\ at metallicities above solar, $\\sim$ 60 to 80 \\% (depending on the evolutionary scenario and age of the region) of the \\hii\\ regions are expected to exhibit WR signatures (Meynet 1995; Schaerer \\& Vacca 1998, hereafter SV98). Thus, to find $\\ga$ 40 regions with WR stars (our initial aim) a sample of at least 5-7 galaxies with $\\ga$ 10 \\hii\\ regions per galaxy needs to be observed. Spectra of high S/N (at least 30) in the continuum are also required to obtain an accurate measure of the WR features. For this propose, we have selected the nearby spiral galaxies NGC 3351, NGC 3521, NGC 4254, NGC 4303 and NGC4321, which have have sufficient number of disk \\hii\\ regions of high-metallicity, as known from earlier studies.  Our observations have indeed allowed to find a large number of metal-rich WR \\hii\\ regions. The analysis of their massive star content is the main aim of the present paper. Quite independently of the detailed modeling undertaken below, our sample combined with additional WR regions from Bresolin \\& Kennicutt (2002) allow us to derive a fairly robust {\\em lower limit} on the upper mass cut-off of the IMF in these metal-rich environments (see Sect.\\ \\ref{s_imf}).  The structure of the paper is as follows: The sample selection, observations and data reduction are described in Sect.\\ \\ref{s_obs}. The properties of the \\hii\\ regions are derived in Sect.\\ \\ref{s_props}. Section \\ref{s_wroh} discusses the trends of the WR populations with metallicity. Detailed comparisons of the observed WR features with the evolutionary synthesis models are presented in Sect.\\ \\ref{s_models}. More model independent constraints on \\mup\\ are derived in Sect.\\ \\ref{s_imf}. Our main results and conclusions are summarised in Sect.\\ \\ref{s_conclude}.  %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% ",
        "conclusions": "\\label{s_conclude}  We have obtained high quality FORS1/VLT optical spectra of 85 disk \\hii\\ regions in the nearby spiral galaxies NGC 3351, NGC 3521, NGC 4254, NGC 4303, and NGC 4321. This sample, consisting in particular of a good fraction of objects with oxygen abundances presumably above solar (as estimated from $R_{23}$ using the calibration reported by Kobulnicky \\etal\\ 1999), provides an unprecedented opportunity to study stellar populations, nebular properties and ISM abundances in \\hii\\ regions at the high metallicity end. In this first paper we have presented the observational findings on spectral signatures from massive stars, and compared these with evolutionary synthesis models with the main aim of constraining the upper part of the IMF.  The average metallicity of our \\hii\\ region sample is \\oh\\ $\\sim 8.9 \\pm 0.2$ using the calibration of Kobulnicky \\etal\\ (1999). For 12 regions we are able to determine the electron temperature from the transauroral \\oii\\ $\\lambda$7325 line, yielding lower limits on O/H (Sect.\\ \\ref{s_props}). For 6 regions we have been able to confirm a high metallicity (\\oh\\ $\\ga$ 8.8--8.9). Detailed photoionisation modeling will be undertaken in the future to improve our abundance determinations and to include the full sample of \\hii\\ regions.  The spectra of a large number (27) of regions show clear signatures of the presence of Wolf-Rayet (WR) stars as indicated by broad emission in the blue WR bump ($\\sim$ 4680 \\AA). Including previous studies (Castellanos 2001, Bresolin \\& Kennicutt 2002, Castellanos \\etal\\ 2002b) our observations now nearly quadrupel the number of metal-rich \\hii\\ regions where WR stars are known. % Approximately half (14) of the WR regions also show broad \\Civ\\ emission attributed to WR stars of the WC subtype. The simultaneous detection of \\Ciii\\ emission in 8 of them allows us to determine an average late WC subtype ($\\sim$ WC7-WC8) compatible with expectations for high metallicities (Sect.\\ \\ref{s_props}).  Combined with existing observations of WR regions and WR galaxies at sub-solar our data confirm the continuation of previously known trends of increasing WRbump/\\hb\\ intensity with metallicity, establish also such a trend for \\wwr, and allow us to estimate the trend of the WC/WN ratio with \\oh\\ in extra-galactic \\hii\\ regions (Sect.\\ \\ref{s_props})  The observed strength of the blue WR bump (relative line intensities and equivalent widths) shows quite clear trends with \\whb. Both \\wwr\\ and \\iwr\\ are found to be smaller than ``standard'' predictions from state-of-the-art evolutionary synthesis models (Schaerer \\& Vacca 1998) at corresponding metallicities. Various possibilities (including deviations of the IMF from a Salpeter slope and a ``normal'' high upper mass cut-off) which could explain this discrepancy have been discussed. The most likely solution is found with an improved prescription to predict the line emission from WN stars in synthesis models (Sect.\\ \\ref{s_models}). % Using this new prescription the observed WR features are found to be broadly consistent with short bursts and a ``standard'' Salpeter IMF extending to high masses, as indicated by earlier studies at sub-solar metallicities.  Independently of the difficulties encountered to model the WR features in detail, the availability of a fairly large sample of metal-rich WR regions allows us to improve existing estimates (Schaerer \\etal\\ 2000, Bresolin \\& Kennicutt 2002) of the upper mass cut-off of the IMF. Independently of the exact tracks and metallicity we derive a {\\bf lower limit for \\mup\\ of 60--90 \\msun} in the case of a Salpeter slope, and larger values for steeper IMF slopes, from the observed maximum \\hb\\ equivalent width of the WR regions. This constitutes a lower limit on \\mup\\ as all observational effects known to affect potentially the \\hb\\ equivalent width (loss of photons in slit or leakage, dust inside \\hii\\ regions, differential extinction, underlying population) can only reduce the observed \\whb. % From our probe of the massive star content we therefore conclude that there is at present no direct evidence for systematic variations of the upper mass cut-off of the IMF in metal-rich environments, in contrast to some claims based on indirect nebular diagnostics (e.g.\\ Goldader \\etal\\ 1997, Bresolin \\etal\\ 1999, Coziol \\etal\\ 2001). % What the origin of this ``universality'' of the IMF is, remains an open question.   %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%"
    },
    "0208/astro-ph0208242_arXiv.txt": {
        "abstract": "We present narrow-band H$\\alpha$, [\\ion{S}{2}], and [\\ion{O}{1}] Hubble Space Telescope images of the young planetary nebula GL 618. This object is a compact, bipolar nebula that is currently undergoing the transition from asymptotic giant branch star to planetary nebula. Our images confirm the presence of at least three highly collimated outflows emanating from the central regions of GL 618. We also detect H$\\alpha$ emission close to the central dust lane and in an extended scattered light halo. The three outflows are occurring simultaneously in this object, as opposed to being the result of a precessing jet. We derive an inclination for the brightest outflow in the East lobe of 39\\degr $\\pm$ 4\\degr. This differs from the previous estimate of 45\\degr. In addition, our results indicate that the outflows seen in GL 618 are probably not coplanar. Line strengths derived from the narrow-band images indicate a shock velocity in the range of 40 $-$ 100 km s$^{-1}$. Based on the shock velocity we estimate that the age of the outflows is less than 500 years. The outflows seen in the optical images of GL 618 are related to features seen in near-IR, CO and CS maps of this object. This relationship indicates that the outflows are playing a major role in the morphological evolution of this young planetary nebula, interacting with and shaping the neutral envelope surrounding GL 618. We discuss the implications of these jets and their interaction with the neutral envelope in the context of current models of planetary nebula formation. ",
        "introduction": "\\citet{kwo78} introduced the interacting winds model to account for the development of spherical planetary nebulae (PN). In this model a massive, low-velocity wind removes a large fraction of the envelope material during the asymptotic giant branch (AGB) phase of evolution, and is followed by a higher-velocity, lower-mass wind. The high-velocity wind snowplows into the previously ejected AGB wind material and sweeps up gas to form a PN. The majority of PN exhibit aspherical structure, ranging from slightly elliptical to bipolar \\citep{gre71,zuc86,bal87,sch92,man96}. Other investigators have extended the interacting winds model to explain the formation of these aspherical objects \\citep{kha85,bal87,sok89,ick92,fra94,dwa96}.  Understanding the origin of asymmetric structure in PN has been the focus of extensive research in recent years. To achieve asymmetric PN morphologies, the interacting winds model requires that mass loss during the early phases of the development of these objects is aspherical. This aspherical structure is then amplified through the wind-shaping that occurs as the PN evolves. Observationally, the time that aspherical structure becomes important is becoming clear. Mass loss during the AGB is predominantly spherical \\citep{ner98,mei98}, while the morphologies of proto-planetary nebulae are typically aspherical \\citep{tra94,mei97,day98,kwo98,uet00}. These findings indicate that during the late-AGB and/or post-AGB phase, the mass loss becomes aspherical \\citep[for reviews]{kas00}. The mechanism that triggers this change in the character of the mass loss is not well understood. Suggestions for this trigger range from binary star evolution \\citep{mor81,sok94} to the effects of magnetic fields \\citep{gar97}.  A recent HST narrow-band H$\\alpha$ survey of the morphologies of low excitation PN \\citep{sah98a} revealed a dazzling array of complex structures in these objects including jets, knots, bubbles, and loops. The large variety of structures seen in these images cannot be explained by the interacting winds model alone. Sahai \\& Trauger suggested that phenomena such as jets acting during the early post-AGB phase may play a role in creating the complex structures seen in these objects. Recent CO observations of several young PN with collimated outflows have demonstrated that these outflows are significantly disrupting and interacting with the neutral shells surrounding these objects \\citep{for98,cox00,hug00,bac00}. Both the HST imaging and the CO observations suggest that jets could have a significant impact on the way we think about the evolution of PN. The interaction of the jet with previously ejected material may influence the morphological evolution of PN in a way that is not considered in current models of PN formation. Further, the phenomena driving the change from spherical to aspherical mass loss in the early stages of PN development are not clear. The formation of jets may play a role in the change to aspherical mass loss in some objects.   If jets are playing a major role in the shaping of PN, their influence will be most significant at a time when the PN still has a large neutral envelope and before the fast wind develops. We present HST observations of GL 618, a young PN undergoing the stage of evolution immediately preceding the PN phase, during which recently lost material has begun to disperse but the fast wind has not fully developed. GL 618 is surrounded by a shell of dust and neutral gas that extends well beyond the optical nebula \\citep[and references therein]{spe00,mei98}. Our new observations have allowed us to confirm that jets are present in GL 618 and to better understand the importance of these outflows in development of this object.  Ground-based optical and near-IR imaging of GL 618 reveal two lobes of emission (each about 3$\\arcsec$ in extent) separated by a dark lane. The central regions of the nebula are hidden from direct view at optical and near-IR wavelengths by the lane of obscuring material. The spectrum of GL 618 is composed of a faint continuum and a variety of low-excitation emission lines. \\citet{sch81} and \\citet{tra93} used spectropolarimetry to study GL 618 and found that the continuum and part of the permitted line emission are reflected from deep in the nebula. The low-excitation, forbidden line flux and remainder of the permitted line emission are produced in the bipolar lobes. Trammell et al. demonstrated that shock-excited emission dominates the spectra of the lobes of GL 618 with the emission indicative of shock velocities in the range of 50$-$100 km s$^{-1}$. Long-slit optical spectroscopy of GL 618 suggests that the shock emission is associated with out-flowing gas \\citep{car82}. \\citet{san02} present a recent detailed ground-based spectroscopic study of GL 618. Their analysis of the emission line profiles and ratios suggest shock velocities in the range 75$-$200 km s$^{-1}$. The largest velocities correspond to the tips of the bullets that we see in our HST images. Near-IR spectroscopy of GL 618 has revealed the presence of thermally excited H$_2$ emission \\citep{thr81,lat92}. This object also exhibits [\\ion{Fe}{2}] emission thought to be associated with the shock-heated gas \\citep{kel92} and more recent observations establish that at least part of this emission is associated with an outflow \\citep{kas01}. \\citet{uet01} presented high spatial resolution near-IR images of GL 618 taken with the Subaru Telescope. The structure observed in the near-IR extends beyond the optical nebula and these images reveal the presence of \"horns\" and bullet-like structures associated with the bipolar lobes of GL 618. Ueta et al. suggest that the [\\ion{Fe}{2}] emission is associated with these features. These features are coincident with the tips of the outflows that we see in our HST images. These authors also found high velocity CO clumps at the positions of the IR bullets when they re-examined existing $^{12}$CO J=1-0 interferometry data. Our new HST observations demonstrate that the source of the shock-excited emission in GL 618 is a set of highly collimated outflows originating in the central regions of the object. ",
        "conclusions": "Our observations demonstrate that emission from highly collimated outflows dominates the optical emission and morphology of GL 618. A comparison of the optical images with previous molecular line studies indicates that the jets are interacting with the neutral envelope surrounding GL 618. This interaction is shaping the neutral envelope of GL 618 and will determine the morphological evolution of this object. GL 618 is undergoing the early stages of PN development. Only a small region of photoionized gas is present in GL 618 and this ionized gas is found only in the immediate vicinity of the central star \\citep{kwo81}. The fast wind described in the interacting winds model \\citep{kwo78} has not fully developed in GL 618. The interacting winds model suggests that it is the interaction of this fast wind with the surrounding neutral envelope that dictates the morphological evolution of a PN. In GL 618 the fast wind will be interacting with a neutral envelope that has already been significantly altered by the action of the jets.  The interaction of the jets with the neutral envelope seen in GL 618 is not unique. Recent observations of several other young PN indicate that jets may have a much more significant role in the overall development of these objects. CO observations of PN known to contain collimated outflows have shown that these outflows are having a major impact on the surrounding neutral shells \\citep[e.g.]{for98,cox00,hug00,bac00}. For example, in BD+30\\degr3639 the CO observations reveal a pair of high velocity molecular knots just outside the ionized nebula. This is very similar to relationship between the molecular material and optical jets seen in GL 618. In all of these objects the jet's interaction with the neutral shell, not interacting winds, could be the dominant mechanism in determining the overall morphology of the PN.  The complex, multi-polar geometry observed in GL 618 is not unique. \\citet{sah00} presented HST images of two young PN, He 2-47 and M1-37 (dubbed the Starfish Twins). These objects exhibit multi-polar morphology that is similar to that seen in GL 618. Sahai suggests that jets established the complex geometry of these nebulae. He estimates that the jets were active within the past few hundred years and suggests that multiple outflows occurred simultaneously in these objects. While the overall morphology of these objects is similar to GL 618, there are some differences. Knots or bullets of emission are prominent at the tips of the outflows in GL 618. These types of structures are not evident in the images of He 2-47 and M1-37. These bright spots mark the point where active jets collide with ambient material in GL 618. The jets are no longer active in the Starfish Twins so that the regions that correspond to tips have cooled significantly and do not appear as bright spots.  CO observations and HST imaging strongly suggest that jets play a important or dominant role in the shaping of some PN. These jets appear to be short-lived and they seem to operate during the post-AGB and/or PPN phases of evolution. Early jet action sets the stage for the complex morphologies we observed in some PN. High spatial resolution spectroscopy of these objects is needed to better understand the dynamics of the outflows and their impact of the surrounding material.  The overall morphology of the neutral envelope of GL 618, as traced by CO, is spherical \\citep{mei98}. The more recent mass outflows in this object are clearly aspherical. As is the case in other PN, the mass loss in GL 618 appears to have switched from spherical to aspherical soon after the star left the AGB. The kinematic age of the outflows suggest that the jets turned on sometime shortly after the star left the AGB. The trigger for the switch from spherical to aspherical mass loss in GL 618 appears to be the formation of collimated outflows in this object.  The debate concerning the origin of collimated outflows and also the formation of aspherical PN in general, centers on whether binary or single stars are responsible for producing aspherical mass loss. Both models of binary star interaction \\citep{mor81,sok94} and magnetic confinement \\citep{gar97,mat00}, while providing a scheme for producing the overall aspherical structure in PN, may also provide mechanisms to produce the highly collimated outflows. The complex, multi-polar outflow geometry seen in GL 618 may be difficult for either of these types of models to explain."
    },
    "0208/astro-ph0208074_arXiv.txt": {
        "abstract": "We discuss the relevant processes for the relativistic electrons in the ICM and the possible mechanisms responsible for the production of these electrons. We focus on the origin of the radio halos giving some of the observational diagnostics which may help in discriminating among the different models proposed so far. Finally, we briefly discuss the discrepancy between the value of the magnetic field assuming an inverse Compton (IC) origin of the hard X--ray emission (HXR) and that obtained from Faraday Rotation Measurements (RM). ",
        "introduction": "The most important evidence for relativistic electrons in clusters of galaxies comes from the diffuse synchrotron radio emission observed in about 35 \\% of the clusters selected with X--ray luminosity $> 10^{45}$erg s$^{-1}$ (e.g., Giovannini, Tordi, Feretti, 1999; Giovannini \\& Feretti 2001). The diffuse emissions are referred to as radio halos and/or radio mini--halos when they appear projected on the center of the cluster, while they are called relics when they are found in the cluster periphery.  The difficulty in explaining the extended radio halos arises from the combination of their $\\sim$Mpc size, and the short radiative lifetime of the radio emitting electrons. Indeed, the diffusion time necessary to the radio electrons to cover such distances is orders of magnitude greater than their radiative lifetime. To solve this problem Jaffe (1977) proposed continuous in situ reacceleration of the relativistic electrons. The in situ reacceleration scenario was quantitatively reconsidered by Schlickeiser et al.(1987) who successfully reproduce the integrated radio spectrum of the Coma halo. In the framework of the in situ reacceleration model, Harris et al.(1980) first suggested that cluster mergers may provide the energetics necessary to reaccelerate the relativistic particles. The role of mergers in particle acceleration and in the amplification of the magnetic fields was than investigated in more detail by De Young (1992) and Tribble (1993). Alternatively, to avoid the energy loss problem of the radio electrons, Dennison (1980) first suggested that the radio emission in radio halos may be emitted by a population of secondary electrons continuously injected by hadronic interactions. {\\it Primary} and the {\\it secondary} electron models still constitute the basis of the more recent theoretical works developed on the argument (see Sect.3). Additional evidence for the presence of non--thermal phenomena in clusters of galaxies comes from the detection in a number of cases of EUV excess emission (e.g., Bowyer et al. 1996; Lieu et al., 1996; Bergh\\\"ofer et al., 2000; Bonamente et al., 2001) and of HXR excess emission in the case of the Coma cluster and A2256 (Fusco--Femiano et al., 1999, 2000; Rephaeli et al., 1999; Rephaeli \\& Gruber, 2002; Fusco--Femiano, this meeting). While, with the exception of the Coma and of the Virgo clusters, the EUV detections are still controversial (e.g., Bergh\\\"ofer et al., 2000; Bergh\\\"ofer, Bowyer, Nevalainen, this meeting), the HXR detections are quite robust as they are independently obtained by different groups and with different X--ray observatories (BeppoSAX, RXTE). On the other hand, while there is agreement on the IC origin at least of the most significant cases of EUV excess emission (e.g., Hwang 1997; Bowyer \\& Berg\\\"ofer 1998; Ensslin \\& Bierman 1998; Sarazin \\& Lieu 1998; Atoyan \\& V\\\"olk 2000; Brunetti et al. 2001b, Petrosian 2001, Tsay et al., 2002), the origin of the XHR is still debated. The XHR excess may be generated by IC scattering of relativistic electrons off the CMB photons (Fusco--Femiano et al., 1999,2000; Rephaeli et al., 1999; V\\\"olk \\& Atoyan 1999; Brunetti et al.2001a; Petrosian 2001; Fujita \\& Sarazin 2001). Alternatively XHR might also result from bremsstrahlung emission from a population of supra--thermal electrons (e.g., Ensslin, Lieu, Biermann 1999; Blasi 2000; Dogiel 2000; Sarazin \\& Kempner 2000). Both the IC model and the bremsstrahlung interpretation have problems : the first one would require cluster magnetic field strengths smaller than that inferred from RM observations (e.g., Clarke, Kronberg, B\\\"oringer 2001), the second one would require a too large amount of energy to mantain a substantial fraction of the thermal electrons far from the thermal equilibrium for more than $10^8$yrs (e.g., Petrosian, 2001). In this contribution we will describe the populations of relativistic electrons expected in clusters of galaxies and we will focus on the origin of radio halos and HXR emission from galaxy clusters. We refer to the contribution by Ensslin for the origin of radio relics and to the contributions by Bowyer for a review on the EUV excesses. ",
        "conclusions": "Highly relativistic electrons (i.e., $\\gamma > 10^3$) can be injected in clusters of galaxies by several processes : they can be accelerated by merger shocks (Pop A), they can be relic relativistic electrons (i.e., $\\gamma \\sim 10-100$) reaccelerated by cluster turbulence (Pop B), and they can be secondary electrons injected by hadronic collisions (Pop C). We examine three diagnostic which can help us to better understand the origin of the relativistic electrons producing the observed radio synchrotron emission :  \\begin{itemize} \\item[{\\it i)}] Electrons accelerated by strong merger shocks (Pop. A) cannot produce synchrotron emission diffuse on $\\geq$Mpc scale as that of classical radio halos. This is due to the short radiative lifetime of the electrons after being accelerated in the shock region. A possibility to accomodate the Mpc sizes within the Pop. A is given by very fast (${\\cal M} > 5$) - but unlikely - shocks crossing the cluster center or by the presence of cluster turbulence in addition to the merger shocks. \\item[{\\it ii)}] The comparison between the radio and the soft X--ray brightness of a number of radio halos indicates that the profile of the radio emission is broader than that of the X--ray thermal emission. This appears to be difficult to be accomodated within secondary models (Pop. C) which would yield narrower radio profiles. A possibility to skip this problem is to admit an {\\it ad hoc} increasing fraction of energy density of the relativistic protons with radius. However, at least in some cases, this would imply an energetics of the relativistic protons higher than that of the thermal pool. \\item[{\\it iii)}] The spectral cut--off and radial spectral steepenings observed in the case of Coma (and in the mini--halo in the Perseus cluster) strongly point to the presence of a cut--off in the spectrum of the emitting electrons. This cut--off may be naturally accounted for if the synchrotron emission is produced by reaccelerated (Pop.B) electrons, whereas it is not expected in the case of secondary electrons (Pop.C). Future studies will clarify how much synchrotron spectral cut--offs and radial steepenings are common in radio halos. \\end{itemize}  \\noindent Points i)--iii) would suggest that radio halos are powered by the synchrotron emission from electrons reaccelerated (Pop.B) in the cluster volume during merger events. This conclusion is, however, based on detailed studies of only few radio halos. As a consequence, detailed observations are still required to better understand the origin of radio halos.   The origin of the HXR emission detected in few clusters of galaxies is still matter of debate. It could be IC emission from relativistic electrons belonging to the same population of electrons responsible for the large scale radio emission. Alternatively HXR emission might result from bremsstrahlung emission from a supra--thermal tail of electrons. Both these hypothesis have problems: the IC emission would require a magnetic field value in apparent disagreement with that inferred from RM observations, while the supra--thermal bremsstrahlung requires a too large amount of energy if emitted for $> 10^8$yrs. We have shown that the discrepancy between the field value obtained from the IC assumption and that from the RM observations can be significantly reduced. It is now clear that the combination of spatial trends and inhomogenities in the thermal gas and magnetic field distribution with the presence of a cut--off in the electron spectrum at the energies of the radio emitting electrons would allow the IC magnetic field to be in better agreement with that from the RM observations. Although these assumptions are {\\it a posteriori} and might seem a sort of {\\it conjuring tricks}, it should be noticed that the presence of a high energy cut--off at $\\gamma_{\\rm c} \\sim 10^{4}$ would really results from models of radio halos invoking the reacceleration of relic relativistic electrons. In addition a decreasing radial profile of the magnetic field strength is naturally expected. In the framework of Pop.B models, we have shown that assuming the same model parameters necessary to reproduce the general radio properties of the Coma halo (brightness profile, integrated radio spectrum, and radial spectral steepening), the resulting IC HXR would easily be about $\\sim 30-100$\\% of that observed. In this case, the resulting magnetic field strength in the cluster core is of the order of $\\sim 1 \\mu G$ which is in rough agreement with the RM observations especially when the field strength from the RM data is calculated assuming a power spectrum of the field itself."
    },
    "0208/astro-ph0208304_arXiv.txt": {
        "abstract": "We show that a common evolutionary history can produce the black hole binaries in the Galaxy in which the black holes have masses of $\\sim 5-10\\msun$. In with low-mass, $\\lsim 2.5\\msun$, ZAMS (zero age main sequence) companions, the latter remain in main sequence during the active stage of soft X-ray transients (SXTs), most of them being of K or M classification. In two intermediate cases, IL Lupi and Nova Scorpii with ZAMS $\\sim 2.5\\msun$ companions the orbits are greatly widened because of large mass loss in the explosion forming the black hole, and whereas these companions are in late main sequence evolution, they are close to evolving. Binaries with companion ZAMS masses $\\gsim 3\\msun$ are initially ``silent\" until the companion begins evolving across the Herzsprung gap.  We provide evidence that the narrower, shorter period binaries, with companions now in main sequence, are fossil remnants of gamma ray bursters (GRBs). We also show that the GRB is generally accompanied by a hypernova explosion (a very energetic supernova explosion). We further show that the binaries with evolved companions are good models for some of the ultraluminous X-ray sources (ULXs) recently seen by Chandra in other galaxies.  The great regularity in our evolutionary history, especially the fact that most of the companions of ZAMS mass $\\lsim 2.5\\msun$ remain in main sequences as K or M stars can be explained by the mass loss in common envelope evolution to be Case C; i.g., to occur only after core He burning has finished. Since our argument for Case C mass transfer is not generally understood in the community, we add an appendix, showing that with certain assumptions which we outline we can reproduce the regularities in the evolution of black hole binaries by Case C mass transfer. ",
        "introduction": "\\label{intro}  The discovery of afterglows to gamma-ray bursts (GRBs) has greatly increased the possibility of studying their physics. Recent observations strongly suggest a connection between GRBs and supernovae, with indication that the supernovae in question are especially energetic and of type Ib/c, i.e., core collapses of massive stars which have lost their hydrogen envelope (see Van Paradijs et al.\\cite{paradijs00}, and references therein). This supports suggestions by Woosley\\cite{Woosley93B} and Paczy\\'nski\\cite{paczynski98} for the origin of GRBs in stellar core collapses. The hydrodynamics of a jet escaping from a star and causing its explosion was explored in detail by MacFadyen \\& Woosley\\cite{MacFadyen99}, who showed that contrary to accepted wisdom, a fairly baryon-free, ultra-relativistic jet could plow through the collapsing star and emerge with large Lorentz factors. The powering of the outflow by coupling of high magnetic fields to the rotation of the black hole~\\cite{bz77}, first suggested by Paczy\\'nski\\cite{paczynski98} in the context of GRBs, was worked out in detail by Van Putten\\cite{putten99,putten01}. Li has also discussed the deposition of energy from a black hole into the accretion disk in a recent series of papers\\cite{Li2000}.   Building on these thoughts, we have modeled both the powering of a GRB by black-hole rotation and the stellar evolution pathways that set up favorable conditions for that mechanism~\\cite{grb2000}. An essential ingredient in this model is a rapidly rotating black hole, and it is this aspect that we focus on in the present paper.    A massive star in a close binary will spin faster than a massive single star for a number of reasons: first, when the hydrogen envelope is lifted off by spiral-in, it will cease to serve as a sink of angular momentum for the core. Second, the tidal friction concomitant with the spiral-in process will spin up the inner region, giving it a larger angular momentum than the same region in a single star~\\cite{Rasio96}. Third, tidal coupling in the close binary will tend to bring the primary into corotation with the orbital period. This latter process is not very efficient in the short post spiral-in life of the binaries we consider, but its effect does probably matter to the outer layers of the helium star, which can be important for our work. With its more rapid rotation, the helium star then forms a black hole with a large Kerr parameter, which immediately after its formation (in a few seconds) begins to input power into its surroundings at a very high rate. This, then, powers both a GRB~\\cite{grb2000} and the expulsion of the material that was centrifugally prevented from falling into the black hole. In fact, Van Putten\\cite{putten99,putten01} estimates that the power input into that material exceeds that into the GRB and Li\\cite{Li2000b} also finds that more energy can be extracted by the disk than by the GRB. It should be noted that an initially less rapidly rotating black hole could be spun up by disk accretion quite rapidly, and start a similar process after some accretion has taken place~\\cite{MacFadyen99,grb2000}. Some implications of such more complicated sequences of events are discussed by Lee et al.\\cite{Lee2001}.  More than dozen soft X-ray transient (SXT) black hole binaries were observed in our Galaxy. By far the most famous one is Nova Scorpii 94. Israelian et al.\\cite{Israelian99} found the overabundances of heavy elements in the subgiant companion star in Nova Scorpii, and suggested that it is a relic of hypernova explosion. Observed high system velocity and the quasi periodic oscillations support the hypernova explosion associated with rapidly rotating black hole. In a more recent analysis~\\cite{Lee2002}, we found the correlation between the black hole masses and the reconstructed pre-explosion orbital periods of SXTs. In binaries with preexplosion orbital period less than 12 hours, rapidly rotating black holes are formed in the core of helium stars by the prompt collapse, and the outer part of the helium star is held from immediate collapse by the augular momentum support. Based on these, we suggest that the SXTs with preexplosion orbital period less than or similar to 12 hours are the relics of GRBs and the hypernovae.   Since the afterglows have thus far only been seen for long GRBs (duration $\\gsim 2$\\,s), we shall concentrate on the mechanism for this subclass. The shorter bursts (duration $\\lsim 2$\\,s) may have a different origin; specifically, it has been suggested that they are the result of compact-object mergers and therefore offer the intriguing possibility of associated outbursts of gravity waves. (Traditionally, binary neutron stars have been considered in this category~\\cite{Eichler89,Janka99}. More recently, Bethe \\& Brown~\\cite{Bethe98} have shown that low-mass black-hole, neutron-star binaries, which have a ten times greater formation rate and are stronger gravity-wave emitters, may be the more promising source of this kind.)   The plan of this paper is as follows. In Sec.~\\ref{grbhypernova}, the observational indications of the GRB and supernova/hypernova association are discussed.  In Sec.~\\ref{energy}, we give the simple argument for the energetics of GRBs and Hypernovae. Our simple argument is based on the Blandford-Znajek mechanism, and we discuss various channels of the power output from the black hole-accretion disk system.  Woosley's  Collapsar model is discussed as a progenitor of GRBs and Hypernovae.  In Sec.~\\ref{sxts}, we summarize the observational indications of SXTs, especially Nova Scorpii which is by far the most interesting SXT. We also discuss our recent discoveries on the empirical correlation between the orbital periods and black hole masses of X-ray transient black hole systems.  The evolution of the SXTs is discussed in Sec.~\\ref{evol}. Based on the Case C mass transfer, which is essential part of the evolution, we could pin down the common envelope efficiency.  Using the evolution scenario in the previous section, in Sec.~\\ref{reconst}, we reconstructed the pre-explosion orbital period and the black hole masses at the time of formation. We found a regularity in the reconstructed mass-period relations.  In Sec.~\\ref{rotate}, we discuss a schematic model which explains the correlation between the black hole masses and the preexplosion periods. In SXTs with short pre-explosion orbital periods, less than or similar to 12 hours, rapidily rotating black holes are formed. Based on these results, we suggest that the progenitors of black holes in SXTs are the sources of GRBs and Hypernovae if their preexplosion spin periods are less than or similar to 12 hours. In this model we assumed that the tidal interaction synchronized the orbital period and spin period of the black hole progenitor.  In Sec.~\\ref{other}, we discuss other observational issues of SXTs. We give estimates on the chemical composition of a few X-ray transients. We discuss the untraluminous X-ray sources in our Galaxy. We also discuss the population synthesis of SXTs and GRBs.  Our final discussion and conclusion follows in Sec.~\\ref{conclu}.  In the Appendix, we discuss the Case C mass transfer in detail, which is essential part of the evolution of the black hole progenitors. We also show when there will be deviation from Case C, indicating that these will not affect our main results. ",
        "conclusions": "\\label{conclu}  Our work here has been based on the Blandford-Znajek mechanism of extracting rotational energies of black holes spun up by accreting matter from a helium star. We present it using the simple circuitry of {\\it ``The Membrane Paradigm\"}\\cite{Thorne86}. Energy delivered into the loading region up the rotational axis of the black hole is used to power a GRB. The energy delivered into the accretion disk powers a SN Ib explosion.  We also discussed black-hole transient sources, high-mass black holes with low-mass companions, as possible relics for both GRBs and Type Ib supernova explosions, since there are indications that they underwent mass loss in a supernova explosion. In Nova Sco 1994 there is evidence from the atmosphere of the companion star that a very powerful supernova explosion (`hypernova') occurred.  We have shown that there is an observed correlation between orbital period and black-hole mass in SXTs. We have modeled this correlation as resulting from the spin of the helium star progenitor of the black hole: if the pre-explosion orbit has a short period, the helium star spins rapidly. This means that some part of its outer envelope is centrifugally prevented from falling into the black hole that forms at the core. This material is then expelled swiftly, leading to a black hole mass less than the helium star mass.  As the orbital period is lengthened, the centrifugal support wanes, leading to a more massive black hole.  The reason for swift expulsion of material held up by a centrifugal barrier is the fact that black holes formed in our scenario naturally have high Kerr parameters (Fig.~\\ref{FIG12}). This implies that they input very high energy fluxes into their surrounding medium via the Blandford-Znajek mechanism, and thus power both a GRB and the expulsion of the material that does not immediately fall in.  However, because the correlation is induced between the orbital period before explosion and the black-hole mass, its manifestation in the observed correlation between BH mass and present orbital period is weakened due to post-explosion evolution of the binaries. We therefore considered the evolution in some detail, and for a subset of the systems were able to reconstruct the pre-explosion orbital periods. The correlation between pre-explosion period and black hole mass (Fig.~\\ref{FIG11}) is in much better agreement with our model than the original one between present period and black hole mass (Fig.~\\ref{FIG1}). We developed a quantitative model for the relation between period and mass, and showed that it fits the subset of reconstructible SXT orbits.   Nova Scorpii stands out as the most extreme case of mass loss, nearly half of the total system mass, and, therefore, a great widening in the orbit which gets its period well beyond the gap between shrinking and expanding orbits. {}From Fig.~\\ref{FIG11} we see that its black hole mass is far below the polytropic line for its $M_{\\rm He}=11\\msun$ progenitor. We believe that in the case of this binary a short central engine time of several seconds was able to furnish angular momentum and energy to the disk quickly enough to stop the infall of some of the interior matter not initially supported by centrifugal force; i.e., the angular momentum was provided in less than a dynamical time. In other words, the Blandford-Znajek mechanism that drives the GRB not only expelled the matter initially supported for a viscous time by angular momentum, but actually stopped the infall within a dynamical time.  Since we can also compute the Kerr parameters of the black holes formed via our model, we find that the short-period systems should have formed black holes with Kerr parameters in the range 0.7--0.9. This makes them prime candidates for energetic hypernovae and GRBs, and thus provides further support for our earlier study in which we posited that SXTs with black-hole primaries are the descendants of GRBs. We can now also refine this statement: SXTs {\\it with short orbital periods\\/} before the formation of the black hole have given rise to a GRB in the past.   We estimate the progenitors of transient sources to be formed at a rate of 200 GEM (Galactic Events per Megayear).  Since this is much greater than the observed rate of GRBs, there must be strong beaming and possible selection of high magnetic fields in order to explain the discrepancy.  We believe that there are strong reasons that a GRB must be associated with a black hole, at least those of duration several seconds or more discussed here. Firstly, neutrinos can deliver energy from a stellar collapse for at most a few seconds, and sufficient power for at most a second or two. Our quantitative estimates show that the rotating black hole can easily supply the energy as it is braked, provided the ambient magnetic field is sufficiently strong. The black hole also solves the baryon pollution problem: we need the ejecta that give rise to the GRB to be accelerated to a Lorentz factor of 100 or more, whereas the natural scale for any particle near a black hole is less than its mass. Consequently, we have a distillation problem of taking all the energy released and putting it into a small fraction of the total mass. The use of a Poynting flux from a black hole in a magnetic field\\cite{bz77} does not require the presence of much mass, and uses the rotation energy of the black hole, so it provides naturally clean power.  Of course, nature is extremely inventive, and we do not claim that all GRBs will fit into the framework outlined here. We would not expect to see all of the highly beamed jets following from the BZ mechanism head on, the jets may encounter some remaining hydrogen envelope in some cases, jets from lower magnetic fields than we have considered here may be much weaker and delivered over longer times, etc., so we speculate that a continuum of phenomena may exist between normal supernovae and extreme hypernovae/GRBs."
    },
    "0208/astro-ph0208132_arXiv.txt": {
        "abstract": "We investigate the kinematics of the central gas disk of the radio-loud elliptical galaxy NGC 4335, derived from HST/STIS long-slit spectroscopic observations of {\\HalphaNII} along 3 parallel slit positions. The observed mean velocities are consistent with a rotating thin disk. We model the gas disk in the customary way, taking into account the combined potential of the galaxy and a putative black hole with mass $\\Mbh$, as well as the influence on the observed kinematics of the point spread function and finite slit width. This sets a 3$\\sigma$ upper limit of $10^8 \\Msun$ on $\\Mbh$. The velocity dispersion at $r \\lta 0.5''$ is in excess of that predicted by the thin rotating disk model. This does not invalidate the model, if the excess dispersion is caused by localized turbulent motion in addition to bulk circular rotation. However, if instead the dispersion is caused by the BH potential then the thin disk model provides an underestimate of $\\Mbh$. A BH mass $\\Mbh \\sim 6 \\times 10^8 \\Msun$ is inferred by modeling the central gas dispersion as due to an isotropic spherical distribution of collisionless gas cloudlets. The stellar kinematics for NGC 4335 are derived from a ground-based (WHT/ISIS) long-slit observation along the galaxy major axis. A two-integral model of the stellar dynamics yields $\\Mbh \\gta 3 \\times 10^9 \\Msun$.  However, there is reason to believe that this model overestimates $\\Mbh$.  Reported correlations between black hole mass and inner stellar velocity dispersion $\\sigma$ predict $\\Mbh$ to be $\\geq 5.4 \\times 10^8 \\Msun$ in NGC 4335. If our standard thin disk modeling of the gas kinematics is valid, then NGC 4335 has an unusually low $\\Mbh$ for its velocity dispersion. If, on the other hand, this approach is flawed, and provides an underestimate of $\\Mbh$, then black hole masses for other galaxies derived from HST gas kinematics with the same assumptions should be treated with caution.  In general, a precise determination of the $\\Mbh - \\sigma$ relation and its scatter will benefit from (i) joint measurements of $\\Mbh$ from gas and stellar kinematics in the same galaxies and (ii) a better understanding of the physical origin of the excess velocity dispersion commonly observed in nuclear gas disks of elliptical galaxies. ",
        "introduction": "\\label{s:intro}  The Hubble Space Telescope has made it possible to measure the masses of black holes (BHs) in the centers of many nearby active and quiescent galaxies using stellar and/or gaseous kinematics (for reviews, see e.g., Richstone \\etal 1998; van der Marel 1999; Ho 1999; de Zeeuw 2001; Kormendy \\& Gebhardt 2001). To date, BH masses have been measured in about 40 galaxies, both spirals and ellipticals, and this number continues to increase. The BH masses correlate loosely with host spheroid luminosity (Kormendy \\& Richstone 1995) and more tightly with inner stellar velocity dispersion (Gebhardt \\etal 2000; Ferrarese \\& Merritt 2000).  Central emission-line gas is detected in virtually all nearby radio galaxies, defined here as galaxies which harbor kpc-scale radio-jets. The equivalent widths of the gas emission lines are generally much larger than those of the absorption lines in the integrated stellar light, so that measurement of the kinematics of nuclear emission-line gas is an efficient way to determine the central gravitational potential and BH mass (e.g., Harms \\etal 1994; Ferrarese, Ford \\& Jaffe 1996; Macchetto \\etal 1997; van der Marel \\& van den Bosch 1998; Ferrarese \\& Ford 1999; Verdoes Kleijn \\etal 2000, hereafter VK00; Sarzi \\etal 2001; Barth \\etal 2001). A drawback of this approach is that the gas kinematics might be affected by non-gravitational motions. Nevertheless, using the gas kinematics to determine accurate BH masses is particularly interesting for radio galaxies because determining the BH mass and the properties of the gas disk can advance our understanding of radio-jet formation and evolution. In turn, the radio-jet offers an extra diagnostic of the BH accretion and immediate BH surroundings.  Current questions include: what is the lower-limit to BH masses that can form kpc-scale jets? Is this lower-limit higher or lower than typical BH masses in spirals which (almost) never show kpc-scale jets? Is there a correlation between BH mass and jet properties, such as total power or jet velocity?  For these and other reasons we are performing a systematic study of a complete sample of nearby radio galaxies with Fanaroff \\& Riley (1974) Type I radio jets, the `UGC FR-I sample' (Verdoes Kleijn \\etal 1999, hereafter VK99; Xu \\etal 2000) using observations at multiple wavelengths. In particular, we have performed a STIS spectroscopic survey of the inner gas distributions to measure the kinematics and the physical state of the gas (Noel-Storr \\etal 2002).  Here we concentrate on one galaxy from our sample, NGC 4335, which is a relatively unknown isolated giant elliptical ($M_B=-20.7^m$; Paturel \\etal 1997) at 66 Mpc. The gas appears embedded in a well-defined dust disk (diameter $\\sim 750 \\pc$) and gas kinematics can be traced sufficiently far out along the three slits to allow detailed gas dynamical modeling. In addition we perform stellar dynamical modeling for NGC 4335 using the stellar kinematics derived from a WHT/ISIS long-slit observation.  The paper layout is as follows. Sections~\\ref{s:wfpc2} and \\ref{s:spec} present HST/WFPC2 broad- and narrowband imaging, HST/STIS gas emission-line spectroscopy and ground-based WHT/ISIS stellar absorption-line spectroscopy, including the basic data reduction and derivation of the gaseous and stellar kinematics. Section~\\ref{s:modelH} describes the modeling of the gas disk flux distribution, the derivation of the stellar mass distribution and the fits to the observed gaseous kinematics to estimate the BH mass. Section~\\ref{s:starkin} describes two-integral modeling of the WHT stellar kinematics to determine the stellar mass-to-light ratio and to constrain the BH mass independently.  Section~\\ref{s:discon} discusses the implications of the BH mass measurements in NGC 4335 for our understanding of BH demography, and for the techniques used to measure black hole masses.  We adopt $H_0 = 70 \\kms \\Mpc^{-1}$ throughout this paper. This does not directly influence the data-model comparison for any of our models, but does set the length, mass and luminosity scales of the models in physical units. Specifically, distances, lengths and masses scale as $H_0^{-1}$, while mass-to-light ratios scale as $H_0$. ",
        "conclusions": "\\label{s:discon}  Before analyzing the implications of the inferred black hole mass we first discuss the two main differences between our gas disk dynamical modeling and the modeling used for NGC 3245 presented by Barth \\etal (2001) which constitutes the current state-of-the-art.  First, instead of the observed emission-line flux distribution, we have used a double exponential fit to the flux distribution. For NGC 3245 the use of an exponential fit changes the inferred $\\Mbh$ by $30\\%$.  Moreover, it accounted less well for the wiggles in the velocity profile. In our case the deviations of the exponential model from the observed photometric and spectroscopic emission-line fluxes are estimated to have a $\\sim 25\\%$ effect on BH mass (see Section~\\ref{s:bestfit}). Second, we do not incorporate the velocity shifts due to asymmetric illumination of the slit by the gas disk in the dispersion direction. In other galaxies this could be important to derive the BH mass as they produce surface brightness 'caustics' (Maciejewski \\& Binney 2001). These caustics are not observed in NGC 4335. STIS observations of stars by one of us show that the velocity shift amounts to at most $\\sim 30\\kms$ for a $0.2''$ wide slit in the extreme case of a star at the edge of the slit. Only the (central) apertures in the adjacent slits have an asymmetric flux gradient in the dispersion direction. The kinematics from these apertures are not taken into account in deriving the upper limit to $\\Mbh$. Hence we conclude that the velocity shift has negligible effect on the derived $\\Mbh$.  Black hole mass $\\Mbh$ correlates in general rather tightly with observed velocity dispersion $\\sigma$ in the inner region of galaxies (Gebhardt \\etal 2000; Ferrarese \\& Merritt 2000). Recently, Tremaine \\etal (2002) completed a detailed analysis of the correlation and we will use their results in what follows. They find a best fit correlation of the form $\\log\\Mbh = \\alpha + \\beta\\log(\\sigma/\\sigma_0)$ with $\\alpha=8.13\\pm0.06$ and $\\beta=4.02\\pm0.32$ for $\\sigma_0=200\\kms$. The observed spread in the correlation indicates that the intrinsic dispersion in $\\log\\Mbh$ is 0.3 dex (perhaps smaller if observational errors are underestimated). Measuring the central $\\sigma$ in similar fashion as in Tremaine \\etal (2002) gives $\\sigma=282\\kms$ for NGC 4335 (cf.\\ Section~\\ref{s:stellarkin}) which corresponds to a predicted $\\Mbh=5.4 \\times 10^8 \\Msun$. The $3\\sigma$ upper-limit $\\Mbh < 10^8 \\Msun$ from the thin disk modeling of the gas mean velocities (Section~\\ref{s:bestfit}) falls well below the relation, even when including the reported intrinsic dispersion in $\\Mbh$. The residuals are $-0.73$ dex and $-0.43$ dex for the upper-limit of $\\Mbh=1.0 \\times 10^8 \\Msun$ (best-fit model) and $\\Mbh=2.0 \\times 10^8 \\Msun$ (for a maximally face-on gas disk), respectively (see Figure~\\ref{f:bhsigma}). The residuals are even larger if we use the best-fit relation as determined by Ferrarese (2002) who infers a larger $\\beta=4.58$.  The $\\Mbh \\gta 3 \\times 10^9\\Msun$ derived from stellar dynamics corresponds to an equally large, but positive residual of $\\gta 0.7$ dex.  Finally, the gravitational modeling of the gas velocity dispersions yields $\\Mbh \\sim 6 \\times 10^8 \\Msun$ corresponding to a residual of $0.05$ dex, well within the reported intrinsic dispersion of the correlation.  Which BH mass are we to trust?  The analysis strongly supports a thin rotating gas disk model at $r \\gta 0.5''$.  The model provides a reasonable fit to the gas mean velocities and dispersions, the dust disk morphology and gas disk kinematics indicate the same inclination and PA, and the $\\Upsilon$ from stellar and gas kinematics is consistent.  However, there are doubts for the validity of the model at $r \\lta 0.5''$. The WFPC2 and STIS fluxes are consistent with a thin disk surface brightness profile but the signal-to-noise of the WFPC2 image is too low to rule out a more spherical distribution.  An excess velocity dispersion with an irregular profile is observed at $r \\lta 0.5''$. Only an ad hoc explanation of localized random motion exists for this excess. It is not clear if such quasi-stationary turbulence in a thin disk is physically viable (e.g., Wada , Meurer \\& Norman 2002; and references therein). If the excess velocity dispersion is (partly) due to gravitational motion around the BH, then the thin disk model underestimates the true BH mass. Ascribing all gas kinetic energy, including the excess velocity dispersion, as counterbalancing the gravitational potential in a simple manner of isotropically moving collisionless cloudlets yields a BH mass which agrees well with the $\\Mbh - \\sigma$ correlation (Tremaine \\etal 2002).  Ascribing the nuclear velocity dispersion to rotational motion from a spatially {\\sl unresolved} disk yields an equally good agreement (Section~\\ref{s:dispersion}).  As discussed in more detail in Section~\\ref{s:starkin}, it is not too unreasonable to assume that the stellar dynamical modeling overestimates the BH mass by a factor of $\\sim 5$. This would be due to radial anisotropy (often observed in bright ellitpicals such as NGC 4335) which the two-integral modeling does not take into account. In conclusion, the gas spheroidal model infers a BH mass in accordance with the empirical $\\Mbh - \\sigma$ relation, but remains very simplistic. If doubts about the validity of the thin disk modeling for $r \\lta 0.5''$ were proven true, its inferred BH mass is expected to be an underestimate, driving the expected true BH mass to higher values in better agreement with the $\\Mbh - \\sigma$ relation. Similarly, the expected corrections for the stellar dynamical model bring its predicted BH mass in better agreement with the relation.  What are the implications if in reality NGC 4335 indeed harbors a BH with $\\Mbh \\sim 6.0 \\times 10^8\\Msun$?  Our results then suggest that gas dynamical modeling assuming thin rotating disks cannot be used to derive accurate black hole masses, {\\sl even when all of the following are true}: (i) the gas mean velocities very clearly suggest rotation; (ii) the surrounding dust disk appears regular; (iii) the inclination and position angle of the inner gas disk are well-constrained due to the use of three adjacent slits and these angles are both indicated by the gas disk kinematics and independent methods; and (iv) the derived $\\Upsilon_I$ from gas and stars agree, ruling out asymmetric drift at $r \\gta 0.5''$. This then would cast doubt on other $\\Mbh$ values determined from gas kinematics using similar models.  Figure~\\ref{f:bhsigma} shows that these measurements have a large influence on the upper-end of the $\\Mbh$ detections and hence on the slope of the correlation. Moreover, to date all measurements of $\\Mbh$ in nearby radio galaxies are based on gas kinematics.  However, the fact that the inferred BH masses using this method follow the best-fit relation more closely than the stellar dynamical measurements argues against this worry of the validity of the gas dynamical modeling in general. The $\\chi^2$ per degree of freedom as used in deriving the best-fit relation (cf.\\ Equation 3 in Tremaine \\etal 2002) is 0.27 for the gas dynamical measurements while for the stellar dynamical measurements it is 1.30 (1.17 if one excludes the Milky Way BH mass measurement which is obtained from a stellar dynamical model which is completely different from those used for external galaxies). Moreover, the average residual of the gas dynamical BH mass measurements is positive, counter to what is expected if the gas mean velocities systematically underestimate the circular velocity and hence the inferred BH mass.  Independent determinations of $\\Mbh$ in galaxies from gaseous and stellar kinematics observed at high spatial resolution are crucial to address the worries about the gas dynamical modeling. This has been performed for IC 1459 (Verdoes Kleijn \\etal 2000; Cappellari \\etal 2002). Verdoes Kleijn \\etal derive from the gaseous kinematics at 6 FOS pointings a black hole mass ranging from $\\Mbh=1.5 \\times 10^8\\Msun$ (thin disk model) to $6 \\times 10^8\\Msun$ (isotropic spheroidal model). However, Cappellari \\etal (2002) infer $\\Mbh=(2.6 \\pm 1.1) \\times 10^9\\Msun$ based on three-integral modeling of combined ground-based and HST/STIS stellar kinematics. Unfortunately, the HST stellar kinematics cannot be measured accurately inside the sphere of influence for this BH mass.  Cappellari \\etal also present a more complete view of the gas kinematics from a recent STIS long-slit observation. This indicates that while the data are consistent with the earlier FOS measurements, and the inferred BH mass is similar (modeled with independently developed software), the gas mean velocities are rather perturbed in this particular case. Moreover, IC1459 has an irregular dust distribution.  Thus, the gas and dust properties in IC 1459 are quite different from the photometrically and kinematically well-behaved central disk in NGC 4335. An independent $\\Mbh$ measurement based on the stellar kinematics for NGC 4335 at HST resolution, and for other galaxies with similar gas disk kinematics, would test the validity of the gas dynamical modeling in well-behaved gas disks as opposed to irregular gas disks.  It is similarly crucial to understand the origin of the excess velocity dispersion commonly seen in nuclear gas disks. Presently, it is not clear under which circumstances the thin gas disks become locally turbulent (for example, gravitational or MHD instabilities) and if they remain globally stable (e.g., Wada, Meurer \\& Norman 2002; and references therein). A better idea for the origin of the excess is needed to improve on the highly idealized collisionless spherical model discussed in this paper.  In this respect, it is also useful to determine the ionization mechanism of the emission-line gas in nuclear disks. If it is shocks we should be wary about the modeling with unperturbed, infinitely thin rotating disks. Furthermore, the disks of gas and dust might perhaps interact with ambient hot X-ray gas which is often present in the centers of bright ellipticals (e.g., Gunn 1979). The ultimate goal for gas dynamical modeling is thus to explain self-consistently the density, ionization state and dynamical state of the central gas distributions. This requires high S/N two-dimensional photometry, kinematics and line ratios to determine the dynamics, ionization state, density and temperature as a function of disk radius. This will provide not only accurate BH masses but also a vast improvement in our understanding of the fueling of BHs.  Kiloparsec-scale radio jets are only seen in early-type galaxies but never in spirals (with one possible exception known to the authors; Ledlow \\etal 2001). For instance, the UGC FR-I galaxies all have Hubble types E-S0 (cf.\\ VK99). Prior to this study, only BH masses $\\Mbh \\geq 2 \\times 10^8\\Msun$ have been reported in nearby radio galaxies (i.e., NGC 4261, NGC 4374, M87, NGC 5128, NGC 6251 and NGC 7052, see Tremaine \\etal 2002 for references). Interestingly, none of the seven galaxies with a Hubble type later than S0 in the the sample compiled by Tremaine \\etal (2002) have $\\Mbh > 1 \\times 10^8\\Msun$. We are aware of only one galaxy with a Hubble type later than S0 and a $\\Mbh > 2 \\times 10^8\\Msun$, which is the Sombrero galaxy (M104, Hubble type Sa) with $\\Mbh = 1.0 \\times 10^9\\Msun$, but this measurement is based on models less general than three-integral stellar dynamical models.  By contrast, all BH mass upper limits and detections in a sample of 16 mostly early-type disk galaxies (Hubble type S0-Sb; Sarzi \\etal 2001) are below $2 \\times 10^8 \\Msun$. These results suggest that the differences in BH mass might be the underlying factor for this host preference of radio-jets. However, if our determination of a $3\\sigma$ upper limit of $\\Mbh < 10^8\\Msun$ on the BH mass of NGC 4335 is correct, then NGC 4335 would illustrate that BHs with $\\Mbh < 10^8\\Msun$ are also capable of producing FR-I radio jets. This would argue against BH mass being the (only) parameter underlying the host morphology preference of radio galaxies."
    },
    "0208/astro-ph0208418_arXiv.txt": {
        "abstract": "We have performed hydrodynamical simulations to investigate the effects of galactic winds on the high-redshift ($z=3$) universe. Strong winds suppress the formation of low-mass galaxies significantly, and the metals carried by them produce \\Cfour{} absorption lines with properties in reasonable agreement with observations. The winds have little effect on the statistics of the \\h{}-absorption lines, because the hot gas bubbles blown by the winds fill only a small fraction of the volume and because they tend to escape into the voids, thereby leaving the filaments that produce these lines intact. ",
        "introduction": "Feedback from star formation is thought to play an important role in the formation of galaxies. For example, theoretical models require feedback in order not to overproduce the number of low mass galaxies (White \\& Rees 1978) and the fraction of baryons that cools (e.g., Balogh et al.\\ 2001). Models without feedback also predict disk galaxies that are too small (e.g., Navarro \\& Steinmetz 1997) and an X-ray background that is too strong (e.g., Pen 1999).  Observations of galactic winds indicate that strong feedback processes do indeed occur. Observations in X-rays, and of optical and UV lines, indicate that most local starbursts (e.g., Heckman 2000), as well as high-redshift Lyman Break galaxies (e.g., Pettini et al.\\ 2001), drive winds with a mass loss rate comparable to their star formation rate. The wind speeds are high, 100-1000~\\kms, and so the conversion of supernova (SN) energy into kinetic energy must be quite efficient. This process is currently not well understood because of the complications that arise from the multiphase nature of the interstellar medium (ISM, e.g., McKee \\& Ostriker 1977; Efstathiou 2000). Simulations that take some of these complications into account, have shown that SNe can indeed plausibly power a wind (e.g., Mac Low \\& Ferrara 1999).  Although successful models of galaxies appear to require feedback, the same is not true for models of the high-redshift ($z \\ga 2$) intergalactic medium (IGM). The IGM can be studied in great detail from the properties of the hydrogen absorption lines, seen in the spectra of background quasars (see Rauch 1998 for a review). Hydrodynamical simulations of the IGM (see Efstathiou, Schaye \\& Theuns 2000 for a recent review) as well as semi-analytic (Bi \\& Davidsen 1997; Viel et al.\\ 2002) and analytic (Schaye 2001) models, have been very successful in reproducing the statistical properties of the observed \\h{} lines. These models, which generally do \\emph{not} take feedback into account (but see Cen \\& Ostriker 1999), suggest that the Ly$\\alpha$ forest absorption arises in a network of voids and filaments, with the higher column density absorbers located at the intersections of filaments. The low column density absorbers are extended structures with densities around the cosmic mean, which contain a large fraction of the baryons in the universe.  Although feedback has so far largely been ignored in models of the IGM, the detection of metals in the IGM suggests that it {\\em does} play a role. The higher column density ($N_\\h \\ga 10^{14.5}~\\cm^{-2}$) Ly$\\alpha$ absorption systems generally have detectable absorption by \\Cfour\\ (Cowie et al.\\ 1995) and, at least at $z\\la 2.5$, \\Osix\\ (Carswell, Schaye, \\& Kim 2002). Furthermore, there is statistical evidence that metals are also present at somewhat lower densities (Cowie \\& Songaila 1998; Ellison et al.\\ 2000; Schaye et al.\\ 2000a). Simple photo-ionization models indicate that the absorbers have a metallicity of order 0.1 -- 1 per cent solar (e.g., Cowie et al.\\ 1995; Rauch, Haehnelt, \\& Steinmetz 1997; Hellsten et al.\\ 1997; Carswell et al.\\ 2002).  If galactic winds are ubiquitous and able to transport mass to large distances, then they may be responsible for enriching the IGM with metals. Indeed, numerical simulations of galactic winds in a cosmological setting (e.g., Gnedin 1998; Cen \\& Ostriker 1999; Aguirre et al.\\ 2001a, 2001b; Thacker, Scannapieco, \\& Davis 2002; Springel \\& Hernquist 2002) suggest that winds could enrich a substantial volume fraction of the IGM to the inferred levels.  Therefore, both observational and theoretical considerations suggest that some fraction of galaxies may undergo an episode in which they blow a strong wind into their surroundings. This may have observational effects on the \\lya{} forest in quasar spectra (Theuns, Mo, \\& Schaye 2000; Croft et al.\\ 2002), which may have already been detected (Rauch et al.\\ 2001; Adelberger et al.\\ 2002).  In this {\\em Letter}, we use hydrodynamical simulations to investigate whether galactic winds that strongly influence the properties of small galaxies, and are effective in enriching the IGM with metals, can do so without undermining the success of current models of the IGM. \\begin{figure*} \\label{fig:reion} \\centerline{\\resizebox{1.92\\colwidth}{!}{\\includegraphics{f1_color_idl.ps}}} \\figcaption{ {\\it Top panels:} Density (a, b) and temperature (c, d) for a slice through the simulation with feedback (a, c) and the simulation without feedback (b, d), at redshift $z=3$. The simulation box is 5.0$h^{-1}$ (co-moving) \\Mpc{}. The slice is chosen to go through the most massive object in the simulation. In the feedback simulation, hot bubbles of gas surround the galaxies. These tend to expand into the voids, thereby leaving the filamentary network of higher density regions unaffected. The four numbered horizontal lines (from top to bottom) are sight lines for which the \\lya{} spectrum, temperature, and density, are shown in the bottom panel (from left to right, offset for clarity). The feedback and no feedback cases are shown as blue and red lines, respectively. Only the strongest lines are significantly affected by feedback.} \\end{figure*} ",
        "conclusions": ""
    },
    "0208/hep-th0208169_arXiv.txt": {
        "abstract": "\\noindent We detail the global structure of the five-dimensional bulk for the cosmological evolution of Dvali-Gabadadze-Porrati braneworlds.  The picture articulated here provides a framework and intuition for understanding how metric perturbations leave (and possibly reenter) the brane universe.  A bulk observer sees the braneworld as a relativistically expanding bubble, viewed either from the interior (in the case of the Friedmann--Lema\\^{\\i}tre--Robertson--Walker phase) or the exterior (the self-accelerating phase).  Shortcuts through the bulk in the first phase can lead to an apparent brane causality violation and provide an opportunity for the evasion of the horizon problem found in conventional four-dimensional cosmologies.  Features of the global geometry in the latter phase anticipate a depletion of power for linear metric perturbations on large scales. ",
        "introduction": "The gravity theory of Dvali--Gabadadze--Porrati (DGP) is a braneworld theory with a metastable four-dimensional graviton \\cite{Dvali:2000hr}.  The graviton is pinned to a four-dimensional braneworld by intrinsic curvature terms induced by quantum matter fluctuations; but as it propagates over large distances, the graviton eventually evaporates off the brane into an infinite volume, five-dimensional Minkowski bulk.  As a result, the DGP braneworld theory is a model in a class of theories in which gravity deviates from conventional Einstein gravity not at short distances (as in more familiar braneworld theories), but rather at long distances.  Such a model has both intriguing phenomenological \\cite{Dvali:2001gm,Dvali:2001gx,Lue:2001gc,Gruzinov:2001hp} as well as cosmological consequences \\cite{Deffayet,Deffayet:2001pu,Deffayet:2002sp,Alcaniz:2002qh,Jain:2002di,Alcaniz:2002qm}.  A braneworld model of the sort where gravity is modified at extremely large scales is motivated by the desire to ascertain how our understanding of cosmology may be refined by the presence of extra dimensions.  Observational pillars of the standard cosmological model, including the cosmic microwave background and large scale structure, provide important tests of theories that seriously modify physics at cosmological scales.  Deffayet's cosmological equations for the DGP model already point to important deviations from the standard model \\cite{Deffayet:2001pu,Deffayet:2002sp,Alcaniz:2002qh,Jain:2002di,Alcaniz:2002qm}. In order to constrain the DGP theory further, understanding the development of large scale structure is necessary.  Leakage of gravitational energy into the bulk is a key aspect of the DGP braneworld model, and one expects such leakage to modify the spectrum of density fluctuations at the largest observable cosmological scales. However, a detailed analysis, or even an intuitive understanding, of how spectral power of metric perturbations on the brane fills the bulk and (potentially) reenters the brane depends on the global structure of the brane worldsheet.  In this paper, we articulate the global structure of the five-dimensional bulk for the cosmological evolution of DGP braneworlds as a first step in understanding how large scale structure in the universe is modified.  After reviewing the particulars of the model, we take the equations laid out in \\cite{Deffayet} and show how one may interpret the evolution of the braneworld as a relativistically expanding bubble, viewed either from the interior or the exterior, depending on the specific phase of the theory.  We then go on to examine the cosmological time foliations of the bulk and show how peculiarities arise, such as shortcuts through the bulk (leading to effective brane causality violation\\footnote{Such shortcuts seem to be prevalent in braneworld theories.  For previous studies of this phenomenon and how such shortcuts relate to four-dimensional Lorentz symmetry violation, see for example \\cite{Chung:1999xg,Csaki:2000dm,Caldwell:2001ja,Deffayet:2001aw}.}), and breakdowns of that foliation occur in different regimes of the theory, and discuss how understanding the global geometry of DGP braneworld evolution may offer insight into cosmological perturbations. ",
        "conclusions": "We examined the global structure of early universe cosmologies for Dvali--Gabadadze--Porrati (DGP) braneworlds.  Two distinct phases exist: the Friedmann--Lema\\^{\\i}tre--Robertson--Walker (FLRW) phase, where late time evolution follows five-dimensional FLRW behavior, and the self-accelerating phase, where the brane asymptotes to late time deSitter expansion with an empty brane.  A bulk observer sees the brane as a relativistically expanding, roughly hyperspherical bubble emerging from a pointlike big bang.  Depending on the spatial curvature of the {\\em internal} brane cosmology, the brane is strictly hyperspherical (positive spatial curvature, $k=1$), has one residual big bang point on the hypersphere moving at exactly the speed of light (flat spatial geometry, $k=0$), or two diametrically opposed residual big bang points on the hypersphere, also moving at exactly the speed of light (negative spatial curvature, $k=-1$).  Note that a bulk observer perceives the brane as compact, even when a cosmological brane observer would not (i.e., the $k=0,-1$ cases).  The bulk is two identical copies of the space interior to this compact brane, glued across a ${\\cal Z}_2$--symmetric brane when in the FLRW phase. Correspondingly, the bulk space is two copies of the space exterior to the brane when in the self-accelerating phase.  The FLRW phase exhibits lightlike shortcuts through the bulk that connect different events on the brane.  This phenomena can lead to apparent brane causality violation and provides an opportunity for the evasion of the horizon problem.  Unlike the big bang of conventional four-dimensional FLRW cosmology, the past (bulk) lightcone of every spacetime event on the brane contains the entire big bang for all comoving coordinate radii $r < \\infty$.  Phrasing this statement another way is that the entire brane worldsheet is in the future lightcone of the big bang for $r < \\infty$, where the locus of spacetime events of the big bang with $r < \\infty$ is strictly pointlike.  Thus, gravitons emitted from a time arbitrarily close to the big bang may travel through the bulk and may transmit thermal information to all parts of the brane during the evolution of the early universe.  The difficulty in this mechanism for evading the horizon problem is that bulk propagation of gravitons is at its least significance precisely at those times (i.e., the early big bang) when we wish to transmit them through the bulk.  The self-accelerating phase does not possess lightlike worldline that connect different events on the brane.  This observation implies that density perturbations leaving the brane via the bulk cannot reenter the brane at some later time.  At sufficiently large scales, where brane fluctuations are strongly coupled to bulk modes, one anticipates that only depletion of power can occur for metric fluctuations, since there is no mechanism for replenishment of power at these scales. This argument implies a deficit in the power spectrum of linear density perturbations at large scales.  The magnitude of this deficit, as well as the validity of this analysis motivated by linearized perturbations, are subject to more quantitative inquiry. Nevertheless, one can see the advantage of developing an understanding of the global structure of DGP braneworlds for gaining insight into their phenomenological consequences."
    },
    "0208/astro-ph0208148_arXiv.txt": {
        "abstract": "We discuss plausible mechanisms to produce bullet-like ejecta from the precessing disk in the SS 433 system. We show that non-steady shocks in the sub-Keplerian accretion flow can provide the basic timescale of the ejection interval while the magnetic rubber-band effect of the toroidal flux tubes in this disk can yield flaring events. ",
        "introduction": "SS 433 remains one of the most enigmatic objects in the sky. Even twenty-five years after its first appearance in the catalogue of Stephanson \\& Sanduleak (1977), it is not clear whether the compact object is a black hole or a neutron star. However, there is ample evidence that the companion is an OB type star with an orbital period of 13.1 days, which is losing mass at the rate of about $10^{-4} M_\\sun ~{\\rm yr}^{-1}$ (van den Heuvel 1981), corresponding to extremely super-Eddington accretion regardless of the mass of the compact object.  One of the most curious properties of the jets of SS 433, which first made their presence distinctly felt through the emission of variable H$\\alpha$ lines, is that they are apparently ejected as bullets (e.g.\\ Borisov \\& Fabrika 1987;\\ Vermeulen et al.\\ 1993; Paragi et al.\\ 1999; 2002; Gies et al.\\ 2002), with a surprisingly nearly constant radial velocity of about $0.26c$. The absence of a significant intrinsic rotational velocity (i.e., $v_\\phi$) component is clear from the fact that the kinematic model (e.g.\\ Abell and Margon 1979), which assumes only radial injection, quite accurately explains the time variation of the red- and blue-shifts of the H$\\alpha$ emission from the jets with a period of 162 days, which is attributed to the precession of the accretion disk about the compact object. The radial velocity is less than the maximum allowed sound speed of $c/\\sqrt{3}$ and thus hydrodynamic acceleration could, in principle, explain it. Therefore one may not require a magnetic or electrodynamic acceleration processes (e.g.\\ Belcher \\& MacGregor 1976; Lovelace 1976). However, the rather good collimation (Margon 1984; Paragi et al.\\ 1999) supports the hypothesis that a substantial degree of confinement produced by toroidal flux tubes may be present. Gies et al.\\ (2002) showed that the ratios of the H$\\alpha$ emission equivalent widths from the approaching and receding jets as a function of precessional phase only could be nicely fit if these emission components are bullet-like. Indeed, the recent Chandra X-ray Observatory discovery of X-rays at a distance of about $10^{17}$cm from the center may result from the collision of such bullets (Migliari,  Fender, \\& Mendez  2002).  SS 433 poses another interesting problem: it was pointed out by Chakrabarti (1999) and Das \\& Chakrabarti (1999) that significant outflows are produced only when the accretion rate is such that the X-ray source is in a low/hard state, and all the observational indications in other micro-quasars also suggest that the jets are indeed produced in low/hard states (Corbel et al.\\ 2001; Klein-Wolt et al.\\ 2002). However, it is difficult to imagine how SS 433 manages to remain in the low/hard state with $10^{-4}M_\\sun ~{\\rm yr}^{-1}$ of wind matter ejected from its companion. The answer to this quandary probably lies in the recent results of Paragi et al.\\ (1999) and Blundell et al.\\ (2000), whose high resolution radio maps show that there is a large region of roughly $50 {\\rm AU}$ in radius which is filled with enough gas and dust to obscure the accretion disk and the base of the jets. They also found an equatorial outflow.  Gies et al.\\ (2002) present additional evidence from observations of the ``stationary'' H$\\alpha$ and He I lines for an extended ``disk wind''. So it is distinctly possible that most of the matter from the donor is rejected either by centrifugal force (Chakrabarti 2002) or by radiation force far outside the central accretion disk, and thus the compact object receives only a few times the Eddington rate ($\\dot M_{\\rm Ed}$) of its companion's wind matter to accrete. This consideration finds further support from the fact that the kinematic luminosity of the jet itself is around $10^{39}$ erg s$^{-1}$ (Margon 1984), which corresponds to about one Eddington rate for a $10M_\\odot$ compact object.  In numerical simulations of supercritical winds by Eggum, Coroniti \\& Katz (1985) designed to model SS 433, it was shown that only a fraction of a percent of the infalling matter is ejected from a radiation pressure supported Keplerian disk, which indicates that the accretion rate must be at least $100 \\dot M_{\\rm Ed}$ if the accretion takes place through a Keplerian disk. On the other hand, numerical simulations of a sub-Keplerian disk by Molteni, Lanzafame \\& Chakrabarti (1994) suggest that about $15-20$ percent of matter is ejected as an outflow, indicating that the accretion rate onto the compact object in SS 433 need be at most a few $\\dot M_{\\rm Ed}$. Similar simulations with different parameters yield situations where no steady shocks can form, even though two saddle-type sonic points are present (Ryu, Chakrabarti, \\& Molteni 1997, hereafter RCM); under these conditions large scale shock oscillations produce intermittent outflows instead of continuous outflows. Since the compact object is a wind accretor, a low-angular momentum, sub-Keplerian flow is the most likely description of the accretion flow. Indeed, the presence of sub-Keplerian flows in several other high mass X-ray binaries has now been verified (Smith, Heindl \\& Swank 2002).  In this {\\it Letter}, we present  a few scenarios leading to ejection of matter as bullets in SS 433. We discuss four possible ways to create blobs of matter emerging from  the disk and conclude that periodic ejection of the blobs by the large scale oscillation of an accretion shock (something like a  piston) may be the fundamental production mechanism of the ``normal'' bullets.  The irregularly observed rapid flaring (Vermeulen et al. 1993) could be understood in terms of the catastrophic collapse of toroidal magnetic flux tubes, very similar to what has been argued to be occurring in GRS 1915+105 (Vadawale et al. 2001; Nandi et al.\\ 2001). In the next section we discuss these processes and their suitability or unsuitability for SS 433. In \\S 3, we present concluding remarks. ",
        "conclusions": "In this {\\it Letter}, we have studied various competing processes for the creation of bullets which move ballistically in the jet of SS 433. We showed that blobs may be separated by: (1) Comptonization; (2) shock oscillations due to resonance; (3) oscillations due to inherent unsteady accretion solutions; (4) intense magnetic tension of the toroidal flux tubes. We reject the first possibility because it requires a large Keplerian disk, which is unlikely.  We are unable to distinguish at this stage which type of shock oscillation is more capable of producing bullet formation in SS 433, but we prefer the third possibility due to its impulsive and generic nature and smaller involved region. We believe that the fourth possibility of the inner disk evacuation should produce flaring events, but will occur rather rarely, perhaps only once in a single precession period, when the magnetic field of the companion is preferentially tilted towards the accretion disk during precessional motion. This fourth mechanism  gives rise to an anti-correlation between radio and X-rays, perhaps already observed in SS 433 (Safi-Harb \\& Kotani 2002).  SKC, AN, and SD acknowledge a grant from the Department of Science and Technology, India, and  financial support from CEA/Saclay where part of this work was performed. PJW is grateful for hospitality at the Department of Astrophysical Sciences, Princeton University, and for support from the Research Program Enhancement program at Georgia State University."
    },
    "0208/astro-ph0208462_arXiv.txt": {
        "abstract": "With the example of Proxima Centauri we discuss the feasibility of detecting terrestrial planets (1 to a few ${\\rm M}_{\\oplus}$) using the high precision radial velocity ($RV$) technique. If a very high $RV$ precision for M stars is achieved even planets with these extremely low masses become detectable. For Proxima Cen (M5V), one of the prime targets of our M-stars planet search program using the UVES spectrograph \\& iodine cell at the ESO VLT UT2, we obtain a long term $RV$ precision of $2.5~{\\rm m\\,s}^{-1}$. Based on numerical simulations we determine that this level of precision would have already allowed us to detect planets with $m\\sin i= 4$ to $6~{\\rm M}_{\\oplus}$ inside the habitable zone of Proxima Cen. ",
        "introduction": "All extrasolar planets orbiting main-sequence stars known to date are giant planets ranging in mass from $m \\sin i = 0.12~{\\rm M}_{\\rm Jupiter}$ to $17~{\\rm M}_{\\rm Jup}$ (with $i$ the unknown angle between the orbital plane and the sky). It is virtually impossible to detect terrestrial planets (rocky objects of 1 to a few ${\\rm M}_{\\oplus}$) around F, G and K-type stars using the high-precision $RV$ technique, since Earth-mass planets induce only negligible reflex-motions on their host stars. This scenario changes in the faint and low-mass regime of the Hertzsprung-Russell diagram. In the case of M-dwarf stars the low stellar primary mass leads to detectable $RV$ amplitudes even for planets of a few ${\\rm M}_{\\oplus}$ and less in short-period orbits. Fig.1 shows the $RV$ signatures of planets with $m \\sin i = 1$ to $2.5~{\\rm M}_{\\oplus}$ orbiting an M5V star with a mass of $0.1~{\\rm M}_{\\sun}$. Due to its intrinsic faintness the habitable zone is located very close to the star (Kasting, Whitmire, \\& Reynolds 1993). This, too, favors the detection by the $RV$ technique, since shorter periods also mean higher $RV$ amplitudes. Although these nearby planets are probably tidally locked into synchronous rotation, a 3-D climate study by Joshi, Haberle, \\& Reynolds (1997) demonstrated that these planets are still likely to be habitable.  \\begin{figure} \\plotfiddle{fig1.eps}{7.0cm}{270}{40}{40}{-180}{220} \\caption{Simulation of radial velocity variations ($RV$ semi-amplitude K) for a low-mass M-dwarf ($0.1~{\\rm M}_{\\sun}$) due to orbiting terrestrial planets plotted vs. orbtial separation. The three curves display the K-amplitudes for planets with $m \\sin i = 1~{\\rm M}_{\\oplus}$ (solid), $1.5~{\\rm M}_{\\oplus}$ (dashed) and $2~{\\rm M}_{\\oplus}$ (dotted curve) residing in circular orbits. The vertical dashed lines show the borders of the habitable zone for this type of star after Kasting et al. (1993). } \\end{figure} ",
        "conclusions": "Measuring radial velocities of M-dwarfs with the appropriate high level of precision allows us to detect extremely low-mass planets in short period orbits. With the 22 $RV$ measurments we obtained so far for Proxima Cen, we could have already detected {\\it all} planets with $m\\sin i= 4$ to $6~{\\rm M}_{\\oplus}$ inside the habitable zone. So the answer to the title question is: {\\it not yet}, but we get pretty close, and our sensitivity will further improve by extending the monitoring time span."
    },
    "0208/astro-ph0208181_arXiv.txt": {
        "abstract": "In this paper I present some recent and new results regarding the effects of the cluster environment on the star formation history of galaxies. Three main aspects are discussed: the differences in the stellar population ages of cluster ellipticals and S0 galaxies; the comparison of the spectroscopic properties of galaxies in distant clusters and in Coma; and the incidence and properties of faint post-starburst/post-starforming galaxies in the Coma cluster. ",
        "introduction": "In the quest for undertanding how galaxy evolution is affected by environmental processes, galaxy optical spectra provide precious informations regarding the evolutionary histories and stellar population content of galaxies. Here I summarize some of the latest results obtained from spectroscopy of cluster galaxies, both at low and high redshift. ",
        "conclusions": ""
    },
    "0208/astro-ph0208238_arXiv.txt": {
        "abstract": "We are presenting a detailed parameter study of the time-dependent electron injection and kinematics and the self-consistent radiation transport in jets of intermediate and low-frequency peaked BL~Lac objects. Using a time-dependent, combined synchrotron-self-Compton and external-Compton jet model, we study the influence of variations of several essential model parameters, such as the electron injection compactness, the relative contribution of synchrotron to external soft photons to the soft photon compactness, the electron-injection spectral index, and the details of the time profiles of the electron injection episodes giving rise to flaring activity. In the analysis of our results, we focus on the expected X-ray spectral variability signatures in a region of parameter space particularly well suited to reproduce the broadband spectral energy distributions of intermediate and low-frequency peaked BL Lac objects. We demonstrate that SSC- and external-Compton dominated models for the $\\gamma$-ray emission from blazars are producing significantly different signatures in the X-ray variability, in particular in the soft X-ray light curves and the spectral hysteresis at soft X-ray energies, which can be used as a powerful diagnostic to unveil the nature of the high-energy emission from BL~Lac objects. ",
        "introduction": "Introduction}  The class of objects referred to as blazars consists of the most extreme examples of active galactic nuclei (AGNs), namely $\\gamma$-ray loud, flat-spectrum radio quasars (FSRQs), and BL~Lac objects. They have been observed in all wavelength bands --- from radio through very-high energy (VHE) $\\gamma$-ray frequencies. More than 65 blazars have been identified as sources of $> 100$~MeV emission detected by the EGRET telescope on board the {\\it Compton Gamma-Ray Observatory} (CGRO) \\citep{hartman99}, and at least 6 blazars have now been detected at VHE $\\gamma$-rays ($> 350$~GeV) by ground-based air \\v Cerenkov telescopes (for a recent review see, e.g., \\cite{buckley01}). Blazars exhibit variability at all wavelengths \\citep{vm95,mukherjee97} on time scales --- in some cases --- down to less than an hour \\citep{gaidos96}.  The broadband continuum spectra of blazars are dominated by non-thermal emission and consist of at least two clearly distinct, broad spectral components. A sequence of sub-classes of blazars can be defined through increasing peak frequencies and a decreasing dominance of the $\\gamma$-ray output in terms of $\\nu F_{\\nu}$ peak flux along a sequence from FSRQs via low-frequency BL~Lac objects (LBLs) to high-frequency peaked BL~Lac objects (HBLs), which is also correlated with a decreasing inferred bolometric luminosity of the sources \\citep{fossati98}. For recent reviews of the observational properties of blazars see, e.g., \\cite{sambruna00}, \\cite{pu01}, or \\cite{boettcher02}.  Although all extragalactic sources detected by ground-based air \\v Cerenkov telescope facilities to date are HBLs, the steadily improving flux sensitivities and decreasing energy thresholds of those instruments provide a growing potential to extend their blazar source list towards intermediate and even low-frequency peaked BL~Lac objects. The detection of such objects at energies $\\sim 40$ -- 100~GeV might provide an opportunity to probe the intrinsic high-energy cutoff of their spectral energy distributions (SEDs) since at those energies, $\\gamma\\gamma$ absorption due to the intergalactic infrared background is expected to be negligible at redshifts of $z \\lesssim 0.2$ \\citep{djs02}. Such detections should significantly further our understanding of the relevant radiation mechanisms responsible for the high-energy emission of blazars and the underlying particle acceleration mechanisms.  The low-energy component of blazar SEDs is well understood as synchrotron emission from ultrarelativistic electrons in a relativistic jet directed at a small angle with respect to the line of sight. In the framework of leptonic models (for a review of the alternative class of hadronic jet models, see, e.g., \\cite{rachen00}), high-energy emission will result from Compton scattering of lower-frequency photons off the relativistic electrons. Possible target photon fields for Compton scattering are the synchrotron photons produced within the jet (the SSC process; \\cite{mg85,maraschi92,bm96}), or external photons (the EC process). Sources of external seed photons include the UV -- soft X-ray emission from the disk --- either entering the jet directly \\citep{dsm92,ds93} or after reprocessing in the broad line region (BLR) or other circumnuclear material \\citep{sikora94,bl95,dss97} ---, jet synchrotron radiation reflected at the BLR \\citep{gm96,bednarek98,bd98}, or the infrared emission from circumnuclear dust \\citep{blaz00,arbeiter02}.  According to the now well-established AGN unification scheme \\citep{up95}, blazars can be unified with other classes of AGN, in particular radio galaxies, through orientation effects. However, \\cite{sambruna96} have pointed out that such orientation effects can not explain the differences between different blazar sub-classes. Instead, it has been suggested that the sequence of spectral properties of blazars from HBLs via LBLs to FSRQs can be interpreted in terms of an increasing total power input into non-thermal electrons in the jet, accompanied by an increasing contribution of external photons to the seed photon field for Compton upscattering \\citep{madejski98,ghisellini98}. It has been suggested that this may be related to an evolutionary effect due to the gradual depletion of the circumnuclear material being accreted onto the central black hole \\citep{dc00,cd02,bd02}. Detailed modeling of blazars in the different sub-classes (FSRQs, LBLs and HBLs) seems to confirm this conjecture (for a recent review, see, e.g., \\cite{boettcher02}).  As mentioned earlier, blazars tend to exhibit rapid flux and spectral variability. The variability is most dramatic and occurs on the shortest time scales at the high-energy ends of the two nonthermal spectral components of their broadband SEDs. Particularly interesting variability patterns could be observed at X-ray energies for those blazars whose X-ray emission is dominated by synchrotron emission. Observational studies of X-ray variability in blazars have so far focused on HBLs and, in particular, on the attempt to identify clear patterns of time lags between hard and soft X-rays. However, such studies have yielded rather inconclusive and often contradictory results (e.g., for Mrk~421: \\cite{takahashi96,fossati00,takahashi00}; or PKS~2155-304: \\cite{chiapetti99,zheng99,kataoka00,edelson01}). Instead, the so-called ``spectral hysteresis'' of blazar X-ray spectral variability may prove to be a more promising diagnostic of the physical nature of acceleration and cooling processes in blazar jets: When plotting the X-ray spectral hardness vs. the X-ray flux (hardness-intensity diagrams = HIDs), some HBLs (e.g., Mrk~421 and PKS~2155-304) have been observed to trace out characteristic, clockwise loops (\\cite{takahashi96,kataoka00}). In terms of pure SSC jet models, such spectral hysteresis can be understood as the synchrotron radiation signature of gradual injection and/or acceleration of ultrarelativistic electrons into the emitting region, and subsequent radiative cooling \\citep{kirk98,gm98,kataoka00,kusunose00,lk00}. However, interestingly, such spectral hysteresis could not be confirmed in a recent series of {\\em XMM-Newton} observations of Mrk~421 \\citep{sembay02}.  In LBLs, the soft X-ray emission is also sometimes dominated by the high-energy end of the synchrotron component \\citep{tagliaferri00,ravasio02}, so similar spectral hysteresis phenomena should in principle be observable. However, those objects are generally much fainter at X-ray energies than their high-frequency peaked counterparts, making the extraction of time-dependent spectral information an observationally very challenging task (see, e.g., \\cite{bllac02}), which may require the new generation of X-ray telescopes such as {\\em Chandra} or {\\em XMM-Newton}. Extracting the physical information contained in the rich X-ray variability patterns exhibited by BL~Lac objects requires detailed theoretical modeling of the time-dependent particle acceleration and radiation transport processes in the jets of blazars. Previous analyses of these processes \\citep{kirk98,gm98,cg99,kataoka00,kusunose00,lk00,krawczynski02} have led to significant progress in our understanding of the particle acceleration and radiation mechanisms in HBLs, but were restricted to pure SSC models, with parameter choices specifically targeted towards HBLs. Consequently, those results may not be directly applicable to intermediate or low-frequency peaked BL~Lac objects or even FSRQs. A notable exception is a recent study by \\cite{sikora01} (see also \\cite{moderski02}), who included a significant contribution of external Compton radiation to the high-energy emission of blazars, and focused on the modeling of photon-energy dependent light curves and time lags between different frequency bands. They applied their results to the FSRQ 3C~279, and concluded that the correlated X-ray/$\\gamma$-ray variability of this quasar was inconsistent with X-rays and $\\gamma$-rays being produced by the same radiation mechanism because otherwise significant systematic time lags between the $\\gamma$-ray and X-ray flaring behaviour would be expected, contrary to the observations (e.g., \\cite{hartman01}).  In the present paper, we describe a newly developed combined SSC + ERC jet radiation transfer code, accounting for time-dependent particle acceleration and injection, radiative cooling, and escape, coupled to the self-consistent treatment of the relevant photon emission, absorption, and escape processes. In \\S \\ref{model} we give a brief description of the underlying blazar jet model. The numerical procedure used in our code will be outlined in \\S \\ref{numerics}. We present results of a detailed parameter study, relevant for application to intermediate and low-frequency peaked BL~Lac objects, in \\S \\ref{results}. We summarize in \\S \\ref{summary}. ",
        "conclusions": ""
    },
    "0208/astro-ph0208524_arXiv.txt": {
        "abstract": "{I discuss the  role of angular momentum in  the formation of disk  galaxies, and  describe  the  results of  two  studies aimed  at testing the standard paradigm for disk formation.}    \\addkeyword{dark      matter}     \\addkeyword{galaxies:     formation} \\addkeyword{galaxies: structure}   \\begin{document} ",
        "introduction": "\\label{sec:intro}  The  current  paradigm for  disk  formation  contains three  important ingredients:  (i) the  angular momentum  originates  from cosmological torques (ii)  the gas and  dark matter within virialized  systems have initial angular  momentum distributions (AMDs) that  are identical and (iii) the  gas conserves its  specific angular momentum  when cooling. Under these  assumptions the predicted scale lengths  of disk galaxies are in excellent agreement with observations (Fall \\& Efstathiou 1980; de Jong \\&  Lacey 2000), which has motivated  the construction of more detailed models,  but always under the three  assumptions listed above (e.g., Mo,  Mao \\& White 1998;  van den Bosch 1998,  2000, 2001, 2002; Firmani \\& Avila-Reese 2000).  Because of  the overall success of  these models in  explaining a wide range of observed  properties of disk galaxies, it  has generally been assumed  that the  aforementioned assumptions  are  correct.  However, several  recent results  have started  to cast  some doubt  as  to the validity  of   this  standard  framework.   First   of  all,  detailed hydro-dynamical simulations  of disk formation  in a cold  dark matter (CDM) Universe  yield disks that are  an order of  magnitude too small (e.g., Steinmetz \\& Navarro 1999).  This problem, known as the angular momentum catastrophe,  is a consequence of  the hierarchical formation of galaxies which causes the baryons to lose a large fraction of their angular momentum to the dark matter.  Secondly, under assumption (iii)  the density distribution of disks is a direct  reflection of  the AMD in  the proto-galaxy.   Bullock \\etal (2001, hereafter  B01) determined the  AMDs of individual  dark matter halos, which according to assumption  (ii) should be identical to that of  the  gas,   and  thus  to  that  of   the  disk.   However,  these distributions seem to have far  too much low angular momentum material for consistency with the  typical exponential density distributions of disk galaxies (B01; van den Bosch 2001). \\begin{figure}[!t] \\includegraphics[width=\\columnwidth]{fvdbosch_fig1.ps} \\caption{A comparison of  the AMDs of 14 disk  galaxies (thin lines) with  those  of CDM  halos  (thick  line)  as parameterized  by  the `universal'  profile  introduced  by  B01.  Whereas  the  latter  is normalized to unity, the former  are normalized to the ratio of disk mass to  expected baryon mass  (i.e., the universal  baryon fraction times the total  virial mass). As is evident,  only a small fraction of the baryons within the halo's  virial radius have ended up in the disk,  and  with  a AMD  that  strongly  differs  from that  of  the `universal' distribution for CDM halos.}  \\label{fig:amd} \\end{figure} ",
        "conclusions": "Since the seminal paper by Fall \\& Efstathiou (1980), a standard model for the formation of disk galaxies has been around that describes disk formation  in  terms  of   the  way  gas  acquires,  and  subsequently conserves, specific angular momentum. Surprisingly enough, very little attention  has been  paid to  testing the  validity of  the underlying assumptions. Via  numerical simulations we  have shown, for  the first time, that in accord with this standard framework, gas and dark matter acquire  identical AMDs.   However, the  fact that  these  AMDs reveal large  mass fractions  with {\\it  negative} specific  angular momentum implies  that the  gas cannot  conserve its  detailed  distribution of specific  angular  momentum  when  cooling  to form  the  disk.   This crumples  one  of  the  main  pillars of  our  standard  picture,  and indicates that a new spin  on the angular momentum acquisition of disk galaxies may be required.  Additional puzzles come  from the fact that the  AMDs of observed disk galaxies are  dramatically different than those of  dark matter halos: disk predominantly  lack low angular  momentum material.  Furthermore, detailed numerical simulations of  disk formation that include cooling find a  large transfer of  angular momentum from  the gas to  the dark matter, resulting in  disks that are an order  of magnitude too small. It is  currently unclear whether these puzzles  indicate a fundamental problem  for the  theory or  merely for  the particular  way  in which feedback  processes  are  implemented  (or  ignored)  in  the  current simulations and/or  models for  disk galaxy formation.   A first-order attempt to  address the  impact of feedback  processes on  the angular momentum  of the  gas in  proto-galaxies  is presented  in Abel  \\etal (2002),  where  it is  shown  that  any  feedback process  capable  of expelling baryons  from dark matter  halos, is likely to  decouple the angular momentum of the gas from that of the baryons."
    },
    "0208/astro-ph0208263_arXiv.txt": {
        "abstract": "Using a self-consistent atmosphere code, we construct a new model of the atmosphere of the transiting extrasolar giant planet HD 209458b to investigate the disparity between the observed strength of the sodium absorption feature at 589 nm and the predictions of previous models.  For the atmospheric temperature-pressure profile we derive, silicate and iron clouds reside at a pressure of several mbar in the planet's atmosphere.  These clouds have significant vertical extent and optical depth due to our slant viewing geometry and lead to increased absorption in bands directly adjacent to the sodium line core.  Using a non-LTE sodium ionization model that includes photoionization by stellar UV flux, collisional processes with H$_2$, and radiative recombination, we show that the ionization depth in the planet's atmosphere reaches $\\sim$1/2 mbar at the day/night terminator. Ionization leads to a slight weakening of the sodium feature.  We present our baseline model, including ionization and clouds, which falls near the observational error bars.  The sensitivity of our conclusions to the derived atmospheric temperature-pressure profile is discussed. ",
        "introduction": "The planet in the system HD 209458 has been an object of intense study since the first observations of its transit \\citep{2000ApJ...529L..41H, 2000ApJ...529L..45C}.  These observations led to a determination of the planet's radius and physical mass, confirming that it must be a gas giant. At wavelengths where the atmospheric opacity is high, the planet blocks more light, yielding a deeper transit light curve that makes the planet appear physically larger.  The location and strength of these absorption features serve as diagnostics of the temperature and chemistry of the planet's atmosphere. \\citet{2000ApJ...537..916S} predicted variations in the transmission spectrum of HD 209458b due to neutral Na and K and to singly-ionized He. \\citet{2001ApJ...553.1006B} explored effects such as ionization and winds and predicted large variations due to H$_{2}$O, CO, and CH$_{4}$ in the infrared. \\citet{2001ApJ...560..413H} explored a variety of physical effects that could be found in transits, including refraction and the angular redistribution of photons due to Rayleigh scattering.  These were found to be minimal for HD 209458b; \\citet{2001ApJ...560..413H} went on to derive the planet radius as a function of wavelength from 300 to 2500 nm.  Confirmation that absorption features due to the opacity of gaseous species could be observed with current technology was obtained by \\citet{2001ApJ...552..699B}.  They observed the transit of HD 209458b from 581 to 638 nm with the STIS instrument on \\textit{HST} and obtained photometric accuracy near 100 micro-magnitudes. This spectral region encompasses the Na-D doublet at 589 nm.  Using these data, \\citet{2002ApJ...568..377C} found that the transit was deeper by $(2.32 \\pm 0.57) \\times 10^{-4}$ in a narrow band centered on the star's Na-D lines than in adjoining bands at shorter and longer wavelengths. Importantly, the observed difference in transit depth in and out of the Na-D line is smaller than previously predicted \\citep{2002ApJ...568..377C}.  The nominal model of \\citet{2001ApJ...560..413H} predicted a difference between these two bands of $4.7 \\times 10^{-4}$.  \\citet{2002ApJ...568..377C}, using the models of \\citet{2001ApJ...553.1006B}, performed a parameter study to investigate a variety of possible reasons for this difference.  These included a bulk underabundance of Na, the sequestering of atomic Na in condensates and/or molecules, clouds very high in the planet's atmosphere, and photoionization of atomic Na due to the UV flux from the parent star. \\citet{2002ApJ...569L..51B} suggested that Na is not in local thermodynamic equilibrium (LTE).  They speculated this leads to a core reversal of the 589-nm absorption doublet in the planet's emission spectrum, but they did not calculate its effect on a transit light curve.  Here we explore the effects of cloud formation and ionization of Na in HD 209458b's atmosphere.  We find that cloud opacity is the dominant effect that determines the depth of the Na-D feature, but ionization of Na also leads to a non-negligible weakening of the Na-D doublet.  In our analysis, following \\citet{2002ApJ...568..377C}, we define a narrow ``in feature\" wavelength band as the wavelength range 588.7-589.9 nm, and an ``out of feature\" band as the combination of the 518.8-588.7 and 589.9-596.8 nm wavelength ranges. The ``in feature\" band includes the wavelength extent of the easy-to-observe HD 209458A (stellar) Na absorption doublet.  When quoting a depth for the Na-D feature, we simply mean the difference in transit depth (the fractional change in flux) between the ``in feature\" and ``out of feature\" bands at midtransit.  The fractional change in flux at midtransit across the entire 581 to 638 nm wavelength range is -0.0164, meaning 1.64\\% of the star's light is blocked. ",
        "conclusions": "We find that HD 209458b's atmosphere is cool enough for most heavy elements to have been incorporated into condensates but still warm enough for Na to remain in atomic form.  As seen in Fig.~\\ref{figure:pt}, there is a range of $\\sim$600 K in the mbar region of the planet's atmosphere where this situation holds.  High clouds arise naturally in the planet's atmosphere near $\\sim$1-5 mbar.  Although the abundances of condensable vapor of the chief cloud-forming species are small, the particles that form in this radiative region are small and have large scattering cross-sections.  Clouds lead to an increase in the apparent planet radius outside of the Na-D core, and on their own they could easily explain the observations.  In addition, clouds can be so high that the predicted sodium feature is $less$ than observed.  With our simple non-LTE model, we have shown that stellar UV flux will ionize Na to $\\sim$1/2 mbar on the limb in the planet's atmosphere, reducing the strength of the 589-nm absorption feature.  The change in the Na-D feature due to ionization is fairly insensitive to the derived \\emph{T-P} profile.  Ionization is a secondary effect, and it alone cannot explain the weak Na-D features.  This conclusion was also reached by \\citet{2002ApJ...568..377C} in their analysis.  Since the majority of Na on the planet's limb at pressures $\\lesssim$1/2 mbar is ionized by stellar UV flux, the possible effects of neutral Na non-LTE level populations found by \\citet{2002ApJ...569L..51B} are reduced.  In addition, since the \\emph{T-P} profile we derive is cooler, a region where Na LTE level populations may not hold will be pushed out to lower pressure.  For our \\emph{T-P} profile, we find an opaque forsterite cloud base at a pressure of several mbar.  Predicting the exact location of the cloud base and vertical distribution of cloud material is challenging, but the Na-D feature is quite senstive to these factors.  An atmosphere incorporating the forsterite cloud at 1.8 mbar and ionization gives a Na-D feature less than observed (due to clouds, not ionization), but for a cloud base of at 5.4 mbar, we obtain a feature depth that matches observations with or without ionization.  More advanced modeling of photochemistry may reveal other non-equilibrium compounds, perhaps including a photochemical haze, which could both increase the importance of particulate absorption and decrease the extent to which the atmosphere is ionized. The potassium feature at 770 nm should be more sensitive to ionization than is Na, because the former's ionization energy is 0.8 eV less than that of Na.  We are currently updating our UV opacity database and collecting all available collsional parameters.  Once this is completed, it will be worthwhile to revisit the sodium ionization problem, taking into account many levels of Na I, and computing consistent non-LTE level populations of sodium together with the radiation field. More advanced global circulation models for \\hd\\ will be of great interest, as they will give a better indication of the temperatures expected on the limb of the planet.  It is likely that observations will push modeling in new directions, and data obtained across many wavelengths, both from the ground and from space, including predicted K, H$_2$O, and CO features, will help gauge the importance of processes at work in HD 209458b's atmosphere and decide between explanations of the Na-D feature depth."
    },
    "0208/astro-ph0208439_arXiv.txt": {
        "abstract": "When a gravitationally lensed source crosses a caustic, a pair of images is created or destroyed. We calculate the mean number of such pairs of micro-images $\\langle n\\rangle$ for a given macro-image of a gravitationally lensed point source, due to microlensing by the stars of the lensing galaxy. This quantity was calculated by Wambsganss, Witt \\& Schneider (1992) for the case of zero external shear, $\\gamma=0$, at the location of the macro-image. Since in realistic lens models a non-zero shear is expected to be induced by the lensing galaxy, we extend this calculation to a general value of $\\gamma$. We find a complex behavior of $\\langle n\\rangle$ as a function of $\\gamma$ and the normalized surface mass density in stars $\\kappa_*$. Specifically, we find that at high magnifications, where the average total magnification of the macro-image is $\\langle\\mu\\rangle=|(1-\\kappa_*)^2-\\gamma^2|^{-1}\\gg 1$, $\\langle n\\rangle$ becomes correspondingly large, and is proportional to $\\langle\\mu\\rangle$. The ratio $\\langle n\\rangle/\\langle\\mu\\rangle$ is largest near the line $\\gamma=1-\\kappa_*$ where the magnification $\\langle\\mu\\rangle$ becomes infinite, and its maximal value is 0.306. We compare our semi-analytic results for $\\langle n\\rangle$ to the results of numerical simulations and find good agreement. We find that the probability distribution for the number of extra micro-image pairs is reasonably described by a Poisson distribution with a mean value of $\\langle n\\rangle$, and that the width of the macro-image magnification distribution tends to be largest for $\\langle n\\rangle\\sim 1$. ",
        "introduction": "\\label{sec:intro}   Gravitational microlensing by the stars of a lensing galaxy can have a large effect on the magnification of lensed sources (Chang \\& Refsdal 1979; Young 1981; Paczy\\'nski 1986). As the macro-images of multiply imaged sources are typically located in relatively dense star fields of the lensing galaxy, microlensing is quite common in such systems. The typical angular separation between the micro-images of a cosmological source due to a stellar mass micro-lens is of order $1$ micro-arc-second, which is far too small to be resolved. For the time being, the only observable manifestation of microlensing is to change the magnification of the macro-image relative to the average magnification that is predicted for the smoothed out surface mass density profile of the galaxy.  The first observational evidence for quasar microlensing was found by Irwin et al. (1989) in the quadruple system Q2237+0305, which subsequently  has been monitored by many groups (Corrigan et al. 1991;  Burud et al. 1998; Lewis et al. 1998, Wo{\\'z}niak et al. 2000a,b). In particular the latest results show that all four quasar images vary dramatically, going up and down by more than one magnitude on timescales of less than a year. The fact that individual (caustic-crossing) events can be clearly distinguished allows to put upper limits on the source  size (Wambsganss, Paczy\\'nski \\& Schneider 1990; Yonehara 1999, 2001; Wyithe et al. 2000).  In the double quasar Q0957+561,  originally there was an almost  linear change detected in the (time-shifted) brightness ratio between the two images ($ \\Delta m_{\\rm AB} \\approx 0.25\\,{\\rm mag}$ over 5 years), which was interpreted as microlensing by solar type stars. But since about 1991, this ratio stayed more or less ``constant\" within about 0.05 mag, so not much microlensing was going on in this system recently (Schild 1996; Pelt et al. 1998). However, even the ``lack of microlensing'' in this system can be used to put limits on compact dark matter in the halo of the lensing galaxy (Wambsganss et al. 2000).  A number of other multiple quasar systems are being monitored more or less regularly, with some showing indications of microlensing, e.g. H1413+117 ({\\O}stensen et al. 1997), B0218+357 (Jackson et al. 2000), or HE1104-1805 (Gil-Merino et al. 2002, Schechter et al. 2002). For a recent review  on quasar microlensing see Wambsganss (2001).  Microlensing has recently been suggested as the source of short time scale low-level variability in the time delayed light curves of multiple images of quasars (Wyithe \\& Loeb 2002), and might help study the properties of broad-line clouds in quasars. A better understanding of the microlensing by stars in galaxies would provide a better handle on the probability distribution for microlensing of cosmological sources (Wyithe \\& Turner 2002) that is relevant for Gamma-Ray Bursts (GRBs), especially in light of a possible microlensing event that was observed in the optical and near IR light curve of the afterglow of GRB 000301C (Garnavich, Loeb \\& Stanek 2000; Gaudi, Granot \\& Loeb 2001; Koopmans \\& Wambsganss 2001)  Another example where microlensing may play an important role is in explaining the flux ratio anomalies observed in close pairs of images of quadruply lensed quasars (Schechter \\& Wambsganss 2002). Such systems are usually modeled using a simple smooth surface mass density profile for the galaxy, possibly with the addition of an external shear. While these models successfully reproduce the observed locations of the macro-images, the flux ratios they predict are quite often in poor agreement with observations. Specifically, the theoretical flux ratio for a close pair of highly magnified macro-images is 1:1, while observations show a difference of up to one magnitude. For example, MG0414+0534 has an observed flux ratio of 2:1 in the optical (Hewitt et al. 1992; Schechter \\& Moore 1993), while in the radio the flux ratio is 1:1 (Trotter et al. 2000). There are also alternative explanations for these flux ratio anomalies, such as intervening dust (Lawrence et al. 1995) or milli-lensing by galactic sub-structure (Mao \\& Schneider 1998; Metcalf \\& Madau 2001; Dalal \\& Kochanek 2002; Chiba 2002). The study of microlensing can help distinguish between these alternative explanations, and may be useful in constraining the sub-structure of galaxies.  In many cases the study of microlensing cannot be done analytically, and much of the work is done using numerical simulations. The magnification distributions of the macro-image (MDMs) are particularly important for understanding the observed properties of lensed systems. The latter have been calculated for a wide range of parameters using numerical simulations (Wambsganss 1992; Rauch et al. 1992; Lewis \\& Irwin 1995, 1996; Schechter \\& Wambsganss 2002). While simulations are applicable to a wide range of problems, and are in many cases the only available technique, they are usually time consuming and do not always provide a qualitative understanding of the results. Specifically, they do not seem to provide an explanation for the detailed structure that is present in the MDMs that they produce.  In a recent paper, Schechter and Wambsganss (2002) explained the flux ratio anomalies as resulting from a different qualitative behavior of the MDMs for macro-minima and macro-saddlepoints in the arrival time surface, that results in a larger probability for de-magnification (relative to the average magnification) of saddlepoints, compared to minima. This is in agreement with the observations of MG0414+0534 and four other recently discovered quadruple systems (Reimers et al. 2002; Inada et al. 2002; Burles et al, in preparation; Schechter et al., in preparation) in all of which the fainter image in a pair corresponds to a saddlepoint, while the brighter image corresponds to a minimum.  The structure of the MDMs seems to be tightly related to the probability distribution for the number of extra micro-image pairs (EIPs) corresponding to a given macro-image (Rauch et al. 1992). This may be expected since the magnification of the macro-image is simply the sum of the magnifications of all the micro-images that it is composed of. An analytic expression for the mean number of EIPs was derived by Wambsganss, Witt \\& Schneider (1992, hereafter WWS) for the case of zero external shear. In \\S \\ref{AR} we generalize this result for arbitrary values of the external shear, and obtain a semi-analytic expression. More detailed expressions are provided in Appendix A, along with an analytic result for the case of equal shear and convergence in stars. In \\S \\ref{num} we compare our analytic results to numerical simulations, and find good agreement. We also show that the probability distribution for the number of extra micro-image pairs is reasonably described by a Poisson distribution. % The possible implications of our results are discussed in \\S \\ref{discussion}. ",
        "conclusions": "\\label{discussion}  We have calculated the mean number of extra micro-image pairs $\\langle n\\rangle$ of a point source, as a function of $\\kappa_*$ and $\\gamma$. The results are shown in Figure \\ref{contour_plots} and Table \\ref{table1}. One dimensional integrals for general values of $(\\kappa_*,\\gamma)$ and an analytic result for the case $\\gamma=\\kappa_*$ are presented in Appendix A. Near the lines of infinite magnification in the $\\gamma-\\kappa_*$ plane ($\\gamma=|1-\\kappa_*|$), $\\langle n\\rangle$ diverges, and is proportional to the mean macro-magnification $\\langle\\mu\\rangle$. The ratio $\\langle n\\rangle/\\langle\\mu\\rangle$ is continuous along these lines and varies smoothly along the line $\\gamma=\\kappa_*-1$, while its derivative is discontinuous along the line $\\gamma=1-\\kappa_*$ in directions that  are not along this line. This creates a ``ridge'' in $\\langle n\\rangle/\\langle\\mu\\rangle$ along the line $\\gamma=1-\\kappa_*$, where it also peaks at $(\\kappa_*,\\gamma)=(0.379,0.621)$ with $(\\langle n\\rangle /\\langle\\mu\\rangle)_{\\rm max}=0.306$. The analytic results for $\\langle n\\rangle$ are in good agreement with the results of numerical simulations we performed for $\\kappa_*=\\gamma=0.333,\\,0.400$ and $0.666$ (as can be seen in Table \\ref{table2}).  We find that the probability distribution $p_n$ for the number of extra image pairs (EIPs) $n$, that is calculated from numerical simulations, may be reasonably described by a Poisson distribution. This result holds both for the numerical simulations performed in this paper (e.g. Table \\ref{table3} and Figures \\ref{n_fig} and \\ref{p_n_fig}), and for numerical simulations from previous works (Rauch et al. 1992). Furthermore, the shape of the magnification distribution $p_n(\\mu)$ of regions with a given $n$ appears to be similar for different values of $n$, where only the overall normalization $p_n$ and mean magnification $\\langle\\mu\\rangle_n$ depend on $n$ (see Figure \\ref{hist}).  The mean number of EIPs $\\langle n\\rangle$ can serve as a rough measure for the width of $p(\\mu)$, the magnification distributions of the macro-image (MDM). For $\\langle n\\rangle\\ll 1$ there is little contribution to the MDM from regions in the source plane with $n>0$, since these regions cover only a small fraction of the source plane. For $\\langle n\\rangle\\gg 1$ the Poisson distribution $p_n$, for $n$, approaches a Gaussian distribution with a mean value of $\\langle n\\rangle$ and a standard deviation of $\\sigma=\\langle n\\rangle^{1/2}$, so that only $\\sim\\langle n\\rangle^{1/2}$ different values of $n$, around $n\\approx\\langle n\\rangle$ will have a noticeable contribution to the MDM. We also note that the average magnification from regions with $n$ EIPs, $\\langle\\mu\\rangle_n$, is approximately linear in $n$ (see Table \\ref{table3}), so that we expect $\\Delta\\mu/\\langle\\mu\\rangle\\sim\\langle n\\rangle^{-1/2} \\ll 1$. Therefore, the width of the MDM is expected to be largest for $\\langle n\\rangle\\sim 1$. This seems to be in rough agreement with the results of numerical simulations (Wambsganss 1992; Irwin \\& Lewis 1995; Schechter \\& Wambsganss 2002)."
    },
    "0208/astro-ph0208113_arXiv.txt": {
        "abstract": "The main requirements for fueling an active galactic nucleus and to form massive black holes are reviewed. Low-luminosity AGN can be fueled easily from the local star clusters, near the nucleus, and the various stellar processes are described. Above a certain luminosity (and therefore accretion rate) large-scale gas flows from galactic scales are required. These can be driven by gravity torques of non-axisymmetric perturbations, such as bars, spirals, galaxy interactions. Observational evidence that these mechanisms are in action is found for high enough luminosities. It is very frequent that starbursts are also triggered through the same mechanisms, and the dense nuclear star clusters formed provide fuel for the AGN over a longer time-scale. Secular internal evolution and more violent evolution through interactions and mergers contribute to grow both a massive black hole and a bulge, and this could explain the observed proportionality relation between the mass of these two components. ",
        "introduction": "From the observed AGN luminosities, and an assumed conversion efficiency to transform the gravitational energy into radiation, the order of magnitudes of the accretion rates can be derived. Luminosities can be typically of the order or higher than 10$^{46}$ erg/s. If we assume a mass-to-energy conversion efficiency $\\epsilon \\sim$ 10\\%  (L = dM/dt c$^2$ $\\epsilon$), then the mass accretion rate dM/dt should be:  dM/dt $\\sim$ 1.7 (0.1/$\\epsilon$)  (L/10$^{46}$ erg/s)  M$_\\odot$/yr  \\noindent If the duty cycle of the AGN is of the order of 10$^8$ yr, then a total mass up to 2 10$^8$ M$_\\odot$ should be available. It is a significant fraction of the gas content of a typical galaxy, like the Milky Way! The time-scale to drive such a large mass to the center is likely to be larger than 1 Gyr.  For the mass to infall into the center, it must lose its angular momentum. Could this be due to viscous torques? In a geometrically thin accretion disk, one can consider the gas subsonic viscosity, where the viscous stress is modelled proportional ($\\alpha$) to the internal pressure, with a factor $\\alpha < 1$. This can only gather in 1 Gyr the gas within 4 $\\alpha$ pc typically (e.g. Shlosman et al 1989, Phinney 1994). This shows that viscous torques will not couple the large-scale galaxy to the nucleus, only the very nuclear regions could play a role through viscous torques.  \\subsection{Stars as the AGN Fuel}   The stars themselves could provide gas to the nucleus, through their mass loss, if there is a local stellar cluster, dense and compact enough (core radius R$_c$ of less than a pc, core mass M$_{core}$ of the order of 10$^{8}$ M$_\\odot$). However, the mass loss rate derived from normal stellar evolution gives only 10$^{-11}$ M$_\\odot$/yr/M$_\\odot$, orders of magnitude below the required rate of a few  M$_\\odot$ /yr. The contribution will be significant, only if a massive stellar cluster (4 10$^9$ M$_\\odot$) has just formed through a starburst (Norman \\& Scoville 1988). A coeval cluster can liberate 10$^9$ M$_\\odot$ on 10$^8$ yr, since mainly massive stars evolve together in the beginning. Thus the existence of a starburst in the first place solves also the problem of the AGN fueling, as in the symbiotic model of Williams et al (1999). The angular momentum problem is now passed on to the starburst fueling, and could be solved only through large-scale dynamical processes.  It is also possible that stars themselves are directly swallowed by the black hole, the various processes that have been studied are: \\begin{itemize} \\item Bloated stars, a phenomenon that makes mass loss more efficient (Edwards 1980, Alexander \\& Netzer 1994, 1997), \\item Tidal disruption of stars (Hills 1975, Frank \\& Rees 1976), \\item Star Collisions (Spitzer \\& Saslaw 1966, Colgate 1967, Courvoisier et al 1996, Rauch 1999). \\end{itemize}    \\begin{figure}[t] \\rotatebox{-90}{\\includegraphics[width=.90\\textwidth]{combesf-f1.ps}} \\caption{ Characteristic radii, corresponding to the various physical phenomena, as a function of black hole mass M$_\\bullet$: from top to bottom, R$_a$, the accretion radius, under which the BH dominates the dynamics; R$_{coll}$, the collision radius, under which stellar collisions are disruptive, i.e. the free-fall velocity around the black hole (GM$_\\bullet$/r)$^{1/2}$ is equal to the escape speed v$_*$ of a typical individual star (GM$_*$/r$_*$)$^{1/2}$; R$_E$, the Eddington radius, under which a star receives more light than its Eddington luminosity; R$_T$, the tidal radius, under which a star is disrupted by the tidal forces of the BH; R$_g$, the gravitational radius. Loss-cone effects are important inside the critical radius r$_{crit}$.} \\label{char} \\end{figure}  The characteristic radii associated to the prevalence of these processes are displayed in figure \\ref{char}. As the black hole horizon grows faster with  M$_\\bullet$ than the tidal radius, there is a limit, when M$\\sim$ 3 10$^8$ M$_\\odot$, above which the star disruption occurs inside the black hole, and there is no gaseous release or AGN activity (but the black hole might grow even more rapidly). This is the Hills limit (Hills 1975).    \\begin{figure} \\rotatebox{-90}{\\includegraphics[width=.45\\textwidth]{combesf-f2a.ps}} \\rotatebox{-90}{\\includegraphics[width=.45\\textwidth]{combesf-f2b.ps}} \\caption{ {\\bf (left)} A schematic view of the activity phase of a typical quasar: during the growth phase, the AGN does not radiate efficiently, although the fueling rate is larger than Eddington. After its luminous phase of  $\\sim$ 4 10$^7$ yr, the fuel is exhausted and the AGN fades away. {\\bf (right)} Growth of a supermassive black hole in two simple models: accretion at Eddington luminosity (time-scale t$_E$ and corresponding luminosity L$_E$, as a function of black hole mass M$_{BH}$), and when accretion is limited by diffusion (t$_D$ and L$_D$) (from Hills, 1975). } \\label{hills} \\end{figure}   \\subsection{Growth of Black Holes}  There is now a consensus to recognize that AGN derive their power from supermassive black holes, but their formation history, their demography, their activity time-scales are still debated. The two extreme hypotheses have been explored: either only a small percentage of galaxies become quasars, and they are continuously fueled, and active over Gyrs, or a massive black hole exists in almost every galaxy, but they have active periods of only a few 10$^7$ yr.  In the first hypothesis, there should exist black holes with masses 100 times higher than the maximum observed today, and accretion rates much lower than the Eddington rate, and this is not supported (e.g. Cavaliere \\& Padovani 1989). Models with a duty cycle of 4 10$^7$ yr are favored, and many galaxies today should host a starving black hole (Haehnelt \\& Rees 1993). Figure \\ref{hills}a schematizes the period of activity and growth of a typical massive black hole.  It is also possible that further gas accretion (during a galaxy interaction for instance) triggers a new activity phase for the black hole.  If we assume that the availability of the fuel around the black hole is not a problem, then the black hole can at maximum accrete mass at the Eddington rate, and radiate at Eddington luminosity (above which the radiation pressure prevents the material to fall in). A typical growth rate for the black hole is then given by the time required to reach the critical mass M$_c$ where R$_T$ = R$_g$, above which stars are swallowed by the black hole without any gas radiation (M$_c$ = 3 10$^8$ M$_\\odot$). The Eddington luminosity is: L$_E$ = 3.2 10$^4$ (M/M$_\\odot$) L$_\\odot$. For a mass M$_c$, the maximum is 10$^{13}$ L$_\\odot$ (close to the peak luminosity of QSOs). Then the corresponding accretion rate, assuming an efficiency of $\\epsilon$ =10-20\\%  is dM/dt$_E$ = 1.1 10$^{-8}$ (M/M$_\\odot$) M$_\\odot$/yr. The growth rate of the black hole in this regime t$_E$ is then exponential; it takes only 1.6 10$^9$ yr to grow from a stellar black hole of 10 M$_\\odot$ to M$_c$:  t$_E$= 9.3 10$^7$ ln(M$_c$/M) yr  \\noindent Note that this very simple scheme would lead to  a maximum at z=2.8 of the number of quasars. This maximum rate, however, is not realistic, since the black hole quickly gets short of fuel, as the neighbouring stars  (in particular at low angular momentum) are depleted.  Then it is necessary to consider a growth limited by stellar density $\\rho_s$:  DM/dt = $\\rho_s \\sigma$ V  \\noindent where $\\sigma$ is the accretion cross-section, and V the typical stellar velocity. The corresponding time-scale to grow from M to M$_c$ is  \\noindent t$_D$ = 1.7 10$^{15}$ yr ($\\rho_s$/M$_\\odot$pc$^{-3}$)$^{-1}$ M/M$_\\odot^{-1/3}$ (1 - M/M$_c^{1/3}$) $<$V$^2>^{1/2}$ (km/s) \\noindent Typically in galaxy nuclei, $\\rho_s$ = 10$^7$ M$_\\odot$/pc$^3$,  $<$V$^2>^{1/2}$  = 225 km/s. A black hole could grow up to M$_c$ in a Hubble time, and the luminosity at the end could be of the order of 10$^{46}$ erg/s (see figure \\ref{hills}b).  In more details, the first stars to disappear, being swallowed by the black hole, are those with the least angular momentum, with a small pericenter. The only mechanism to replenish the stellar density near the BH is the two-body relaxation, with a time-scale $t_R$. The relevant two-body relaxation time t$_R$ is dependent on the number of bodies N in the system, as  t$_R$/t$_c$ = N/logN, where  t$_c$ is the crossing time= r$_c$/V. For a galactic center, with a volumic density of stars of 10$^7$M$_\\odot$/pc$^3$, this relaxation time is 3 10$^8$ yr.  Since the angular momentum diffuse faster than the energy, the low anglular momentum stars will re-appear faster, this is the loss-cone effect, that increases the accretion rate by a factor $(1-e^2)^{-1}$, where $e$ is the exentricity of the orbits. This is significant inside a critical radius r$_{crit}$, where the loss-cone angle becomes larger than the diffusion angle $\\theta_D \\sim (t_{dyn}/t_R)^{1/2}$. This critical radius is also plotted in figure \\ref{char}.  \\subsection{Formation of a Cusp of Stars around the BH} From a numerical resolution of the time-dependent Boltzmann equation, with the relevant diffusion coefficients, it can be shown that around a black hole at the center of a globular cluster, the stellar density should be of a power-law shape, with a slope of 7/4, cf fig \\ref{duncan}a (Bahcall \\& Wolf 1976).  The distribution of stars around a black hole can be described, according to the distance to the center: \\begin{itemize} \\item first for the stars not bound to the black hole, at r $>$ R$_a$, their velocity distribution is Maxwellian, and their density profile has the isothermal power law in r$^{-2}$. There are also unbound stars inside R$_a$, but with a density in  r$^{-1/2}$. This allows to compute the penetration rate of these unbound stars in the tidal or collision radius, to estimate the accretion rate. With a core stellar mass of M$_{core}$ = 10$^7$- 3 10$^8$ M$_\\odot$, a density 10$^7$ pc$^{-3}$, the accretion rate is, by tidal disruption:  dM/dt$_{tide}$ = 1 M$_\\odot$/yr M$_8^{4/3}$  \\noindent and by stellar collisions:  dM/dt$_{coll}$ = 0.1 M$_\\odot$/yr M$_8^3$  \\item the orbits bound to the black hole  r$<$ R$_a$: due to the cusp, their density is in  r$^{-7/4}$, there is an excess of stars inside R$_{coll}$, that favors stellar collisions. \\end{itemize}  However, detailed numerical show that the stellar cluster cannot fuel the black hole indefinitely (Duncan \\& Shapiro 1983). The growth rate of the black hole and its luminosity decreases as 1/time (cf fig \\ref{duncan}b). The loss-cone theory and the simulations are in agreement: the accretion rate due to tidal disruptions  is M$_{core}$/t$_R$, typically of 10$^{-2}$ M$_\\odot$/yr, with a maximum lower than 1 M$_\\odot$/yr; this cannot explain the luminosity of QSOs. QSOs might be explained only when stellar collisions are included, the corresponding accretion rate is typically a hundred times higher.  \\begin{figure} \\rotatebox{0}{\\includegraphics[width=.40\\textwidth]{combesf-f3a.ps}} \\rotatebox{-0}{\\includegraphics[width=.60\\textwidth]{combesf-f3b.ps}} \\caption{ {\\bf (right)} A schematic representation of the radial profile of stellar density in a spheroid with a central massive black hole. (I) the cusp, with a slope of r$^{-7/4}$, (II) an isothermal core (in r$^{-2}$), (III) a halo (in r$^{-7/2}$), and (IV) an isothermal tail (in r$^{-2}$). {\\bf (left)} the black hole growth rate, as a function of time, according to a model where the number od stars in the core is initially 2.7 10$^8$, and the velocity dispersion in the center 2.9 x 350 = 1015 km/s, parameters fitted to a ``quasar'' environment; the dash-curve is the initial rate predicted analytically (from Duncan \\& Shapiro, 1983).} \\label{duncan} \\end{figure}    Triaxial deviations from spherical symmetry of only 5\\%  (due to a bar or binary black hole) can repopulate the loss-cone, increasing tidal disruption to QSOs levels. However, t$_{coll} <$ t$_R$, and collisions may destroy the cusp (Norman \\& Silk 1983).  Stellar collisions help to refill the loss-cone, although they flatten the stellar cusp (Rauch 1999). The collisions rate is comparable to the diffusion rate, that refill the central core.  In summary, for low density nuclei, stellar evolution and tidal disruption is the main mechanism to bring matter to the black hole, and for high density nuclei, stellar collisions dominate the gas fueling. The evolution of the stellar density through these processes is then opposite, and accentuates the differences:  -- for n $<$ 10$^7$/pc$^3$ the core then expands, due to heating that results from the settling of a small population of stars into orbits tightly bound to the black hole;  -- for  n$>$ 10$^7$/pc$^3$, the core shrinks due to the removal of kinetic energy by collisions. To give an order of magnitude, the nuclear density in our own Galaxy is estimated at 10$^8$ M$_\\odot$/pc$^3$ (Eckart et al 1993).  These mechanisms produce differing power-law slopes in the resulting stellar density cusp surrounding the black hole, -7/4 and -1/2 for low- and high-density nuclei, respectively (Murphy et al 1991, Rauch 1999). In simulations however (Rauch 1999), collisions tend to produce a flat core, instead of r$^{-1/2}$ law in Fokker-PLank studies which imply isotropy (and are unable to treat sparse regions).  The cusps observed in nearby galaxies are consistent with the hypothesis of adiabatical black hole growth in homogeneous isothermal core (e.g. Young 1980) and with initial conditions for the cores following the scaling relations of the fundamental plane (van der Marel 1999).  One must also take into account that the merging of black holes, and the dynamical effects of binary black-holes, flatten the cusp slopes (Nakano \\& Makino, 1999; Milosavljevic \\& Merritt, 2001).    \\subsection{Conclusion for AGN Fueling by Stars}  Stars are sufficient to fuel some low luminosity AGN: there is first mass loss from a nuclear cluster, which provides only low accretion rates, then tidal disruption of stars themselves, which depletes the center, replenished through two-body relaxation. At high densities stellar collisions also replenish the central density, and the AGN can reach higher luminosities. According to the central density, the power-law slopes of the density cusp in the center are different (flatter for higher densities). However, the accretion rates reached through fueling by stars only are not sufficient to account for the most luminous AGN and quasars. Large-scale processes to transfer angular momentum of the interstellar gas over a large disk radius are then required. ",
        "conclusions": "The rapid decline of AGN from z=2 to z=0 is parallel to that of the star formation history, and is likely to have the same causes.  Since black hole masses have accumulated in the center of each galaxy at the present time, the main reason for the lower activity is the shortage of fuel. AGN are accreting today at a rate much lower than Eddington rate. The dominant active nuclei are presently low-luminosity AGN (Seyferts), revived by gas accretion from galaxy interactions, much less frequent than in the past. Also the galaxies with higher gas fraction are late-type galaxies, that have small-mass black holes in their nuclei.  Spheroids and massive black holes appear to be formed with the same mechanisms. Mass concentration is enhanced by radial gas flows, driven from the galactic disks, through density waves, such as bars and spirals. Internal gravity torques and non-axisymmetries are also boosted by galaxy interactions and mergers.  Gas fueling towards the center can trigger nuclear starbursts, forming dense compact nuclear stellar clusters, that in turn will fuel a massive black hole.  The detailed processes regulating the successive/alternate fueling of starbursts and BH are not yet completely elucidated, and might involve the stability of galactic disks, the depth of the central potential well.  The required fueling depends on the strength and luminosity of the AGN. For Seyferts, only stars from a dense nuclear cluster are sufficient, through tidal disruptions and stellar collisions. For quasars, big starbursts are required, and the coeval compact cluster just formed can provide the fuel through mass loss of young stars and supernovae.  The relations between BH and bulge masses, or with the central velocity dispersion, are then naturally explained by the complicity between starbursts and AGN."
    },
    "0208/astro-ph0208325_arXiv.txt": {
        "abstract": "We discuss the real-space moments of temperature anisotropies in the cosmic microwave background (CMB) due to weak gravitational lensing by intervening large-scale structure.  We show that if the probability distribution function of primordial temperature anisotropies is Gaussian, then it remains unchanged after gravitational lensing. With finite resolution, however, non-zero higher-order cumulants are generated both by lensing autocorrelations and by cross-correlations between the lensing potential and secondary anisotropies in the CMB such as the Sunayev-Zel'dovich (SZ) effect.  Skewness is produced by these lensing-SZ correlations, while kurtosis receives contributions from both lensing alone and lensing-SZ correlations.  We show that if the projected lensing potential is Gaussian, all cumulants of higher-order than the kurtosis vanish. While recent results raise the possibility of detection of the skewness in upcoming data, the kurtosis will likely remain undetected. ",
        "introduction": "Weak gravitational lensing deflects the paths of cosmic microwave background (CMB) photons propagating from the surface of last scattering. One result of this lensing is the transfer of power from large angular scales associated with acoustic-peak structures to small angular scales in the damping tail of the anisotropy power spectrum \\cite{Blaetal87,Hu00}. This transfer only results in a few-percent modification of the power associated with the acoustic-peak structure, and the increase in power along the damping tail is significantly smaller than that generated by secondary anisotropies due to reionization \\cite{Coo02}. To indentify the effect of gravitational lensing on CMB data, it is necessary to consider signatures beyond that in the angular power spectrum of temperature fluctuations. The existence of non-vanishing higher order cumulants is one such non-Gaussian signature lensing can generate.  Since gravitational lensing conserves surface brightness, CMB fluctuations from lensing are at the second order in temperature fluctuations and result in non-Gaussian behavior through non-linear mode coupling. Though lensing alone does not lead to a three-point correlation function, the correlation between lensing and other secondary anisotropies can lead to such a contribution. This three-point correlation has been widely discussed in the literature in terms of its Fourier-space analogue, the bispectrum \\cite{SpeGol99}.  Weak lensing of the primary anisotropies can produce a four-point correlation due to its non-linear mode-coupling nature \\cite{Bern97,Zal00,Coo02a}, as can correlations between lensing and secondary effects \\cite{Coo02a}. When probed appropriately through quadratic statistics such as the power spectrum of the squared-temperature map, the trispectrum due to lensing alone can be used for a model-independent recovery of the projected mass distribution out to the last scattering surface \\cite{Hu01b,CooKes02}. Though these statistics have been shown to be interesting and potentially detectable, measurement of these Fourier-based statistics is challenging and techniques are still underdeveloped for this purpose.  Here, we discuss real-space moments of the lensed CMB temperature anisotropies. Real-space statistics are easily measurable from data. The only drawbacks are that they are unlikely to be optimal and only provide limited knowledge of the full non-Gaussian aspect of the temperature distribution. The first attempts to measure non-Gaussianity in the COBE data relied on real-space cumulants \\cite{Hinetal95}, as will attempts using data from its successor experiments such as MAP and Planck. This motivates our emphasis here on the real-space cumulants such as the skewness and kurtosis; we make several remarks on higher-order cumulants as well.  As part of this calculation, we extend a previous discussion of the kurtosis due to lensing in Ref. \\cite{Bern97} and also consider effects related to correlations between lensing and secondary effects such as the Sunyave-Zel'dovich (SZ; \\cite{SunZel80}) effect.  Real-space moments can be derived from the one-point probability distribution function (PDF) of temperature fluctuations, and can conversely be used to constrain the form of this function.  In the case of infinite angular resolution, we conclude that lensing does not modify the PDF of temperature anisotropies produced at the last scattering surface, which is a reflection on the fact that lensing does not create new power but rather transfers power from large to small angular scales. The higher-order moments are only generated in a temperature map by finite-resolution effects such as beam smoothing introduced either experimentally or artificially by explicit filtering.  The paper is organized as follows. In \\S~\\ref{sec:lensing}, we introduce formalism concerning the weak-lensing approximation and define the bispectrum, trispectrum, and corresponding higher-order quantities.  The bispectrum and trispetrum induced in the CMB by lensing and secondary anisotropies are derived in \\S~\\ref{sec:PBT}, and some remarks are made concerning higher-order cumulants as well.  The nonzero bispectrum and trispectrum yield a skewness and kurtosis respectively in the one-point distribution function of the CMB as shown in \\S~\\ref{sec:S&K}. We refer the reader to Ref. \\cite{CooKes02} for additional details related to the effect of lensing on CMB anisotropies. Though we present a general discussion, we illustrate our results in \\S~\\ref{sec:results} using the currently favored $\\Lambda$CDM cosmological model with $\\Omega_b=0.05$, $\\Omega_m=0.35$, $\\Omega_\\Lambda=0.65$, $h$=0.65, and $\\sigma_8 = 0.9$. Results for a model with $\\sigma_8 = 1.2$ as suggested by CBI are also considered. ",
        "conclusions": "\\label{sec:results}  \\subsection{Skewness}  We illustrate in Fig.~(\\ref{fig:skew}) our results for skewness due to the correlation between lensing and the SZ effect.  We calculate this correlation following Ref.~\\cite{Coo01} using the halo approach to large-scale structure \\cite{CooShe02}. The skewness approaches zero at small values of the smoothing scale, consistent with our conclusion  that no non-Gaussian signatures exist in the PDF in the limit of infinite resolution. As shown, skewness due to the lensing-SZ correlation peaks at an angular scale of tens of arcminutes, which is in the range of interest to upcoming experiments such as MAP and Planck.  When calculating expected signal-to-noise ratios for these experiments, we use detector sensitivities and resolutions tabulated in Ref. \\cite{Cooetal00}. For simplicity, we combine information from individual frequency channels to form one estimate of temperature with an overall noise given by inversely weighting individual noise contributions.  The skewness as shown has signal-to-noise ratios slightly less than unity suggesting that its detection may be hard and potentially affected by noise.  However, recent small-scale excess-power detections by experiments such as CBI \\cite{Masetal02} raise the possibility that we may have underestimated the lensing-SZ correlation and thus the skewness.  The lensing-SZ power spectrum $C_l^{\\len\\s}$ is roughly proportional to the fifth power of $\\sigma_8$, the standard deviation of linear mass fluctuations within an $8 h^{-1}$ Mpc sphere.  If we adopt the CBI $1\\sigma$ upper bound of $\\sigma_8 \\leq  1.2$ \\cite{Masetal02} as opposed to the value $\\sigma_8 = 0.9$ suggested by previous studies, our signal increases by a factor of 4.21. In this case, Planck could conceivably detect skewness with a signal-to-noise of 2.5. The potential for detection of the temperature skewness is consistent with previous expectations that the temperature anisotropy bispectrum due to lensing-SZ correlation can be detected in future data \\cite{SpeGol99}. The cumulative signal-to-noise for skewness, however, is significantly smaller than that for the full bispectrum because the skewness is a single number while the bispectrum contains all information related to non-Gaussianities at the three-point level. As described below, we find a similar reduction in signal-to noise for kurtosis when compared to the full trispectrum.  The frequency dependence of the SZ effect allows us to construct an SZ map of the sky as well as a temperature map with the SZ effect removed.  This provides us a unique opportunity to test our understanding of non-Gaussianity at the three-point level.  If skewness is purely a consequence of lensing-SZ correlations as posited in this paper, then the skewness obtained by combining one measurement of the SZ map with two measurements of the SZ-cleaned temperature map at the same location using the estimator in Eq.~(\\ref{E:est}) should be precisely one third that produced by three measurements of the total anisotropy map.  This corresponds to the fact that the composite map will sample only one of the three permutations appearing in Eq.~(\\ref{E:SZbi}).  \\begin{figure}[t] \\centerline{ \\psfig{file=fig2a.ps,width=3.5in,angle=0} \\psfig{file=fig2b.ps,width=3.5in,angle=0} } \\caption{{\\it Left:} The kurtosis $K^\\pp$ due to lensing autocorrelations and $K^{\\len\\s}$ due to lensing-SZ cross-correlations for a perfect (no-noise) experiment (solid line) and Planck (dashed line). The kurtosis due to lensing-SZ correlations is negative at smoothing scales below the kink at $\\sim$ 8 arcminutes and positive thereafter; its absolute value is shown here. {\\it Right:} The signal-to-noise ratio for the detection of kurtosis in CMB data with curves labeled as in the left figure.  We assume full sky-coverage; for partial sky coverage the signal-to-noise ratio scales as $\\sqrt{f_{\\rm sky}}$, where $f_{\\rm sky}$ is the fraction of sky covered.} \\label{fig:kurt} \\end{figure}  \\subsection{Kurtosis}  Both lensing kurtosis $K^\\pp$ and the kurtosis $K^{\\len\\s}$ due to lensing-SZ correlations are undetectable even for a perfect no-noise experiment as illustrated in Fig.~(\\ref{fig:kurt}).  Since the cumulative signal-to-noise ratio for $K^{\\len\\s}$ is well below one, we expect it to remain undetectable despite any uncertainty in our calculation of the SZ effect. Note our prediction of the lensing kurtosis $K^\\pp$ is likely to be more certain since it only depends on the matter power spectrum, with contributions coming mainly from the linear regime. Thus, uncertainties in non-linear aspects of clustering are unlikely to affect our conclusion.  The signal-to-noise value for $K^\\pp$ can be compared to the cumulative signal-to-noise ratio for the direct detection of the full trispectrum due to lensing, which in the case of Planck can be as high as $\\sim$ 55 \\cite{Zal00}. Consequently, although the lensing kurtosis cannot be detected directly from the data, lensing effects associated with this kurtosis can be used to reconstruct the lensing deflection angle as described in Refs.~\\cite{Hu01b,CooKes02}, again with cumulative signal-to-noise ratios significantly greater than that for the kurtosis itself. The higher signal-to-noise ratio in lensing reconstruction is possible for two reasons.  Unlike the kurtosis, which averages indiscriminately over all configurations of the trispectrum as shown in Eq.~(\\ref{E:SBKT}), lensing reconstruction is sensitive to certain configurations of the trispectrum, mainly those that contribute to the power spectrum of squared temperature. This avoids severe positive-negative cancellations that significantly reduce the signature of non-Gaussianity. Secondly, the noise contribution associated with lensing reconstruction is also {\\it a priori} reduced through a filter which is designed to extract information on the lensing potentials optimally.  The low signal-to-noise associated with the kurtosis is also consistent with the fact that real-space moments, in general, suffer from excess noise. Though such statistics are easily measurable in data, they do not provide the most optimal methods to search for the existence of non-Gaussian signatures. While we recommend construction of cumulants such as skewness and kurtosis as a first step in understanding non-Gaussianity from effects such as lensing, we suggest that full measures of quantities such as bispectrum and trispectrum will  be necessary to fully understand the non-Gaussian behavior of lensing. If measurement of such statistics are still cumbersome, we suggest the use of quadratic statistics in real space, such as the squared-temperature--temperature \\cite{Coo02a} and the squared-temperature--squared-temperature \\cite{CooKes02} power spectra which probe certain configurations of the bispectrum and trispectrum."
    },
    "0208/astro-ph0208443_arXiv.txt": {
        "abstract": "Motivated by the prospect of testing inflation from precision cosmic microwave background observations, we present analytic results for scalar and tensor perturbations in single-field inflation models based on the application of uniform approximations. This technique is systematically improvable, possesses controlled error bounds, and does not rely on assuming the slow-roll parameters to be constant. We provide closed-form expressions for the power spectra and the corresponding scalar and tensor spectral indices. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0208/astro-ph0208169_arXiv.txt": {
        "abstract": "Using high-dispersion spectra from the HIRES echelle spectrograph on the Keck I telescope, we measure velocity dispersions for 4 globular clusters in M33. Combining the velocity dispersions with integrated photometry and structural parameters derived from King--Michie model fits to WFPC2 images, we obtain mass-to-light ratios for the clusters. The mean value is $\\mlv = 1.53\\pm0.18$, very similar to the \\mlv\\ of Milky Way and M31 globular clusters. The M33 clusters also fit very well onto the fundamental plane and binding energy -- luminosity relations derived for Milky Way GCs. Dynamically and structurally, the four M33 clusters studied here appear virtually identical to Milky Way and M31 GCs. ",
        "introduction": "In the Local Group, by far the best studied globular cluster (GC) systems are those of the Milky Way and M31. Significant amounts of data have been gathered for globular clusters in these galaxies, with extensive compilations in \\citet{har96} (Milky Way, 147 clusters) and \\citet{barm00} (M31, 435 clusters).  Both of these large spirals are known to contain two distinct GC subpopulations: a metal-poor halo population and a more metal-rich population, associated with the bulge or thick disk \\citep{kin59,zin85,min95,jab98,per02}. The presence of two GC subpopulations presumably indicates different formation processes, and may hold clues to the formation and early evolution of the host galaxies. A particularly controversial issue is whether or not age differences exist between globular clusters of different metallicities.  In M31, \\citet{bh00} found evidence for age differences of 4--8 Gyrs between the metal-rich and metal-poor clusters and \\citet{sara97} concluded that a similar age spread exists among GCs in the Milky Way. Other recent studies have proven less conclusive, but age differences of a few Gyrs cannot be ruled out \\citep{svb96,buo98,ros99}.  The GC system of the third spiral galaxy in the Local Group, M33, is more sparse and consequently more difficult to identify and study.  Early work on the star clusters in this galaxy was done by \\citet{hil60} and \\citet{km60}, who noted that some of them had unusually blue colors. Photometry for about 130 cluster candidates was given by \\citet{cs82,cs88}, who tentatively identified 27 old GC candidates in M33. They also noted that M33 appears to contain a number of ``blue populous'' clusters similar to those in the Large Magellanic Cloud \\citep{hod61}, but without the pronounced age gap that exists between $\\sim3$ Gyrs and 12--15 Gyrs in the LMC \\citep{gir95}. The old GCs exhibit halo-like kinematics, while the younger clusters follow the rotation of the disk \\citep{schom91}. Color-magnitude diagrams for 10 halo GCs in M33, obtained with WFPC2 on board HST, were presented by \\citet{sara98}, who noted that most of the clusters showed no blue horizontal branches in spite of having low metallicities. They suggested that this might be due to a second-parameter effect, indicating that many of the ``old'' M33 clusters may be several Gyrs younger than their Galactic counterparts.  Additional clusters have been identified on WFPC2 images by \\citet{chan01}, and again \\citet{chan02} suggested a larger age spread among the halo GCs compared to the Milky Way.   It is important to establish how much the old globular clusters in different galaxies have in common. A larger age spread among halo GCs in M33, for example, might indicate that this galaxy assembled over a longer period of time, while the structure of individual clusters may provide information about the gas clouds out of which they formed \\citep{ml92,ee97}. In the Milky Way, a number of fundamental correlations are known to exist between parameters such as surface brightness, core radius and velocity dispersion of globular clusters \\citep{korm85,djorg95}. \\citet{mac00} found that all Milky Way GCs are consistent with a constant $V$--band core mass-to-light ratio (within the error margins) of 1.45 $\\msun \\, \\lsun^{-1}$. He also found a tight relation between the binding energy of individual clusters and the total luminosity, with a weak decrease in binding energy as a function of Galactocentric distance. The dependence on Galactocentric distance may hold information about pressure gradients in the early Galactic halo.  Detailed information about the structure of extragalactic GCs, even within the Local Group, is more limited for obvious reasons. At the distance of M31 and M33, a typical ground-based resolution of $\\sim1\\arcsec$ corresponds to a linear scale of about 4 pc, roughly similar to the typical half-light radii of globular clusters. Thus, information about the detailed structure of individual clusters is not easily obtained. However, with the HST the situation is greatly improved, with one pixel on the Planetary Camera (PC) corresponding to about 0.2 pc. This is of great importance not only for studies aiming at a morphological characterization of the clusters, but also for dynamical studies in which velocity dispersions must be tied to knowledge about the cluster structure in order to provide accurate mass-to-light ratios. So far, the mass-to-light ratios and other properties of globular clusters in M31 appear to be identical to those observed in Milky Way GCs \\citep{djorg97,dg97,barm02}. Similarity in the M/L ratios (modulo metallicity effects) is, of course, to be expected if the clusters have similar ages, unless there are variations in the stellar mass function. No information about dynamical masses has yet been published for M33 GCs, but could potentially be used as a tool to independently check whether age differences exist.  In this paper we study 4 globular clusters in M33, selected from the sample of \\citet{sara98}. The clusters were originally selected based on their halo-like kinematics and red colors ($\\bv > 0.6$), and should be as close an analogy to the halo clusters in the Milky Way as possible. We use new high-dispersion spectroscopy from the HIRES spectrograph on the Keck I telescope to measure velocity dispersions for the individual clusters, and combine these with structural information from the same HST images used by \\citet{sara98} to derive mass-to-light ratios and compare with data for Milky Way globular clusters. Throughout the paper we will adopt a distance modulus of 24.84 and the reddenings for each cluster determined by \\citet{sara00}. ",
        "conclusions": "We have measured velocity dispersions for 4 old ``halo'' globular clusters in M33 and combined these with structural parameters from WFPC2 images to obtain mass-to-light ratios for the clusters. Our analysis shows that the M33 GCs have \\mlv\\ ratios essentially identical to those of GCs in the Milky Way and M31, as well as similar structural parameters (core- and effective radii). The M33 clusters also fall on the same ``fundamental plane'' and binding energy vs.\\ luminosity relations as Milky Way and M31 clusters \\citep{djorg95,dg97,mac00}. Unless the stellar IMFs in these clusters have an excess of low-mass stars relative to GCs in the Milky Way, the similarity in M/L ratios and the fundamental plane relations suggest similar high ages.  It would be very interesting to measure velocity dispersions for M33 clusters spanning a wide range of ages and look for the expected variations in M/L ratio."
    },
    "0208/astro-ph0208396_arXiv.txt": {
        "abstract": "The atmosphere is an intrinsic part of any ground based Cherenkov $\\gamma$-ray telescope, and the telescope response is therefore sensitive to unpredictable changes in the atmospheric transparency which are difficult to measure and interpret in the absence of a calibrated beam of high energy $\\gamma$-rays. In this paper, we use the detector response to Cherenkov emission from cosmic ray initiated air showers to obtain a relative calibration for data obtained under different instrumental and atmospheric conditions as well as over a range of source angles to the Zenith. We show that such a relative calibration is useful and efficient for data selection, for correcting the measured $\\gamma$-ray rate and for inter-calibration between the elements of an array of Cherenkov telescopes.  \\vspace{1pc} ",
        "introduction": "Atmospheric Cherenkov detectors cannot be calibrated using a test beam and the estimation of their sensitivity strongly depends on Monte Carlo simulation programs in which are modeled the atmosphere and the various elements of the detectors. Simulations usually assume a set of fixed conditions while the overall efficiency of the experiment can vary in time due to a number of factors. The most important cause of these variations is the atmosphere itself, and measuring and modeling changes in the $\\gamma$-ray detection efficiency due to changing atmospheric conditions is complex. The slow degradation of optical elements until their recoating or replacement or the occasional readjustment of photo-detector gains also affects the sensitivity of the experiment. When measuring the $\\gamma$-ray flux from a source, one must correct for these effects.  In this paper we present a method used for the Whipple Atmospheric Cherenkov Imaging Telescope to estimate an overall relative efficiency factor. We also validate the method using observations of the Crab Nebula and present some applications. In its basic form the method is based on the analysis of data taken toward the Zenith \\cite{Mohanty} and this is presented first. We have realized that the method can be generalized in a way which incorporates the effects of the Zenith angle at which observations are made. While detailed simulations will always be necessary in order to understand variations in telescope sensitivity, a simple correction such as that presented here is a useful tool which may be particularly important when studying the time variability of $\\gamma$-ray sources. An example of this is the case of flaring active galactic nuclei (AGN) which may be observed over a long period of time and a wide range of Zenith angles and atmospheric conditions. We also present the application of a similar method to CELESTE, a Cherenkov wavefront sampling experiment, which illustrates the utility of the technique as a way of obtaining a relative calibration between individual elements of a detector array. ",
        "conclusions": "We have shown that cosmic ray background events observed at fixed Zenith angle can be used to establish a relative calibration for a single atmospheric Cherenkov imaging telescope in order to account for the many unavoidable temporal changes in light collection efficiency, gain and, most importantly, atmospheric conditions. Generalizing the method, we have shown that it can be used for the relative calibration of data obtained at different Zenith angles, taking into account both the geometrical effects due to Zenith angle and the variations in atmospheric conditions.  This calibration method can be used to introduce corrections at various levels. At the most basic level, it can be used to select which data were taken under good conditions. We have also shown that it can be used to rescale the measured $\\gamma$-ray fluxes in order to make observations taken under different conditions more comparable. One method of estimating the background due to cosmic rays for $\\gamma$-ray observations taken without dedicated background control observations is to choose archival background observations taken under conditions as similar as possible to the source observation being considered. The throughput factor can be used as one of the criteria to judge which background runs are most suitable \\citep{Horan02}.  The application of the throughput calibration to CELESTE observations shows how useful it may be when considering telescope arrays. The next generation of Cherenkov imaging telescopes are currently being developed; the VERITAS \\citep{VERITAS}, HESS \\citep{HESS} and CANGAROO III \\citep{CANGAROO} projects all involve using multiple telescopes on the same site. Inter-calibration of these telescopes will be difficult without a dedicated test beam. For VERITAS, simulations indicate that an energy resolution of $15\\%$ should be possible; in practice, this will require a relative calibration accurate to $<15\\%$. The throughput method described here may well prove to be the best solution"
    },
    "0208/astro-ph0208505_arXiv.txt": {
        "abstract": "We investigate FUSE spectra of three PG1159 stars and do not find any evidence for iron lines. From a comparison with NLTE models we conclude a deficiency of 1--1.5\\,dex. We speculate that iron was transformed into heavier elements. A soft X-ray Chandra spectrum of the unique H- and He-deficient star H1504+65 is analyzed. We find high neon and magnesium abundances and confirm that H1504+65 is the bare core of either a C-O or a O-Ne-Mg white dwarf. ",
        "introduction": "The origin of hot hydrogen-deficient post-AGB stars (spectral types [WC] and PG1159) is supposed to be a late He-shell flash. Detailed summaries on their spectroscopic characteristics and quantitative analyses, and relevant evolutionary calculations can be found e.g.\\ in Werner (2001) and Herwig (2001), respectively.  In the last white dwarf workshop, held in Delaware two years ago, we presented for the first time clear evidence for iron deficiency in a PG1159 star (Miksa \\etal 2001). Because PG1159 central stars are very hot (\\Teff\\ $>$75\\,000\\,K), the metals are highly ionized. The dominant ionization stage of iron in the line formation region is \\ion{Fe}{VII} and most of its lines are located in the FUV spectral region. Contrary to our expectation we were not able to detect any \\ion{Fe}{VII} line in the FUSE FUV spectrum of the pulsating central star of K1-16, which means that iron is deficient by at least one dex. This has initiated an effort to utilize archival IUE and HST spectra as well as new FUSE spectra to look for iron in other PG1159 stars (Miksa \\etal 2002). It turned out that IUE and HST spectra do show a hint of iron deficiency in some other objects, but the data quality is not sufficient for reliable quantitative analyses. The same holds for a number of FUSE spectra from PG1159 stars. However, in two more cases, where FUSE spectra with sufficiently high S/N were obtained, we were successful in proving that iron is deficient. The first one is the central star of NGC\\,7094, which is classified as a hybrid-PG1159 star, because it exhibits H Balmer lines. No iron lines were detected and we concluded that iron is depleted by at least 1.5 dex (Miksa \\etal 2002). The next case is the central star of Abell~78, which is classified as one of the rare [WC]--PG1159 transition objects. For Abell~78 we find an iron deficiency of 1.5 dex from the lack of iron lines as well (Werner \\etal 2002a).  What is the origin of the iron deficiency? We think that iron has been transformed to heavier elements. The high C and O abundances in PG1159 stars result from envelope mixing caused by a late He-shell flash. This event also modifies the near-solar abundance ratios of iron-peak elements in the envelope by dredging up matter in which s-process elements were built-up by n-capture on $^{56}$Fe seeds during the AGB phase. This scenario can and will be tested by analyzing the resulting Fe/Ni abundance ratio, because it is significantly changed in the intershell region in favor of Ni by the conversion of $^{56}$Fe into $^{60}$Ni. More quantitative results from nucleosynthesis calculations in appropriate stellar models have been presented recently (Herwig \\etal 2002) and inclusion of nuclear networks in evolutionary model sequences will become available in the near future.  Interestingly, Asplund \\etal (1999) have found that in Sakurai's object, which is thought to undergo a late He-shell flash, iron is reduced to 0.1 solar and Fe/Ni$\\approx$3. This and other s-process signatures should also be exhibited by Wolf-Rayet type central stars and PG1159 stars. And in fact, latest results confirm that iron deficiency among H-deficient post-AGB stars is not restricted to PG1159 stars. It appears that three Wolf-Rayet type central stars are iron deficient, too. Gr\\\"afener \\etal (2002) report a low iron abundance in SMP\\,61, an early type [WC5] central star in the LMC. Its abundance is at least 0.7\\,dex below the LMC metallicity. Crowther \\etal (2002) find evidence for an iron underabundance of 0.3--0.7\\,dex in the Galactic [WC] type central stars stars NGC~40 ([WC8]) and BD+30$^\\circ$3639 ([WC9]). In this context it is also interesting to remark that the hot DO white dwarf \\re\\ is iron deficient and rich in nickel (Barstow \\etal 2000). If this is related to the iron deficiency in PG1159 stars is an open question.  \\begin{figure}[ht] \\epsfxsize=\\textwidth \\epsffile{werner1_f1.eps} \\caption{Detail from the Chandra count spectrum of H1504+65 (thin line). Relative fluxes of two models are overplotted, one including Mg (solid line) and one without Mg (dashed line). Observation and model spectra are smoothed with Gaussians (FWHM 0.02\\AA\\ and 0.03\\AA, respectively). } \\end{figure} ",
        "conclusions": ""
    },
    "0208/astro-ph0208219_arXiv.txt": {
        "abstract": "{Cosmic structure, dark matter, gas dynamics, galaxy formation, computer simulations} A timely combination of new theoretical ideas and observational discoveries has brought about significant advances in our understanding of cosmic evolution. Computer simulations have played a key role in these developments by providing the means to interpret astronomical data in the context of physical and cosmological theory. In the current paradigm, our Universe has a flat geometry, is undergoing accelerated expansion and is gravitationaly dominated by elementary particles that make up cold dark matter. Within this framework, it is possible to simulate in a computer the emergence of galaxies and other structures from small quantum fluctuations imprinted during an epoch of inflationary expansion shortly after the Big Bang. The simulations must take into account the evolution of the dark matter as well as the gaseous processes involved in the formation of stars and other visible components. Although many unresolved questions remain, a coherent picture for the formation of cosmic structure in now beginning to emerge. ",
        "introduction": "The origin of structure in the Universe is a central problem in Physics. Its solution will not only inform our understanding of the processes by which matter became organized into galaxies and clusters, but it will also help uncover the identity of the dark matter, offer insights into events that happened in the early stages of the Big Bang and provide a useful check on the values of the fundamental cosmological parameters estimated by other means.  Because of its non-linear character, lack of symmetry and general complexity, the formation of cosmic structure is best approached theoretically using numerical simulations. The problem is well posed because the initial conditions -- small perturbations in the density and velocity field of matter -- are, in principle, known from Big Bang theory and observations of the early Universe, while the basic physical principles involved are understood. The behaviour of the dark matter is governed primarily by gravity, while the formation of the visible parts of galaxies involves gas dynamics and radiative processes of various kinds.  Using cosmological simulations it is possible to follow the development of structure from primordial pertubations to the point where the model can be compared with observations.  Over the past few years, there has been huge progress in quantifying observationally the properties of galaxies not only in the nearby universe, but also in the very distant universe. Since the clustering pattern of galaxies is rich with information about physics and cosmology, much effort is invested in mapping the distribution of galaxies at different epochs. Two large ongoing surveys, the US-based Sloan Digital Sky Survey (York \\etal 2000), and the Anglo-Australian ``2-degree field galaxy redshift survey'' (2dFGRS, Colless \\etal 2001), are revolutionizing our view of the nearby universe with order of magnitude increases in the amount of available data. Similarly, new data collected in the past five years or so have, for the first time, opened up the high redshift universe\\footnote{In cosmology, distances to galaxies are estimated from the redshift of their spectral lines; higher redshifts correspond to more distant galaxies and thus to earlier epochs.} to detailed statistical study (Steidel \\etal 1996).  The advent of large computers, particularly parallel supercomputers, together with the development of efficient algorithms, has enabled the accuracy and realism of simulations to keep pace with observational progress. With the wealth of data now available, simulations are essential to interpret astronomical data and to link them to physical and cosmological theory. ",
        "conclusions": "Unlike most computational problems in many areas of science, the cosmological problem is blessed with known, well-specified initial conditions. Within a general class of models, it is possible to calculate the properties of primordial perturbations in the cosmic energy density generated by quantum processes during an early inflationary epoch. In a wide family of inflationary models, these perturbations are adiabatic, scale-invariant and have Gaussian-distributed Fourier amplitudes. The model also requires an assumption about the nature of the dark matter and the possibilities have now been narrowed down to non-baryonic candidates of which cold dark matter particles seem the most promising. An empirical test of the initial conditions for the formation of structure predicted by the model is provided by the cosmic microwave background radiation. The tiny temperature fluctuations it exhibits have exactly the properties expected in the model. Furthermore, the CMB data can be used to fix some of the key model parameters such as $\\Omega$ and $\\Omega_b$, while these data, combined with other recent datasets such as the 2dFGRS, allow the determination of many of the remaining parameters such as $\\Omega_m$, $\\Omega_{\\Lambda}$ and $h$.  It this specificity of the cosmological problem that has turned simulations into the primary tool for connecting cosmological theory to astronomical observations.  In addition to well-specified initial conditions, the cosmological dark matter problem has the advantage that the only physical interaction that is important is gravity. The problem can thus be posed as a gravitational N-body problem and approached using the many sophisticated techniques that have been developed over the past two decades to tackle this problem. Although on small scales there remain a number of unresolved issues, it is fair to say that on scales larger than a few megaparsecs, the distribution of dark matter in CDM models is essentially understood. The inner structure of dark matter halos, on the other hand, is still a matter of debate and the mass function of dark matter halos has only been reliably established by simulations down to masses of order $10^{11} \\Msun$. Resolving these outstanding issues is certainly within reach, but this will require carefully designed simulations and large amounts of computing power.  The frontier of the subject at present lies in simulations of the formation, evolution and structure of galaxies. This problem requires first of all a treatment of gas dynamics in a cosmological context and a number of techniques, relying on direct simulations or on semi-analytical approximations, are being explored. There are quite a few different approaches to cosmological gasdynamics, but it is reassuring that they all give similar results in the simplest relevant problem, the evolution of non-radiative gas during the formation of a galaxy cluster. No detailed comparisons exist yet for the more complicated case in which the gas is allowed to cool, but at least one of the gasdynamic simulation techniques, SPH, gives quite similar results to a simple semi-analytic approach. Realistic models of galaxy formation, however, will require much more than a correct treatment of cooling gas. Such models will necessarily have to include a plethora of astrophysical phenomena such as star formation, feedback, metal enrichment, etc. The huge disparity between the submegaparsec scales on which these processes operate and the gigaparsec scale of the large-scale structure makes it impossible to contemplate a comprehensive {\\it ab initio} calculation. The way forward is clearly through a hybrid approach combining direct simulation of processes operating on a limited range of scales with a phenomenological treatment of the others. There is currently a great deal of activity in the phenomenology of galaxy formation.  In spite of the uncertainties that remain, all the indications are that our Universe is well described by a model in which  \\begin{itemize}  \\item[(i)] the overall geometry is flat;  \\item[(ii)] the dominant dynamical components are cold dark matter ($\\sim 30\\%$) and dark energy ($\\sim 70\\%$) with baryons playing very much a supporting role ($\\sim 4\\%$);  \\item[(iii)] the initial conditions are quantum fluctuations in the primordial energy density generated during inflation and  \\item[(iv)] structure has grown primarily as a result of the gravitational instability experienced by mass fluctuations in an expanding universe.  \\end{itemize}  A skeptic is entitled to feel that the current paradigm is odd, to say the least. Not only is there a need to invoke vast amounts of as yet undetected non-baryonic cold dark matter, but there is also the need to account for the dominant presence of a dark energy whose very existence is a mystery within conventional models of fundamental physics.  Odd as it may seem, however, this model accounts remarkably well for a large and diverse collection of empirical facts that span 13 billion years of evolution."
    },
    "0208/astro-ph0208511_arXiv.txt": {
        "abstract": "We discuss fluctuations in the cosmic microwave background (CMB) polarization due to scattering from reionized gas at low redshifts.  Polarization is produced by re-scattering of the primordial temperature anisotropy quadrupole and of the kinematic quadrupole that arises from gas motion transverse to the line of sight.  We show that both effects produce equal E- and B-parity polarization, and are, in general, several orders of magnitude below the dominant polarization contributions at the last scattering surface to E-modes or the gravitational-lensing contribution to B-modes at intermediate redshifts.  These effects are also several orders of magnitude below the B polarization due to lensing even after subtraction with higher-order correlations, and are thus too small to constitute a background for searches for the polarization signature of inflationary gravitational waves. ",
        "introduction": "The angular power spectrum of cosmic microwave background (CMB) temperature fluctuations is now becoming a powerful cosmological probe \\cite{Kno95}, both due to our detailed understanding of physics during the recombination era \\cite{PeeYu70} and progress on the experimental front \\cite{Miletal99}.  In addition to the primary anisotropies generated at the last-scattering surface, CMB photons are also affected by large-scale structure at low redshifts. These latter contributions result from scattering off free electrons in clusters or the reionized IGM and from modifications due the evolving gravitational field associated with the formation of structures \\cite{Coo02}.  The existence of such secondary signals has now become evident with the recent detection of small-scale power in excess of that from primary fluctuations \\cite{Masetal02}.  The simplest and most plausible explanation for this small-scale power is the Sunyaev-Zel'dovich (SZ; \\cite{SunZel80}), re-scattering of CMB photons from hot electrons in unresolved galaxy clusters.  Although the power observed is considerably larger than theoretical expectations, the excess can be accommodated with a relatively small increase in the power-spectrum amplitude \\cite{Bonetal,KomSel,CooMel,Sigetal02}.  In addition to small-scale anisotropies in the temperature, increasing attention is now being devoted to detection of the CMB polarization.  Besides resolving cosmological parameter degeneracies \\cite{Kno95}, the polarization will allow several unique cosmological and astrophysical studies to be carried out. These include a reionization signal \\cite{Zaldarriaga:1996ke}, probes of gravitational lensing \\cite{SeljakZaldarriaga}, and a signature of inflationary gravitational waves (IGW) through its contribution to the B, or curl, modes of the polarization \\cite{EB}.  Given the rapid pace of experimental progress and the rule of thumb that the polarization is typically 10\\% of the temperature anisotropy, it is appropriate to investigate the polarization produced by the secondary effects that have produced the recently detected small-scale power.  This polarization is produced by Thomson scattering of the quadrupole moment of the radiation incident on the scatterer.  The quadrupole moment can be either the primordial quadrupole that the scatterer sees \\cite{KamLoeb} or the kinematic quadrupole that arises from quadratic terms in the Doppler shift when the gas moves transverse to the line of sight \\cite{SunZel80}.  Small-scale angular fluctuations in this polarization are produced by variations in the optical depth as a function of position across the sky.  Since the secondary polarization signals can affect cosmological studies involving the primordial polarization, and, by themselves, may provide important information on astrophysics at late times, it is important that we both quantify and understand the extent to which large-scale structure can be a potential source of CMB polarization.  Here, we discuss the angular power spectra for scattering from reionized electrons and study how these may affect potentially interesting studies with CMB polarization.  Our calculation parallels that of Ref. \\cite{Hu99}, although differs in that we present a simplified derivation of the results based on a flat-sky approximation and use a halo-clustering approach \\cite{CooShe02} to describe fluctuations in the electron density, in analogy to similar recent calculations of small-scale temperature fluctuations from unresolved clusters \\cite{KomSel,KomKit,Coo2000}.  We also include for the first time the frequency dependence in the small-scale polarization using results for the polarization from individual clusters \\cite{SazSun99,Challinor}.  The paper is organized as follows.  In \\S~\\ref{sec:pol}, we introduce polarization signals associated with the scattering of the primordial CMB temperature quadrupole and with the kinematic quadrupole generated by transverse motions. We then display and discuss our results in \\S~\\ref{sec:results}. Though we present a general discussion of the polarization, when illustrating results, we will use the currently favored $\\Lambda$CDM cosmology with matter density (in units of the critical density) $\\Omega_m=0.35$, baryon density $\\Omega_b=0.05$, vacuum-energy density $\\Omega_\\Lambda=0.65$, Hubble constant (in units of 100 km~sec$^{-1}$~Mpc$^{-1}$) $h=0.65$, and spectral $n=1$ for primordial density perturbations. We employ natural units with the speed of light $c=1$ throughout. ",
        "conclusions": "\\label{sec:results}  \\begin{figure}[t] \\centerline{\\psfig{file=fig4.eps,width=3.5in,angle=-90}} \\caption{The fractional contribution to polarization power spectra due to scattering of the primary anisotropy temperature quadrupole as a function of redshift.  Here, we plot $d \\ln C_l/ d \\ln z$, for three specific values of $l$ (10$^2$, 10$^3$, and 10$^4$). We show the total (solid curves) as well as the 1-halo term (dotted curves). Note that contributions come from a broad range in redshift, while, with  increasing  $l$, or decreasing angular scale, fractional contributions in the 1-halo term increase to higher redshifts.} \\label{fig:dCdz} \\end{figure}  \\begin{figure}[t] \\centerline{\\psfig{file=fig5.eps,width=3.5in,angle=-90}} \\caption{The fractional contribution to polarization power spectra due to scattering of the primary anisotropy temperature quadrupole as a function of cluster mass (in terms of solar mass). Here, we plot $d \\ln C_l/ d \\ln M$, for three specific values of $l$ (10$^2$, 10$^3$, and 10$^4$). We show the total (solid line), the 1-halo term (dotted), and the 2-halo term (dashed).} \\label{fig:dCdM} \\end{figure}  We summarize our results on the polarization power spectra in Fig.~\\ref{fig:cl}. Note that the secondary polarization discussed here contributes equally to E- and B-modes.  While the 1-halo term dominates at arcminute angular scales and below, correlations between halos are important and determine the total effect due to secondary polarization at angular scales corresponding to a few degrees.   The dependence of our results on the inclusion of the 2-halo term is consistent with the result obtained for the temperature power spectra from the kinetic SZ effect \\cite{Cooray:2001wa}, while it is inconsistent with the thermal SZ effect, where contributions are dominated by the 1-halo term over the whole range of angular scales. The latter behavior is explained by the fact that the thermal-SZ effect is highly dependent on the most massive halos, while the kinetic SZ effect, and the secondary polarization signals calculated here, are {\\it in}dependent of the gas temperature and thus can depend on halos with a wider mass range.  As shown in Fig.~\\ref{fig:cl}, the secondary E-mode polarization is several orders of magnitude below the E polarization from the surface of last scattering.  The secondary polarization is therefore unlikely to be a source of confusion when interpreting polarization contributions to E modes.  The amplitude of the primary effect in B modes, due to gravitational waves, is highly uncertain and depends on the energy scale of inflation \\cite{KamKos}. For illustration, we show in Fig.~\\ref{fig:cl} the inflationary gravitational wave (IGW) signal assuming an energy scale for inflation of $E_{\\rm infl}=10^{16}$ GeV; the amplitude of the power spectrum scales as $E_{\\rm infl}^4$. At large angular scales the secondary polarization is several orders of magnitude below the peak of this hypothetical IGW polarization signal.  If the energy scale of inflation is lowered considerably, say to $E_{\\rm infl} \\lesssim 10^{15}$ GeV, then we might guess that the secondary polarization could ultimately constitute a background.   As also shown in Fig. \\ref{fig:cl}, however, there is a contribution to the B-mode power spectrum that arises from conversion of the primary E modes to B modes by gravitational lensing \\cite{SeljakZaldarriaga}, and this is considerably larger than the secondary polarization. Moreover, we also show (the dot-dash curve) the contribution to the irreducible B-mode power spectrum that remains even after the lensing has been optimally subtracted with higher-order correlations \\cite{Kesdenetal,Knox,CooKes,Hu:2001fa}.  This residual lensing power spectrum is considerably larger than the polarization from reionization.  If the power spectrum is measured at a frequency $\\nu\\simeq220$ GHz, then the polarization power spectrum from the kinematic effect will be boosted by a factor $\\sim5$ from the $g(x)=1$ power spectrum shown in Fig. \\ref{fig:cl}. Moreover, if $\\sigma_8=1$ (rather than the value $\\sigma_8=0.9$ assumed in Fig.~\\ref{fig:cl}), then both the secondary power spectra will be increased for the same reasons that the temperature power spectra increase by a factor $\\sim3$.\\footnote{However, the boost in the polarization power spectra will be smaller due to the fact that much of the polarization is induced by electrons in smaller halos, rather than the massive clusters that induce the temperature fluctuation.}  Even with the possible frequency and $\\sigma_8$ boosts, the secondary effects we consider here will be unlikely to be a factor for either gravitational-lensing or gravitational-wave studies with B modes.  In Fig.~\\ref{fig:dCdz}, we show the fractional contributions to polarization power spectra associated with the scattering of the temperature quadrupole as a function of redshift.  We show the total and the 1-halo term for three specific values of $l$ corresponding to degree scales to arcminute angular scales.  Though not shown here, we find consistent behavior for the polarization power spectrum generated by the scattering of the kinematic quadrupole.  As shown, contributions come over a broad range in redshift out to the assumed reionization redshift of 10 with a decrease at highest redshifts due to the decreasing abundance of massive halos at higher redshifts. At arcminute scales with $l \\sim 10^4$, the signal arises from halos at $z >1$. A comparison of this behavior to Fig.~\\ref{fig:PowSpec} reveals that at redshifts greater than 1, the primary temperature quadrupole at the cluster positions results from a projection of the SW effect only. Thus, at small angular scales, scattering contributions come only from the quadrupole associated with the SW effect and not the total that includes the ISW effect as well.  In Fig.~\\ref{fig:dCdM}, we show the mass dependence of the secondary polarization signal, again using the scattering of the temperature quadrupole for illustration purposes. We show the total, the 1-halo, and the 2-halo term at three different values of $l$. While the 1-halo term is dominated by halos at the high-mass end of the mass function, the 2-halo term arises from a wide range in halo mass. At tens of arcminute scales equal contributions come from halo masses in the range of 10$^{10}$ to 10$^{14}$ $M_{\\sun}$. This is consistent with the equivalent result for the kinetic SZ effect where a wide range of masses contribute.  Note that in addition to the auto-correlation of polarization, one expects secondary temperature fluctuations due to galaxy clusters to be correlated with that of the secondary polarization involving the E-mode. The temperature-polarization cross-correlation with the B-modes is expected to be zero based on parity considerations.  We considered all combinations between secondary polarization and temperature anisotropies involving thermal and kinetic SZ effects and found them to be zero based on simple geometric arguments.  While our simple flat-sky derivation of the secondary polarization anisotropy from reionization agrees with the all-sky approach of Ref.~\\cite{Hu99}, our calculational method complements the one used there. We use the halo model to describe the non-linear power spectrum of electrons while in Ref.~\\cite{Hu99}, electrons of the intergalactic medium were assumed to trace the dark-matter-density field. Our numerical results agree well with those of Ref.~\\cite{Hu99}, particularly at large angular scales where they should both converge to the same linear-theory calculation.  Here we have neglected to consider the expected smoothing of the electron density on small scales from reheating of the IGM gas.  However, as Ref. \\cite{Hu99} shows, these effects are easily included and reduce the power spectrum substantially only on angular scales $l \\gtrsim10^4$ smaller than those we have considered here.  While our halo-based approach is likely to be affected by uncertainties related to the mass function or the distribution of electrons within halos, we expect our calculations to accurately reflect the polarization anisotropy power at small angular scales. In the case of the kinematic quadrupole, it is likely that we have overestimated the power at scales of a few degrees or more due to our assumption that the velocity field is coherent at such angular scales. We expect this assumption to only affect the 2-halo term of correlations and to result in an overestimate of $C_l$ when $l$ is less than $\\sim$ 1000."
    },
    "0208/astro-ph0208382_arXiv.txt": {
        "abstract": "From {\\sl Hubble Space Telescope\\/} images with $0{\\farcs}05$ resolution we identify four stars brighter than $V = 25$ mag within $2{\\farcs}5$ of SN 1993J in M81 which contaminated previous ground-based brightness estimates for the supernova progenitor.  Correcting for the contamination, we find that the energy distribution of the progenitor is consistent with that of an early K-type supergiant star with $M_V \\approx -7.0 \\pm 0.4$ mag and an initial mass of 13--22 $M_\\odot$.  The brightnesses of the nearby stars are sufficient to account for the excess blue light seen from the ground in pre-explosion observations.  Therefore, the SN 1993J progenitor did not necessarily have a blue companion, although by 2001, fainter blue stars are seen in close proximity to the supernova.  These observations do not strongly limit the mass of a hypothetical companion.  A blue dwarf star with a mass up to 30 $M_\\odot$ could have been orbiting the progenitor without being detected in the ground-based images.  Explosion models and observations show that the SN 1993J progenitor had a helium-rich envelope.  To test whether the helium abundance could influence the energy distribution of the progenitor, we calculated model supergiant atmospheres with a range of plausible helium abundances.  The models show that the pre-supernova colors are not strongly affected by the helium abundance longward of 4000~\\AA, and abundances ranging between solar and 90\\%\\ helium (by number) are all consistent with the observations. ",
        "introduction": "Direct identification of the progenitor star has been scarce for the more than 2000 known historical supernovae (SNe); hence, our knowledge of the types of stars that become SNe is based primarily on models of stellar evolution and observations of the environments in which SNe occur.  Pre-SN stars have been detected for only five events: SN~1987A in the LMC (e.g., Walborn et al.~1987), SN~1961V in NGC 1058 (Goodrich et al. 1989; Filippenko et al. 1995), SN~1978K in NGC 1313 (Ryder et al.~1993), SN~1997bs in NGC 3627 (Van Dyk et al.~1999), and SN~1993J in M81 (Filippenko 1993a; Aldering, Humphreys, \\& Richmond 1994, hereafter AHR; Cohen, Darling, \\& Porter 1995).  Interestingly, the properties of all five of these SNe are atypical of type II supernovae (SNe~II).  In fact, the highly unusual properties of SN 1961V and SN 1997bs (Van Dyk et al.~2000) suggest that they were actually eruptions of extremely massive stars, similar to $\\eta$~Carinae, rather than genuine SNe.  SN~1993J was initially classified as a SN~II, because its optical spectra showed hydrogen lines (Filippenko 1993b; Garnavich \\& Ann 1993), but it soon developed the characteristics of a SN~Ib (Filippenko \\& Matheson 1993; Filippenko, Matheson, \\& Ho 1993; Filippenko, Matheson, \\& Barth 1994) --- hence, the classification as Type IIb.  Several groups (see the review by Wheeler \\& Filippenko 1996) concluded that the progenitor of SN~1993J was a massive star (10--20~M$_\\odot$) that had lost most of its hydrogen envelope before exploding.  The models generally employ a close companion star to strip the progenitor of its envelope, since mass loss through a wind is not efficient at such low initial masses. The companion gains mass and may reach more than 15~M$_\\odot$ before the supernova explosion (Woosley et al. 1994).  H\\\"oflich, Langer, \\& Duschinger (1993) offered an alternative model, in which a single, very massive star loses its envelope through a wind before exploding; the results, however, do not fit the observations as well as do binary star models.  The photometric properties of the SN 1993J progenitor were estimated by AHR, using an extensive ground-based photographic and CCD archive for M81.  Their analysis was affected by the range in quality of the observations and the crowded field near the SN.  AHR concluded that the progenitor was a K0 supergiant with contamination from one or more blue stars projected near the SN and possibly light from an OB association that spawned SN 1993J.  From a set of CCD images obtained in the 1980s to search for novae in M81, Cohen et al. (1995) determined that the SN 1993J progenitor was not variable at the $0.2$ mag level. They also confirmed the AHR estimate of the progenitor color.  Crotts (1995) obtained high-quality CCD images 600 days after maximum brightness and found a red star with $I \\approx 22.7$ mag within 1\\arcsec\\ of SN~1993J.  This star was responsible for the elliptical appearance of the progenitor in the AHR set of images, even under good seeing conditions.  In the course of the Supernova Intensive Study (SINS) collaboration, multiwavelength {\\sl Hubble Space Telescope\\/} ({\\sl HST}) images had been obtained of SN~1993J and the surrounding star field in 1994 and 1995, while the SN was fading.  The SN and the field have also been observed in 2001, though not as the main target, by program GO-9073 (see Liu, Bregman, \\& Seitzer 2002). In this paper, we use all these data to help resolve some questions that were raised by the ground-based data about the progenitor.  In particular, we estimate the contamination of the progenitor light by stars which are unresolved in the ground-based images and attempt to determine the source of the excess blue light seen by AHR.  We also estimate the brightness, color, age, and initial mass of the progenitor.  In \\S 2 we describe the observations and photometric reduction.  The implications of the {\\sl HST\\/} data for the pre-SN observations are discussed in \\S 3, and conclusions are presented in \\S 4. ",
        "conclusions": "We have used {\\sl HST\\/} images of SN~1993J after outburst to search for stars within a ground-based resolution element, which may have contaminated the archival ground-based observations of the SN progenitor.  We find four stars within $2{\\farcs}5$ of the progenitor which were at least partially included in the progenitor energy distribution determined by AHR.  Correcting for this contamination we find the following.  \\begin{description}  \\item[1)] The progenitor may have had apparent $V = 21.6 \\pm 0.3$, $B-V = 0.8 \\pm 0.6$, and $V-I = 1.9 \\pm 0.5$ mag, implying an absolute magnitude $M_V \\approx -7.0 \\pm 0.4$ for $A_V = 0.75$ mag (the uncertainty in the absolute magnitude includes the estimated photometric uncertainty, plus the uncertainty in the distance modulus).  The brightness and colors are consistent with those of a reddened early K-type supergiant, although possibly somewhat too blue in the blue bands.  The mass of the progenitor can be constrained by the possible mass for the nearby red supergiant, Star A, i.e., $\\sim$13 $M_\\odot$, and the age and mass of a hypothetical progenitor suggested by its brightness and colors, i.e., $\\sim$22 $M_\\odot$.  This range, 13--22 $M_\\odot$, is consistent with other estimates for the progenitor mass in the literature (e.g., Podsiadlowski et al.~1993; Woosley et al.~1994; Young, Baron, \\& Branch 1995; Iwamoto et al.~1997).  \\item[2)] Model atmospheres with a wide range of helium abundances fit the observed progenitor energy distribution redward of 4000 \\AA.  Although at shorter wavelengths the models show a greater spread in color with helium fraction, the data are insufficient to differentiate between them.  \\item[3)] The blue excess seen by AHR can be fully accounted for by the nearby stars.  The progenitor was not a member of a significant OB association.  Any companion to the progenitor must have had $M_B > -5.6$ mag for reasonable assumptions of the reddening and distance to the SN.  The mass of the companion is not very strongly constrained, but must be less than 30 $M_\\odot$. This is consistent with the binary mass-transfer scenario, even for conservative mass exchange (e.g., Podsiadlowski et al. 1993).  \\end{description}  SN 1993J appears to be interacting with circumstellar material ejected by the progenitor during its evolution (e.g., Filippenko et al.~1994; Van Dyk et al.~1994; Patat, Chugai, \\& Mazzali 1995; Matheson et al.~2000).  The energy from this interaction continues to power the optical light curve, which in 1995 slowed its decline to less than 2 millimag d$^{-1}$ (Garnavich et al.~1995). Comparing the magnitudes derived from the {\\it HST\\/} WFPC2 images in 1995 and 2001, the SN is still declining at only ${\\sim}$0.2 mag yr$^{-1}$.  Recent ACS multi-band images have been obtained with {\\sl HST\\/} by program GO-9353, but even with the superior sensitivity and resolution of ACS, the SN is likely still too bright to successfully isolate a companion or other stars in the immediate SN environment (${\\le}0{\\farcs}1$--$0{\\farcs}2$).  Only when the SN sufficiently fades, probably not for a few more {\\sl HST\\/} cycles, can the high-resolution imaging be used to search for a companion to the SN progenitor and better determine the progenitor's nature."
    },
    "0208/astro-ph0208457_arXiv.txt": {
        "abstract": "We present a comparison between the observed properties of damped Ly$\\alpha$ systems (DLAs) and the predictions of simple models for the evolution of present day disk galaxies, including both low and high surface brightness galaxies. We focus in particular on the number density, column density distribution and gas density of DLAs, which have now been measured in relatively large samples of absorbers. From the comparison we estimate the contribution of present day disk galaxies to the population of damped Ly$\\alpha$ systems, and how it varies with redshift. Based on the differences between the models and the observations, we also speculate on the nature of the fraction of DLAs which apparently do not arise in disk galaxies. ",
        "introduction": "The true nature of the galaxies responsible for high redshift damped Ly$\\alpha$ systems (DLAs, absorbers seen in quasar spectra with H~I column densities $N$(H~I)\\,$\\geq 2 \\times 10^{20}$~cm$^{-2}$) is largely unconstrained. At present, there are two main competing scenarios for their origin. One school of thought sees DLAs as the (large) progenitors of massive {\\it spiral disks} (Wolfe \\etal 1986; Lanzetta \\etal 1991). The gas disks would have formed at $z  >5$ through monolithic collapse, and this gas is converted to stars over a Hubble time. In support of this picture Prochaska \\& Wolfe (1998) argued that the kinematics of the metal absorption lines in DLAs are best explained if they are formed in thick, large, and rapidly rotating galactic disks, with circular velocities $V_C  \\simgt 200$\\,km~s$^{-1}$. However, this large disk hypothesis runs counter to currently popular models of hierarchical structure formation in which present day galaxies are assembled from virialized sub-units over a protracted time interval ($z \\sim 1-5$).   Haehnelt \\etal (1998) argued that hydrodynamic N-body simulations are able to reproduce the velocity structure of the absorption lines with infalling sub-galactic clumps in collapsing dark matter haloes with small virial velocities ($V \\sim 100$\\,km~s$^{-1}$).   More generally, the kinematics and metallicites of DLAs have been interpreted as evidence that they arise in spiral galaxies (Fritze-V.Alvensleben et al. 2001; Hou, Boissier \\& Prantzos 2001), low surface brightness galaxies (Jimenez, Bowen \\& Matteucci 1999), dwarf galaxies (Matteucci, Molaro \\& Vladilo 1997), the progenitors of globular clusters (Burgarella, Kissler-Patig, \\& Buat 2001), the building blocks of current galaxies (Tissera et al. 2001), galaxies undergoing tidal stripping and mergers (Maller et al. 2001), and outflows from dwarf galaxies (Schaye 2001a).  Observationally, imaging studies of the fields of QSOs with damped systems have shown conclusively that the absorbers are a very `mixed bag', which includes galaxies of different luminosities and surface brightnesses, down to objects with apparently no associated stellar populations which remain undetected even in very deep images (Steidel \\etal 1994, 1995; Le Brun \\etal 1997; Lanzetta et al. 1997; Fynbo \\etal 1999; Pettini \\etal 2000; Turnshek \\etal 2001; Bowen, Tripp, \\& Jenkins 2001; Kulkarni et al. 2000, 2001; Colbert \\& Malkan 2002). While it seems clear that selection based on H~I absorption cross-section picks out a variety of galaxies, it is of interest to establish if one particular class of objects dominates and, if so, whether the dominant population changes with cosmic epoch.  In this paper we assess the contribution of the progenitors of today's disk galaxies to the DLA population at different redshifts by comparing the most up to date observational determinations of several properties of DLAs with models of the chemical and spectrophotometric evolution of disk galaxies, including low surface brightness (LSB) galaxies. Specifically, we have compiled recent data on the number density per unit redshift, the column density distribution and its integral, which gives the total mass of H~I traced by DLA, for a large sample of DLAs. These data are presented in \\S2, while the models of disk galaxies are discussed in \\S3. In \\S4 we compare our predictions with the statistical properties of DLAs, as well as with available imaging observations of the galaxies identified as DLA absorbers. We further speculate on what the differences we find at high redshift between models and observations may be telling us about the earliest population of DLAs. We summarise and discuss our results in \\S6.  Throughout the paper we adopt the currently favoured cosmology $H_0 = 65 $\\,km~s$^{-1}$~Mpc$^{-1}$, $\\Omega_M = 0.3$, $\\Omega_{\\Lambda} = 0.7$. ",
        "conclusions": "\\label{summary}  In this paper we have used the models developed by Boissier \\& Prantzos (2000), which describe the chemical and spectrophotometric evolution of disk galaxies of different surface brightnesses, to assess whether the progenitors of present day spirals can account for the population of DLA absorbers, as is often assumed. To this end, we have compared the model predictions with several, recently determined, statistical properties of DLAs, in particular their number density per unit redshift, their column density distribution, and its integral which gives $\\Omega_{\\rm DLA}$, the total mass of neutral gas traced by DLAs. In addition we have brought together available imaging data on these absorbers at $z < 1$ from the literature, to compare with the morphologies, sizes and luminosities of the model galaxies. Our main results can be summarised as follows.  1. The models are reasonably successful at reproducing many of the properties of DLAs at redshifts $z \\simlt 2$, within the uncertainties in the measurements. Specifically, the number density of absorbers per unit redshift and about half of the neutral gas mass they trace can be reproduced with evolving disk galaxies in our models. Dwarf galaxies, which have not been considered here, may account for some of the DLA absorbers, but this population is not the dominant one.  2. Normal spirals and low surface brightness galaxies make comparable contributions in our models to both the numbers of DLAs and their neutral gas mass. Turning this statement around, it is only with the inclusion of LSBs that it is possible to reproduce the statistics of DLAs in our models. While LSBs may not be as numerous as high surface brightness galaxies, their large dimensions and high gas content combine to make a significant contribution to the overall cross section for H~I absorption. The scant imaging data available at $z < 1$ is broadly in agreement with this conclusion, although the observed distribution of impact parameters $D$ seems to be narrower, and more peaked towards lower values of $D$, than our models predict. The distributions of luminosities agree at the bright end, but the data also include a few cases of very faint DLA-producing galaxies which are not predicted by the models. More uniform and extensive imaging surveys are required to reach firm conclusions on both of these points. Furthermore, the properties of LSBs, which have been included in our models according to a rather simple recipe, need to be investigated further for a proper comparison with DLAs.  3. We have investigated the effects of a possible dust-induced bias in current DLA samples, which may lead to an underestimate of the relative number of sightlines through galactic regions where the column densities of gas and metals are both high. We have found this potential problem to have a relatively minor effect, particularly on $\\Omega_{\\rm DLA}$, in agreement with the initial results from the CORALS survey by Ellison et al. (2001).  4. As we move from $z \\sim 2$ to higher redshifts, models based on extrapolating back in time the properties of today's disk galaxies are progressively less successful in reproducing the statistics of DLAs. The data can be interpreted as evidence for the existence of an additional population of DLAs, of generally lower column density, which dominate the number density of absorbers over the progenitors of today's disks beyond $z \\simeq 3$. Possibly these are sub-units destined to merge with each other and eventually with more massive galaxies by $z \\simeq 2$. This interpretation is supported by the failure up to now to detect the galaxies producing DLAs at $z \\simeq 4$ in deep images reaching down to $\\simlt 1/4 L^{\\ast}$. The history of assembly of today's galactic disks may well be reflected in the redshift evolution of the number density of DLAs.\\\\   Samuel Boissier would like to acknowledge the support of a Framework 5 Marie Curie fellowship through contract number HPMF-CT-2000-00521."
    },
    "0208/astro-ph0208331_arXiv.txt": {
        "abstract": "We present the results of high-resolution mid-infrared observations of the source NGC3576 IRS 1. Near diffraction-limited images were taken at the Gemini South Observatory\\footnote{Based on observations obtained at the Gemini Observatory, which is operated by the Association of Universities for Research in Astronomy, Inc., under a cooperative agreement with the NSF on behalf of the Gemini partnership: the National Science Foundation (United States), the Particle Physics and Astronomy Research Council (United Kingdom), the National Research Council (Canada), CONICYT (Chile), the Australian Research Council (Australia), CNPq (Brazil) and CONICET (Argentina).} through OSCIR's\\footnote{This paper is based on observations obtained with the mid-infrared camera OSCIR, developed by the University of Florida with support from the National Aeronautics and Space Administration, and operated jointly by Gemini and the University of Florida Infrared Astrophysics Group.} filters N (10.8 $\\mu$m), 7.9, 9.8, 12.5 and IHW18 (18.2 $\\mu$m). The source IRS1 was resolved into 3 sources for the first time at mid-infrared wavelengths. For each source we constructed the SED from 1.25 to 18 $\\mu$m, as well the color temperature and the spatial distribution of the dust in the region. The optical depth of the silicate absorption feature at 9.8 $\\mu$m is presented also. ",
        "introduction": "The formation mechanism of massive stars is essentially unknown. This is mostly due to observational difficulties in finding and establishing an evolutionary sequence for young stellar objects (YSOs). It is believed that during the accretion phase, YSOs remain heavily enshrouded in dusty cocoons, behind hundreds of magnitudes of extinction at visual wavelengths (even at near-infrared wavelengths for the earliest phases).  A very interesting group of YSOs has been identified in the Galactic giant HII regions (GHII) M17 (Hanson, Horwarth \\& Conti 1997), W43 (Blum, Damineli \\& Conti 1999), W42 (Blum, Conti \\& Damineli 2000), W31 (Blum, Damineli \\& Conti 2001) and NGC 3576 (Figueredo et al. 2002). These are luminous YSOs with excess emission in the K band and have color indexes H-K$>$2. J, H and K spectra of these objects typically show a featureless continuum. In some cases the CO 2.3 $\\mu$m bandhead is seen in emission or absorption, and in others FeII and/or H$_{2}$ are seen in emission.  Figueredo et al. (2002) identified two massive YSO candidates in the position of IRS 1 in NGC 3576. This source was discovered by Lacy, Beck \\& Geballe (1982) and it was recently observed at 10 $\\mu$m by Walsh et al. (2001), but none of their images have enough spatial resolution to resolve the source. ",
        "conclusions": "We presented high-resolution mid-infrared images of NGC 3576 IRS 1, that has been resolved into 3 sources for the first time at mid-infrared wavelengths. The SEDs of each source were constructed from 1.25 to 18 $\\mu$m with data taken from literature. Each SED shows increasing fluxes toward to longer wavelengths. Hillenbrand et al. (1992) found similar SEDs for a group of less massive young stars in process of accreting mass via an accretion disk embedded in a dusty cocoon. This similarity suggests the same interpretation for our results. Finally we presented maps of dust distribution, temperature and optical depth as well.\\\\ \\\\ \\\\ \\\\ We thank Chris De Pree for kindly providing the FITS file of the 3.4 cm map of NGC 3576.\\\\ CLB and AD acknowledge the financial support from PROAP, PRONEX and FAPESP.\\\\ PC appreciates continuing support from the National Science Foundation."
    },
    "0208/astro-ph0208107_arXiv.txt": {
        "abstract": "Data from the Tibet-III air shower array (with energies around 3 TeV) and from the Tibet-II array (with energies around 10 TeV) have been searched for diffuse gamma rays from the Galactic plane.  These arrays have an angular resolution of about 0.9 degrees.  The sky regions searched are the inner Galaxy, $20^\\circ \\leq l \\leq 55^\\circ$, and outer Galaxy, $140^\\circ \\leq l \\leq 225^\\circ$, and $|b| \\leq 2^\\circ$ or $\\leq 5^\\circ$. No significant Galactic plane gamma-ray excess was observed. The 99\\% confidence level upper limits for gamma-ray intensity obtained are (for $|b| \\leq 2^\\circ$) 1.1 $\\times 10^{-15}$ cm$^{-2}$s$^{-1}$sr$^{-1}$MeV$^{-1}$ at 3 TeV and 4.1 $\\times 10^{-17}$ cm$^{-2}$s$^{-1}$sr$^{-1}$MeV$^{-1}$ at 10 TeV for the inner Galaxy, and 3.6 $\\times 10^{-16}$ cm$^{-2}$s$^{-1}$sr$^{-1}$MeV$^{-1}$ at 3 TeV and 1.3 $\\times 10^{-17}$ cm$^{-2}$s$^{-1}$sr$^{-1}$MeV$^{-1}$ at 10 TeV for the outer Galaxy, assuming a differential spectral index of 2.4.  The upper limits are significant in the multi-TeV region when compared to those from Cherenkov telescopes in the lower energy region and other air shower arrays in the higher energy region; however, the results are not sufficient to rule out the inverse Compton model with a source electron spectral index of 2.0. ",
        "introduction": "Detection of diffuse gamma rays from the Galactic plane is considered to be a promising way to understand spatial distributions of cosmic-ray acceleration regions, of interstellar matter (ISM), and interstellar photon field (ISPF) densities.  Experimental data in the energy region below 10 GeV with the EGRET instrument \\cite{hunt97}, are fairly well interpreted in terms of interactions of cosmic-ray hadrons and electrons with ISM and ISPF. However, a portion of the contribution of each elementary process of interaction is uncertain due to assumed intensities and spectral indices of cosmic-ray hadrons, electrons, and densities of ISM and ISPF. With experiments in the energy region above 100 GeV, only upper limits for diffuse gamma-ray intensities have been obtained, and these results provide some constraints on the parameters in the models.  In order to constrain the parameter values more severely in this energy region, it is necessary to obtain more observational data.  \\par In the early stage of cosmic-ray astrophysics, the production rate of gamma rays by Compton scattering of starlight photons by cosmic-ray electrons was calculated by Feenberg \\& Primakoff (1948), and gamma rays arising from neutral pions through cosmic-ray collision with ISM was studied by Hayakawa (1952).  Morrison (1958) advocated gamma-ray astronomy because of the known production mechanisms and the straight trajectories from the origin. Ginzburg \\& Syrovatskii (1964) summarized the elementary process for gamma-ray production and discussed diffusion and propagation of cosmic rays in the Galactic disc.  As a result, it is quite natural to expect gamma rays from the Galactic plane, however was twenty years after the pioneering work by Feenberg \\& Primakoff (1948) until the intense band of gamma-ray intensity along the Galactic plane was observed with OSO~3 \\cite{krau72} and a balloon-borne detector \\cite{fich72}. Advanced observations were carried out by detectors borne on SAS~2 (Fichtel et al. 1975; Hartman et al. 1979) and COS~B (Mayer-Hasselwander et al. 1980, 1982).  These observations revealed that the intensity profile of high-energy gamma rays is quite relevant to the structure of the Galaxy, and have stimulated much theoretical work in gamma-ray astronomy below 10 GeV (see, Bertsch et al. 1993 and references therein).  \\par A detailed intensity distribution of high-energy gamma rays coming from the Galactic plane was given by Hunter et al. (1997) based on EGRET observations. In the energy region above 1 GeV, the gamma-ray intensity from the inner Galaxy is higher than the COS~B data by a factor of about 3.  It is also higher than the conventional model predictions (e.g., Hunter et al. 1997; Bertsch et al. 1993) by a factor of 1.7, where the neutral pion production is based on the calculations of Stecker (1988), assuming the power law proton spectrum with spectral index of 2.75 \\cite{derm86}.   Mori (1997) showed that the latter discrepancy can not be attributed to a calculation with inaccurate accelerator data on neutral pion production. It can, however, be interpreted by adopting a harder proton spectral index of 2.45 for the EGRET excess within a plane thickness of $|b| \\leq 10^\\circ$. Webber (1999) also showed that the EGRET excess in $|b| \\leq 5^\\circ$ can be reproduced by assuming a source proton spectral index of 2.25.  \\par Following an indication by Schlickeiser (1979) of the importance of using the Klein-Nishina cross section, Protheroe \\& Wolfendale (1980) showed possible dominance of inverse Compton (IC) gamma rays over neutral-pion decay gamma rays for a range of electron injection spectra. Detailed calculations of IC gamma rays above 50 MeV were made by Chi et al. (1989), which were more than 50 \\% of the total diffuse intensity at a medium galactic latitude of $|b|=10^\\circ - 20^\\circ$.  Porter \\& Protheroe (1997) indicated that in such a high-energy region, cosmic-ray electrons may create a significant part of the diffuse gamma rays, depending on their injection spectral index and acceleration cutoff energy. Pohl \\& Esposito (1998) remarked that most of radio synchrotron spectra of SNRs are well represented by power law indices around 0.5, corresponding to an electron injection index of about 2.0 \\cite{gree95}. They argued that if this injection electron index is employed, the EGRET excess above 1 GeV can be well explained by IC scattering.  They also argued that an electron injection index of 2.0 is within the expected Poisson fluctuations and is reasonable in the direction toward the Galactic center with the line of sight passing through the vicinity of many SNRs, taking into account the diffusion coefficient and the observed local electron index of around 3.0 below 1 TeV.   Recently, the energy spectrum of local electrons has been obtained with balloon-borne instruments HEAT \\cite{barw98} and BETS \\cite{tori01}.  \\par Some models of diffuse Galactic gamma-ray continuum radiation below 10 GeV were synthetically discussed by Strong, Moskalenko, \\& Reimer (2000) with respect to the injection spectra, including a harder nucleon and electron spectral indices, and also with respect to different ISM and ISPF densities.  At the higher energies of the TeV-PeV region, the diffuse gamma-ray emission was calculated by Berezinsky et al. (1993) in terms of cosmic-ray interaction with ISM, and also by Ingelman \\& Thunman (1996) using the current models for high energy particle interaction. Broad-band diffuse gamma-ray emission, covering the MeV-PeV region, was comprehensively calculated by Aharonian \\& Atoyan (2001), and they discussed propagation of hadron and electron components and their injection spectra. Assuming that the hadron spectral index is 2.15 (or 2.0) in SNRs, Berezhko \\& V$\\ddot{\\rm o}$lk (2000) showed that the averaged contribution to the diffuse gamma-ray flux should exceed the current model predictions (Hunter et al. 1997; Bertsch et al 1993) by a factor 5 (or 29) at 1 TeV.  \\par Recently, several groups gave upper limits for diffuse gamma-ray fluxes (Borione et al. 1998) above 100 TeV from the outer Galactic plane, using large ground-based air shower arrays and muon detectors, and above 500 GeV \\cite{lebo00} and around 1 TeV \\cite{aha01} from the inner Galactic plane, using imaging atmospheric Cherenkov telescopes (IACT).  Using a small scintillation counter array, Tibet-I, upper limits \\cite{amen97} were given at 10 TeV from the inner and outer Galactic planes.  In this paper we report new upper limits from both the enlarged Tibet-II array and the high detector density array of Tibet-III. ",
        "conclusions": "The significance of an on-plane excess, $(E-B)/\\sqrt{B}$ measured in standard deviation of the number of on-plane events, is calculated for each distribution shown in Fig. 6 and the results are summarized in Table 2. In Table 2, the results thus obtained are given for the regions of IG ($20^\\circ \\leq l \\leq 55^\\circ$) and OG ($140^\\circ \\leq l \\leq 225^\\circ$). Mode energies 3 TeV and 10 TeV indicate the analyses of the Tibet-III and Tibet-II data, respectively. As given in this table, no significant excess is found, although an excess of +2.52 $\\sigma$ is marginal for IG at 3 TeV in the 4$^\\circ$ bin analysis of the Tibet-III data.  We calculate the upper limits for the gamma-ray intensity using the methods given by Helene (1983) and Protheroe (1984), specifically for a small excess or deficit on the Galactic plane.  In this table, $J_\\gamma/J_{\\rm CR}$ means the flux ratio at 1 $\\sigma$ excess, which is identical to $1/\\sqrt{B}$, of the diffuse gamma rays versus the galactic cosmic rays in the energy region above 3 TeV and above 10 TeV.  A minor component of the isotropic diffuse gamma rays is included in the galactic cosmic rays because separating them is impossible in the Tibet air shower array due to the lack of any equipment to reduce the background hadron initiated air showers.  \\vspace{-3mm}  \\begin{center} TABLE 2\\\\ LIMITS TO DIFFUSE GAMMA RAYS \\end{center}  \\begin{tabular}{c|c c c c c c}\\hline \\hline\\\\[-5mm] Inner or Outer&        & Mode  & Signifi- &$\\frac{J_\\gamma(>E)}{J_{\\rm CR}(>E)}$ {\\small at 1$\\sigma$}&\\multicolumn{2}{c}{$E^2 \\frac{dJ_\\gamma(>E)}{dE}$}\\hspace{0.1mm}{\\scriptsize (cm$^{-2}$s$^{-1}$sr$^{-1}$MeV)}\\\\ Galactic Plane& Region & Energy&cance&({$\\small \\equiv 1/\\sqrt{B}$}) & \\hspace*{3.5mm}90\\% CL & 99\\% CL\\\\ (Region of\\hspace{2mm}$l$)&of\\hspace{2mm}$b$&(TeV)&($\\sigma$)&(10$^{-4}$)&\\hspace*{3.5mm}(10$^{-3}$)&(10$^{-3}$)\\\\\\hline &        & 3 &  +2.52&1.95 & 7.6 & 9.6\\\\[-3mm] &$|b|<2^\\circ$ & & && &\\\\[-3mm] I G      &        & 10 &+1.71&2.43 & 3.0 & 4.0\\\\[1mm] {\\small (20$^\\circ\\leq l\\leq 55^\\circ$)}& & 3 &+1.88&1.23 & 4.0 & 5.3\\\\[-3mm] &$|b|<5^\\circ$& & &&& \\\\[-3mm] &        & 10 &+0.81&1.54 & 1.4 & 2.0\\\\\\hline &        & 3  &+0.25&1.16 & 2.1 & 3.3\\\\[-3mm] &$|b|<2^\\circ$ & & &&&\\\\[-3mm] O G      &       & 10 &$-0.63$&1.45 & 0.78 & 1.3\\\\[1mm] {\\small (140$^\\circ\\leq l\\leq 225^\\circ$)}&& 3 &+1.78&0.737& 2.3 & 3.1\\\\[-3mm] &$|b|<5^\\circ$& & &&&\\\\[-3mm] &        & 10 &$-0.66$&0.936 & 0.50 & 0.83 \\\\\\hline \\end{tabular}  \\vspace*{3mm}  \\par In Table 2, the intensity upper limits are also given as 90\\% and 99\\% confidence level (CL), calculated from the above flux ratio and assuming a differential spectral index of 2.4 for the diffuse gamma rays, and utlizing the all-particle energy spectrum of the galactic cosmic rays recently compiled by Apanasenko et al. (2001).  \\par Figure 7 shows the 99\\% CL upper limits thus obtained for diffuse gamma rays from the inner Galactic plane, $20^\\circ \\leq l \\leq 55^\\circ$ and $|b| \\leq 2^\\circ$, at energies around 3 TeV (T3 for Tibet-III) and 10 TeV (T2 for Tibet-II). In this figure the EGRET data \\cite{hunt97}, $315^\\circ \\leq l \\leq 45^\\circ$ and $|b| \\leq 2^\\circ$, are plotted.  The Cherenkov data are also plotted, including Whipple's 99.9\\% CL upper limit above 500 GeV \\cite{lebo00} at the region of $38.5^\\circ \\leq l \\leq 41.5^\\circ$ and $|b| \\leq 2^\\circ$, and HEGRA-IACT's 99\\% CL upper limit above 1 TeV \\cite{aha01} in a similar region of $38^\\circ \\leq l \\leq 43^\\circ$ and $|b| \\leq 2^\\circ$. The theoretical curve calculated by Berezinsky et al. (1993) for $\\pi^\\circ\\rightarrow2\\gamma$ due to the collision of cosmic-ray hadrons with ISM is drawn by a solid curve (BGHS) which is for the region of $20^\\circ \\leq l \\leq 55^\\circ$ and $|b| \\leq 2^\\circ$, deduced from their original paper which is based on the matter density distribution compiled by Fichtel \\& Kniffen (1984) and Bloemen et al. (1984). \\par For inverse Compton gamma rays induced by energetic electrons, the calculation by Porter \\& Protheroe (1996) is shown by dashed curves for source electron spectral indices of 2.0 (PP2.0) and 2.4 (PP2.4) in the direction $l=0^\\circ$ and $b=0^\\circ$, the Galactic center.  Similar theoretical curves calculated by Tateyama \\& Nishimura (2001) are shown by dot-dashed lines with source spectral indices of 2.0 (TN2.0) and 2.4 (TN2.4) in the direction $l= 0^\\circ$ and $|b| \\leq 2^\\circ$. The density distribution of ISPF in their calculations is based on the compilations by Bloemen (1985) and Mathis et al. (1983).  The curves from Tateyama \\& Nishimura (2001) are consistent with the estimations by Porter \\& Protheroe (1996) considering the different region of $|b|$.  The present Tibet data, especially at 10 TeV, together with the HEGRA-IACT data, give the most stringent upper limit for the IC model, although these data can not clearly rule out the IC model with a source electron spectral index of 2.0.  \\placefigure{fig7}  Figure 8 shows the present results as 99\\% CL upper limits (upper bars of T3 and T2) for diffuse gamma rays from the outer part of the Galactic plane, $140^\\circ \\leq l \\leq 225^\\circ$ and $|b| \\leq 2^\\circ$, for the energy ranges around 3 TeV (T3 for Tibet-III) and 10 TeV (T2 for Tibet-II). In this figure, the 90\\% CL upper limits (lower bars) are compared with the CASA-MIA 90\\% CL upper limits, which are based upon muon-poor air showers \\cite{bori98} at about 140 TeV-1.3 PeV, from the OG plane of $50^\\circ \\leq l \\leq 200^\\circ$ and $|b| \\leq 2^\\circ$.  The CASA-MIA data can rule out the IC model with index 2.0 without acceleration energy cutoff, but can not rule out the case with an energy cutoff at 100 TeV.  In this figure, the theoretical curve by Berezinsky et al. (1993) is also shown for $\\pi^\\circ \\rightarrow 2\\gamma$ component (BGHS) in the region of $140^\\circ \\leq l \\leq 225^\\circ$ and $|b| \\leq 2^\\circ$. The IC gamma rays calculated by Porter \\& Protheroe (1996) for the region $50^\\circ \\leq l \\leq 200^\\circ$ and $|b| \\leq 10^\\circ$ for source spectral indices of 2.0 (PP2.0) and 2.4 (PP2.4) are shown, as well as the ones by Tateyama \\& Nishimura (2001) in the region of $l=180^\\circ$ and $|b| \\leq 2^\\circ$ for source spectral indices of 2.0 (TN2.0) and 2.4 (TN2.4). The present data are also not sufficient to rule out the IC model with a spectral index of 2.0.  \\placefigure{fig8}  Next we discuss some considerations regarding our method of data analysis and its results.  First, a difference of the average shower size between gamma-ray and proton initiated showers with the same energy at the Yangbajing site (606 gcm$^{-2}$) will produce different initial proton and gamma-ray energies for the same observed shower size. By employing the subroutine package GENAS of Kasahara \\& Torii (1991), the average incident gamma-ray energy is roughly estimated to be lower by 18\\% at 3 TeV and lower by 23\\% at 10 TeV than that for protons of the same shower size, taking the median zenith angle 24.1$^\\circ$ (atmospheric depth 664 gcm$^{-2}$) into account for generally observed shower events.  The median zenith angles are 27.2$^\\circ$ and 22.8$^\\circ$ (681 gcm$^{-2}$ and 657 gcm$^{-2}$) for the showers arriving from the inner and outer Galactic plane, respectively. Thus, the energy decline rate in gamma-initiated showers is somewhat reduced for the IG plane and magnified a little for the OG plane.  On the other hand, the spectral index of the diffuse gamma rays is probably smaller than the 2.7 of the galactic cosmic rays at energies around 10 TeV.  This produces an opposite offset of the observed gamma-ray energy. which shifts to a higher value, depending on the spectral index.  If the gamma-ray spectral index is 2.4, as assumed in the derivation of the intensity upper limit in Table 2, gamma-ray energy goes up about 10\\% or more.  The effects just described act to offset each other.  Thus we use the observed energies, 3 TeV for the Tibet-III and 10 TeV for the Tibet-II, for the primary energies determined for the generally observed air showers at Yangbajing site.  Second, the sky regions searched for Galactic diffuse gamma rays have a shape like a convex lens along the Galactic plane as shown in Fig. 2.  Such a lens shaped region is inevitable in our method, employing the warped belts along the Galactic plane in equatorial coordinates.  The maximum thickness of the lens is 3.6$^\\circ$ or 8.9$^\\circ$, but the simple mean of the range used is 3.5$^\\circ$ or 8.7$^\\circ$ for IG and 2.7$^\\circ$ or 6.8$^\\circ$ for OG in the 4$^\\circ$ bin and 10$^\\circ$ bin analyses, respectively.  Taking the number density of events into account, the weighted mean thickness becomes 3.5$^\\circ$ or 8.7$^\\circ$ for IG and 2.9$^\\circ$ or 7.2$^\\circ$ for OG, respectively.  As already described, the data in the two 2$^\\circ$ bin belts has been excluded from the off-plane region for OG in order to minimize the influence of the Crab Nebula.  Other strong gamma-ray sources, Geminga ($l=195.03^\\circ, b=4.83^\\circ$) and IC443 ($l=188.83^\\circ, b=3.07^\\circ$) are involved in the on-plane region of the OG plane in the 10$^\\circ$ bin analysis. No TeV gamma-ray signal has, however, been obtained from these candidate sources by any surface air shower arrays or Cherenkov telescopes. Therefore, no correction for the influence of these candidates is necessary.  Third, according to the EGRET data \\cite{hunt97} with $|b| \\leq 2^\\circ$, gamma-ray intensity shows a decline in the galactic longitude region from $l=25^\\circ$ to $65^\\circ$ in IG for every energy ranges, 30-100 MeV, 100-300 MeV, 300-1000 MeV and above 1000 MeV.  This tendency is also seen in SAS-2 and COS-B data (see Bertsch et al. 1993).  The intensity shows a rather flat top in the central range, $330^\\circ \\leq l \\leq 25^\\circ$.  In our analysis, the inner Galactic region is located just at this region of declining intensity. The Whipple \\cite{lebo00} and HEGRA \\cite{aha01} data also lie partially in this region, at around $l=40^\\circ$.  If the intensity decline around 1 GeV is caused by the density distribution of ISPF, it is expected that the gamma-ray intensity shows similar behavior in the TeV region.  We should compare the experimental data with the reduced intensity of about 80 \\% of the theoretical curves, which have been calculated for the central flat top region.  This is the reason that the experimental data are not sufficient to rule out the inverse Compton model with a source spectral index of 2.0.  Theoretical calculations give an indication that, if the source electron spectral index in the 10 GeV to 10 TeV energy region is smaller than 2.4, the diffuse gamma rays in the TeV region are mainly generated by IC scattering. In that case, it is essential to fix the diffuse gamma-ray intensity to determine the source electron spectrum in the Galactic plane.  This can suggest the strength of shock acceleration, and clarifies the electron propagation process in the Galactic disc through a comparison with the direct observation of local electrons, and also gives an estimate of average magnetic field in the source region by examining the consistency with the radio synchrotron intensity.  The extension of the Tibet-III array is expected to be completed by the end of 2002; its effective inner area will become about 1.5 times larger than at present, and 1.15 times larger than the Tibet-II inner area.  If the new Tibet-III array continues for running three years without long suspension, statistics will increase to 4-5 times the present data at both 3 TeV and 10 TeV.  The resulting statistical reduction in the upper limits, by a factor of 2 or more, will be more closely comparable with theoretical models, and can also give a significant upper limit at even higher energies, e.g., at 20 TeV, where the upper limit is relatively more sensitive to the acceleration energy cutoff in the IC model."
    },
    "0208/astro-ph0208145.txt": {
        "abstract": "We present and use new spectra and narrow-band images, along with previously published broad-band images, of stars in the Arches cluster to extract photometry, astrometry, equivalent width, and velocity information. The data are interpreted with a wind/atmosphere code to determine stellar temperatures, luminosities, mass-loss rates, and abundances. We have doubled the number of known emission-line stars, and we have also made the first spectroscopic identification of the main sequence for any population in the Galactic Center. We conclude that the most massive stars are bona-fide Wolf-Rayet (WR) stars and are some of the most massive stars known, having \\Minit~$>$100~\\Msun, and prodigious winds, \\Mdot~$>$10$^{-5}$~\\Msunyr, that are enriched with helium and nitrogen; with these identifications, the Arches cluster contains about 5\\% of all known WR stars in the Galaxy. We find an upper limit to the velocity dispersion of 22~\\kms, implying an upper limit to the cluster mass of 7(10$^4$)~\\Msun\\ within a radius of 0.23~pc; we also estimate the bulk heliocentric velocity of the cluster to be v$_{\\rm cluster,\\odot}\\approx+95$~\\kms. Taken together, these results suggest that the Arches cluster was formed in a short, but massive, burst of star formation about 2.5$\\pm$0.5~\\Myr\\ ago, from a molecular cloud which is no longer present. The cluster happens to be approaching and ionizing the surface of a background molecular cloud, thus producing the Thermal Arched Filaments. We estimate that the cluster produces 4(10$^{51}$)~ionizing photons~s$^{-1}$, more than enough to account for the observed thermal radio flux from the nearby cloud, 3(10$^{49}$)~ionizing photons~s$^{-1}$. Commensurately, it produces 10$^{7.8}$~\\Lsun\\ in total luminosity, providing the heating source for the nearby molecular cloud, L$_{\\rm cloud}\\approx10^7$~\\Lsun. These interactions between a cluster of hot stars and a wayward molecular cloud are similar to those seen in the ``Quintuplet/Sickle'' region. The small spread of formation times for the known young clusters in the Galactic Center, and the relative lack of intermediate-age stars ($\\tau_{\\rm age}$=10$^{7.0}$ to 10$^{7.3}$~yrs), suggest that the Galactic Center has recently been host to a burst of star formation. Finally, we have made new identifications of near-infrared sources that are counterparts to recently identified x-ray and radio sources. ",
        "introduction": "The Arches cluster is an extraordinarily massive and dense young cluster of stars near the Galactic Center. First discovered about 10 years ago as a compact collection of a dozen or so emission-line stars \\citep{cot92,nag95,fig95a,cot95,cot96}, the cluster contains thousands of stars, including at least 160 O stars  \\citep{ser98,fig99a}. \\citet{fig99a} used HST/NICMOS observations to estimate a total cluster mass ($\\gtrsim$10$^4$~\\Msun) and radius (0.2~pc) to arrive at an average mass density of 3(10$^5$)~\\Msun~pc$^{-3}$ in stars, suggesting that the Arches cluster is the densest, and one of the most massive, young clusters in the Galaxy. They further used these data to estimate an initial mass function (IMF) which is very flat ($\\Gamma$~$\\sim-$0.6$\\pm$0.1) with respect to what has been found for the solar neighborhood \\citep[$\\Gamma$~$\\sim-$1.35]{sal55} and other Galactic clusters \\citep{sca98}. They also estimated an age of 2$\\pm$1~\\Myr, based on the magnitudes, colors, and mix of spectral types, which makes the cluster ideal for testing massive stellar-evolution models.  Given its extraordinary nature, the Arches cluster has been a target for many new observations. \\citet{sto02} recently verified a flat IMF slope for the Arches cluster, finding $\\Gamma = -0.8$, using both adaptive optics imaging with the Gemini North telescope and the HST/NICMOS data presented in \\citet{fig99a}. \\citet{blu01} used adaptive optics imaging at the CFHT and HST/NICMOS data (also presented in this paper) to identify several new emission-line stars and estimate an age for the cluster of 2$-$4.5~\\Myr. \\citet{lan01a} detected eight radio sources, seven of which have thermal spectral indices and stellar counterparts, within 10\\arcsec\\ of the center of the cluster. They suggest that the stellar winds from the counterparts produce the radio emission via free-free emission, consistent with earlier indications from near-infrared narrow-band imaging \\citep{nag95} and spectroscopy \\citep{cot96}. In a related study, \\citet{lan01b} argued that the hot stars in the Arches cluster are responsible for ionizing the surface of a nearby molecular cloud to produce the arches filaments, as originally suggested by \\citet{cot96} and \\citet{ser98}, but in contrast to earlier suggestions \\citep{mor89,dav94,col96}. \\citet{zad01} used the Chandra telescope to detect three x-ray components that they associate with the cluster, claiming that hot (10$^7$~K) x-ray emitting gas is produced by an interaction between material expelled by the massive stellar winds and the local interstellar medium.  The Arches cluster has also been the target of several theoretical studies regarding dynamical evolution of compact young clusters.  \\citet{kim99} used Fokker-Planck  models and  \\citet{kim00} used N-body models to simulate the Arches cluster, assuming the presence of the gravitational field of the Galactic Center. They found that such a cluster will disperse through two-body interactions over a 10~\\Myr\\ timescale. \\citet{zwa01a} performed a similar study and found a similar result, although they note the possibility that the Arches cluster is located in front of the plane containing the Galactic Center. Finally, \\citet{ger01} considered the possibility that compact clusters formed outside the central parsec will plunge into the Galactic center as a result of dynamical friction, eventually becoming similar in appearance to the young cluster currently residing there; \\citet{kim02} further consider this possibility.  In this paper, we use new and existing observations to determine the stellar properties of the most massive stars in the Arches cluster. We present astrometry and photometry of stars with estimated initial masses greater than 20~\\Msun\\ (the theoretical minimum mass of O stars), based upon HST/NICMOS narrow-band and broad-band imaging. We also present {\\it K}-band high-resolution spectra of the emission-line stars, based upon Keck/NIRSPEC observations. We couple these data with previously-reported radio and x-ray data to infer stellar wind/atmosphere properties using a modeling code. Finally, we compare our results to those reported in recent observational and theoretical papers. ",
        "conclusions": "In this section, we compare our measurements to those in previous papers and use measurements at other wavelengths to determine the physical parameters of the observed stars. Finally, we discuss how the Arches cluster interacts with its local environment to create heating and ionization of a nearby cloud.  \\subsection{Comparison to Previous Near-infrared Measurements} Table~3 lists over 30 probable emission-line stars, albeit the faintest having relatively weak emission lines; Figure~3a confirms that there are roughly this number of stars with reliable emission-line excesses. This list contains over a factor of two increase in the number of emission-line stars previously identified in the cluster \\citep{blu01,nag95,cot96}. The line and continuum fluxes presented here largely agree with earlier results \\citep{nag95,cot96}. The \\Fnp\\ and \\Fnc\\ fluxes reported in this paper are similar to those reported in \\citet{blu01}, after correcting for differences in the assumed zero points, the fact that we corrected for the difference in extinction at the two narrow-band wavelengths, and that we also corrected for the intrinsic shape of the stellar continuum; in addition, our extinction estimates are higher in many cases than those used in \\citet{blu01}.  The spectra in this paper are consistent with the narrow-band photometry in \\citet{nag95} and \\citet{blu01} and the spectra in \\citet{cot96}, although our high-resolution spectroscopy shows that the photometry is significantly affected by blending of absorption and emission features in P-Cygni profiles.  We confirm the discovery of a new bright emission-line star (\\#5, B22, N9) near the southern edge of the cluster, reported in \\citet{blu01}, and note that it is a counterpart of the x-ray source, ``AR8,'' in \\citet{lan01a}. We also confirm that star \\#16 (B19) is an emission-line star, as suspected by \\citet{blu01}. \\citet{blu01} listed some additional candidate emission-line stars. We confirm that the following stars from that list are, indeed, emission-line stars (their designations in parentheses): \\#15 (B8), \\#27 (B16), \\#17 (B29), \\#10 (B30), \\#10 (B20), \\#13 (B31).  \\subsection{Comparison to X-ray Flux Measurements} \\citet{zad01} reported Chandra X-ray observations of a region including the Arches cluster.  They detected three extended sources, one (A1) near the center of the Arches cluster, another (A2) located to the North and West of the center by about 7\\arcsec\\, and a third weaker source (A3) about 90\\arcsec\\ $\\times$  60\\arcsec\\ in size underlying the first two.  The centroid of source A2 coincides within 1 arcsecond with an emission-line star, \\#9 in Table~3.  The apparent spatial coincidence of the X-ray sources and Arches cluster strongly suggests that the X-ray sources are physically associated with the cluster.  Yusef-Zadeh et al. estimate the total X-ray luminosity between 0.2 and 10 keV to be 3.3, 0.8 and 0.16 (10$^{35}$) ergs s$^{-1}$ for A1, A2 and A3, respectively. They attribute the emission from A1 and A2 to either colliding winds in binary systems or to the winds from single stars interacting with the collective wind from the entire cluster.  The coincidence of source A2 with an emission-line source is very interesting in the context of the latter scenario.  A3, on the other hand, has roughly the characteristics expected from shock-heated gas created by the collisions of the multitude of 1000-\\kms\\ stellar winds emanating from the stars in the rich, dense cluster \\citep{oze97,can00}.  Because the X-ray sources are extended, it is unlikely that they can be attributed to single X-ray binary systems. However, the rough coincidence of source A1 with the core of the cluster raises the possibility that it may be comprised of many relatively weak stellar X-ray sources, binary or single, residing in the cluster core, and unresolved spatially from each other.  \\subsection{Comparison to Radio Flux Measurements} From Table~3 we see that one of the objects analysed in this work, \\#8, has also been detected at 8.5~GHz \\citep{lan01a}. Our derived mass-loss rate is consistent with the observed radio flux (0.23~mJy) only if the outer wind regions are unclumped. Such a behavior for the clumping law has been suggested by \\citet{nug98} from analysis of galactic WR stars. They found that the  observed infrared to radio fluxes of WR stars are well reproduced by a clumping law where the filling factor is unity close to the stellar surface, increasing to a minimum at 5 to 10 R$_*$ and returning again to unity in the outer wind where the radio flux forms. Note, however, that the line fluxes of the weaker lines like  Br$\\gamma$ or \\ion{He}{1} remain unaltered with this new description of the clumping law, but the line fluxes of the strongest lines, such as Pa$\\alpha$, formed in the outer wind  can be significantly reduced. From Table~4 we see that our models are fully consistent with the observed equivalent-width of Pa$\\alpha$.  We consider now the possible correlation between line fluxes and radio-continuum flux analogous to the one discussed above for {\\it K}-band fluxes (see Figure~3a). In principle, we also expect the near-infrared emission line strengths to scale with the free-free emission detected at radio wavelengths \\citep{nug98,lei97}.  However, we do not find such a correlation, as can be seen in Figures~8a,b. A similar result was obtained by \\citet{bie82} for a sample of eight WR stars. We believe this apparent absence of correlation between Pa$\\alpha$ line-strength and radio flux is caused by both  observational and  physical effects. The observational effect is related to the fact that the radio measurements are picking up only the tip of the iceberg, i.e., those stars with the densest winds of the cluster. The physical effect is related to the fact that all three components contributing to EW$_{\\rm 1.87\\micron}$ (\\ion{H}{1}, \\ion{He}{1}, and \\ion{He}{2}) are very sensitive to changes in temperature in the parameter domain appropriate to these objects. Further, both the line and continuum fluxes depend strongly not only on the mass-loss rate but also on the shape of the velocity field and the clumping law. Therefore, such strong dependence of the Pa$\\alpha$ line flux on several stellar parameters introduces a large scatter in the expected line-strength vs radio-flux relationship.  The radio fluxes of the most massive Arches stars are comparable to those of WNL stars, but not to those of O~If$^+$ stars. The WN8 star WR105 \\citep{van01} would emit 0.14 mJy at the distance of the Arches cluster, comfortably within the range of fluxes measured for the Arches stars. Similar values are reported for WR stars in \\citet{bie82}. On the other hand, HD 16691 (O4~If$^+$) emits 0.3~mJy at 4.9~GHz, according to \\citet{wen95}, implying an expected flux of 1.7~$\\mu$Jy at the distance of the Arches cluster, assuming that the star has a parallax of 1.7~mas \\citep{per97}. The expected flux is two orders of magnitude below the flux levels of the brightest Arches stars \\citep{lan01a}. No doubt, this difference is due to the relatively low mass-loss rate for HD 16691, about 1/20 of that of the bright Arches stars. A similar trend can be seen in Figure~6 where the emission lines in the spectra of HD 16691 are shown to be much weaker than those in the spectra of the Arches emission-line stars. Again, weak winds produce weak emission lines and weak free-free emission.  Finally, we report several additional radio sources having emission-line star counterparts. They were found by comparing the radio continuum contour plot in \\citet{lan01a} with the difference image in Figure~2; they are marked by squares in this figure. We have designated the four near-infrared counterparts to these newly identified radio sources in Table~3.  \\subsection{Evolutionary Status of the Massive Stars in the Arches Cluster} The emission-line stars appear to contain significant amounts of hydrogen, while also exhibiting considerable helium content. We believe that this can be explained by the most massive stellar models in \\citet{mey94}. For the brightest 10 or so stars in Table~3, the observations can be fit by these models for \\Minit$\\gtrsim$120~\\Msun\\ stars that have evolved to cool temperatures while retaining hydrogen. In particular, star \\#8 can be fit by a \\Minit$\\sim$120~\\Msun\\ star with solar abundance, standard mass-loss rates, age of 2.4~\\Myr\\ to 2.5~\\Myr, and present-day mass of 72~\\Msun\\ to 76~\\Msun\\ \\citep{sch92}.  \\subsection{Relation to the Nearby Molecular Cloud (M0.10+0.03)} It appears that the Arches cluster heats and ionizes the surface of M0.10+0.03, the nearby molecular cloud \\citep{ser87,bro84}, given that the cluster can easily provide the necessary flux to account for the infrared emission and recombination-line flux from the cloud.  The relative heliocentric velocity between the cluster stars (+95$\\pm8$~\\kms) and the ionized gas on the surface of the cloud ($-$20 to $-$50~\\kms) suggests that the physical association is accidental and that the cluster stars are ionizing the surface of the cloud. This difference in velocity is reminiscient of that observed between the Quintuplet (+130~\\kms) \\citep{fig95a,fig99a} and the Sickle cloud, M0.10+0.03 (+30~\\kms) \\citep{lan97}. In both cases, it appears that young clusters happen to lie near molecular clouds whose surfaces are ionized by the photons from the hot stars in the clusters. The following shows that the ionizing flux and energy required to heat the cloud can be provided by the Arches cluster. Note that a differential velocity of 100~\\kms\\ would produce a relative drift of 100~pc in 1~\\Myr, a distance that would bring a cluster within the vicinity of a few clouds, given the spatial distribution of clouds in the central few hundred parsecs.  \\subsubsection{Ionization and Heating} Even before the discovery of the Arches cluster, \\citet{ser87}, \\citet{gen90}, and \\citet{miz94} suggested that the Thermal Arched Filaments are photoionized by nearby hot stars. After the discovery of the cluster, many authors considered the possibility that the cluster is ionizing the cloud. One problem with this idea is the fact that the filaments are very large, and have roughly constant surface brightness and excitation conditions \\citep{eri91,col96}, indicating that the ionizing source is either evenly distributed over many parsecs or is relatively far away. Given the new ionizing flux estimates in this paper and in \\citep{ser98}, the cluster would produce enough flux to account for the filaments, even if 20~pc away, far enough to allow for the even illumination that is observed. Indeed, \\citet{tim96} predicted this result, and \\citet{lan01b} present a detailed analysis that confirms it.  \\citet{cot96} give estimates for the total ionizing flux of 2$-$5(10$^{50}$) photons s$^{-1}$, depending on whether one models the spectral energy distributions for the emission-line stars with \\citet{kur79} atmospheres or blackbody functions; however, this estimate includes only flux from the dozen or so emission-line stars that were known at the time. Nonetheless, the results of this paper suggest that the cluster ionizes the cloud.  The heating in the cloud produces an infrared luminosity of 10$^7$~\\Lsun\\ \\citep{mor95}. Assuming a covering fraction of $\\approx$10\\%, we find that the Arches cluster can deliver about this much luminosity, within a factor of two.  \\subsubsection{Location along the line of sight} Given the Br$\\gamma$ flux from the filaments measured by \\citet{fig95a}, we know that the line emission is extincted by about \\AK$\\sim$3. This implies that the filaments are on the near side of the cloud, since such an extinction corresponds only to the typical foreground extinction to the Galactic Center, and precludes any substantial additonal extinction. This information leads to the conclusion that the Arches cluster is on the near side of, and is approaching, the wayward molecular cloud that is moving in opposition to the bulk motion of stars and gas around the Galactic Center \\citep{mcg89}, consistent with the geometry described in \\citet{lan01b}.  \\subsection{Dynamical Evolution and Uniqueness of the Arches and Quintuplet Clusters} The temporal coincidence of the star formation events that produced the massive clusters in the Galactic Center, and the lack of older red supergiants, suggest that the Galactic Center has been host to a recent burst of star formation. \\citet{kim99} and \\citet{kim00} predicted that compact young clusters in the Galactic Center would evaporate on short timescales, i.e.\\ a few \\Myr. \\citet{zwa01a} argue that other clusters similar to, yet somewhat older than, the Arches and Quintuplet clusters exist in the central 100~pc. This argument is based upon a dynamical analysis which predicts that such clusters evaporate after 55~\\Myr, and further that the clusters' projected surface number density in stars drops below the limit of detectability in a few \\Myr. The statement that members of dispersed clusters could have gone undetected is incorrect. Such stars would easily be detectable for their extreme brightness, i.e. there would be hundreds of stars as bright as IRS~7 (in the central parsec) strewn about the central 100~pc for each Arches/Quintuplet-like cluster between the age of 5 and 30~\\Myr. Given the claim in Portegies-Zwart et al.\\ about the expected number of ``hidden'' young clusters in the central hundred parsecs, we would expect to see of order ten thousand red supergiants in this region. Only a few are seen, as demonstrated by surveys for such stars \\citep{fig95a,cot95}.  Portegies-Zwart et al.\\ suggest that clusters could be ``hiding'' near bright stars due to limitations in dynamic range, but these arguments are specious, since the dynamic range of array-based detections are obviously not limited by the digitization of a single read if multiple coadds are used (e.g.\\ \\citet{fig99a} reach a dynamic range of over 10$^5$). Thus in agreement with Kim et al., we conclude that the clusters must disperse rapidly  \\subsection{Comparison to NGC~3603 and R136 in 30 Dor} The Arches cluster is similar in age and content to NGC~3603 and R136 in 30 Dor, and is surrounded by a giant \\ion{H}{2} region as is R136. In contrast to these clusters, we do not see WN5h or WN6h stars \\citep{cro98}, suggesting that the Arches cluster is older. This is consistent with our age determination, as suggested by other means described earlier.  While our spectra exhibit no primary diagnostic lines to estimate metallicity, we may use our estimates for the helium and nitrogen abundances in object \\#8 in conjunction with the evolutionary model for 120~\\Msun\\ to infer metallicity. For Z(He)=0.7, the models predict the star to have already reached its maximum nitrogen surface mass fraction. Hence, we may compare our derived nitrogen mass fraction, Z(N)=0.016, with the evolutionary models values at different metallicities \\citep{sch92}. We see that this value is met for solar metallicity. A more detailed analysis of the metallicity of the Arches stars will be presented in \\citet{naj02}."
    },
    "0208/astro-ph0208277_arXiv.txt": {
        "abstract": "The giant jets represent a fundamental trace of the historical evolution of the outflow activity over timescales of $\\sim 10^{4}$ yr, i.e. a timescale comparable to the accretion time of the outflow sources in their main protostellar phase. The study of such huge jets provides the possibility of retrieving important elements related to the life of the outflow sources. In this paper, we study the role of precession (combined with jet velocity-variability and the resulting enhanced interaction with the surrounding environment) as a deceleration mechanism for giant jets using a numerical approach. This thesis was proposed for the first time by Devine et al. (1997) but it could not be numerically explored until now because it is intrinsically difficult to reproduce, at the same time, the large range of scales from $\\sim 100$~AU up to a few parsecs. In the present paper, we obtain predictions of H$\\alpha$ intensity maps and position-velocity diagrams from 3D simulations of the giant HH~34 jet (including an appropriate ejection velocity time-variability and a precession of the outflow axis), and we compare them with previously published observations of this object. Our simulations represent a step forward from previous numerical studies of HH objects, in that the use of a 7-level, binary adaptive grid has allowed us to compute models which appropiately cover all relevant scales of a giant jet, from the $\\sim 100$~AU jet radius close to the source to the $\\sim 1$~pc length of the outflow. A good qualitative and quantitative agreement is found between the model predictions and the observations, indicating that a precession of the jet axis can indeed be the probable cause of the deceleration of the giant jets. Moreover, we show that a critical parameter for obtaining a better or worse agreement with the observations is the ratio $\\rho_{j}/\\rho_{a}$ between the jet and the environmental densities. The implications of this result in the context of the current star formation models are discussed ",
        "introduction": "\\label{intr} Herbig-Haro (HH) objects are the optical manifestations of outflows from young stellar objects (YSOs). Following the discovery of jet-like structures in HH objects (Dopita et al. 1982; Mundt \\& Fried 1983), many of these HH jets were observed in the Orion (Reipurth et al. 1986; Mundt et al. 1987; Reipurth 1989a, Reipurth 1989b) and Taurus (Mundt et al 1988) star formation regions. These objects present a characteristic morphology of aligned knots extending over $\\sim 0.3$~pc. It appears that the kinematics and morphologies of these jets depend simultaneously on the time-dependent nature of the outflow activity and on the interaction of the hypersonic flows with the surrounding interstellar medium.  It has recently been discovered (Bally \\& Devine 1994; Reipurth et al. 1997; Devine et al. 1997) that a few HH jets extend over distances of a few parsecs. For example, HH~111 shows a total extent of $\\approx 7.7$~pc, HH~34 of $\\approx 3$~pc and HH~355 a total extent of $\\approx 1.55$~pc.  Apart from their alignments, the main evidence that the knots belong to the same jet (and not to other, smaller outflows) is their kinematic association with red- and blue-shifted bipolar lobes. From the radial velocity, the proper motions and the distance of the knots from the source it has been possible to estimate a typical dynamical age of $\\sim 10^4$~yr for these ``giant jets''.  An important characteristic of the giant jets is that they appear to slow down for increasing distances from the outflow source. This effect is seen in the HH~34 (Devine et al. 1997) and in the HH~111 giant jets (Reipurth et al. 1997; Rosado et al. 1999). The present paper is concerned with the possible theoretical interpretations of the deceleration effect. This is a critical point in the determination of the jet's age and also in the identification of the physical properties of the central engine that feeds the outflows.  The potential causes that might produce such a deceleration can be divided into two categories~: \\begin{itemize} \\item internal causes, i.e. that the mechanism is intrinsically related to the properties of the outflow, for example to a temporal variability of the ejection velocity, \\item external causes, i.e. that the deceleration is due to the drag effect resulting from the interaction of the hypersonic flow with the interstellar medium. \\end{itemize}  Previous studies have considered both of these possibilities. Cabrit \\& Raga (2000) considered the case of an ejection velocity which monotonically grows as a function of time, and tried to fit the observed, position-dependent jet velocity with different, parameterized forms of the ejection velocity time-dependence. These authors concluded that the only way to fit the observed kinematics of the HH~34 giant jet is with an ejection velocity that slowly increases over $\\sim {5\\times 10^{4}}$~yr, followed by a very strongly increasing ejection velocity over the last $\\sim 10^4$~yr.  Cabrit \\& Raga (2000) argued that this very dramatic increase in the ejection velocity at recent times appeared to be unlikely, and then studied an alternative scenario. Following the idea proposed by Devine et al. (1997), they considered the knots along the HH~34 giant jet as scattered ``bullets'' resulting from the combination of an ejection velocity variability and a precession of the jet axis. Raga \\& Biro (1993) have carried out a theoretical study of this kind of flow, obtaining an analytic description of this ``machine gun jet'' flow and comparing this model with a numerical simulation of a radiative, 2D ``slab'' jet with a time-dependent ejection velocity and direction.  One could argue that the jet/counterjet symmetry observed in the HH~34 giant jet (see Devine et al. 1997) goes against the ``environmental drag'' scenario for the deceleration of this object. As there is no reason to suppose that the environments within which the jet and the counterjet are traveling have identical densities, one would think that the drag would introduce asymmetries between the two outflow lobes. However, the deceleration induced by the environmental drag is proportional to ${\\rho_a}^{1/3}$ (where $\\rho_a$ is the environmental density, see Cabrit \\& Raga 2000), so that these asymmetries might not be so important.  Also, de Gouveia Dal Pino (2001) presented 3D simulations (done with the Lagrangian SPH method) of HH~34 assuming a sinusoidal ejection velocity variability of the kind used by Raga \\& Noriega-Crespo (1998) and studying the two cases of a pressured and overpressured jet without considering the precession contribution. To reproduce the large spatial working surface structures the author used a half-amplitude of the velocity modulation of $\\approx 100$~km~s$^{-1}$ and a period of $760$ yr. The author obtained encouraging results showing that a deceleration of the jet velocity was obtained from the model (lower decelerations being produced for initially overpressured jets than for pressure matched jets). She concluded from the simulations that the deceleration was related to the temporal velocity variability of the jet at injection and was mainly caused by progressive momentum transfer sideways into the surrounding medium by the expelled gas from the travelling working surfaces. She also found that a steady state jet, with similar initial conditions to those of the pulsed jet, experienced, on the contrary, an initial acceleration followed by a constant velocity propagation regime, which was an additional  indication that the primary source of deceleration in giant flows could not be attributed to simple breaking of the jet head against the external medium. These models were run over a distance of only $0.3$ pc, well below the size of the giant HH~34 jet (see Devine et al. 1997).  In the present paper, we carry out a numerical study of the effect of a precession of the outflow axis on the deceleration of a giant jet. The existence of such a phenomenon is suggested by the observed morphology of the flow, which shows evidence of long period (of the order of $10^4$ yr) precession (Devine et al. 1997). The precession is generally ascribed to tidal forces produced by a companion in a binary or multiple system. Even though in the case of HH~34 the binary source has noy yet resolved, there are elements that indicate that this source could be a binary such as the discovery of a second outflow (HH~534) emanating from the source as well as the abrupt change of direction of the jet axis near the knots B (Reipurth et al. 2002). Less evident is the explanation of the existence of a precession period of the order of $10^{4}$ yr.  As reported in Terquem et al. (1999), the precession period ($\\tau_{p}$) depends on the orbital parameters and the spatial extent of the accretion disk. The precession period is generally at least one order of magnitude larger than the orbital period. Therefore, to justify such a large $\\tau_{p}$ we need a value of the ratio between the accretion radius disk $R$ and the orbital radius $r_{0}$ ($\\sigma$ $=$ $R$/$r_{0}$ ) of the order of $10^{-3}$ (see Masciadri \\& Raga, 2002).  Reipurth (2000) has also proposed that perturbations on an accretion disk due to close passages (i.~e., at perihelion) of a binary companion in an elliptical orbit could be responsible for producing a time-variability in the ejection of the outflow. It is of course unclear whether or not such a mechanism could produce a variability in the ejection velocity such as the one included in our jet models.  We carry out the simulations of only one of the two lobes of HH~34 (with and without precession) conserving the same geometrical and physical parameters. However, our numerical simulations could correspond to any of the two lobes of the HH~34 giant jet. Actually, our simulations are made to reproduce the morphology of the northern lobe of the HH~34 outflow, and in order to compare the predicted maps with the southern lobe it is necessary to carry out a point reflection of the predicted maps with respect to the position of the outflow source. This point symmetry of a precessing jet/counterjet system is clearly seen in the observations of the HH~34 giant outflow (Devine et al. 1997).  We underline that this study cannot be applied in a simple way to other giant jets. The two lobes of HH~111, for example, show quite straight paths. HH~355, on the contrary, seems to precess with a half-opening angle of $\\sim 13^\\circ$ and a period of $\\sim 1500$-$2000$ yr (Reipurth et al. 2002). Further work should be carried out to study the properties of these other giant flows.  We find that the impact of the precession on the deceleration mechanism is quite considerable, and that the dynamical age of the jet grows by $\\sim 3000$~yr when including a precession. We also find that H$\\alpha$ intensity maps and position-velocity diagrams obtained from models with precession reproduce the observations of the HH~34 giant jet in a qualitatively successful way (an agreement which is not found for models without precession).  We add that, from a numerical point of view it is not a simple exercise to reproduce the evolution of outflows over such a large spatial and temporal extent particularly given the small initial radius of the beam. Indeed, in order to cover the whole domain of $1.5$ pc, previously published simulations used $r_{j}$ $=$ $10^{16}$ cm (de Gouveia Dal Pino 2001) and $r_{j}$ $=$ $10^{17}$ cm jet radii (Cabrit \\& Raga 2000), which are $1$-$2$ order of magnitude greater than the width of the jet as observed in HST images (Reipurth \\& Raga 1999). One of the goals of our study is to simulate the HH~34 giant jet over its full extent ($\\sim$ $1.5$ pc) using the correct $r_{j}$ $=$ $3\\times10^{15}$ cm initial jet radius corresponding to $0\\arcsec.4$ at $460$ pc. We underline that the radius is measured at $\\sim 10\\arcsec$ distance away from the HH~34 source.  In Section \\ref{param}, we describe the parameters and the ejection velocity time-variability used in our models of the HH~34 jet. In Section \\ref{numer} the numerical simulations are discussed, and H$\\alpha$ maps and position-velocity diagrams predicted from models with and without precession are presented and compared with the corresponding observations of the HH~34 giant jet. In Section \\ref{conc} we summarize the conclusions of this study. ",
        "conclusions": "\\label{conc}  In this paper, we explore the role of a precession of the outflow axis on the deceleration of giant HH jets. In particular, we try to simulate the HH~34 giant jet, and we compare the results of our numerical simulations with previously published observations of this object.  Our simulations represent a step forward from previous numerical studies of HH objects, in that the use of a 7-level, binary adaptive grid has allowed us to compute models which appropriately cover all of the relevant scales of a giant jet, from the $\\sim 100$~AU jet radius close to the source to the $\\sim 1$~pc length of the outflow. Previous simulations of giant jets either did not cover the length of a real flow (de Gouveia dal Pino 2001), or else had a very large jet radius (Cabrit \\& Raga 2000 and de Gouveia dal Pino 2001).  A set of simulations done with and without precession of the outflow axis are presented, and predictions of H$\\alpha$ maps, proper motions and radial velocities are compared with the observations of Devine et al. (1997). The principal conclusions of our study are the following~: \\begin{itemize} \\item we see that the morphology and kinematics of the HH~34 giant jet can be reproduced with a model of a jet with a sinusoidal ejection velocity variability (with a mean velocity of 300~km~s$^{-1}$, a half-amplitude of 100~km~s$^{-1}$ and a period of 1010~yr) and a precession of the outflow axis (with a half-angle of 6$^\\circ$ and a 12000~yr period). The simulated and measured H$\\alpha$ maps and radial velocities show a good qualitative as well as quantitative agreement, \\item comparing simulations done with and without precession we showed that the simple precession can give differences in the jet age estimations of the order of about $3000$ yr. This proves that the drag effect produced by the external medium on the working surfaces is not negligible with respect to the $\\sim 10^{4}$ yr dynamical timescale of the outflow. \\item we proved that the $\\rho_{j}/\\rho_{a}$ ratio is a critical parameter in the determination of the deceleration rate of the jet, and that it has to be properly adjusted in order to be able to fit the observed properties of a giant HH flow. \\end{itemize}  The results of our study do not exclude that, in other giant jets, a correct jet deceleration could be attained without the precession. More models should be tested with different $\\rho_{j}/\\rho_{a}$ ratios to obtain more definite conclusions. Besides this, in the case of HH~34, the results seem to indicate that the precession has a fundamental role in the deceleration mechanism. We underline that the non-precessing, velocity-variable jet of Model C also decelerates, however it has radial velocities which are larger than the observed ones in HH34 (Devine et al. 1997). Previous 3D modeling of time-variable, non-precessing giant outflows (de Gouveia Dal Pino 2001) had also detected jet deceleration that reproduced the observations only qualitatively. Therefore, the results of the present work indicate that in the case of HH34, it is the combined effect of both, the jet temporal velocity variability and the precession (along with the appropriate choice of the ratio $\\rho_j/\\rho_a$) that reproduces the observed deceleration pattern in that source (as in Model B). One could argue that Model C with different $\\rho_{j}/\\rho_{a}$ ratios or velocity variability law could reproduce the correct deceleration. On the other hand we observe that a change of the $\\rho_{j}/\\rho_{a}$ ratio seems to produce a modification in the deceleration rate (Fig.~\\ref{mapaha_tot1} right hand side). A different velocity variability would produce a different distribution of the working surfaces along the jet trajectory and an H${\\alpha}$ map characterized by a different emission. We also recall that the emission of the H${\\alpha}$ maps strongly depends on the $\\rho_{j}/\\rho_{a}$ ratio. A smaller $\\rho_{j}/\\rho_{a}$ ratio in Model C would probably reproduce a H${\\alpha}$ map characterized by an emission level which is too low (see Section \\ref{alfa}). It therefore appears that these high radial velocities are due neither to an incorrect $\\rho_{j}/\\rho_{a}$ ratio nor to a velocity variability law different from the one that we have considered.  In our simulations of the HH~34 giant jet, all of the structure of the outflow is due to a velocity variability with a single, sinusoidal mode and a well ordered precession. In our model, both the velocity variability and the precession last for all of the life of the outflow. Furthermore, the ejection velocity variability that we have used agrees with the one determined by Raga \\& Noriega-Crespo (1998) for the region between the source and HH~34S, so that there is evidence that this variability is continuing to at least quite close to the present time.  The existence of a precession is an indication that the source belongs to a binary or multiple system. This is in agreement with Reipurth (2000), who argued that the sources of giant HH jets are binary or multiple systems.  However, our results seem to be less consistent with the thesis proposed by Reipurth (2000) that giant jets are fossil records of the evolution of orbital motions in disintregating multiple systems. This process has three distinct phases~: (1) a non-hierarchical state (called {\\it interplay}) in which the multiple system performs a random motion, (2) a close triple approach in which a close binary is formed and a low mass star or embryo moves over to a larger orbit and (3) an ejection phase in which the low mass embryo is ejected from the nucleus of the system.  In our models, the structure of the HH~34 giant jet is reproduced without the need of having different properties of the ejection at different times, reflecting qualitative changes in the outflow source resulting from the three phases of a disintegrating multiple system (see above). Therefore, we conclude that if the HH~34 system does correspond to an outflow history with distinct phases, the evidence for this appears to have been lost in the complexities of the interaction between the jet and the surrounding environment."
    },
    "0208/astro-ph0208088_arXiv.txt": {
        "abstract": "The timing of 503~solar flares observed simultaneously in hard \\mbox{X-rays}, soft X-rays and H$\\alpha$ is analyzed. We investigated the start and the peak time differences in different wavelengths, as well as the differences between the end of the hard X-ray emission and the maximum of the soft \\mbox{X-ray} and H$\\alpha$ emission. In more than 90\\% of the analyzed events, a thermal preheating seen in soft X-rays is present prior to the impulsive flare phase. On average, the soft X-ray emission starts 3~min before the hard X-ray and the H$\\alpha$~emission. No correlation between the duration of the preheating phase and the importance of the subsequent flare is found. Furthermore, the duration of the preheating phase does not differ for impulsive and gradual flares. For at least half of the events, the end of the nonthermal emission coincides well with the maximum of the thermal emission, consistent with the beam-driven evaporation model. On the other hand, for $\\sim$25\\% of the events there is strong evidence for prolonged evaporation beyond the end of the hard \\mbox{X-rays}. For these events, the presence of an additional energy transport mechanism, most probably thermal conduction, seems to play an important role. ",
        "introduction": "In this paper we investigate the timing behavior of solar flares, observed simultaneously in hard X-ray (HXR), soft X-ray (SXR) and H$\\alpha$ emission. The main items we address are: a) the flare onset in different wavelengths, b) the timing of different flare emissions with respect to the electron-heated chromospheric evaporation model. Based on a sample of 503~solar flares observed simultaneously in HXR, SXR and H$\\alpha$, we aim to determine how common is the preheating prior to the impulsive phase and is it different in different types of events. Furthermore, we will investigate whether the electron beam-driven evaporation model is consistent with the majority of solar flares, considering the predicted coincidence between the end of the nonthermal (HXR) and the maximum of the thermal (SXR and H$\\alpha$) flare emission.  Electron beam-driven evaporation is usually supposed to be a dominant energy transport mechanism during solar flares. According to the thick-target model (Brown, 1971), the HXR emission is electron-ion bremsstrahlung produced by electron beams encountering the dense layers of the lower corona, the transition region, and the chromosphere. The model assumes that only a small fraction of the energy of the nonthermal electrons is lost through radiation; most of the energy is transferred to heating of the ambient plasma. Due to the rapid deposition of energy by the electron beams, the energy cannot be radiated away sufficiently fast. Thus, a strong pressure imbalance develops, and the heated plasma explosively expands up into the corona in a process known as chromospheric evaporation (Antonucci, Gabriel and Dennis, 1984; Fisher, Canfield and McClymont, 1985; Antonucci {\\it et al.}, 1999). The hot dense plasma that has been convected into the corona gives rise to enhanced soft SXR emission via thermal bremsstrahlung. Thus, the model predicts that the hard X-ray emission is directly related to the flux of the accelerated electrons, whereas the soft X-ray emission is related to the accumulated energy deposited by the same nonthermal electron population up to a given time.  However, the model is questioned by several authors. E.g., Simnett (1986) and Plunkett and Simnett (1994) proposed that protons accelerated at the energy release site, not electrons, are the primary energy carrier in solar flares  (see also the review by Simnett, 1995). Another controversial issue is the role of thermal conduction versus electron beams (e.g., Doschek {\\it et al.}, 1989). Furthermore, it has been questioned (e.g., Feldman, 1990) whether chromospheric evaporation is a ``real'' phenomenon at all (see the reviews by Doschek {\\it et al.}, 1989; Antonucci {\\it et al.}, 1999).  From high time resolution observations it is known that during the impulsive phase, the fast time structures seen in H$\\alpha$ are correlated with the hard X-rays and microwaves (e.g., K\\\"ampfer and Magun, 1983; Kurokawa, Takakura and Ohki, 1988; W\\\"ulser and Marti, 1989; Trottet {\\it et al.}, 2000). This suggests that nonthermal particle beams directly heat the chromospheric plasma, giving rise to the impulsive H$\\alpha$ emission. Numerical simulations of the chromospheric response to pulse beam heating on time scales of less than 1~s have been performed by Heinzel (1991). On the other hand, the H$\\alpha$ emission during the main phase of a flare is likely due to heating of the chromosphere by thermal conduction from the hot SXR emitting plasma in the flare loop (e.g., Phillips, 1991). Veronig {\\it et al.} (2001) have shown that there is a distinct correlation between the SXR flux, $F_{\\rm SXR}$, and the H$\\alpha$ area, $A_{\\rm H\\alpha}$, at the time of the flare maximum, close to the relation: $F_{\\rm SXR} \\propto (A_{\\rm H\\alpha})^{3/2}$. This means that the measured H$\\alpha$~area can be understood as an intersection at chromospheric level of the volume of evaporated plasma responsible for the enhanced SXR emission.  Under the assumption that exclusively accelerated electrons contribute to the evaporation and that the cooling time of the plasma is significantly longer than the impulsive HXR emission (see also Dennis, 1991), it is expected that the SXR as well as the H$\\alpha$ emission do not further increase after the HXR emission, i.e. the electron input, has stopped. In a previous paper (Veronig {\\it et al.}, 2002a), the timing of the SXR peak emission relative to the end of the HXR emission, has been investigated. In this study we additionally include H$\\alpha$ measurements as a further indicator of thermal flare emission. Considering H$\\alpha$ observations complementary to soft X-rays is of particular interest, since, as shown by McTiernan, Fisher and Li (1999), the temporal behavior of the SXR emission depends on the temperature response of the SXR detector used.  There are several papers that investigate the start of soft X-rays relative to hard X-rays. Kahler (1979) did not find systematic brightenings in soft X-rays before the impulsive flare phase. On the other hand, Machado, Orwig and Antonucci (1986) and Schmahl {\\it et al.} (1989) reported frequent strong SXR emission before the impulsive phase. These authors have shown that, on average, the SXR emission precedes the onset of the HXR emission by $\\sim$2~min. The fact that a gradual rise of SXR emission is present before the onset of the hard X-rays suggests a thermal origin of the first phase of a flare (e.g., \\v{S}vestka, 1976; Schmahl {\\it et al.}, 1989). Such a gradual heating before the impulsive particle acceleration is supposed to be related to the re-arrangement of the magnetic fields preceding a flare. Furthermore, this initial phase may also determine the subsequent impulsive phase of the flare. For example, Emslie, Li and Mariska (1992) have shown that preheating of the flare atmosphere influences the subsequent evaporation process. We stress that this initial phase of the flare observed in soft X-rays has to be discriminated from SXR precursors, which occur several tens of minutes before the actual flare and not necessarily at the flare site. Furthermore, there is a distinct fall of intensity between the precursor event and the flare itself (e.g., Tappin 1991). In this paper we do not concentrate on SXR precursors, whose existence is still a controversial issue, but refer to the papers by Webb (1983), Tappin (1991), F\\'{a}rn\\'{\\i}k, Hudson and Watanabe (1996), and F\\'{a}rn\\'{\\i}k and Savy (1998).  Statistical studies of the timing behavior of solar flares observed at different wavelengths have been presented in several papers. However, in general only the emissions at two wavelengths are compared. Investigations on the timing of the SXR and H$\\alpha$ emission during a flare have been carried out by Thomas and Teske (1971), Datlowe, Hudson and Peterson  (1974), Falciani {\\it et al.} (1977), Zirin {\\it et al.} (1981), Verma and Pande (1985a) and Veronig {\\it et al.} (2002b). The reported results are quite contradictory. There is neither a consensus whether the SXR emission starts before the H$\\alpha$ emission or vice versa, nor in which succession the peaks occur. These conflicting results are presumably related to the fact that apart from the studies by Thomas and Teske (1971), Datlowe, Hudson and Peterson (1974) and Veronig {\\it et al.} (2002b), rather small data sets have been used for the analysis, which may cause large statistical errors. A statistical study of the timing of the HXR emission relative to the H$\\alpha$ emission has been performed by Verma and Pande (1985b), finding that most impulsive flares produce HXR emission up to 1~min before and up to 2~min after the start of the H$\\alpha$ emission. For previous studies of the relative timing of the SXR and HXR flare emission we refer to Machado, Orwig and Antonucci (1986) and Schmahl {\\it et al.} (1989), and references therein. ",
        "conclusions": "We found in almost all of 503~studied flares ($\\sim$90\\%) a thermal preheating of the flare atmosphere, seen in soft X-rays prior to the impulsive particle acceleration. On average, the SXR emission starts $\\sim$3~min before the HXR and the H$\\alpha$ emission. The duration of the preheating phase is not related to the importance of the subsequent flare. Moreover, there is no evidence that the duration of the preheating phase differs for impulsive and gradual flares. The H$\\alpha$ and the HXR emission start preferentially simultaneously, indicating that the onset of the H$\\alpha$~emission is related to the impulsive phase of particle acceleration.   The thermal (SXR and H$\\alpha$) emissions predominantly peak after the nonthermal (HXR) emission. This outcome provides a necessary condition for the electron-heated chromospheric evaporation model, in which the thermal flare emission is caused by thermalization of the same nonthermal electrons that are emitting in hard X-rays. For more than half of the events, the end of the nonthermal emission coincides well with the maximum of the thermal emission ($\\Delta t \\le 1$~min), as predicted from the beam-driven evaporation model. However, for  $\\sim$25\\% of the events, there is strong evidence for a prolonged evaporation beyond the end of the nonthermal emission. On average, these events are characterized by a weak and short HXR emission. The extended thermal emission beyond the hard \\mbox{X-rays} suggests the presence of an additional energy transport mechanism from the energy release site other than particle beams, most probably thermal conduction. Events, in which the thermal emission is found to peak prior to the end of the hard \\mbox{X-rays}, are preferentially of long duration. This effect can be explained, within the electron-heated evaporation model, by instantaneous cooling of the plasma that dominates over the energy supply by evaporated material during the decay phase of long-duration events."
    },
    "0208/astro-ph0208041_arXiv.txt": {
        "abstract": "We study the formation of the Milky Way's halo in a $\\Lambda$CDM cosmology by scaling down a high resolution simulation of the formation of a cluster of galaxies. We determine how much phase-space substructure is left over from the objects that merge to build up the present galaxy. We study the debris streams originating from such objects and find that their evolution can be explained simply in terms of the conservation of phase-space density.  Analysing the mass growth history of our halo we find that its inner regions have been in place for more than 10 Gyr, but that the growth of the halo as a whole is more gradual, in agreement with other high resolution simulations of dark-matter halos. Recent accretion contributes to the inner 10 kpc of the halo only at the 10$^{-4}$ level. Finally we determine the number of dark-matter streams as a function of distance from the centre of the halo. In the equivalent of the ``Solar vicinity'', we find that the dark-matter is smoothly distributed in space, and that the velocity ellipsoid is formed by hundreds of thousands of streams, most of which have velocity dispersions of the order of 1 \\kms~ or less. ",
        "introduction": "Over the last twenty years, the hierarchical paradigm has emerged as the standard model to describe the formation of structure in the Universe. As embodied in the current ``concordance\" $\\Lambda$CDM model it appears to be consistent with a very wide range of cosmological data ranging from fluctuations in the Cosmic Microwave Background through the structure of Ly$\\alpha$ forest absorption in QSO spectra and the gravitational shear induced by dark-matter structures to the observed large scale structure in the galaxy distribution. An important characteristic of such models is that they are based on a set of well-defined and testable assumptions. This renders possible the detailed modelling of the formation and evolution of galactic systems, and a later comparison to observations of the properties of these systems as a function of environment or redshift (e.g. Diaferio et al. 2001; Benson et al. 2001; Somerville, Primack \\& Faber 2001).  It is also possible to test the hierarchical paradigm on our Galaxy (e.g. Hern\\'andez, Avila-Reese \\& Firmani 2001).  Several groups (Moore et al. 1999; Klypin et al. 1999; Klypin, Zhao \\& Somerville 2002) have focused on the properties of dark halos, hoping to constrain the nature of dark-matter. These groups performed high resolution simulations of a galactic size halo in CDM cosmologies. They confirmed earlier analytic claims (Kauffmann, White \\& Guiderdoni 1993) that the predicted number of satellites exceeds the known population in the Local Group by a factor of ten.  Some attempts have been made to account for the disagreement, by changing the nature of the dark-matter (Spergel \\& Steinhardt 2000; Bode, Ostriker \\& Turok 2001), by modifying the initial power spectrum of density fluctuations (Kamionkowski \\& Liddle 2000) or by taking into account the effects of a reionising background which may inhibit star formation in the smallest mass halos (Kauffmann et al. 1993; Bullock, Kravtsov \\& Weinberg 2001; Benson et al. 2002). The recent results by Kleyna et al. (2002) on the mass distribution in the Draco dSph (see also Mateo 1997 and Lokas 2001 for a similar study on Fornax) favour an astrophysical explanation since the actual circular velocities of the other satellite galaxies of the Milky Way are in fact several tens of \\kms~ larger than previously thought, and agree with those expected for the most massive substructures in a $\\Lambda$CDM universe (Stoehr et al. 2002).  Broadly speaking, the hierarchical paradigm predicts that the Milky Way formed through mergers of smaller systems (White \\& Rees 1978). These systems would contribute to the dark halo, the spheroid (the bulge and the stellar halo) and to the Galactic gas reservoir. It may be very difficult to determine the relative gas contribution of these progenitor objects to the present Galaxy, since gas ``easily forgets'' its site of origin. However for collisionless stars and dark-matter the situation can be quite different. If the dynamical mixing timescales are sufficiently long (i.e. longer than the age of the Universe) it may be possible to ``break-up'' the Galactic spheroid (stars and may be even dark matter particles) into coherent structures in phase-space directly related to the systems that merged to form the Milky Way we observe today.  A first attempt at determining whether the merging history of the Milky Way may be imprinted in the phase-space structure of {\\it nearby} halo stars, and thus be recovered, was made by Helmi \\& White (1999; hereafter HW). They studied the infall of satellites onto a fixed Galactic potential, and the evolution of the debris in phase-space. They found that after 10 Gyr stars having a common origin are distributed smoothly in space, but appear very clumped in velocity space, where they define streams with very small velocity dispersions. The expected number of such streams scales with the initial size $r$, velocity dispersion $\\sigma$ and orbital period $P$, of the disrupted object: \\begin{equation} N_{\\rm stream} \\sim 10 \\D\\left(\\frac{r}{1~{\\rm kpc}}\\right)^2 \\left(\\frac{\\sigma}{15~{\\mbox \\kms}}\\right) \\left(\\frac{P}{0.23~{\\rm Gyr}}\\right)^{-3}. \\end{equation} The total number of stars associated with the object is $N_* \\propto r\\sigma^2$ (from the virial theorem) while the volume $V$ over which they are spread scales with the cube of the size of the orbit, and so approximately as $P^3$. Hence the number of stars per stream in the Solar neighbourhood scales as $N_*/VN_{\\rm stream} \\sim \\sigma /r$; objects with large initial velocity dispersion and small initial size should produce the most easily detectable streams with little dependence on initial period.  Such arguments suggest that the Solar neighbourhood velocity ellipsoid is composed of $300-500$ kinematically coherent structures which originated in past merger and accretion events. A pair of halo streams that can perhaps be directly linked to a disrupted satellite were detected in the Solar neighbourhood by Helmi et al.(1999).  The progenitor of these two streams was probably similar to the dwarf galaxy Fornax. Substructure in the outer halo also appears to be ubiquitous, and has been found by several surveys over the last few years (e.g. Ivezic et al.  (2000) and Yanny et al. (2000) for the SDSS; Dohm-Palmer et al. (2001) for the SPS; Vivas et al. (2001) for QUEST).  Most of these recently discovered structures can be associated to just one of Milky Way's satellites: the Sagittarius dwarf galaxy which is in the process of being completely disrupted (Ibata et al. 2001; Mart\\'{\\i}nez-Delgado et al. 2001; Helmi \\& White 2001).   A weak point of the HW analysis and of similar studies (e.g. Johnston, Hernquist \\& Bolte 1996; Johnston 1998), is the assumption of a fixed, smooth potential onto which galaxies are accreted. In hierarchical clustering, galaxy potentials are constantly changing, and can vary very violently during mergers. Large numbers of clumps orbit the centre of even a ``virialised\" halo. These clumps may have substantial effects on the structure of debris streams (e.g Johnston, Spergel \\& Haydn 2002; Ibata et al. 2002; Mayer et al. 2002).  The main goal of the present paper is to understand the phase-space structure of cold dark-matter halos. In particular, we want to study the evolution of satellite debris, and to quantify the expected amount of substructure. We also want to determine to what extent previous results are valid in the truly hierarchical regime of the build--up of a galaxy.  We tackle these problems by scaling down to galactic size a high-resolution simulation of the formation of a cluster in a $\\Lambda$CDM cosmology (Springel et al. 2001).  The paper is organised as follows. In Sec.2 we describe the simulations, in Sec.3 we follow the evolution in phase-space of debris streams, and compare to the analytic model of HW in Section 3.2. Sec.4 describes the mass-growth history of the simulated dark-matter halo, and in Section 5 we determine the number of streams and their internal properties as function of distance from the dark-matter halo centre. We leave the summary and discussion of our results for Section 6. ",
        "conclusions": "We have studied the phase-space evolution of debris from the progenitors that merge to build up a dark-matter halo in a $\\Lambda$CDM cosmology. Our analysis has shown that the debris streams originating in progenitors of different sizes and orbital characteristics all behave in a similar way: with velocity dispersions and local space density decreasing in time. The evolution of the debris streams that we were able to follow until the present time is consistent with phase mixing. Even for halos that we could not follow for a very long time -- because of their smaller initial number of particles or their shorter orbital timescales-- we find the debris to show similar behaviour. On the scales of the fine-grained distribution function, mixing is apparently not strongly chaotic. On the contrary, the phase-space evolution appears to be quite organised and simple, very similar to the mixing of streams orbiting in an idealised static potential.   In principle, a dark-matter halo formed in a $\\Lambda$CDM cosmology is not a smooth entity. Not only do dark-matter halos contain a large number of dark satellites, they also have large amounts of substructure in the form of streams.  We predict, however, that dark matter in the Solar neighbourhood should be clumped in a few hundred thousand streams, producing a velocity ellipsoid which is close to a multivariate Gaussian (Helmi, White \\& Springel 2002). These streams have their origin in the different halos that merged to form the dark halo of the Galaxy. Most of these halos give rise to a large number of intersecting streams in the inner Galaxy.  Determining the characteristics of these hundred thousand streams is difficult even with the high-resolution simulation used here. We are mostly limited by the number of particles. Although the simulation as a whole has 66 million particles, inside a 4~kpc box centred on the ``Sun'', we find only a couple of thousand particles.  In such boxes we find on average two hundred streams with more than one particle, and typically each has only two particles!  The internal stream velocity dispersion that we measure at the present-day is extremely small, of the order of only 1 \\kms.  It is encouraging that we find reasonable agreement between the behaviour of debris streams in static potentials and that observed in this high resolution simulation. This implies that our earlier estimates of the number of stellar streams in the vicinity of the Sun (Helmi \\& White 1999) may indeed apply. Since the initial stellar phase-space distribution in the progenitor objects was probably of lower dimensionality than we here assume -- stars tend to form in disks so that the distribution in at least two of the six phase-space coordinates collapses -- star streams may be colder than the dark matter streams we have analysed, and so may be more easily distinguishable.  It is also interesting to note that because the material that ends up populating the inner galaxy was already in place 10 Gyr ago, the oldest stars are predicted to be near the galactic centre (White \\& Springel 2000). Because this material comes from only a few objects, one might expect the stellar populations to be quite homogeneous, although this of course depends on whether the stars themselves formed in these few massive objects, or whether they were accreted into these objects in the first place.  A crude estimate of the stellar content of a stream in the Solar neighbourhood can be obtained as follows. Let us first estimate the ``mass-to-light '' ratio per particle $[M_{\\rm DM}/L_*]$ as the ratio of dark-matter mass to stellar halo light enclosed in a shell of thickness 2 kpc at the location of the Sun. We use the NFW profile of Eq.(\\ref{eq:rho_NFW}) for the dark-matter distribution. For the stellar distribution we assume a density profile $\\rho_*(r) = \\rho_\\odot r_\\odot^2 / ( r^3 + r_c^3)$, where $\\rho_\\odot \\sim 3\\times 10^4 \\sm$~kpc$^{-3}$, and where $r_c$ is the core radius, and is much smaller than $r_\\odot$. We obtain for the Solar Neighbourhood $[M_{\\rm DM}/L_*] \\sim 625$, for $\\Upsilon_* = 2.5 \\Upsilon_\\odot$. For one of our most massive streams (with 3 particles), this would imply an average $\\sim 0.54 \\sm$ in stars in a sphere of 100 pc radius centred on the Sun. Assuming a Salpeter initial mass function, and down to an absolute magnitude of $M_V = 7^m$ (Bergbusch \\& VandenBerg 1992), i.e. $V = 12^m$ at a distance of 100 pc, this corresponds to approximately 2 stars per stream. On the other hand, streams originate from very localised regions of phase-space in the progenitor objects, so it seems more likely that only about one stream in 250 has any stars at all, that these streams have $\\Upsilon_* = 2.5$ characteristic of the ``stellar'' regions of their progenitors, and that all the other streams are dark. In this case a massive, ``luminous'' stream might contain $135 \\sm$ in stars in a sphere of 100 pc radius centred on the Sun, implying approximately $450$ stars down to $M_V = 7^m$ for a Salpeter IMF.  More reliable estimates of the stellar content and number of streams, as well as considerably more insight into the properties of the Galactic stellar halo would be obtained by combining semi-analytic techniques (e.g Kauffmann et al. 1993) with high-resolution simulations (see, for example, Springel et al. 2001).  This would enable one to predict trends in the chemical composition, age, spatial distribution and kinematics of halo stars as a function of position throughout the Galaxy."
    },
    "0208/astro-ph0208367_arXiv.txt": {
        "abstract": "Observations of the 297~s X-ray pulsar \\onee\\ with the {\\em Rossi X-Ray Timing Explorer} have revealed its 14.4~d eccentric orbit around its B supergiant companion.  The best-fit orbital elements are: $P_{\\rm orb} = 14.365(2)$~d, $a_x\\sin i=99.4(18)$~light~s, $e=0.20(3)$.  No eclipses are detected, indicating that the binary inclination is $\\lesssim 55^\\circ$. ",
        "introduction": "It is a remarkable coincidence that there are two unrelated X-ray pulsars in Centaurus with nearly identical spin periods separated by only 15\\arcmin\\ on the sky \\citep{wps78,lmh+80}.  One of the sources is the 292~s transient X-ray pulsar \\twos, which is associated with the main sequence Be~V companion Hen~715 (HD 102567) \\citep{dab+78} and is 1.5~kpc distant.  The pulsar exhibits periodic outbursts at 186.5~d intervals, which are believed to occur during periastron passage in the neutron star's eccentric orbit.  The X-ray flux in quiescence is typically $\\la$ 3 mcrab but the periastron flares reach a flux of several hundred mcrab.  The second source is the 297~s X-ray pulsar \\onee, which is associated with the B2~Iae supergiant companion V830~Cen \\citep{hcc81,dc82} and is $8.5 \\pm 1.5$ kpc distant.  The pulsar appears to be persistent and steady, with a typical X-ray flux of a few mcrab, corresponding to a luminosity of order $10^{36}$ erg~s$^{-1}$.  Such a low luminosity is inconsistent with Roche-lobe overflow and indicates that the pulsar is almost certainly accreting from the wind of V830~Cen.   A binary period of $\\gtrsim 6$~d is required for the B~supergiant to fit within the pulsar's orbit \\citep{dc82}, and previous authors have proposed periods ranging from 5.6~d to 12.1~d based on optical photometric and spectroscopic studies \\citep{icm82,hcct87}.  For supergiant X-ray binaries with orbital period $P_{\\rm orb}\\lesssim 20$~d, there is a significant \\textit{a priori} probability of an X-ray eclipse (corresponding to a critical binary inclination) given approximately by \\begin{equation} \\Pr \\simeq 0.38 \\left(\\frac{P_{\\rm orb}}{\\mbox{\\rm 15\\ d}}\\right)^{-2/3} \\left(\\frac{M_c}{15 M_\\odot}\\right)^{-1/3} \\left(\\frac{R_c}{24 R_\\odot}\\right) , \\end{equation} where $M_c$ and $R_c$ are the companion's mass and radius, and we have assumed a circular orbit.  Eclipsing X-ray pulsars provide important constraints on the neutron star mass range \\citep{vkvpz95}.  Since only 8 such systems are known \\citep{cwj+02,cm02}, it is of significant interest to increase the sample.  Thus motivated, we observed \\onee\\ with the {\\em Rossi X-Ray Timing Explorer (RXTE)} in an effort to determine the pulsar's orbital period and search for X-ray eclipses.  In this paper, we report our discovery of the 14.4~d orbit of \\onee.  We found no evidence for an X-ray eclipse. ",
        "conclusions": "We have found that the supergiant X-ray pulsar \\onee\\ is in a moderately eccentric 14.4 d orbit.  On a $P_{\\rm spin}$-$P_{\\rm orb}$ diagram, the system resides among the wind-fed supergiant X-ray binaries \\citep{cor86,wv89,bcc+97}, as expected from its low X-ray luminosity.  The absence of an X-ray eclipse may be used to constrain the binary inclination (and hence the supergiant mass) for a given companion radius $R_c$.  Neglecting the moderate orbital eccentricity, the no-eclipse condition is $R_c/(a_x\\sin i) < \\cot i$, where we have assumed that the companion mass is much larger than the pulsar mass. For the range of radii consistent with a B2~Ia supergiant (30--60 $R_\\odot$; see de Jager \\& Nieuwenhuijzen 1987)\\nocite{dn87}, this gives a maximum inclination angle ranging from 55$^\\circ$ to 40$^\\circ$.  Assuming a neutron star mass of 1.4 $M_\\odot$, this implies a minimum companion mass of 11--22 $M_\\odot$.  \\citet{prps02} have noted that massive X-ray binaries can be divided into three broad groups: (i) moderately wide ($P_{\\rm orb}\\simeq$ 20--100 d) binaries with a significant ($e>0.3$) eccentricity caused by a neutron star ``kick'' at birth; (ii) short-period, ($P_{\\rm orb}\\lesssim 10$ systems in which tidal circularization has resulted in low-eccentricity ($e\\lesssim 0.1$) orbits;  and (iii) wide ($P_{\\rm orb}> 30$ d) binaries with low ($e<0.2$) eccentricities, which may have experienced a weaker kick at birth.  We note that \\onee\\ does not fit comfortably into any of these categories, although it appears to be intermediate between groups (i) and (ii).  As a 14.4-d binary, it is possible that tidal torques have played some role in reducing the eccentricity caused by the neutron star birth, although this is unlikely to have been a strong effect. It would be interesting to  identify more binaries in the 10--20~d range, in order to better understand at what point tidal torques play a substantial role."
    },
    "0208/astro-ph0208198_arXiv.txt": {
        "abstract": "{ The 22~GHz H$_2$O masers in the circumstellar envelope of the Mira variable star U~Her have been observed with MERLIN using a phase referencing technique to determine accurate astrometric positions. The positions were compared with the optical positions obtained with the Hipparcos satellite to an accuracy of $18$~mas. The absolute radio position of the brightest H$_2$O maser spot is found to match the optical position, indicating that this spot is the stellar image amplified by the maser screen in front of it.  The occurrence of an amplified image in the 22 GHz maser can be used to accurately determine the positions of the H$_2$O with respect to the star as well as with respect to the SiO and OH masers. Our observations seem to indicate that the star is not in the centre of the distribution of maser spots, which has been interpreted as a ring. ",
        "introduction": "The circumstellar envelopes (CSEs) around late-type stars contain several maser species that are excellent probes of the dynamics of the outflowing material. Astrometric observations are an important tool to study the locations and motions of the circumstellar masers with respect to the central star. This understanding is essential to reach the main goal of maser astrometry, which is to determine the distances to heavily enshrouded stars, that are too faint for their parallax to be determined directly.  \\subsection{Circumstellar H$_2$O masers}  Interferometric observations of the 22 GHz H$_2$O masers in the CSEs of Mira-variable stars with MERLIN, the VLA and Very Long Baseline Interferometry (VLBI) indicate that the H$_2$O masers are found up to a few hundred AU from the star (e.g. Lane et al. 1987). This is generally inside the OH maser shell, which is located at up to several 1000 AU.  The H$_2$O masers often show an a-spherical distribution, and the size of the maser region is thought to increase with mass-loss rate (Cooke \\& Elitzur 1985). The maser is expected to be pumped due to collisions (Neufeld \\& Melnick 1991). The 22 GHz maser can then be easily excited in the inner parts of the CSE at temperatures of 400 to 1000 K and H$_2$ number densities of $10^9$ cm$^{-3}$. In this region the outflow is still being accelerated. Therefore, as shown by Rosen at al. (1978) the velocity coherent paths through the masing medium are of approximately equal length in both the radial direction as well as the direction tangential to the star, and thus the H$_2$O maser beaming is expected to be both radial and tangential. The radial beaming results in the maser occurring in front of the star, tangential beaming would display a ring within a narrow velocity range close to the stellar velocity (e.g. Reid \\& Menten 1990).  H$_2$O masers are significantly more variable than their OH cousins. They exhibit strong variability in intensity, which seems to indicate that the masers are at least partially unsaturated, since an unsaturated maser is strongly influenced by changes in the local conditions. An analysis of the H$_2$O maser line-widths and line-shapes also indicates that the masers are not completely saturated (e.g. Vlemmings et al. 2002).  Whereas semi-regular stars are observed to have H$_2$O maser spectra that can change shape rapidly, Mira-variable stars typically show no large profile changes over several years. However, the individual features can still show significant changes in intensity (Engels et al. 1988).  \\subsection{Amplified Stellar Image}  Interferometric observations of OH masers have revealed that the most compact features were only found at the blue-shifted side of the spectrum (Norris et al. 1984). It was argued that this was due to amplification of the continuum maser emission from the underlying star by the maser screen in front of it. This is called the {\\it Amplified Stellar Image Theory}. Amplification of the stellar emission results in a high brightness maser spot at the most blue-shifted side of the OH maser spectrum. This spot should coincide at the different OH maser transitions, and is expected to be persistent over a long period of time. Several observations have confirmed this hypothesis (e.g. Sivagnanam et al. 1990).  According to the amplified stellar image theory, the compact, most blue-shifted, spot is necessarily fixed to the stellar position. Therefore, high resolution astrometric observations of this spot can be used to determine the stellar trajectory. This hypothesis has been tested for the Mira-variable star U~Her by van Langevelde et al. (2000, hereafter vL00). The {\\it absolute} positions of the OH maser spots were determined with respect to the radio-reference frame using extra-galactic phase reference sources. The positions were compared with the optical Hipparcos positions with unprecedented accuracy. It was shown that the most blue-shifted spot was indeed located in front of the stellar radio-photosphere. The size of this spot was found to be $\\approx 20$~mas, which is comparable with the expected size of the radio-photosphere, which is thought to be twice the size of the star, as proposed by Reid \\& Menten (1997).  Although the H$_2$O masers generally show a great number of spots over an area of several hundred mas, it has been argued that in some cases one of the H$_2$O maser spots corresponds to the stellar image (Reid \\& Menten 1990; Marvel 1996; Colomer et al. 2000). However, because the distribution of the H$_2$O masers is considerably less spherical than that of the OH masers, it is not straightforward to assume that the stellar image underlies the most blue-shifted spot. Also, because the maser brightness depends strongly on local effects such as density or pumping inhomogeneities, several bright spots can be observed and an H$_2$O maser stellar image could be less conspicuous or persistent than the OH stellar image.  Here we present phase referencing observations of the H$_2$O masers around U~Her used to determine accurate maser spot positions. These have been compared with the Hipparcos optical position and the positions obtained for the OH masers in vL00. ",
        "conclusions": "Our observations have shown that the circumstellar H$_2$O maser can amplify the stellar image and produce a strong stellar image, a phenomenon previously detected at the OH maser transitions.  Although this effect is not necessarily strong in all H$_2$O masers, it can be very valuable for astrometric purposes. Accurate astrometry of the stellar image can be used to determine the location of the various maser species in the CSE with respect to each other and the star. Simultaneous, high resolution observations significantly improve our understanding of the kinematics in the CSEs. Using the stellar image in H$_2$O masers, it will also be possible to determine the stellar trajectory and distance with a higher accuracy than with OH masers. However, because of the high variability of H$_2$O masers additional monitoring will have to be performed to show if the stellar image is persistent enough for a long term monitoring campaign.  {\\it Acknowledgments:} This project is supported by NWO grant 614-21-007. MERLIN is a National Facility operated by the University of Manchester at Jodrell Bank Observatory on behalf of PPARC. We also thank the referee Mark Reid for valuable input."
    },
    "0208/astro-ph0208221_arXiv.txt": {
        "abstract": "We obtained near--infrared spectra of a sample of very low mass objects as a function of age in order to investigate the temperature and surface gravity sensitivity of several features in the J-- and K--bands. ",
        "introduction": "In order to study the initial mass function in young open clusters at the low mass end, one needs to have a fast and reliable way of separating pre-MS members from foreground dwarfs and background giants. These objects occupy same place on the color-magnitude diagram, but have different surface gravities. Using this fact, we decided to explore the  gravity-sensitive features in the near-infrared spectra of the objects of spectral types M--L, corresponding to the range of masses near sub-stellar boundary. ",
        "conclusions": ""
    },
    "0208/astro-ph0208017_arXiv.txt": {
        "abstract": "As part of a program to study the evolution of active galactic nuclei (AGN) in clusters of galaxies, we present our results for Abell~2104. A deep \\chandra\\ observation of this massive, $z = 0.154$ cluster reveals a significant X-ray point source excess over the expectations of blank fields, including eight X-ray counterparts with $R<20$ mag. Our spectroscopy shows that all six X-ray sources associated with red counterparts are cluster members and their X-ray properties are consistent with all of them being AGN. Only one of the six has the emission lines characteristic of optically selected AGN; the remaining five would not have been classified as AGN based on their optical spectra. This suggests the existence of a large population of obscured, or at least optically unremarkable, AGN in clusters of galaxies. These six sources correspond to a lower limit of $\\sim 5$\\% of the AGN fraction in cluster galaxies with $R<20$ mag (rest-frame $M_V = -19.5$ mag) and is comparable to the blue galaxy fraction in the cluster. Such an obscured AGN population in clusters of galaxies has many implications for cluster galaxy evolution, the hidden growth of their central, supermassive black holes, estimates of the star formation rate at infrared and radio wavelengths, and the observed variance in the hard X-ray background. ",
        "introduction": "Both star formation and accretion onto supermassive black holes require reservoirs of cold gas. The presence of nuclear activity in cluster galaxies is therefore an indication of the efficiency with which these galaxies were stripped of their cold interstellar medium, and to what extent their central, supermassive black holes may continue to grow in cluster environments. The evolution of the fraction of cluster galaxies which host active galactic nuclei (AGN) is therefore an important component of galaxy evolution in clusters. Determination of the connection between AGN and the history of infall and star formation for host galaxies may also provide interesting constraints on the fueling and lifetimes of nuclear activity. AGN in clusters of galaxies may be a significant contribution to the radio and infrared sources in clusters, and need to be identified in order to use such measurements to infer star formation rates \\citep[\\eg][]{duc02}.  It is not yet known if the AGN fraction in clusters of galaxies shows any evolution with redshift, although there is clear evidence for evolution in cluster galaxies. The fraction of galaxies with recent star formation decreases significantly from $z \\sim 0.5$ to the present \\citep[the Butcher-Oemler effect; ][]{butcher78,butcher84} and the fraction of poststarburst galaxies also has decreased dramatically over this same redshift range \\citep{dressler88,dressler99} and much more so than in the field \\citep{zabludoff96,dressler99}. The current best estimate of the mean cluster AGN fraction is $\\sim 1$\\% \\citep[\\eg][]{dressler83,dressler99}, a factor of a few less than the bright AGN fraction in the field \\citep[\\eg][]{huchra92}. These measurements of the AGN fraction in clusters and the field are the result of extremely large, spectroscopic surveys of galaxies. However, the fraction of all galaxies that are bright AGN is small; this makes spectroscopic searches inefficient and any evolution in the host galaxy population difficult to detect. A related problem is that AGN may not be obvious at visible wavelengths, whether due to moderate line strengths compared to the host galaxy light, or obscuration.  A better way to search for AGN is at X-ray wavelengths. While AGN comprise only a small minority of all galaxies at visible wavelengths, they are the dominant contribution to the luminous, hard (2-10 keV) X-ray point source population. We have begun a program to study the AGN content of clusters of galaxies that have deep observations with the {\\it Chandra X-ray Observatory}. As all luminous AGN produce significant hard X-ray emission, we use the \\chandra\\ data to select AGN for spectroscopic observations to determine cluster membership. \\chandra\\ data are therefore uniquely suited to identify these AGN due to its superior angular resolution compared to all other X-ray satellites. In this {\\it Letter} we present our discovery of six previously unknown AGN in Abell~2104, a rich cluster \\citep[Abell class 2;][]{allen92} of galaxies at $z = 0.154$ \\citep{liang00}. ",
        "conclusions": "Only one of the six AGN in Abell~2104, or $\\sim 1$\\% of the $\\sim 100$ $R<20$ mag (rest-frame $M_V = -19.5$ mag) cluster members, is an obvious AGN in our optical spectra. This is in good agreement with the mean cluster AGN fraction of 1\\% from optical spectroscopic surveys \\citep{dressler99}. When \\chandra\\ observations are used to identify AGN, the AGN fraction appears to be a factor of five higher, or $\\sim 5$\\%. To evaluate the significance of this result, we use a simple binomial distribution test. Given an expected AGN fraction of 1\\%, the variance in the number of AGN in a sample of 100 galaxies is $100 \\times 0.01 \\times 0.99$. We discovered five more AGN than expected, which is significant at the 5/0.99 or $5\\sigma$ level. In fact, an AGN fraction of 5\\% is a lower limit to the true AGN fraction as our observations are not sensitive to more highly obscured AGN. For example, if the lower-luminosity AGN (such as \\#5 and \\#6) had a factor of five to ten higher absorbing columns, we would not have detected them; such higher absorption columns are not unusual among nearby field AGN with similar luminosities. There is additional evidence for a larger number of AGN in other clusters of galaxies in recent \\chandra\\ observations of point source excesses toward several other clusters \\citep{cappi01,sun02,molnar02}.  Most of the AGN in Abell~2104 lack strong emission lines, which may be due to dilution by host galaxy light. For example, if we use the ratio of [OIII]$\\lambda5007$\\AA\\ to hard X-ray flux for source \\#1 to predict the equivalent widths for the other five sources, the equivalent widths would not be detectable in our spectroscopy. The [OIII]$\\lambda5007$\\AA\\ to hard X-ray flux ratio for source \\#1 is $\\sim 10^{-3}$ and is comparable to many Seyfert 2s \\citep{polletta96}, although this ratio varies by several orders of magnitude in AGN. Our nondetection of emission lines in these objects could therefore be due to the low signal-to-noise of our spectroscopy, or due to some obscuration of the emission line region. It is not likely that the two X-ray sources with the lowest luminosities are powered by star formation, rather than AGN. While the most luminous local starbursts, such as Arp 220 or NGC 3256, have hard X-ray luminosities comparable to the two faintest sources in Abell 2104 \\citep[$\\sim 10^{41}$ \\ergs][]{moran99,iwasawa99}, our low luminosity sources do not show detectable emission lines indicative of massive star formation. Furthermore, both Arp 220 and NGC 3256 would be 0.3 mag bluer in rest-frame $B-V$ (or 0.6 mag in observed $B-R$) than our A2104 optical counterparts. It is therefore more likely that these X-ray sources are powered by nuclear activity. Even if these two X-ray sources do not include an AGN component, the significance of our result changes to (4-1)/0.99 or a $3\\sigma$ excess over expectations. These AGN, without detectable emission lines, appear similar to the optically bright, hard X-ray population that contribute approximately 40\\% of the hard X-ray background \\citep{mushotzky00}. If a significant number of such sources reside in clusters, they may be an important constituent of the hard X-ray background. Their association with the highly biased cluster galaxy population may also explain the variation in the hard X-ray background intensity from field to field discussed by \\citet{cowie02}.  The unexpectedly high AGN fraction in Abell 2104 is comparable to the fraction of galaxies that qualify as Butcher--Oemler galaxies \\citep{butcher84} or that are emission-line galaxies \\citep{liang00}. This AGN fraction, and the fact that it is a lower limit, suggests that obscured AGN may make a significant contribution to infrared and radio sources in cluster galaxies. For example, \\citet{duc02} concluded that most of their FIR and radio-selected galaxies in Abell~1689 are powered by star formation because they only found optical emission line ratios consistent with AGN in one galaxy. Our result suggests that many of their sources could be AGN that lack strong emission-line features.  Surprisingly, the host galaxies of these AGN all fall near or on the cluster color--magnitude relation \\citep{sandage78}, which is presumably composed of old, quiescent galaxies without the reservoirs of cold gas necessary to fuel AGN. From our ground-based data ($1''$ seeing FWHM), approximately half of these host galaxies appear to have disk morphologies and half appear to have elliptical or S0 morphologies. They also span approximately three magnitudes in luminosity. Our naive expectation was that the AGN population in clusters should be associated with the blue, starforming galaxies as they have both significant reservoirs of cold gas and are interacting with the cluster potential for the first time. If the AGN lifetime were longer than the typical lifetime of $\\sim 0.1$ Gyr for star forming galaxies in clusters \\citep{poggianti99}, or there were a long delay before the onset of nuclear activity, AGN might instead be associated with the post-starburst population. In Abell~2104, the brightest three AGN shown in Figure~\\ref{fig:cmd} fall on the cluster color--magnitude relation; while the fainter three are slightly bluer, they are not as blue as Butcher-Oemler galaxies. None of these AGN host galaxies have been completely stripped of their cold gas by the cluster potential \\citep[\\eg][]{gunn72}. These galaxies appear to have retained a reservoir of cold gas at the center of their potential wells similar to the dust disks observed in local cluster ellipticals \\citep[\\eg][]{jaffe94,martel99}.  The association of these six AGN in Abell~2104 with red cluster galaxies could be because these observations only probe the inner 1~Mpc of the cluster, which is dominated by old, red galaxies. The population of blue, starforming galaxies in clusters is known to be more spatially extended \\citep{butcher84} and have higher velocity dispersions \\citep{dressler99} than the red, passively evolving galaxies. While the six AGN in Abell~2104 are photometrically and spectroscopically similar to old, red galaxies, their mean redshift is offset by $\\sim 1000$ \\kms\\ from the cluster members measured by \\citet{liang00}. At least some of the AGN may therefore be falling into the cluster for the first time. The asymmetric appearance of the cluster galaxy distribution in Figure~\\ref{fig:image} suggests that this cluster is dynamically evolving. The spatial and kinematic distribution of AGN hosts to large cluster radii would help identify whether AGN are predominantly associated with the infalling galaxy population.  The AGN host galaxy population in clusters could provide some interesting constraints on the fueling and lifetimes of AGN. Their association with blue galaxies would suggest that the AGN lifetime is less than the 0.1 Gyr lifetime of this population, consistent with current estimates for the lifetime of nearby AGN in the field \\citep{martini02b}, and that nuclear activity is quenched by the cluster potential. If AGN are instead found in cluster galaxy populations of different ages, this suggests that nuclear activity is also episodic in the cluster environment and galaxies retain a central reservoir of cold gas for several Gyr after they enter the cluster potential. In either scenario, the presence of this large population of obscured AGN in cluster galaxies indicates that their central, supermassive black holes continue to grow in the cluster environment."
    },
    "0208/astro-ph0208292_arXiv.txt": {
        "abstract": "A flood of reliable seismic data will soon arrive. The migration to larger telescopes on the ground may free up 4-m class instruments for multi-site campaigns, and several forthcoming satellite missions promise to yield nearly uninterrupted long-term coverage of many pulsating stars. We will then face the challenge of determining the fundamental properties of these stars from the data, by trying to match them with the output of our computer models. The traditional approach to this task is to make informed guesses for each of the model parameters, and then adjust them iteratively until an adequate match is found. The trouble is: how do we know that our solution is unique, or that some other combination of parameters will not do even better? Computers are now sufficiently powerful and inexpensive that we can produce large grids of models and simply compare {\\it all} of them to the observations. The question then becomes: what range of parameters do we want to consider, and how many models do we want to calculate? This can minimize the subjective nature of the process, but it may not be the most efficient approach and it may give us a false sense of security that the final result is {\\it correct}, when it is really just {\\it optimal}. I discuss these issues in the context of recent advances in the asteroseismological analysis of white dwarf stars. ",
        "introduction": "Most scientists are familiar with the concept of Occam's razor---the idea that if you have to choose between competing explanations for some physical phenomenon, the simplest explanation is most likely to be correct. My thesis supervisor, Ed Nather, told me a story about another less widely known scientific tool that may be just as important as Occam's razor. He calls it ``Wampler's screwdriver'' \\cite{nat95}.  In the early 1970's, Ed was attending a conference of the Astronomical Society of the Pacific in California, and Joe Wampler was giving a presentation about the first discovery of a double quasar \\cite{wam73}. The standard procedure at the time was to identify blue objects inside the relatively large positional error box of a newly discovered radio point source, and then take spectra of them, one by one, until you found one with a big redshift. What Joe decided to do was go back to the fields where quasars had been discovered in this way, and take spectra of {\\it all} of the blue objects, even after he found one of them to be a quasar. Joe's double quasar turned out to be an accidental alignment of two quasars at different distances, but later on others repeated what he had done and found a double quasar that was the result of gravitational lensing---so his method was an important contribution to the field. At the end of Joe's presentation, he posed a simple question to the audience: ``Why do you always find a lost screwdriver in the last place you look?'' The answer, of course, is because you stop looking.  In this paper, I will review a method of fitting models to seismological data that {\\it keeps looking}, even after it has found a pretty good fit to the observations. This is drawn from work I have been doing over the past few years to develop a model-fitting method based on a genetic algorithm, helping us to learn more about pulsating white dwarf stars (\\opencite{mnw00}; \\citeyear{mwc01}). In section~\\ref{wdsec}, I will discuss what this method has allowed us to learn about white dwarfs, but for the bulk of this review I will focus on the method and related issues.  \\begin{figure} \\centerline{\\includegraphics[width=12truecm]{fig1.eps}} \\caption[]{A representation of the logical distinction between reality, observations, and models of reality. Note that the ultimate goal of this process is an improved understanding of the constitutive physics.} \\label{fig1} \\end{figure}  \\subsection{Reality, Models, \\& Physics}  Before I get to that, I would like to step back and take a philosophical look at the general process that we use to learn anything about the objects we study (see Figure~\\ref{fig1}). Out there somewhere there is a ``real world'' that we pass through various filters and selection effects to get ``observations''. As observers, we try our best to compensate for every effect between the real world and the data point, but it's important to realize that we use models in this process too.  With the observations in hand, we devise computer models to try to explain them, doing our best to include all of the relevant physical processes that are in principle detectable. Generally these models have a number of tunable parameters, and we do our best to adjust them until the predictions of the model agree as closely as possible with the observations---in our case, generally the pulsation periods of a star. When we have found a model that adequately reproduces the observations, we assume that the values of the parameters tell us something about the properties of the actual star. But we should never forget that what we are actually dealing with are {\\it models} of reality, and not reality itself. When we derive values of $\\log g$ and $T_{\\rm eff}$ from spectral lines, for example, we are not measuring the mass and temperature of the star---we are (at best) deriving the optimal match between our models of stellar atmospheres and the extracted spectrum over a finite wavelength interval. We call this approach the ``forward method'', and it can only tell us what is best {\\it within the context of the models we use}. It cannot tell us that we are using the wrong models, unless or until we actually try different models.  \\begin{figure} \\centerline{\\includegraphics[width=12truecm]{fig2.eps}} \\caption[]{Example model-spaces that would be easy (left) and hard (right) for traditional optimization methods. Note how the likely outcome in the two cases would differ if the search were confined to the region specified by the grey rectangle in the contour plots (Adapted from \\opencite{cha95}).} \\label{fig2} \\end{figure} ",
        "conclusions": "I hope that I have convinced you that genetic algorithms are potentially a very powerful tool for asteroseismology. They can provide objective global optimization for problems with more than a few parameters, and in the process they yield fairly detailed maps of the model-space, which we can use to judge the uniqueness of the final result. To convince yourself that a genetic algorithm is more efficient than a large grid, and to convince others that the final result can be trusted, you should definitely perform a ``Hare \\& Hound'' exercise---trying to match your models to the observables from another model, and demonstrating that it can be done. Remember that the speed of your computer effectively determines the size of the problems you can attack. If you want to solve a specific problem, you can and should determine how much computing power you will need to solve it in a reasonable time. Finally, the payoff can be quite high, as it was when we applied this method to our white dwarf models.  I have seen this method begin to transform white dwarf asteroseismology from a field where the theory was being driven by the observations, to one where new observations are being driven by the theory. I hope that it can help launch revolutions in the seismological analysis of other types of stars too."
    },
    "0208/astro-ph0208547_arXiv.txt": {
        "abstract": " ",
        "introduction": "Adaptive SPH and N-body simulations were carried out to study the collapse and evolution of dark matter halos that result from the gravitational instability and fragmentation of cosmological pancakes. Such halos resemble those formed by hierarchical clustering from realistic initial conditions in a CDM universe and, therefore, serve as a convenient test-bed model for studying halo dynamics. Our halos are in approximate virial equilibrium and roughly isothermal, as in CDM simulations.  Their density profiles agree quite well with the fit to N-body results for CDM halos by Navarro, Frenk, \\& White (1997; NFW).  This test-bed model enables us to study the evolution of individual halos. The masses of our halos evolve in three stages: an initial collapse involving rapid mass assembly, an intermediate stage of continuous infall, and a final stage in which infall tapers off as a result of finite mass supply. In the intermediate stage, halo mass grows at the rate expected for self-similar spherical infall, with $M(a) \\propto a$.  After the end of initial collapse at ($a\\equiv a_0$), the concentration parameter grows linearly with the cosmic scale factor $a$, $c(a)\\simeq 4(a/a_0)$. The virial ratio $2T/|W|$ just after virialization is about 1.35, a value close to that of the $N$-body results for CDM halos, as predicted by the truncated isothermal sphere model (TIS) (Shapiro, Iliev, \\& Raga 1999) and consistent with the value expected for a virialized halo in which mass infall contributes an effective surface pressure.  Thereafter, the virial ratio evolves towards the value expected for an isolated halo, $2T/|W|\\simeq 1$, as the mass infall rate declines. This mass accretion history and evolution of concentration parameter are very similar to those reported recently in $N$-body simulations of CDM analyzed by following the evolution of individual halos.  We therefore conclude that the fundamental properties of halo collapse, virialization, structure, and evolution are generic to the formation of cosmological halos by gravitational instability and are not limited to hierarchical collapse scenarios or even to Gaussian-random-noise initial conditions.  \\begin{figure}[!t] \\begin{center} \\includegraphics[width=2.5in]{malvarez_fig1.eps} \\end{center} \\caption{ Dark matter particles at $a/a_c=3$. } \\label{fig:simple} \\end{figure} ",
        "conclusions": ""
    },
    "0208/astro-ph0208401_arXiv.txt": {
        "abstract": "We have imaged a sample of 20 spiral galaxies in H$\\alpha$ and in the near--infrared {\\em K} band (2.2$\\mu$m), in order to determine the location and strength of star formation in these objects with respect to perturbations in the old stellar population. We have found that star formation rates are significantly enhanced in the vicinity of {\\em K} band arms. We have also found that this enhancement in star formation rate in arm regions correlates well with a quantity that measures the relative strengths of shocks in arms. Assuming that the {\\em K} band light is dominated by emission from the old stellar population, this shows that density waves trigger star formation in the vicinity of spiral arms. ",
        "introduction": "Two of the leading theories of star formation in spiral galaxies use the concept of a density wave, either as the actual triggering mechanism, or as a means of organisation of star--forming material. The first of these theories has been termed the large scale galactic shock scenario (Roberts 1969; see also Shu et al. 1972; Tosa 1973; Woodward 1975; Nelson \\& Matsuda 1977). This model hypothesises that the gas settles into a quasi--stationary state, with a velocity and density distribution that is driven by the gravitational field of the galaxy. The gas response can be non--linear to an imposed azimuthal sinusoidal potential, if relative motion between the density wave and the cold interstellar medium (ISM) is supersonic (Binney \\& Tremaine 1987). This leads to the formation of a shock near the trailing edge of spiral arms, assuming that the region is inside the corotation radius, which compresses the gas to densities at which stars can form. In observations of spiral galaxies the shock is thought to be characterised by dust--lanes seen on the trailing edges of arms. The time delay needed for the onset of star formation after the compression of the gas implies that star--forming regions should be seen towards the leading edges of arms.  Many of the recent advances in this area have arisen from studies of the atomic and molecular gas in nearby spiral galaxies, through HI and CO line emission (e.g Nakai et al. 1994; Rand 1993, 1995). These studies reveal streaming velocities of gas through the spiral arms and find offsets between the peaks of the gas density and the old stellar population, in agreement with the predictions of the large scale shock scenario.  The alternative picture to this form of triggered star formation is stochastic star formation. \\\"Opik (1953) first hypothesised that a supernova explosion could trigger star formation. A model proposed and developed by several authors (e.g. Gerola \\& Seiden 1978; Seiden \\& Gerola 1982; Seiden 1983; Jungwiert \\& Palous 1994; Sleath \\& Alexander 1995, 1996) suggests that the dominant process for forming stars is stochastic self--propagating star formation, and not density wave triggering. In this model, density waves are only responsible for the organisation of the ISM and stars, and for concentrating new HII regions along spiral arms (Elmegreen \\& Elmegreen 1986; Elmegreen 1993). Thus, star formation rate {\\em efficiency} (i.e. normalised to unit mass of disc material) should be unaffected by location in arm or interarm regions, in this model.  This is in clear distinction to the predictions of the large--scale shock model.  Previous H$\\alpha$ studies of spiral galaxies (Kennicutt 1989, 1998a; Kennicutt, Tamblyn \\& Congdon 1994; and the review by Kennicutt 1998b) have mainly looked at global star--formation, particularly the form of the Schmidt law (Schmidt 1959, 1963). In this paper we are looking at localised star formation in disc galaxies as well as global properties of star formation, and comparing the distribution of star formation with the underlying old stellar population. The main aim of this paper is the analysis of star formation efficiencies in arm and interarm regions in spiral galaxies.  Other studies of star formation efficiencies include Lord \\& Young (1990) and Tacconi \\& Young (1986). Lord \\& Young (1990) looked at the molecular, neutral, and ionized hydrogen distributions in M51. They compared a ratio of massive star formation rates (MSFR) to gas surface density between arm regions and interarm regions and found a higher ratio in the arm regions. Tacconi \\& Young (1986) performed a similar analysis for NGC 6946 and also found that star formation is more efficient in arm regions than in interarm regions. Cepa \\& Beckman (1990) and Knapen et al. (1992) again looked at star formation efficiencies in arm and interarm regions, but with high spatial resolution. Cepa \\& Beckman (1990) compare the H$\\alpha$/$\\sigma_{HI}$ ratio (where $\\sigma_{HI}$ is the HI surface brightness) in the arm and interarm regions of NGC 3992 and NGC 628. Knapen et al. (1992) compare the H$\\alpha$/$\\sigma(H_2+HI)$ ratio in arm and interarm regions. Both Knapen et al. (1992) and Cepa \\& Beckman (1990) explicitly find evidence for spiral arm triggering of star formation, via a non--linear dependence of SFR on gas density. Wyder, Dolphin \\& Hodge (1998) obtained HST WFPC2  V band (F555W) and I band (F702W) data of the spiral galaxy NGC 4321. From the I--V colour of the arm and interarm regions, they concluded that the SFR over the last 5 Myr has been approximately 4 times larger in the arm regions than in the surrounding interarm regions. Also, Knapen (1998) performed a study of HII regions in M100, using a new H$\\alpha$ image. He found that the arms collect HII regions in a similar way to that described by Elmegreen \\& Elmegreen (1986). He concluded that spiral arms may trigger some star formation, but do not affect the HII region luminosity function or mass distribution. Finally, Alonso--Herrero \\& Knapen (2001) used the specific SFR or SFR per stellar mass (i.e. a comparison of Pa$\\alpha$ flux with {\\em H} band continuum emission), and found that while the specific SFR in the central region does not vary statistically with galaxy type, it is higher in barred than in non-barred galaxies.  The remainder of this paper is arranged as follows. Section 2 describes the observations and data analysis; section 3 describes local star--formation rates with respect to properties of spiral structure; and section 4 contains our conclusions. ",
        "conclusions": "We have found an increase in H$\\alpha$ flux in the vicinity of {\\em K} band arms in this sample. If the {\\em K} band arms are dominated by light from the old stellar population, then this can be interpreted as star formation triggered by a density wave. This result is in agreement with Lord \\& Young (1990), who found a higher star formation efficiency in arm regions. It also agrees with Cepa \\& Beckman (1990) and Knapen et al. (1992) who found evidence for spiral arm triggering of star formation via a non--linear dependence of SFR on gas density. This result therefore agrees with the large scale shock scenario (Roberts 1969; Shu et al. 1972; Tosa 1973; Woodward 1975; Nelson \\& Matsuda 1977), which predicts a higher star formation efficiency in arm regions. It does not support the ideas of stochastic self--propagation star formation (e.g. Gerola \\& Seiden 1978; Seiden \\& Gerola 1982; Seiden 1983; Jungwiert \\& Palous 1994; Sleath \\& Alexander 1995, 1996) which predict no change in the star formation efficiency from interarm to arm regions."
    },
    "0208/astro-ph0208386.txt": {
        "abstract": "{ This paper describes a lossy method for compressing raw images produced by CCDs or similar devices. The method is very simple: lossy quantization followed by lossless compression using general-purpose compression tools such as gzip and bzip2. A key feature of the method is that compressed images can be converted to FITS files simply by decompressing with gunzip or bunzip2, and this is a significant advantage for distributing compressed files. The degree of quantization is chosen to eliminate low-order bits that over-sample the noise, contain no information, and are difficult or impossible to compress. The method is lossy but gives guarantees on the maximum absolute difference, the expected mean difference, and the expected RMS difference between the compressed and original images; these guarantees make it suitable for use on raw images. The method consistently compresses images to roughly 1/5 of their original size with a quantization such that no value changes by more than 1/2 of a standard deviation in the background. This is a dramatic improvement on lossless compression. It appears that bzip2 compresses the quantized images to within a few tens of percent of the theoretical limit. } ",
        "introduction": "Optical and infrared instruments now routinely produce huge amounts of image data. The largest current common-user CCD mosaic has $\\rm 12k \\times 8k$ pixels (Veillet 1998); a single image from such a mosaic is 192 MB in size. The largest current common-user infrared detector mosaic has $\\rm 2k \\times 2k$ pixels (Beckett et al.\\ 1998), but makes up for its smaller size by being read more frequently. More than a few nights of data from such instruments can easily overwhelm workstation-class computers. Compression is a solution to some of the problems generated by these large quantities of data. Similarly, compression can improve the effective bandwidth to remote observatories (in particular space observatories), remote data archives, and even local storage devices.  This paper describes in detail a lossy method for compressing raw images and presents a quantitative comparison to other lossy methods. It is organized as follows: \\S~2 reviews the limitation of lossless compression and the motivation for lossy compression; \\S~3 briefly summarizes the most relevant previous work on lossy compression; \\S~4 describes the new method; \\S~5 presents results on the distribution of differences between the original and compressed images; \\S~6 investigates the performance of the method with particular reference to hcomp; \\S~7 compares the performance of the method to other similar methods; \\S~8 discusses the suitability of the method for compressing raw data and investigates some consequences of such use; \\S~9 discusses how the method might be improved; and \\S~10 presents a brief summary. ",
        "conclusions": "\\label{section-discussion}  \\subsection{Suitability for Compressing Raw Data}  The quantization compression method described here, qcomp, was designed to compress raw data. With the results of the tests and comparisons in \\S~\\ref{section-performance} and \\S~7, we can now critically evaluate its suitability for this task. We will see that that qcomp is better suited for compressing raw data than either hcomp (White 1992) or the quantization methods proposed by White and Greenfield (1999) and Nieto-Santisteban et al.\\ (1999).  The fundamental advantage of qcomp over hcomp is that its straightforward nature allows firm limits to be placed on the distribution of differences (see \\S~\\ref{section-differences}). So, for example, an image can be compressed with the knowledge that no pixel will change by more than a specified amount. Furthermore, the distribution of the errors is predictable. Hcomp does give such firm guarantees on the distribution of differences.  However, this advantage would be moot if qcomp were inferior in compression ratio and speed. The most interesting range of $q$ for compressing raw images is 0.5--2, that is, quantizations that change the values in the image by at most 0.25--1 standard deviations. In this range, qcomp/bzip2 gives mean compression ratios of 0.27--0.13 and gives roughly the same RMS difference as hcomp for similar compression ratios but with smaller maximum absolute differences and mean differences (see Figure \\ref{figure-merit}). Furthermore, its decompression speed is very similar to hcomp. Its only disadvantage is that its compression speed is only one third that of hcomp, although it is still adequately fast; its compression rate of roughly 800 kpixel/s on a 1.9 GHz Pentium IV is much faster than the typical read rate of 40 kpixel/s for single-output CCDs and 160 kpixel/s for quad-output CCDs.  It is also worth considering qcomp/gzip, which achieves lower compression ratios of 0.38--0.23 for $q$ in 0.5--2, while compressing slightly faster and decompressing very much faster than hcomp. It also has the great advantage of portability; gzip is installed almost universally on workstations.  The current implementation of qcomp has a further small advantage over the current implementation of hcomp in it distinguishes between the data and overscan sections of an image and quantizes each appropriately; the current implementation of hcomp treats both identically, and so the two have to be separated and compressed individually to avoid either drastic over-compression of the overscan section or under-compression of the data section.  The advantages of the qcomp over the other quantization methods are that the manner in which White and Greenfield (1999) determine the quantization makes their method unsuitable for use with raw data (and indeed such a claim was never made) and the manner in which Nieto-Santisteban et al.\\ (1999) quantize introduces a small bias that is not present with qcomp. When using the same lossless compression method, all of the quantization methods should be similar in speed. The one disadvantage of qcomp is that is requires that the noise in the background be roughly constant within fixed regions, whereas the method of Nieto-Santisteban et al.\\ (1999) estimates the noise on a pixel-by-pixel basis and makes no such requirement. However, this may not be too restrictive as the noise varies only as the square root of the background (and more slowly if read noise is significant), so changes in the background produce much smaller changes in the noise.  Obviously, confidence in the use of qcomp to compress raw data would be significantly improved if end-to-end tests on compressed data gave results that were statistically indistinguishable from those derived from the original data. Such tests should include astrometry of stars, aperture and PSF-fitting photometry of stars, and photometry of low surface brightness sources. These tests are especially important for applications which have stringent requirements on the absence of biases, such as far infrared and sub-millimeter imaging and ultra low surface brightness studies at other wavelengths. Such tests will be presented in Paper II (Watson, in preparation).  \\subsection{Distribution}  An important advantage of compressing files with qcomp/gzip is that they can be distributed with the expectation that no additional software will be needed to decompress them; gzip is installed almost universally on workstations. As bzip2 becomes more widespread (it already forms part of several Linux distributions), it too will gain this advantage. This is a major advantage over software that uses special-purpose compression methods.  \\subsection{Media Capacities}  If we take a compression ratio of 0.2 as typical for qcomp/bzip2 with $q=1$, then media have effective capacities that are 5 times larger than their raw capacities. Thus, a 650 MB CD-ROM has an effective capacity of 3 GB, a 12 GB DAT DDS-3 tape has an effective capacity of 60 GB, and an 80 GB disk has a capacity of 400 GB. This is roughly equivalent to a generation of technology; in terms of capacity, a CD-ROM is almost a DVD-ROM, a DAT tape is effectively a DLT tape, and a single disk is effectively a large RAID array. Thus, compression can allow one to work with large data sets without acquiring the devices often considered essential.  \\subsection{Bandwidth} \\label{section-bandwidth}  \\begin{table*}[tp] \\caption{Device and Transport Mechanism Speed} \\begin{center} \\begin{tabular}{lrrrr} \\hline Device                  &\\multicolumn{4}{c}{Bandwidth (Mpixel/s)}\\\\ &Raw    &Current Read  &Current Write  &Maximum    \\\\ \\hline Slow single disk        &4      &12            &3              &13         \\\\ Fast single disk        &7      &12            &3              &23         \\\\ Fast RAID disk array    &25     &12            &3              &83         \\\\ $50 \\times$ CD-ROM      &4      &12            &               &13         \\\\ $4 \\times$ CD-R         &0.3    &              &1              &1          \\\\ DAT DDS-3               &0.6    &2             &2              &2          \\\\ DLT                     &2.5    &8             &3              &8          \\\\ 56 kbps link            &0.003  &0.01          &0.01           &0.01       \\\\ 10 Mbps link            &0.5    &1.7           &1.7            &1.7        \\\\ 100 Mbps link           &5      &12            &3              &17         \\\\ 1 Gbps link             &50     &12            &3              &150        \\\\ \\hline \\end{tabular} \\end{center} \\end{table*}  If images can be compressed and decompressed sufficiently quickly, compression increases the effective bandwidth of a device or transport mechanism by a factor equal to the inverse of the compression ratio. Thus, it may be faster to store images in their compressed form, even though there is an overhead in compressing or decompressing them.  Table 4 shows for several devices or transport mechanisms the raw bandwidths and the current and maximum effective bandwidths for compressed data, assuming a compression ratio of 0.3 appropriate for compression with qcomp/gzip with $q=1$. The maximum bandwidths assume an arbitrarily fast processor, so the overheads of compression and decompression drop to zero. The current read and write bandwidths were measured or estimated for qcomp/gzip on an early-2002 computer with single 1.9 GHz Pentium IV processor, which can compress at about 3 Mpixel/s and decompress at about 12 Mpixel/s. (A similar table could be constructed for qcomp/bzip2; the maximum effective bandwidths would be higher and the current effective bandwidths would be lower, except for the 56 kbps link which would have a current effective bandwidth of 0.015 Mpixel/s.)  Table 4 shows that even now compression results in an improvement in effective bandwidth for CD-ROMs, tapes, single disks, and all but the very fastest network connections. As processors become faster, the actual bandwidths will approach the maximum bandwidths, and, for fast enough processors, all devices will benefit. For example, achieving 83 Mpixel/s read bandwidth from a fast RAID array will require processors only 7 times faster than a 1.9 GHz Pentium IV; if processors continue to double in speed every 18 months, according to Moore's law, such processors should be available in about 2006. Achieving 83 Mpixel/s write speed will require processors 28 times faster which should be available in about 2009. Alternatively, if we allow 8-way parallelism (see \\S~\\ref{section-speed}), such speeds should be possible now for reading and in about 2005 for writing. (Again assuming 8-way parallelism, qcomp/bzip2 should give 125 Mpixel/s read bandwidth around 2006 and 125 Mpixel/s write bandwidth around 2008.)   \\label{section-summary}  A lossy method for compressing raw images has been presented. The method is very simple; it consists of lossy quantization (resampling in brightness) followed by lossless compression. The degree of quantization is chosen to eliminate the low-order bits that over-sample the noise, contain no information, and are difficult or impossible to compress.  The method is lossy but gives certain guarantees about the distribution of differences between the compressed and original images. In particular, it gives guarantees on the maximum absolute value, the expected mean value, and the expected RMS value of the difference. These guarantees make it suitable for use on raw data.  The method consistently reduces images to 1/5 of their original size while changing no value by more than 1/2 of a standard deviation in the background. This is a dramatic improvement on the compression ratios achieved by lossless compression. The method is adequately fast on current computers and would be relatively simple to parallelize.  A key feature of the method is that data can be uncompressed using tools that are widely available on modern workstations, which means that one can distribute compressed data and expect that it can be used without the need to install specialized software. This is achieved by writing the quantized image as a normal FITS file and compressing it with gzip and bzip2, which widely available general-purpose compression tools. It appears that bzip2 is compressing the data within a few tens of percent of optimally.  The next step in the development of this method is real-world testing with compressed raw data to ensure that the method does not degrade the results of astronomical analyses. Such end-to-end tests will be presented in Paper II (Watson, in preparation)."
    },
    "0208/hep-ph0208072_arXiv.txt": {
        "abstract": "\\vspace{1cm}  Since the thermal history of the Universe is unknown before the epoch of primordial nucleosynthesis, the largest temperature of the radiation dominated phase (the reheating temperature) might have been as low as 1 MeV.  We perform a quantitative study of supersymmetric dark matter relic abundance in cosmological scenarios with low reheating temperature. We show that, for values of the reheating temperature smaller than about 30 GeV, the domains of the supergravity parameter space which are compatible with the hypothesis that dark matter is composed by neutralinos are largely enhanced. We also find a lower bound on the reheating temperature: if the latter is smaller than about 1 GeV neutralinos cannot be efficiently produced in the early Universe and then they are not able to explain the present amount of dark matter. ",
        "introduction": "\\label{sec:intro}  Supersymmetric dark matter provides one of the hottest topics at the border between Cosmology and Particle Physics. This is due to the fact that in $R$--parity conserving Supersymmetric theories the Lightest Supersymmetric Particle (LSP) is stable and may provide the cold dark matter, whose existence is inferred by a large number of independent observations \\cite{DM,cosmo_params}. Among the different supersymmetric candidates, the neutralino turns out to be a perfect dark matter particle, since it has neither charge nor colour, its only interactions being of the weak type.  The present abundance of neutralinos depends on the thermal history of the Universe. In the early Universe interactions may keep neutralinos in thermal equilibrium with the radiation bath until their abundance freezes out at a temperature $T_F$. The neutralino mass is constrained by accelerator data to be heavier then a few tens of GeV. This implies that it decouples in the early Universe when it is non relativistic, at $T_F$ in the GeV range.  This picture is correct if the maximum temperature in the radiation-dominated era, which from now on we will refer to as the reheating temperature $T_{RH}$, is much larger than the freeze-out temperature $T_F$. If this is the case, the neutralino relic abundance turns out to be: \\begin{equation} \\Omega_{\\chi} h^{2}\\equiv \\frac{\\rho_{\\chi}}{\\rho_c}h^2 \\propto \\frac{2.6 \\cdot 10^{-10} {\\rm GeV^{-2}}}{\\langle \\sigma_{\\rm ann} v \\rangle} \\label{eq:omega_intro} \\end{equation} where $\\rho_c\\equiv 1.8791 h^2 \\times 10^{-29}$ g cm$^{-3}$ is the critical density, $h$ is the Hubble constant in units of 100 km sec$^{-1}$ pc$^{-1}$, while $\\sigma_{\\rm ann}$ is the WIMP pair--annihilation cross section, $v$ is the relative velocity between the two annihilating particles, and brackets denote thermal average. Indeed, what specifically makes the neutralino an ideal dark matter candidate is that in Eq. (\\ref{eq:omega_intro}) the value of the annihilation cross section, calculated in a wide variety of susy models, may yield a result for $\\Omega_\\chi h^2$ which falls in the correct interval suggested by present day observations for the amount of non--baryonic dark matter in the Universe\\cite{DM,cosmo_params}: \\begin{equation} 0.05 \\lsim \\Omega_{\\rm M} h^2 \\lsim 0.3. \\label{eq:omega_interval} \\end{equation}  When exploring the neutralino parameter space and its chances of discovery both in accelerator and non--accelerator searches, this argument has been usually turned the other way around, and Eq. (\\ref{eq:omega_interval}) used as a constraint on the neutralino parameter space. Depending on the particular supersymmetric scenario, this may have important consequences on the allowed supersymmetric configurations.  In particular, in Supergravity--inspired models (SUGRA), the allowed neutralino parameter space turns out to be severely reduced by requiring that $\\Omega_\\chi h^2$ falls inside the interval defined by Eq. (\\ref{eq:omega_interval}) \\cite{Berezinsky:1995cj,Berezinsky:1996ga,noi_sugra,Ellis.HOW,Ellis.FGO,Ellis:2002wv,Ellis.more,Feng,Arnowitt,Nath,Roszkowski}.  The robustness of these constraints relies on the cosmological assumptions that lead to Eq. ({\\ref{eq:omega_intro}). Indeed, the thermal history of the Universe before the epoch of nucleosynthesis is unknown. The maximum temperature in the radiation-dominated era $T_{RH}$ may have been as low as 1 MeV (but not smaller in order not to spoil the nucleosynthesis predictions).  The possibility of a low reheating temperature of the Universe has been recently discussed in Ref. \\cite{Giudice:2000ex}. There it was shown that a low reheating temperature has important implications for many topics in cosmology such as axion physics, leptogenesis and nucleosynthesis constraints on decaying particles. In particular, it was shown that stable weakly interacting massive particles may be produced even if the reheating temperature is much smaller than the freeze-out temperature of the dark matter particles, $T_{RH}< T_F$, and that the dependence of the present abundance on the mass and the annihilation cross section of the dark matter particle may differ drastically from standard results\\footnote{Low reheating scenarios lead as well to a new perspective on baryogenesis \\cite{Davidson:2000dw} and to the possibility that massive neutrinos may play the role of warm dark matter \\cite{Giudice:2000dp}.}.  The goal of this paper is twofold: first, we wish to extend the analysis of Ref. \\cite{Giudice:2000ex} and perform a quantitative study of the case of neutralinos in SUGRA scenarios, analyzing in detail the impact that a low $T_{RH}$ may have for the present neutralino relic abundance; secondly, we aim at providing a lower bound on the reheating temperature. The logic is the following.  All matter is produced at the end of inflation \\cite{reviewinf} when all the vacuum energy stored into the inflaton field is released and the Universe becomes radiation-dominated with the initial temperature $T_{RH}$ .  During the reheating process, particles are generated through thermal scatterings and quickly thermalize.  Among them, dark matter particles may be also produced but their final number depends strongly on the reheating temperature. If the latter is too small, the thermal bath does not give rise to a number of neutralinos large enough to make them good candidates for dark matter. This leads to a lower limit on $T_{RH}$. We will find that the reheating temperature needs to be larger than about 1 GeV for neutralinos to be good dark matter candidates\\footnote{In this paper we suppose that neutralinos are produced during the reheating process only through thermal scatterings. Another source might be the direct decay of the inflaton field into neutralinos \\cite{Moroi:1999zb}. This introduces though another unknown parameter, the decay rate of the inflaton field into neutralinos, and we do not consider this possibility any further.}.   The plan of the paper is as follows. In Section II we briefly recall the calculation of a WIMP relic abundance both in the low--reheating scenario and in the standard case. In Section III the specific supersymmetric models which will be considered in our analysis are introduced. In Section IV we will discuss our results. Section V is devoted to our conclusions. ",
        "conclusions": "\\label{sec:conclusions}  In standard cosmology it is usually assumed that the temperature $T_{RH}$ of the Universe at the beginning of the radiation--dominated era is much higher than the supersymmetry breaking scale. Moreover, neutralinos decouple from the thermal bath after the reheating phase, which followed the end of inflation, has terminated. Under these assumptions, the allowed parameter space of SUGRA models turns out to be severely constrained by the requirement that the neutralino relic density does not exceed the maximal value of the matter density of the Universe deduced from observations.  In this paper we have performed a quantitative study of the neutralino relic abundance in cosmological scenarios with a low reheating temperature \\cite{Giudice:2000ex}. This is a viable possibility since the only robust lower bound on the reheating temperature $T_{RH}$ can be set at the MeV scale, in order not to spoil nucleosynthesis predictions. The suppression on $\\Omega_\\chi h^2$ is originated by the fact that neutralinos decouple from the thermal bath before the end of the reheating phase. In this case the neutralino number density is diluted by entropy production and by a higher expansion rate than in the radiation--dominated era.  For values of $T_{RH} \\lsim 30$ GeV the domains of the SUGRA parameter space which are compatible with dominant relic neutralinos are largely enhanced with respect to the standard cosmological case. These domains depend on $T_{RH}$ and we have shown their evolution as a function of the reheating temperature. For $T_{RH} \\lsim 1$ GeV all the SUGRA parameter space becomes compatible with the bounds on the dark matter relic abundance (even though the neutralino relic abundance for these low values of $T_{RH}$ is strongly suppressed).  Since lower $T_{RH}$ imply smaller relic densities, the assumption that neutralinos provide a major contribution to the dark matter of the Universe implies a lower limit on $T_{RH}$. This constraint ranges from 0.6 GeV for neutralino masses of the order of few tens of GeV, up to 20 GeV for neutralino masses in the TeV range. This bound on $T_{RH}$, subject to the request of explaining the dark matter content of the Universe only in terms of relic neutralinos in SUGRA schemes, is much stronger than the limit on $T_{RH}$ coming from nucleosynthesis. Similar conclusions occur also for non--universal SUGRA models."
    },
    "0208/nucl-th0208067_arXiv.txt": {
        "abstract": "The possibility of structured mixed phases at first order phase transitions in neutron stars is re-examined by taking into account charge screening and surface effects. The transition from the hadron $npe$ phase to the quark phase is  studied. Two possibilities, the mixed phase and  two separate phases given by the double-tangent (Maxwell) construction are considered. Inhomogeneous profiles of the electric potential and their contribution to the energy are analytically calculated. The electric field configurations determine the droplet size and the geometry of structures. ",
        "introduction": "It is now commonly accepted that different phase transitions may occur in neutron star interiors. The possibilities of pion and kaon condensate states, and quark matter state were studied by many authors during the last thirty years, see \\cite{M78,MSTV90,TPL,T95} and refs therein. It has been argued that these transitions are of first order. They were described in terms of two spatially separated phases using the Maxwell construction. Glendenning raised the question whether mixed phases in systems composed of charged particles exist instead of the configuration described by the Maxwell construction \\cite{G92}. In particular, the possibilities of hadron - kaon condensate ($npe$ - $npeK_{cond}$) and hadron - quark (H-Q) mixed phases were discussed. The existence of such kind of mixed phases in dense neutron star interiors would have important consequences for the equation of state, also affecting neutrino emissivities \\cite{RBP00}, glitch phenomena and $r$ modes, cf. \\cite{BGP00,G01}.  Basing on the validity of the Gibbs conditions, in particular on the equality of the electron chemical potentials of the phases, refs \\cite{G92,GS99,CG00,G01} further argued that the Maxwell construction is {\\em{always}} unstable. It is due to the inequality of the electron chemical potentials of two phases and, thus, due to a possibility for particles to fall down from the higher energetic levels characterized by the higher electron chemical potential of the one phase to the lower levels of the other phase. Thereby, refs \\cite{G92,GS99,CG00,CGS00,G01} argued that, if first order phase transitions, such as quark, kaon condensate and pion condensate transitions, indeed, occur in neutron star interiors, the existence of a wide region of the mixed phase is inevitable. On explicit examples of $npe$ - $npeK_{cond}$ and H-Q phase transitions authors demonstrated the energetic preference of the mixed phase.  However, strictly speaking, such an argumentation is in contradiction to the results of some other works. It has been recently observed that in some models the Gibbs condition of equality of electron chemical potentials of two phases can't be fulfilled at all \\cite{PREPL00,MYTT,ARRW}, whereas conditions for the Maxwell construction are fulfilled. The latter construction is stable in these specific cases. A critics of the bulk calculations which ignore finite size effects was given in ref. \\cite{NR00}. To include these effects authors used a relaxation procedure in which they start with an initial guess for the shape of the electric potential in the Wigner-Seitz cell, solve for the kaon equation of motion to obtain the charged particle profiles, and then recalculate the electric potential using the new profiles. This is then repeated until convergence. Authors found that  with inclusion of inhomogeneity effects the region of the kaon condensate mixed phase is significantly narrowed compared to that obtained using standard Gibbs conditions, disregarding finite size effects. Please notice that in order the scheme to be completely self-consistent the electric field, as the new degree of freedom, should obey the own equation of motion, and it enters equations of motions of other fields (including protons and electrons) via the gauge shift of the chemical potentials of all charged particles. All the charged particle densities are affected by the inhomogeneous electric field even in the case, if one artificially suppressed the contribution of the electric field to the charged kaon density.  For small size droplets, the problem of the construction of the mixed phase is analogous to that studied somewhat earlier for matter at sub-nuclear densities, cf. \\cite{RPW83}. The possibility of the structure, as well as its geometry, are determined by competition between the Coulomb energy and the surface energy of droplets. Ref. \\cite{HPS93} applied these ideas to the description of the mixed phase for the H-Q phase transition. The Coulomb plus surface energy per droplet of the new phase always has a minimum as a function of the droplet radius. On the one hand, this radius should be not too small in order for the droplet to have rather large baryon number ($A\\gg 1$)\\footnote{This condition, $A\\gg 1$, is assumed to be fulfilled. Otherwise quantum effects, like shell effects, may affect the consideration.} and, on the other hand, it should be not too large (less than the Debye screening length) in order one could ignore  screening effects. The Debye screening length for the quark matter was evaluated as $\\lambda_{\\rm D} \\simeq 5~$fm. The value of the droplet radius $R_{\\rm min}$ depends on a poorly known surface tension parameter $\\sigma$.  With a small value of the surface tension $\\sigma \\simeq 10~$ MeV$\\cdot$fm$^{-2}$ the droplet size was  estimated as $R_{\\rm min} \\geq 3.1~$fm and with $\\sigma \\simeq 100~$ MeV$\\cdot$fm$^{-2}$, as $R_{\\rm min} \\geq 6.6~$fm $>\\lambda_{\\rm D}$. Thereby, it was argued that the mixed phase is not permitted in the latter case. Corrections  to the Coulomb solutions\\footnote{By the Coulomb solutions we call solutions with the charged density profiles in the form of combination of step-functions. Screening effects are disregarded in this case.} due to screening effects were not considered, although one could intuitively expect that for such a narrow interval of available values of droplet radii, being of the order of the Debye screening length, the screening may significantly affect the results. Also the effect of the inhomogeneity of the field profiles in the strong interaction part of the energy was disregarded. Therefore, further detailed study of the effects of inhomogeneous field profiles, such as screening and surface effects, seems to be of prime importance.      The problem of the construction of the charged density profiles by taking into account screening effects is in some sense analogous to that considered previously in refs \\cite{MPV77,VSCh77,VCh78} for abnormal pion condensate nuclei. Results can be used also for the description of the charge distributions in strangelets and kaon condensate droplets.  The value of the surface tension is poorly known. It has the meaning only if there exists a related shortest scale in the problem, being much smaller than the typical scale of the change of the electric field. In the case of the  kaon condensate phase transition there is no necessity to introduce the surface tension.  One can explicitly solve the equations of motion for all  the mean meson fields, cf. refs \\cite{CGS00,NR00}. However one should also include the equation of motion for the electric field. The existence of the additional typical scale $\\lambda_{\\rm D}$ appearing in this equation may affect the solution. In the case of the H-Q phase transition there is a natural minimal scale in the problem related to the confinement $\\lsim 0.5$~fm. Since the latter scale is much shorter than $\\lambda_{\\rm D}$ the introduction of the surface tension is well established in the given case.  With this paper we show that, firstly, the Maxwell construction does not contradict  the Gibbs condition of equality of the electron chemical potentials, if one properly incorporates the electric potential. Secondly, inclusion of finite size effects, such as screening and surface tension, is crucial for understanding the mixed phase existence and its description. We will consistently incorporate these effects. Part of the results of this paper was announced in the Letter \\cite{VYT01}.   Our consideration is rather general and main effects do not depend on what the concrete system is studied. However we need to consider a concrete example to demonstrate the quantitative effects. For that aim we concentrate on the description of the H-Q phase transition. Thus we avoid the discussion of other possibilities, such as the charged pion condensation, charged kaon condensation, see \\cite{MSTV90,G01} and refs therein. We will consider the simplest case of the H-Q phase transition between normal phases. Thus we also avoid the consideration of the diquark condensates and diquark condensates together with the pion and kaon condensates, see \\cite{SS00} and refs therein. With the inclusion of the diquark superfluid gap, the corrections to the effective energy functional are expected to be as small as $(\\Delta /\\mu_B )^2$, where $\\Delta$ is the pairing gap and $\\mu_B$ is the baryon chemical potential, cf. \\cite{ARRW,IB02}. However the response to the electric field can be a more specific. E.g., in the case of the phase transition to the color-flavor-locked (CFL) phase the quark sub-system is enforced to be electrically neutral without electrons, cf. \\cite{ARRW}. There can be even more involved effects, which we do not consider, like the modification of the quark and gluon condensates with the increase of the baryon density, a modification of masses and widths of all particles, that needs a special quantum treatment, etc.         In sect. \\ref{gcap} we critically review the application of the Gibbs conditions to the finite size structures and we discuss the stability of the Maxwell construction. Since the problem of interpreting the Gibbs conditions and the Maxwell construction is related to the counting of the electric potential from different levels we discuss different choices of the gauge in  sect. \\ref{gcap}. Starting from  sect.  \\ref{gen} we present our results in an arbitrary gauge. In sect.  \\ref{gen} we develop a general formalism based on the thermodynamic potential. It serves  as a generating functional, variation of which in the corresponding independent variables determines the equations of motion. We demonstrate how one can explicitly treat electric field effects.  Chemical equilibrium conditions and the charged particle densities are presented in sect. \\ref{Eq}. In sect. \\ref{sph}  we analytically solve the equation of motion for the electric potential of the new phase droplets placed in the Wigner-Seitz lattice and assuming spherical geometry we recover corresponding contributions to the energy and thermodynamic potential. Slab structures are studied in sect. \\ref{plane}. Sect. \\ref{Comparison} discusses which mixed phase structures, spherical droplets or slabs, are realized in dependence on the value of the quark fraction volume. The discussion on the specifics of the description of the H-Q system is deferred to the Appendices. In Appendix \\ref{en} we introduce the model to calculate the volume part of the energy of quark and hadron phases. In Appendix \\ref{surf} we discuss surface effects. In Appendix \\ref{Free}  we develop  a formalism based on the Gibbs potential, as generating functional. In Appendix \\ref{example} we study the role of the nonlinear correction terms. In Appendix \\ref{Coul-pec} we discuss peculiarities of the Coulomb limit.  All over the paper we use units $\\hbar =c=1$. ",
        "conclusions": "Summarizing, in this paper we have discussed  the possibility of the presence of structured mixed phases at first order phase transitions in multi-component systems of charged particles.  Using the thermodynamical potential as the generating functional we have developed a formalism, which allows to explicitly treat electric field effects including the Debye screening.  We have studied a ``contradiction'' between the Gibbs conditions and the Maxwell construction extensively discussed in previous works. We have demonstrated that this contradiction is resolved if one takes into account the difference in the treatment of the chemical potentials used in the two approaches. The different values of the electron chemical potentials in the Maxwell construction and the ones used in the Gibbs conditions do not contradict each other if one takes into account the electric field arising at the boundaries of the finite size structures of the mixed phase and at the boundary between phases within the Maxwell construction.  As an example, our formalism was applied to the hadron-quark structured mixed phase. Using a linearization procedure we analytically solved the Poisson equation for the electric potential and found the electric field profiles of the structured mixed phase in cases of spherical and one-dimensional geometries. We have demonstrated that the charge screening effect greatly modifies the description of the mixed phase, changing its parameters and affecting the possibility of its existence. With screening effects included it has been proven that the  mixed phase structures may exist even if their sizes are significantly larger than the Debye screening lengths. The characteristics of the structures essentially depend on the value of the ratio of the Debye screening lengths in the interior and exterior of the structure and on the value of the surface tension on its boundary. We showed the important role played by the correlation energy terms omitted in previous works. For realistic values of the parameters these correlation contributions are larger than the electromagnetic contribution to the energy discussed in previous works.  The result, which structure is energetically favorable, essentially depends on the values of parameters $\\alpha_0$ and $\\beta_1$. This our conclusion essentially differs from that was previously done for the Coulomb case, when one ignored the screening effects for any $\\beta_1$ and assumed the step-function charge-density profile. We recover that limit case result from our general expressions as well, but only for tiny values of $\\beta_1$ and $\\alpha_0 =1$.     We support the conclusion of \\cite{RPW83,HPS93} that the mixed phase may arise only if the surface tension is smaller than a critical value. We also evaluated the surface tension between the hadron and the quark phases. However this evaluation remains  rather rough. An uncertainty in the value of the surface tension does not allow to conclude whether the mixed H-Q phase exists or not. Both possibilities, the mixed phase and the Maxwell construction, can be studied within different models. An advantage of our study is that the characteristics of the mixed phase are expressed in terms of only few parameters. Thus knowing these parameters for a given model one may apply our results for the discussion of the mixed phase, the Maxwell construction, the geometry of the structures, the energy contributions of the structures and the electric field profiles.  In absence of a mixed phase our charged distributions describe the boundary layer between two separated phases existing within the double-tangent (Maxwell) construction. The previously obtained Coulomb limits (when one omitted the screening effects) are naturally recovered from our general expressions but only in the limiting case of tiny size droplets. This limiting case, however, is never realized for realistic values of the parameters.  In this paper we considered only droplets and slabs and their interplay for $f<0.5$ at zero temperature. More detailed analysis of a possible interplay between different structures within the whole interval $0<f<1$ and at finite temperatures will be performed in a forthcoming paper.     \\paragraph{Acknowledgements.}  Research of D.N.V. at Yukawa Institute for Theoretical Physics was sponsored by the COE program of the Ministry of Education, Science, Sports and Culture of Japan. D.N.V. acknowledges this support and kind hospitality of the Yukawa Institute for Theoretical Physics. He thanks Department of Physics at Kyoto University for the warm hospitality and also acknowledges hospitality and support of GSI Darmstadt. The present research of T.T. is partially supported by the REIMEI Research Resources of Japan Atomic Energy Research Institute, and by the Japanese Grant-in-Aid for Scientific Research Fund of the Ministry of Education, Culture, Sports, Science and Technology (11640272, 13640282). Authors greatly acknowledge E.E. Kolomeitsev for the help and discussions and M. Lutz for valuable remarks.  \\newpage \\appendix \\begin{center} \\begin{eqnarray} \\nonumber\\\\\\nonumber \\end{eqnarray} {\\bf{APPENDICES}}  \\end{center}"
    },
    "0208/astro-ph0208415_arXiv.txt": {
        "abstract": "We report the first measurement of proper motions in the SN~1006 remnant (G327.6+14.6) based entirely on digital images. CCD images from three epochs spanning a period of 11 years are used: 1987 from Las Campanas, and 1991 and 1998 from CTIO. The filaments in SN~1006 are non-radiative, appearing only in the Balmer lines of hydrogen.  Delicate filaments---probably thin sheets seen in projection---delineate the shock front along the northwest rim of the remnant.  We use the center of curvature of the filaments to define a convenient geometry, and integrate along short segments to obtain one-dimensional profiles.  These remain unchanged, within seeing and statistical errors, from one epoch to another.  Measuring the shift in a direction perpendicular to the filaments, we obtain proper motions of $280 \\pm 8 \\masy$\\ along the entire length where the filaments are well defined, with little systematic variation along the filaments.  In addition to the measurements of the well-defined filaments in the NW of SN~1006, we also report very deep \\ha\\ imaging observations of the entire remnant that clearly show very faint emission surrounding almost the entire  shell, as well as some diffuse emission regions in the (projected) interior.  Combining the proper motion measurement with a recent measurement of the shock velocity based on spectra of the same filaments by Ghavamian \\etal\\ leads to a distance of $2.17\\pm 0.08\\; {\\rm kpc}$ to SN~1006. Several lines of argument suggest that SN~1006 was a Type Ia event, so the improved distance measurement can be combined with the peak luminosity for SNe~Ia, as recently determined for events in galaxies with Cepheid-based distances, to calculate the apparent brightness of the spectacular event that drew wide attention in the eleventh century.  The result, $V_{max} = -7.5 \\pm 0.4$, lies squarely in the middle of the wide range of estimates based on the historical observations. ",
        "introduction": "Early in the eleventh century, the supernova of 1006 C.E. attracted wide attention, and it is probably the brightest stellar event in recorded human history \\citep{ste77}.  Almost a millennium later, the supernova remnant (SNR) that this event produced has gradually emerged as an object of great astrophysical interest. The remnant of SN~1006, G327.6+14.6, was first identified through its radio emission by \\citet{gar65}.  X-ray emission was detected by \\citet{win76}, and very soon afterward \\citet{vdb76} reported faint optical filaments associated with SN~1006.   More recent radio maps \\citep{rey86, mof93} show a somewhat barrel-shaped shell, $30\\arcmin$\\ in diameter and brightest along the NE and SW limbs. X-ray images \\citep[][hereafter WL97]{wil96, win97a} show a similar structure to that seen in the radio, while optical emission is limited almost entirely to delicate filaments extending some $15\\arcmin$\\ along the NW rim of the remnant shell---a region that is generally faint in both X-rays and radio. The optical spectrum shows only Balmer lines of hydrogen \\citep{sch78}, indicative of a nonradiative shock  propagating into partially neutral interstellar material \\citep[\\eg][]{mck80}.  \\begin{table*}[t] \\include{tab1a} \\end{table*}  Renewed interest in the  SN~1006 remnant was stimulated with the discovery by \\citet{koy95}, based on {\\it ASCA} data, that the strongest X-ray emission, from the NE and SW limbs, has a hard, featureless, non-thermal spectrum characteristic of synchrotron radiation, while the X-rays from the interior and from the NW and SE limbs are much softer and are thermal in character.  These data strongly suggested that in the NE and SW, electrons are accelerated to TeV energies by the first-order Fermi mechanism in the supernova shock. Furthermore, since the same mechanism will accelerate nuclei as well, the \\citet{koy95} data provided the first solid observational evidence that supernova shocks produce cosmic rays.  The supernova-cosmic ray connection was solidified with the demonstration of detailed correspondence between X-ray and radio morphology along the NE and SW limbs of SN~1006 (WL97) and especially by the discovery of $\\gamma$-ray emission from the NE limb of SN~1006 by \\citet{tan98}.   The optical filaments of SN~1006 provide a rich opportunity to investigate the nature of the nonradiative shock and the environment into which it propagates. The Balmer line profile consists of two components: a broad component that results from neutral atoms that penetrate the shock and then undergo charge exchange with hot, post-shock protons to produce a population of hot neutrals, and a narrow component from simple collisional excitation of the ambient neutral atoms \\citep{che78, che80}.  \\citet{gha02} present detailed spectra and models  that lead to a precise measurement of the shock velocity in the NW region of SN~1006: $v_s = 2890\\pm 100\\ \\kms$.   The present paper presents a measurement of the proper motion of the NW filaments of SN~1006 with a precision of $\\sim 2\\%$, comparable to that of the recent shock velocity measurement. The geometry of the NW filaments---apparently thin sheets seen almost exactly edge-on---is ideally suited to measuring the remnant's expansion. Proper motion of the optical filaments in SN~1006 was first reported only five years after their discovery \\citep{hes81}, but a reasonably precise measurement had to wait for a longer baseline. \\citet[hereafter LBV88]{lon88} combined first-epoch photographic plates and CCD images  taken 11 years later to obtain a measurement of $0.30 \\pm 0.04\\asy$.  We use the same 1987-epoch CCD image used by LBV88, together with subsequent CCD images from epochs 1991 and 1998, to obtain a much more precise measurement---the first relying entirely on digital images.   Section 2 of the paper describes the observations themselves.  Section 3 discusses two different algorithms we have employed to measure the filamentary motion and presents the results.  Section 4 presents new, deep observations of very faint \\ha\\ emission that outlines the entire shell of the SN~1006 remnant, as first reported by WL97.  In the final three sections we discuss some implications of the present measurements, including a comparison with proper motion measurements carried out in other regimes of the electromagnetic spectrum.  The combination of our new proper-motion measurements and the shock-velocity determined by \\citet{gha02} leads directly to a distance of $2.17\\pm 0.08\\; {\\rm kpc}$\\  pc to SN~1006. This distance, together with recent determinations of the peak absolute magnitudes for Type Ia supernovae, provides an estimate of the apparent brightness of SN~1006 at maximum that is significantly more precise than earlier ones. ",
        "conclusions": ""
    },
    "0208/astro-ph0208309_arXiv.txt": {
        "abstract": "At the heart of a successful theory of galaxy formation must be a detailed physical understanding of the dissipational processes which form spiral galaxies.  To what extent can we unravel the events that produced the Galaxy as we see it today?  Could some of the residual inhomogeneities from prehistory have escaped the dissipative process at an early stage?  To make a comprehensive inventory of surviving inhomogeneities would require a vast catalog of stellar properties that is presently out of reach.  The Gaia space astrometry mission, set to launch at the end of the decade, will acquire detailed phase space coordinates for about one billion stars, within a sphere of diameter 20 kpc -- the Gaiasphere.  Here we look forward to a time when all stars within the Gaiasphere have complete chemical abundance measurements (including $\\alpha$, $s$ and $r$ process elements).  Even with such a vast increase in information, there may exist fundamental $-$ but unproven $-$ limits to unravelling the observed complexity. ",
        "introduction": "Eddington once remarked that `the contemplation in natural science of a wider domain than the actual leads to a far better understanding of the actual.' Our intuition is that any dynamical phase early on in the history of the Galaxy which is dominated by relaxation or dissipation probably removes more information than it retains.  This remains largely true for controlled experiments within terrestrial laboratories for the reason that it is exceedingly difficult to track fluid particles, for example, in order to unravel the process under study.  However, at least in principle, the situation may be more tractable in astrophysics in that individual stars carry information about the birth site at the time of their formation.  Nature offers us a vast untapped reservoir of information in the detailed chemical abundances stored within stellar atmospheres.  The fossil evidence of galaxy formation is not restricted to chemical abundances alone. In our review, ``The New Galaxy -- Signatures of its Formation,'' we stress the importance of new observations across a wide range of parameter space (Freeman \\& Bland-Hawthorn 2002).  We are coming into a new era of galactic investigation, in which one can study the fossil remnants of the early days of the Galaxy in a broader and more focussed way, not only in the halo but throughout the major luminous components of the Galaxy.  The goal of these studies is to reconstruct as much as possible of the early galactic history.  What do we mean by the reconstruction of early galactic history? We seek a detailed physical understanding of the sequence of events which led to the Milky Way. Ideally, we would want to tag (i.e. associate) components of the Galaxy to elements of the protocloud -- the baryon reservoir which fueled the stars in the Galaxy.  Our approach is anchored to observations of the Galaxy, interpreted within the broad scope of the CDM hierarchy. Many of the observables in the Galaxy relate to events which occurred long ago, at high redshift. Fig.~\\ref{fig1} shows the relationship between look-back time and redshift in the context of the $\\Lambda$CDM model: the redshift range (z $\\lta$ 6) of discrete sources in contemporary observational cosmology matches closely the known ages of the oldest components in the Galaxy.  The Galaxy (near-field cosmology) provides a link to the distant universe (far-field cosmology).  Here we address the prospect of chemically `tagging' stars to common sites of formation, in order to reconstruct star clusters which have long since dispersed.  High resolution spectroscopy at high signal-to-noise ratio of many stellar types reveals an extraordinarily complex pattern of spectral lines.  The spectral lines carry key information on element abundances that make up the stellar atmosphere. Some of these elements cannot arise through normal stellar evolution, and therefore must reflect conditions in the progenitor cloud at the time of its formation. A cornerstone of near-field cosmology must be to explore how much of the Galaxy's past can be reconstructed from the chemical signatures in stellar atmospheres. ",
        "conclusions": ""
    },
    "0208/astro-ph0208553_arXiv.txt": {
        "abstract": "During a 9-month campaign (1996--1997), the Midcourse Space Experiment (MSX) satellite mapped the Galactic Plane at mid-infrared wavelengths (4.3--21.3$\\mu$m).  The much greater spatial resolution and sensitivity of the MSX 8.3$\\mu$m band compared to IRAS at 12$\\mu$m has revealed diffuse emission not seen before. Here we report evidence for a spectacular limb-brightened, bipolar structure at the Galactic Center extending more than a degree (170~pc at 8.0~kpc) on either side of the plane. The 8.3$\\mu$m filamentary emission appears confined to a dome-shaped shell which extends perpendicular to the plane, and shows a tight correlation with the 3, 6 and 11~cm continuum emission over the same scales.  The most likely scenario is that the extraplanar 8.3$\\mu$m emission arises from dust entrained in a large-scale bipolar wind powered by a central starburst.  The inferred energy injection at the source is $\\sim$10$^{54}/\\kappa$ erg for which $\\kappa$ is the covering fraction of the dusty shell ($\\kappa \\lta 0.1$).  In accord with the stellar record in the Galactic Center, we infer that a powerful nuclear starburst has taken place within the last several million years.  The inferred energies require $\\gta 10^4/\\varepsilon$ supernovae, where $\\varepsilon$ is the thermalization efficiency of the supernova ejecta.  There is observational evidence for a galactic wind on much larger scales, presumably from the same central source which produced the bipolar shell seen by MSX.  Sofue has long argued that the North Polar Spur -- a thermal x-ray/radio loop which extends from the Galactic Plane to $b = +80$\\deg -- was powered by a nuclear explosion ($1-30\\times 10^{55}$ erg) roughly 15~Myr ago, although his published models did not explain the projected loop structure.  We demonstrate that an open-ended bipolar model when viewed in projection {\\it in the near field} provides the most natural explanation for the observed loop structure. The ROSAT 1.5~keV diffuse x-ray map over the inner 45\\deg\\ provides compelling evidence for this interpretation. Since the faint bipolar emission would be very difficult to detect beyond the Galaxy, {\\it the phenomenon of large-scale galactic winds may be far more common than has been observed to date.} The derived energetics for the bipolar wind are of order $10^{55}$ erg, in line with the energy requirement of the MSX observations.  We infer that the Galactic Center is driving large-scale winds into the halo every $\\sim 10-15$~Myr or so. ",
        "introduction": "Nuclear starburst-driven winds are now observed in dozens of external galaxies across the electromagnetic spectrum (Strickland 2001; Martin 1999; Heckman 2002; Veilleux 2002). For nearby galaxies, detailed studies reveal that the superwind energetics are in the range 10$^{53}$ to 10$^{56}$ erg.  Galaxy-scale winds are suspected to be important in both galaxy evolution and for the state of the intergalactic medium. In the Galaxy, early enrichment of the halo and the outer disk through powerful nuclear winds may account for the observed abundances of the thick disk and globular clusters (Freeman \\& Bland-Hawthorn 2002). Moreover, nuclear winds can regulate the growth of the central bulge (Carlberg 1999) and moderate the evolution of dwarf satellites (Irwin\\etal\\ 1987).  At a distance of only 8.0$\\pm$0.5 kpc (Reid 1993), the Galactic Center shows a remarkable range of energetic activity at infrared, radio, x-ray and $\\gamma$-ray wavelengths (Morris \\& Serabyn 1996 -- hereafter MS; Yusef-Zadeh\\etal\\ 2000; Cheng\\etal\\ 1997). While this activity has proved difficult to disentangle, there is now solid evidence on scales of arcminutes to tens of degrees for powerful mass ejections from the Galactic Center.  Potential sources of energy are young star clusters or the $2-3\\times 10^6$ M$_\\odot$ central black hole (Oort 1977; Frogel 1988; Genzel\\etal\\ 1996; Ghez\\etal\\ 1998).  Individual star clusters have ages ranging from 5~Myr (Krabbe\\etal\\ 1995) to 20~Myr (Eckart\\etal\\ 1999; Figer\\etal\\ 2000; Gezari\\etal\\ 2002).  While the star formation history is undoubtedly complicated, there is now abundant evidence that the Galactic Center has experienced starburst episodes at different times in the past (e.g.  Tamblyn \\& Rieke 1993; Sjouwerman\\etal\\ 1998; Simpson\\etal\\ 1999).  Detailed modelling of the PAH and fine structure features (Lutz 1998) suggests a starburst event some 7~Myr ago, supported by a census of individual stars (Genzel\\etal\\ 1994; Krabbe\\etal\\ 1995; Najarro\\etal\\ 1997).  An acute shortcoming of galaxy wind studies to date has been deriving reliable energetics from multi-wavelength observations. We will argue that the proximity of the Galactic Center makes this one of the most important arenas for studying the wind process.  But current estimates of the central energetics span a huge range.  Sofue \\& Handa (1984) presented evidence of a radio lobe on scales of $\\sim 200$ pc (now referred to as the Galactic Center Lobe) with implied thermal energetics of roughly $3\\times 10^{51}$ erg. Chevalier (1992) argued for a higher energy content ($\\sim 2\\times 10^{52}$ erg) arising from stellar winds driven by hot young stars over a 30 Myr duration. Several authors have deduced that an explosive nuclear event ($4-8\\times 10^{53}$ erg) must have taken place from the high temperatures implied by the ASCA detection of 6.7~keV K$\\alpha$ line emission from He-like Fe~XXV (Koyama\\etal\\ 1996; Yamauchi\\etal\\ 1990). On larger scales, Sofue (1977; 1984; 1994; 2000) has long argued that the North Polar Spur is the byproduct of a central explosion which took place 15 Myr ago. In his model, the scale of the blast is 10$-$20 kpc, with explosive energies in the range 10$^{55}$ to 3$\\times 10^{56}$ erg.  The source of fuel for the observed activity appears to be the central molecular zone (CMZ), a molecular `ring' with radius 180~pc and total mass M$_{\\rm cmz} \\sim 8\\times 10^6$ M$_{\\odot}$.  Recent estimates of mass inflow to the Galactic Center are of order 1 M$_\\odot$ per year (MS).   This is sufficient to trigger starbursts and nuclear activity associated with powerful Seyfert galaxies ($\\sim$10$^{43}$ erg s$^{-1}$). The kinematics suggest the gas is kept on $x_2$ orbits by a strong central bar, which guarantees cloud-cloud collisions on timescales of $2\\pi/\\omega_c \\approx 10^8$ yrs.  Of particular interest here, a central explosion of order $10^{55-56}$ erg would provide a central gas mass M$_{\\rm cmz}$  with sufficient radial impulse to produce the observed ring (Sanders 1989; Saito 1990).  Sophisticated hydrodynamical simulations (e.g.  Suchkov\\etal\\ 1996; Strickland \\& Stevens 2000) show that such rings are indeed formed by central explosions and that disk material is entrained by the expanding wind.  So what are the true energetics of the Galactic Center wind, and is it a periodic or continuous flow? Is the wind driven by the central black hole or by bursts of star formation, or both?  Here, we present the first clear evidence that the nuclear wind which gave rise to the Galactic Center Lobe (GCL) is entraining large amounts of molecular gas and dust in the flow. The newly derived energetics, which are much larger than earlier estimates, are of the same order needed to explain the North Polar Spur (NPS) in terms of a central explosion, which strongly suggests that the Galactic Center undergoes periodic bursts of nuclear activity, as we show.  In \\S~2, the new MSX observations are presented and compared to existing far-infrared and radio data. In \\S~3, we find that the elevated dust temperature in the GCL requires a local heat source. In \\S~4, we derive the basic properties of the nuclear wind. In \\S~5, we revisit Sofue's earlier wind model for the NPS in light of new observations.  In \\S~6, we argue that the Galactic Center arena is of extreme importance in our quest to understand large-scale winds. ",
        "conclusions": ""
    },
    "0208/astro-ph0208079_arXiv.txt": {
        "abstract": "The recent detection of X-ray emission from HH~2 and HH~154 with the Chandra and XMM-Newton satellites (respectively) have opened up an interesting, new observational possibility in the field of Herbig-Haro objects. In order to be able to plan further X-ray observations of other HH objects, it is now of interest to be able to estimate their X-ray luminosities in order to choose which objects to observe. This paper describes a simple, analytic model for predicting the X-ray luminosity of a bow shock from the parameters of the flow (i. e., the size of the bow shock, its velocity, and the pre-shock density). The accuracy of the analytic model is analyzed through a comparison with the predictions obtained from axisymmetric, gasdynamic simulations of the leading working surface of an HH jet. We find that our analytic model reproduces the observed X-ray luminosities of HH~2 and HH~154, and we propose that HH~80/81 is a good candidate for future observations with Chandra. ",
        "introduction": "More than two decades ago, Ortolani \\& D'Odorico (1980) detected the UV emission of the Herbig-Haro object HH~1 with the International Ultraviolet Explorer (IUE). This observation opened up the new possibility of carrying out ultraviolet observations of HH objects, which resulted in a large number of papers describing results obtained with IUE (see, e.~g., Moro-Mart\\'\\i n et al. 1996), the Hopkins Ultraviolet Telescope (Raymond et al. 1997) and the Hubble Space Telescope (HST, Curiel et al. 1995).  Very recently, Pravdo et al. (2001) have reported Chandra observations of HH~2 and Favata et al. (2002) have reported XMM-Newton observations of HH~154 which are the first X-ray detections ever of HH objects. Even though HH~2 and HH~154 are detected only in a marginal way, these observations open up the new possibility of analyzing the X-ray properties of HH objects. This is an exciting development in observations of outflows from young stars, because it gives us the possibility of detecting fast, non-radiative shocks which could be associated with the outflows but have not been previously detected. The results of future X-ray observations of HH objects, however, will depend on whether or not other HH objects are bright in the 0.1 - 10 keV window observed by Chandra.  In the present paper, we derive a simple, analytic model which gives the X-ray luminosity of a bow shock as a function of the flow parameters (\\S 2). We then compare this analytic model with predictions obtained from axisymmetric simulations of the leading working surface of a jet (\\S 3) in order to evaluate its accuracy. Finally, in \\S 4 we compare the luminosity predicted from our model with the HH~2 observations of Pravdo et al. (2001), and suggest other objects which appear to be good candidates for future Chandra observations. ",
        "conclusions": "We have derived a simple, analytic model for predicting the X-ray luminosity of HH bow shocks. We have tested this model against HH jet numerical simulations, showing that it is applicable for bow shocks with shock velocities in the $120\\to 600$~km~s$^{-1}$ range.  We have then applied the analytic model to obtain predictions of the X-ray luminosities of HH~2H, HH~154D and HH~81, and obtain the following results~: \\begin{itemize} \\item the luminosity predicted for HH~2H is in good agreement with the Chandra observation of this object by Pravdo et al. (2001), \\item the luminosity predicted for HH~154D is in good agreement with the XMM-Newton observation of this object by Favata et al. (2002), \\item the X-ray luminosity predicted for HH~81 is a factor of $\\sim 1000$ larger than the one of HH~2. \\end{itemize} Because of the very large difference in X-ray luminosities predicted for HH~81 and for HH~2, even though HH~81 is a factor of $\\sim 3$ more distant and more highly extinguished than HH~2, we expect it to be brighter than HH~2 by a factor of 10 to 100."
    },
    "0208/astro-ph0208286_arXiv.txt": {
        "abstract": "\\vspace{0.2cm} Viscous redistribution of mass in Milky Way-type galactic disks is an appealing way of generating an exponential stellar profile over many scale lengths, almost independent of initial conditions, requiring only that the viscous timescale and star formation timescale are approximately equal.  However, galaxies with solid-body rotation curves cannot undergo viscous evolution.  Low surface brightness (LSB) galaxies have exponential surface brightness profiles, yet have slowly rising, nearly solid-body, rotation curves.  Because of this, viscous evolution may be inefficient in LSB galaxies: the exponential profiles, instead would give important insight into initial conditions for galaxy disk formation.  Using star formation laws from the literature, and tuning the efficiency of viscous processes to reproduce an exponential stellar profile in Milky Way-type galaxies, I test the role of viscous evolution in LSB galaxies.  Under the conservative and not unreasonable condition that LSB galaxies are gravitationally unstable for at least a part of their lives, I find that it is impossible to rule out a significant r\\^ole for viscous evolution.  This type of model still offers an attractive way of producing exponential disks, even in LSB galaxies with slowly-rising rotation curves.  \\vspace{0.2cm} ",
        "introduction": "\\label{sec:intro}  An exponential stellar light profile over 4-6 disk scale lengths is an almost universal observational feature of disk galaxies \\citep[e.g.,][]{freeman70,dejong96}. Yet, the production of a stellar exponential disk over more than 3 scale-lengths during the early stages of galaxy formation can prove highly challenging \\citep[e.g.,][although see also Contardo, Steinmetz \\& Fritze-von Alvensleben 1998]{dalc97,vdb01,fer01}. Ever since their inception \\citep{lin87}, viscous evolution models have endured as an attractive way of producing exponential stellar disks over $\\ga$4 disk scale lengths, {\\it almost independent of the initial density distribution}.  In a differentially rotating gas disk, viscosity caused by non-circular gas motions and turbulence transports angular momentum outwards as mass flows inwards \\citep{pringle81,lin87}. To prevent viscous evolution from reaching its logical endpoint (all the mass at the origin, all the angular momentum at infinity), the star formation (SF) timescale $t_*$ should be within half an order of magnitude of the viscous timescale\\footnote{This expression for $t_{\\nu}$ only applies in the case of a rotation curve which is {\\it not} close to solid body.} $t_{\\nu} \\sim r^2/\\nu$ to `freeze in' a nearly exponential stellar surface brightness profile over many disk scale-lengths \\citep[e.g.,][]{lin87,clarke89,yoshii89,olivier91,hellsten92,firmani96,fer01,slyz01}.  In 1D, the evolution of the gas density $\\Sigma_g$ at a given radius $r$ is given by: \\begin{equation} \\frac{\\partial\\Sigma_g}{\\partial t} = - \\frac{1}{r} \\frac{\\partial}{\\partial r} \\left\\{  \\frac{(\\partial/\\partial r)[\\nu \\Sigma_g r^3 d\\Omega/dr]}{d/dr (r^2 \\Omega)} \\right\\} - \\psi_*,  \\label{eqn:visc} \\end{equation} where $t$ is the time, $\\nu$ is the viscosity, $\\Omega$ is the angular velocity, and $\\psi_*$ is the SF rate \\citep{lin87}.  A brief inspection of Eqn.\\ \\ref{eqn:visc} shows a well-known result: in a galaxy without shear (a solid-body rotation curve with $d\\Omega/dr = 0$) there can be {\\it no viscous evolution}.  Bearing this in mind, it is interesting to note that recently published high-resolution ($\\la$2\\arcsec) \\ha rotation curves for low surface brightness (LSB) galaxies are slowly rising, and in a few cases are nearly solid-body, within the optical extent of the galaxy \\citep[e.g.,][]{deblok01,deblok02,matthews02}. Solid-body rotation curves have no shear, meaning that viscous evolution would not occur.  Yet, many LSB galaxies have exponential surface brightness profiles over 4 or more disk scale lengths \\citep[e.g.,][]{mcgaugh94,deblok95,bell00}. This could argue against a significant role for viscous evolution in generating the exponential stellar disks of LSB galaxies, implying that the structure of (at least) LSB disks offers important insight into the initial conditions of galactic disk formation.  To understand if constraints on galaxy formation models can be gleaned from the structure of LSB galaxy disks, it is important to quantitatively explore this issue.  The key questions are: {\\it i)} how close to solid-body are the rotation curves of LSB galaxies? and {\\it ii)} will the same kind of viscous evolution processes postulated to affect Milky Way-type galaxies be able to capitalize on any small departures from solid-body behavior, and be able to generate exponential stellar disks in LSB galaxies?  To address this problem, I set out in this paper to estimate the possible effect of viscous evolution in LSB galaxies with known, well-constrained surface brightness profiles and rotation curves. I make the central assumption that viscous processes in LSB galaxies and normal Milky Way-like galaxies have the same efficiency, and that any differences in viscous timescales are the {\\it result of different rotation curves and gas densities only}. Comparing these viscous timescales for LSB galaxies with realistic SF timescales \\citep[e.g.,][]{bell00}, one can say whether the same sources of viscosity that are applicable in normal galaxies could affect the evolution of LSB galaxies.  The plan of this paper is as follows. In \\S \\ref{sec:visc}, a brief description of the model is given. In \\S \\ref{sec:mw}, the model is calibrated to produce an exponential stellar light distribution for the Milky Way.  In \\S \\ref{sec:lsb}, the Milky Way-calibrated model is applied to five LSB galaxies.  I present my conclusions in \\S \\ref{sec:conc}.  As an aside, I discuss viscosity from gas cloud collisions in Appendix \\ref{app}. ",
        "conclusions": "\\label{sec:conc}  Models of galaxy evolution which include viscosity due to non-circular and turbulent gas motions have, for the last 15 years, provided a relatively appealing way of producing accurately exponential stellar disks in large spiral galaxies almost independent of initial conditions.  The primary assumption is that the viscous timescale and star formation timescale should be approximately equal. However, galaxies with solid-body rotation curves cannot undergo viscous evolution.  Low surface brightness (LSB) galaxies have exponential surface brightness profiles, yet have slowly rising, nearly solid-body, rotation curves.  Thus, it could be that LSB galaxies have had little redistribution of gas due to viscous evolution during their lifetimes.  Their surface brightness profiles would give unique insight into the initial conditions of galaxy formation.  In this paper, I have constrained the importance of viscous evolution during the life of LSB galaxies.  I have adopted star formation laws from the literature, and have tuned the efficiency of viscous processes to reproduce an exponential stellar profile in Milky Way-type galaxies.  Conservatively, I have assumed that the physics of turbulence are the same in LSB and Milky Way-type galaxies: any differences in the importance of viscous evolution in the two types of galaxy are the product of differences in rotation curve and gas density only.  I then apply this model to LSB galaxies with observed rotation curves and surface brightness profiles. Under the conservative and not unreasonable condition that LSB galaxies are gravitationally unstable for at least part of their lives, I find that viscous evolution from gravitational instabilities is quite effective. Viscous evolution models still offer an attractive way of producing an exponential disk over four or more disk scale lengths, almost independent of initial density distribution, even for LSB galaxies with slowly-rising rotation curves.  Does this have implications for the power of surface brightness profiles to constrain galaxy formation theory?  It is impossible at present to predict the degree of viscous evolution that {\\it will} occur, as the physical mechanisms driving viscous evolution are so poorly constrained.  However, this paper has shown that it {\\it can} plausibly occur in both low and high surface density disk galaxies, and \\citet{yoshii89} have shown that it would modify most intial density profiles into an exponential disk. This reduces substantially the ability of observations to rule out models of galaxy formation based on surface brightness profiles.  Indeed, the assembly of a disk with a reasonably centrally concentrated profile with roughly the right amount of angular momentum is all that is required of a galaxy formation model.  However, it is impossible to reduce the central matter density of a galaxy through viscous processes; therefore, models which overpredict the central mass density of galaxy disks cannot escape the need for revision or modification \\citep[see, e.g.,][]{vdb01}."
    },
    "0208/astro-ph0208529_arXiv.txt": {
        "abstract": "Stellar micro-variability poses a serious threat to the capacities of space-based planet-finding missions such as \\emph{Kepler} or \\emph{Eddington}. The solar lightcurves obtained by the VIRGO PMO6 and SPM instruments on board SoHO from 1996 to 2001 have been studied in order to follow variability changes through the activity cycle. In all datasets, active regions-induced variability, below $2~\\mu\\mathrm{Hz}$, is closely correlated to the BBSO Ca~{\\sc II} K-line index. The PMO6 (total irradiance) data shows evidence for a meso-granulation component around $\\tau \\simeq 8 \\times 10^3~\\mathrm{s}$, while all narrow-band SPM datasets (red, green and blue) show super-granulation ($\\tau \\simeq 5 \\times 10^4 ~\\mathrm{s}$) but no meso-granulation. Both actvity and granulation related components have significantly smaller amplitudes in the red than in the blue channel. These results, coupled with available stellar data, allow us to generate simulated lightcurves with enhanced variability as a testbed for pre-processing and detection methods, and influence the case for using colour information in this kind of mission. ",
        "introduction": "Transit detection algorithms in photometric time series, developed by the COROT (Defa{\\\" y}, Deleuil, \\& Barge 2001), \\emph{Eddington} (Aigrain \\& Favata 2002) and \\emph{Kepler} (Jenkins, Caldwell, \\& Borucki 2002) teams are very effective in the presence of white noise, and show that the latter two missions could detect habitable Earth-sized planets in data with white noise only. However, the non-Gaussian, ill-known stellar variability can have amplitudes over an order of magnitude larger than an Earth-analogue transit. Simulations using PMO6 data and the Bayesian algorithm developed for \\emph{Eddington} show that the minimum detectable planet size changes to $3~R_{\\oplus}$ when Sun-like variability is introduced without sophisitcated filtering techniques\\footnote{Simulations for a K5V star with $V=14.5$, 4 transits in the light curve} (Aigrain, Gilmore, \\& Favata, 2001). Variability will thus impact the choice of pre-processing techniques as well as target selection and colour information issues.  Pre-processing methods are already under investigation. After testing a highpass filter with some success, the \\emph{Eddington} team have developed a more optimized filter, based on a priori knowledge of the transit shape. Initial tests suggest it will be very successful, but realistic lightcurves with various levels of activity are needed for a quantitative assessment. The only star monitored with sufficient photometric precision and sampling is the Sun, which we use as a starting point. ",
        "conclusions": "We have shown that indicators of chromospheric activity such as the Ca~{\\sc II} K-line index provide a good proxy to the amplitude of the Sun's variability at low frequencies. As similar indicators are commonly measured in other stars, we may infer their expected variability level up to a few $\\mu\\mathrm{Hz}$. However, little theoretical or observational insight is yet available to relate granulation-type phenomena to stellar observables. Relating the observed components in the power spectrum to spectral and/or surface features may allow us to identify some new indicators of this type of variability.  In the mean time, a lightcurve `simulator' is being developed, using known scaling laws between chromospheric indicators, rotation period and colour, in which we can realistically scale up the active regions component. In the absence of more information a variety of scaling laws will be tried for the granulation component. This will provide a thorough test of pre-processing techniques under development.  In the longer term missions such as MOST, MONS and COROT will provide an ideal dataset to calibrate a general micro-variability model\\footnote{A proposal for the COROT Parallel Science Working Group, entitled \\emph{Hours-timescale stellar background variability and exo-planet transit detection prospects} (PI A. Collier-Cameron) has recently been accepted.}. Our understanding of the mechanisms behind low frequency, low amplitude stellar variability promises to improve vastly over the next few years.  Work is underway to quantify any increase in detection power gained from concentrating on the red part of the spectrum where we observe smaller variability, and from optimizing the filter bandwidth to exclude particularly variable spectral regions. These must be weighed against loss of photometric accuracy."
    },
    "0208/astro-ph0208003_arXiv.txt": {
        "abstract": "We have used STIS aboard HST to search for \\lya\\ absorption lines in the outer regions of eight nearby galaxies using background QSOs and AGN as probes. \\lya\\ lines are detected within a few hundred ~\\kms\\ of the systemic velocity of the galaxy in all cases. We conclude that a background line of sight which passes within $26-200$~\\h\\ of a foreground galaxy is likely to intercept low column density neutral hydrogen with $\\log N$(H~I)$\\apg\\: 13.0$. The ubiquity of detections implies a covering factor of $\\simeq 100$~\\% for low $N$(H~I) gas around galaxies within 200~\\h . We find, however, that the \\lya\\ lines are usually composed of individual components spread out in velocity over ranges of $300-900$~\\kms .  Two sightlines show components which are unusually broad for low-redshift \\lya\\ systems, with Doppler parameters $\\sim 150$~\\kms. These may arise in intragroup gas at temperatures of $1-2\\times10^{6}$~K.  We discuss the difficulty in trying to associate individual absorption components with the selected galaxies and their neighbors, but show that by degrading our STIS data to lower resolutions, we are able to reproduce the anti-correlation of \\lya\\ equivalent width and impact parameter found at higher redshift. The anti-correlation does not improve by correcting for the absolute magnitude of a galaxy in the same way as found at higher-$z$. We also show that the equivalent width and column density of \\lya\\ complexes (when individual components are summed over $\\sim 1000$~\\kms ) correlate well with a simple estimate of the volume density of galaxies brighter than $M_B = -17.5$ at the same redshift as a \\lya\\ complex. We do not reject the hypothesis that the selected galaxies are directly responsible for the observed \\lya\\ lines, but our analysis indicates that absorption by clumpy intragroup gas is an equally likely explanation. ",
        "introduction": "The initial detection of $z<< 1$ \\lya -forest absorption lines in HST QSO spectra \\citep{Bah91, morr91, KP1} not only demonstrated the existence and evolution of diffuse H~I clouds over a significant fraction of the age of the universe, but made it possible to search for the origin of the clouds themselves. Although \\lya\\ absorbers were generally accepted to be intergalactic clouds at high-redshift \\citep{sarg80}, the success at detecting galaxies associated with higher H~I column density Mg~II systems at $z < 1$ \\citep{jb1,crist87,BergBois,SteiMg2,stei95conf} fuelled a keen interest in investigating whether or not low column density \\lya\\ lines might also arise in the regions of individual galaxies.  The mapping of galaxies around the sightline towards 3C~273 \\citep{salz92, Morr93, salp95} found little evidence for a direct association between galaxy and \\lya -absorber. \\citet{Morr93} concluded that \\lya\\ clouds are not distributed at random with respect to galaxies, nor do they cluster as strongly as galaxies cluster with each other, and could only be associated with galaxies on scales of $\\sim 0.5-1\\: h^{-1}$~Mpc\\footnote{$h\\:=\\:H_0/100$, where $H_0$ is the Hubble constant, and $q_0\\:=\\:0$ is assumed throughout this paper}. However, in a study of six different fields, \\citet[][hereafter LBTW]{lbtw} found that {\\it i)} the majority of normal, luminous galaxies possess extended \\lya -absorbing halos or disks of radii $\\sim 160$~\\h , and {\\it ii)} between one and two thirds of all \\lya\\ lines arise in such galaxies. Combined with other similar studies \\citep{Stoc95,LeBr96} it appeared that one possible way to reconcile these disparate results was if strong \\lya\\ lines (those with equivalent widths $W\\apg 0.3$~\\AA ) were associated with galaxies, while weak ones ($W\\apl 0.3$~\\AA ) arose in the intergalactic medium. Such a claim was quickly rejected by other authors \\citep[e.g.][]{Trip98}.  These initial results were soon re-evaluated in light of the rapid development in hydrodynamical and semi-analytic simulations of hierarchical cold dark matter structure formation.  Simulations showed  gas following the same density fluctuations that are gravitationally induced by dark matter distributions, resulting in a `web' of intersecting filaments and sheets of gas. Analyses of artificial spectra, generated by shooting random sightlines through the simulations, were extremely successful in reproducing the observed properties of the high-redshift \\lya -forest \\citep{cen94,zhang95,hern96,mira96,bryan99}. In particular, they showed that low column density lines can be produced predominantly in the sheets of gas, while higher column density lines can arise from denser gas in a virialized halo, i.e., the same regions in which a high galaxy density might be expected.  More recently, \\citet[hereafter CLWB]{chen98} and \\citet{chen01} have extended the original observations of LBTW, and continue to find that the strength of the absorption depends on QSO-galaxy separation (`impact parameter') as well as galaxy luminosity, suggesting a strong link between galaxy and absorber.  Meanwhile, the simulators have advanced their models to $z\\sim 0$, and are again able to reproduce many features of the observed \\lya -forest \\citep{theuns98,dave+tripp}. \\citet{dave99} have identified clumps of gas and stars in their simulations which are likely to correspond to galaxies, and impressively, have been able to reproduce the anti-correlation of line strength and impact parameter.    In a previous paper \\citep{BBP96} we used Archival HST Faint Object Spectrograph (FOS) data to search for \\lya\\ lines from {\\it present-day} galaxies.  Studying the absorption properties of galaxies in the local universe is useful because a galaxy's properties are more easily determined and because its environment can be examined in detail, without the difficulties which naturally arise from observing higher redshift galaxies.  In the previous paper, we found that nearby galaxies do not possess \\lya -absorbing halos beyond 300~\\h\\ in radius, and --- assuming that an individual galaxy was indeed responsible for the absorption --- that the covering factor of galaxies between 50 and 300~\\h\\ was $\\sim 40$~\\% at an equivalent width limit $\\geq 0.3$~\\AA . However, we found no anti-correlation of \\lya\\ equivalent width with impact parameter or with galaxy luminosity, and questioned whether the galaxies were indeed responsible for the absorption lines. We instead concluded that our results supported the emerging picture that \\lya\\ lines arise in the gaseous sheets and filaments discussed above.  Unfortunately, our analysis suffered from two deficiencies: we probed few galaxies within the canonical 160~\\h\\ found originally by LBTW and confirmed by CLWB; and we were restricted to looking only for strong lines in the low resolution FOS data. In this paper we seek to remedy these inadequacies by describing the results of a program designed specifically to search for {\\it weak} \\lya\\ lines {\\it within} 160~\\h\\ of a nearby galaxy.  The experiment was not originally designed to address the origin of {\\it all} \\lya\\ absorbers, since we start by identifying a suitable galaxy and then search for \\lya\\ absorption at the galaxy's systemic redshift.  As we shall see, however, this simple attempt to define galaxy-absorber associations is not so straightforward, since the high spectral resolution of our STIS observations not only enables us to identify lines weaker than those found in FOS data, but also to resolve multiple components within an absorption complex on a scale of $\\approx 15$~\\kms. As we will discuss, resolving individual components in \\lya\\ systems makes it surprisingly difficult to tie absorption lines to individual galaxies.  The paper is ordered as follows: We describe the selection of the probed galaxies in \\S\\ref{sect_select}, and the data reduction and analysis in \\S\\ref{sect_analysis}. \\S\\ref{sect_results} details the geography of individual galaxy fields and describes the \\lya\\ absorption lines detected in the STIS spectra. \\S\\ref{sect_corrs} attempts to interpret the data obtained as a whole, and \\S\\ref{sect_conclusions} summarizes our results and places them in context with the higher-redshift systems discussed above. ",
        "conclusions": "}  By selecting background QSOs whose sightlines pass close to low redshift galaxies in the local universe, we have detected \\lya\\ absorption lines within a few hundred ~\\kms\\ of the systemic velocity of a galaxy in all cases. We conclude that a line of sight which passes within an impact parameter of $26-200$~\\h\\ of a galaxy is likely to intercept low column density neutral hydrogen with $\\log N$(H~I)$\\:\\apg\\: 13$. The ubiquity of detections implies a covering factor of $\\simeq 100$~\\% around galaxies selected in this way. We find that \\lya\\ absorption lines are actually composed of individual components spread out in velocity over a range sometimes as large as 800$-$900~\\kms .  Along the sightlines towards two probes, the high resolution and high S/N of our STIS data reveal components which are unusually broad for low-redshift \\lya\\ lines, with Doppler parameters $\\sim 150$~\\kms. If these widths reflect the kinetic temperature of the absorbing gas, then the gas must be at temperatures of $1-2\\times10^{6}$~K.  In general, there are two possible explanations for the \\lya\\ absorption we see. The first is that absorbing gas may be directly associated with the galaxies we chose to probe. Deposition of gas far from stellar populations of a galaxy may be due to internal interstellar processes within galaxies (e.g. superwinds from episodes of intense star formation, or gas cycling between the disk and halo in a galactic fountain) or perhaps from more external, dynamical processes, such as galaxy-galaxy interactions, or galaxy-dwarf accretion. The second explanation is that some or all of the absorbing components are not physically located around the probed galaxy at all. Instead, the absorbing gas simply traces the same large-scale gravitational structures that the galaxies inhabit.  The amount of data presented in this paper is insufficient to distinguish categorically between these two scenarios. However, several of our results appear to indicate that the second of these interpretations is at least as plausible as the first.  Although it seems relatively straightforward to match an individual absorption line with a given galaxy, we have discussed in \\S\\ref{sect_corrs} how such a one-to-one assignment is made uncertain by the multi-component structure of many of the \\lya\\ lines detected. Furthermore, galaxies are often found in loose groups and clusters. Although we targeted galaxies which appeared isolated, we subsequently discovered that there is usually more than one galaxy within our imposed impact parameter of 200~\\h\\ from the QSO sightline. By degrading our STIS data to the resolution of the FOS, the multicomponent nature of our absorption systems is largely removed. By matching these degraded lines (effectively summing individual component equivalent widths and column densities over 1000~\\kms\\ to produce an equivalent width $W_T$ and column density $N_T$) to the galaxy closest to the sightline, and ignoring any companions, we are able to reproduce the anti-correlation of \\lya\\ equivalent width and impact parameter found at higher redshift. This suggests that the population of absorbing galaxies studied at high-$z$ is no different than the nearby galaxies we selected to probe.  We find that our anti-correlation is weakened by correcting for galaxy luminosity in the manner prescribed for the higher redshift systems, $W\\propto L^{0.4}$, yet appears to improve for $W\\propto L^{-0.3}$. Such an relationship is surprising. It is possible to envisage a scenario whereby brighter galaxies have used up all the available neutral hydrogen in their immediate vicinity, resulting in brighter galaxies being associated with weaker \\lya\\ lines at a particular radius. Nevertheless, such a behaviour is significantly at odds with the results at higher redshift, and would imply that our population of absorbing galaxies had little to do with those found at higher-$z$. It also seems largely incompatible with the result discussed below and \\S\\ref{sect_lss}, that $W_T$ correlates with the volume density of moderately bright galaxies.  We note that the improved correlation becomes significant largely due to the correction for a single galaxy, UGCA~226 in the field of ESO~438-G009 (compare the top and bottom panels of Figure~\\ref{fig_corrs1}). Also, the rank correlation tests suggest the correlation is only marginally significant, at somewhere between the 2.2$\\sigma$ (Spearman) and 2.8$\\sigma$ (Kendall) level. We suspect therefore that the improvement in the correlation after correcting for galaxy luminosity is coincidental, a result of small number statistics, but we cannot rule out the relationship is real. Obvioulsy, further data are required to explore this issue in more detail.   Collating all available galaxy redshifts along each line of sight has shown that many of our selected galaxies are actually members of moderately rich groups, and we could equally well conclude that low redshift \\lya\\ lines are associated with groups of galaxies. We have investigated this alternative interpretation quantitatively, by defining a simple measure of the volume density of galaxies, $n$, evaluated at the same distance as a given galaxy/\\lya-complex. Assuming that the RC3 provides a magnitude limited sample (to the depths determined in Table~\\ref{tab_full}),  a correlation appears to exist between $n$ and both the total \\lya\\ equivalent width and H~I column density of an absorption complex (Figures~\\ref{fig_lss} and \\ref{fig_lss2}). This is highly suggestive evidence for an association of the \\lya\\ absorption lines with larger-scale structures rather than with individual galaxies, as predicted by hydrodynamical simulations of the growth of structure in currently favoured cosmologies. The evidence is strengthened by the fact that the same correlation seems to apply irrespective of whether there is a bright galaxy within 200~\\h\\ of a sightline.  This second scenario has a parallel at high redshift, where \\citet{Adel02} also see a strong correlation between the optical depth in the \\lya\\ forest and the overdensity of galaxies.  From their observations, they find that the forest is actually cleared out in the immediate vicinity of galaxies (at impact parameters of 100--200~\\h ) which appears to be different from the galaxies studied in our sample (presumably from outflows generated by active star-forming galaxies at that epoch).  However, on the same scales that we have been considering, 1--2~\\mpc , there is an excess of \\lya\\ absorption closely related to the density of galaxies in the high redshift sample. The parallel is intriguing because the high-$z$ and low-$z$ \\lya\\ forest have been considered to be different populations, given their markedly different redshift evolutions \\citep[e.g.][]{KP14}. Possibly, the two populations are closely related, as has been suggested by results from the hydrodynamical simulations of \\citet{dave99}.  Can our conclusions be reconciled with other results from studies of $z<1$ \\lya\\ lines?  One corollary would be that the strongest lines seen in FOS spectra should be associated with the highest density regions of galaxies, in rich groups or clusters.  (Of course, at sufficiently small impact parameters we would expect the outer regions of individual galaxy disks to be intercepted directly.)  It should be possible to test this prediction by completing the redshift surveys which have been underway for several years, obtaining redshifts of {\\it all} galaxies within a reasonably large radius (e.g. 2~\\mpc\\ $\\equiv 11$ arcmin at $z=0.3$) down to a specified magnitude limit. Unfortunately, the amount of telescope time required to fully sample a field may be hard to come-by. A field would need to be sampled to a depth where the galaxy luminosity function was adequately sampled at the absorber redshift.  For example, if the field luminosity function used in \\S\\ref{sect_lss} is valid at higher redshift, there would be only half a dozen or so $M=-18$ galaxies in the cylindrical volume we used to measure galaxy volume density. Unfortunately, an $M=-18$ galaxy would have an observed magnitude of $m\\sim22$ at $z=0.3$, meaning that several hundred galaxies with magnitudes brighter than this limit (which would include foreground and background galaxies) would have to be surveyed to properly sample the field at the required redshift.  The anti-correlation of equivalent width and impact parameter found at higher redshift might also be understood in terms of galaxy density: stronger lines would arise in regions of higher galaxy density, where the probability of finding a galaxy close to the sightline would be high because of the increase in the number of galaxies per unit area. It is likely that this effect has already been seen in the simulations of, e.g., \\citet{dave99}.  If galaxy density principally governs \\lya\\ equivalent width, then why should \\lya\\ equivalent width correlate with the luminosity of the galaxy closest to the QSO line of sight \\citep[Figure 2 of][]{chen01}?  One possibility is that galaxies in high density regions (giving rise to strong \\lya\\ lines) may be brighter than those in low density regions (which produce only weak \\lya\\ lines). So QSO sightlines which intercept richer groups of galaxies are more likely to find brighter galaxies, which are themselves likely to be closer to a sightline due to the higher surface density of galaxies.  The exact shape of the luminosity function in groups and clusters, and its variation between environments of different densities, is still unclear, but there is evidence for an increase in the luminosity of galaxies in denser environments \\citep[e.g.][]{garilli99, yagi02, norberg02}.   Whether these effects are significant enough to recreate the correlations seem for the high redshift galaxy sample remains uncertain. Only more comprehensive redshift surveys around high-$z$ QSO fields, and more detailed studies of the absorbing population of galaxies in the nearby universe, might eventually help us to understand whether galaxy density and environment really are important factors in governing the existence and conditions of \\lya\\ clouds."
    },
    "0208/astro-ph0208235_arXiv.txt": {
        "abstract": "The identification and extraction of non-Gaussian signals is one of the main cosmological challenges facing future experimental measurements of the cosmic microwave background temperature pattern. We present a generalized statistical measure based on a novel technique representation of Fourier phases using the return map. We show that this method is both robust and powerful in comparison, for example, with morphological measures. ",
        "introduction": "One of the key challenges facing modern cosmology is the complete statistical characterization of the primordial density fluctuations believed to be the seeds of the large-scale structure of the Universe we see today. According to the prevailing orthodoxy, these initial perturbations were generated as quantum phenomena during an inflationary epoch \\cite{inflation}. If this is the case they should display Gaussian statistics. Among other potentially testable consequences of this is that the small angular variations of the cosmic microwave background (CMB) sky should have the same statistical form \\cite{bbks,be}. Testing the hypothesis of primordial Gaussianity using maps of the CMB thus provides an opportunity to make experimental tests of ideas about the physics of the early universe. It is also necessary to devise robust statistical descriptors  for the identification of foreground contaminations and systematic instrumental artifacts.  Searches for non-Gaussian (NG) signals within the COBE-DMR data have yielded positive detections that are at least partly explained by systematics \\cite{nongaussian}. The need to devise more powerful statistical probes has been recognized and acted upon, with key ideas focusing on the bispectrum \\cite{bispectrum} and higher-order polyspectra \\cite{trispectrum,second}. Some of these methods have been developed as far as practical implementation on real data sets \\cite{bisp_measured}. These techniques are sensitive only to particular types of NG signal \\cite{wc}, so in this paper we present a new method that is both extremely general and extremely robust. ",
        "conclusions": ""
    },
    "0208/astro-ph0208328_arXiv.txt": {
        "abstract": "General arguments for optimized coronagraphy in the search for planets are presented.  First, off-axis telescopes provide the best telescopic platforms for use with coronagraphy, and telescope fabrication technology now allows the fabrication of such telescopes with diameters of up to 6.5 m.  We show that in certain circumstances a smaller telescope with an off-axis primary has a signal-to-noise advantage compared with larger Cassegrain telescopes.  Second, to fully exploit the advantages of the coronagraph for suppressing stray light, it is necessary to use a high Strehl ratio adaptive optics system.  This can be best achieved initially with modest aperture telescopes of 3--4 m in diameter.  Third, application of simultaneous differential imaging and simultaneous polarimetric techniques are required to reach the photon-limit of coronagraphic imaging.  These three developments, if pursued together, will yield significant improvements in the search for planets. ",
        "introduction": "There has been much interest recently in the pursuit of planet detection using 8 and 10 m telescopes, primarily to take advantage of the higher angular resolution and greater light gathering power.  The primary problem to overcome is the large amount of scattered light from the star in a region where planets might be found (within 1--2\\arcsec\\ radius of the star).  Since the reflected light from a Jovian-size planet is about 10$^{-9}$ that of the primary star's intensity, it is necessary to employ techniques to reduce the scattered light arising from Earth's atmosphere, secondary mirror spiders, the hole in the primary mirror, and from optical surfaces such as the primary and secondary mirrors of the telescope.\\\\  In this paper we review the advantages of off-axis telescopes and the advantages of searching for planets with a high-order adaptive optics (AO) systems on 3--4 m class telescopes.  It is worth noting that some of the most significant astronomical discoveries in the past decade were made using modest aperture telescopes (about 2 m or smaller), including  \\begin{itemize} \\item The discovery of the first brown dwarfs (Nakajima et al. 1995; Rebolo et al. 1995) and the subsequent discovery of field brown dwarfs by the 2MASS and DENIS surveys. \\item The discovery of planetary mass objects (Mayor \\& Queloz 1995) \\item The discovery of Kuiper Belt objects (Jewitt \\& Luu 1993) \\end{itemize}  We suggest here an approach for the direct detection of planets by using modest aperture telescopes to develop techniques that can be scaled up to larger telescopes.  \\\\ ",
        "conclusions": "We have summarized ideas that could significantly advance the detection of planets around nearby stars.  Realization of these ideas would allow observations near bright stars that are limited by photon noise rather than scattered light and speckle noise.  There are two key points:  \\begin{enumerate} \\item Unobscured apertures (off-axis telescopes) provide higher throughput and a cleaner PSF.  \\item High-order AO systems greatly enhance the performance of coronagraphs. Implementation of high-order AO systems on modest aperture telescopes (3--4 m) will initially provide an observational niche for detecting planets.  Techniques to suppress speckle noise will be required to approach the photon noise limit. \\end{enumerate}"
    },
    "0208/astro-ph0208434_arXiv.txt": {
        "abstract": "We present the results from a sensitive multi-wavelength analysis of the properties of Extremely Red Objects (EROs).  Our analysis employs deep $RIzJHK$ photometry of a $8.5'\\times 8.5'$ region to select a sample of 68 EROs with $(R-K)\\geq 5.3$ and brighter than $K=20.5$ ($5\\sigma$).  We combine this photometric dataset with an extremely deep 1.4-GHz radio map of the field obtained from the VLA.  This map reaches a 1-$\\sigma$ limiting flux density of 3.5$\\mu$Jy making it the deepest 1.4-GHz map taken and is sensitive enough to detect an active galaxy with $L_{1.4}\\gs 10^{23}$\\,W\\,Hz$^{-1}$ at $z>1$. If powered by a starburst, this radio luminosity is equivalent to a star-formation rate of $\\gs 25$\\,M$_\\odot$\\,yr$^{-1}$ for stars more massive than 5\\,M$_\\odot$.  We identify radio counterparts to 21 of the EROs in this field with radio fluxes above 12.6$\\mu$Jy and resolve a third of these with our 1.6$''$ FWHM beam. The spectral energy distributions of the majority of these galaxies are consistent with those expected for dust-reddened starbursts at $z\\sim 1$.  At these redshifts the radio luminosities of these galaxies indicate a median far-infrared luminosity of this population of $L_{FIR}\\gs 10^{12}L_\\odot$, meaning half of the radio-detected sample are ultraluminous infrared galaxies (ULIRGs).  We conclude that $\\gs 16\\pm 5$\\% of the ERO population brighter than $K=20.5$ are luminous infrared galaxies (LIRGs) at $z\\sim 1$. We also use photometric classification of the colors of the EROs to investigate the mix of dusty active and evolved passive systems in the remaining ERO population which is undetected in our radio map. Based on this we suggest that at least 30\\% and possibly up to $\\sim 60$\\% of {\\it all} EROs with $(R-K)\\geq 5.3$ and $K\\leq 20.5$ are dusty, star-forming systems at $z\\gs 1$.  Our best estimate of the star formation density in this highly-obscured and optically faint ($R\\gs 26$) population is $\\stackrel{.}{\\rho_*}(0.1$--100\\,M$_\\odot)=0.11\\pm 0.03$\\,M$_\\odot$\\,yr$^{-1}$\\,Mpc$^{-3}$, comparable to estimates of that in H$\\alpha$ emitting galaxies at $z\\sim1$, and greater than the estimates from UV-selected samples at these epochs.  This lends support to the claims of a strong increase in the contribution from obscured systems to the star formation density at high redshifts.  Using the observed counts of the radio-detected ERO population we model the apparent break in the $K$-band number counts of the whole ERO population at $K\\sim 19$--20 and propose that the passive ERO class dominates the total population in a relatively narrow magnitude range around $K\\ls 20$, with dusty, active EROs making up the bulk of the population at fainter limits. ",
        "introduction": "The last five years have seen a growing appreciation of the diversity of galaxy properties at $z\\gs 1$--3.  In part this has arisen from an impressive improvement in the quality of the observations of galaxies at these redshifts -- driven by the availability of powerful new instruments on 8-m class telescopes (e.g.\\ NIRSPEC on Keck and ISAAC on the VLT).  However, an equal role has been played by the realization of the necessity of a multi-wavelength approach to studies of galaxy evolution at $z\\gs 1$.  Thus a range of new surveys spanning wavebands from the X-ray, near- and mid-infrared, submillimeter and out to the radio, have complimented the traditional view of distant galaxies based on UV/optical observations.  These new studies have all tended to stress the role of dust obscuration in censoring our view of the galaxy population at high redshifts, and especially in disguising the extent of activity in the most luminous systems, both AGN and star-forming. Indeed, our current, incomplete, knowledge of the evolution of dusty galaxies suggests that some of the most active galaxies in the distant Universe may be so obscured as to be undetectable in even the most sensitive UV/optical observations (e.g.\\ Haarsma et al.\\ 2000; Smail et al.\\ 2002).  One central theme to appear from this new multi-wavelength view of galaxy formation is the ubiquity of optically-faint, but bright near-infrared, counterparts to sources identified in many wavebands: in the X-ray (Alexander et al.\\ 2001, 2002; Cowie et al.\\ 2001; Page et al.\\ 2002; Stevens et al.\\ 2002; Mainieri et al.\\ 2002), the mid-infrared (Pierre et al.\\ 2001; Smith et al.\\ 2001; Franceschini et al.\\ 2002), submillimeter (Smail et al.\\ 1999a; Lutz et al.\\ 2001; Ivison et al.\\ 2002) and radio bands (Richards et al.\\ 1999; Chapman et al.\\ 2001).  This has resulted in a renaissance in interest in such Extremely Red Objects (EROs) -- a class of galaxies which had previously been viewed as a curiosity with little relevance to our understanding of galaxy formation and evolution, comprising as they do a mere $\\sim 5$--10\\% of the population at $K\\leq 20$.  The ERO class is photometrically-defined -- the most frequently used definition is now $(R-K)\\geq 5.3$.  This very red optical-near-infrared color is intended to isolate two broad populations of galaxies at $z\\gs 1$: those which are red by virtue of the presence of large amounts of dust (resulting from active star formation), as well as passive systems whose red colors arise from the dominance of old, evolved stars in their stellar populations.  The very different natures of these two sub-classes has prompted efforts to disentangle their relative contributions to the ERO population (Pozzetti \\& Mannucci 2000; Mannucci et al.\\ 2002) so that well-defined samples can be used to test galaxy formation models (Daddi et al.\\ 2000; Smith et al.\\ 2002a; Firth et al.\\ 2002).  Studies of the ERO population have turned up hints of the diverse nature of this population, including an apparent break in the number counts of EROs at $K\\sim 19$--20 (McCarthy et al.\\ 2001; Smith et al.\\ 2002a).  This feature is roughly 2\\,mags fainter than the well known break in the total $K$-band counts of the faint field population (Gardner et al.\\ 1993). The cumulative count slope of EROs ($\\log_{10}N(<K) \\propto \\alpha K$) drops from $\\alpha= 0.81\\pm 0.05$ at bright magnitudes to $\\alpha= 0.32\\pm 0.05$ for the fainter EROs (e.g.\\ Smith et al.\\ 2002a; Chen et al.\\ 2002; Fig.~2).  This break in the counts could either result from a relatively narrow redshift range probed by the selection criteria of EROs -- with the change in the integrated counts then being simply related to the form of the luminosity function of the ERO population, or it may reflect differing evolutionary behavior in sub-populations -- where one class of EROs declines rapidly beyond $K\\sim 19$.  To test this suggestion we need to track the relative mix of galaxy types within the ERO population to $K>19$--20.  Techniques to separate the passive, evolved and obscured, active populations include near-infrared photometric classifications (Pozzetti \\& Mannucci 2000; Smith et al.\\ 2002a; Mannucci et al.\\ 2002), submillimeter emission (Dey et al.\\ 1999; Mohan et al.\\ 2002), X-ray emission (Alexander et al.\\ 2001, 2002; Brusa et al.\\ 2002) and optical/near-infrared spectroscopy (Cimatti et al.\\ 2002; Smith et al.\\ 2001, 2002b).  Of these approaches, by far the largest and most influential project is the ``K20'' survey undertaken by Cimatti et al.\\ (2002) who obtained spectroscopic classifications for 29 EROs with $K<19.2$ and $(R-K)\\geq 5$ from a full sample of 45. The median redshift of their $K<19.2$ sample is $z=1.1\\pm 0.2$, with galaxies spread across $z=0.7$--1.4 (see Daddi et al.\\ 2001) and they find that their sample is split equally into passive, evolved and active, dusty EROs: 31--64\\% versus 67--33\\%, where the ranges reflect the uncertainty due to the incompleteness in their survey.  The K20 survey has significantly increased our knowledge of the ERO population, however, it is reliant on identifying emission lines in the restframe UV spectra of possibly dusty galaxies, which results in an incomplete and perhaps biased view of the mix of galaxies within the population. Moreover, the magnitude limit achieved for the high completeness sample, $K<19.2$, means that it can't be used to track the variation in the properties of the population across the break in the counts of EROs.  Here we apply a new technique to investigate the mix of the ERO population at faint magnitudes: using very deep radio observations at 1.4\\,GHz which are both sensitive enough to identify strongly star forming galaxies out to high redshifts, $z\\gs 1$--2, and yet insensitive to the effects of dust obscuration (Mohan et al.\\ 2002; Chapman et al.\\ 2002a).  Radio surveys to sub-mJy flux limits have shown the usefulness of this technique for identifying star-forming galaxies in the distant Universe -- morphological classifications from {\\it Hubble Space Telescope} ({\\it HST}) imaging show many are blue disk galaxies (Windhorst et al.\\ 1994; Richards et al.\\ 1998; Richards 2000), with optical spectroscopy confirming that these are apparently normal star forming galaxies at intermediate redshifts (Benn et al.\\ 1993; Mobasher et al.\\ 1999; Roche, Lowenthal \\& Koo 2002), with a small proportion of AGN (20\\% based on their radio morphologies in very high-resolution maps from combined VLA and Merlin observations, Muxlow et al.\\ 1999, with most of these having blue colors).  However, the spectral properties of the galaxies in these surveys also show signatures of the presence of dust, which may be cloaking the strength of the activity in these systems and making them appear more mundane than they should (Hammer et al.\\ 1995; Poggianti \\& Wu 2000; Smail et al.\\ 1999b).  More recently, samples of faint radio sources have begun to be used to attempt to identify obscured star-forming galaxies at even higher redshifts. By focusing on those radio sources with faint or undetectable counterparts in the optical these surveys attempt to isolate the most distant star forming galaxies in the radio samples (Richards et al.\\ 1999; Barger et al.\\ 2000; Chapman et al.\\ 2001, 2002b).  Sensitive submillimeter observations of these optically-faint radio sources confirm that a large fraction of them are luminous, dusty systems at redshifts of $z\\sim 1$--3 -- although, as with the whole submillimeter galaxy population, there is still considerable uncertainty over their exact redshifts (Chapman et al.\\ 2002b).  The relationship between these highly obscured, but very luminous galaxies and the perhaps less-obscured systems selected through Lyman-$\\alpha$ emission and the Lyman-break technique is a crucial issue for a complete understanding of the formation and evolution of galaxies (Smail et al.\\ 2002, 2003).  The project described here seeks to trace the population of the luminous, dusty galaxies detected in the submillimeter at $z\\gs 2$ to lower redshifts and lower luminosities. These samples should be more amenable to detailed study and so provide a clearer connection between the evolution of the obscured and unobscured galaxies over the lifetime of the Universe (Chapman et al.\\ 2002b).  In this paper we use an extremely deep VLA radio map to investigate the radio emission from a sample of ERO galaxies at a sensitivity limit sufficient to detect strongly star forming galaxies at $z\\gs 1$--2.  We also present extensive optical/near-infrared imaging of this ERO sample.  We exploit this deep $RIzJHK$ photometry to analyse the spectral energy distributions (SEDs) of the EROs and investigate their properties in detail.  We present our observations, their reduction and cataloging in the next section, describe our analysis and results in \\S3, and discuss these in \\S4 before giving our conclusions in \\S5.  Throughout we assume a cosmological model with $q_o=0.5$ and $H_o=50$\\,km\\,s$^{-1}$\\,Mpc$^{-1}$.  If we instead adopted the currently fashionable $H_o=70$\\,km\\,s$^{-1}$\\,Mpc$^{-1}$, $\\Omega_\\Lambda=0.7$, $\\Omega_m=0.3$ cosmology, then for a source at $z=1$ all the linear sizes we derive would be 6\\% smaller, the luminosities would be 12\\% fainter and the volume densities would be 12\\% lower. ",
        "conclusions": "The main conclusions of this work are:  \\noindent{1)} We catalog 68 EROs within a 72.8 arcmin$^2$ field which has the deepest 1.4-GHz observations of any region on the sky.  We find that 31\\% of these EROs are detected in the radio map down to a flux limit of 12.6\\,$\\mu$Jy, of these radio-detected EROs, 33\\% are resolved in our VLA map with a typical FWHM of $1.4\\pm 0.4''$ -- equivalent to $12\\pm 3$\\,kpc at $z\\sim 1$.  Moreover, the radio emission in these EROs mirrors their $R$-band morphologies, suggesting that star formation is responsible for the bulk of the radio flux.  \\noindent{2)} Using our deep $RIzJHK$ imaging we photometrically classify the radio-detected and undetected EROs into dusty or evolved systems based on simple SED models.  We first analyse the colors of those radio-detected with high quality photometry and conclude that the majority of these are adequately described by a highly dust-reddened starburst spectrum.  This provides support for applying this analysis to the radio-undetected EROs to search for similar obscured, active galaxies.  \\noindent{3)} Using our simple photometric models we estimate that more than 32\\% of the radio-undetected EROs with high quality photometry, or a minimum of 20\\% of all radio-undetected EROs, have the colors of dusty starbursts, similar to the radio-detected EROs.  We estimate the rate of mis-classification of EROs in our analysis assuming that most of the radio-detected EROs are actually dusty starbursts, and based on this propose that $\\sim 45$\\% of the EROs with radio fluxes $<12.6\\mu$Jy have dusty SEDs.  ~From the continuity of the number counts of radio-detected EROs we would expect that most of these sources would have 1.4-GHz fluxes around 5\\,$\\mu$Jy.  We also find that one of the 68 EROs has an SED consistent with galactic star, while another apparent ERO appears to be a transient source.  \\noindent{4)} Combining the samples of radio-detected and undetected EROs which are photometrically classified as dusty we place a firm lower limit of $\\geq 30$\\% on the proportion of dusty starbursts in the whole ERO population.  To obtain a more realistic estimate we adopt a correction for mis-classification and assume the resulting proportions of dusty and evolved EROs are representative of the full sample (not just those with the best photometry) we estimate that at least 45--50\\%, and possibly up to $\\sim 60\\pm15$\\%, of the {\\it whole} ERO population with $(R-K)\\geq 5.3$ and $K\\leq 20.5$ are probably dusty, active galaxies.  Our fitting procedure also suggests a median redshift of $z\\sim 1.0\\pm 0.3$ with a quartile range of $z=0.8$--1.5.  This is in reasonable agreement with the spectroscopic survey of somewhat brighter and bluer EROs by Cimatti et al.\\ (2002).  Assuming that AGN do not dominate the radio emission from these galaxies, we estimate a volume density of EROs with far-infrared luminosities of $\\gs 10^{12} L_\\odot$ (i.e.\\ ULIRGs) of $0.5\\times 10^{-4}$\\,Mpc$^{-3}$, this is over an order of magnitude higher than for equivalent luminosity events at $z=0$ and confirms the strong evolution of the most dusty, active galaxies out to $z\\gs 1$ (Blain et al.\\ 1999a).  \\noindent{5)} We use the local far-infrared--radio correlation to estimate star formation rates from the radio fluxes of the dusty EROs, assuming that their radio emission is purely powered by massive star formation.  Combining these estimates with the probable volume spanned by our dusty ERO subsample we can calculate the star formation density in this obscured population.  Adopting conservative assumptions we find that the star formation density in star forming EROs at $z\\sim 1$ probably exceeds $\\stackrel{.}{\\rho_*}($0.1--100\\,M$_\\odot)>0.073\\pm 0.018$\\,M$_\\odot$\\,yr$^{-1}$\\,Mpc$^{-3}$ and maybe as high as $\\stackrel{.}{\\rho_*}=0.11\\pm 0.03$\\,M$_\\odot$\\,yr$^{-1}$\\,Mpc$^{-3}$. This measurement is comparable to estimates of the star formation density in H$\\alpha$-selected galaxies at these epochs, and greater than that seen in optically--selected galaxies, showing that the activity in obscured systems may make a significant contribution to the total star formation at $z\\sim 1$.  We also estimate that the dusty ERO population may produce up to 5--10\\% of the far-infrared background at its peak at $\\sim 100\\mu$m, underlining the cosmological significance of these events for understanding the history of star formation in the Universe.  \\noindent{6)} We propose that the apparent break in the integrated counts of the EROs at $K\\sim 19$--20 reflects the peak of the contribution from passive, evolved galaxies to this population.  At magnitudes fainter than $K\\sim 20$ we predict that the population will be increasing dominated by dusty, active galaxies.  The continuity in the properties of these faint dusty EROs connects them to samples of submm-selected obscured starbursts at higher redshifts and fainter $K$-band magnitudes.  This proposal can be tested by studies of the multiwavelength properties of extremely faint samples of EROs, $K\\sim 23$, and their relationship to the blank-field populations of radio sources.  \\noindent{7)} By identifying the likely contribution from dusty EROs to the total ERO population we estimate that the volume density of passive, evolved galaxies brighter than $K\\leq 20.5$ is $\\gs 4\\times 10^{-4}$\\,Mpc$^{-3}$. Recent results point to a morphological diversity in the passive ERO population which would suggest that these galaxies should not be simply viewed as passively evolving ellipticals (Smith et al.\\ 2002b).  We conclude that previously well-fitting pure luminosity evolution models, which described the number counts of the whole ERO sample as due to a passively evolving elliptical population formed at high redshifts, will probably over predict the number of truly passive EROs and hence there is scope for the recent (trans)formation of some local elliptical galaxies.  Nevertheless, the presence of luminous, passive galaxies at $z\\sim 1$ does point to a phase of massive spheroid formation at higher redshifts, consistent with the interpretation of the SCUBA population as the formation phase of massive, proto-ellipticals (Lilly et al.\\ 1999; Smail et al.\\ 2002, 2003).  The typical, reddening-corrected luminosities of the radio-detected ERO population at $z\\sim 1$ suggests that these galaxies will evolve into sub-$L^\\ast$ systems by the present-day unless their star formation activity continues at a high rate for $\\gg 10^8$\\,yrs.  The area analysed in this work covers a small fraction of the sensitive field of our VLA map -- we expect to be able to quadruple the ERO sample available for analysis in the near future.  In addition we are undertaking a full photometric analysis of the $\\sim 1500$ optically-bright $\\mu$Jy radio sources in this field using our panoramic $U\\!BV\\!RIzJHK$ imaging dataset (Owen et al.\\ 2002), constrained by deep spectroscopy from GMOS on Gemini."
    },
    "0208/astro-ph0208058_arXiv.txt": {
        "abstract": "Recent advances in the modelling of stellar winds driven by radiation pressure make it possible to fit many wind-sensitive features in the UV spectra of hot stars, opening the way for a hydrodynamically consistent determination of stellar radii, masses, and luminosities from the UV spectrum alone. It is thus no longer necessary to assume a theoretical mass--luminosity relation. As the method has been shown to work for massive O~stars, we are now able to test predictions from the post-AGB evolutionary calculations quantitatively for the first time. Here we present the first rather surprising consequences of using the new generation of model atmospheres for the analysis of a sample of central stars of planetary nebulae. ",
        "introduction": "A lot of work on model atmospheres of PN central stars (CSPNs in what follows) has been motivated by the desire to obtain information about the basic properties of CSPNs (surface temperature, mass, luminosity, abundances), so as to be able to test predictions from post-AGB evolutionary calculations. The earlier efforts, based on plane-parallel non-LTE models, could not achieve a completely independent test, in the following sense: since the plane-parallel model fits to H and He photospheric absorption lines can only produce information about surface temperature, He abundance and $\\log g$, we cannot derive stellar masses or luminosities, but only $L/M$ ratios. This is exactly the same problem we face when dealing with low-gravity early-type ``supergiant'' stars at high Galactic latitudes: are they luminous and massive, or are they evolving away from the AGB? We need some independent evidence to settle the issue -- for example, the distance to the star. Unfortunately, we lack reliable distances to most CSPNs.  So what could be done was to plot the positions of CSPNs in the $\\log g$--$\\log \\Teff$ diagram, and compare them with plots of post-AGB tracks, translated from the $\\log L$--$\\log \\Teff$ diagram. After doing this translation it is possible to read the stellar mass in the $\\log g$--$\\log \\Teff$ diagram. From this, we can derive $L$ and, if we know the visual dereddened apparent magnitude, a so-called ``spectroscopic distance''. All this work, however, is based on {\\em assuming\\/} that the evolutionary models give us the correct relation between stellar mass and luminosity. It is not a real test of the evolutionary models, but only a consistency check.  In the last 10 years there has been a lot of progress in the modelling of stellar winds driven by radiation pressure. Many CSPNs show spectroscopic evidence of winds, in the form of emission lines in the visible spectrum and especially P-Cygni-type profiles of resonance lines in the ultraviolet between 1000 and 2000\\,\\AA.  The existence of these wind features provides both a challenge and an opportunity. The challenge is to model them. The opportunity is to use the information about the geometrical extension of the atmosphere and the forces of gravity and radiative pressure, implicit in the wind profiles (from which the terminal velocity and mass loss rate can be derived), to obtain the physical size of the star, which is the key to derive the stellar luminosity and mass. (The idea is described together with a first application by Pauldrach et al.~1988.) Thus the successful modelling of the wind features opens the way for a real test of the mass--luminosity relation of CSPNs.  In this review we would like to present the current situation of the project and the rather surprising results obtained up to now. Section 2 describes a model analysis of a massive Population I star, using state-of-the-art hydrodynamically consistent, spherically symmetric model atmospheres to demonstrate the power of the technique and to show how successfully we can reproduce the ionizing fluxes and observed spectra of such stars. In Section 3 we introduce the wind-momentum--luminosity relation and describe a previous attempt to determine whether or not CSPNs follow this relation. In Section 4 we show two examples of the application of our hydrodynamically consistent wind models to CSPNs (fits to IUE and HST spectra and derived parameters). In Section 5 we add the results from 6 other CSPNs similarly analyzed and discuss the consequences. ",
        "conclusions": "\\begin{table}[t] \\caption{Parameters of eight CSPNs derived by our analysis of the UV spectra using our model atmospheres, compared to the values found by Kudritzki et al.~1997. \\label{tbl:params}} \\begin{center} \\input{table.tex} \\end{center} \\end{table}  Table~\\ref{tbl:params} shows the result of applying the method and UV analysis described in the previous section to eight CSPNs, including the two exemplary objects above. The problem which Table~\\ref{tbl:params} points to is obvious: according to the post-AGB evolutionary timescales we should not find so many extremely luminous CSPNs, because according to this theory they are expected to fade very quickly. We shall note, however, that the sample of objects chosen here is most likely not a representative one.  \\begin{figure}[t] \\centerline{\\includegraphics[height=4.1in,angle=-90]{ml.ps}} \\caption{ Luminosity vs.\\ mass for the evolutionary tracks (open squares) compared to the observed quantities determined with our method (filled squares). Although the luminosities deduced from the UV spectra lie in the expected range, a much larger spread in the masses (from $0.4$ to $1.4\\,\\Msun$) is obtained. No well-defined relation between CSPN mass and luminosity can be made out. \\label{fig:ml}} \\end{figure}  Figure~\\ref{fig:ml} shows the relation between stellar mass and luminosity obtained from our model atmosphere analyses, in comparison with the mass--luminosity relation of the evolutionary tracks, represented by the values from Kudritzki et al.~1997. This plot shows that the problem already indicated is indeed very disturbing: the derived masses and luminosities do not agree with the classical post-AGB mass--luminosity relation. {\\em There is a very large spread in masses, between 0.4 and 1.4\\,$\\Msun$, and there is no well-defined relation between CSPN mass and luminosity.}  In Figure~\\ref{fig:wml2} we now again show the wind-momentum--luminosity relation for both massive hot stars and CSPNs, but this time based on the parameters derived in our analysis. Our models give wind momenta of the right order of magnitude and within the expected luminosity range (there may be too many CSPNs at $\\log L/\\Lsun > 4$, but not so many as in Kudritzki et al.~1997). The CSPNs are found along the extrapolation of the wind-momentum--luminosity relation defined by the massive hot stars, and the CSPNs show a smaller dispersion, i.\\,e., a tighter correlation of wind-momentum with luminosity, than was the case in Kudritzki et al.~(1997). And, most important, this was achieved by fitting the multitude of diagnostic features in the CSPN UV spectra by means of up-to-date hydrodynamically consistent models.  How then does our wind-momentum--luminosity relation compare to that found by Kudritzki et~al.? The answer is, quite favorably. If we drop the assumption made by Kudritzki et~al.\\ that the stars obey the theoretical post-AGB mass--luminosity relation, and instead scale their mass loss rates to our radii\\footnote{ Additionally allowing for their different effective temperatures by requiring that the observed visual flux ($\\sim R^2 \\Teff$) stay constant.} -- keeping $Q$, the real observational quantity, fixed -- then their wind momenta match ours to within about a factor of two. Furthermore, their sample with the radii thus scaled now also shows a much tighter correlation of the wind momentum to luminosity than before (see Figure~\\ref{fig:wml2}).  \\begin{figure}[t] \\includegraphics[height=3.0in,angle=-90]{wml2.ps} \\hfill\\parbox[t]{2.5in}{ \\caption{ The wind-mo\\-mentum--luminosity relation for CSPNs (lower left) based on our values determined from the UV spectra (filled squares). The open squares are the values from Kudritzki et~al.\\ with our radii applied (see text). Compared to Figure~\\ref{fig:wml1} the result is striking. \\label{fig:wml2}}} \\end{figure}  All this is strong evidence that all these winds are radiatively driven. It would be extremely difficult to explain the wind-momentum--luminosity relation if there was another mechanism driving the winds.  What makes the {\\em CSPN mass discrepancy problem\\/} found most intriguing is the fact that the same model atmospheres, with exactly the same physics, obviously work also perfectly well for massive O stars, as we have shown. We are reluctant to conclude that there is something basic we do not understand about either the winds and photospheres of O-type stars, or how to produce CSPNs and what their internal structure is. Both alternatives are difficult to believe. But we need to explain why we obtain such a spread in the masses.  We cannot offer a fair solution to this paradox; all we can do right now is to present our method, the results obtained from our analysis, and the corresponding problem in the clearest possible way, which is usually the first step along the road that leads to the solution."
    },
    "0208/astro-ph0208091_arXiv.txt": {
        "abstract": "{\\small We apply a model of magnetically dominated coronae above standard accretion discs to the low/hard state of galactic black holes. When the disc-corona coupling is accounted for self-consistently assuming that magneto-rotational instability is at work in the disc, and that the corona is generated by buoyant escape of disc magnetic structures, then the model predicts powerful, X-ray emitting coronae at low accretion rates. A main consequence is discussed: the possibility that the corona itself is the launching site of powerful, MHD driven jets/outflows. This depends crucially of the coronal scaleheight. Finally, we present the first radial profiles of a corona a different accretion rates, and discuss their implications for high frequency variability. } ",
        "introduction": "In \\cite{mf02} we have shown how it is possible to build a physically self-consistent model for an accretion disc corona: assuming that the turbulent magnetic stresses generated by MRI are responsible for angular momentum transport in the disc, and that the field saturates mainly due to buoyancy (as should be the case if strong coronae are to be generated), then the fraction of power released into the corona as a function of radius, $f(R)$, can be uniquely determined by solving the algebraic equations for accretion disc structure. For {\\it uniform} discs (which may not be the case in the radiation pressure dominated parts, see \\cite{tss02}), we have \\begin{equation} \\label{eq_fr} f(R) \\simeq \\sqrt{\\alpha} \\left(1+\\frac{P_{\\rm rad}(R)}{P_{\\rm gas}(R)}\\right)^{-1/4}, \\end{equation} where $\\alpha$ is a constant of the order of unity and $P_{\\rm rad}$ and $P_{\\rm gas}$ represent radiation and gas pressure in the disc, respectively.  A major consequence of such a disc-corona coupling, which is of interest in the context of microquasars, is that coronae are more powerful at low accretion rates, in particular below the critical rate at which the radiation pressure dominated part of the disc disappears altogether. Indeed, if the $\\alpha$ viscosity parameter is high enough, $f$ can approach unity, and we are therefore left with a flow in which most of the energy goes into the magnetically dominated corona. The cold, geometrically thin disc, although very dim, may manifest itself as a reprocessor of X-ray coronal radiation. This makes our solution different from a simple magnetically dominated non radiative accretion flow (NRAF): the two may be distinguished, for example, by the presence of relativistically smeared reflection features. ",
        "conclusions": "If a standard, geometrically thin and optically thick accretion disc is coupled to a corona through buoyancy of magnetic flux generated by MRI in the disc itself, then a powerful corona is generated whenever gas pressure dominates over radiation pressure, i.e. at low accretion rates. This implies that microquasar in the low/hard state may be characterized by strong coronae on top of very dim, cold discs, that may be seen only as reprocessors. Depending on the detailed coronal geometry, and in particular on its vertical extent, low accretion rate systems may produce strong MHD driven jets/outflows whose total kinetic energy flux may exceed the radiated one. Beside their radio emission, such outflows should manifest themselves through their influence on the variability properties (noise and QPOs), for example reducing the intensity and narrowness of such timing features. In order to understand the complex interplay between coronal physics and dynamics and the variability properties of microquasars, full time- and radial-dependent models are needed. As a first illustrative step in this direction, we have shown here a stationary model for the radial profiles of the different spectral components emerging from a disc-corona system at various accretion rates."
    },
    "0208/astro-ph0208214_arXiv.txt": {
        "abstract": "We use ultraviolet spectra of Capella from the {\\em Hubble Space Telescope} (HST) and {\\em Far Ultraviolet Spectroscopic Explorer} (FUSE) satellites to study interstellar absorption lines from the Local Interstellar Cloud (LIC).  Measurements of these lines are used to empirically determine the ionization states of carbon, nitrogen, and silicon in the LIC, for comparison with the predictions of theoretical photoionization models.  We find that the observed ionization states are consistent with previously published photoionization predictions.  Total abundances are determined for the elements mentioned above, and others, for comparison with solar abundances. Magnesium, aluminum, silicon, and iron are all depleted by at least a factor of 10 toward Capella.  The abundances of carbon, nitrogen, and oxygen are essentially solar, although the error bars are large enough to also allow depletions of about a factor of 2 for these elements. ",
        "introduction": "The Sun is located inside a warm, partially ionized interstellar cloud called the Local Interstellar Cloud (LIC).  Ultraviolet spectra containing LIC absorption lines have been used to study the properties of the LIC.  Its average temperature appears to be about $T=8000\\pm 1000$~K, although there is some evidence for spatial variation within the LIC \\citep{jll95,ard97,np97,bew98,bew02a}. The average H~I density is $n({\\rm H~I})\\approx 0.1$ cm$^{-3}$ \\citep{jll00}. Hydrogen must be roughly half-ionized since the average electron density is $n_{e}=0.11^{+0.12}_{-0.06}$ cm$^{-3}$ \\citep{bew97,jbh99}. Estimates of $n({\\rm H~I})$ from observations of interstellar atoms passing through the heliosphere suggest somewhat higher H~I densities of $n({\\rm H~I})\\approx 0.2$ cm$^{-3}$ \\citep{eq94,vvi99}, possibly suggesting the presence of density and/or hydrogen ionization state variations within the LIC. Ionization state variations are to be expected since different parts of the LIC will be shielded from photoionization sources to different extents \\citep{fcb88,kpc90,jds02}.  Measurements of interstellar material flowing through the heliosphere and local interstellar medium (LISM) absorption line studies all suggest that in a heliocentric rest frame the LIC appears to be flowing towards Galactic coordinates $l=186.1^{\\circ}$ and $b=-16.4^{\\circ}$, with a speed of about 25.7 km~s$^{-1}$ \\citep{mw93,mw96,rl92,rl95,eq00}.  Probably the best line of sight for studying absorption by gas in the LIC is that toward Capella, which is a spectroscopic binary system (G8~III+G1~III) located 12.9~pc away, with Galactic coordinates $l=162.6^{\\circ}$ and $b=+4.6^{\\circ}$.  There are several reasons why this line of sight is ideal.  First of all, there is only one interstellar velocity component observed for this short line of sight, that of the LIC.  Short lines of sight are preferable for studying the LIC to avoid a complicated, multi-component ISM structure that is often difficult to resolve into individual components.  Secondly, although cool stars are not as bright in the UV as hot stars, Capella is the brightest cool star in the sky in the ultraviolet, providing sufficient background in the continuum and bright emission lines to observe numerous LIC absorption lines.  There are few nearby hot stars (including hot white dwarfs) that can be observed for LIC studies, which means one must generally observe cool stars like Capella.  The UV emission lines from the Capella stars are quite broad, making it easy to distinguish narrow ISM lines located within the broad emission profiles, which serve as the continuum for measuring the ISM lines.  Another advantage of the Capella line of sight is that LIC column densities are particularly high in this direction, allowing us to detect weak absorption lines that would be undetectable in other directions.  This is illustrated by Figure~1, a schematic picture of the structure of the LISM in the Galactic plane, showing that the Capella line of sight passes through the entire length of the LIC. The LIC outline in Figure~1 is from the model of \\citet{sr00}. The shape of the nearby ``G cloud'' is a crude estimate from \\citet{bew00}, and there is some question as to whether this cloud is truly distinct from the LIC \\citep*[see][]{bew02a}.  Also shown is an equally crude estimate of the shape of a cloud detected toward Sirius (dotted line), which might be the same cloud as the ``Hyades Cloud'' identified by \\citet{sr01} in the direction of the Hyades Cluster.  The lines of sight to the stars shown in Figure~1 have all been studied by the {\\em Hubble Space Telescope} (HST) to better understand the properties of the LISM.  Since all the stars are $<10^{\\circ}$ from the Galactic plane, distortions due to the projection effects in showing the three-dimensional LISM as a two-dimensional figure are not severe.  The direction toward the B2~II star $\\epsilon$~CMa is indicated in Figure~1.  This star is important because it is the dominant stellar source of ionizing photons irradiating the LIC \\citep{jvv95}.  Several studies of the Capella line of sight have been published using data from the Goddard High Resolution Spectrograph (GHRS) formerly aboard HST.  \\citet{jll93,jll95} analyzed LIC absorption lines of H~I, D~I, Mg~II, and Fe~II toward Capella, with the primary goal being to measure the D/H ratio within the LIC.  \\citet{avm98} also analyzed the HST/GHRS data and found close agreement with the results of \\citet{jll93,jll95}, although \\citet{avm02} claim that the analysis of H~I should have larger uncertainties due to the possible existence of undetected hot H~I absorption components.  \\citet{bew97} measured the electron density toward Capella using observations of C~II $\\lambda$1334.5 and C~II$^{*}$ $\\lambda$1335.7.  One problem with the GHRS instrument is that its one dimensional detector could only observe a relatively narrow wavelength region for each exposure.  However, in 1997 the GHRS was replaced with the Space Telescope Imaging Spectrograph (STIS).  The STIS instrument has capabilities similar to the GHRS in terms of spectral resolution and sensitivity, but it has a two-dimensional detector providing much broader wavelength coverage for each observation.  On 1999 September 12, HST/STIS observed the $1170-1710$~\\AA\\ spectrum of Capella using the E140M grating.  This observation provides access to many LIC absorption lines unavailable in the GHRS data set.  The {\\em Far Ultraviolet Spectroscopic Explorer} (FUSE) has also observed Capella recently.  The FUSE satellite obtains spectra at wavelengths shorter than those accessible to HST, from $905-1187$~\\AA, allowing access to still more ISM absorption features.  In this paper, we use the new FUSE and HST/STIS data to provide a complete analysis of UV absorption lines detected towards Capella. Our primary goal is to measure column densities for as many atomic species as possible to establish the ionization states and abundances of various elements within the LIC. ",
        "conclusions": "We have analyzed LIC absorption lines in HST/STIS and FUSE observations of Capella to measure column densities for as many atomic species as possible.  Our results are summarized as follows: \\begin{description} \\item[1.] We detect D~I and H~I Ly$\\beta$ absorption in the FUSE data, but the quality of the data do not allow us to improve on previous measurements of D~I and H~I from HST observations of Ly$\\alpha$. \\item[2.] The combined FUSE and HST/STIS data sets allow us to measure or estimate low upper limits for column densities of all three of the lowest ionization states of C, N, and Si.  This allows us to empirically establish the ionization states of these elements within the LIC.  At least 95\\% of C is in the form of C~II, at least 90\\% of Si is in the form of Si~II, and N is roughly half-ionized. \\item[3.] The C, N, and Si ionization states are consistent with the predictions of steady state photoionization models for the LIC computed by \\citet{jds02}, providing some support for the contention that photoionization alone can account for the observed ionization level of the LIC.  However, more must be known about the diffuse EUV background before this can truly be established. \\item[4.] Based on our column density measurements and previous ones, we measure total abundances for seven elements.  The heavy elements Mg, Al, Si, and Fe are all depleted relative to solar abundances by factors of about $10-30$, presumably due to the incorporation of these elements into dust grains.  The abundances of carbon, nitrogen, and oxygen are close to solar, although the error bars are large enough to also allow depletions of about a factor of 2 for these elements. Our measurements of O/H and N/H are consistent with previous LISM measurements \\citep{dmm97,dmm98,ml02,hwm02}. \\end{description}"
    },
    "0208/astro-ph0208022_arXiv.txt": {
        "abstract": "{\\small We have observed  microquasar \\grs at 1.28 GHz for 8 days from June 18 to July 1, 2001 using Giant Metrewave  Radio Telescope (GMRT). We have seen several isolated radio baby flares of varying intensity and  duration.  We have also observed  broad composite flares with  rise and decay times of few hours on June 28-29, 2001 few days prior to when source went to the ``plateau radio state'' on 3rd July, 2001. These broad radio flares consist of several overlapping baby radio flares. The source was in the low-hard X-ray state during this period. We compare these results with 15 GHz radio data from the Ryle telescope.} ",
        "introduction": "The long term  monitoring of microquasar  \\grs has  shown broad  correlation between the   X-ray and non-thermal radio emission \\cite{herm97}. The radio emission in \\grs can be  classified into three classes; (i) the relativistic superluminal radio jets of flux density $\\sim$ 1 Jy with decay time-scales of several days (\\cite{mira94}, \\cite{fend99}), (ii) the baby jets of 20 $-$ 40 min durations with flux density of 20 $-$ 200 mJy both in infrared (IR) and radio (\\cite{pool97}, \\cite{eike98}), and (iii) the plateau state with persistent radio emission of 20 $-$ 100 mJy for extended durations \\cite{muno01}. In the case of superluminal jets, the radio emission has steep spectra and are observed at large distances (400 $-$ 5000 AU) from the accretion disk \\cite{fend99}, \\cite{dhaw00}. The radio emission during other two classes has flat spectra and they occur close to the accretion disk (within a few tens of AU)\\cite{dhaw00}. The multi-wavelength studies have indicated strong disk-jet connection for class 2(\\cite{mira98}, \\cite{eike98}, \\cite{yada01}).   In this paper, we present our radio observations at 1.28 and 15 GHz during June 18 to July 3, 2001 (just prior to when  source went  to the plateau state on July 3). These observations include faint radio emission, isolated baby radio flares riding above smooth as well as slowly rising/decaying radio emission (broad flares), and large radio flare ($\\sim$ 200 mJy on June 20). \\begin{figure}[h] \\centering \\psfig{file=yadav1_1c.ps,width=10cm,angle=270} \\caption{The RXTE/ASM flux (left axis) and hardness ratio (right axis) during June 17 to July 3, 2001(top panel). The observed 1.28 GHz flux from GMRT and 15 GHz flux from Ryle telescope are shown in bottom panel (5min bin) for the same duration.} \\label{fig:ex} \\end{figure} ",
        "conclusions": "The 1.28 GHz radio observations were carried out  with a bandwidth of 16 MHz using the GMRT at Pune, India \\cite{swar91} on 2001 June 18, 22, 23, 27, 28, 29, 30 and July 1.  The flux density scale is set by observing primary calibrator 3C286 or 3C48. The integration time of 32 s was chosen. The data recorded from GMRT has been converted to FITS and was analysed using Astronomical Image Processing System ({\\tt AIPS}).  The details of observations/analysis is  given elsewhere \\cite{ishw02}. The radio observations at 15 GHz are from the Ryle telescope at Cambridge. Details of the instrument and analysis  can be found  elsewhere \\cite{pool97}.  The RXTE/ASM flux and X-ray hardness ratio (5-12 keV/1.5-5 keV) are  shown in the top panel of Figure 1. The radio lightcurves at 1.28 and 15 GHz produced at 5 min interval are shown in the bottom panel of Figure 1.  \\grs exhibited significant radio emission on all days, except on June 27,  when the source showed much weaker radio emission ($\\sim$ 5 mJy). Figure 1 brings out following important points: \\begin{enumerate} \\item The radio fluxes at 1.28 and 15 GHz are consistent and suggest a flat radio spectrum during these observations except during the large radio flare on June 20-24. The flux and decay time of this flare suggest that it belongs to the class of relativistic jets. The higher flux at 1.28 GHz is consistent with steep spectrum during the decay of this flare. \\item The X-ray hardness ratio changed from $\\sim$ 0.7 to $\\sim$ 0.85 around June 30 prior to when source went  to the plateau state on July 3. The RXTE/ASM flux  becomes stable at higher hardness ratio. The X-ray hardness ratio suggests that the source remains in the low-hard state during this period. \\end{enumerate}  \\begin{figure}[t] \\centering \\psfig{file=yadav1_2.ps,width=7cm,angle=270} \\caption{The radio light curve from GMRT at 1.28 GHz (mJy) observed on June 28, 2001 (5 min. bin).} \\label{fig:ex} \\end{figure} The radio light curve observed on June 28 is shown in Figure 2. The flux density was $\\sim$ 5 mJy at about UTC 17 hour and reached gradually to 50 mJy at UTC 24 hour. When the observations resumed on 29 (not shown here), the source was \"caught\" at 70 mJy at UTC 16 hour, and the flux started decaying slowly to the value of 10 mJy at UTC 24 hour. This is probably the first detail observation of the precursor flares to the plateau state with rise and decay times of over  six hours. It is also important to note that these broad flares and the change in the hardness ratio around June 30 set the stage for the plateau state.  These broad flares   consist of mini (baby) radio flares riding above rising/decaying radio emission. In contrast, the radio lightcurve observed on June 30 shows mini (baby) flares riding above almost smooth radio emission (Figure 3).  \\begin{figure}[h] \\centering \\psfig{file=yadav1_3.ps,width=7cm,angle=270} \\caption{The radio light curve from GMRT at 1.28 GHz (mJy)observed on June 30, 2001 ( 5 min. bin).} \\label{fig:ex} \\end{figure}  These isolated baby flares are modeled as adiabatically expanding synchrotron clouds ejected from the accretion disk \\cite{ishw02}. One such flare along with model fit data is shown in Figure 4. This model provides estimate of the spectrum index from single frequency observations and successfully explains the observed delay times between different frequencies. The radio emission during these observations is consistent with flat radio spectrum which is in agreement with observed flux at 15 GHz. \\begin{figure}[h] \\centering \\psfig{file=yadav1_4.ps,width=7cm,angle=270} \\caption{Isolated radio flare observed with GMRT at 1.28 GHz on 30 June, 2001 with model fit (for details see text).} \\label{fig:ex} \\end{figure}"
    },
    "0208/astro-ph0208352_arXiv.txt": {
        "abstract": "A problem still unsolved in cosmology is the identification of the sources of radiation able to reionize \\HI in the intergalactic medium (IGM) by $z \\sim 6$.  Theoretical works and observations seem to indicate that the fraction, \\fesc, of \\HI ionizing radiation emitted from galaxies that escapes into the IGM is small in the local universe (\\fesc$\\simlt 10$\\%). At high redshift galaxies are more compact and probably gas rich implying smaller values of \\fesc from their disks or spheroids. But if the sites of star formation are displaced from the disk or spheroid and the star formation efficiency of the proto-clusters is high, then \\fesc should be about one.  This star formation scenario is consistent with several models for globular clusters formation. Using simple arguments based on the observed number of globular cluster systems in the local universe and assuming that the oldest globular clusters formed before reionization and had \\fesc$\\sim 1$, I show that they produced enough ionizing photons to reionize the IGM at $z \\approx 6$. ",
        "introduction": "Observation of Ly$\\alpha$ absorption systems toward newly found high-redshift quasars \\citep{Becker:01, Djorgovski:01} indicate that the redshift of reionization of the intergalactic medium (IGM) should be close to $z=6$ \\citep{Gnedin:02, Songaila:02}. Perhaps the recent identification of a lensed galaxy at $z=6.56$ points to a somewhat earlier redshift of reionization \\citep{Hu:02}. Although quasars play a dominant role in photoionizing the IGM at $z \\approx 3$ \\citep{Meiksin:93}, their dwindling numbers at $z > 4$ suggest the need for another ionization source.  Unless a hidden population of quasars is found, radiation emitted by high-redshift massive stars seems necessary to reionize the universe. A key ingredient in determining the effectiveness by which galaxies photoionize the surrounding IGM is the parameter \\fesc, defined here as the mean fraction of Lyc photons escaping from galaxy halos into the IGM. To be an important source of ionizing photons and rival with quasars, a substantial fraction ($\\sim 10$\\%) of them must escape the gas layers of the galaxies \\citep{Madau:96}.  Cosmological simulations and semi-analytical models of IGM reionization by stellar sources find that the ionizing background rises steeply at the redshift of reionization. Unfortunately a direct comparison between models is difficult because of different recipes used for star formation, clumping of the IGM or the definition of \\fesc. But a result common to all the models is that, in order to reionize the IGM by $z =6 - 7$, the escape fraction must be relatively large: \\fesc$\\simgt 10$\\% assuming a Salpeter initial mass function (IMF) and the standard $\\Lambda$CDM cosmological model. \\cite{Benson:02} finds that \\fesc should be about 15\\% for reionization at $z = 6$, but a smaller value \\fesc$ \\simlt 10$\\% is consistent with the observed ionizing background at $z \\sim 3$. \\cite{Gnedin:02} finds that assuming a primordial power spectrum index $n=0.93$, the preferred value from CMB and LSS data, reionization at $z \\simgt 6$ requires a large \\fesc; but this assumption produces an ionizing background at $z \\simlt 4$, that is too large. The common assumption of a universal star formation efficiency (SFE) (\\eg, the coefficient in front of the Schmidt-Law in some models or in others the fraction $f_*$ of baryons converted into stars) is consistent with the observed values of the star formation rate (SFR) at $0<z<5$ and total star fraction $\\Omega_*$ at $z =0$. However, the assumption of a constant \\fesc does not seem to be consistent with observations.  An escape fraction \\fesc$\\sim 1$ is required for reionization at $z \\sim 6$ but the ionizing background at $z \\sim 3$ is consistent with \\fesc$\\simlt 10$\\% \\citep{Bianchi:01}.  Small values of \\fesc at $z \\simlt 3$ are also supported by direct observations of the Lyc emission from Lyman-break and starburst galaxies.  \\cite{Giallongo:02} find an upper limit \\fesc$< 16$\\% at $z \\sim 3$ \\citep[but see][]{Steidel:01} and observations of low-redshift starbursts are consistent with \\fesc upper limits ranging from a few percent up to 10\\% \\citep{Hurwitz:97,Deharveng:01}.  Calculations of \\fesc from first principles are difficult. The main complications arise in simulating a realistic interstellar medium (ISM) that includes small-scale physics and feedback processes. Moreover the mean \\fesc results from the contribution of a variety of galaxies whose ISM properties are largely unknown at high redshift. Theoretical models \\citep{Dove:00, Ciardi:02} for the radiative transfer of ionizing radiation through the disk layer of spiral galaxies similar to the Milky Way find \\fesc$\\sim 6-10$\\%. At high redshift the mean value of \\fesc is expected to decrease almost exponentially with increasing redshift \\citep{RicottiS:00, Wood:00}; at $z > 6$, \\fesc$\\simlt 0.1-1$\\% even assuming star formation rates typical of starburst galaxies (\\eg, SFR$\\sim 10$ times that of the Milky Way).  Using Monte-Carlo simulations, \\cite{RicottiS:00} have studied how \\fesc depends on galactic parameters.  Assuming gas density profile in hydrostatic equilibrium in the dark matter (DM) potential, star density proportional to the gas density and a power law for the luminosity function of the OB associations, they found that \\fesc$\\propto(\\epsilon f_g M_{DM})^{-1/3}\\exp[-(z_{vir}+1) \\epsilon^{-1/3}]$. Here $\\epsilon$ is proportional to the SFE, $f_g$ is the fraction of collapsed gas, $z_{vir}$ is the virialization redshift and $M_{DM}$ is the DM halo mass. The majority of photons that escape the halo come from the most luminous OB associations located in the outermost parts of the galaxy. Indeed \\cite{RicottiS:00} have shown that changing the luminosity function of the OB association and the density distribution of the stars has major effects on \\fesc (see their Figs. 8 and 9). In the aforementioned models, \\fesc should be regarded as an upper limit since dust extinction and absorption of ionizing radiation from the molecular cloud in which OB associations are born are neglected.  The theoretical suggestion of a decreasing \\fesc with increasing redshift is in contrast with models for reionization that require \\fesc$\\sim 1$ at $z = 6$. A different star formation mode, with very luminous OB associations forming in the outer parts of galaxy halos, could explain the large \\fesc required for reionization. Globular clusters (GCs) are possible observable relics of such a star formation mode.  Their redshift of formation is compatible with redshift of reionization \\citep{GnedinR:02}.  Because of their large star density they survived tidal destruction and represent the most luminous tail of the luminosity distribution of primordial OB associations.  In \\S~\\ref{ssec:mods} I explain that several models for the formation of proto-GCs imply an \\fesc$\\sim 1$.  I will also show that the total amount of stars in GCs observed today is sufficient to reionize the universe at $z \\sim 6$ if their \\fesc$\\sim 1$. This conclusion is reinforced if the GCs we observe today are only a fraction, $1/f_{di}$, of primordial GCs as a consequence of mass segregation and tidal stripping.  The paper is organized as follows. In \\S~\\ref{sec:rev} I briefly review recent progress in our understanding of GC properties and formation theories; in \\S~\\ref{sec:meth} I discuss the model assumptions in light of GC observations and present the results.  In \\S~\\ref{sec:disc} I present my conclusions. ",
        "conclusions": "\\label{sec:disc}  Observed Lyman break galaxies at $z \\sim 3$ are probably the most luminous starburst galaxies of a population that produced the bulk of the stars in our universe. Their formation epoch corresponds to the assembly of the bulges of spirals and ellipticals. Nevertheless the observed upper limit on \\fesc from Lyman break galaxies is \\fesc$\\simlt 10$\\%, insufficient to reionize the IGM according to numerical simulations.  Recently \\cite{Ferguson:02}, using different arguments, have claimed that the radiation emitted from Lyman break galaxies is insufficient to reionize the IGM assuming a continuous star formation mode.  I propose that GCs could produce enough ionizing photons to reionize the IGM.  Assuming $f_{di}=2$ (\\ie, during their evolution GCs have lost half of their original mass), I find $\\omega^f_{gc} \\approx 0.1$\\%, small compared to the total $\\omega^f_* \\sim 10$\\% at $z=0$. But GCs are around 12-13 Gyr old and, if they formed between $7<z<5$ (in about 0.5 Gyr), the expected total $\\omega^f_*$ formed during this time period is about $\\omega^f_* \\sim 1$\\%, only 10 times larger than $\\omega^f_{gc}$. Assuming $f_{di}=20$, expected from the results of N-body simulations, I find $\\omega^f_{gc} \\approx 1$\\%, suggesting that GC formation is an important mode of star formation at high-redshift.  The special star formation mode required to explain the formation of GCs suggests an \\fesc$\\sim 1$ from these objects. This is because \\fesc is dominated by the most luminous OB associations and GCs are extremely luminous, emitting $S \\sim 3 \\times 10^{53}$ s$^{-1}$ ionizing photons in bursts lasting 4 Myrs.  Moreover, according to many models, GCs form in the hot, collisionally-ionized galaxy halo, from which all the ionizing radiation emitted can escape into the IGM.  Therefore it is not too surprising, if GCs started forming before $z = 6$, that their contribution to reionization is large. I find that the number of ionizing photons per baryon emitted in a Hubble time at $z=6$ by GCs is ${\\cal N}^{gc}_{ph}=(5.1^{+4.3}_{-3.2})f_{di}/\\Delta t_{gc}>1$, therefore sufficient to reionize the IGM even if we assume $f_{di}=1$. Here, $\\Delta t_{gc} \\sim 0.5-2$ is the period of formation of the bulk of old GCs in Gyrs. Using simple calculations based on Press-Schechter formalism (see Fig.~\\ref{fig:nph}) I find that, if normal star formation in galaxies have \\fesc$ \\simlt 5$\\%, GC contribution to reionization should be important. If GCs formed by thermal instability in the halo of $T_{vir} \\sim 10^5$ K galaxies (case (iii)), the ionizing sources are located in rare peaks of the initial density field. Therefore, the mean size of intergalactic \\HII regions before overlap is large and reionization is inhomogeneous on large scales.  In this letter I have considered the possibility that an increasing \\fesc at $z \\sim 6$ due to GCs formation could explain IGM reionization and still be consistent with the observed values of the ionizing background at $z <3$. Alternatively an increasing production of ionizing photons per baryon converted into stars, due to a varying IMF, would have similar effects on the IGM. Chemical evolution studies should be able to distinguish between these two scenarios."
    },
    "0208/astro-ph0208487_arXiv.txt": {
        "abstract": "Most astrophysical systems, e.g. stellar winds, the diffuse interstellar medium, molecular clouds, are magnetized with magnetic fields that influence almost all of their properties. One of the most informative techniques of magnetic field studies is based on the use of starlight polarization and polarized emission arising from aligned dust. How reliable the interpretation of the polarization maps in terms of magnetic fields is the issue that the grain alignment theory addresses. Although grain alignment is a problem of half a century standing, recent progress achieved in the field makes us believe that we are approaching the solution of this mystery. I review basic physical processes involved in grain alignment and discuss the niches for different alignment mechanisms. I show why mechanisms that were favored for decades do not look so promising right now, while the radiative torque mechanism ignored for more than 20 years looks so attractive. I define the observational tests and outline the circumstances when grain alignment theory predicts that new yet untapped information of magnetic field structure is available through polarimetry. In particular, I touch upon mapping magnetic fields in circumstellar regions, interplanetary space and in comet comae. ",
        "introduction": "Magnetic fields are of utmost importance most astrophysical systems. Conducting matter is entrained on magnetic field lines and magnetic pressure and tension are very important for its dynamics. For instance, galactic magnetic fields play key role in many processes, including star formation, mediating shocks, influencing heat and mass transport, modifying turbulence etc. Aligned dust grains trace the magnetic field and provide a unique source of information about magnetic field structure. How reliable is this source of information? What are the prospects of the polarimetric research? This review addresses those questions while dealing with the problem of grain alignment theory.   Grain alignment of interstellar dust has been discovered more than half a century ago. Hall (1949) and Hiltner (1949) reported polarization that was attributed to the differential extinction of starlight by dust particles with longer axes preferentially aligned. Very soon it was realized that the alignment happens with respect to the interstellar magnetic field\\footnote{The relation between grain alignment direction and that of magnetic field is clear from a comparison of synchrotron polarization maps and those of galactic starlight polarization (see Serkowski, Mathewson \\& Ford 1975). Recent measurements of polarization in external galaxies (see Jones 2000) makes this relation even more obvious.} Starting from that moment polarized starlight and later the polarized emission by aligned grains have become the principal technique of studying magnetic field morphology in molecular clouds. As magnetic fields are thought to control star formation (see Savier, McKee \\& Stahler 1997) the value of the technique is difficult to overestimate. However, to what extend the polarization maps trace magnetic fields is a non-trivial question that the grain alignment theory deals with.   For many years grain alignment theory had a very limited predictive power and was an issue of hot debates. This caused somewhat cynical approach to the theory among some of the polarimetry practitioners who preferred to be guided in their work by the following rules of thumb: {\\it All grains are always aligned and the alignment happens with the longer grain axes perpendicular to magnetic field.} This simple recipe was shattered, however, by observational data which indicated that \\\\ I. Grains of sizes smaller than the critical size are either not aligned or marginally aligned (Mathis 1986, Kim \\& Martin 1995).\\\\ II. Carbonaceous grains are not aligned, but silicate grains are aligned (see Mathis 1986).\\\\ III. Substantial part of grains deep within molecular clouds are not aligned (Goodman et al. 1995, Lazarian, Goodman \\& Myers 1997).\\\\ VI. Grains might be aligned with longer axes parallel to magnetic fields\\footnote{A simple, but not always clearly understood property of grain alignment in interstellar medium is that it always happens  in respect to magnetic field. It can be shown that the fast (compared with other time scales) Larmor precession of grains makes the magnetic field the reference axis. Note, however. that grains may align with their longer axes {\\it perpendicular} or {\\it parallel} to magnetic field direction. Similarly, magnetic fields may change their configuration and orientation in space (e.g. due to Alfven waves), but if the time for such a change is much longer than the Larmor period the alignment of grains {\\it in respect to the field lines} persists as the consequence of preservation of the adiabatic invariant.} (Rao et al 1998).   These facts could persuade even the most stubborn types that the interpretation of interstellar polarimetric data does require an adequate theory. A further boost of the interest to grain alignment  came from the search of Cosmic Microwave Background (CMB) polarization (see Lazarian \\& Prunet 2002, for a review). Aligned dust in this case acts as a source of a ubiquitous foreground that is necessary to remove from the data. It is clear that understanding of grain alignment is the key element for such a removal.  With the present level of interest to the CMB polarization we are bound to have a lot of microwave and far infrared polarimetry data. It is important to understand to what extend this data reflects the structure of magnetic field in the Galaxy and whether this data can be used to get insight into the processes of galactic magnetic field generation and into interstellar turbulence\\footnote{Velocity and magnetic field statistics provide the most clear insight in what is going on with the turbulence. With velocity statistics available through the recently developed Velocity Channel Analysis (VCA) technique ( Lazarian \\& Pogosyan 2000) magnetic fields statistics is the missing element. Polarized starlight and emission from aligned grains provide the easiest way to get such a statistics.}.  While the alignment of interstellar dust is a generally accepted fact, the alignment of dust in conditions other than interstellar has not been fully appreciated. The common explanation of light polarization from comets or circumstellar regions is based on light scattering by randomly oriented particles. The low efficiency and slow rates of alignment were quoted to justify such an approach (see Bastien 1988). This point of view is common in spite of the mounting evidence in favor of grain alignment (see Briggs \\& Aitken 1986, Aitken et al. 1995, Tamura et al. 1995). However, recent advances in understanding of grain alignment show that it is an efficient and rapid process. Therefore, we do expect to have circumstellar, interplanetary and comet dust aligned. This opens new exciting avenues for polarimetry.  Traditionally linear starlight polarimetry was used. These days far infrared polarimetry of dust emission has become the major source of molecular field structure data (see Hildebrand 2000). It is possible that circular polarization may become an important means of probing magnetic fields in circumstellar regions and comets.  In this review I claim that the modern grain alignment theory allows us to solve most of the existing puzzles and can be used successfully to interpret polarimetry in terms of magnetic field. A substantial part of the review is devoted to the physics of grain alignment, which is deep and exciting. It is enough to say that its study resulted in a discovery of a few new solid state effects. The rich physics of grain alignment (see Fig~1a for an illustration of motion complexity) presents a problem, however, for its presentation. Therefore I shall describe first the genesis of ideas that form the basis of the present-day grain alignment theory. The references to the original papers should help the interested reader to get the in-depth coverage of the topic. Earlier reviews on the subject include Hildebrand (1988), Roberge (1996), Lazarian, Goodman \\& Myers (1997), Lazarian (2000).  In what follows we show how the properties of polarized radiation is related to the statistics of aligned grains (section~2), analyze the major alignment mechanisms (section~3), discuss observational data that allows to distinguish between different alignment processes (section~4) and outline the prospects of using grain alignment to study circumstellar, interplanetary magnetic fields (section~5). A discussion and summary are provided in sections~6 and 7. ",
        "conclusions": "I anticipate a number of questions that can worry the reader. For instance:  $\\bullet$ Does the review cover all the astrophysically important situations when grain alignment is important? It has become clear recently that grain alignment should happen in various astrophysical conditions. Polarized radiation from neighboring galaxies (Jones 2000), galactic nuclei (see Tadhunter et al 2001), AGNs, Seyfet galaxies (see Lumsden et al 2001), accretion discs (see Aitken et al 2002) can be partially due to aligned particles. Revealing this contribution would allow to study magnetic fields in those and other interesting objects.  $\\bullet$ To what degree do aligned grains reveal magnetic field geometry/topology during star formation? It is generally accepted that star formation starts with the accumulation of interstellar gas that is caused by turbulence and gravity. Aligned grains allow to trace magnetic fields during this preliminary stage. At some point of evolution the conditions within molecular clouds approach equilibrium with the alignment being shut down (see LGM). Finding out exactly when this happens is extremely important and this requires the quantitative description of grain alignment processes. Consider, for instance, radiative torques. Realistic clumpy, fractal-type structure of molecular clouds allows photons to penetrate much deeper into clouds compared with the idealized uniform structure frequently assumed in theoretical modeling. Therefore we expect grains within skin layers of the clumps to be aligned and to reveal magnetic field up to a substantial optical depth. Simulations in (Padoan et al. 2001) support this argument. As  protostars are formed in molecular clouds their light induces grain alignment in their neighborhood. The size of this neighborhood also depends on the cloud inhomogeneity in the protostar vicinity as well as on the radiative torque efficiency as a function of wavelength. The fact that grains in molecular clouds are larger than their counterparts in diffuse media allows for a more efficient alignment by starlight reddened by dust extinction; this increases the neighborhood volume. Therefore we expect to be able to trace magnetic evolution via polarimetry through important stages of star formation. Additional information can be available through microwave emission of the aligned PAH-type tiny grains, which rotate non-thermally due to their collisions with ions (see Draine \\& Lazarian 1998b). The abundance of such grains in molecular clouds is poorly known, however.  $\\bullet$ What is the advantage of far-infrared polarimetry for studies of magnetic field in molecular clouds compared to optical and near infrared ones? The trivial answer is that far infrared polarimetry reveals aligned grains near newly born stars unaccessible by optical and near infrared photons. A more subtle but essential effect is that photons, as we discussed earlier, can align grains within skin layers of clumps rather far into molecular clouds. Those aligned grains are only accessible by far infrared polarimetry. This, for instance, makes SOFIA airborn observatory so desirable for studies of magnetic fields in molecular clouds. Additional advantage of far infrared spectropolarimetry  stems from the fact that it allows us to separate contributions from different parts of the cloud (see Hildebrand 2000). This enables tomography of magnetic field structure.  $\\bullet$ What is the future of optical and near infrared polarimetry? It would be wrong to think that with the advent of far infrared polarimetry there is a bleak future for extinction polarimetry at shorter wavelengths. In fact, its potential for studies of magnetic fields in the Galaxy is enormous (see Fosalba et al. 2002, Cho \\& Lazarian 2002a). The possibility of using stars at different distances from the observer allows to get an insight into the 3D distribution of magnetic fields. In general, however, it is extremely advantageous to combine optical/near infrared and far infrared polarimetry. For instance, it may be pretty challenging to trace the connection of Giant Molecular Clouds (GMCs) with the ambient interstellar medium using just far-infrared measurement. However, if extinction polarimetry of the nearby stars is included, the task gets feasible. Similarly testing modern concepts of MHD turbulence (Goldreich \\& Shridhar 1995, Lithwick \\& Goldreich 2001, Cho \\& Lazarian 2002b) and turbulent cloud support (see reviews by McKee 1999, and Cho, Lazarian \\& Vishniac 2002) would require data from both  diffuse and dense media.  $\\bullet$ What is the advantage of doing polarimetry for different wavelengths? The list of advantages is pretty long. It is clear that aligned grains can be successfully used as pick up devices for various physical and chemical processes, provided that we understand the causes of alignment. Differences in alignment of grains of different chemical composition (see Smith et al. 2000) provides a unique source of the valuable information. Comets present another case in support of simultaneous multifrequency studies. There the properties of dust evolve in a poorly understood fashion and this makes an unambiguous interpretation of optical polarimetry rather difficult. Degrees and directions of dust alignment that can be obtained that can be obtained via far infrared polarimetry can be used to get a self-consistent picture of the dust evolution and grain alignment.  $\\bullet$ Do we need grain alignment theory to deal with polarized CMB foregrounds? Polarized emission spectrum arising from aligned dust may be very complex if grains having different temperatures exhibit different degrees of alignment. In this situation the use of the naive power-law templates may result in huge errors unless we understand grain alignment properly. Needless to say that grain alignment theory is necessary to predict the spectrum of polarized emission from PAHs in the range of 10-100~GHz.   $\\bullet$ What is the future of grain alignment theory? Although the recent progress in understanding grain alignment is really encouraging, it would be a mistake to think that grain alignment theory does not require intensive work any more. For instance, radiative torques alignment in the presence of starlight anisotropy should be treated as an experimental fact obtained via simulations rather than a theoretically understood effect. Moreover, crossover dynamics must be added to the existing code to get the simulations more realistic and frequency dependence of radiative torques should be quantified. More special cases of alignment should be studied. The simultaneous action of various processes, e.g. grain streaming together with the action of radiative torques must be investigated. Some additional processes, e.g. mechanical alignment of helical grains (see table~1. in LGM) must be quantified. Alignment of tiny PAH grains, in particular, is an essentially unexplored field that requires more studies of relaxation processes in minute quantum mechanical samples as well as plasma and magnetic turbulence interactions with grains. More observational testings are necessary as well. For instance, comets allow to trace grain alignment in time. More systematic studies that include not only linear polarimetry, but circular polarimetry as well, should be made. All in all, grain alignment has become a predictive theory, but there is more work, both observational and theoretical to be done."
    },
    "0208/astro-ph0208164_arXiv.txt": {
        "abstract": "The discovery of two accreting millisecond X-ray pulsars in binaries with $\\approx 43$ minute orbital periods allows for a new probe of the donor's structure. For \\xtegsfc, only a hot white dwarf (WD) can fill the Roche Lobe. A cold He WD is a possible solution for \\xtemit, though I will show that evolutionary arguments make a hot WD more likely. In addition to being larger than the $T=0$ models, these finite entropy, low-mass ($M_c<0.03M_\\odot$) WDs have a minimum mass for a fixed core temperature. If they remain hot as they lose mass and expand, they can ``evaporate'' to leave an isolated millisecond radio pulsar. They also adiabatically expand upon mass loss at a rate faster than the growth of the Roche radius if the angular momentum deposited in the disk is not returned to the donor.  If the timescale of the resulting runaway mass transfer is shorter than the viscous timescale in the outer disk, then the mass transfer instability of Ruderman and Shaham for He WDs would be realized. However, my estimates of these timescales still makes the instability unlikely for adiabatic responses. I close by noting the possible impact of finite $T$ WDs on our understanding of AM CVn binaries. ",
        "introduction": "The recent discovery by the {\\it Rossi X-Ray Timing Explorer} of two accreting millisecond pulsar transients \\xtegsfc \\ ($\\nu_s\\approx 435 \\ {\\rm Hz}$; Markwardt et al. 2002) and \\xtemit \\ ($\\nu_s\\approx 185 \\ {\\rm Hz}$; Galloway et al. 2002) has allowed us to learn about the donors in these ultracompact binaries. At $P_{\\rm orb}\\approx 43$ minute orbital periods, H-rich donors are ruled out (Nelson, Rappaport \\& Joss 1986), and He-rich stars (see Podsiadlowski, Rappaport \\& Pfahl 2002 for an updated discussion of the range of H/He ratios) or WDs are filling the Roche lobe (RL). If a degenerate WD, then the orbital period increases as mass is transferred at the rate set by angular momentum losses from gravity waves, $\\dot J_{\\rm GR}$ (see Verbunt 1993).  The measured pulsar orbital parameters yield the RL filling companion's mass, $M_{\\rm c}$, and the minimum mass transfer rate, $\\dot M_{\\rm GR}=3M_{\\rm c}\\dot J_{\\rm GR}/2J$, which are shown in Figure~1 for neutron stars of $M_{x}=1.4-2.0 M_\\odot$.  The low $\\dot M$'s are the likely cause for the transient behavior, as the steady-state outer disk temperature is below these element's ionization temperature (Tsugawa \\& Osaki 1997; Menou et al. 2002), even if X-ray heating is included at the rate inferred in other X-ray binaries (Dubus, Hameury \\& Lasota 2001). If identical WDs are the donors in both of these binaries, then their identical orbital periods would require the same RL filling solutions. This would constrain the inclination for \\xtemit \\ to less than 37 degrees and $M_c>0.013M_\\odot$ for both \\xtemit \\ and \\xtegsfc.  The dotted lines in Figure 2 show the mass-radius relation for RL filling donors of \\xtegsfc\\ (lower line) and \\xtemit\\ (upper line). The solid and dashed lines (which exhibit a maximum radius due to the onset of Coulomb physics) are the cold ($T=0$) WDs of Zapolsky \\& Salpeter (1969, hereafter ZS) for pure He and C, respectively. A cold He WD will fill the RL for \\xtemit \\ (Galloway et al. 2002), whereas there are no cold WD solutions that fill the RL for \\xtegsfc \\ (Markwardt et al. 2002).  This evidence for finite $T_c$ WDs motivated my calculations shown in Figure 2 by the solid and dashed lines that diverge at low $M_c$. These models (see \\S 2) are for He at $T_c=10^5\\ {\\rm K}$ and $10^6 \\ {\\rm K}$ and C at $T_c=10^6\\ {\\rm K}$ and $T_c=3\\times 10^6 \\ {\\rm K}$. These provide RL filling solutions, but don't {\\it a priori} differentiate between He or C or any other element that might dominate the donor star. I consider C WDs throughout this paper, as Schulz et al. (2001) have measured a high Ne to O ratio in the matter transferred onto NSs in ultracompact X-ray binaries (confirmed by Juett, Psaltis \\& Chakrabarty 2002; Homer et al. 2002; Juett \\& Chakrabarty 2002). These measurements led Schulz et al. (2001) and Juett et al. (2001) to suggest that the donors in these binaries are the cores of previously crystallized C/O WDs.  In \\S 2 I justify (on evolutionary grounds) that hot WDs are expected, describe the construction of their $R_c(M_c)$ relations, and point out the existence of a minimum $M_c$ solution for a given $T_c$.  I put the models in the context of mass donors in \\S 3 and calculate \\begin{equation} \\label{eq:nad} n_{\\rm Ad}={d\\ln R_c\\over d\\ln M_c}, \\end{equation} the adiabatic exponent needed to evaluate the stability of mass transfer. Though uncertainties remain regarding the ability of the accretion disk to store angular momentum on long timescales (Verbunt \\& Rappaport 1988), my initial work finds that the expansion of a He WD donor under mass transfer can exceed that of the RL, possibly allowing for the Ruderman \\& Shaham (1983) instability. ",
        "conclusions": "The timing of two accreting millisecond pulsars (Markwardt et al. 2002; Galloway et al. 2002) in ultracompact binaries has probed the WD donor properties to new levels and shown that they are of finite entropy. This motivated my calculations of low-mass WDs of finite $T_c$ that allow for $T_c$ to be constrained.  For He WDs, the implied $T_c$ are nearly that expected just from adiabatic expansion of the initially hot WD that filled the RL. Only a small amount of tidal heating is needed. More tidal heating is needed to make a C/O WD fill the RL.  These finite $T$ solutions allowed for a re-evaluation of Ruderman \\& Shaham's (1983) scenario for making isolated millisecond radio pulsars via a mass transfer instability. I find that the adiabatic mass transfer instability can occur for a hot He WD as long as the angular momentum leaving the RL filling star is not returned to the orbit. I have thus eliminated one criticism of their model, though the question of angular momentum elimination remains a serious one. The physics of hot, low-mass WDs also yields a minimum mass WD solution for a fixed $T_c$ so that a mass transfer instability can occur if the donor remains isothermal under mass loss.  I thus speculate that an evaporative or mass transfer instability endpoint might occur as long as tidal (or other) heating persists at 40-80 minute orbital periods.  This work also impacts AM CVn binaries, where a low mass He star donates material to a more massive WD (see Solheim 1995 for a review). The larger WD radii lead to more GW emission and a higher $\\dot M_c$ than expected for a given $P_{\\rm orb}$. Hence, models which track the WD entropy will fall between the degenerate and non-degenerate models in Nelemans et al. (2001) and depend on both the age of the system when RL filling occurs (as this fixes the initial WD entropy) and any tidal heating that occurs during the mass transfer. If either of the instabilities discussed above occur, then the endpoint of AM CVn's could well be a DB WD (e.g. Tutukov \\& Yungelson 1996)."
    },
    "0208/astro-ph0208493_arXiv.txt": {
        "abstract": " ",
        "introduction": "The first images of the nebula around $\\eta$ Carinae were made by Gaviola (1946, 1950) and Thackeray (1949, 1950). Gaviola named the nebula according to the geometry he identified at that time the {\\it Homunculus}. It had a size of somewhat larger than 10\\arcsec. Nowadays we know that the Homunculus is highly symmetric---bipolar---and is larger. The Homunculus is only the central part of the nebula around $\\eta$ Carinae (e.g.  Walborn 1976), which in total extends to a diameter of 60\\arcsec\\ (0.67\\,pc). While we still call the central bipolar structure the Homunculus, all outer filaments are combined into what is known as the {\\it  outer ejecta}. The sizes and morphology of structures in the outer ejecta are manifold. In contrary to the Homunculus is the outer ejecta not symmetric, nor is it a coherent object but consists out of quite a number of filaments, bullets or knots. Fig. \\ref{figure1} (left) shows an HST image taken with the F658N filter of the nebula around $\\eta$ Carinae. Besides the different morphology of the Homunculus and the outer ejecta, it illustrates also the large difference in brightness. Therefore the Homunculus is additionally plotted with contours. If instead we compare the kinematics of Homunculus and outer ejecta, they seem more alike. The Homuculus, according to his bipolarity expands with about 650\\,\\kms (Davidson \\& Humphreys 1997, Currie et al. 1996) and with the south-east lobe approaching and the north-west lobe receeding. The outer ejecta has on average similar velocities, the majority of the structures expanding with $|v_{\\rm exp}|$ = 600\\,\\kms (e.g. Meaburn et al. 1996, Weis et al. 2001a). But note that a significant fraction of the filaments move much faster, they reach velocities as high as 2000\\,\\kms (Weis 2001a,b). So in velocity space nevertheless the outer ejecta are more ordered than expected from the morphology. In the south-eastern region most filaments are blueshifted, while in the north-west the clumps are redhifted, see right panel in Fig. \\ref{figure1}. Compared to the movement of the Homunculus the expansion of the outer ejecta is along a very similar symmetry axis. As the Homunculus the outer ejecta has a bi-directional (bipolar) movement.  \\begin{figure} \\includegraphics[width=8.4cm]{hstbild.ps}\\vspace{0.3cm} \\includegraphics[width=8.4cm]{overlay_velhst.ps} \\caption{{\\it Left:} An HST/F658N image of the nebula around $\\eta$ Carinae, the brightest regions, the Homunculus, are additionally overplotted with intensity contours, for a better illustration and comparison. {\\it Right:} Same HST image but with radial velocities overplotted at certain regions. The font size is larger for larger velocities. Red and blue (plus underlined) colors indicate red- (positive) and blueshifted (negative) radial velocities. }\\label{figure1} \\end{figure} ",
        "conclusions": "The nebula around $\\eta$ Carinae can be divided into two quite different parts, the Homunculus a bipolar reflection nebula about 0.2\\,pc across and the 0.67\\,pc large outer ejecta, a filamentary emission nebula. We have shown that the kinematics of the outer ejecta indicate that also this part of the nebula expands bi-directional, or bipolar. The symmetry axis is similar to that of the Homunculus. \\\\Historically we know that $\\eta$ Carinae und its nebula emit X-rays. With CHANDRAs unprecedented spatial resolution a very accurate comparison of the X-ray emission with the optical emission could be made. The X-ray emission is more homogeneous and smoother than the optical nebula, which is more of a accumulation of filaments, knots and bullets. While the sizes match, that is we find X-ray emission in regions were there is optical emission we barely find an agreement of individual knots, except for the S condensation. Comparing the intensity maxima of the X-ray emission   with the velocities of the optical filaments yields a much better consensus. We conclude that in the case of $\\eta$ Carinae's outer ejecta the faster moving filaments are able to form stronger shocks and therefore stronger X-ray emission. The temperature of 0.65\\,keV indicates post-shock velocities of 750\\,\\kms, in agreement with the measurements.  \\\\ From the new HST-STIS spectra of the Strings we obtain a first analysis of new parameters of the Strings. We can see that the Strings starts abruptly and does not extend back into the Homunculus. The slowest velocity of String\\,1 was denoted with $-290$\\,\\kms. The fastest is similar  to the previous measurements with about $-950$\\,\\kms. We determined several line ratios for String\\,1, the most interesting of which is the [S\\,{\\sc ii}] ratio, a density indicator. For String\\,1 this ratio is about 0.5 $\\pm$ 0.1. Within the errors the ratio is steady along the String. The ratio is close---but clearly not always at---the high density limit of the line, so we still can determine an electron density of 10$^4$\\,cm$^{-3}$. The density, as the line ratios in general are for all Strings alike. The Strings are more of a denser steady flow rather than ablating bullets, shadows or knots on chains. Since the HST-STIS spectra we have taken for the Strings also contain information of the immediate surrounding---that is the outer ejecta, mainly the S ridge and {\\it W\\,Arc}---we could also determine the density of several filaments in this region. Measurements in these spectra show that the filaments in the outer ejecta have a density of the order of 10$^4$\\,cm$^{-3}$, thus about the same as the Strings. This value does not change significantly---at least in the regions which are covered by out HST-STIS observations. If we assume that all filaments in the outer ejecta have roughly this density, and take a reasonable filling factor for the knots in the ejecta of 1\\% within the total (0.67)$^3$\\,pc$^3$ volume of the outer ejecta, the total mass is at least 0.5\\,M$_{\\sun}$.\\\\  {\\bf  \\noindent Acknowledgments:} The author thanks Michael F. Corcoran and Kris Davidson for help and discussion on the CHANDRA data and Ted Gull for providing a much-better-than-pipeline HST-STIS data reduction and advice on technical aspects concerning these data. Special thanks go to Dominik Bomans for discussion on the subject and commending on the manuscript.\\\\"
    },
    "0208/astro-ph0208346_arXiv.txt": {
        "abstract": "{\\sl Identified radio supernova remnants (SNRs) in the Galaxy comprise an incomplete sample of the SNR population due to various selection effects. ROSAT performed the first all-sky survey with an imaging X-ray telescope, and thus provides another window for finding SNRs and compact objects that may reside within them. Performing a search for extended X-ray sources in the ROSAT all-sky survey database about 350 objects were identified as SNR candidates in recent years (Busser 1998). Continuing this systematic search, we have reanalysed the ROSAT all-sky survey (RASS) data of these candidates and correlated the results with radio surveys like NVSS, ATNF, Molonglo and Effelsberg. A further correlation with SIMBAD and NED was performed for subsequent identification purposes. About 50 of the 350 candidates turned out to be likely galaxies or clusters of galaxies. We found 14 RASS sources which are very promising SNR candidates and are currently subject of further follow-up studies. We will provide the details of the identification campaign and present first results. } ",
        "introduction": "Supernovae (SNe) are rare events, believed to occur at intervals of $\\sim$30-50~years in the Galaxy (van den Berg \\& Tamman 1991; Tamman et al. 1994). However, in the past 2000 years, only 7 Galactic SNe have been observed~--- SN 185 (RCW86), SN 386 (G11.2-0.3), SN 1006, SN 1181 (3C58), Crab SN, Tycho SN, and Kepler SN. Most Galactic SNe appear unobserved owing to visible-band extinction by interstellar dust. When observational techniques in the radio band became available and the SNR Cas~A was found to be the brightest radio source in the sky, several surveys and directed searches for SNR were performed at decimeter-wavelengths. Of course, radio observations are unhampered by interstellar dust. Currrently, about 250 SNRs have been identified in the radio band (Green 2000 and references therein). Although this catalog represents the result of more than 40 years of intensive search with the largest radio telescopes, it is incomplete and strongly biased by mainly two selection effects: (i) the surface brightness of the remnant must be above the sensitivity limit of the observations, and (ii) the angular size of the remnant must be at least several times the resolution of the observations. These effects mean that not only are old faint remnants missing in the current catalogues, but there is also a deficit of young but distant SNRs (Green 1991). This is only a small fraction of SNRs of the total number expected according to the galactic SNe rate and the life-time of their remnants. \\newline The RASS was performed from 1990 June to 1991 February and is the first (and only) all-sky survey done with an imaging X-ray telescope and thus provides another window that can be exploited to find SNRs as well as the compact objects that may reside within them (Aschenbach 1996). ROSAT was sensitive at soft X-ray energies (0.1-2.4~keV), had an angular resolution in survey mode of $\\sim$$96^{''}$ and a limiting survey sensitivity of $f_x\\sim 3 \\times 10^{-13}$ erg~cm$^{-2}$~s$^{-1}$ (Voges et al. 1999). The exposure time varied between about 400 s and 40,000 s in the ecliptic plane and poles respectively. ",
        "conclusions": "About 70 galactic SNRs are listed in Green's SNR Catalogue (Green 2000) being detected in X-rays. Making use of the spectro-imaging capability provided by ROSAT in its all-sky survey, we reanalysed a sample of about 350 extended X-ray sources, which were proposed by (Busser 1998) as SNR candidates. Performing a more dedicated imaging analysis and making use of recently updated databases like SIMBAD and the NED  we reduced the number of SNR candidates to be about 230. 14 of these targets turn out to be promising SNR candidates based on available data and are subject of follow-up observations for clarification. Two candidates were recently observed with the Chandra observatory and another six candidates with the Effelsberg 100-m telescope. The analysis of these data is currently in progress. We will continue this campaign providing additional observations in X-rays with XMM-Newton and Chandra and in radio with Effelsberg and ATNF telescopes. Optical observations will complete our multi-wavelength studies."
    },
    "0208/astro-ph0208036_arXiv.txt": {
        "abstract": "We discuss the potential of future sub-10 GeV threshold imaging atmospheric Cherenkov telescope arrays for exploring the physics of rotation  powered pulsars and their interactions with the  ambient medium  through relativistic winds and termination shocks. One such telescope is the high-altitude concept called ``5@5'' recently suggested by  Aharonian et al. (2001). 5@5,  with its enormous detection area exceeding  $10^{4} \\, \\rm m^2$ at the threshold energy of about 5 GeV,   combines two  distinct features  of the current satellite-borne (large photon  fluxes at GeV  energies) and ground-based  (large detection areas at TeV energies) gamma-ray astronomies.  Such an  instrument  would allow comprehensive studies of  temporal and spectral  characteristics of $\\gamma$-ray pulsars in the crucial   5 to 30 GeV energy interval. An equally  important topic   in the program of pulsar studies by 5@5  would   be  the search for GeV $\\gamma$-radiation from other radio pulsars at a few mVela level.   And finally, the searches  for pulsed  radiation components in the spectra of a large fraction of   unidentified EGRET  sources  (suspected to be pulsars)  without invoking information from  lower  (radio, optical, X-ray) frequency domains,  seems to be another  important issue,  because the periodic signals at  lower energies  could be significantly  suppressed in  many  cases. The  detection rate of $\\gamma$-rays from   ``standard''  EGRET sources by 5@5 is expected to exceed several  events per one second.  This should  provide an adequate photon statistics for the search  for periodic signals at the flux level of 3 mVela within the observation time of 3 h or so  (a time resolution below which any change of a signal's phase can be ignored). The spectral coverage by 5@5 and its  flux sensitivity are  nicely suited  for studying  other  aspects of  pulsar physics and astrophysics, in particular for detecting unshocked relativistic pulsar winds,  as well as  for quantifying characteristics of  pulsar driven synchrotron  nebulae through the inverse   Compton radiation   at energies  between  several GeV and several 100 GeV.  The Vela pulsar, the brightest $\\gamma$-ray source on the sky,  is  an ideal laboratory for  practical realization of  these  unique observational  possibilities. ",
        "introduction": "The number of cataloged  radio pulsars  -- single neutron stars powered by fast  rotation -- exceeds 1200 (Lorimer 2001). Only seven of them (with two other possible candidates) are reported by the EGRET team as GeV $\\gamma$-ray sources (Thompson 1999, Kaspi et al. 2000) . Actually, this is not a big surprise, and can be explained by   modest flux sensitivities of $\\gamma$-ray instruments. For example, the minimum detectable flux by EGRET of about $10^{-11} \\, \\rm erg/cm^2 s$ is  5 orders of magnitude larger than the sensitivity  achieved by radio telescopes at GHz frequencies.  This striking difference in sensitivities is compensated, to a certain extent,   by much larger energy fluxes  in $\\gamma$-rays. While the  pulsars emit in the radio band a tiny $\\leq 10^{-6}$ fraction of their rotational energy (Manchester \\& Taylor 1977), the $\\gamma$-ray  luminosities   of some of the  EGRET pulsars above $100 \\, \\rm MeV$ exceed one per cent of their  spin-down luminosities. It is expected that  the Gamma-ray Large Area Space telescope (GLAST),   the future major satellite-based high energy $\\gamma$-ray detector with flux sensitivity as good as $10^{-12} \\, \\rm erg/cm^2 s$ (Gehrels \\& Michelson 1999),   will  increase the number of {\\it radio} pulsars seen also in  $\\gamma$-rays  by at least an order of magnitude  (Thompson 2001). Moreover,   the radio pulsars, which are observable only when their  radio beams  cross  the Earth orbit,  are only  a part of a larger source  population called Rotation Powered  Pulsars (RPPs, for short).  The \\gr beams  are believed  to be significantly wider than the radio beams and   they should not necessarily  overlap with  each other (Yadigaroglu \\& Romani 1995). If so,  the RPPs would have improved chances, in principle,  to be detected, in high energy $\\gamma$-rays rather than  at radio wavelengths. This could be the case of some of the  unidentified EGRET sources, a large fraction of which  is believed to be radio-quiet pulsars (see e.g. Grenier 2000).  The  unambiguous and straightforward   proof of this hypothesis would be the discovery of $\\gamma$-ray pulsations  from these objects. In this regard, it is crucial to search for the  pulsed $\\gamma$-ray components  without relying on observations at other  energy bands.  Although GLAST will  be able to perform  effective  searches  for strong unidentified $\\gamma$-ray sources (Thompson 2001),  its  potential in this regard  will be  still limited, at least  for marginally detected weak  sources which  most likely  will appear in the  GLAST source catalog (future ``unidentified   GLAST sources''). An effective realization  of this  important approach which relies only on $\\gamma$-ray astronomical observations requires large photon statistics, and therefore much larger detector areas. The current and forthcoming atmospheric imaging Cherenkov telescopes do provide huge (as large as $10^5 \\, \\rm m^2$) detection areas. But most of these telescopes operate in the energy region $\\geq 100 \\, \\rm GeV$ which  lies, most probably,  well  beyond the cutoffs expected in the pulsar spectra.  In contrast, the recently suggested concept ``5@5''  by Aharonian et al. (2001) --5 GeV energy threshold array of imaging atmospheric Cherenkov telescopes at 5 km altitude--  which combines the most important features of the current  space-based (large fluxes at GeV energies) and ground-based  (large detection areas at TeV energies)  astronomies, could serve as an ideal tool for pulsar studies. Remarkably, the analysis of the EGRET pulsars shows that with  age of pulsars  the fraction of the spin-down luminosity converted into $\\gamma$-rays increases,  and the peak in  the spectral energy distribution $\\nu S_\\nu$ shifts towards 10 GeV (Nel et al. 1996).  If this tendency extends to fainter pulsars, the chances for detection of many pulsars by 5@5 would be increased dramatically.  The potential of the 5@5 concept is not limited to the discovery of  faint  $\\gamma$-ray pulsars. In addition, 5@5 can  provide  detailed spectroscopy in one of the key or, perhaps even the most informative, energy regions of above several GeV for pulsar physics. It is worth noting that  the EGRET data indicate  that some important changes might take place at multi-GeV energies. In particular, the light curves of all EGRET  pulsars  at energies above 5 GeV  seem to be  different  than at lower  energies (Thompson 2001). Detailed temporal  and spectroscopic measurements in this transition region  by 5@5 will hopefully remove many uncertainties we presently face in the physics of pulsar magnetospheres.  Accurate measurements of the energy spectrum of strong  EGRET pulsars in the cutoff region expected between 5 and 20 GeV would allow, in particular,   to distinguish between  two currently popular models  of high energy $\\gamma$-radiation  ---  {\\it polar cap}  (Daugherty \\& Harding 1982, 1996;  Usov \\& Melrose 1995)  and {\\it outer gap} (Cheng et al. 1986;   Romani 1996;  Hirotani \\&  Shibata 1999)  models. These models generally predict different fractions in the population of radio-quiet versus  radio-loud $\\gamma$-ray pulsars, as well as different spectral characteristics, especially at the multi-GeV energy range.  These differences are quite significant, even with  large uncertainties in  model  parameters (Harding 2001). Thus  5@5 should allow distinction among these models, or perhaps even to challenge most scenarios. Remarkably, this intermediate energy region   can be effectively studied by both GLAST and 5@5--- a combination of which would provide high reliability of results. On one hand  GLAST will  extend the studies to low energies, down to $\\sim 20 \\, \\rm MeV$, while 5@5 can perfectly cover the energy region well above 10 GeV where GLAST most probably would run out of photons. Presently several ground-based telescopes such as MAGIC, CELESTE and STACEE  are pushing  their energy thresholds  below 100 GeV (e.g.  Buckley et al.  2001; Krennrich 2001), however  none of these is optimized for energies in the crucial 10 GeV range.  The flux sensitivity,  angular resolution, and the spectral coverage  of 5@5 perfectly match other important aspects of the pulsar physics and  astrophysics. In particular,  the search for inverse Compton  $\\gamma$-radiation from unshocked  ultrarelativistic  winds caused by  illumination of the wind by the IR to X-ray radiation emitted by the pulsar/neutron star itself (Bogovalov \\& Aharonian 2000) or, in the case of binary  pulsar systems, with illumination provided by the companion star  (Ball \\& Kirk 2000), would be an important topic in the program of pulsar studies by 5@5. Because  this radiation is produced  in deep Klein-Nishina regime, it has a rather specific shape -   unusually narrow   spectra peaked at   $E_\\gamma \\sim  \\Gamma_{\\rm w} m_{\\rm e}c^2$, where   $\\Gamma_{\\rm w}$ is the wind Lorentz factor. The detection of this radiation  would be the first {\\it observational} test for  existence of  ultrarelativistic pulsar winds, and would give unique information  about  the Lorentz  factor and the site of  creation of the  kinetic energy dominated (KED) wind.  The large dynamical energy range of 5@5, extending from several GeV to  several  hundred GeV,  would allow such a search  for pulsar winds in  the most relevant range of Lorentz factors  between  $10^4$ to $10^6$.  Another, more traditional,  objective for 5@5 would  be  the detection of inverse Compton $\\gamma$-rays from  synchrotron  nebulae  surrounding pulsars - the regions powered by the wind  termination shocks (plerions). The TeV radiation detected from  the Crab Nebula by many groups (for review  see e.g.  Catanese \\& Weekes  1999) agrees reasonably well with  model calculations  (De Jager \\& Harding 1992; De Jager et al. 1996; Atoyan \\& Aharonian 1996; Aharonian \\& Atoyan 1998; Hillas et al. 1998) performed within the  MHD model of Kennel \\& Coroniti (1984). According to this model,  the wind is terminated by a standing reverse shock at a distance of about r = 0.1 pc, the shock in turn accelerates electrons up to energies exceeding  $10^{15} \\, \\rm eV$ and randomizes their pitch angles.  The inverse Compton $\\gamma$-radiation at GeV/TeV energies,  combined with  synchrotron optical and  X-ray emission, contains  information about the  relativistic  electrons and the nebular magnetic  fields in the  downstream  region of the shock. But the expected $\\gamma$-ray fluxes from  other pulsar-driven nebulae, based on the  {\\it observed}  synchrotron X-ray fluxes  as well as   on the  {\\it model assumptions} concerning  the nebular magnetic fields, contain significant uncertainties. Nevertheless, we believe that the future low-energy IACT arrays should be  able to reveal and allow study of the spectral  and spatial characteristics of  this component of radiation  from the synchrotron nebula surrounding the Vela pulsar,  and hopefully also from some other pulsars  with  ``spin-down  fluxes'' $\\dot{E}/4 \\pi d^2 \\geq 10^{36} \\, \\rm erg/s \\ kpc^2$ (provided that a noticeable part of the spin-down luminosity can be transformed eventually into the shock-accelerated  TeV electrons). ",
        "conclusions": "The high flux sensitivity of 5@5,  supported with unprecedented photon statistics in one of the most informative windows of electromagnetic radiation for the pulsar physics, $E \\geq 5 \\ \\rm GeV$, should allow detailed studies of spectral and temporal characteristics of  the \\gr pulsars detected by EGRET,   as well as  very effective searches for pulsed components from   unidentified EGRET sources.  The energy and  angular resolutions of 5@5 can provide unique studies of other radiation components of pulsars associated with the unshocked winds and the synchrotron nebulae (plerions) in a broad energy band extending from several GeV to 1 TeV. We believe that the new generation sub-10 GeV ground-based \\gr instruments like 5@5 can provide  crucial insight into the physics of pulsars and their interactions with surrounding interstellar medium."
    },
    "0208/astro-ph0208200_arXiv.txt": {
        "abstract": "{ We simulate numerically the surface flow of a gas-supplying companion star in a semi-detached binary system. Calculations are carried out for a region including only the mass-losing star, thus not the mass accreting star. The equation of state is that of an ideal gas characterized by a specific heat ratio $\\gamma$, and the case with $\\gamma=5/3$ is mainly studied.  A system of eddies appears on the surface of the companion star: an eddy in the low pressure region near the L1 point, one around the high pressure at the north pole, and one or two eddies around the low pressure at the opposite side of the L1 point. Gas elements starting near the pole region rotate clockwise around the north pole (here the binary system rotates counter-clockwise as seen from the north pole). Because of viscosity, the gas drifts to the equatorial region, switches to the counter-clockwise eddy near the L1 point and flows through the L1 point to finally form the L1 stream.  The flow field in the L1 region and the structure of the L1 stream are also considered. ",
        "introduction": "A semi-detached binary is a system consisting of a primary star and a Roche-lobe filling secondary star. In cataclysmic variables (CVs), for instance, the primary star is a white dwarf, while the Roche-lobe filling star is a low-mass main-sequence star, from which gas is supplied to the primary star via an accretion disc (see Warner 1995 for a review).  Accretion discs have been the main target of astrophysical research, while gas supplying companion stars has not been investigated fully from both theoretical and observational viewpoints. Knowledge of the properties of the companion star is required in order to understand the evolutionary mechanism of the binary systems. One reason that the companions have not attracted much attention from astrophysicists is that they had little observable constraints. Recently, however, Dhillon and Watson (\\cite{dhi00}) have demonstrated the technique of imaging the companion stars in CVs using Roche tomography.  The companion stars in CVs are similar to low-mass main-sequence stars in their gross properties. However, companion stars exist in a different environment from those of isolated stars. They might be influenced by effects like irradiation from the accretion disc around the primary star, its rapid rotation, Roche-lobe shape, mass loss, and so on. These effects could in turn affect the evolution of the binary systems.  \\subsection{Astrostrophic wind of Lubow and Shu}  Lubow and Shu (\\cite{lub75}, \\cite{lub76}) conducted pioneering work in the study of gas dynamics in semi-detached binaries. In their paper, they analyzed the surface layer of the companion star, as well as the L1 region (Lubow and Shu \\cite{lub75}) and the L1 stream (Lubow and Shu \\cite{lub76}). In their analysis they solved the hydrodynamic equations by a matched expansion technique and reduced partial differential equations to ordinary differential equations. Their method was superior to numerical methods at that time considering the poor capability of computers. Since then, the progress in computer power has been considerable, so that it is worthwhile to investigate the same problem numerically. Our main target in the present paper is to investigate the surface layer, the L1 region and the L1 stream numerically. The region around the accreting compact star has been investigated separately by our group (Makita, Miyawaki and Matsuda \\cite{mak00}, Matsuda et al. \\cite{mat00}, Fujiwara et al. \\cite{fuj01}).  Lubow and Shu (\\cite{lub75}) stated that ``the horizontal component of the fluid velocity is parallel to isobars in the astrostrophic approximation. Like the large-scale circulation patterns in the atmosphere and oceans of the Earth, the flow does not simply proceed from high pressures to low, but is directed around the contours of equal pressure. In the northern hemi-lobe, the flow proceeds by keeping the region of high (low) pressure to the right (left); in the southern hemi-lobe, by keeping high (low) pressure to the left (right).'' On a given equi-potential surface they expected ``the pressure to be largest at the poles (for the same reason that meridional circulation occurs in rotating stars) and smallest at L1 (because of the mass-loss flow)''; therefore ``the excess circulation should be counter to the sense of the orbit rotation.''  They predicted that ``on each equipotential surface, the flow may be astrostrophic, i.e., parallel to isobars, but a net transfer of matter within the entire surface layer to the equator is possible ... if there is an outward flux of matter at the bottom of the surface layer.'' They also considered that ``another effect which reinforces the concept that there must be a slow drift of matter from the poles to the equator, and thence to L1, occurs if the envelope of the contact component is convective. In this case, the effective frictional drag exerted on the upper layers by the lower layers must produce the ``Ekman effect\" (Ekman \\cite{ekm22}) where the induced secondary flow corresponds to a drift across isobars from high pressures to low.''  One of the aims of the present paper is to investigate the flow pattern of the gas in the surface layer of the companion numerically and to compare the results with the qualitative prediction of Lubow and Shu (\\cite{lub75}). This kind of problems may be termed {\\it stellar meteorology}. The structure of the L1 region and the L1 stream are also considered.  \\subsection{Our previous work}  Two-dimensional finite-volume hydrodynamic simulations of flows in a semi-detached binary were done by Sawada, Matsuda and Hachisu (\\cite{saw86}, \\cite{saw87}). Three-dimensional calculation was done by Sawada and Matsuda (\\cite{saw92}). They were concerned mainly by the accretion disc and did not pay much attention to the surface flow on the secondary companion.  Recently, Fujiwara et al. (\\cite{fuj01}) performed three-dimensional hydrodynamic simulations of a semi-detached binary system including the companion star. The equation of state is that of an ideal gas characterized by the specific heat ratio $\\gamma$ and the case with $\\gamma=1.01$, meaning that the gas is almost isothermal, was mainly studied. The reason for the choice of $\\gamma=1.01$ is to mimic a radiative cooling effect occurring in an accretion disc. Although their main concern was the hydrodynamics of the accretion disc, they also obtained the flow pattern on the surface of the companion.  Their results showed that gas flows from the pole region to the equatorial region rather directly, although small eddy patterns appeared on the surface of the companion star. One, however, has to be careful to select the values of the specific heat ratio. In the surface region of the companion, where the gas is expanding outward, the temperature of the gas is reduced due to the adiabatic expansion. Thus the case with $\\gamma=1.01$, which results in the input of energy into the expanding gas, may not be appropriate for the present purpose.  This paper is organized as follows. In ${\\S}$ 2 we describe the numerical method and physical assumptions. In ${\\S}$ 3 we show the results of our numerical simulations. In ${\\S}$ 4 we discuss the dynamics of the obtained eddy system. A summary is given in ${\\S}$ 5. ",
        "conclusions": "In our simulations, we obtained a system of eddies: the H eddy rotating clockwise, and counterclockwise rotating  L1 and L2 eddies. We now discuss what we believe is the reason that this eddy system is formed (see Fig. \\ref{fig10}).  If we consider a time sequence, the L1 low pressure is formed first, because of the mass loss through the L1 point. Since the gas near the equator does not feel much resistance in flowing towards the L1 point, the equatorial region becomes a low pressure zone, and a high pressure zone builds up at the north pole: the H eddy is thus formed.  The above explanation does not explain the appearance of the third eddy: the L2 eddy. This can be explained as follows. Consider a force exerted by an eddy on another eddy.  The L1 eddy tries to convect the H eddy to negative $y$-direction, while the H eddy tries to convect the L1 eddy to negative $y$-direction as well. However, the L1 eddy must be fixed to the L1 region, so that the H eddy must drift to negative $y$-direction. The system is therefore not steady. In order to stabilize the eddy system, we need the third eddy, which tries to convect the H eddy to positive $y$-direction. The L2 eddy thus appears and evolves until the three eddy system is stabilized. The exact location and the strength of these eddies can be obtained only through numerical simulations.  It is interesting to note that Davey and Smith (\\cite{dav92}) have found that the trailing hemisphere has stronger line absorption than the leading hemisphere in a dwarf nova (see also Dhillon and Watson \\cite{dhi00}). This asymmetry can be naturally explained in terms of the H-eddy current, which convects heat due to irradiation to the leading hemisphere.  If this kind of astrostrophic wind conveys heat effectively from an irradiated hemisphere to the other, the unirradiated side of the companion becomes hotter than previously thought. Such an example may be observed in the 2000 outburst of the recurrent nova CI Aql because, in nova outbursts, a companion star should be strongly irradiated by a hot white dwarf. The inclination angle of the CI Aql orbit is about 80 degrees and the eclipse is almost total in the sense that the companion completely occults the bright accretion disc around the white dwarf (e.g., Hachisu \\& Kato 2001).  In eclipse minima, therefore, we see only the unirradiated side of the companion. Hachisu, Kato and Schaefer (2002) showed that the eclipse minima of CI Aql in outburst is about 0.6 mag brighter than in quiescence and could reasonably reproduce the orbital light curve (folded by the orbital period) when the temperature of the unirradiated hemisphere is about 1000 K hotter than that in quiescence.  In the present study, we do not take into account such complex effects as (real) viscosity, magnetic fields, and irradiation from the accretion disc. Activities of magnetic fields on the companion surface would lead to star-spots, and the irradiation from the accretion disc would lead to an increase of the gas temperature of the companion. These effects might more or less change the surface flow patterns. The present study is a first step to examine the surface flow structure. Thus to study these phenomena is beyond the scope of the present paper, and is left for future work."
    },
    "0208/astro-ph0208085_arXiv.txt": {
        "abstract": "The Local Group Census is a narrow- and broad-band survey of all the galaxies of the Local Group above $\\delta =-30^\\circ$, in progress at the 2.5m Isaac Newton telescope on La Palma.  We discuss here the ability of the survey to detect symbiotic star candidates in the Local Group, by deriving detection limits in each of the narrow- and broad-band frames used in the survey, and by estimating the total number of objects expected in each galaxy. We present two diagnostic diagrams, based on the adopted photometric filters, to discriminate between symbiotic stars and other emission-line objects such as planetary nebulae. ",
        "introduction": "The Local Group Census (LGC) is a narrow- and broad-band survey of all the galaxies of the Local Group (LG) above $\\delta =-30^\\circ$ (http://www.ing.iac.es/ WFS/LGC/).  It is carried out as part of the Isaac Newton Group's Wide Field Survey program and its observations are being obtained with the Wide Field Camera at the 2.5m Isaac Newton telescope, equipped with a 4-CCD mosaic covering a field of view of $34^\\prime \\times 34^\\prime$.\\\\  The LG galaxies have been observed through narrow-band filters (\\oiii\\ $\\lambda$5007\\AA, \\ha, \\heii\\ $\\lambda$4686\\AA, and \\sii\\ $\\lambda$6725\\AA) and broad band filters (Sloan \\gfilter, \\rfilter, \\ifilter, and \\stry).  One of the goals of the LGC is the search for symbiotic star candidates in galactic systems other than the Milky Way, using the broad-band color index \\gfilter --\\ifilter\\ to detect the red giants and the narrow-band \\ha\\ filter to select those with a strong emission from circumbinary gas. ",
        "conclusions": ""
    },
    "0208/astro-ph0208420_arXiv.txt": {
        "abstract": "The gravitational lens CLASS B1608+656 is the only four-image lens system for which all three independent time delays have been measured. This makes the system an excellent candidate for a high-quality determination of $H_0$ at cosmological distances.  However, the original measurements of the time delays had large (12--20\\%) uncertainties, due to the low level of variability of the background source during the monitoring campaign.  In this paper, we present results from two additional VLA monitoring campaigns.  In contrast to the $\\sim$5\\% variations seen during the first season of monitoring, the source flux density changed by 25--30\\% in each of the subsequent two seasons.  We analyzed the combined data set from all three seasons of monitoring to improve significantly the precision of the time delay measurements; the delays are consistent with those found in the original measurements, but the uncertainties have decreased by factors of two to three.  We combined the delays with revised isothermal mass models to derive a measurement of $H_0$.  Depending on the positions of the galaxy centroids, which vary by up to 0\\farcs1 in HST images obtained with different filters, we obtain $H_0 =$ 61--65 \\ksm, for $(\\Omega_M,\\Omega_\\Lambda) = (0.3,0.7)$.  The value of $H_0$ decreases by 6\\% if $(\\Omega_M,\\Omega_\\Lambda) = (1.0,0.0)$.  The formal uncertainties on $H_0$ due to the time delay measurements are $\\pm 1$ ($\\pm 2$) \\ksm\\ for the 1$\\sigma$ (2$\\sigma$) confidence limits. Thus, the systematic uncertainties due to the lens model, which are on the order of $\\pm$15 \\ksm, now dominate the error budget for this system.  In order to improve the measurement of $H_0$ with this lens, new models that incorporate the constraints provided by stellar dynamics and the optical/infrared Einstein ring seen in HST images must be developed. ",
        "introduction": "Gravitational lenses provide excellent tools for the study of cosmology.  In particular, the method developed by \\citet{refsdal} can be used to determine the Hubble Constant at cosmological distances. This method requires a lens system for which both lens and source redshifts have been measured, for which a well-constrained model of the gravitational potential of the lensing mass distribution has been determined, and for which the time delays between the lensed images have been measured.  To date, measurements of time delays have been reported in the literature for 11 lens systems: 0957+561        \\citep{tk0957_1,tk0957_2}, PG~1115+080     \\citep{pls1115,barkana1115}, JVAS~B0218+357  \\citep{adb0218,cohen0218}, PKS~1830-211    \\citep{lovell1830,tw1830}, CLASS~B1608+656 \\citep[][hereafter Paper I]{paper1}, CLASS~B1600+434 \\citep{lvek1600,burud1600}, JVAS~B1422+231  \\citep{1422delay}, HE~1104-1805    \\citep[see, however, Pelt, Refsdal, \\& Stabell 2002]{gilmerino1104}, HE~2149-2745    \\citep{burud2149}, RX~J0911.4+0551 \\citep{hjorth0911}, and SBS~1520+530    \\citep{burud1520}. Of these lens systems, CLASS B1608+656 is the only four-image system for which all three independent time delays have been unambiguously measured.  The CLASS B1608+656 lens system consists of the core of a radio-loud poststarburst galaxy at a redshift of $z = 1.394$ \\citep{zs1608} being lensed by a $z = 0.630$ pair of galaxies \\citep{sm1608}.  Radio maps of the system show four images of the background source arranged in a typical lens geometry (Figure~\\ref{fig_1608map}). Models of the lens system predict that, if the background source is variable, image B should show the variation first, followed by components A, C, and D in turn \\citep[hereafter Paper~II]{sm1608,paper2}.  The time delays determined in Paper I were based on radio-wavelength light curves obtained with the Very Large Array (VLA\\footnote{The National Radio Astronomy Observatory is a facility of the National Science Foundation operated under cooperative agreement by Associated Universities, Inc.})  between 1996 October and 1997 May. During that time, the background source 8.5~GHz flux density varied by $\\sim$5\\%.  Although the measured time delays were robust, the small level of variability meant that the uncertainties on the delays were large.  The estimated uncertainties from the Paper I data ranged from $\\sim$12\\% for the B$\\rightarrow$D time delay ($\\tau_{BD}$) to $\\sim$20\\% for the other two measured delays (95\\% CL).  These uncertainties translated to uncertainties on the derived value of the Hubble Constant of $\\sim$15\\% (Paper~I).  In addition, the lens modeling contributes an uncertainty of $\\sim$30\\% to the derived $H_0$, with the largest contribution coming from the uncertainty in the radial profile of the mass distribution in the lensing galaxies (Paper~II).  In order to achieve a more precise determination of $H_0$ from this lens system, the uncertainties on both the time delays and the modeling need to be significantly reduced.  This paper addresses the first of those issues.  \\begin{figure} \\plotone{f1.eps} \\caption{Typical radio map of B1608+656 from the season 3 monitoring campaign.  These data were obtained on 1999 August 08.  Both the grayscale and the contours represent radio brightness.  The contours are set at $-$2.5, 2.5, 5, 10, 20, 40, 80, and 160 times the RMS noise level of 0.13 mJy~beam$^{-1}$. \\label{fig_1608map}} \\end{figure} ",
        "conclusions": "\\subsection{Evidence for Microlensing and Lens Substructure\\label{microlens}}  As discussed in \\S\\ref{sec_comb_delay}, the component relative magnifications were not constant from seasons 1 to 3.  In particular, $\\mu_{AB}$ and $\\mu_{DB}$ decreased by 2\\% or more during the course of the observations (Figure~\\ref{fig_mulens}).  Although not large, these changes exceed the 95\\% confidence limits on the relative magnifications derived from the Monte Carlo simulations (Table~\\ref{tab_mcresults}), and thus appear to be real.  Similar or more extreme changes in flux ratios are seen in optical monitoring of lens systems \\citep[e.g.,][]{2237micro,burud1600}.  These changes are attributed to microlensing events, where stars or other massive compact objects in the lensing galaxy change the magnifications of the lensed images as they move through the galaxy.  For many years, it was thought that radio observations of gravitational lens systems should be unaffected by microlensing because the angular sizes of compact radio cores, typically on the order of a milliarcsecond, are much larger than the microarcsecond lensing cross sections of stars. However, radio monitoring of the lens CLASS B1600+434 has revealed changes in the component flux ratio that has been attributed to microlensing \\citep{1600micro1}.  In the case of B1600+434 the unexpected microlensing was interpreted as being due to the superluminal motion of a microarcsecond-sized component in the jet of the background quasar across the complex caustic structure produced by compact objects in the lens galaxy halo \\citep{1600micro1}.  The lensed source in the B1608+656 system is the core of a classical radio double source \\citep{snellen1608}.  Thus, it is certainly possible that there are extremely compact jet components associated with B1608+656 as well.  On the other hand, the changes in the flux density ratios may have other explanations, such as scintillation.  However, because we have monitored the system at only one frequency, we do not have the clear discriminant between microlensing and scintillation that multifrequency monitoring provides \\citep{1600micro1}.  With properly designed future observations, it may be possible to determine the cause of the changing flux density ratios in this system.  \\begin{figure} \\figurenum{8} \\plotone{f8.eps} \\caption{Relative magnifications with respect to component B as a function of the season of observations.  For each point, the inner error bars are the 68\\% confidence regions from the Monte Carlo simulations, while the outer error bars are the 95\\% confidence regions. \\label{fig_mulens}} \\end{figure}  We note that, whatever the cause of the variability in the relative magnifications, it may also be necessary to invoke the presence of substructure in the B1608+656 lensing galaxies to explain the observed flux densities of the images.  Simple, or even fairly complex, lens models of the B1608+656 system have not been able to properly reproduce all of the observed flux density ratios \\citep[Paper II;][]{sm1608}.  As we noted in Paper II, this discrepant magnification could be produced by perturbations to the smooth mass distributions assumed for the lensing galaxies.  The presence of such substructure has been invoked to explain similarly discrepant flux ratios in other four-image lenses, although in those cases the incongruous ratios were those between bright merging images.  In some cases, the substructure has been interpreted as globular clusters or plane density waves \\citep{substruct}.  Recently, several papers have suggested that the substructure is due to the small dark matter satellite halos expected from CDM structure-formation models \\citep[e.g.,][]{cdmstruct1,cdmstruct2,cdmstruct3}.  \\subsection{Determination of $H_0$ with B1608+656}  To determine $H_0$ from a lens system, the measured time delays are combined with those predicted by the model.  The predicted delays depend on the values of the fundamental cosmological parameters, so one must assume a cosmological model.  Specifically, the model delays are proportional to $(D_\\ell D_s) / D_{\\ell s}$ where $D_\\ell$, $D_s$, and $D_{\\ell s}$ are the angular diameter distances from the observer to the lens, from the observer to the source, and from the lens to the source, respectively.  The angular diameter distances are functions of $H_0$, $\\Omega_m$, and $\\Omega_\\Lambda$, as well as the redshifts of the lens and the background source.  We assume $(\\Omega_M, \\Omega_\\Lambda)$ = (0.3,0.7) for the rest of this paper.  In comparison to this assumed cosmology, the value of $H_0$ derived from this lens system will change by no more than $\\sim$6\\% in both flat and open cosmological models with $0.1 < \\Omega_M < 1.0$.  As specific examples, the values of $H_0$ discussed below should be scaled by a factor of 1.02 if $(\\Omega_M,\\Omega_\\Lambda) = (0.3,0.0)$ and a factor of 0.94 if $(\\Omega_M,\\Omega_\\Lambda) = (1.0,0.0)$.  The considerable improvement in the accuracy of the measured time delays between the three image pairs in B1608+656 warrants an update of the lens model of this system, first presented in Paper~II. Multiple time delays in a single lens system add additional constraints on the lens model, and an improvement in their accuracy could therefore lead to a change of the lens model and a change of the inferred Hubble Constant.  Our models are based on the {\\em lensmodel} package developed by \\citet{lensmodel}, but the results are consistent with those based on the code used in Paper~II.  We emphasize that the models presented here are only an update of the isothermal lens mass models in Paper~II and that a much more detailed mass model will be presented in a forthcoming publication, where we fully exploit additional data on the system (see \\S\\ref{summary}).  For the lens modeling in this paper, we use the same constraints as in Paper~II. We use the VLBI image positions with formal 1$\\sigma$ errors listed in Paper~II. Because the flux density ratios appear to slowly change in time (\\S\\ref{sec_comb_delay}), even after the time-delay correction of the light curves, we widen the error bars to 20\\% on the flux density ratios to err on the side of caution. We use the improved time delays presented in this paper with 1.5~d errors (68\\% CL). We use singular isothermal ellipsoid (SIE) mass models for both lens galaxies (G1 and G2; see Paper~II for details) and center them on the galaxy centroids measured from {\\em Hubble Space Telescope} (HST) images obtained with the F555W, F814W, and F160W filters. The three centroids differ by amounts that are significant, which could be due to the presence of dust extinction and/or PSF problems.  In addition, it is possible that the two merging lens galaxies are contained within a common dark matter halo and, thus, that the luminous material may not provide an accurate representation of the shape and centroid(s) of the mass distribution.  These issues will be examined in more detail in the forthcoming publication. Here, we present models for the centroids from the F814W and F160W images; we reject the model based on the F555W centroids because it produces an extra image of the background source that is not seen in the data.  As already shown in Paper~II, the differences in the centroids do not strongly affect the inferred value of $H_0$, although the cause of the wavelength-dependent shifts in position is something that requires further study.  The mass distributions of G1 and G2 are allowed to have a free mass scale (i.e., velocity dispersion), position angle, and ellipticity. At this point no external shear is allowed.  Both lens galaxies are assumed to be at the same redshift. The results of the updated lens models and inferred values of $H_0$ are listed in Table~\\ref{tab_modparams}.  We note that these models are equivalent to models I and II in Table~3 of Paper~II, in spite of using completely independent modeling codes in the two papers.  We also note that, in order to compare the models produced by the two codes, it is important to understand how the mass scales in the models depend on the projected axial ratio of the lensing galaxies, $q$.  We find that the relation between the quantities representing the lens strengths, $\\sigma$ in Paper II and $b^\\prime$ in Table~\\ref{tab_modparams}, is given by $b^\\prime \\sqrt{1+q^2} \\propto \\sqrt{q}\\,\\sigma^2$. Using this relation, we find that the mass scales presented in Paper II and in Table~6 are also equivalent and that the mass ratio between galaxies G1 and G2 is $(M_1/M_2) \\sim (b^\\prime_1/b^\\prime_2)^2 \\sim3$. This large mass ratio avoids the creation of additional images between galaxies G1 and G2.  We have also modeled the system with an allowance for an external shear ($\\gamma_{\\rm ext}$).  The addition of shear to the models in Table~\\ref{tab_modparams} leads to a considerable decrease in the value of $\\chi^2$, for $\\gamma_{\\rm ext}\\sim0.1$.  However, the number of free parameters in the models increases to 11 (including $H_0$), and the models are less well constrained than the models with no shear.  Additionally, although there are several small galaxies in the field around B1608+656, there is no evidence for a more massive group or cluster that could yield a 10\\% external shear, as for example in the case of PG1115+080 \\citep{kundic1115,tonry11151422}. Thus, we think that the true external shear (i.e., a constant shear not due to the lens galaxies themselves) is unlikely to be as high as 10\\% and that a revision of the galaxy mass models is more likely required.  The formal statistical errors on $H_0$ in the models are $\\pm$1~\\ksm\\ and $\\pm$2~\\ksm\\ for the 1$\\sigma$ and 2$\\sigma$ confidence limits, respectively. In Paper~II we estimated that the range of reasonable variations in the shapes of the mass profiles of the lens galaxies contributed an additional uncertainty of 30\\% to the determination of $H_0$ from this system.  This estimate may be overly conservative, given the small scatter in mass profile slopes seen in other lens galaxies (\\S\\ref{sec_otherlens}).  However, given the complex nature of the lens system and the uncertain positions of the lensing galaxies, we will use this estimate of the systematic modeling uncertainties in this paper.  In \\S\\ref{summary}, we discuss methods by which the modeling uncertainties for this system may be reduced.  We note that an alternative and independent approach to lens modeling, namely the non-parametric method developed by \\citet{nonparam}, produces estimates of $H_0$ and its systematic uncertainties that are similar to those determined in this paper.  In comparison to the more frequently used analytic modeling approaches, the non-parametric method has the advantage of making fewer assumptions about the form of the mass distribution in the lensing galaxy or galaxies.  Thus, it is possible to explore a larger region of the model space than is usually done with single analytic models and, as a consequence, to obtain a perhaps more realistic estimate of the uncertainties due to modeling. On the other hand, given the additional freedom in model parameters, the non-parametric approach may produce some descriptions of the lensing galaxy that do not have a physical meaning, in spite of the constraints introduced to produce realistic galaxy models. Therefore, a straightforward interpretation of the results may be difficult. \\citet{nonparam} applied their approach to the B1608+656 system, using the image positions and time delays as constraints.  The time delays used by \\citet{nonparam} were from an early analysis of the Season 1 data and differ slightly from those found in this paper. However, these small differences should not significantly affect the resulting values of $H_0$.  \\citet{nonparam} found that $\\sim$90\\% of their model reconstructions produced values of $H_0$ between 50 and 100~\\ksm.  They also applied their non-parametric method to PG1115+080 and the results were combined with the B1608+656 results to produce $H_0 = 61 \\pm 11\\ (\\pm 18)$ \\ksm at 68\\% (90\\%) confidence, for $(\\Omega_M,\\Omega_\\Lambda) = (1.0,0.0)$.  For the cosmological model adopted in this paper, their value of $H_0$ would change to 65 \\ksm, consistent with our results.  \\begin{center} \\begin{deluxetable}{cccccccccr} \\tablenum{6} \\tablewidth{0pt} \\scriptsize \\tablecaption{Mass Model Parameters \\label{tab_modparams}} \\tablehead{  & \\colhead{$x_c$\\tablenotemark{a}} & \\colhead{$y_c$\\tablenotemark{a}} & \\colhead{$b^\\prime$} & & \\colhead{P.A.\\tablenotemark{b}} & \\colhead{$x_s$\\tablenotemark{a}} & \\colhead{$y_s$\\tablenotemark{a}} & \\colhead{$H_0$} &  \\\\ \\colhead{Filter} & \\colhead{(arcsec)} & \\colhead{(arcsec)} & \\colhead{(arcsec)} & \\colhead{$q$} & \\colhead{(deg)} & \\colhead{(arcsec)} & \\colhead{(arcsec)} & \\colhead{(\\ksm)} & \\colhead{$\\chi^2$} } \\startdata F814W & $+$0.521 & $-$1.062 & 0.68 & 0.88 & $+$71.8 & $+$0.060 & $-$1.090 & 65 & 43  \\\\ & $-$0.293 & $-$0.965 & 0.39 & 0.39 & $+$53.1 &          &          &    &     \\\\ F160W & $+$0.446 & $-$1.063 & 0.69 & 0.91 & $-$58.0 & $+$0.058 & $-$1.103 & 61 & 196 \\\\ & $-$0.276 & $-$0.937 & 0.36 & 0.31 & $+$53.8 &          &          &    &     \\\\ \\enddata  \\tablenotetext{a}{Positions are given in Cartesian rather than astronomical coordinates.} \\tablenotetext{b}{Position angles are defined in the astronomical convention, in degrees measured east of north.}  \\tablecomments{Lens models for B1608+656 consisting of two SIE mass distributions with no external shear. Columns 2--3 indicate the galaxy centroids measured in the HST images obtained with the listed filter; columns 4--6 indicate the lens strength (see alpha models in Keeton 2001), axial ratio, and position angle of the SIE mass distribution; columns 7--8 indicate the source position; columns 9--10 indicate the inferred value of $H_0$ and the model $\\chi^2$ value. The values of $H_0$ are quoted for $(\\Omega_M,\\Omega_\\Lambda) = (0.3,0.7)$. Every first/second line indicates the parameters for galaxy G1/G1.} \\end{deluxetable} \\end{center}  \\subsection{Comparison to Determinations of $H_0$ with Other Lens Systems \\label{sec_otherlens} }  Time delays have been measured in 11 lens systems, with the B1608+656 delays being among those with the smallest uncertainties. Unfortunately, the determination of $H_0$ from lenses is not as clean as one might have hoped.  For nearly every lens with a measured time delay, more than one model has been used to fit the data.  Often the resulting values of $H_0$ are quoted with small formal errors that, when taken at face value, make different determinations of $H_0$ with the same lens system mutually exclusive at high levels of significance.  These discrepant values of $H_0$ arise because observations of most lens systems do not tightly constrain the mass distribution of the lens, and there is a degeneracy between the steepness of the lens mass profile and the derived value of $H_0$ \\citep[e.g., Paper~II;][]{nonparam,witt_delays}.  Although it may not be possible to determine the radial mass profile in many individual lens systems, there are indications that lensing galaxies may follow a nearly universal mass profile.  It is, for example, possible to assume that all lenses must give consistent determinations of $H_0$ and then use the time-delay measurements to obtain information about the mass distribution in the lenses. \\citet{cskdelays} has performed this experiment on a sample of five lens systems with measured delays.  He finds that, in order to produce the same value of $H_0$, the mass surface density properties of the lensing galaxies must differ by only very small amounts. Furthermore, several independent lines of investigation indicate that not only are the mass distributions in many lenses similar, but that the mass profiles are close to isothermal.  These approaches have used models that incorporate more data than the standard two or four image positions and fluxes.  The additional inputs make it possible to place tight limits on the slope of the mass distribution.  The additional constraints may come from full or partial Einstein rings \\citep[e.g.,][]{csk1654,keeton0957,cskrings}, complex structure in the background source that in lensed into a ring-like configuration \\citep{cohn1933}, stellar dynamics \\citep{tk2016,kt0047}, or the orientations of mas-scale jets seen in very high angular resolution radio maps \\citep{rusin1152}.  Although it should not be concluded that all lens galaxies have nearly isothermal mass distributions, it may be that the range of mass profile slopes present in lensing galaxies is significantly smaller than what is allowed by the constraints in most individual lens systems with time delay measurements.  If further observations strengthen these conclusions, then the contribution of a major source of systematic error in lens-derived measurements of $H_0$ may be significantly decreased.  The mass profile of the main lensing galaxy is not the only source of uncertainty in the determination of $H_0$ from a lens system.  Another factor enters if there is a cluster or massive group associated with the primary lensing galaxy.  The inclusion of the effects of the cluster can vastly complicate the lens model.  The case of the first lens to be discovered, Q0957+561, is instructive.  Despite many years of intensive observational and modeling efforts, no definitive model has been developed.  For an excellent discussion of the wealth of historical models of this system, see \\citet{keeton0957}.  Clusters have also been discovered in association with RX~J0911.4+0551 \\citep{kneib0911,morgan0911} and possibly SBS~1520+530 \\citep{faure1520}.  In addition, compact groups of galaxies have been discovered in association with several lens systems \\citep[e.g.,][]{kundic1115,kundic1422,tonry11151422,tonry0751,tonry1131,fl0712,rusin1359}, although their effects on the lensing properties of the systems are much smaller than those of clusters.  Still another problem in the determination of $H_0$ from gravitational lens systems arises if the location of the lensing galaxy is not known, as is the case in B0218+357 \\citep{lehar_2lenses}.  In spite of the possible problems mentioned above, the gravitational lens method is attractive because many of the systematic errors affecting the determination of $H_0$ with one lens system are different from those affecting other systems.  Thus, it should be possible to obtain a global measurement of $H_0$ with small uncertainties by averaging the measurements from many lens systems. This is in contrast to the distance-ladder techniques in which many of the systematic uncertainties affect all distance determinations in the same sense.  The major problem with the gravitational-lens method is that the number of lens systems with measured time delays is still small.  Even though the number of systems for which delays have been measured has increased substantially over the last few years, some of the delays are not measured to high precision.  In addition, lens models for some systems are viewed with suspicion due to problems such as those mentioned in the previous paragraph.  Thus, efforts to determine a global value of $H_0$ often exclude a significant fraction of systems with time-delay measurements.  Recent attempts have been limited to samples consisting of five or fewer lenses, and have obtained global values of $H_0$ ranging from $\\sim$50 to $\\sim$75~\\ksm\\ \\citep[e.g., Paper~II;][]{schechreview,cskH0}.  With such small sample sizes, the mean $H_0$ obtained can easily be biased by unknown factors affecting one or two of the lens systems.  It is thus crucial to measure time delays in more lens systems in order to obtain a robust global determination of $H_0$ from lenses.  \\subsection{Comparison to Determinations of $H_0$ with Other Methods}  There are, of course, methods for measuring $H_0$ that are completely independent of the lens-derived values.  These include the traditional distance-ladder methods in which Cepheid-based distances are used to calibrate secondary distance indicators.  The HST Key Project combined several secondary distance measurement methods to obtain a final value of $H_0 = 72 \\pm 3 (1 \\sigma) \\pm 7$~(sys) \\ksm\\ \\citep{keyproject}. Using an analysis of Cepheid-calibrated distances to Type Ia supernovae, \\citet{parodi_h0} derived $H_0 = 59 \\pm 6$\\ksm\\ ($2 \\sigma$ random errors combined with estimated systematic effects). The use of the Sunyaev-Zeldovich (SZ) effect is another approach that, like the gravitational lens method, is independent of the distance-ladder approach.  In recent work by \\citet{szh0}, the average measurement obtained from a flux-limited sample of clusters was $H_0 = 66^{+14}_{-11} (1 \\sigma) \\pm 14$~(sys) \\ksm.  The global values of $H_0$ determined from lenses, the SZ effect, and traditional distance-ladder methods are broadly consistent with each other and with the $H_0$ determined from B1608+656.  The sources of systematic error in each of these methods are different.  Therefore, if the methods all produced values of $H_0$ that were in good, rather than broad, agreement, the confidence that the correct value had been measured would be increased.  It is important that steps be taken to understand the systematic errors affecting each method and to attempt to reduce those errors.  The systematic errors associated with the gravitational lens method may be reduced if time delays can be measured and mass profiles can be determined in significantly larger samples of lens systems."
    },
    "0208/astro-ph0208029_arXiv.txt": {
        "abstract": "We present results of a new investigation (Carraro et al. 2002) aimed at clarifying the mutual relationship between the most prominent young open clusters close to $\\eta$ Carin\\ae~, namely Trumpler~16, Trumpler~14 and Collinder~232. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0208/astro-ph0208445_arXiv.txt": {
        "abstract": " ",
        "introduction": "On 22~September~2001~UT, NASA's Deep Space~1 (DS1) spacecraft made a close flyby of the nucleus of comet 19P/Borrelly, obtaining high-resolution images, infrared spectra and particles and fields measurements within about 12 hours of closest approach (Soderblom {\\it et al.}~2002).  The images that were obtained offer an unprecedented look at the nucleus of this comet and promise to reveal many details about the innermost region of the coma as well as the topology and albedo of the nucleus.  However, due to the rapid velocity and short duration of the encounter, additional information is needed to provide a more global interpretation of the spacecraft measurements and how they relate to observations of the entire coma.  In support of this mission, we utilized the 2.1-m and 2.7-m telescopes of McDonald Observatory to observe comet Borrelly around the time of the DS1 encounter and in subsequent months.  We obtained both moderate-resolution spectroscopy and broad band imaging in R and V filters.  In this paper we describe our observations and discuss how they were used to analyze the morphology of the coma, probe the asymmetric distributions of the gas and dust, and derive the reflectivity of the dust.  We then highlight some unique inherent physical characteristics of the comet and discuss how they were used to determine the orientation of the spin axis.  Finally, we present an analysis in which we used our pole determination and the comet's nongravitational forces to constrain the mass and density of the nucleus.  We also compare our results to those found from earlier apparitions (Cochran and Barker 1999) and to other observers' results from this apparition, including the DS1 measurements. ",
        "conclusions": "  \\begin{itemize} \\item We utilized the nongravitational accelerations of comet Borrelly to compute a mass of the nucleus of $3.3\\times10^{16}$~g and a density of 0.49~g\\,cm$^{-3}$ (with a range of $0.29 < \\rho < 0.83$~g\\,cm$^{-3}$).  Because the direction of the reaction force and the water production rates are both well-known (and highly repeatable from one apparition to the next), and because the dimensions of the nucleus were measured in situ, this is the least model-dependent comet density known to date.  \\item The strong jet seen in the DS1 images that emanates from the waist of the comet is aligned with the comet's rotation axis.  We determined the orientation of the pole to be $\\alpha=214^\\circ,~\\delta = -5^\\circ$, with an uncertainty of 4$^\\circ$, which is consistent with other solutions, including the DS1 estimate.  Given this orientation, the jet was pointed about 40$^\\circ$ from the Sun at the time of the DS1 encounter.  There is also evidence that the pole orientation changed by 5-10$^\\circ$ between the 1911 and 1994 apparitions.  \\item The position of the pole results in a strong seasonal effect in the activity levels of the jets.  As the comet receded from the Sun, the primary jet at the pole received less and less illumination. Eventually the primary jet turned off and a secondary, much weaker jet, turned on.  The secondary jet is located on the opposite hemisphere from the primary jet and probably lies within 30-40$^\\circ$ of the pole.  \\item The distribution of the gas and dust in the coma is quite asymmetric in the sunward/anti-sunward directions.  However, perpendicular to this direction, the gas seems to be quite symmetrically distributed.  The distribution of C$_{2}$ gas in the sunward direction in November 2001 is quite uniform with cometocentric distance out to 50,000\\,km.  Such a distribution cannot be easily reproduced with simple two-component models.  \\item A comparison of the C$_{2}$ and CN gas distributions in the coma on September 2001 and November 1994 shows a remarkable similarity. Except for the geocentric distance, the viewing geometries from these dates were nearly identical.  The comet shows the same asymmetries in both apparitions and the gas column densities are the same.  This points to a very stable gas production and is another piece of evidence that the comet must be in simple rotation.  \\item The comet is mildly depleted in C$_{2}$ and C$_{3}$ relative to CN.  \\item The dust in the coma is very red, with the tailward region being much redder than the sunward jet. This suggests that the particles in the primary jet are, on average, smaller than those in the rest of the coma and tail.  However, residual particles from the primary jet are still seen in February, which indicates that the particle size distribution, even in the primary jet, contains many large grains. The jet appears to exhibit a steady-state outflow while the tail and perpendicular regions show evidence for radiation pressure acting on the dust.  \\end{itemize} \\vspace{0.5in} \\begin{center}Acknowledgements\\end{center} This research was supported by NASA Grants NAG5-9003 and NAG5-4384. We thank Drs. David Schleicher, Laura Woodney and Nalin Samarasinha for helpful discussions, Dr. Laurence Soderblom for communicating the Deep Space 1 pole solution to us prior to publication, and Dr. Beatrice Mueller for her comments on the manuscript.  \\newpage"
    },
    "0208/astro-ph0208390_arXiv.txt": {
        "abstract": "A new analysis of the connection between black-hole mass and radio luminosity in radio-selected flat-spectrum quasars (FSQ) is presented. In contrast to recent claims in the literature, we find no evidence that the black-hole masses of radio-selected FSQ are systematically lower that those of luminous optically-selected radio-loud quasars. The black-hole masses of the FSQ are estimated via the virial black-hole mass estimator which utilizes the line-width of the H$\\beta$ emission line as a tracer of the central gravitational potential. By correcting for the inevitable effects of inclination, incurred due to the FSQ being viewed close to the line of sight, we find that the black-hole masses of the FSQ with intrinsically powerful radio jets are confined, virtually exclusively, to $M_{bh} >10^{8}$~M$_{\\odot}$. This is in good agreement with previous studies of optically selected FSQ and steep-spectrum radio-loud quasars.  Finally, following the application of a realistic Doppler boosting correction, we find that the FSQ occupy a wide range in intrinsic radio luminosity, and that many sources would be more accurately classified as radio-intermediate or radio-quiet quasars. This range in radio luminosity suggests that the FSQ are fully consistent with an upper boundary on radio power of the form $L_{\\rm 5GHz} \\propto M_{bh}^{2.5}$. ",
        "introduction": "It is now widely believed that the energy emitted by active galactic nuclei (AGN) is a consequence of mass accretion onto a central supermassive black hole. It is possible that the mass of the central black hole may be a crucial fundamental parameter in understanding the physics of AGN.  In particular, one question which has recently received a great deal of attention in the literature is whether or not the mass of an AGN's black hole is strongly related to it's radio luminosity. This question is of importance, because if it is established that radio-loud and radio-quiet quasars have different black-hole mass distributions, it may help explain why quasars of comparable optical luminosities can differ in their radio luminosity by many orders of magnitude. On the contrary, if radio-loud and radio-quiet quasars are found to have essentially identical black-hole mass distributions, then the search for the origin of radio loudness must move to some other physical parameter such as black-hole spin.  Recent studies of the black-hole mass -- radio luminosity relation ($M_{bh}-L_{\\rm rad}$) relation have produced apparently contradictory results. Franceschini, Vercellone \\& Fabian (1998) found that high frequency (5~GHz) radio luminosity correlated strongly with the mass of the central black hole in a sample of nearby inactive galaxies from the work of Kormendy \\& Richstone (1995). This surprisingly tight correlation had the form $L_{\\rm 5GHz} \\propto M_{bh}^{2.5}$, which they proposed was indicative of advection dominated accretion being the primary mechanism in controlling the radio output of objects with a low accretion rate.  Using the spectral data of Boroson \\& Green (1992), Laor (2000) investigated the relation between black-hole mass and radio luminosity in the Palomar-Green quasar sample using the virial black-hole mass estimator. The results from this analysis pointed to an apparent bi-modality in black-hole mass, with virtually all of the radio-loud quasars containing black holes with masses $M_{bh} > 10^{9}~$M$_{\\odot}$, whereas the majority of quasars with black hole masses $M_{bh} < 3 \\times 10^{8}~$M$_{\\odot}$ were radio quiet.  A similar result was arrived at by McLure \\& Dunlop (2002) using a sample of radio-loud and radio-quiet quasars matched in terms of both redshift and optical luminosity. The results of this study indicated that the median black-hole mass of the radio-loud quasars was a factor of $\\sim2$ larger than that of their radio-quiet counterparts. However, the substantial over-lap between the black-hole mass distributions of the two quasar samples indicated in addition that black-hole mass could not be the sole parameter controlling radio power.  The FIRST Bright Quasar Survey (FBQS; Gregg et al. 1996; White et al. 2000) has also been used to probe the relationship between radio luminosity and black-hole mass. The radio-loud quasars in the FBQS fill the gap between radio-quiet and radio-loud quasars and are thus ideal probes of medium power radio sources. From their study of the FBQS, Lacy et al. (2001) found a continuous variation of radio luminosity with black hole mass, in addition to evidence supporting the view that radio power also depends on the accretion rate relative to the Eddington limit.  However, two studies have recently questioned whether there is any real connection between black-hole mass and radio power. Using a compilation of objects which ranged from nearby quiescent galaxies to low-redshift quasars, Ho (2002) re-examined the relationship between black-hole mass and radio luminosity. In contrast to the studies outlined above, the results of this analysis suggested that there was no clear relationship between radio power and black-hole mass, leading the author to conclude that radio-loud AGN could be powered by black holes with a large range of masses ($10^{6}\\rightarrow 10^{9}{\\rm M}_{\\odot}$).  At least part of the disagreement about what black-hole mass is required to produce a radio-loud AGN is due to different methods of classifying what constitutes `radio-loud'. For example, the classification adopted by Ho (2002) is the so-called radio-loudness parameter ${\\mathcal R}$, the ratio of radio-to-optical luminosity, under which any object satisfying ${\\mathcal R}>10$ is classified as radio-loud. Although there can be little doubt that objects with black-hole masses of $10^{6}\\rightarrow 10^{8}{\\rm M}_{\\odot}$ can be radio-loud under this classification, it is worth noting that only one object in the sample of Ho (2002) with an estimated black-hole mass of $<10^{8}{\\rm M}_{\\odot}$, would also satisfy the alternative radio-loudness criterion of $L_{\\rm 5GHz}>10^{24}$W~Hz$^{-1}$sr$^{-1}$ (Miller et al. 1990).  Following their study of the black-hole masses and host-galaxy properties of low redshift radio-loud and radio-quiet quasars, Dunlop et al. (2002) proposed an alternative view of the $M_{bh}-L_{\\rm rad}$ plane. They argue that the location of both active and non-active galaxies on the $M_{bh}-L_{\\rm rad}$ plane appears to be consistent with the existence of an upper and lower envelope, both of the approximate form $L_{\\rm 5GHz} \\propto M_{bh}^{2.5}$, but separated by some 5 orders of magnitude in radio power. In this scheme the upper and lower envelopes delineate the maximum and minimum radio luminosity capable of being produced by a black hole of a given mass. Dunlop \\& McLure (2002) demonstrate that this scenario naturally describes the FBQS data of Lacy et al. (2001), and we note here that it is also consistent with the findings of Ho (2002).  However, the recent study by Oshlack, Webster \\& Whiting (2002; hereafter OWW02) of the black-hole masses of a sample of flat-spectrum radio-loud quasars from the Parkes Half-Jansky Flat Spectrum sample of Drinkwater et al. (1997), casts doubt on the existence of any upper threshold in the $M_{bh}-L_{\\rm rad}$ plane. OWW02 found that their flat-spectrum quasars, which are securely radio-loud with respect to either classification mentioned above, harbour black-hole masses in the range $10^{6}\\rightarrow 10^{9}M_{\\odot}$, and therefore lie well above the upper $L_{\\rm 5GHz} \\propto M_{bh}^{2.5}$ boundary proposed by Dunlop et al. (2002). The conclusion reached by OWW02 following this result was that previous studies have actively selected against including powerful radio sources with relatively low black-hole masses, due to their concentration on luminous, optically selected radio-loud quasars.  In this paper we use the OWW02 sample to re-examine the position of these flat-spectrum radio-loud objects on the $M_{bh}-L_{\\rm rad}$ plane when both Doppler boosting effects and the likely geometry of the broad-line region are taken into account. The motivation behind this re-analysis is to determine, once the inevitable complications associated with flat-spectrum quasars are considered, whether this sample of objects does genuinely violate the apparent upper boundary to radio luminosity suggested by previous studies.  The paper is set out as follows, in Section~\\ref{sec:virial} we briefly summarise the main aspects of the virial black-hole mass estimate. In Section~\\ref{sec:sample} we provide a brief description of the OWW02 sample. In Section~\\ref{sec:doppler} we discuss the amount of Doppler boosting one might expect in flat-spectrum radio sources, and in Section~\\ref{sec:reanalysis2} we consider what effect a BLR with a flattened disk-like geometry will have on the virial black-hole mass estimates. In Section~\\ref{sec:reanalysis} we briefly discuss how the combination of these may affect the $L_{\\rm rad} - M_{bh}$ relation for flat-spectrum quasars.The implications of this work are discussed in Section~\\ref{sec:implications}.  All cosmological calculations presented in this paper assume $\\Omega_{\\rm M} = 0.3$, $\\Omega_{\\Lambda} = 0.7$, $H_{\\circ} = 70$~\\kmpspMpc.  \\begin{figure} \\includegraphics[width=0.48\\textwidth,angle=0]{fig1.ps} \\caption{Rest-frame 5~GHz radio luminosity versus redshift for the flat-spectrum sources analysed by OWW02. The solid line represents the flux-density limit of the Parkes half-Jy flat-spectrum sample assuming a spectral index of $\\alpha = 0.5$. Some objects appear below this line because the remeasured spectral index using the whole radio spectrum is steeper than $\\alpha = 0.5$  } \\label{fig:pzplane} \\end{figure} ",
        "conclusions": "\\label{sec:implications} We have re-analysed the data of Oshlack et al. (2002) on a sample of flat-spectrum radio-loud quasars. Contrary to their conclusions we find that, by correcting for the effects of inclination upon both the radio luminosity and estimated black-hole mass, the black holes harboured by intrinsically powerful flat-spectrum quasars are of comparable mass to those found in other quasars of similar {\\it intrinsic} radio luminosity, i.e. $M_{bh} > 10^{8}$~M$_{\\circ}$.  We also find that although many of the flat-spectrum quasars occupy the region of intrinsic radio luminosity comparable to the FRII radio sources (Fanaroff \\& Riley 1974) found in low-frequency selected radio surveys, some of the sources may occupy the lower-luminosity regime of radio-intermediate and radio-quiet quasars. Therefore, we conclude that by consideration of source inclination and intrinsic radio power, flat-spectrum quasars may well be consistent with the $L_{\\rm rad} \\propto M_{bh}^{2.5}$ relation found in previous studies.  Further work is obviously essential to make firm statements regarding the black-hole masses in flat-spectrum radio-loud quasars. This may be achieved by utilizing the bulge luminosity versus black-hole mass correlation to determine the black-hole mass independent of any orientation biases, and this is investigated in a subsequent paper (Jarvis, McLure \\& Rawlings in prep.)."
    },
    "0208/astro-ph0208359_arXiv.txt": {
        "abstract": "Using archival HST/WFPC2 imaging of $7$ LMC globular clusters, and following the methods outlined in our previous study, we have reached the tightest constrain so far on their age dispersion, which cannot be greater than $\\approx 0.5$~Gyr. We also confirm earlier results that their average age is comparable to that of the metal-poor Galactic globulars. Evidence is also provided that NGC~1841 is younger than the rest of LMC globulars. ",
        "introduction": "The LMC is the closest disk galaxy (at \\( \\sim 50 \\) kpc) with a large ensamble of star clusters. At least \\( 11 \\) of them can be considered \\emph{ bona fide} counterparts (e.g. Olszewski et al. 1996) of Galactic globular clusters (GGC). Their absolute ages are comparable to those of the Milky Way  metal-poor ones (Brocato et al. 1996; Olsen et al. 1998; Johnson et al. 1999), however, their kinematics is strikingly different, since they show a definite disk-like rotation of $\\approx 80~\\rm km~s^{-1}$ (Freeman et al. 1983). It is of great interest to understand the reasons of the simultaneous formation of the two systems in such dynamically different protogalaxies.  A key input to this problem is the age dispersion of LMC globulars. ",
        "conclusions": "We have found that both the mean age and age dispersion of the LMC GCs is comparable to that of the metal-poor Galactic GCs, and clear evidence has been obtained in favor of a younger age for NGC~1841.  Since it is the most metal-poor LMC cluster, it means that the interstellar medium (ISM) where it formed, has had a lower degree of chemical evolution. NGC~1841 could have been formed in a relatively isolated fragment of the proto-LMC, or it could have been part of an independent system now disrupted. Further support to the latter hypothesis could be the fact that NGC~1841 is the farthest cluster from the LMC center ($\\sim 10$~kpc) and  that its radial velocity is incompatible with the  rotation of the old LMC halo (Freeman et al. 1983). The fact that LMC globulars formed in a rotating protogalaxy with perfect synchronization with the GGCs, seems to require a joint formation scenario, possibly induced by the Milky Way early SF (e.g. Taniguchi et al. 1999)."
    },
    "0208/astro-ph0208115_arXiv.txt": {
        "abstract": "While symbiotic Miras and planetary nebulae are hard to distinguish by optical spectroscopy, their near infrared colors differ. We propose the near infrared two-color diagram to be an excellent tool to easily distinguish these two classes of objects. ",
        "introduction": "Nebulae around symbiotic Miras are very similar to planetary nebulae (PNe) in terms of morphology, excitation conditions, and chemical abundances, although they are formed in a slightly different way (Corradi 2002). While they are hard to distinguish by means of optical spectroscopy, their near infrared (NIR) colors differ noticeably. We obtained NIR photometry of a sample of 155 PNe and (known or suspected) nebulae around symbiotic Miras observed with DENIS (Deep Near Infrared Southern Sky Survey; Epchtein et al.\\ 1997) in Gunn-$I$ (0.82~\\micron), $J$ (1.25~\\micron), and $K{\\rm _s}$ (2.15~\\micron). The catalog is presented in Schmeja \\& Kimeswenger (2002a, 2003), details on the selected sample and on the data reduction are given in Schmeja \\& Kimeswenger (2001). ",
        "conclusions": "The DENIS NIR two-color diagram is an excellent and easily applicable tool to distinguish nebulae around symbiotic Miras from genuine PNe. Unfortunately, it does not apply to yellow symbiotics, however. Several (but certainly not all) bipolar PNe, like Mz~3 or M~2-9 turn out to be in fact symbiotic Miras. A widely ignored fact is the inconsistent classification of the discussed objects: Whereas many nebulae around symbiotic Miras (e.\\,g. He~2-104, He~2-147) are classified as PNe, others, like BI~Cru, that shows a similar nebula, are not. Whether those objects are accepted or rejected as PNe, or whether a new designation like ``symbiotic proto-PNe'' (Kohoutek 2001) is introduced, at least this should be done consistently for all objects of that kind."
    },
    "0208/astro-ph0208323_arXiv.txt": {
        "abstract": "We have studied the effect of the mass of the central star (CS) on the gas evolution during the planetary nebula (PN) phase. We have performed numerical simulations of PN formation using CS tracks for six stellar core masses corresponding to initial masses from 1 to 5 \\Mso. The gas structure resulting from the previous asymptotic giant branch (AGB) evolution is used as the starting configuration. The formation of multiple shells is discussed in the light of our models, and the density, velocity and \\ha emission brightness profiles are shown for each stellar mass considered. We have computed the evolution of the different shells in terms of radius, expansion velocity, and \\ha peak emissivity. We find that the evolution of the main shell is controlled by the ionization front rather than by the thermal pressure provided by the hot bubble during the early PN stages. This effect explains why the kinematical ages overestimate the age in young CSs. At later stages in the evolution and for low mass progenitors the kinematical ages severely underestimate the CS age. Large (up to 2.3 {\\rm pc}), low surface brightness shells (less than 2000 times the brightness of the main shell) are formed in all of our models (with the exception of the 5~\\Mso\\ model). These PN halos contain most of the ionized mass in PNe, which we find is greatly underestimated by the observations because of the low surface brightness of the halos. ",
        "introduction": "Low-and intermediate-mass stars experience high mass loss rates during the asymptotic giant branch (AGB) phase. Owing to the high mass loss rates during this phase, stars which have masses between 1 and $\\sim$8 \\Mso~end their lives as white dwarfs with final masses below the Chandrasekhar mass limit (1.44~\\Mso).  After an uncertain phase of evolution called the transition time (the timescales are not completely known), the stellar remnant\\footnote{Central star (CS) from now onwards.} becomes hot enough to ionize the previously ejected envelope. In the meantime, the wind velocity increases and shapes the inner parts of the envelope according to the so-called interacting stellar winds \\citep{Kpf:78}. This leads to the formation of a planetary nebula (PN). The evolution of the nebular gas from then onward depends on the energy provided by the CS through the wind and the radiation field, and both depend on the stellar luminosity and effective temperature. The post-AGB evolution of the CS is mainly determined by its core mass (\\citealt{Pac:71}; \\citealt{Wf:86}; \\citealt{Vw:94} [hereafter VW94]) and by the previous AGB evolution \\citep{Blo:95}.  The relationship between the CS evolution and that of the nebular shells has been the subject of many observational studies (e.g., \\citealt{Scs:93}; \\citealt{Fvb:94}; \\citealt{Sp:95}; \\citealt{Getal:96}; \\citealt{Hetal:97};\\citealt{Setal:02}). However, most of the numerical studies of PN evolution in the literature have been restricted to a $\\sim$0.6~\\Mso~post-AGB evolutionary track, without allowing for the CS mass range. This lack of consideration for the CS mass range is not the only problem with the existing numerical models of PN formation. Classically, the complex AGB evolution have been modeled by a simple $r^{-2}$ density law \\citep{Oetal:85,Sk:87,Fbr:90, Ms:91,Mel:94}. In the series of related papers of \\cite{Setal:97}, \\cite{Petal:98}, and \\cite{Cetal:00}, the previous AGB history is included; however, the grids were truncated to $\\sim$0.65 pc, and hence only the recent mass loss history can be completely followed up. Thus, the consequences of the long term thermal-pulse AGB evolution on PN structure at large scales still need to be tested.  Our aim in this paper is to study the role that the stellar evolution and the stellar progenitor mass play on PN formation. In a previous paper (\\citealt{Vgm:02} [hereafter, Paper I]) we described the dynamical evolution of the stellar wind during the AGB for low- and intermediate-mass stars. In this paper we use the gas structure resulting from the previous AGB evolution as the starting configuration. The PN formation is then followed by using post-AGB tracks that comprise the mass range of 0.57 to 0.9~\\Mso, which correspond to main sequence masses of between 1 and 5~\\Mso \\citep{Vw:93}. In $\\S2$ we describe the numerical method. The results of our models for different PN progenitors and grid sizes are presented in Section 3. In $\\S4$ we describe the observational properties derived from our models and we compare them with observations. Our conclusions are summarized in Section~5. ",
        "conclusions": "We have investigated PN formation considering the gas structure resulting from the preceding AGB as the starting point for our models, and using hydrogen-burning post-AGB tracks taken from VW94 with core masses 0.569, 0.597, 0.633, 0.677, 0.754, and 0.9 \\Mso, and solar metallicity.  As the result of our models we highlight the importance of the dynamical effects of ionization in the shell's evolution. During the early stages of the PN evolution the main shell is formed by the IF and is not driven by the hot bubble. We find that, with the exception of the 5 \\Mso~star model, multiple ionized shells are present in all the stellar models. The emission line profiles that characterize the attached shells result from a previous density distribution present in the circumstellar gas prior to the onset of the PN phase. The halos have sizes up to 2.3 {\\rm pc} (more than twice the size of the main shell) and are formed during the AGB phase,  with \\ha emissivities between 10 and 5000 times fainter than the main shell. We find that intermediate detached shells are formed in our models by the IF.  We have studied the dynamical evolution of the main shell and found that for young CSs, the kinematical ages tend to overestimate the CS ages for all the masses. At later stages of evolution, the kinematical ages underestimate the evolutionary status of the CSs when the sample is biased towards low mass progenitors. The details of the evolution of the main shell for the different masses accounts for the difference in timescales between low- and intermediate-mass progenitors.  According to our models the observations severely underestimate the ionized mass present in PNe as most of the ionized mass in PNe is contained in the the detached halos, which are not usually detected because of their faintness.  We thank M. L. Norman and the Laboratory for Computational Astrophysics for the use of ZEUS-3D. We also want to thank Letizia Stanghellini, Martin Guerrero, and Tariq Shahbaz for their careful reading of the manuscript and their valuable comments. The work of EV and AM is supported by Spanish grant PB97-1435-C02-01. GGS is partially supported by grants from DGAPA-UNAM (IN130698, IN117799, and IN114199) and CONACyT (32214-E)."
    },
    "0208/astro-ph0208053_arXiv.txt": {
        "abstract": "Mass segregation in a star cluster is studied in an analytical manner. We consider a two-component cluster, which consists of two types of stars with different masses. Plummer's model is used for the initial condition. We trace the overall behaviors of the probability distribution functions of the two components and obtain the timescale of mass segregation as a simple function of the cluster parameters. The result is used to discuss the origin of a black hole with mass of $\\gtrsim 10^3\\,M_{\\sun}$ found in the starburst galaxy M82. ",
        "introduction": "Within a star cluster, massive objects segregate into the cluster core as they lose kinetic energies to the less massive objects during approach toward energy equipartition. The result is an increase of the number density of the massive objects in the core. This phenomenon of mass segregation has long been known (Spitzer 1969; Binney \\& Tremaine 1987, p. 531) but has attracted new interests in recent years. The large number density induces successive mergers of those massive objects, which could lead to the formation of an exotic object such as a very massive black hole.  The important parameter of mass segregation is its timescale. In order to explain an observation in terms of mass segregation, the timescale has to be less than the age of the star cluster.  However, mass segregation has been studied only for specific cases using numerical methods, i.e., Fokker-Planck, Monte Carlo, and $N$-body simulations. The exceptional analytical works that are available are those of Spitzer (1969) and Tremaine, Ostriker, \\& Spitzer (1975). Spitzer (1969) considered an isothermal cluster of uniform density and estimated the timescale of mass segregation as the timescale of energy equipartition between stars of different masses. This model is too idealized for a star cluster, which is not isothermal or of uniform density. Tremaine et al. (1975) studied a decaying circular motion of an object in the singular isothermal sphere. This is originally a model for a motion of a globular cluster around the center of a galaxy, and it is not a model for random motions of stars in a cluster. Therefore, an analytical study of a more appropriate situation is desirable.  We analytically study a two-component cluster, which consists of two masses $m_0$ and $m$ ($m_0 < m$). The major component is the less massive stars. Their total mass $M_0$ dominates over the total mass $M$ of the more massive stars, $M_0 \\gg M$. Here the subscript 0 is used to indicate that the quantity is relevant to the major component. The cluster is  spherically symmetric in all its properties and completely isotropic in the velocity space.  Since the timescale of mass segregation is expected to be less than the relaxation timescale (Spitzer 1969), the star cluster is regarded as almost collisionless. The individual stars move almost freely in the mean potential. The gravitational encounters occur only in a stochastic manner. Our study is accordingly based on the overall behaviors of the probability distribution functions (PDFs) of the major and minor components. They are approximated by PDFs of completely collisionless systems. The minor-component PDF is from a parameterized family of steady-state PDFs for which the degree of mass segregation is allowed to change. The major-component PDF is kept the same (\\S2). These PDFs are used to obtain the energy loss per unit time from the minor component due to encounters with the major component (\\S3). The energy loss rate yields the rate of change of the minor-component PDF, which in turn yields the timescale of mass segregation (\\S4, eq. [\\ref{eq26}]).  We compare our result with those for relevant timescales, including the pioneering work of Spitzer (1969). We also use our result to discuss the origin of a black hole with mass of $\\gtrsim 10^3\\,M_{\\sun}$ found in a young star cluster of the starburst galaxy M82 (\\S5). ",
        "conclusions": "\\subsection{Comparison with Relevant Timescales}  The timescale of mass segregation is greater than the core crossing timescale. Since the core radius is $r_{\\rm c}$ and the central velocity dispersion is $(GM_0/2r_{\\rm c})^{1/2}$, the core crossing timescale $\\tau _{{\\rm cc}}$ is \\begin{equation} \\label{eq27} \\tau _{{\\rm cc}} = \\left( \\frac{2r_{\\rm c}^3}{GM_0} \\right) ^{1/2}. \\end{equation} Then equation (\\ref{eq26}) for the timescale of mass segregation $\\tau _{{\\rm ms}}$ is rewritten as \\begin{equation} \\label{eq28} \\tau _{{\\rm ms}} = \\frac{\\pi}{6} \\frac{m_0 N_0 /(m-m_0)}{\\ln \\left[ m_0 N_0 /(m+m_0) \\right]} \\tau _{{\\rm cc}} \\gg \\tau _{{\\rm cc}}, \\end{equation} where we have assumed $m_0/m \\simeq 10^{-1}$--$10^{-2}$ and $N_0 \\simeq 10^5$--$10^6$ as a practical situation. On the other hand, the timescale of mass segregation is less than the relaxation timescale $\\tau_{{\\rm rlx}}$ of the major component, which is about $N_0 / 8\\ln N_0$ times the crossing timescale $\\tau_{{\\rm cc}}$ (Binney \\& Tremaine 1987, p. 190): \\begin{equation} \\label{eq29} \\tau_{{\\rm rlx}} \\simeq \\frac{1}{8} \\frac{N_0}{\\ln N_0} \\tau_{{\\rm cc}} > \\tau_{{\\rm ms}}. \\end{equation} These results justifies our assumption that the star cluster is almost collisionless and the major-component PDF remains the same during mass segregation.  The timescale of mass segregation is close to the timescale of energy loss from the minor component, $\\tau_{{\\rm el}} = E/(dE/dt)$ at $t=0$. Using equations (\\ref{eq10}) and (\\ref{eq16}), we obtain \\begin{eqnarray} \\label{eq30} \\tau_{{\\rm el}} &=& \\frac{3^3 \\times 5 \\times 7^2 \\pi \\sqrt{\\pi}}{2^{23}} \\frac{19!!}{16!!} \\frac{\\Gamma (15/4)}{\\Gamma (13/4)} \\frac{1}{G^2 \\ln \\Lambda} \\frac{1}{m-m_0} \\left( \\frac{3M_0}{4 \\pi r_{\\rm c}^3} \\right) ^{-1} \\left( \\frac{GM_0}{2r_{\\rm c}} \\right) ^{3/2}. \\end{eqnarray} The dependence on physical quantities is the same as that of equation (\\ref{eq26}). This is because the energy loss from the minor component and the increase of its central concentration are of the same phenomenon. The numerical factor of equation (\\ref{eq30}) is about 0.4832, which happens to be in agreement with the numerical factor 2$^{-1}$ of equation (\\ref{eq26}). Since this energy loss timescale is obtained without any assumption on the evolution of the minor-component PDF, it is a robust measure.    \\subsection{Comparison with Spitzer (1969)}  Spitzer (1969) considered an isothermal cluster of uniform density and estimated the timescale of mass segregation as the timescale of approach toward energy equipartition between the major and minor components,\\footnote{ Defined as the time it takes for the difference between $m \\sigma^2$ and $m_0 \\sigma_0^2$ to decrease by a factor of $e^{-1}$.} which was about $m_0/m$ times the relaxation timescale $\\tau_{{\\rm rlx}}$ of the major component. Using equation (\\ref{eq29}), we obtain the corresponding timescale in our model cluster: \\begin{equation} \\label{eq31} \\tau_{{\\rm ee\\,(Spitzer)}} \\simeq \\frac{1}{8} \\frac{m_0N_0/m}{\\ln N_0} \\tau_{{\\rm cc}}. \\end{equation} If $m \\gg m_0$, this energy equipartition timescale is by definition equivalent to our energy loss timescale $\\tau_{{\\rm el}}$ in equation (\\ref{eq30}). Since the numerical factor is $2^{-1}$, the equation is rewritten as \\begin{equation} \\label{eq32} \\tau _{{\\rm el}} = \\frac{\\pi}{6} \\frac{m_0 N_0 /m}{\\ln \\left( m_0 N_0 /m \\right)} \\tau _{{\\rm cc}}, \\end{equation} where we have replaced $m \\pm m_0$ with $m$. Except for an unimportant difference in the argument of the logarithm, the timescales (\\ref{eq31}) and (\\ref{eq32}) have the same dependence on physical quantities. On the other hand, the numerical factors are different. This difference is not serious because there is ambiguity in the definition of the relaxation timescale. Our relaxation timescale (\\ref{eq29}) is based on the core crossing timescale $\\tau_{{\\rm cc}}$ and might not be sufficiently large to yield the energy equipartition timescale for the whole cluster. Overall, although Spitzer (1969) considered a very idealized cluster, his timescale is expected to be valid for energy equipartition in a real star cluster at least as an order-of-magnitude estimation. This is also the case for mass segregation, which occurs during approach toward energy equipartition.  Spitzer (1969) also predicted that, if the total mass of the minor component is much less than that of the major component, $M \\ll M_0$, the two components eventually achieve energy equipartition, $m \\sigma^2 = m_0 \\sigma_0^2$. His prediction has been confirmed at least for the cluster core by Fokker-Planck and Monte Carlo simulations (Inagaki \\& Wiyanto 1984; Watters, Joshi, \\& Rasio 2000: Fregeau et al. 2002). Since it is required that both the major and minor components evolve to have thermal velocity distributions (\\S3), the energy equipartition is achieved at $t \\gtrsim \\tau_{{\\rm rlx}}$.\\footnote{ This energy equipartition in the cluster core is not exact. The difference between $m \\sigma^2$ and $m_0 \\sigma_0^2$ is very small but still existent ($\\ga 1$\\%; Inagaki \\& Wiyanto 1984; Watters et al. 2000). The reason is that the velocity distributions do not become exactly thermal even in the cluster core.} This is beyond the limit of our model ($t < \\tau_{{\\rm rlx}})$, where the cluster is assumed to be almost collisionless. The major component is kept to follow Plummer's nonthermal model and hence does not achieve energy equipartition with the minor component, regardless of the total mass ratio $M/M_0$. This fact is not serious to our estimation of the timescale of mass segregation, which is less than the relaxation timescale $\\tau_{{\\rm rlx}}$ and has been defined at the initial time of the evolution $t = 0$ in equation (\\ref{eq19}).    \\subsection{Application to Observations} If the central mass density $\\rho_0(0)$ and central velocity dispersion $\\sigma_0(0)$ have values typical of globular clusters (Binney \\& Tremaine 1987, p. 26, see also p. 423 for the cutoff factor $\\Lambda$), equation (\\ref{eq26}) is written as \\begin{equation} \\label{eq33} \\tau _{{\\rm ms}} = 3 \\times 10^7\\,{\\rm yr} \\left( \\frac{\\ln \\Lambda}{10} \\right) ^{-1} \\left( \\frac{m}{10\\,M_{\\sun}} \\right) ^{-1} \\left( \\frac{\\rho_0(0)}{10^4\\,M_{\\sun}\\,{\\rm pc}^{-3}} \\right) ^{-1} \\left( \\frac{\\sigma_0(0)}{10\\,{\\rm km}\\,{\\rm s}^{-1}} \\right) ^3. \\end{equation} Here we have assumed $m \\gg m_0$ and replaced $m-m_0$ with $m$. If we instead use the typical values of the core radius $r_{\\rm c}$ and central velocity dispersion $\\sigma_0(0)$, both of which are observable quantities, equation (\\ref{eq26}) is rewritten as \\begin{equation} \\label{eq34} \\tau _{{\\rm ms}} = 2 \\times 10^7\\,{\\rm yr} \\left( \\frac{\\ln \\Lambda}{10} \\right) ^{-1} \\left( \\frac{m}{10\\,M_{\\sun}} \\right) ^{-1} \\left( \\frac{r_{\\rm c}}{1\\,{\\rm pc}} \\right) ^2 \\left( \\frac{\\sigma_0(0)}{10\\,{\\rm km}\\,{\\rm s}^{-1}} \\right). \\end{equation} Thus mass segregation in a typical star cluster has the timescale $\\tau_{{\\rm ms}}$ of order $10^7$ yr.  The above timescale of mass segregation is used to discuss the origin of a very massive black hole found in the starburst galaxy M82 (Kaaret et al. 2001; Matsumoto et al. 2001; see also Matsushita et al. 2000). At the 2 \\micron\\ secondary peak, i.e., an active site of star formation, there is a source of compact X-ray emission. The observed strong variability implies that the source is an accreting black hole. The observed luminosity of $10^{41}$\\,ergs\\,s$^{-1}$ implies that the mass is greater than $10^3$\\,$M_{\\sun}$ if the emission is isotropic and its luminosity is below the Eddington limit. This black hole should have been formed via successive mergers of massive stars and black holes that had segregated into the core of a star cluster (Taniguchi et al. 2000). Since the 2 \\micron\\ secondary peak has a starburst age of about $1 \\times 10^7$\\,yr (Satyapal et al. 1997), the mass segregation should have occurred well within this short duration. The timescales given in equations (\\ref{eq33}) and (\\ref{eq34}) are too long. We suspect that the star cluster is exceptionally dense and compact. For example, the young star cluster R136 in the Large Magellanic Cloud has $\\rho_0(0) \\simeq 10^6\\,M_{\\sun}$\\,pc$^{-3}$, $\\sigma_0(0) \\simeq 10^1$\\,km\\,s$^{-1}$, and $r_{\\rm c} \\simeq 10^{-2}$--$10^{-1}$\\,pc (Portegies Zwart et al. 1999), which yields $\\tau _{{\\rm ms}} \\simeq 10^5$\\,yr. In fact, only one black hole heavier than $10^3\\,M_{\\sun}$ has been found among more than 100 young star clusters of M82. The objects that had segregated into the cluster core should have merged in a runaway manner (Quinlan \\& Shapiro 1989; Portegies Zwart et al. 1999; Mouri \\& Taniguchi 2002) because the merger probability is higher for more massive objects."
    },
    "0208/astro-ph0208271_arXiv.txt": {
        "abstract": "{ Structure of mean pulsar radiation patterns is discussed within the nested-cones and patchy beam models. Observational predictions of both these models are analyzed and compared with available data on pulsar waveforms. It is argued that observational properties of pulsar waveforms are highly consistent with the nested-cone model and, in general, inconsistent with the patchy beam model. ",
        "introduction": "One of the important questions in pulsar research is what the overall structure of the mean pulsar beam is and how this structure is related to highly fluctuating instantaneous pulsar radiation. It is difficult to reveal this structure as pulsar observations represent one-dimensional cuts through two-dimensional beams. However, some indirect methods have been applied in an attempt to resolve this problem and two major models of pulsar beams have emerged from this work. \\citet{r93}, \\citet{gks93} and \\citet{kwj94} calculated the opening angles $\\rho$ of emission corresponding to a pulse width $W$ measured at 10 and 50 percent of the maximum intensity. As a result, they obtained a binomial distribution of these angles, that is, for a given period $P$ one of the two preferred values was possible, following however a general $P^{-1/2}$ dependence. Such distribution is most naturally interpreted as an indication of two nested cones in the structure of mean pulsar beams. This interpretation is called a conal model of pulsar beams. An alternative model postulates that the mean pulsar beam is patchy \\citep[][LM88 hreafter]{lm88}, with different components randomly distributed within an almost circular ``window function'' \\citep{m95,hm01}. Such a model is apparently inconsistent with the binomial distribution of the opening angles inferred from measured pulse widths. In fact, unless putative patches are distributed along nested-circular patterns, the distribution of corresponding opening angles should be (for any given period) random rather than binomial.  Recently, \\citet[][MD99 hereafter]{md99} attempted to test both these rival models. They distributed locations of the profile components (measured as the peak-to-peak separation of the outer conal components in complex profile pulsars) on one quadrant of the beam represented on the common normalised scale (with $x$-axis and $y$-axis representing longitudes $\\varphi$ and impact angles $\\beta$, respectively; refer to Fig.~4 of MD99). They found that most of the peak intensity points are concentrated in narrow sections of the beam. This feature is a strong indication of the conal structure of the pulsar beam. It can be argued that if, indeed, the pulsar beams are patchy, then in such a case there is a high probability for the beam to be uniformly filled with peak intensity locations. Further in their analysis, they excluded the so-called conal single and conal double profiles \\citep{r83}, which are thought to be exclusively grazing cuts of the line-of-sight at the beam boundary. They found that such exclusion in their sample led to the absence of points at high impact angles $\\beta$, which is perfectly consistent with the nested cone model and inconsistent with the patchy beam model. In fact, within the patchy beam model there is no reason why the single and double profiles occur exclusively at high impact angles. Moreover, the midpoint of single and double profiles usually coincides with the fiducial phase, at which the multifrequency profiles align after being corrected for cold plasma dispersive delays. This property is natural within the conal model and inconsistent (in general) with the patchy beam model, since patches would have to be placed symmetrically with respect to the fiducial plane, containing both magnetic ${\\bf m}$ and spin ${\\bf\\Omega}$ axes (Fig.~\\ref{fig1}). As argued by MD and independently by \\citet[][GS00 hereafter]{gs00} the ``mean'' average pulsar beam consists of up to three nested cones, centered on the global magnetic dipole axis.   \\begin{figure*} \\centering \\includegraphics[height=18cm]{jkfigure1.eps} \\caption{The beam emission patterns for both the conal model (illustrated on the right-hand side), and for the patchy beam model (illustrated on the left-hand side) are presented in this composite picture. The geometry of observation is determined in both models by the inclination angle $\\alpha$ between the magnetic $\\bf{m}$ and the spin $\\bf\\Omega$ axes and the impact angle $\\beta$ of the closest approach of the line-of-sight (observer ${\\bf O}$) to the magnetic axis $\\bf{m}$. The fiducial plane containing the spin axis $\\bf\\Omega$, the magnetic axis $\\bf{m}$ and the observer $\\bf{O}$ defines the fiducial longitude $\\varphi=0^\\circ$ which is common for both models. All pulse phases ($\\varphi_s$ - longitude of subpulse peak, $\\varphi_p$ - longitude of profile component peak) are measured from this fiducial longitude, to the left from it in the patchy model and to the right from it in the conal model. In this paper we ignore the dispersive delays which are the only cause of frequency variation at or near the fiducial plane. The frequency-dependent position of any observed feature in the beam pattern is described by two angles: the opening angle $\\rho(\\nu)$ between the ${\\bf m}$ axis and the line-of-sight (l-of-s) and by the azimuthal angle $\\sigma$ between the fiducial plane and the plane of dipolar field lines associated with a particular feature. Within the conal model the subpulse emission is associated with the frequency dependent subpulse spots $S(\\nu)$, which move circumpherentially to form a cone $C(\\nu)$. Thus, the profile components are associated with the frequency dependent cones in the conal model, while within the patchy model the subpulse-associated spots $SP(\\nu)$ occupy the patchy areas $PA(\\nu)$ related to the  profile components. The longitudes of subpulse peaks $\\varphi_s(\\nu)$ and profile peaks $\\varphi_p(\\nu)$ are marked, with $\\nu_1>\\nu_2$, in both models. \\label{fig1}} \\end{figure*}  Observations of single pulses in strong pulsars show that longitudes of subpulses are weakly dependent on frequency as compared with longitudes of  corresponding profile components \\citep{i93,gggk02}. Within the conal model, the longitudes of profile components are determined by the intersection of the line of sight trajectory with the frequency-dependent cones of the maximum average intensity, while the longitudes of subpulses are determined by the intersection of the line-of-sight trajectory with subpulse-associated emission beams, which move across the average cones as frequency changes \\citep{gk96}. We demonstrate in this paper that the different frequency dependence of subpulse and profile component longitudes is a natural property of the conal model, and that both subpulses and profile components should demonstrate the same frequency dependence of their longitudes within the patchy model. We present both general qualitative arguments and detailed quantitative model calculations to support the above statements. For better understanding of our arguments, this paper should be studied along with the paper by \\citet[][GGGK hereafter]{gggk02}, in which details of frequency dependence of emission patterns in PSR B0329$+$54 are discussed within the conal model of pulsar beams. ",
        "conclusions": "In this paper we explore a geometrical method of pulsar radiation simulation, based on two  well-justified assumptions: (i) the elementary coherent radio emission is narrow-band, and the emission altitude depends on both the frequency and the pulsar period, (ii) the emission is relativistically beamed tangently to dipolar magnetic field lines. We have considered two competitive models of the organization of pulsar emission beams: the conal model, in which enhancements related to subpulse emission in single pulses are distributed along the cones corresponding to maximum average intensity, and the patchy beam model in which subpulse enhancements corresponding to the component of the mean profile are confined to the patchy area limited both in azimuthal and radial dimensions. We examined the consistency of these rival models with the variety of observational data.   We have argued that a number of observational properties of pulsar radio emission, namely: (i) binomial distribution of the opening angles \\citep{r93,gks93,kwj94}; (ii) high impact angles corresponding to single and double profile pulsars (MD), and (iii) different frequency dependence of a subpulse and corresponding profile component longitudes \\citep{i93,gk96,gggk02}, strongly support the conal model of pulsar beams. The alternative patchy beam model is inconsistent with these observational properties of pulsar radiation.   We have also demonstrated that the beam reconstruction technique developed by \\citet{hm01} is not capable of  revealing the true structure of pulsar beams. In fact, their formalism assumes implicitely that neither the radio emission altitude nor the number of putative nested-cones and their locations within the pulsar depends on the pulsar period. The lack of an apparent conal structure in their ''global beam'' does not exclude the conal beam model. Thus, the results of the HM01 analysis provide no strong evidence of  patchy beam structure in pulsars.  We tend to favour the version of the conal model in which the relationship between the subpulse-associated beams and cones of the average emission is established through the phenomenon of the ${\\bf E}\\times{\\bf B}$ drift \\citep[][GS00]{rs75,dr99,dr01}, which forces the spark filaments of plasma to rotate slowly around the magnetic axis. This ``spark model'' of radio pulsars was recently tested statistically by \\citet{fan01}. They showed, by means of  Monte Carlo simulations, that various pulsar parameters can be reproduced if both the spark dimension and their mutual separation are approximately equal to the height $h$ of the polar gap \\citep[][GS00]{rs75}, or consequently, the maximum number of sparks along the diameter of the polar cap with the radius $r_p$ is $N_{max}\\sim r_p/h$. Thus, their conclusions are consistent with the assumptions used in this paper.    We note that  evidence of a relationship between drifting subpulses and the conal structure of mean pulsar beams already exists  in the literature. \\citet{hw87} examined three pulsars with triple average profiles showing subpulses drifting across the full pulse window (including the central component). They found that these pulsars are consistent with two nested cones of emission, each associated with a prominent subpulse drift. Moreover, \\citet{pw86} showed that in complex profile pulsars there is a strong correlation between drifting subpulses associated with different profile components. Such correlations are natural within the ${\\bf E}\\times{\\bf B}$ induced conal model, but inconsistent with the patchy model of pulsar beams.  Finally, we suggest that more pulsars should be observed in the single pulse, simultanenous dual frequency mode. Our simulation method illustrated and described in Fig. 2 can be adapted to the analysis of such data in order to ultimately discriminate  between conal and patchy beam models in pulsars."
    },
    "0208/astro-ph0208047_arXiv.txt": {
        "abstract": "Theoretical integrated broad-band colors ranging from far-UV to near-IR have been computed for old stellar systems from our evolutionary population synthesis code. These models take into account, for the first time, the detailed systematic variation of horizontal-branch (HB) morphology with age and metallicity. Our models show that some temperature-sensitive color indices are significantly affected by the presence of blue HB stars. In particular, $B$ $-$ $V$ does not become monotonically redder as metallicity increases at given ages, but becomes bluer by as much as $\\sim$ 0.15 mag because of the contribution from blue HB stars. Similar trends are also found in the Washington photometric system. In addition to appropriate age-sensitive spectrophotometric indices, the use of far-UV to optical colors is proposed as a powerful age diagnostic for old stellar systems with differing HB morphologies.  Our models are calibrated in the $B$ $-$ $V$, $V$ $-$ $I$, $C$ $-$ $T_{1}$, and $M$ $-$ $T_{1}$ vs. [Fe/H] planes, using low-reddened Galactic globular clusters (GCs) [$E$($B$ $-$ $V$) $<$ 0.2] and the relative age difference between the older inner halo Galactic GCs and younger outer halo counterparts is well reproduced. Several empirical linear color-metallicity transformation relations are assessed with our models and it is noted that they may not be safely used to estimate metallicity if there are sizable age differences amongst GCs within and between galaxies. M31 GCs are found to be fundamentally similar to those in the Milky Way, not only in the optical to near-IR range, but also in the UV range. For globular cluster systems in two nearby giant ellipticals, M87 and NGC 1399, the current available photometric data in the literature does not appear sufficient to provide robust age discrimination. It is anticipated, however, that the detailed population models presented here coupled with further precise spectrophotometric observations of globular cluster systems in external galaxies from the large ground-based telescopes and space UV facilities will enable us to accurately estimate their ages and metallicities. ",
        "introduction": "When stellar systems can be resolved into individual stars such as open and globular star clusters in the Milky Way and Local Group galaxies, one can derive their ages and metallicities via isochrone fitting to the main sequence turnoff and red giant branch in the color-magnitude diagram and fitting luminosity functions to the white dwarf cooling sequence (Hansen et al. 2002). This luxury does not exist for more distant systems, however, so one must rely upon their integrated colors or spectra. Calibrating theoretical spectrophotometric quantities using local, resolved systems is then surely a necessary step (e.g., Gibson et al. 1999).  Since local and nearby extragalactic globular clusters (GCs) are easily detected and feasible for analysis due to their rather simple nature, there have been many photometric studies of globular cluster systems in external galaxies which have revealed important information regarding the link between GC systems and their host galaxies (see Ashman \\& Zepf 1998 for a recent review). Information regarding their relative ages within a given galaxy or between galaxies is particularly relevant for studying the hierarchical formation of galaxies. As globular cluster ages and metallicities become better constrained, the opportunities for understanding the process of galaxy formation and its subsequent evolution can only increase.  In recent years, there have been many efforts to develop evolutionary population synthesis models (e.g., Bruzual \\& Charlot 1993; Worthey 1994; Buzzoni 1995; Vazdekis et al. 1996; Maraston 1998; Lee 2001a, 2001b) in order to analyze the integrated spectrophotometric quantities from globular clusters and galaxies and to derive their mean ages and metallicities. It is true that those integrated quantities are indeed determined by age and metallicity of those stellar systems, but it also should be kept in mind that they are the two most arguably dominant parameters that determine globular clusters' horizontal-branch (HB) morphologies. Therefore, the integrated colors of GCs should be affected to some degree by differing HB morphologies at the similar metallicity (the `second parameter' effect). It is now generally accepted that age is the {\\it global} second parameter that controls HB morphology [after the first parameter, metallicity -- Lee, Demarque, \\& Zinn 1994; Sarajedini, Chaboyer, \\& Demarque 1997; Salaris \\& Weiss 2002], although for some GCs a third (or more) parameter may be needed to explain their peculiar HB morphologies [such as blue tail phenomenon -- Recio-Blanco et al. 2002].  We are extending this second parameter analysis to broader age ranges in this work, applicable even to potentially older stellar systems than we have in the Milky Way, and investigate how variation of HB morphology influences the integrated photometric quantities. With a better understanding of post-main sequence stellar evolution, it is now feasible to study the systematic effects of those evolved stars on the integrated spectrophotometric values. One vivid example has already appeared in Lee, Yoon, \\& Lee (2000; hereafter Paper I) for the realistic assessment of ages of old stellar systems using the H$\\beta$ index. A unique aspect of these models lies in its treatment of the systematic variation of HB morphology as a function of age and metallicity. In this paper, we present a detailed quantitative analysis toward the effects of HB morphology on theoretical integrated broad-band colors ranging from far-UV to near-IR at various stages of age and metallicity. In addition, integrated broad-band colors that have been used to estimate the metallicity of globular clusters are examined with our models in order to check the veracity of the so-called color-metallicity transformation relations that are widely used, but based solely upon Galactic GCs.  In Section 2, our population models developed with and without consideration of HB stars are presented and compared with those of other groups. We first consider optical to near-IR colors, including those using the Washington photometric system, an investigation of UV to optical colors is then presented. Section 3 corroborates the validity of our models with HB stars by calibrating them using a sample of relatively low-reddened Galactic GCs and our results are then used to assess several color-metallicity transformation relations. We compare our models with available UV to near-IR data for GC systems in M31, M87, and NGC 1399 (e.g., Cohen, Blakeslee, \\& Ryzhov 1998; Kissler-Patig et al. 1998; Forbes et al. 2001). In most cases, we have specially sought age-sensitive colors, to combine with spectroscopic metallicity information. A discussion of the implications of our work is provided in Section 4. ",
        "conclusions": "The primary goal of our paper was to investigate the effects of horizontal-branch stars on the integrated broad-band colors of old, simple, stellar populations. To do so, we have employed a unique, self-consistent, treatment of HB morphology as a function of age and metallicity. We have found that some temperature-sensitive integrated broad-band colors are significantly affected by the presence of blue HB stars within our investigated age range of $-$ 4 Gyr $\\leq$ $\\Delta${\\it t} $\\leq$ + 4 Gyr (i.e., 8 Gyr $\\leq$ t $\\leq$ 16 Gyr). The close agreement between our models and the relatively low-reddened inner halo Galactic GCs in both the Johnson-Cousins and Washington filter systems is encouraging. The use of far-UV to optical colors is suggested to be a powerful age-dating regime for old stellar systems, when coupled with realistic HB morphologies (Lee 2001a, 2001b). Future observations of GC systems in external galaxies from large ground-based telescopes and space UV facilities will enable us to quantify any systematic age differences between the various systems. A more sophisticated understanding of the theory governing mass loss and helium enrichment will also contribute to a better understanding of stellar age determination using far-UV photometry (O'Connell 1999).  In the case of the M31 GC system, our work suggests that it is not fundamentally different from the Galactic system, from the UV through to the near-IR. This is in contrast to several studies which have claimed that the metal-rich M31 GCs could be much younger than the metal-poor sample (e.g., Burstein et al. 1984; Barmby \\& Huchra 2000). Future far-UV datasets to be provided, for example, by the {\\it GALEX} mission should aid in resolving this controversy. Further Washington system photometry will also be valuable (Lee et al. 2001).  The study of extragalactic GC systems today is being driven primarily by an attempt at understanding the bimodal color distributions seen from many early-type galaxies (e.g., Larsen et al. 2001; Kundu \\& Whitmore 2001) as well as from some spirals (Forbes, Brodie, \\& Larsen 2001). The origin of these blue and red subpopulations and the implication for the formation of their host galaxies remains unclear. If a relative age difference exists between the subpopulations, an important piece of the galaxy formation puzzle will have been found. Identifying useful age discriminants remains an important component of cosmology and galaxy formation. The broad-band photometric predictions described here, in combination with the spectroscopic discriminants of Paper I, should provide observers with the necessary tools to discriminate age differences amongst GC subpopulations. Finally, we conclude that several of the canonical {\\it linear} color-metallicity transformation relations should be used with caution if there are sizable age differences amongst globular clusters within and between galaxies. It remains to be seen if bimodal color distributions can be interpreted as simple metallicity or age differences, or whether a more complicated interplay between metallicity and age must be involved."
    },
    "0208/astro-ph0208337_arXiv.txt": {
        "abstract": "We present a statistical method to derive the mass functions of open clusters using sky survey data such as the 2 Micron All Sky Survey (2MASS) and the Guide Star Catalogue (GSC). We have used this method to derive the mass functions in the stellar/substellar regime of three young, nearby open clusters, namely IC 348, $\\sigma$ Orionis and Pleiades. The mass function in the low mass range (M$<0.50 M_\\odot$) is appreciably flatter than the stellar Salpeter function for all three open clusters. The contribution of objects below 0.5~M$_\\odot$ to the total mass of the cluster is $\\sim$40\\% and the contribution of objects below 0.08~M$_\\odot$ to the total is $\\sim$4\\%. ",
        "introduction": "Recent surveys have found a significant population of low-mass stars, brown dwarfs and planetary-mass objects in young open clusters. Since low-mass objects evolve little over the lifetime of the Universe, the present day mass function of these objects is a good representation of the Initial Mass Function (IMF). The mass function in this low-mass regime is however poorly known due to faintness of these objects and also due to uncertainty in the mass-luminosity relations. Low-mass objects at or below the Hydrogen Burning Mass Limit (HBML) of 0.08~M$_\\odot$ are known to be warmer and hence more luminous  when young although they cool rapidly and fade with age (Baraffe et al. 1998). The combination of youth and proximity in some open clusters make them ideal targets for searches of low-mass objects below the HBML particularly at infrared wavelengths. In the present study we have adopted a statistical approach to determine the mass function (dN/dM $\\propto$ M$^{-\\alpha}$) of objects in the mass range $0.5~M_\\odot$ to 0.025--0.05~M$_\\odot$ using data of three open clusters namely IC 348, $\\sigma$ Orionis and Pleiades. ",
        "conclusions": ""
    },
    "0208/astro-ph0208101_arXiv.txt": {
        "abstract": "Recent studies have found the Type II-plateau supernova (SN) 1999gi to be highly polarized ($p_{\\rm max} = 5.8 \\%$, where $p_{\\rm max}$ is the highest degree of polarization measured in the optical bandpass; Leonard \\& Filippenko 2001) and minimally reddened ($\\EBV = 0.21 \\pm 0.09$ mag; Leonard et al. 2002). From multiple lines of evidence, including the convincing fit of a ``Serkowski'' interstellar polarization (ISP) curve to the continuum polarization shape, we conclude that the bulk of the observed polarization is likely due to dust along the line of sight (l-o-s), and is not intrinsic to SN~1999gi.  We present new spectropolarimetric observations of four distant Galactic stars close to the l-o-s to SN~1999gi (two are within $0.02^{\\circ}$), and find that all are null to within $0.2\\%$, effectively eliminating Galactic dust as the cause of the high polarization.  The high ISP coupled with the low reddening implies an extraordinarily high polarization efficiency for the dust along this l-o-s in NGC~3184: ${\\rm ISP}/\\EBV = 31^{+22}_{-9}\\% {\\rm\\ mag}^{-1}$.  This is inconsistent with the empirical Galactic limit (${\\rm ISP}/\\EBV < 9\\% {\\rm\\ mag}^{-1}$), and represents the highest polarization efficiency yet confirmed for a single sight line in either the Milky Way or an external galaxy. ",
        "introduction": "\\label{sec:introduction}  It has long been understood that dust in the interstellar medium of the Milky Way (MW) is responsible for the linear polarization of starlight from intrinsically unpolarized Galactic stars (see, e.g., Greenberg 1968, and references therein). The accepted model for the origin of the interstellar polarization (ISP) is linear dichroism (directional extinction), which results from aspherical dust grains aligned by some mechanism such that their optic axes have a preferred direction.  By means of Mie theory computations (e.g., Greenberg 1968), the extinction efficiency factors for a specific grain size, shape, and refractive index may be calculated.  Typically, the dust grains are modeled as dielectric cylinders with refractive index near $1.6$, the value for silicates. The basic result is that an aspherical dust grain extinguishes slightly more light that is propagating with electric vector parallel to its long axis than it does light with electric vector perpendicular to its long axis. The resulting polarization direction is therefore perpendicular to the direction of the grain's alignment.  The predicted degree of polarization produced by linear dichroism is strongly wavelength dependent, with maximum polarization occurring when the characteristic size of the grains (the cylinder's radius) is of the same order as the wavelength of the light.  Unlike the behavior of dust extinction, the expected spectral dependence of the ISP therefore rapidly decreases towards either longer {\\it or shorter} wavelengths away from the peak.  Such dust polarization models have been quite successful at reproducing the empirically derived ``Serkowski law'' ISP curve (Serkowski 1973; Wilking, Lebofsky, \\& Rieke 1982; Whittet et al. 1992) in the optical and near-infrared bandpasses (see, e.g., Mathis 1986).  Naturally, the same dust that polarizes the light is expected to redden it as well, a prediction observationally verified for MW stars.  The degree of polarization produced for a given amount of extinction (or reddening) is referred to as the ``polarization efficiency'' of the intervening dust grains. The most efficient polarization medium conceivable is obtained by modeling the dust grains as infinite cylinders (length $\\gg$ radius) with diameters comparable to the wavelength of the incident light, perfectly aligned with their long axes parallel to one another and perpendicular to the line-of-sight (l-o-s).  For such a model, Mie calculations place a theoretical upper limit on the polarization efficiency of the grains due to directional extinction at visual wavelengths of (Whittet 1992, and references therein) \\begin{equation} \\frac{\\rm ISP}{\\Ebv} <  R_V \\times 13.82\\%\\ {\\rm mag}^{-1} , \\label{eq:3} \\end{equation} \\noindent where ${\\rm ISP}$ is the maximum degree of interstellar polarization measured in the optical band, \\ebv\\ is the reddening, and $R_V$ is the ratio of total to selective extinction (e.g., Savage \\& Mathis 1979).  This upper limit on the polarization efficiency does not vary greatly with the refractive index of the grains considered, and is therefore rather insensitive to uncertainty in grain composition (Spitzer 1978, p. 175).  Adopting the canonical value of $R_V = 3.1$ for MW dust leads to a maximum theoretical polarization efficiency along typical lines of sight in the MW of ${\\rm ISP}/\\Ebv < 43\\%\\ {\\rm mag}^{-1}$.  This may be compared with the empirically derived upper bound (Serkowski, Mathewson, \\& Ford 1975) resulting from polarization studies of reddened Galactic stars of \\begin{equation} \\frac{\\rm ISP}{\\Ebv} < 9.0\\%\\ {\\rm mag}^{-1} . \\label{eq:4} \\end{equation} \\noindent The fact that the observed polarization efficiency in the MW is more than a factor of 4 less than theory allows is generally taken as evidence that the alignment of dust grains is not total (or has multiple preferred orientations due to non-uniformity of the magnetic field along the l-o-s), and/or that the grains are only moderately elongated particles (rather than infinite cylinders) that may be irregularly shaped.  The empirical limit given by Equation~(\\ref{eq:4}) is found to hold for thousands of lines of sight to Galactic stars.  To be sure, there are several stars in the comprehensive stellar polarization catalog by Heiles (2000) that exceed this limit (indeed, several polarized stars have claimed values of $\\Ebv \\leq 0.0$ mag, formally creating an infinite polarization efficiency), but the colors and spectral types of the stars are not generally thought to be determined with sufficient accuracy to claim a clear violation of the inequality given by Equation~(\\ref{eq:4}).  Somewhat surprisingly, there have been very few measures of the polarization efficiency of dust in external galaxies.  By far the best way to determine the polarization efficiency of dust in an external galaxy is to measure the ISP along the l-o-s to a discrete source with a well-determined reddening.  Although this has been a very rare exercise (see Hough 1996 for a review of the four studies known at that time, and Rodrigues et al. 1997 for a subsequent study of the polarization and reddening of stars in the Small Magellanic Cloud), the results of the investigations that have been carried out are generally consistent with the upper limit given by Equation~(\\ref{eq:4}); while a few of the Large Magellanic Cloud stars studied by Clayton, Martin, \\& Thompson (1983) do exceed the limit, the discrepancy has been attributed to incorrect spectral classification or intrinsic colors (Hough 1996).  As part of a campaign to study the physical geometry of young supernovae (SNe; Leonard, Filippenko, \\& Matheson 2000b; Leonard et al. 2000a, 2001, 2002a), Leonard \\& Filippenko (2001; hereafter LF01) present a single epoch of optical spectropolarimetry of the bright, Type II-plateau supernova (SN) 1999gi in NGC~3184 during the late photospheric phase (107 days after discovery).  They find an extraordinarily high degree of linear polarization, $p_{\\rm max} = 5.8\\%$, where $p_{\\rm max}$ is the highest level of polarization observed in the optical bandpass.  If intrinsic to SN~1999gi, such polarization is without precedent for this type of event (see Wheeler 2000; Leonard et al. 2001; LF01). It implies an enormous departure from spherical symmetry, since even the most strongly aspherical theoretical models (axis ratio of 5-to-1, favorably viewed edge-on) of H\\\"{o}flich (1991) fail to produce polarizations of more than $\\sim 5\\%$.  However, there is compelling evidence that much of the observed polarization is, in fact, due to interstellar dust.  One clue pointing toward an interstellar origin is that the polarization is not flat with wavelength, but rather has a broad, asymmetric peak that gently decreases on either side (Fig.~\\ref{fig:1}).  This is unlike the wavelength-independent polarization expected from an aspherical electron-scattering supernova atmosphere, but is very similar to the polarization observed for reddened Galactic stars (Serkowski 1973; Wilking et al. 1982; Whittet et al. 1992).  Since the same dust that polarizes the SN light should also redden it, the natural expectation is for significant reddening.  But, the Galactic foreground reddening being only $\\Ebv = 0.017$ mag (Schlegel, Finkbeiner, \\& Davis 1998), LF01 contend that nearly all of the dust along the l-o-s to SN~1999gi must reside in the host galaxy, NGC~3184.  \\begin{figure} \\ssp \\vskip -0.5in \\hskip -0.3in \\begin{center} \\rotatebox{0}{ \\scalebox{0.8}{ \\plotone{fig1.ps} } } \\end{center} \\caption{Observed polarization (calculated as the rotated Stokes parameter; see Leonard et al. 2001) of SN~1999gi ({\\it thin line}) on 2000 Mar. 25 (107 days after discovery; UT dates are used throughout this paper) compared with the total flux spectrum from the same date ({\\it thick line}), arbitrarily scaled and offset for comparison of features.  A Serkowski ISP curve characterized by $p_{\\rm max} = 5.78\\%, \\theta = 154^\\circ, {\\rm\\ and\\ } \\lambda_{\\rm max} = 5100$ \\AA\\ is shown for comparison ({\\it dashed line}; see \\S~\\ref{sec:alternativestoisp}).  Prominent features in the spectropolarimetry and total flux spectra are labeled. \\label{fig:1}} \\end{figure}   Leonard et al. (2002c; hereafter L02) present optical spectra and photometry of SN~1999gi spanning the first six months after discovery as part of a study to derive the distance to SN~1999gi through the expanding photosphere method. Surprisingly, L02 conclude that SN~1999gi is only minimally reddened: Estimates resulting from five independent techniques are all consistent with a hard upper limit of $\\Ebv < 0.45$ mag established by comparing the early-time color of SN~1999gi with that of an infinitely hot blackbody, and yield a probable reddening of $\\Ebv = 0.21 \\pm 0.09$ mag.  In this paper, we further investigate the nature of the observed polarization of SN~1999gi reported by LF01 in light of the recent evidence for low reddening found by L02.  In \\S~\\ref{sec:observations} we present new spectropolarimetric observations of four distant Galactic ``probe'' stars close to the l-o-s of SN~1999gi in order to test the assertion by LF01 that the polarization is not due to Galactic dust, and discuss the results in \\S~\\ref{sec:galacticpolarization}.  In \\S~\\ref{sec:alternativestoisp} we consider alternatives to ISP as the cause of the polarization, and introduce a new technique to quantify the component of SN polarization that lies in the direction perpendicular to the ISP direction in the Stokes parameter $q$-$u$ plane.  In \\S~\\ref{sec:polarizationefficiencyofdustinngc3184} we derive the polarization efficiency for dust in NGC~3184 along the l-o-s to SN~1999gi.  We discuss the results in \\S~\\ref{sec:discussion} and summarize our conclusions in \\S~\\ref{sec:conclusions}. ",
        "conclusions": "\\label{sec:conclusions}  The single spectropolarimetric epoch obtained for the Type II-P SN~1999gi by LF01 indicates a very high observed polarization ($p_{\\rm max} = 5.8\\%$), with a shape that is well characterized by a Serkowski law ISP curve.  The null result obtained by spectropolarimetric observations of four Galactic stars near to the l-o-s to SN~1999gi and sufficiently distant to fully sample Galactic dust convincingly eliminates Galactic ISP as a significant contributor to the observed polarization. It is also difficult to reconcile intrinsic SN polarization, newly formed dust in the ejecta, or dust reflection by off-axis dust blobs with the observed polarization characteristics.  We therefore conclude that ISP produced by dust in NGC 3184 is largely responsible for the observed polarization, and that it lies in the range $4.3\\% \\leq {\\rm ISP} \\leq 7.3\\%$.  Since L02 find SN~1999gi to be only minimally reddened, $\\Ebv = 0.21 \\pm 0.09$ mag (with all reddening estimates supporting a firm upper limit of $\\EBV < 0.45$ mag, which was derived through the comparison between the early-time continuum shape with that of an infinitely hot blackbody), we derive a polarization efficiency for the dust along this l-o-s in NGC~3184 of $31^{+22}_{-9}\\% {\\rm\\ mag}^{-1}$.  Even by assuming the limiting reddening value of $\\Ebv = 0.45$ mag and the lower ISP limit of ISP = 4.3\\%, the derived polarization efficiency does not conform to the empirical Galactic limit of ${\\rm ISP}/\\EBV < 9\\% {\\rm\\ mag}^{-1}$, and these limiting values are thought to be extremely unlikely.  Therefore, unless the estimates for the ranges of either the reddening or the ISP are grossly in error, this measurement is the highest polarization efficiency yet confirmed for a single sight line in either the MW or an external galaxy.  Although the polarization efficiency for dust along the l-o-s to SN~1999gi is inconsistent with the empirically derived maximum observed Galactic polarization efficiency, it can be accommodated by the maximum theoretical efficiency derived from Mie scattering considerations (${\\rm ISP}/ \\EBV < 40\\%\\ {\\rm mag}^{-1}$).  In addition, the dust grain polarization model of Jones et al. (1992) predicts high polarization efficiency for lines of sight to sources with low reddening ($A_V < 1$ mag).  From an examination of the Galactic stellar polarization agglomeration catalog of Heiles (2000), we also find some evidence for unusually high polarization efficiency for dust in dark clouds, although the current lack of specific extinction information for these regions precludes definitive conclusions.  Since the low-reddening regime has been only sparsely sampled by previous studies, increased scrutiny of the polarimetry of discrete Galactic and extragalactic sources with low reddening certainly seems warranted to test whether the polarization efficiency inferred for dust along the l-o-s to SN~1999gi in NGC~3184 is indeed common in the low-reddening regime."
    },
    "0208/astro-ph0208451_arXiv.txt": {
        "abstract": "The effects of a wind on the emerging spectrum from an inefficiently-radiating accretion flow in a global magnetic field are examined, based on the analytic solution obtained recently by one of the present authors. The results exhibit the steepening of the negative slope appearing in the intermediate frequency range of bremsstrahlung spectrum and the decrease in the luminosity ratio of thermal synchrotron to bremsstrahlung, in accordance with the increasing wind strength. Both effects are due to a suppressed mass accretion rate in the inner disk, caused by a mass loss in terms of wind.  In order to demonstrate the reliability of this model, Sagittarius A$^*$ (Sgr A$^*$) and the nucleus of M\\ 31, both of which have been resolved in an X-ray band by {\\it Chandra}, are taken up as the best candidates for the broadband spectral fittings. Although the observed X-ray data are reproduced for these objects by both of the inverse-Compton and the bremsstrahlung fittings, some evidence of preference for the latter are recognized. The wind effects are clearly seen in the latter fitting case, in which we can conclude that a widely extending accretion disk is present in each nucleus, with no or only weak wind in Sgr A$^*$ and with a considerably strong wind in the nuclear region of M\\ 31. Especially in Sgr A$^*$, the inferred mass accretion rates are much smaller than the Bondi rate, whose estimate has become reliable due to {\\it Chandra}. This fact strongly suggests that the accretion in this object does not proceed like Bondi's prediction, though its extent almost reaches the Bondi radius. ",
        "introduction": "The broadband spectrum of Sgr A$^*$, from radio to hard X-ray bands, was reproduced fairly well for the first time by Narayan, Yi and Mahadevan (1995) based on an optically thin ADAF (advection-dominated accretion flow) model. In contrast to the standard accretion-disk model (Shakura \\& Sunyaev 1973), the optically thin ADAF model could reproduce the wide spread of the observed spectrum. Namely, the rather narrow peak in the radio band was explained by thermal synchrotron radiation from the inner part of a disk with relativistic temperature, and X-ray luminosity, by the bremsstrahlung from the whole disk. Another great advantage of this model is in its low efficiency in producing radiative fluxes. The latter feature could make the model possible to explain the low luminosity of Sgr A$^*$ compared with that expected from a standard disk of about the Bondi accretion rate (see, e.g., Melia \\& Falcke 2001).  This type of models for optically thin ADAFs has been developed by many authors (e.g., Ichimaru 1977; Rees et al. 1982; Narayan \\& Yi 1994, 1995a, b; Abramowicz et al. 1995; Blandford \\& Begelman 1999). Since, in this category, the viscosity of the accreting plasma plays a dominant role in both energy dissipation and angular-momentum extraction processes, we call it the viscous ADAF model. The model looked successful in explaining the broadband spectra not only of Sgr A$^*$ (Narayan, Yi \\& Mahadevan 1995; Manmoto, Mineshige \\& Kusunose 1997; and Narayan et al. 1998b), but also of Galactic X-ray sources and of low-luminosity active galactic nuclei (LLAGNs; for a review, see Narayan, Mahadevan \\& Quataert 1998a).  Meanwhile, if the presence of ordered magnetic fields in the nuclear regions of galaxies is taken seriously, anther type of optically thin ADAF model can be constructed (Kaburaki 2000, for brevity, K00). Since, in this model, the gravitational energy of accreting matter is liberated through the plasma's electric resistivity it will be called the resistive ADAF model. Angular momentum, on the other hand, is extracted from the accreting matter by an ordered magnetic field that is penetrating the disk and is twisted, to a certain extent, by the rotational motion of the accreting matter. The observed spectrum of Sgr A$^*$ could be reproduced also by this model (Kino, Kaburaki \\& Yamazaki 2000, for brevity, KKY) as satisfactorily as the viscous ADAF models could.  Recently, however, a fairly drastic change of the above stated situation has come with the appearance of the results of the X-ray telescope, {\\it Chandra}. Owing to its high resolution, the apparent nuclear sources of a few nearby galaxies have been further resolved into several point sources including a true nuclear source in each case (e.g., Garcia et al. 2000 for M31; Baganoff et al. 2001a for Sgr A$^*$). The intrinsic X-ray luminosity of these true nuclear sources are therefore considerably lower in fact, and moreover, their spectra have tuned out to be much softer than previously believed.  Especially, this softness of the spectra requires a critical reconsideration of the broadband spectral fittings of these objects by optically thin ADAF models, because this fact may exclude the hitherto accepted bremsstrahlung fitting to the X-ray band of the spectra. Further, as already been demonstrated by some authors (e.g., Quataert \\& Narayan 1999; Di Matteo et al. 2000), the presence of winds emanating from accretion disks can alter the ratio of X-ray luminosity to radio luminosity. Therefore, it is also necessary to include the presence of winds into the basic models based on which the spectral fittings are performed. In this context, the resistive ADAF model has been developed to include winds emanating from the disk surfaces (Kaburaki 2001, for brevity, K01).  The present paper is devoted to describe the general predictions on the broadband spectra radiated from such accretion disks as described by the analytic model of K01, and to report the results of applications to Sgr A$^*$ and the nucleus of M31. Although Di Matteo et al. (2000) insist the presence of winds in the nuclei of some nearby elliptical galaxies on the basis of their broadband spectral fittings by a wind-version of the viscous ADAF model, the results are still uncertain because the X-ray fluxes they used are obtained by {\\it ASCA} and hence the nuclear sources are not resolved (see also Quataert \\& Narayan 1999, for a Galactic X-ray transient and Sgr A$^*$). We therefore restrict our applications only to such objects as have been resolved by {\\it Chandra} and the results of observations have been open to the public. ",
        "conclusions": "We have examined the expected effects of a wind on the emerging spectrum from an ADAF in a global magnetic field, based on the recently proposed resistive ADAF model including winds (K01). The main effects are seen both in the spectral index (the power of $\\nu$) appearing in the intermediate frequency range of the thermal bremsstrahlung, and in the luminosity ratio of the thermal synchrotron emission to the bremsstrahlung. These two values decrease according to the strength of a wind. This fact can be explained by a suppressed mass accretion rate in the inner disk caused by wind loss.  In order to test the plausibility of the resistive ADAF model, we have fitted the observed broadband spectra of Sgr A$^*$ and of the nucleus of M\\ 31, by this model. For each observed spectrum, there are two possible types of fittings. One is the Compton fitting in which the negative X-ray slopes in the spectral energy distribution, which are obtained by {\\it Chandra} for both objects, are reproduced by the synchrotron self-Compton process, and the other is the bremsung fitting in which the negative slopes are reproduced by the intermediate frequency range of the thermal bremsstrahlung.  On the grounds of the goodness only of the fitting to the observational data points currently available, it is difficult to clearly distinguish the superiority of the one type of fitting to the other, because of the shortage of the observational data. However, we prefer the bremsung fittings for both objects. The main reasons for that are uncomfortably large values required for the strength of the seed magnetic field in the Compton fittings (0.5-3 G for both objects) and the very wide extension of accretion disks in the bremsung fittings, which seems favorable to the X-ray observations in the case of Sgr A$^*$.  If the bremsung fittings are more plausible, we can conclude that the ADAFs extend so far as to reach the Bondi accretion radii (for both objects $r > (1$ - $5)\\times10^5\\ r_{\\rm S}$), with no or very weak wind in the Sgr A$^*$ case, and with fairly strong wind in the M\\ 31 case. The resulting mass accretion rates for both objects are smaller than the Bondi rates by more than an order of magnitude, and this fact strongly suggests that the actual accretion processes in these objects are certainly different from Bondi's spherical accretion.  The major concern of our model in its present form is in the point that it largely under estimates the central masses. This fact seems to come from the over estimation of the electron temperature near the disk's inner edge. Therefore, the improvement of the model's accuracy especially near the inner boundary is strongly desired.  \\par"
    },
    "0208/astro-ph0208517_arXiv.txt": {
        "abstract": "We report on the first phase of our study of slightly rotating accretion flows onto black holes. We consider inviscid accretion flows with a spherically symmetric density distribution at the outer boundary, but with  spherical symmetry broken by the introduction of a small, latitude-dependent angular momentum. We study accretion flows by means of numerical 2D, axisymmetric, hydrodynamical simulations. Our  main result is that the properties of the accretion flow do not depend as much on the outer boundary conditions (i.e., the amount as well as distribution of the angular momentum) as on the geometry of the non-accreting matter. The material that has too much angular momentum to be accreted forms a thick torus near the equator. Consequently, the geometry of the polar region, where material is accreted (the funnel), and the mass accretion rate through it are constrained by the size and shape  of the torus. Our results show one way in which the mass accretion rate of slightly rotating gas can be significantly reduced compared to the accretion of non-rotating gas (i.e., the Bondi rate), and set the stage for calculations that will take into account the transport of angular momentum and energy. ",
        "introduction": "Some of the most dramatic phenomena of astrophysics, such as quasars and powerful radio galaxies, are most likely powered by accretion onto supermassive black holes (SMBHs). Nevertheless, SMBHs appear to spend most of their time in  a remarkably quiescent state. SMBHs are embedded in the relatively dense environments of galactic nuclei, and  it is natural to suppose that the gravity due to an SMBH will draw in matter  at high rates, leading to a high system luminosity. However, this simple prediction often fails as many systems are much dimmer than one would expect.  To illustrate a key problem in constructing theoretical models for accretion onto a black hole, let us express the luminosity due to accretion as \\begin{equation} L= \\eta c^2 \\MDOT_a, \\end{equation} where we invoke the simplest assumption, that the luminosity is proportional to the mass accretion rate, $\\MDOT_a$, and an efficiency factor, $\\eta$. The accretion luminosity is very uncertain because $\\eta$ is uncertain: $\\eta$ ranges from $\\sim 10^{-1}$ in a standard, radiatively efficient thin disk, to $\\sim 10^{-11}$ for spherically symmetric accretion from a low density medium (e.g., Shakura \\& Sunyaev 1973; Shapiro 1973; M\\'{e}sz\\'{a}ros 1975). The mass accretion rate  is also a source of uncertainty  in estimates of the accretion luminosity because  $\\MDOT_a$ depends on the physical conditions and geometry at large distances from the black hole. Nevertheless, it is customary to adopt the analytic formula due to Bondi (1952) to estimate the mass accretion rate. In his classic paper, Bondi (1952) considered spherically symmetric accretion from a non-rotating polytropic gas with uniform density $\\rho_\\infty$ and sound speed $c_\\infty$ at infinity. Under these assumptions, a steady state solution to the equations of mass and momentum conservation exists with a mass accretion rate of \\begin{equation} \\MDOT_B= \\lambda 4 \\pi R^2_B \\rho_\\infty c_\\infty, \\end{equation} where $\\lambda$ is a dimensionless parameter that, for the Newtonian potential, depends only on the adiabatic index. The Bondi radius, $R_B$, is defined as \\begin{equation} R_B=\\frac{G M}{c^2_\\infty}, \\end{equation} where $G$ is the gravitational constant and $M$ is the mass of the accretor.   The Bondi accretion formula predicts that SMBHs in typical galaxies should be more luminous than observations indicate when $\\eta$ is assumed to be as large as in a standard, radiatively efficient thin disk (e.g., Di Matteo et al. 1999, 2000, 2001; Loewenstein et al. 2001; Baganoff et al. 2001). In the context of equation (1),  one possible explanation for this disagreement is that the black hole accretion flow can be radiatively inefficient because binding energy dissipated in the gas is advected through the event horizon before being radiated (Ichimaru 1977; Rees et al. 1982; Narayan \\& Yi 1994, 1995; Abramowicz et al. 1995). However, pure advection-dominated inflow may not be the whole story. Even before recent observations forced us to confront very low SMBH luminosities, theorists had begun to realize that rotating, radiatively inefficient hydrodynamical (HD) flows are subject to strong convection (Begelman \\& Meier 1982; Narayan \\& Yi 1995), which can fundamentally change the flow pattern and its radiative properties (Blandford \\& Begelman 1999; Quataert \\& Narayan 1999; Narayan, Igumenshchev \\& Abramowicz 2000; Quataert \\& Gruzinov 2000). The theoretical studies showed that convection alters the steep ($\\propto r^{-3/2}$) density profiles of the advection-dominated flow into a much flatter ($\\propto r^{-1/2}$) profile, which can explain the faintness of many SMBHs because it predicts relatively low density close to the black hole (i.e., $\\MDOT_a$ is low in eq.~2). Similar structural changes occur in the magnetohydrodynamical (MHD) limit (Stone \\& Pringle 2001; Hawley, Balbus, \\& Stone 2001; Machida, Matsumoto \\& Mineshige 2001; Igumenshchev \\& Narayan 2002; Hawley \\& Balbus 2002), although here the turbulence is probably driven by  magnetorotational instability  (MRI) rather than thermal convection (Balbus \\& Hawley 2002; but see Abramowicz et al. 2002 and Narayan et al. 2002 for alternative views).  The turbulent character of both HD and MHD models does not settle the issue of what happens to the energy and angular momentum that must be transported away. There are two possibilities: (i) turbulent transport effectively shuts off the accretion flow, turning it into a closed circulation (Narayan et al. 2000; Quataert \\& Gruzinov 2000) or (ii) turbulent transport drives powerful outflows that can strongly modify the black hole's environment (Narayan \\& Yi 1994, 1995; Blandford \\& Begelman 1999). Recent MHD simulations bring new insights that may help us to resolve this issue. For example, Hawley \\& Balbus's (2002) three-dimensional MHD simulations show that, with and without resistive heating, mass and energy in  nonradiative accretion flows are carried off by an outflow in keeping with the outline of the second possibility.  Another possible solution to the problem of very low SMBH luminosity is that mass is captured into the accretion flow at a rate that is far lower that $\\MDOT_B$. Thus, a complete formulation of the accretion flow must also consider the outer boundary conditions. SMBHs draw matter from an extended medium and most authors assume that the Bondi (1952) formula provides an adequate approximation for the rate of mass supply. There has  been little systematic work done to demonstrate that this assumption is justified and correct. Igumenshchev \\& Narayan (2002) showed that even non-rotating Bondi accretion can be altered, in particular that the mass accretion rate can be reduced below the Bondi rate. The cause of the $\\MDOT_a$ reduction in Igumenshchev \\& Narayan's simulations is the local generation of energy through  by magnetic reconnection, which leads to the development of efficient convection. However, in  Igumenshchev \\& Narayan's simulations the flow was supersonically injected into the computational grid through the outer boundary, at  a specified rate (determined from the Bondi formula). Their calculations did not address self-consistently the problem of the mass accretion and supply rate because the latter was fixed at the outer boundary. As we mention above, the Bondi formula has been derived under the assumption that the gas is non-rotating and only under the influence of the central gravity. Thus, for a given gravitational field, the gas internal energy determines the accretion rate. By relaxing this assumption, introducing additional forces or sources of energy, one may find that the mass supply rate is much lower than the one predicted by the Bondi formula. For example, the rate at which matter is captured by a black hole can be severely limited when the matter is heated by X-rays produced near the black hole (Ostriker et al. 1976) or by  mass outflow from the central region (Di Matteo et al. 2002). In these two cases, the gas internal energy is increased. Introducing kinetic energy to the gas at infinity may have a similar effect: although the flow outside the Bondi accretion radius often can be described as nonrotating, even a tiny amount of angular momentum, $l$ --- when followed inward --- could severely limit the rate at which matter is captured by the black hole.  Our focus shall be on assessing the gross properties of rotating accretion flows onto  black holes. We consider a classic Bondi accretion flow with the only modifications being the introduction of a small, latitude-dependent angular momentum at infinity and a pseudo-Newtonian gravitational potential. The imposed angular momentum is small enough  to have a negligible effect on the density distribution at the outer boundary, which remains spherically symmetric. Therefore, matter near the rotational axis can be accreted. We thus consider a very simple model of an accretion flow, far simpler than those occurring in nature, as we neglect the gravitational field due to the host galaxy, radiative heating and cooling, viscosity and MHD effects. For example, we will not consider here the transport of energy and angular momentum outward, as needed to accrete matter with a specific angular momentum higher than $2 R_S c$ (where $R_S=2 G M/c^2$ is the radius of a Schwarzschild black hole). Nevertheless, the results presented here provide a useful exploratory study of accretion onto black holes. In particular, our results constitute a ``baseline'' for evaluating the effects of dissipative and transport processes in subsequent work. They also serve as a ``proof-of-concept'' for the reduction of the mass accretion rate due to a small angular momentum in the accretion flow.  \\subsection{Expectations}  Before we embark on a detailed analysis and numerical HD simulations, we first consider the problem of  accretion of  low-$l$ material in a general way. If the matter far from the SMBH (well outside the Bondi radius) has a uniform density and a specific angular momentum, $l$, which exceeds $2R_Sc$ --- a tiny value compared to the Keplerian angular momentum at $R_B$ --- then no accretion will take place without angular momentum transport. After a transient episode of infall, the gas will pile up outside the black hole and settle into a nearly steady state atmosphere bounded by a centrifugal barrier near the rotation axis. Realistically, matter far from the SMBH will have a range of angular momenta, and in a steady state with axisymmetry, there will always be low-$l$ material close to the axis that can accrete steadily through a funnel along the rotational axis.  For this highly idealized problem, one would expect that the mass accretion rate should scale with the dependence of the angular momentum on the polar angle, $\\theta$, at the outer radius, $r_o$: the larger the amount of the material with $l > 2 R_S c$ at $r_o$, the lower the mass accretion rate. This geometrical argument on the nature of  the $\\MDOT_a$ $vs.$ $l$ relation can be quantified as follows. If the angular momentum  depends on $\\theta$ as \\begin{equation} l(\\theta)=l_0 f(\\theta), \\end{equation} where  $f=1$ on the equator ($\\theta=90^\\circ$) and monotonically decreases to zero at the poles ($\\theta=0^\\circ$ and $180^\\circ$), then a naive expectation would be that  $\\MDOT_a/\\MDOT_B$ scales with the solid angle within which $l< 2 R_S c$: \\begin{equation} \\frac{\\MDOT_a}{\\MDOT_B}=\\frac{\\Delta \\Omega_o}{4\\pi}= 1 - \\cos{\\theta_o}, \\end{equation} where $\\theta_o$ is the width of the angular distribution for which $l \\le 2 R_S c$. The latter can be formally defined as \\begin{equation} \\theta_o \\equiv f^{-1} \\left[\\min\\left(1,\\frac{2R_Sc}{l_0}\\right)\\right], \\end{equation} where $f^{-1}$ represents the functional  inverse  of $f$. This simple geometrical argument is based on the assumption of radial flow and implies that if $l\\leq 2 R_S c$ the material will be accreted approximately at the Bondi rate. If relation (6) were true then the accretion rate should decrease with increasing $l_0$ for a fixed $R_S$. Additionally, one would expect that $\\MDOT_a/\\MDOT_B$ should be independent of $R_B$ for fixed $R_S$ and $l_0$. However, to determine the actual mass accretion rate even in this idealized case we need to perform numerical simulations, as just HD effects of the inviscid fluid make the above geometrical argument invalid both quantitatively and qualitatively.   The  main result of our numerical HD calculations is that the properties of the accretion flow do not depend as much on the outer boundary conditions (i.e., the amount as well as distribution of the angular momentum) as on the geometry of the {\\it non-accreting matter}. Material with $l \\simgreat 2 R_S c$ cannot accrete and forms a thick torus near the equator. This thick torus and its formation have been a subject of numerous studies (see below).  Our simulations  show that the dependence of angular momentum on $\\theta$ in the torus gets weaker with decreasing radius. The material with $l \\simgreat 2 R_S c$ inflows in the polar region, turns around as it reaches a centrifugal barrier, and then starts to outflow  along the equator. As a result, a thick torus of nearly uniform specific angular momentum forms as gas of the highest angular momentum ($l \\approx l_0 > 2 R_S c$) near the equator is replaced by gas of lower angular momentum ($l \\simgreat 2 R_S c$). The geometry of the polar region, where material is accreted (the funnel) and the mass accretion rate through it are constrained by the size and shape of the torus. We describe the size of the torus by an angle, $\\theta_t$, between the torus's upper envelope and the pole at a characteristic radius.  Our HD models show that the $\\MDOT_a$ $vs.$ $l$ relation has three regimes for a given $f(\\theta)$: (i) for low $l$ (i.e., $l < 2 R_S c$ for all $\\theta$ at large radii), the torus does not form and $\\MDOT_a= \\MDOT_B$, (ii) for intermediate $l$, or more appropriately where there is a narrow range of $\\theta$ for which $l>2 R_S c$ so $\\theta_o > \\theta_t$, $\\MDOT_a \\sim const$ with the actual value of the  constant depending on the ratio $R_S/R_B$ and (iii) for  high $l$, or in the case for which $l> 2 R_S c$ at nearly all $\\theta$ so $\\theta_o < \\theta_t$, $\\MDOT_a$  decreases with increasing $l_0$.   Comparing the $\\MDOT_a$ $vs.$ $l$ relation based on  our HD models with that described by equation (5), we find that the two relations agree exactly in the first regime, disagree qualitatively and quantitatively in the second regime [the HD models predict $\\MDOT_a$ lower than eq. (5)], and agree again but only qualitatively in the third region [the HD models predict $\\MDOT_a$ higher than eq. (5)]. Thus the geometrical argument used above does not hold. However, we can use a modified version to describe the key aspects of  the $\\MDOT_a$ $vs.$ $l$ relation. The modification to the geometrical argument is to replace the solid angle within which $l< 2 R_S c$ at the outer boundary, $\\Delta \\Omega_o$, with the solid angle within which $l< 2 R_S c$ at {\\it a characteristic radius comparable with the sonic radius}, $\\Delta \\Omega_f$ (i.e., the solid angle of the accretion funnel). Thus, $\\MDOT_a/\\MDOT_B \\approx \\Delta \\Omega_f/ 4\\pi$. The insensitivity of $\\MDOT_a$  to the angular momentum at infinity, in the second regime, can be attributed to the relative insensitivity of the torus, and thus the funnel, to the angular momentum distribution at infinity. In terms of the solid angle within which $l < 2 R_S c$, this corresponds to $\\Delta \\Omega_f=constant$  for variable $l_0$, provided that $\\Delta \\Omega_f < \\Delta \\Omega_o$ (i.e., $\\theta_o > \\theta_t$). On the other hand, the decrease of $\\MDOT_a$ with increasing $l_0$, in the third regime, can be attributed to the fact that  $\\Delta \\Omega_f$ decreases with increasing $l_0$, provided that $\\Delta \\Omega_f > \\Delta \\Omega_o$ ($\\theta_o < \\theta_t$). We find that the mass accretion rate decreases with increasing $l_0$ more slowly  than predicted by eq. (5) in the third regime because the sonic point radius starts to increase as the funnel gets narrower.  \\subsection{Previous Work}  Similar calculations have been performed before. For example, the formation of rotationally supported thick tori from inviscid accretion of gas with various initial angular momentum distributions has been reported (Hawley, Smarr \\& Wilson 1984a; 1984b; Clarke, Karpik \\& Henriksen 1985; Hawley 1986; Molteni,  Lanzafame \\& Chakrabarti 1994; Ryu et al. 1995; Chen et al.  1997). However, there is one key  difference between our work and some past work: our outer radial boundary is located outside the Bondi radius and we adopt subsonic, Bondi-like outer radial conditions whereas  Molteni et al. 1994, Ryu et al. 1995, and Chen et al.  1997 (see also Toropin et al. 1999; Kryukov et al. 2000; and Igumenshchev \\& Narayan 2002) imposed  outer boundary conditions inside the Bondi radius or even inside the sonic radius. The latter approach allows one to study cases where $R_S/R_B$ is as low as in some real systems (e.g., $R_S/R_B=10^{-5}$ in Chen et al. 1997) but this approach is not suitable for addressing our main issue: what is the mass supply rate. The approach adopted by Hawley et al. (1984a, 1984b), Clarke et al. (1985) and Hawley (1986) is much closer to ours as they also used subsonic outer boundary conditions. However, these authors did not consider how the accretion rate onto the black hole depends on the angular momentum distribution beyond the Bondi radius but rather focused on the formation of the thick torus. As far as the treatment of the outer radial boundary is concerned, our simulations are also similar for those of Ruffert (1994), who studied three-dimensional hydrodynamic Bondi-Hoyle accretion.  Other studies are also relevant to our work. Several authors considered accretion onto black holes with a focus on the evolution of rotationally supported thick tori including the transport of angular momentum and energy (e.g., Igumenshchev \\& Abramowicz 1999; Stone, Pringle \\& Begelman 1999; Stone \\& Pringle 2001; Machida et al. 2001; Hawley \\& Balbus 2002). The main difference between our work and these studies is that the other authors adopt the point of view that virtually all of the material at large radii has too much angular momentum to be accreted without the transport of angular momentum. They assume that material with zero or very low angular momentum is unimportant dynamically, and that accretion is dominated by  angular momentum and energy transport processes. For example, for their initial conditions Stone et al. (1999) and Hawley \\& Balbus (2002) adopted a bounded torus in hydrostatic equilibrium with constant angular momentum, embedded in zero angular momentum ambient gas which is also in  hydrostatic equilibrium. Thus, these  calculations were set up so that, if not for the transport of angular momentum, there would be neither time evolution nor  mass accretion. We recognize that transport processes are essential, but assert that the zero or very low angular momentum material also deserves a rigorous treatment because it can play an important role in determining the total mass supply  and accretion rate.  In this paper, we consider a far simpler case of an accretion flow than those occurring in nature (see Section~4). For example, we neglect viscosity and MHD effects. In particular, the MRI has been shown to be a very robust and universal mechanism to produce turbulence and the transport of angular momentum in disks at all radii (Balbus \\& Hawley 1998).   The outline of this paper is as follows. We describe our calculations in Section 2. In Section 3, we present our results. We summarize our results and discuss them together with their limitations in Section 4. ",
        "conclusions": "This paper presents the first phase of our study of slightly rotating accretion flows onto  black holes. We decided to consider a far simpler case of an accretion flow than those occurring in nature. In particular, we neglected  the gravitational field due to the host galaxy, radiative heating and cooling, viscosity and MHD effects. Perhaps the most important simplification we made is neglecting the transport of energy and angular momentum outward as needed to accrete matter with a specific angular momentum higher than $2 R_S c$. Nevertheless, our results  provide a useful exploratory study of accretion onto black holes as they have revealed unexpected properties and complexity of accretion flows in even this simple case. Clearly, a lot more  work is needed to give a definitive answer to the question of whether slow rotation of gas at large radii is really enough to reduce the mass accretion rate to the level required by observations. In what follows we will summarize our results,  briefly  review the limitations of our work, and discuss how the physical effects neglected here may change the results.  We have performed numerical 2D, axisymmetric, hydrodynamical simulations of slightly rotating, inviscid accretion flows onto a black hole. We attempt to mimic the boundary conditions of classic Bondi accretion flows with the only modifications being the introduction of a small, latitude-dependent angular momentum at the outer boundary and a pseudo-Newtonian gravitational potential. The distribution of $l$ with latitude allows the density distribution at infinity to approach spherical symmetry. The main result of our calculations is that the properties of the accretion flow do not depend as much on the outer boundary conditions (i.e., the amount as well as distribution of the angular momentum) as on the geometry of the {\\it non-accreting matter}. Additionally, we find that the mass accretion rate $vs.$ angular momentum relation for a given angular distribution of $l$ has three regimes: (i) for low $l$ (i.e., $l < 2 R_S c$ for all $\\theta$ at large radii), a torus does not form and $\\MDOT_a= \\MDOT_B$, (ii) for intermediate $l$, or more appropriately where there is a narrow range of $\\theta$ for which $l>2 R_S c$ so that $\\theta_o > \\theta_t$, $\\MDOT_a \\sim const$ and (iii) for  high $l$, or in the case for which $l> 2 R_S c$ at nearly all $\\theta$ so that $\\theta_o < \\theta_t$, $\\MDOT_a$  decreases with increasing $l_0$. Our limited data suggest that in the second regime, the actual value of the constant is dependent on the ratio $R_S/R_B$  for large $R_S/R_B$ but becomes independent on the ratio for small $R_S/R_B$ (i.e., for $R_S/R_B\\simless 10^{-3}$). We conclude that the inclusion of even slow rotational motion of the inviscid flow at large radii can significantly reduce $\\MDOT_a$ compared to the Bondi rate, as the $\\MDOT_a$ $vs.$ $l$ relation in the third regime indicates. For $R_S/R_B \\ge 10^{-3.5}$, our results show that $\\MDOT_a$ can be reduced by $\\sim 1.5$ orders of magnitude compared to the Bondi rate. To reduce the mass accretion rate more, the accretion funnel needs to be much narrower than the  funnels allowed by our assumption of low angular momentum (eq. 27) for $R_S/R_B \\ge 10^{-3.5}$. It remains to be seen whether simulations for values of $R_S/R_B$ as low as those observed in astrophysical systems (i.e., $R_S/R_B \\simless 10^{-6}$) will confirm our predictions  that the accretion funnel can be very narrow and subsequently the mass accretion rate very small.  We note that the discontinuous change from quasi-spherical to a disk-like flow was first found by  Abramowicz  \\& Zurek (1981). Abramowicz \\& Zurek studied analytically the adiabatic accretion of a radial flow onto a black hole using the PW potential. They considered  the case of constant specific angular momentum and also found that for small $l$, the flow is quasi-spherical and becomes transonic at $x_s \\gg R'_S$ while for sufficiently large $l$, the flow has a disk-like pattern and $x_s \\leq 3 R'_S$. Applying Abramowicz  \\& Zurek's solution for the sonic point, we find very good agreement between the location of the sonic point on the equator and their prediction. In particular, for $n\\equiv1/(\\gamma-1)=3/2$, $l=2 R_Sc$, and zero total energy, eq. 3.5 in Abramowicz \\& Zurek (1981) yields that the sonic point equals $1+\\sqrt{3}$, which is in a good  agreement with our result for $\\theta=90^\\circ$ where the sonic radius is $\\approx 2.5 R_S$ (e.g., see Figure~4). Our simulations also agree with Abramowicz  \\& Zurek's result that the transition between quasi-spherical and disk-like accretion occurs at $l\\approx 2 R_S c$.  We find that the shape of the polar funnel through which matter is accreted is partially constrained by the torus, which is made of matter that cannot accrete. The torus consists of material with roughly uniform angular momentum and its structure does not depend much on the outer conditions. The accretion funnel is therefore also not very sensitive to the outer conditions in the limit that it is broad at large radii (i.e., large $\\theta_o$). This is the key reason for the insensitivity of $\\MDOT_a$ to $l$ in the second regime and its weak sensitivity in the third regime: the mass accretion rate depends on the geometry of the sonic surface at radii where the presence of the torus is important, and not only on the geometry at the outer boundary. Thus, the accretion flow in the funnel should depend on the physics that controls the flow in the torus. In particular, the introduction of energy dissipation, and the transport of energy and angular momentum in the torus, may change the shape of the torus and its effect on the polar regions, where material can accrete without the transport of angular momentum. One could argue that the total mass accretion rate (via both the funnel and torus) onto the black hole should increase when transport of  angular momentum and consequently accretion via the torus are allowed. However, it is not clear by how much $\\MDOT_a$ will increase, if at all, because the energy and angular momentum from the torus are likely to be deposited in the polar region (see e.g., Stone, Pringle \\& Begelman 1999; Blandford \\& Begelman 1999; Blandford \\& Begelman 2002a, 2002b; Hawley \\& Balbus 2002). Indeed, outflow from the torus may interfere with the inflow in the funnel and $\\MDOT_a$ may  well {\\it decrease}. There is also a possibility that accretion via the funnel will decrease, not because of the dynamical effects due to outflowing material but because a torus in which energy is dissipated should be hotter and  thicker than a torus in which energy cannot be dissipated. On the other hand, it is not clear how accretion via the torus will change when supersonic accretion in the funnel is present. Our simulations show that the range of $\\theta$ that is occupied by the accretion funnel increases with decreasing distance from the black hole. In particular, accretion through the inner radius occurs for the entire range of $\\theta$. We expect that the presence of a supersonic accretion flow near the equator at small radii may cause accretion via the torus to be less effective compared to the situation in which the material in the polar region is static. Most simulations of the formation of the torus have assumed that the material near the poles, outside the initially hydrostatic rotating torus, is static or subsonic, and therefore unimportant dynamically.  Even within our simple framework of inviscid flow, we anticipate that our results may change if we relax some of our assumptions about the geometry at the outer boundary. The assumptions we adopted are as simple as possible because we do not know the magnitude of the angular momentum of  material in the environment of SMBHs, nor the angular distribution of $l$. In particular, it is rather unlikely that the $l$ distribution is axisymmetric. Therefore, fully 3D calculations are required to explore how $\\MDOT_a$ responds to non-axisymmetric $l$ distributions. In the context of inviscid HD calculations similar to ours, one would expect that 3D effects may reduce $\\MDOT_a$ compared to 2D axisymmetric calculations. For example, it is possible that  rotation at large radii occurs not just around one axis but around two or more axes. It is also possible that  rotation occurs around an axis that changes with $r$. Additionally it is plausible that there are ``winds'' flowing past the SMBH or that the entropy in the SMBH environment is variable. In such a case,  material with too high $l$ to be accreted may occupy nearly the entire range of $\\theta$ at small radii (near or inside the sonic radius) and prevent low-$l$ material from accreting. In terms of accretion funnels, this would correspond to the situation in which the `funnel' is very narrow and maybe not lined up with a radial vector.  Perhaps the best studied  massive black hole with a very low luminosity is the black hole in the Galactic center. Many models and ideas for how to explain very low SMBH luminosities  have been explored in the context of  Sgr~$\\rm{A^\\ast}$. In particular, Melia (1992; 1994) proposed a spherical accretion model in which the accretion flow is assumed to be in free-fall until a Keplerian disk is formed within a small circularization radius. Coker \\& Melia (2000) looked at the problem of spherical accretion but with the magnetic field being in subequipartition. The latter results in a reduced bremsstrahlung emissivity and can help to explain the low luminosity of Sgr~$\\rm{A^\\ast}$. A relatively small distance to Sgr~$\\rm{A^\\ast}$ allows us to map the vicinity of the black hole at the Galactic center. The complexity of such maps (e.g., in radio) motivate three dimensional HD simulations. For example, Coker \\& Melia (1997) performed three-dimensional simulations of Bondi-Hoyle accretion of  stellar winds onto a black hole. Clearly Sgr~$\\rm{A^\\ast}$ shows us that accretion onto black holes is a complex phenomenon and the Bondi accretion formula should be used with great caution because the assumption of spherical accretion is most certainly an oversimplification.  We finish with the observation that after three decades of studying the Bondi problem with slight rotation at large radii (see, e.g., Henriksen \\& Heaton 1975, Lynden-Bell 1978; Cassen \\& Pettibone 1976; Sparke \\& Shu 1980; Sparke 1982; Abramowicz \\& Zurek 1981, for analytic attempts to solve this problem and see references in Section~1 for 2D and 3D numerical simulations), we still find new complexities in the behavior of  accretion flows. In our next phase of studying these flows, we will consider 3D MHD models  of radiatively inefficient flows.   ACKNOWLEDGMENTS: We thank J.M. Stone for useful discussions. We acknowledge partial support from NSF grant AST-9876887. DP also acknowledges partial support from NASA grant NAG5-11736. Some computations were  performed at Imperial College Parallel Computing Center.   \\newpage"
    },
    "0208/astro-ph0208585_arXiv.txt": {
        "abstract": "The thermometer and thermal control system, for the Absolute Radiometer for Cosmology, Astrophysics, and Diffuse Emission (ARCADE) experiment, is described, including the design, testing, and results from the first flight of ARCADE. The noise is equivalent to about 1~$\\Omega$ or 0.15 mK in a second for the RuO$_2$ resistive thermometers at 2.7~K. The average power dissipation in each thermometer is 1~nW. The control system can take full advantage of the thermometers to maintain stable temperatures. Systematic effects are still under investigation, but the measured precision and accuracy are sufficient to allow measurement of the cosmic background spectrum. ",
        "introduction": "The Absolute Radiometer for Cosmology, Astrophysics, and Diffuse Emission (ARCADE) experiment$^1$ is designed to detect or limit spectral distortions in the Rayleigh-Jeans tail of the Cosmic Microwave Background. The key to this experiment is the external calibrator. The absolute temperature of the calibrator is required to compare to other experiments such as Far InfraRed Absolute Spectrophotometer (FIRAS)$^{2,3}$. It is more critical, however, that the calibrator be isothermal and remain at a constant temperature while it is shifted between the various radiometers of ARCADE. The other parts of the ARCADE instrument (loads, horns, switches etc) must remain at a stable temperature, but the absolute accuracy requirement is not severe because the external calibrator will calibrate all of these terms to first order. The measurement and thermal control must be performed at high altitudes ($\\sim30$~km) while the instrument is suspended from a balloon.  In order to eliminate the reflections and emission of a window the ARCADE radiometers run without a window. The target, similar in many respects to that of FIRAS, is moved from one horn to the other to provide an external blackbody reference to compare to the sky. A blackbody internal reference reduces the dynamic range of the amplifier signal. With this arrangement, it is important to maintain the temperature of critical components in the radiometer such as the target, the internal reference, the amplifier and the horn antenna while the target is moved from one radiometer to another.  Cryogenic thermometers are often used in applications where low noise is desired and low power is required. Low level signals from the thermometers are susceptible to noise pick-up and degradation by the capacitance of long lines. Long term measurement stability is required, so it is desirable to have a thermometer system with built in calibration.  Most of the ARCADE thermometers are used to maintain thermal control of the instrument; low noise and stability are more important than knowledge of the absolute temperature. However, key thermometers are imbedded within the microwave absorber of the external calibration target. The ARCADE science goals require an isothermal target, which in turn requires precise cross-calibration for the target thermometers. ",
        "conclusions": ""
    },
    "0208/astro-ph0208066_arXiv.txt": {
        "abstract": "The black-hole/accretion-disk paradigm for active galactic nuclei (AGNs) is now reasonably secure, but there are still important unresolved issues, some of which will require the capabilities of an 8 to 10-m class UV/optical space-based telescope. Imaging spectroscopy with a diffraction-limited large telescope will be required to measure AGN black-hole masses from stellar dynamics for direct comparison with reverberation mapping-based masses.  High spectral resolution in the UV is required to determine the mass and kinetic energy of the outflows observed in the absorption spectra of AGNs and to understand the energetics of the accretion process. As with ground-based astronomy, however, effective use of a large UV/optical space telescope requires complementary smaller facility instruments; a meter-class UV spectroscopic telescope, for example, can fit into a Medium Explorer budget. ",
        "introduction": "Not long after the discovery of quasars, it was realized that the fundamental source of energy for these objects must in fact be gravitation, and fairly straightforward arguments led to the long-standing paradigm of a supermassive black hole (SMBH) surrounded by an accretion disk. Observational evidence has been rather ambiguous, although there were early strong clues, such as the near-UV/optical ``big blue bump'' (Shields 1978; Malkan \\& Sargent 1982) and the rapid X-ray variability, that supported the model if only because they defied other explanation. But it is only within the last several years that the circumstantial evidence has accumulated to the point that few doubters remain. While there is now general, though not unanimous, agreement about the fundamental nature of AGNs, we cannot claim any real understanding of the quasar phenomenon until we successfully address a number of key questions, including the following: \\begin{enumerate} \\item {\\em What are the masses of AGN black holes?} As described below, we are making progress, but there are still areas where our understanding is dangerously incomplete, especially with regard to the magnitude of possible systematic errors. \\item {\\em What are the energetics of the accretion process?} In particular, for the various types of AGNs what are the accretion rates and radiative efficiencies and how do these scale with luminosity? How much of the output is in the form of kinetic energy (e.g., jets and absorbing gas) as opposed to radiation? \\item {\\em How does the AGN mass function evolve over time?} Does the accretion process contribute significantly to black-hole growth, and how do black-hole demographics evolve with time? \\item {\\em What is the nature of the line-emitting and absorbing gas in AGNs?} There is good reason to believe that these are somehow related to the accretion process, but there is no standard paradigm for the origin and role of these components in AGNs. This is one of the remaining outstanding mysteries in AGN structure. \\end{enumerate} ",
        "conclusions": "The most dramatic impact that an 8-m space-based UV/optical telescope would have on AGNs would be to enable stellar-dynamical mass measurements out to a distance of $\\sim 100$\\,Mpc, a volume large enough to include several AGNs with reverberation-based mass measurements. The increase in collecting area relative to \\HST\\ will enable far more detailed studies of the poorly understood, but energetically important, massive outflows seen in AGN spectra. The order-of-magnitude improvement in spatial resolution afforded by \\HST\\ relative to ground-based observations has provided us with a wealth of information on the inner structure of AGNs (e.g., Pogge \\& Martini 2002) and on the evolution of AGN host galaxies. Certainly, another factor of a few improvement in resolution and the much larger collecting area will allow us push these frontiers forward."
    },
    "0208/astro-ph0208250_arXiv.txt": {
        "abstract": "{\\small The detection of a massive jet-ejection event from SS~433 with RXTE is reported.   SS~433 in its high state has been monitored with RXTE from 2001/11/09 (MJD = 52222) to 2001/11/25 (MJD = 52238), following a radio flare on 2001/11/02 (MJD = 52215).  An irregular temporal variation with time scales of $10^2-10^3$ s appears in the light curve, and the amplitude increases day by day.  This is the first detection of such a fast variation from the source. In addition to the fast variations, the daily light curve scatters with a time scale of $\\sim$day from 2001/11/17 (MJD = 52230).  Following the scatter, another radio flare has been detected on 2001/11/22 (MJD = 52235), which has been obviously formed during the X-ray scatter.  This is a preliminary report on a massive jet-ejection event witnessed in X-ray band for the first time. } ",
        "introduction": "The famous microquasar SS~433 shows two distinctive states; the quiescent state in which the continuous jet flow is emanated, and the high state in which massive jet blobs are successively ejected\\cite{fiedler87}.  While the former state has been well studied with numerous X-ray missions, few massive jet-ejection events were observed in X-ray band, except for a possible snapshot or two taken with {\\it Einstein}\\cite{band89} and RXTE \\cite{safi-harb02,band02}. Because the ejection of a massive jet blob is a rare (2.6 yr$^{-1}$) and short (a few days) event, it is difficult to observe with an X-ray mission unless the observation is specially coordinated for that purpose.  The situation is same for other microquasars such as GRS~1915+105. While minor jet-ejection events in GRS~1915+105 have been observed with RXTE in multi-wavelength campaigns and several other occasions (e.g., \\cite{mirabel98,greiner96}), there are few X-ray observations of a massive ejection event like the one reported in the famous paper by Mirabel \\& Rodir\\'{\\i}guez\\cite{mirabel94}.  X-ray data of a massive jet-ejection event in microquasars are needed to fill the void in our understanding.  With such data, we can measure the mass of the massive blob, determine the energy budget of the system, etc.  We planed TOO monitoring observations of SS~433 with RXTE\\@.  In the plan, a long-term monitoring is triggered with a radio flare, which indicates that the source enters its high state.  In the high state, the source is expected to experience a second flare within 30 days.  If the second flare occurs, we can observe it from the onset.  Our plan is different from the common TOO strategy for transient sources aiming at the first flare.  The formation of a massive jet blob would not be observed with an X-ray pointing observation after the detection of the flare.  SS~433's radio activity is monitored with the RATAN-600 radio telescope, which performs such a several-month-long monitoring observation of SS~433 occasionally. ",
        "conclusions": "The X-ray emitting part of the continuous jet flow with a speed of 0.26 c was estimated to be as long as $\\sim 10^{13}$ cm, and the cooling time of the X-ray emitting plasma as short as $\\sim 10^3$ s~\\cite{kotani97}. According to this picture, an X-ray variation faster than $10^3$ s is difficult to detect.  So, the discovery of the fast variation clearly seen in figure~\\ref{fig:blowup} is surprising, and our understanding of the system must be somehow changed.  The fast variation suggests that either the X-ray emitting part of the jet was as short as $\\sim 10^{12}$ cm, or other source than the jet, e.g., the inner part of the accretion disk or the surface of the compact object was seen.  To distinguish the two possibilities, a temporal analysis of the Doppler-shifted iron lines is quite promising.  If the jet itself was flickering, the iron line would also show flickering, and if the variation was caused by some instability of the accretion disk, the iron line would be stable through the variation.  This report is preliminary and spectrally-resolved temporal study is not yet  done.  The data taken on 2001/11/19 has been divided into two, data with count rate larger than 220 counts s$^{-1}$ and with count rate less than 220 counts s$^{-1}$, and a spectrum has been made from each data.  The fit results are not different significantly.  So far, there is no evidence of an X-ray source other than the jet.  And from the quick-look analysis, it can be said that the fast variation is not periodic; no pulsation or QPO has been found from the power spectra.  We can not determine whether the variation comes from the accretion disk or a neutron star at this stage of analysis.  If we assume that the fast variation is resulted from the sudden change of the power of the jet, the parameters of the jet would be derived. The shortest time scale of the fast variation seems to be $\\sim$50 s, i.e., $4\\times10^{11}$ cm, which corresponds to an initial electron density of $10^{14}$ cm$^{-3}$, assuming an initial temperature of 20 keV~\\cite{kotani97}.  Assuming a X-ray luminosity of $6\\times10^{35}$ erg s$^{-1}$, the mass outflow rate of the jet would be $10^{-6}$ M$_\\odot$ yr$^{-1}$, or a kinematic luminosity of $10^{38}$ erg s$^{-1}$.  It should be noted that these estimated parameters are rather smaller than those based on the data taken in the quiescent state.  Is the jet weaker in the high state than in the quiescent state?  Or the previous estimations are simply wrong and the jet is always short and weak? Further analysis is going on."
    },
    "0208/astro-ph0208120_arXiv.txt": {
        "abstract": "{We present a detailed calculation of the mechanism by which the Accretion-Ejection Instability can extract accretion energy and angular momentum from a magnetized disk, and redirect them to its corona.  In a disk threaded by a poloidal magnetic field of the order of equipartition with the gas pressure, the instability is composed of a spiral wave (analogous to galactic ones) and a Rossby vortex.  The mechanism detailed here describes how the vortex, twisting the footpoints of field lines threading the disk, generates Alfv\\'en waves propagating to the corona.  We find that this is a very efficient mechanism, providing to the corona (where it could feed a jet or a wind) a substantial fraction of the accretion energy. ",
        "introduction": "\\label{sec:Intro} MHD models have shown that jets can be very efficient to carry away the accretion energy and angular momentum extracted by turbulence from accretion disks (Blandford and Payne, 1982; Lovelace {\\em et al.}, 1987; Pelletier and Pudritz, 1992), if the disk is threaded by a poloidal magnetic field.  This fits with the observation that accretion and ejection are intimately connected in objects ranging from protostellar disks to X-ray binaries and AGNs.  However these models, based on self-similar analytical computations or on numerical simulations, most often start at the upper surface of the disk.  Although more recent works (Ferreira and Pelletier, 1993a, 1993b, 1995; Casse and Ferreira, 2000) find solutions connecting continuously the disk and the jet, these solutions are heavily constrained by conflicting requirements.  These can be traced, in good part, to the fact that disk models, whether they rely on specific instability mechanisms or on the assumption of a turbulent viscosity, imply that the accretion energy and angular momentum are transported {\\em radially outward}.  They must thus somehow be redirected {\\em upward} to feed the jet.  The Accretion-Ejection Instability (AEI) of magnetized accretion disks, presented by Tagger and Pellat (1999, hereafter TP99), could provide a solution to this difficulty.  It occurs in the inner region of the disk, in the configuration assumed by the MHD models of jets (\\ie a disk threaded by a poloidal field of the order of equipartition with the gas thermal pressure), and it has the unique property that energy and momentum extracted from the disk can be emitted {\\em vertically} as Alfv\\'en waves propagating along magnetic field lines to the corona of the disk.  Thus they could provide a source for a jet or a wind formed in the corona.   \\correct{This ability to emit the energy and angular momentum as Alfven wave} was recognized in TP99, and indeed justified the name given to the instability.  However this possibility % was shown only in a WKB approximation, valid away from the region (the corotation radius, where the wave rotates at the same velocity as the gas) where Alfv\\'en wave emission is most efficient.  The WKB result was found divergent at corotation, providing a good indication that this mechanism of vertical emission could be quite efficient.    The goal of this paper is to present a more general derivation, valid in the corotation region.  We use a description of the waves in three dimensions (whereas TP99 was basically a model averaged over the disk thickness).  We are thus able to give an explicit computation of the Alfv\\'en wave emission mechanism and of its efficiency.  The main unknown to be solved for at this stage is the fraction of the accretion energy and angular momentum, extracted from the disk by the instability, which will end up emitted to the corona.  The result we present is quite limited: the constraints of giving an analytical derivation force us to use a very simplified magnetic field geometry\\correct{, namely an initially constant and vertical field,} and a more realistic one would certainly affect the result. \\correct{Appendix \\ref{an:Br} is dedicated to the case of a radially varying $B_z$ field. We show that the present computation may be apply in the case of a slowly variable $B$ field.} On the other hand, the result we obtain is interesting in itself: we will show that, in linear theory, the flux of the Alfv\\'en waves is again divergent at corotation.  Although we discuss how it can be regularized, our interpretation will be that the efficiency of the mechanism is indeed quite high, but that we are reaching the limits of linear theory and that the true result will most certainly be determined by self-consistent non-linear effects.  The present work should thus be viewed as an exploration of the complex physics involved and of its potential efficiency, which will then have to be treated by non-linear simulations. \\begin{figure*}[htbp] \\centering \\epsfig{file=ms2259.fig1.eps,width=\\textwidth} \\caption{The structure of the instability is shown here schematically as a function of radius.  It is formed of a standing spiral density wave in the inner part of the disk, coupled to a Rossby vortex it excites at its corotation radius.  The spiral grows by extracting energy and angular momentum from the disk, and depositing them in the Rossby vortex; the latter in turn generates Alfv\\'en waves propagating toward the corona of the disk.} \\label{fig:cavite} \\end{figure*} The paper is structured as follows: in the next section we will briefly review the main properties of the instability, and its interest to explain the low-frequency Quasi-Periodic Oscillation (QPO) of X-ray binaries.  Section \\ref{sec:par} will present the system of equations to be solved, and section \\ref{sec:sys} their combination into a variational form containing the physics of the problem.  In section \\ref{sec:kz} we will compute the Alfv\\'en wave flux, and we will discuss the significance of this result in section \\ref{sec:conc}. \\correct{In appendix \\ref{an:Br} we will present the variational form we obtain in a more general geometry with a radially varying $B_z$ and the limitation it implies.} ",
        "conclusions": "\\label{sec:conc}  We have presented a computation of the flux of Alfv\\'en waves emitted to the corona of a magnetized disk by the Accretion-Ejection Instability. This means that we have justified here the name chosen by TP99: the instability is a spiral density wave, which grows by extracting energy and angular momentum from the disk (thus causing accretion) and transferring them \\emph{radially outward} to the Rossby vortex at corotation; a significant fraction, given by equation (\\ref{eq:fluxratio}), of this flux is then transmitted \\emph{vertically} to the corona as Alfv\\'en waves.  We expect that, if the Alfv\\'en waves can deposit their energy and momentum in the corona, this would be an ideal mechanism to feed a wind or jet directly from the accretion process in the disk.  The amplification of the wave (and thus the flux deposited by the spiral in the vortex) and the flux transmitted to Alfv\\'en waves are both linked to the singularity of the vortex.  This allows us to give in a very simple form a result of paramount importance in the physics of accretion disks and jets: an estimate of the fraction of the accretion energy, extracted from the inner region of the disk, which will end up in the corona where it might feed a jet.  This fraction is of the order of unity if the coronal density is not too low (typically $10^{-4}$ of the density in the disk would be sufficient, in an X-ray binary).  In order to obtain analytically a physically consistent result, we have had to use a very artificial configuration of the equilibrium magnetic field, vertical and independent of $r$.  On the other hand this has allowed us, proceeding rigorously by perturbation of a variational form, to obtain an exact result clarifying the role and the physical nature of the singularity of the Rossby vortex at corotation. We can thus expect that these results would survive less stringent assumptions on the equilibrium configuration.  However this must be taken carefully: our final result is in fact divergent at the corotation radius, and regularized by the effect of the finite thickness of the disk, or by the growth rate of the instability.  In both cases, it depends on the density in the corona of the disk. Thus we believe that the end result should be obtained from a self-consistent, non-linear description where the growth of the instability itself affects the evolution of the magnetic geometry and the mass loading of the corona.  In this respect it is worth mentioning one of the results of stationary MHD jets models: in these models, as the gas is accelerated along the field lines it passes a slow magnetosonic point where the field lines bend outwards.  Magnetocentrifugal acceleration can then proceed and leads to the formation, higher up and further out, of an alfv\\'enic point.  The slow magnetosonic point is thus associated with the mass loading of the field lines, and the alfv\\'enic point to the acceleration.  By analogy we can thus expect that, while the Alfv\\'en waves described in the present work allow to accelerate the gas, the instability can also generate slow magnetosonic waves which will lift the gas above the disk.  The coupling of the instability to the slow wave will be the object of a forthcoming paper.  \\appendix"
    },
    "0208/astro-ph0208316_arXiv.txt": {
        "abstract": "Deep, wide-range and accurate UBVRI photometric sequences have been established around more than 80 symbiotic stars, to assist current photometry as well as measurement of old photographic plates. Sequences for 40 symbiotic stars have already been published; the observations for the others have already been secured. ",
        "introduction": "Availability of suitable UBVRI photometric sequences around symbiotic stars is an essential step in promoting a large-scale documentation effort of present time multi-band lightcurves as well as reconstruction of the historical behavior.  To serve for measuring archival plates, the sequence stars have to be well isolated from neighbors to avoid blending on short focal length patrol plates, have to be grouped close to the symbiotic variable so as to fall within the same eyepiece field of view, and must extend over a wide range in brightness to easily cover both faint states in quiescence (like eclipses) as well as bright outbursts.  To assist present time CCD photometry (carried out mostly by amateurs with telescopes of 1-2 m focal lengths), the sequences must also cover a wide range in colors (to allow simultaneous derivation of transformation coefficients), be compact enough to easily fit onto a small CCD ($\\la$5 arcmin in diameter) and be accurately placed on the UBVRI system.  The UBVRI sequences on the Johnson-Cousins system that we have calibrated around more than 80 symbiotic stars follow the above prescriptions and are discussed in this paper. Their use on Asiago archive plates is described by Jurdana-Sepic and Munari (2002, and references therein). ",
        "conclusions": ""
    },
    "0208/astro-ph0208193_arXiv.txt": {
        "abstract": "High spectral resolution Fabry-P\\'erot observations of the \\OI 63.2 and 145.5~$\\mu$m and \\CII 157.7 $\\mu$m fine structure lines are presented for the center of the Sagittarius B2 complex (Sgr~B2). The data were obtained with the Long Wavelength Spectrometer on board the Infrared Space Observatory (ISO). Both the \\OI 63.2~$\\mu$m and the \\CII 157.7~$\\mu$m lines are detected in absorption. The upper state level of atomic oxygen at 145.5~$\\mu$m is in emission. Whereas the \\OI 63.2~$\\mu$m line is seen in absorption over the entire wavelength range $-200$ to 100~km~s$^{-1}$, the \\CII 157.7~$\\mu$m line displays a more complex profile: absorption occurs at velocities $<$ 20~km~s$^{-1}$ and emission comes from the Sgr~B2 complex at velocities greater than 20~km~s$^{-1}$. Using observations of the CO isotopes and of the \\HI lines, absorption components can be associated with many clouds along the Sagittarius B2 line of sight. From these data, we were able to disentangle three different layers which contain atomic oxygen. These layers, as predicted by PDR models, are characterized by different forms of carbon in the gas phase, i.e. the C$^+$ external layer, the C$^+$ to C$^0$ transition and the CO internal layer. We derive lower limits for the column densities of atomic carbon and oxygen of the order of $\\sim$ 10$^{18}$~cm$^{-2}$ and 3 $\\times$ 10$^{19}$~cm$^{-2}$, respectively. An O$^0$/CO ratio of around 2.5 is computed in the internal cores of the clouds lying along the line of sight, which means that $\\sim$ 70\\% of gaseous oxygen is in the atomic form and not locked into CO. The fact that the \\CII 157.7 $\\mu$m line is detected in absorption implies that the main cooling line of the interstellar medium can be optically thick especially in the direction of large star-forming complexes or in the nuclei of galaxies. This could partially account for the  deficiency in the \\CII 157.7 $\\mu$m line which has been recently found toward infrared bright galaxies in ISO data. ",
        "introduction": "Observations have suggested that, in some molecular clouds, most of the gas-phase oxygen might be in atomic form, in contradiction with predictions of steady state chemical models (e.g., Lee, Bettens \\& Herbst 1996) which yield CO, O, and O$_2$ as the major oxygen bearing species in molecular clouds. First suggestions of a high atomic oxygen abundance were made by Jacq et al. (1990) and Schulz et al. (1991) in order to interpret their HDO observations of hot cores and quiescent clouds. These results are in accord with models which take into account cosmic-ray induced photo-dissociation (e.g., Jacq et al. 1990; Wannier et al. 1991). Further evidence has been obtained from observations of the \\OI 63.2~$\\mu$m fine structure line which has been detected in absorption against the far-infrared continuum of bright galactic sources, namely: DR~21 (Poglitsch et al. 1996), Sgr~B2 (Baluteau et al. 1997) and NGC~6334V (Kraemer et al. 1998). These observations indicate that the absorbing material is predominantly in foreground cloud(s) where the abundance of the atomic oxygen is within a factor of a few of the cosmic abundance.  Recent studies have strengthened the above results. Based on ISO-LWS Fabry-P\\'erot data, Vastel et al. (2000) have modelled the \\OI 63.2~$\\mu$m line absorption components in the direction of the compact \\HII region W49~N. Combining these observations with molecular (CO and its isotopes) and \\HI observations, they showed that both molecular and atomic clouds absorb the strong continuum at 63~$\\mu$m, and disentangled the absorption due to the molecular clouds from the absorption due to the atomic (\\HI) clouds. They concluded that the major part of the 63.2~$\\mu$m \\OI absorption is due to the cold molecular clouds along the line of sight and, through the computation of O$^0$/CO ratio, that in these clouds the gaseous oxygen is almost totally in atomic form.  Similar results were obtained toward Sgr~B2 by Lis et al. (2001). They found three \\OI 63.2~$\\mu$m absorption components corresponding to foreground clouds, for which the oxygen content of the atomic halo gas could be estimated based on \\HI observations. Lis et al. (2001) found that the remaining O$^0$ column density is correlated with the observed $^{13}$CO column density, corresponding to an average O$^0$/CO ratio of about 9 and to an atomic oxygen abundance of 2.7 $\\times$ 10$^{-4}$ in the dense gas phase.  The full Fabry-P\\'erot ISO-LWS spectrum which was obtained on Sgr~B2 as part of the ISO Central Program (see Baluteau et al. {\\it in preparation}, for a detailed description) enables us to study, at high spectral resolution, the atomic fine-structure of oxygen and carbon in this source and to analyze in a consistent way the profiles of the \\OI lines at 63.2 and 145.5~$\\mu$m and of the \\CII 157.7~$\\mu$m line. Sgr~B2 is the most massive star-forming region of an ensemble of dense cloud cores in the central ($\\approx \\, 500$~pc) region of the Galaxy. Its estimated mass is larger than $\\rm 5 \\times 10^6 \\, M_{\\odot}$ (Lis, Carlstrom \\& Keene 1991), and the high opacities found toward Srg~B2 makes it one of the best candidate for absorption studies. Sgr~B2 is located at about 8.5~kpc from the Sun (adopting the IAU distance) and has a projected distance of $\\rm \\sim 100 \\, pc$ from the Galactic Center. Ground-based spectroscopic observations of Sgr~B2 have shown a very complex pattern of absorption features, with numerous components of foreground gas associated with clouds along the line of sight, covering a wide range of velocities (see Sect.~3.3).  In this paper we present the observational data of the three main far-infrared cooling lines and of the isotopic CO lines in Sect.~2. The absorption lines arising from foreground clouds, which have no physical connection with Sgr~B2, are modeled in Sect.~3, where we try to disentangle the absorption due to the molecular cores from that due to the external layers of these clouds. We discuss these results and summarize them in Sect.~4 and 5 respectively. ",
        "conclusions": "Using \\CII 157.7 $\\mu m$, \\OI 63.2 and 145.5 $\\mu m$ line observations, we were able to distinguish between the contributions of the different layers within the galactic clouds along the line of sight to Sgr~B2. We separate the layers of the atomic diffuse surface and of the molecular core of these clouds. This is a major improvement over the previous analysis of this line of sight by Lis et al. (2001), which was based on observations of the \\OI 63 $\\mu$m line only. We were able to associate atomic oxygen with three layers through the clouds characterized by different forms of carbon in the gas phase, as predicted by standard PDR models: i.e. the major form of carbon changing from C$^+$ to C$^0$ and finally to CO.  From the line shape modelling presented here, a total column density of atomic oxygen in the line of sight to Sgr B2 of about 3.1 $\\times$ 10$^{19}$ cm$^{-2}$ is derived within the clouds with velocities between $-$120 km~s$^{-1}$ and +10 km~s$^{-1}$. Less than 30 \\% of this total O$^0$ column density is found to be due to the external layers of the clouds (where C$^+$ is the major form of carbon).  The method used in this study leads to an estimate of the oxygen content in the intermediate layer where C$^0$ is the dominant form of carbon. The atomic carbon column densities derived here for the galactic clouds, with a mean value about 2.4 $\\times$ 10$^{17}$ cm$^{-2}$, are in good agreement with recent observations of the \\CI 492 GHz line. The derived C$^0$/CO ratios are indicative of clouds, ranging from diffuse to dense, along the line of sight of the Sgr B2 complex which have been fragmented and illuminated by the galactic interstellar radiation field. C$^+$/C$^0$ ratios close to 2.5 are derived for the clouds at galactocentric distances of 3 - 4 kpc, and ratios less than 0.7 for clouds in the galactic center region.\\\\ The method used to disentangle the different layers of the clouds in the line of sight leads to the accurate computation of the O$^0$/CO ratio ($\\sim$ 2.5) in the internal layers. Therefore, about 70\\% of gaseous oxygen is in the atomic form and not locked into CO in the molecular clouds along the Sgr~B2 line of sight.   \\\\ Future instrumentation will enable the present analysis to be improved. The Herschel project, with its high spectral resolution (HIFI) capability, will allow to better characterize the physical conditions of these clouds along the line of sight to Sgr B2 (and maybe on other lines of sight in direction of the galactic center as well), in particular from the C$^+$ and C$^0$ lines and from high J transitions of CO. SMA and ALMA will provide high spatial resolution C$^0$ observations of these clouds necessary to separate each of them. At high spectral resolution, the fundamental transitions of atomic oxygen will only be accessible to the second instrument generation on board the SOFIA observatory."
    },
    "0208/astro-ph0208470_arXiv.txt": {
        "abstract": "The spectral evolution of the peculiar SN~Ic 2002ap during the first 40 days is presented.  The spectra display very broad absorption features, which are typical of \"hypernovae\".  The maximum expansion velocity measured on the earliest spectra exceeds $ 3 \\times 10^4 $ km s$^{-1}$.  The spectrum of SN~2002ap at the epoch of maximum brightness resembles that of SN~1997ef more than that of SN~1998bw. The spectral evolution of SN~2002ap proceeds at about 1.5 times the rate of SN~1997ef.  The parameterized supernova spectrum synthesis code SYNOW was used to perform line identification and deduce velocity information from the early-phase spectra, which are heavily affected by line blending. The photospheric velocity, as deduced from the fitting results and from the blueshift of the \\ion{Si}{2} $\\lambda$6355 absorption minimum, is lower than in previously studied hypernovae. At advanced epochs, the \\ion{Si}{2} $\\lambda$6355 absorption minimum becomes difficult to distinguish. This may be caused by the growth of [\\ion{O}{1}] $\\lambda\\lambda$6300, 6364 emission. Together with the rapid spectral evolution, this suggests that SN~2002ap should enter the nebular phase sooner than previously studied hypernovae. ",
        "introduction": "The very powerful supernova (SN) explosions called ``hypernovae'' \\citep{IWA1998} are perhaps the most energetic events in the current universe. Their kinetic energy is $\\sim 10^{52}$ ergs, which is about 10 times larger than of normal SNe. SN~1998bw, probably associated with GRB 980425 \\citep{GAL1998}, was the first recognized hypernova. Although classified as a Type Ic SN (SN~Ic), the spectrum of SN~1998bw showed unusual broad spectral features in the early phase. The derived velocities were as large as 30,000 km s$^{-1}$. The kinetic energy of SN~1998bw was deduced to be 3 -- 5 $\\times$ 10$^{52}$ ergs \\citep{IWA1998,NAK2001}. Other Type Ic hypernovae were subsequently recognized based on their broad absorption features.  These include SNe~1997ef (e.g., Iwamoto et al.\\ 2000), 1998ey \\citep{GAR1998}, and the recently discovered 2002bl \\citep{FIL2002}.  Moreover, SN~1997dq was suggested to be a hypernova from its similarity to SN~1997ef at later phases \\citep{MATH2001}. The kinetic energy deduced for SN~1997ef was about $1.8\\times10^{52}$ ergs (Mazzali, Iwamoto, \\& Nomoto 2000), which was smaller than that of SN~1998bw, but still significantly larger than normal SNe. Hypernovae also occur among SNe~IIn, for example, SN~1997cy, one of the brightest SNe ever discovered and characterized by a large kinetic energy, $\\sim 3\\times 10^{52}$ ergs \\citep{TUR2000}, and SN~1999E (Cappellaro, Turatto, \\& Mazzali 1999). The range of properties of hypernovae, and their possible association with GRBs are also topics of great interest (e.g., Nomoto et al. 2001).  SN~2002ap was discovered in M74 on January 29.4 UT by Y.Hirose \\citep{NAK2002}. Spectra obtained on January 30 -- 31 showed very broad absorption features, resembling those in SNe~1998bw and 1997ef. Therefore, SN~2002ap was suggested to be a hypernova \\citep{KIN2002,MEI2002,GAL2002a}. SN~2002ap is the nearest hypernova discovered to date, hence it was a good target even for small telescopes. Various multi-wavelength observations of SN~2002ap have been performed (summarized in Gal-Yam, Ofek, \\& Shemmer 2002). NIR photometry was also performed frequently at Gunma Astronomical Observatory (GAO) (E.\\ Nishihara et al.\\ 2002, in preparation).  We report here on the optical spectroscopy of SN~2002ap performed at GAO. We compare the spectra of SN~2002ap with those of other hypernovae. We also used the parameterized SN synthetic-spectrum code SYNOW to perform line identification and deduce velocity information from the spectra in the early phase, when very broad features resulting from line-blending were present. ",
        "conclusions": "The time-sequence of the calibrated spectra of SN~2002ap is shown in Figure 1. The evolution of the spectra is apparent. As time progresses, most absorption features shift to the red, and the continuum becomes redder.  \\placefigure{fig1}  The first two spectra, taken on the same night, show a blue continuum, with neither deep absorptions nor strong emissions (see also the top of Figure 2). Broad and shallow depressions are seen near 4700, 5700, and 6200 \\AA. These broad features resemble those of SNe~1998bw and 1997ef in the earliest phases \\citep{PAT2001,IWA2000}.  However, SN~2002ap is bluer, possibly because it was caught at an earlier epoch. The fact that the first observations must have been very close to the time of explosion is exemplified by the rapid development of spectral features. The first two spectra also show a broad peak near 5000\\AA. But in the third spectrum taken the following night, this peak has shifted significantly to the red ($\\sim 5100$\\AA), indicating a rapid decrease of the material velocity sampled by the spectra, as the photosphere moves inwards.  Photometric observation at Wise Observatory established that the peak of the light curve occurred on February 7.1 and 8.8 UT in the $B$ and $V$-band, respectively \\citep{GAL2002b}. We designate the epochs using the number of days from $B$ maximum. Figure 2 also shows the spectrum of SN~2002ap taken on day $+$2 (February 9.4 UT), near the time of maximum. We compared this with the near-maximum spectra of other SNe~Ic: SNe~1998bw, 1997ef (hypernovae) and 1994I (a normal SN~Ic; Nomoto et al. 1994). Based on the line width, the spectra can easily be separated into two groups. One is that of ``hypernovae'', including SNe~1998bw, 1997ef, and 2002ap, while the other includes only the normal Type Ic SN~1994I. Therefore, SN~2002ap is easily identified as a hypernova. Moreover, the spectrum of SN~2002ap shows emission-like features around 4600 \\AA\\ and 5300 \\AA, resembling the spectrum of SN~1997ef more than that of SN~1998bw.  \\placefigure{fig2}  In Figure 3 we compare similar-looking spectra of SNe~2002ap and 1997ef. We can see that the spectral evolution of SN~2002ap is about 1.5 times faster. The earliest available spectrum of SN~1997ef, on day  $-$6, appears to have properties intermediate between the spectrum of SN~2002ap on day $-$1 and that on day $-$6, judging from the progressive redshifting of line features. The near-maximum spectrum of SN~1997ef resembles that of SN~2002ap on day $+$2 in all major features. The similarity between the spectrum of SN~1997ef on day  $+$19 and that of SN~2002ap on day $+$15 is striking, both showing minor features of comparable intensities at about 4900, 5900, 6200, 7100 and 7300 \\AA. Most of the new features visible in the spectrum of SN~1997ef on day $+$52, some of which are probably net emissions, are already visible in SN~2002ap on day $+$25, although they are not as developed. Both of these spectra suggest that the SNe are making a transition to the nebular phase, while the spectrum of SN~1997ef on day $+$27 is still essentially photospheric in nature \\citep{MAZ2000}.  \\placefigure{fig3}  Especially in the earliest spectra, most features are so broad because of line blending that it is not easy to identify which lines contribute to them. Thus we used the parameterized SN spectrum synthesis code SYNOW, described by \\citet{FIS2000}, to perform line identification and deduce velocity information. It is simpler and faster than the Monte-Carlo code of \\citet{MAZ2002}, and better suited for our objective, as it does not require the  construction of realistic SN models. In this paper, the radial-dependent line optical depth above the photosphere is assumed to be proportional to $(v/v_{\\rm ph})^{-n}$, where $v$ is the expansion velocity and $v_{\\rm ph}$ is the photospheric velocity. A reasonable fit to the observations could be obtained using just a few ions: \\ion{Ca}{2}, \\ion{O}{1}, \\ion{Si}{2}, \\ion{Fe}{2}, \\ion{Co}{2} and \\ion{Ni}{2}, consistent with the line identification in SNe~1997ef and 1998bw \\citep{BRA2001}.  Synthetic spectra and line identification for day $-$7 and $+$2 are shown in Figure 2.  Either a high $v_{\\rm ph}$ or a small $n$ can make the lines broad. Guided by Branch's (2001) work, we found that both are necessary to account for the extreme line widths. Similar results were obtained by \\citet{MAZ2002}, who used a Monte-Carlo code to synthesize some selected spectra of SN~2002ap. Our best fit value of the power-law index was 3 for the first spectrum.  This index value was used for the radial dependence of the opacity for all lines and at all epochs. As shown in Figure 4, the photospheric velocity obtained from the fitting decrease from 35,000 $\\rm km~s^{-1}$ on day $-7$ to 13,000 $\\rm km~s^{-1}$ on day $+$2, with the uncertainty being about 1000 $\\rm km~s^{-1}$ empirically \\citep{BRA2002}.   Our fits are reasonable for these epochs, allowing us to identify most absorption lines.  However, some problems remain. One is the unwanted ``peak'' around 4100 \\AA, which is actually the remainder of  continuum. We can remove it by introducing both \\ion{Ti}{2} and  \\ion{Sc}{2} lines. However, there is no other argument for \\ion{Sc}{2} in SNe~Ic. Such difficulties may show the limitations of SYNOW. They are probably due to the assumption of a pure resonant scattering source function and on the simple power-law dependence of the optical depth on velocity applied to all ions. The hump around 5200 \\AA\\ is also incorrectly reproduced, probably the consequence of not transferring sufficient flux from the blue ($\\sim 4000$\\AA) in SYNOW. Finally, we introduced \\ion{Na}{1} to fit the absorption around 5700 \\AA. \\ion{He}{1} $\\lambda$5876 may also contributes there. The NIR spectra show a significant feature, possibly due to \\ion{He}{1} $\\lambda$10830 \\citep{MOT2002}. However, that feature in SN~1994I might have been \\ion{C}{1}, \\ion{O}{1}, and/or \\ion{Si}{1} lines. \\citep{BAR1999,MIL1999}.  Among those absorption lines identified, the \\ion{Si}{2} $\\lambda\\lambda$6347, 6371 (collectively called $\\lambda$6355) absorption minimum was easily distinguishable during almost the entire observation period. Therefore, we measured the minimum of the \\ion{Si}{2} absorption as an alternative method of estimating the photospheric velocity. In Figure 1, we marked with an arrow the minimum of the \\ion{Si}{2} absorption. The evolution of the photospheric velocity, as deduced from this method, is also shown in Figure 4. At the earliest phases, the typical statistical error was around $\\pm$ 3000 km s$^{-1}$. Moreover, strong line blending makes the estimate of the velocity somewhat uncertain at those epochs. Near maximum light, although the minima are clearly visible (the statistical error was about $\\pm$ 1500 km s$^{-1}$), the systematic error on the velocity is mostly due to the ambiguity introduced by the uncertain continuum. At late phases, the error (around $\\pm$ 2500 km s$^{-1}$) was mainly due to poor statistics. The difference between the velocities from \\ion{Si}{2} and those from SYNOW fitting may be mostly caused by systematic errors introduced by the uncertainty in the continuum and the heavy line blending.  \\placefigure{fig4}  In Figure 4, we compare the evolution of the photospheric velocity of SN~2002ap, 1998bw and 1997ef \\citep{PAT2001}. We adopted for SN~2002ap an explosion date of January 28.9 UT (day $-$9.2), as proposed by \\citet{MAZ2002}. The photospheric velocity of SN~2002ap declines faster than for the other two hypernovae, suggesting that the ejecta mass of SN~2002ap is smaller. Although the velocity evolution of the three SNe is initially well fitted with exponential functions, both SNe~1997ef and 1998bw eventually seem to flatten out, SN~1998bw at about 7000 km s$^{-1}$ and SN~1997ef at about 2000 km s$^{-1}$. While the measurement for SN~1998bw can be done directly on the spectra, the value for SN~1997ef is derived from model fitting, since, as for SN~2002ap, the \\ion{Si}{2} line becomes indistinguishable at advanced epochs (see spectra on day $+$52 for SN~1997ef and on day $+$25 for SN~2002ap in Figure 3). No sign of the flattening for SN~2002ap was seen over the period of our observations, but it might occur later, which may have to be established through detailed spectral modeling.  The rapid evolution of the photospheric velocity and of the spectra of SN~2002ap suggests that SN~2002ap will develop a nebular spectrum earlier than the other hypernovae presented here. In SN~1997ef nebular emission was observed in the \\ion{Ca}{2} IR triplet on day $+$27. Our spectra of SN~2002ap unfortunately do not extend to that region. However, the spectrum on day $+$25, which is intermediate between the day $+$27 and the day $+$52 spectra of SN~1997ef, may show the onset of weak [\\ion{O}{1}] $\\lambda\\lambda$6300, 6364 net emission. E.\\ Nishihara et al.\\ (2002, in preparation) show that a break occurs in the evolution of the IR colors around February 25, suggesting the onset of nebular emission. The first spectra of SN~2002ap after solar conjunction, taken on day 131 and 140, show that it has already entered the nebular phase \\citep{LEO2002}.  The broad spectral features, suggesting high kinetic energy, distinguish hypernovae from normal SNe~Ic. However, the observations of SN~2002ap suggest that there is a variety of hypernovae, based on the observational features (spectra, spectral evolution, luminosity, light curves, the association with a GRB, etc.). More observations of hypernovae are necessary in order to understand what physical parameters are responsible for the variations among hypernovae and whether there is a gap in properties between hypernovae and normal SNe~Ic (Figure 2 of Nomoto et al.\\ 2002)."
    },
    "0208/astro-ph0208008_arXiv.txt": {
        "abstract": "We present the discovery of GRB~020405 made with the Inter-Planetary Network (IPN).  With a duration of 60 s, the burst appears to be a typical long duration event.  We observed the 75-square acrminute IPN error region with the Mount Stromlo Observatory's 50-inch robotic telescope and discovered a transient source which subsequently decayed and was also associated with a variable radio source.  We identify this source as the afterglow of GRB\\,020405.  Subsequent observations by other groups found varying polarized flux and established a redshift of 0.690 to the host galaxy. Motivated by the low redshift we triggered observations with WFPC2 on-board the {\\it Hubble Space Telescope} (HST).  Modeling the early ground-based data with a jet model, we find a clear red excess over the decaying optical lightcurves that is present between day 10 and day 141 (the last HST epoch).  This ``bump'' has the spectral and temporal features expected of an underlying supernova (SN).  In particular, the red color of the putative SN is similar to that of the SN associated with GRB\\,011121, at late time.  Restricting the sample of GRBs to those with $z<0.7$, a total of five bursts, red bumps at late times are found in GRB\\,970228, GRB\\,011121, and GRB\\,020405. It is possible that the simplest idea, namely that all long duration GRBs have underlying SNe with a modest dispersion in their properties (especially peak luminosity), is sufficient to explain the non detections. ",
        "introduction": "\\label{sec:intro}  In recent years several indirect lines of evidence have emerged connecting the class of long duration $\\gamma$-ray bursts (GRBs) to massive stars.  Every GRB afterglow with a sub-arcsecond position is associated with a star-forming galaxy \\citep{bkd02}.  Some of these galaxies are forming stars copiously with rates of a few hundred M$_\\odot$ yr$^{-1}$ \\citep{bkf01,fbm+02}.  On smaller scales, some afterglows (the so-called ``dark bursts''), show evidence for heavy dust extinction \\citep{dfk+01,pfg+02}.  X-ray and optical observations of some GRBs indicate substantial column densities \\citep{ogo+98,gw01}.  In addition, there is evidence for moderate circumburst densities, $n\\sim 10$ cm$^{-3}$, in some bursts \\citep{hys+01,pk01,yfh+02}.  These indicators are consistent with GRBs originating in gas-rich star-forming regions (i.e.~molecular clouds).  The most direct link between GRBs and massive stars comes from observations on stellar scales, namely the detection of underlying supernovae (SNe) and X-ray spectral features.  X-ray spectral features have been observed in a few GRBs (e.g.~\\citealt{pgg+00,rwo+02}), although the detections generally have low signal-to-noise, and the interpretations are somewhat controversial.  What is generally agreed, however, is that X-ray features would require the presence of high densities of iron on stellar scales.  The discovery of the unusual Type Ic SN~1998bw \\citep{gvv+98} in a nearby ($\\sim$40~Mpc) galaxy, within the small error box of GRB~980425 \\citep{paa+00} suggested that at least some GRBs might be caused by SN explosions.  Despite the fact that GRB~980425 was under-energetic compared to the cosmological GRBs \\citep{fks+01} and may therefore represent an independent class of GRBs, the fact remains that SN~1998bw directly demonstrates that a massive star is capable of producing relativistic ejecta \\citep{kfw+98} --- an essential requirement for producing $\\gamma$-rays.  The first indication of a SN underlying a cosmological GRB came from the observation of a red excess (``bump'') in the rapidly-decaying afterglow of GRB~980326 \\citep{bkd+99}, which had a color and peak time consistent with SN~1998bw shifted to $z\\sim 1$.  However, the lack of a measured redshift for this GRB and the possibility of other explanations (e.g. dust echoes: \\citealt{eb00,reichart01b}) made the identification of the bump uncertain.  Several attempts to identify similar bumps in the afterglows of GRBs with known redshift followed with mixed results; see \\citet{pks+02} for a review.  These earlier results motivated us to successfully propose a large {\\it Hubble Space Telescope} (HST) program to search for SNe underlying GRBs (GO-9180, P.I.: Kulkarni).  HST is ideally suited to this effort since its stable point-spread function and high angular resolution make possible accurate and precise photometry of variable sources embedded on host galaxies.  Low redshift GRBs are particularly important to study, since beyond a redshift of 1.2 the strong absorption in the SNe rest-frame spectra blueward of 4000 \\AA\\ covers the entire observed optical region, thus making searches all but impossible with current instruments.  To date, the best case for a SN underlying a cosmological GRB comes from HST observations of GRB~011121 ($z=0.365$; \\citealt{bkp+02,gsw+02}).  This is based on a bump in the optical afterglow lightcurves between 15 and 75 days, exhibiting a spectral turnover at $\\sim$ 7200 \\AA.  In addition, based on early NIR and radio observations \\citep{pbr+02}, the afterglow of GRB~011121 exhibits clear evidence for a circumburst density, $\\rho\\propto r^{-2}$ (where $r$ is the radial distance from the burst).  Such a density profile is indicative of stellar mass loss.  Hence, from two independent lines of evidence, it can be inferred that the progenitor of GRB~011121 was a massive star.  However, not all GRBs have an underlying SN as bright as that of GRB~011121 (e.g. GRB~010921: \\citealt{pks+02}).  Thus, additional deeper searches for coincident SNe are necessary to determine whether the lack of an observed SN is due to dust obscuration, diversity in the brightness of SNe coincident with GRBs, or to some subset of GRBs having a different progenitor.  So far, our discussion has been motivated by, and based on observations.  However, theorists have studied massive star models for long duration GRBs for more than a decade.  In particular, the ``collapsar'' model posits that GRBs arise when the cores of massive stars with sufficient angular momentum collapse and form black holes \\citep{woosley93,mw99,mwh01} whose accretion powers bursts of $\\gamma$-rays.  From the discussion in this section, it is clear that there is a good observational basis for the collapsar model.  Detailed studies of the underlying SNe (or their absence) will provide much needed observational constraints to the collapsar model, or other models which also require an associated SN event (``supranova'' --- \\citealt{vs98}; ``cannonball'' --- \\citealt{ddd02}).  Here, we present the discovery of the afterglow of GRB~020405 and the subsequent search for and discovery with HST of a red bump in the afterglow that we suggest may be a SN underlying the GRB. ",
        "conclusions": "Here we report the discovery of the nearby ($z\\sim 0.7$) GRB~020405 and the subsequent discovery of the afterglow.  The GRB itself, with a duration of 60~s, appears to be a typical long duration burst. The optical afterglow data, spanning about 10 days, can be fitted with a standard broken power law with a break time of about 1.5 d. Identifying this break with a jet we obtain a beaming-corrected energy release of about $2\\times 10^{50}\\,$erg, typical of that inferred for long duration GRBs.  Motivated by the low redshift, we undertook multi-color observations of the afterglow with HST. We found an excess over the flux predicted by the modeling of the afterglow from ground based data.  The overall broad-band spectrum of the bump as well as its temporal evolution are most simply explained as due to an underlying SN which exploded at about the same time as the GRB.  In \\citet{pks+02}, we summarize the searches for underlying SNe in $z<1.2$ GRBs (the redshift restriction arising from the fact that the searches are conducted in the optical band; see \\S\\ref{sec:intro}). Including GRB~020405, there are 13 GRBs with $z<1.2$.  A strong case for an underlying SN can be made for GRB~011121 ($z=0.365$; \\citealt{bkp+02,gsw+02}) and GRB~020405 ($z=0.690$).  A good case can be made for GRB~970228 ($z=0.695$; \\citealt{reichart99,gtv+00}) and GRB~980326 ($z$ unknown; \\citealt{bkd+99}).  At first blush this appears to be a low yield and suggestive that either there is diversity in the progenitors of long duration GRBs or in the properties of underlying SNe, or both.  However, it is important to bear in mind that one of the two unique signatures for a SN is the spectral rollover at short wavelengths, namely below about 4000 \\AA\\ (see \\citealt{bkd+99}).  Thus for $z\\sim 1$, observations in the $R$ and $I$ bands are critical, whereas at lower redshifts observations in $V$ and $R$ bands are critical. The $I$ band is quite noisy for ground-based observations whereas most afterglows are well observed in $R$ and $V$ bands.  Restricting to $z<0.7$ we find five GRBs (970228, 011121, 020405, 990712 and 010921), of which underlying SNe have been identified in the first three, and possibly in GRB~990712 \\citep{bhj+01}.  The limit for an underlying SN in GRB~010921 is not very stringent \\citep{pks+02}; in particular, an underlying SN fainter by more than 2~mag relative to that of SN 1998bw (about as bright as typical SNe Ib/c) would have escaped identification. It is thus premature to conclude that we need several progenitors to cause GRBs. What we can conclude, though, is that dense sampling in several bands of nearby GRBs is likely to remain a productive activity."
    },
    "0208/astro-ph0208522_arXiv.txt": {
        "abstract": "V838 Mon underwent, after a first nova--like outburst in January and a usual \\hyphenation{decline}decline, a second outburst after one month, and a third weak one again a month later. Moreover a very small increase of the  temperature at the beginning of April gives us a hint on  a physical process with a period of one month. We obtained a $BVRI_{\\rm C}$ time sequence and modelled the photometric behaviour of the object. This leads us to the conclusion that the interstellar foreground extinction has to be \\mbox{0\\fm6 $\\le$ E$_{B-V}$ $\\le$ 0\\fm8} and that the quasi-photosphere had persistently unusually low temperatures for nova--like systems. The photometry was used to follow the dramatic changes of the expansion. While the appearing 10~$\\mu$m excess can be well described by the heating of material ejected during this event, the IRAS emission near the location of the progenitor, originates most likely from dust, which were formed during the previous evolution of the object. Assuming that  the light echoes are coming from circumstellar material, the distance is 640 to 680\\ pc -- smaller than the 790\\ pc given in Munari et al. (2002). In our opinion V838~Mon and V4332~Sgr are manifestations of a new class of eruptive variables. We do not count M31~RV to this class. ",
        "introduction": "V838 Mon was discovered by Brown et al. \\shortcite{iauc7785_1} on 2002 January 6. Pre--discovery images show the target first on 2002 January 1. On February 2 the second 'outburst' \\mbox{($\\Delta_{\\rm V}$ = 3\\fm6)} was monitored in detail (Kimeswenger et al. 2002a). Such a behaviour of an extremely long pre-max halt or even a decline followed by a strong outburst was, according to our knowledge, never observed before in nova--like outburst, although HR Del and V723 Cas showed some similarities (Terzan et al. 1974; D\\\"urbeck 1981, Munari et al. 1996).  Our data were obtained with the Innsbruck 60cm telescope (Kimeswenger 2001; Kimeswenger \\& Lederle 2002) and a direct imaging CCD device. Until March 5 a CompuScope Kodak 0400 CCD (4\\farcm6 $\\times$ 3\\farcm1 field of view) was used. Thereafter and in the night of February 8 an AP7p SITe 502e (8\\farcm36 $\\times$ 8\\farcm36) was attached to the system. \\begin{figure} \\centering \\includegraphics[width=\\columnwidth]{mc438_f1.eps} \\caption{The lightcurves (from top to bottom $I_{\\rm C}$, $R$, $V$ and $B$) of V838 Mon between 2002 January 14 and 2002 April 13. The errors (rms) are typically below 0\\fm025. The lines do not correspond to the real behaviour (especially around the first maximum) but are given to connect the data of different bands properly.} \\label{lightcurve} \\end{figure} \\noindent 1018 images were taken in 31 nights between 2002 January 14 and April 13 using $BVR$ and $I_{\\rm C}$ filters. Average values for each day are given in Table~\\ref{phot_tab} and the resulting light curves shown in Fig.~\\ref{lightcurve}. The complete data table containing all individual measurements will be published elsewhere (Kimeswenger et al. 2002b).  \\begin{table} \\caption{Daily averaged photometry of V838 Mon} \\label{phot_tab} {\\small \\begin{tabular}{lcccccc} JD & day of  & V & B-V & V-R & R-I$_{\\rm C}$ & V-I$_{\\rm C}$ \\\\ & 2002  &  & &  & &  \\\\ \\hline 2452289.39 & 13.89 & 9.80 & 1.83 & 0.92 & 0.89 & 1.80 \\\\ 2452290.36 & 14.86 & 9.85 & 1.89 & 0.88 & 0.93 & 1.81 \\\\ 2452292.34 & 16.84 & 9.96 & 1.90 & 0.97 & 0.91 & 1.88 \\\\ 2452300.30 & 24.80 & 10.24 & 2.03 & 1.04 & 0.93 & 1.97 \\\\ 2452304.36 & 28.86 & 10.51 & 2.05 & 1.05 & 1.02 & 2.06 \\\\ 2452307.33 & 31.83 & 10.71 & 2.10 & 1.06 & 1.02 & 2.08 \\\\ 2452308.33 & 32.83 & 8.14 & 1.41 & 0.76 & 0.81 & 1.57 \\\\ 2452308.42 & 32.92 & 8.03 & 1.38 & 0.78 & 0.79 & 1.57 \\\\ 2452308.50 & 33.00 & 7.94 & 1.34 & 0.80 & 0.77 & 1.57 \\\\ 2452309.32 & 33.82 & 7.48 & 1.33 & 0.78 & 0.75 & 1.53 \\\\ 2452310.38 & 34.88 & 7.12 & 1.23 & 0.74 & 0.77 & 1.51 \\\\ 2452314.37 & 38.87 & 7.20 & 1.16 & 0.80 & 0.72 & 1.52 \\\\ 2452317.32 & 41.82 & 7.61 & 1.41 & 0.79 & 0.77 & 1.55 \\\\ 2452318.32 & 42.82 & 7.68 & 1.50 & 0.80 & 0.77 & 1.56 \\\\ 2452321.38 & 45.88 & 7.84 & 1.67 & 0.92 & 0.87 & 1.79 \\\\ 2452322.29 & 46.79 & 7.78 & 1.79 & 0.89 & 0.88 & 1.76 \\\\ 2452323.34 & 47.84 & 7.83 & 1.76 & 0.95 & 0.83 & 1.78 \\\\ 2452334.26 & 58.76 & 7.95 & 2.26 & 1.07 & 0.98 & 2.05 \\\\ 2452338.36 & 62.86 & 7.41 & 2.02 & 0.97 & 0.90 & 1.87 \\\\ 2452339.27 & 63.77 & 7.39 & 1.91 & 1.03 & 0.93 & 1.95 \\\\ 2452341.31 & 65.81 & 7.21 & 1.88 & 1.01 & 0.90 & 1.91 \\\\ 2452342.31 & 66.81 & 7.14 & 1.89 & 1.00 & 0.90 & 1.90 \\\\ 2452343.28 & 67.78 & 7.12 & 1.90 & 1.01 & 0.90 & 1.90 \\\\ 2452344.28 & 68.78 & 7.11 & 1.93 & 1.01 & 0.88 & 1.90 \\\\ 2452345.29 & 69.79 & 7.14 & 1.97 & 1.02 & 0.89 & 1.92 \\\\ 2452347.26 & 71.76 & 7.20 & 2.05 & 1.06 & 0.90 & 1.96 \\\\ 2452349.27 & 73.77 & 7.35 & 2.15 & 1.11 & 0.96 & 2.07 \\\\ 2452361.31 & 85.81 & 7.60 & 2.48 & 1.29 & 1.05 & 2.34 \\\\ 2452362.29 & 86.79 & 7.62 & 2.51 & 1.30 & 1.08 & 2.38 \\\\ 2452364.30 & 88.80 & 7.66 & 2.57 & 1.33 & 1.09 & 2.42 \\\\ 2452367.30 & 91.80 & 7.84 & 2.51 & 1.45 & 1.10 & 2.55 \\\\ 2452369.30 & 93.80 & 7.79 & 2.58 & 1.33 & 1.19 & 2.52 \\\\ 2452375.30 & 99.80 & 8.33 & 2.64 & 1.50 & 1.32 & 2.82 \\\\ 2452378.30 & 102.80 & 8.71 & 2.66 & 1.56 & 1.44 & 3.00 \\\\ \\hline \\end{tabular}} \\end{table}  Wagner et al. \\shortcite{iauc7785_2} mention that the object on the sky survey plate is a blend of at least two objects. To identify unambiguously the progenitor, CCD frames from January and SuperCOSMOS sky survey plate scans  were astrometrically calibrated. The difference in position is below 30\\ mas (see Fig. \\ref{dssimage}).  \\begin{figure} \\centering \\includegraphics[width=60mm]{mc438_f2.eps} \\caption{The R-band SuperCOSMOS sky survey plate scan. The CCD position is marked by the cross. The error of the position is smaller than the thickness of the lines. The ellipse marks the position of IRAS 07015-0346 (see Sec.~\\ref{sec_ir}).} \\label{dssimage} \\end{figure}  \\begin{figure} \\centering \\includegraphics[width=\\columnwidth]{mc438_f3.eps} \\caption{The calibration of the SuperCOSMOS sky survey plate scans (circles: $B$-band; diamonds: $R$-band) using deep CCD frames. The bars mark the positions of the progenitor for each band.} \\label{dss} \\end{figure}  Furthermore a set of deep $B$ and $R$ images were taken to calibrate the stars surrounding V838 Mon on the SuperCOSMOS frames (Fig.~\\ref{dss}). Using the colour corrections from H{\\\"o}rtnagl et al. (1992), we find the pre--outburst magnitudes $R=14\\fm56\\,\\pm\\,$0\\fm10 and $B=15\\fm87\\,\\pm\\,$0\\fm10. However, the ($B-R$) may be used only with care, since the plate epochs differ by 6.13 years. If the object was already variable, the colour is not meaningful. Our values differ from those given by Munari et al. (2002). This difference in $R$ is higher than expected from the different filter sets used ($R$ vs. $R_{\\rm C}$). As they give no information on the type of plate scan used in their work (DSS, SuperCOSMOS, \\dots) we are unable to compare the results. No information also is given there, how they transformed POSS E and ESO R ($\\lambda_{\\rm eff} = 645$nm). Moreover their conclusion that a red object \\mbox{[(B-V) = 0\\fm9} in their calibration] has the same magnitude on POSS I ($\\lambda_{\\rm eff} = 410$nm) and SERC $J$ ($\\lambda_{\\rm eff} = 480$nm) is confusing. They also include the $B$-band to the fit of the spectral energy distribution (SED), although this band is known to be strongly affected by non grey opacities at temperatures below \\mbox{10\\,000~K}. This leads to a lower temperature and thus later on to a lower interstellar extinction. ",
        "conclusions": "\\paragraph*{CV or Nova--like outburst ?\\medskip\\newline}  Using, although this may be not applicable for this peculiar object, the steady decline after the first primary maximum 2002 January 10, the thick wind models of Hachisu \\& Kato \\shortcite{HaKa} suggest a low mass WD with about 1 M$_\\odot$. The optical and $JHK$ (2MASS) pre-outburst photometry, even if the use of the colours might be dangerous, is either fitted by a single 7800$\\pm$300\\ K blackbody or better by a composite of a hot accretion disk with a \\mbox{4500\\ K} companion. At a distance of 650\\ pc the absolute visual magnitude would be around +3\\fm6. A single early F type main sequence star would fit to these values. Using the better fitting double component model and using the accretion disk parameters from Webbink et al. \\shortcite{We87} we obtain a K2-4 main sequence star and a very reasonable accretion rate of $2 \\times 10^{-8}$~M$_\\odot$~yr$^{-1}$. Due to the higher extinction and the different values of the photometry for the progenitor, we do not need an unusual subluminous star as it is used by Munari et al. (2002). The temperatures during outburst are too low for such a classical scenario. Although the models of M31\\ RV (Iben \\& Tutukov 1992) allow such a red cold outburst in small period CV with a cold low mass WD, it leads to a higher outburst luminosity (M$_{\\rm V} \\approx -10$).  Munari et al. (2002) discuss this scenario with respect to a  thermonuclear runaway of the complete WD. However M31 RV underwent at least two recurrent outbursts within the rather small timescale of 20 years (Sharov 1990). We thus infer  that the nova/CV scenario have to be excluded   \\paragraph*{A late helium flash post-AGB ?\\medskip\\newline}  This scenario (V605 Aql, V4334\\ Sgr, \\dots) was discussed several times in different IAU circulars.  Both, V605 Aql as well as V4334 Sgr, showed a 6000-7000\\ K photosphere throughout more than one year. The light curves were smooth. For V4334~Sgr (outburst in February 1996) we know now, that it stayed at constant bolometric luminosity for at least seven years. V605~Aql has the same bolometric luminosity even 85 years after its outburst. Even if we adopt a much larger distance and thus a more luminous scenario, we should have a much slower evolution after the outburst. Also the lack of strong carbon overabundance in the spectra clearly rules out this model.  \\paragraph*{A twin of V4332\\ Sgr ?\\medskip\\newline}  V4332\\ Sgr underwent a similar outburst in 1994 (Martini et al. 1999 and references therein). Its nature is unclear up to now. While the evolution of the temperature and the expansion of an opaque massive shell and the luminosity at maximum look much the same, V4332\\ Sgr did not show multiple visual maxima like V838\\ Mon. This statement, as well as the statement that the ascent to maximum lasted 200 days (Martini et al. 1999), have to be used with care. V4332~Sgr was not observable before the discovery (daytime object). If we compare the light curve and the temperature sequence of V838~Mon with those of V4332~Sgr, we see a good agreement, starting mid of March 2002. It is thus possible, that the earlier features did also exist there. They were simply not observed. A single thermonuclear event, as discussed for V4332\\ Sgr, still might cause such a behaviour. Also the density of \\mbox{$N_e \\approx 10^{14}$ m$^{-2}$} (Martini et al. 1999) for the late stages fits well to our finding above, if we take into account the 15 times larger radius, compared to the date when we estimated the density. But this does not agree with the main sequence progenitor. Why should a main sequence star undergo a shell flash? Low mass blue horizontal branch (BHB) stars evolve directly to a WD without passing through a luminous AGB phase (Heber et al. 1997). They are stated to be the progenitors of low mass WDs ($\\approx 0.3 M_\\odot$) without planetary nebula. Those stars, at the end of the He-core burning in transition to He-shell burning, pass the main sequence at a similar position in the HR-diagram. This would be a new possible scenario similar to a late He-flash for normal mass progenitors passing through the AGB.   \\bigskip  Although it is too early to draw final conclusions, the photometric behaviour, the temperature and the bolometric luminosity most likely make V838~Mon a twin of V4332~Sgr. Those two objects seem to form a new class of eruptive variables. Unlike Munari et al. (2002), we do not include M31~RV in this group. The photometric and spectroscopic behaviour and especially the luminosity of this object seem to differ significantly. We have to think about possible physical processes causing a star near the main sequence to ignite eruptive thermonuclear events.\\\\ Further observations at all wavelengths of the up to now unstudied remnant of V4332~Sgr and the new remnant expected after V838~Mon emerges from its solar conjunction are urged."
    },
    "0208/astro-ph0208378_arXiv.txt": {
        "abstract": "We consider the physics of free precession of a rotating neutron star with an oblique magnetic field. We show that if the magnetic stresses are large enough, then there is no possibility of steady rotation, and precession is inevitable. Even if the magnetic stresses are not strong enough to prevent steady rotation, we show that the minimum energy state is one in which the star precesses. Since the moment of inertia tensor is inherently triaxial in a magnetic star, the precession is periodic but not sinusoidal in time, in agreement with observations of PSR 1828-11. However, the problem we consider is {\\it not} just precession of a triaxial body. If magnetic stresses dominate, the amplitude of the precession is not set just by the angle between the rotational angular velocity and any principal axis, which allows it to be small without suppressing oscillations of timing residuals at harmonics of the precession frequency. We argue that magnetic distortions can lead to oscillations of timing residuals of the amplitude, period, and relative strength of harmonics observed in PSR 1828-11 if magnetic stresses in its core are about 200 times larger than the classical Maxwell value for its dipole field, and the stellar distortion induced by these enhanced magnetic stresses is about 100-1000 times larger than the deformation of the neutron star's crust. Magnetic stresses this large can arise if the core is a Type II superconductor, or from toroidal fields $\\sim 10^{14}$ G if the core is a normal conductor. The observations of PSR 1828-11 appear to require that the neutron star is slightly prolate. ",
        "introduction": "\\label{intro}  The convincing observation of free precession of PSR 1828-11 (Stairs, Lyne \\& Shemar 2000) poses challenges for theories of neutron stars. Shaham (1977; 1986) argued that vortex line pinning in the neutron star crust should prevent long term precession. Sedrakian, Wasserman \\& Cordes (1999) showed that precession is still prevented if vortex lines are not pinned perfectly but vortex drag is strong. They also showed that even if vortex drag is weak, precession is damped away. Link \\& Cutler (2002) estimated the strength of vortex line pinning forces, and argued that PSR 1828-11 may precess at large enough amplitude to unpin superfluid vortices in the crust. If vortex drag is small this would remove one impediment to free precession, although the free precession would still damp away eventually if it is not excited continuously.  Here, we consider an additional feature of radiopulsars like PSR 1828-11, namely, that they are strongly magnetized, with magnetic axes that are at an angle to their rotation axes. Based on earlier work on magnetic stars by Mestel and collaborators (Mestel \\& Takhar 1972, Mestel et al. 1981, Nittman \\& Wood 1981; see also Spitzer 1958), we argue that precession may be {\\it required} -- there is no equilibrium corresponding to solid body rotation without precession for a rotating star with an oblique magnetic field. For a fluid star, though, we shall see that although the fluid must precess, the magnetic axis rotates uniformly. Although Mestel et al. (Mestel \\& Takhar 1972, Mestel et al. 1981, Nittman \\& Wood 1981) showed that hydrostatic balance also requires fluid motions in addition to the precession which affect the stellar magnetic field, these are too slow to be important observationally.  Once we also take account of the solid crust of a neutron star in addition to its fluid interior, we show that if the stellar distortions due to the magnetic field are larger than the distortion of the crust, then steady state rotation is very unlikely (but not necessarily impossible). If the magnetic stresses inside PSR 1828-11 are simply due to the classical Maxwell stress tensor, evaluated with the inferred dipole magnetic field strength, then they are too weak to require precession at a period $\\sim 1$ year. However, if the interior of the neutron star consists of a Type II superconductor, the effective stress tensor is larger than for a classical magnetic field, according to Jones (1975) and Easson \\& Pethick (1977); as was emphasized by these authors, and Cutler (2002), the magnetic distortion is correspondingly larger. For PSR 1828-11, we estimate that the distortion that would result in a core that contains a Type II superconductor can lead to precession at a period of order 1 year. A sufficiently strong toroidal magnetic field ($B_t\\sim 10^{14}$ G) could also lead to precession at this period without Type II superconductivity (see e.g. Eq. [2.4] in Cutler [2002] with $B_c\\to 0$).  Even if magnetic stresses are not strong enough to prevent steady rotation, magnetic distortion in an oblique rotator alters the physics of free precession qualitatively and quantitatively compared to what one would expect for precession due to axisymmetric crustal distortions. Even if the crust is axisymmetric, misalignment between the symmetry axis of the crust and the magnetic field make the effective stellar moment of inertia inherently triaxial. From a phenomenological viewpoint, one manifestation of this loss of axisymmetry is that although the angular velocity of the star is a {\\it periodic} function of time in the frame rotating with the crust and magnetic field, it is not a {\\it sinusoidal} function of time. This feature is consistent with observations of PSR 1828-11, which reveal behavior at several different harmonically related frequencies (Stairs, Lyne \\& Shemar 2000). The relative strengths of the harmonics depend on the degree of nonaxisymmetry, and, presuming an axisymmetric crust, on the distortions induced by the magnetic field. For the ordinary Maxwell stresses evaluated just with the inferred dipole field strength, the nonaxisymmetry would be small, but distortions resulting from a Type II superconductor, or from a normal core with a large toroidal field, could produce sufficient nonaxisymmetry to lead to comparable amplitudes for at least the first few harmonics of the fundamental precession period, as is observed.  Periodic, but not sinusoidal, precession would also arise if the neutron star crust were simply nonaxisymmetric even if magnetic stresses were negligible. Thus, the detection of harmonic behavior in PSR 1828-11 cannot, by itself, be taken to be evidence for amplified magnetic stresses in its core. However, standard results for triaxial precession, which are reproduced as a byproduct of the calculations we present, show that the oscillations at harmonics of the precession frequency are smaller in amplitude than the oscillation at the fundamental frequency by powers of the precession amplitude. We shall see that this is not the case in models where magnetic stresses predominate: the amplitudes of oscillations at the precessiom frequency and twice the precession frequency may be comparable {\\it even at small amplitude}.  Moreover, it is well known that the minimum energy state for rotation of a nonaxisymmetric body is one in which the angular velocity and angular momentum are aligned with the principal axis of the body with largest eigenvalue of the moment of inertia tensor. There is no precession at all in this state. However, we shall see that even when magnetic stresses are not strong enough to {\\it require} free precession, the minimum energy state for an oblique rotator does {\\it not} correspond to alignment of the angular velocity with any principal axis of the effective moment of inertia tensor. Thus, the minimum energy state is one in which the star precesses.  We shall see that the timing residuals associated with the precession can account for observations of long term oscillations in PSR 1828-11, but that the explanation only works if the core of the neutron star has sufficiently strong magnetic stresses that magnetic distortions are 100-1000 times larger than crustal deformations. Thus, we argue that observations of free precession in PSR 1828-11 offer evidence that the pulsar is in the regime where magnetic stresses dominate, because either its core is a Type II superconductor, or, if it is a normal conductor, possesses a very large toroidal field.  We review the arguments given by Mestel and collaborators (Mestel \\& Takhar 1972, Mestel et al. 1981, Nittman \\& Wood 1981) in \\S \\ref{mestel}. In \\S \\ref{inevitable}, we consider the conditions under which precession of an oblique rotator is {\\it inevitable}. In \\S \\ref{precess}, we consider the modified precession problem for an axisymmetric crust and misaligned magnetic field. There we show that the minimum energy state at a given angular momentum is one in which the star precesses, provided that the magnetic field is not either along or perpendicular to the symmetry axis of the crust. There, we also review classical results for free precession of triaxial bodies; we shall find that there are three distinct cases of interest for a star where magnetic distortions are important. We present limiting results for the timing residuals expected in this model in \\S \\ref{pulsarrivaltimes}, and obtain approximate results for the limit in which crustal distortions dominate in \\S \\ref{pulsibsmall}, and for the opposite limit in which magnetic distortions dominate in \\S \\ref{pulsiblarge}.  Here, our main purpose is to present arguments that an oblique rotator must precess. In \\S \\ref{application}, though, we present brief application of the model to observations of PSR 1828-11. There we argue that only models in which magnetic distortions dominate can account for all of the observed features of the long-term periodic timing residuals from this pulsar. We also suggest that the data favor prolate rather than oblate magnetic distortions.  Several appendices present cumbersome mathematical details needed to derive (and verify!) analytic results presented in the text. ",
        "conclusions": "\\label{discussion}  Here, we have extended previous studies of precession of neutron stars to incorporate the effects of oblique magnetic fields. We have shown that if the magnetic stresses are large enough, then steady rotation is unlikely, and the neutron star must precess. Moreover, even when the magnetic stresses are relatively weak, so that steady rotation is possible irrespective of the obliquity of the magnetic field, the minimum energy state is not the steady state. Thus, even in this case, the neutron star will precess. We argued, in \\S \\ref{precess}, that the minimum energy, precessing state is a local energy minimum that applies at fixed angle between the magnetic and rotational axes. On a longer timescale, we would expect the star to seek its global energy minimum, which should correspond to either aligned or perpendicular magnetic and rotation axes, and no precession. We might expect short timescale dissipative effects to drive the system toward its local minimum, and that the global minimum is only achieved on somewhat longer timescale, perhaps as a result of electromagnetic spindown torques (e.g. Goldreich 1970).  The effective moment of inertia tensor of a neutron star with an inclined magnetic field is inherently triaxial. Consequently, the precession is {\\it periodic} but not {\\it sinusoidal} in time. In general, the solution for the rotational angular velocity of the star can be expanded in a Fourier series involving harmonics of the precession frequency. We have shown that at least the first few terms in such an expansion can have comparable magnitudes provided that the interior magnetic stresses are not very small.  The condition that magnetic stresses play an important role is that the magnetic-induced distortions are comparable to or larger than the distortions of the stellar crust. For precession periods of order years, the implied magnetic stresses exceed those expected from the classical Maxwell stress tensor, evaluated using the inferred dipole magnetic field strength, by a couple of orders of magnitude. However, if the interior of a neutron star contains a Type II superconductor, or else is a normal conductor but possesses large toroidal magnetic fields, the magnetic stresses are larger, and the implied distortions can be of the right order of magnitude (Jones 1975, Easson \\& Pethick 1977, Cutler 2002). Thus, the observation of neutron star precession can be taken as indirect evidence for enhanced magnetic stresses, due to either Type II superconductivity, or large toroidal fields.  We postpone detailed application of the ideas set forth here to PSR 1828-11 to another paper (Akgun, Epstein \\& Wasserman 2002). However, in \\S \\ref{application} we argued that only a model with $\\vert I_B\\vert\\gg C$ can lead to time residuals that oscillate with comparable amplitude at both $\\omega_p$ and $2\\omega_p$. In this case, the amplitude of the observed time residuals is set by the dimensionless ratio $\\Ctil\\equiv C/\\vert I_B\\vert$, not by the tilt of the angular velocity vector away from any principal axis, which may be large.  By contrast, if the stellar distortions associated with magnetic stresses are {\\it smaller than} those associated with the crust, then Eqs. (\\ref{oscibposllc}) and (\\ref{oscibnegllc}) show that the timing residuals oscillate predominantly at $\\omega_p$. Oscillations at $2\\omega_p$ would be down by factors $\\sim\\vert\\Omhat_-\\vert$, where $\\Omhat_-$ is given by Eq. (\\ref{omo}), and is $\\sim\\vert I_B\\vert/C$ in the minimum energy configuration. Moreover, we argued, in \\S \\ref{pulsibsmall} and \\S \\ref{application}, that precession of a triaxial crust alone would probably not, at small precession amplitude, be capable of producing oscillations of comparable magnitude at both $\\omega_p$ and $2\\omega_p$, because the precession amplitude is proportional to $\\vert\\Omhat_-\\vert$ and the oscillations at harmonics of $\\omega_p$ are suppressed by factors $\\sim\\vert\\Omhat_-\\vert$. Thus, a solution in which the crustal distortion is responsible for precession is unlikely to explain the data on PSR 1828-11. The model in which magnetic stresses dominates is not merely precession of a triaxial body because small amplitude phase residuals can arise even when $\\vert\\Omhat_-\\vert$ is not small.  For a precessing star in which the distortion is due to magnetic stresses primarily, the precession period is \\be {P_p}\\simeq{P_0\\Ibar\\over 3I_B\\cos\\chi} \\simeq{492~{\\rm days}\\over\\beta\\cos\\chi(BH)_{27}/5}~, \\ee where the value given is for PSR 1828-11. This is similar to the observed period, about 1000 days, provided that $\\beta\\cos\\chi$ is not very small. Thus, if $\\chi\\sim 60^\\circ$ the timing residuals could oscillate with about the right period, amplitude and relative importance of the fundamental precession period and its first harmonic. In this case, we note that the condition that there is no steady state, $\\vert\\sin 2\\chi\\vert\\geq\\Ctil$ can be satisfied: for $\\chi=60^\\circ$, for example, $\\sin 2\\chi=\\sqrt{3}/2$ whereas we estimated that $\\Ctil\\sim 0.001-0.01$ in \\S \\ref{application}. This possible explanation of the timing residuals for PSR 1828-11 only works if the magnetic stresses in this star are $\\sim 200$ times larger than would be indicated by its dipole magnetic field. Thus, we conclude that either the interior is a Type II superconductor, or is a normal conductor with a toroidal field whose strength is $\\sim 10^{14}$ G. Otherwise, the expected magnetic stresses are far smaller than is needed for this solution to apply.  We also noted, in \\S \\ref{application}, that the ratio of the minimum and maximum timing residuals, and the shape of the variation of $\\Delta P$ and $\\Delta\\dot P$ seen in PSR 1828-11 appear to favor a model in which $I_B>0$, so that magnetic distortions are {\\it prolate}. Prolate distortions would arise naturally from the stresses due to a toroidal field, with or without Type II superconductivity (Cutler 2002), but may also result if magnetic flux tubes have been transported outward in the core and accumulate at its outer boundary (Ruderman, Zhu \\& Chen 1998, Ruderman \\& Chen 1999), which could ``pinch'' the interior.  Crustal distortions are still needed in order for the precessional amplitude to be nonzero. In fact, to be more precise, the precessional amplitude depends on the component of the moment of inertia tensor of the crust that is {\\it not} symmetric about the magnetic axis. This can be seen directly from Eqs. (\\ref{oscibggcpos}) and (\\ref{oscibggcneg}), which show that the oscillating time residuals vanish as $\\Ctil\\sin\\theta\\to 0$, where $\\theta$ is the angle between the magnetic axis $\\bhat$ and, in this axisymmetric distortion model, the symmetry axis of the relevant crustal deformation, $\\khat$. Thus, although all neutron stars precess as a consequence of their magnetic stresses for $\\vert I_B\\vert\\gg C$ in the picture advanced here, only those with sufficiently large nonaligned crustal deformations would have discernible oscillations of their timing residuals. In this sense, PSR 1828-11 may be special.  Although we have treated the basic physics of precession of an oblique rotator in some detail, we have not treated several effects that might play significant roles. We have not explicitly included either vortex line pinning or vortex drag. Link \\& Cutler (2002) have argued that vortex lines can unpin globally at large enough precession amplitude. For $\\vert I_B \\vert\\gg C$, as is required to explain the timing residuals in PSR 1828-11, the precession amplitude is $\\simeq\\sin\\chi$ or $\\cos\\chi$ (depending on the sign of $I_B$), which is not small, so global unpinning is expected. Thus, we may expect that vortex lines are unpinned in neutron stars with magnetic fields that are strong enough to have $\\vert I_B \\vert\\gg C$ except for $\\chi$ very close to either zero or $\\pi/2$, depending on the sign of $I_B$. For these, vortex line drag, if weak enough, may simply serve to bring the neutron star superfluid into corotation with its crust, and drive the rotating star toward its minimum energy state. For large $\\vert I_B\\vert/C$, we have seen that precession is required, so weak vortex drag or other forms of dissipation need not prevent precession. For small values of $\\vert I_B\\vert/C$, the precession amplitude is given by Eq. (\\ref{omo}) and can be very small; in the minimum energy state, Eq. (\\ref{omo}) implies $\\vert\\Omhat_-\\vert\\simeq\\vert I_B\\sin 2\\theta\\vert/C$. In this case, it is possible that precession cannot overcome pinning forces, as discussed in Link \\& Cutler (2002), and so precession does not occur.  Theories of pulsar glitches involve the pinning, unpinning and repinning of crustal superfluid vortex lines (e.g. Anderson and Itoh 1975; Alpar et al. 1981, 1984a,b, 1993; Link, Epstein and Baym 1993). As we have seen, for large $\\vert I_B\\vert/C$, precession amplitudes are large, and vortices may be expected to unpin, but it is possible for vortex lines to remain pinned in neutron stars with $C\\gg\\vert I_B\\vert$. Thus, there could be a dichotomy between pulsars that glitch ($C\\gg\\vert I_B\\vert$) and those that precess ($C\\ll\\vert I_B\\vert$). If, in the course of a glitch, all crustal superfluid vortices were to unpin, then the star might precess briefly. Perhaps that explains the detection of damped, quasisinusoidal timing residuals in the Vela pulsar after and perhaps before its Christmas 1988 glitch (McCulloch et al. 1990).  We have kept the problem of precession of an oblique rotator as simple as possible by considering what happens when the magnetic field is axisymmetric about some axis, and the crustal distortions are also axisymmetric, but about a different axis. More realistically, both of these simplifying assumptions are likely to be violated. Most likely, the crust is not axisymmetric. When magnetic stresses dominate, we do not expect including intrinsic crustal asymmetry to alter the results found here qualitatively, since the effective moment of inertia is already triaxial here. Triaxiality of the crust, in the limit of rather small crustal distortions, would simply rotate the principal axes slightly. Furthermore, the magnetic field may have a more complicated structure than we have assumed. A substantial quadrupolar component would presumably render the contribution to the inertia tensor from magnetic stresses alone triaxial. We shall consider these complications elsewhere.  Although we have included the spindown torque in our evaluations of timing residuals, we did not include near zone electromagnetic torques (Good \\& Ng 1985, Melatos 1997, 1999, 2000). The principal effect of such torques would be to renormalize the moment of inertia tensor of the star. Near zone torques can play a role similar to the magnetic distortions considered here, but are smaller by a factor $\\sim (H/B)(Rc^2/GM)$, which is non-negligible even if $H=B$. However, we note here that the large magnetic distortions we propose would presumably apply to the spindown of magnetars and anomalous X-ray pulsars, in much the same fashion as proposed by Melatos (1999, 2000). We shall pursue this idea elsewhere.  We have also ignored motions of the fluid and crust of the star apart from rigid rotation. Mestel and collaborators (Mestel \\& Takhar 1972, Mestel et al. 1981, Nittman \\& Wood 1981) have pointed out that the equilibrium in a fluid star with oblique magnetic field must involve fluid motions with velocities $\\sim(\\Omega^2R^3/GM)\\omega_pR$. These distort the stellar magnetic field by a fractional amount $\\sim \\Omega^2R^3/GM$ over the precession period $2\\pi/\\omega_p$ (Mestel \\& Takhar 1972, Mestel et al. 1981, Nittman \\& Wood 1981). Similar motions might arise in the crust, but with magnitudes $\\sim(\\Omega^2R^2/c_t^2)\\omega_pR$, where $c_t$ is the sound speed for transverse waves. The expected amplitude of the resulting magnetic wander is $\\sim \\Omega^2R^2/c_t^2$. Although slow, these displacements cause the magnetic field of the star to oscillate about its undisturbed, axisymmetric state, and might influence the long-term behavior of the observed spindown. We shall investigate whether there is any long-term observational signature of these motions elsewhere. In addition, it is likely that magnetic and rotational deformations of the core must also deform the crust, as they exert pressure on its inner boundary. We would expect the magnetic deformation of the core to promote crustal deformation symmetric about $\\bhat$, which would not lead to observable precession, but the rotation-induced deformation need not be symmetric about $\\bhat$, and should be substantial. An important question left unanswered here is whether the crustal distortion, $C_{ij}$, that leads to detectable precession is relatively steady, or is simply due to a seismic fluctuation. We shall explore the important issue of crustal deformations elsewhere.  Finally, we emphasize that any neutron star with strong enough core magnetic stresses ought to precess, but we may not be able to detect their precession because their magnetic axes can still rotate more or less uniformly. This is because, at small values of $C/\\vert I_B\\vert$, the neutron star precesses almost exactly about its magnetic axis, which therefore rotates almost uniformly as seen in the inertial frame. Although the precession may not be detectable readily from timing residuals for most pulsars, gravitational radiation amplitudes would be larger than would arise without enhanced internal magnetic stresses (e.g. Cutler \\& Thorne 2002, Cutler 2002). The distortions required for PSR 1828-11 are still smaller than would be needed for detection by LIGO, even if it were spinning faster (Brady et al. 1998). If there are young, highly magnetized neutron stars rotating rapidly, they would be the brightest emitters of gravitational radiation. Such objects have been hypothesized to be the sources of the highest energy cosmic rays (Blasi, Epstein \\& Olinto 2000, Arons 2002)."
    },
    "0208/astro-ph0208464_arXiv.txt": {
        "abstract": "We investigate the limitations of thermonuclear X-ray bursts as a distance indicator for the weakly-magnetized accreting neutron star 4U~1728$-$34. We measured the unabsorbed peak flux of \\burstnum\\ bursts in public data from the {\\em Rossi X-Ray Timing Explorer}.  The distribution of peak fluxes was bimodal: \\preburst\\ bursts exhibited photospheric radius expansion (presumably reaching the local Eddington limit) and were distributed about a mean bolometric flux of \\fmean, while the remaining (non-radius expansion) bursts reached \\fnopremean, on average. The peak fluxes of the radius-expansion bursts were not constant, exhibiting a standard deviation of \\fstddevp\\% and a total variation of \\netvarp\\%.  These bursts showed significant correlations between their peak flux and the X-ray colors of the persistent emission immediately prior to the burst.  We also found evidence for quasi-periodic variation of the peak fluxes of radius-expansion bursts, with a time scale of $\\simeq 40$~d.  The persistent flux observed with \\xte/ASM over 5.8~yr exhibited quasi-periodic variability on a similar time scale.  We suggest that these variations may have a common origin in reflection from a warped accretion disk.  Once the systematic variation of the peak burst fluxes is subtracted, the residual scatter is only $\\simeq 3$\\%, roughly consistent with the measurement uncertainties.  The narrowness of this distribution strongly suggests that i) the radiation from the neutron star atmosphere during radius-expansion episodes is nearly spherically symmetric, and ii) the radius-expansion bursts reach a common peak flux which may be interpreted as a standard candle intensity. Adopting the minimum peak flux for the radius-expansion bursts as the Eddington flux limit, we derive a distance for the source of 4.4--4.8~kpc (assuming $R_{\\rm NS}=10$~km), with the uncertainty arising from the probable range of the neutron star mass $M_{\\rm NS}=1.4$--2~$M_\\sun$. ",
        "introduction": "Thermonuclear (type I) X-ray bursts manifest as rapid changes in the X-ray intensity of accreting neutron stars in low-mass X-ray binary (LMXB) systems, with rise times between $\\la 1-10$~s and decay times between $\\sim 10-100$~s \\cite[see][for reviews]{lew93,bil98a}. Such bursts have been observed from more than 70 sources \\cite[]{lmxb01} and are caused by unstable nuclear burning of accreted matter on the neutron-star surface.  If the thermonuclear energy during a burst is released sufficiently rapidly, the flux through the neutron star atmosphere may reach the local Eddington limit, at which point the outward radiation force balances gravity.  The excess energy is converted into potential and kinetic energy of the X-ray photosphere, which is lifted above the neutron star surface while the emerging luminosity (measured locally) remains approximately constant and equal to the Eddington limit.  These are the so-called radius-expansion or Eddington-limited bursts.  For spherically symmetric emission, the Eddington luminosity measured by an observer at infinity is given by \\cite[]{lew93} \\begin{eqnarray} L_{\\rm Edd,\\infty} & = & \\frac{8\\pi G m_{\\rm p} M_{\\rm NS} c [1+(\\alpha_{\\rm T}T_{\\rm e})^{0.86}]} {\\xi\\sigma_{\\rm T_{\\rm e}}(1+X)} \\left(1-\\frac{2GM_{\\rm NS}}{Rc^2}\\right)^{1/2} \\nonumber \\\\ & = & 2.5\\times10^{38} \\left(\\frac{M_{\\rm NS}}{M_\\odot}\\right) \\frac{1+(\\alpha_{\\rm T}T_{\\rm e})^{0.86}}{\\xi(1+X)}\\left(1-\\frac{2GM_{\\rm NS}}{Rc^2}\\right)^{1/2}\\ \\eps \\label{ledd} \\end{eqnarray} where $M_{\\rm NS}$ is the mass of the neutron star, $T_{\\rm e}$ is the effective temperature of the atmosphere, $\\alpha_{\\rm T}$ is a coefficient parametrizing the temperature dependence of the electron scattering opacity \\cite[$\\simeq 2.2\\times10^{-9}$~K$^{-1}$;][]{lew93}, $X$ is the mass fraction of hydrogen in the atmosphere ($\\approx0.7$ for cosmic abundances), and the parameter $\\xi$ accounts for possible anisotropy of the burst emission.  The final factor in parentheses represents the gravitational redshift due to the compact nature of the neutron star, and also depends upon the height of the emission above the neutron star surface $R\\ge R_{\\rm NS}$.  Because the Eddington luminosity depends on the ratio $M_{\\rm NS}/R_{\\rm NS}$, measurements of the peak flux of radius-expansion bursts allows in principle the measurement of these fundamental properties \\cite[e.g.][]{damen90,smale01,kuul01}. On the other hand, because the masses and radii of stable neutron stars predicted by any given equation of state span a narrow range of values \\cite[see, e.g.,][]{lp01}, Eddington-limited X-ray bursts can be used as distance indicators.  The validity of the physical picture of Eddington-limited bursts discussed above can be verified observationally in two ways. First, the peak fluxes of such bursts from each individual source should be the same. Second, the peak luminosities inferred for sources with independent distance measurements, such as those in globular clusters, should correspond to the Eddington limit for a neutron star.  Since the original discovery of Eddington-limited bursts, a number of authors have addressed these questions. In early observations, a significant number of radius expansion bursts had been detected from three sources and their peak fluxes were found to be similar to within $\\simeq 20$\\% (4U~1820$-$30: \\citealt{vlvp86,damen90}; 4U~1636$-$536: \\citealt{damen90}; 4U~1728$-$34: \\citealt{bas84}). In particular, the peak luminosities of radius-expansion bursts observed from 4U~1820$-$30, which resides in the gobular cluster NGC~6624 were found to be comparable to the Eddington limit for a neutron star. Although successful in providing support to the model of Eddington-limited bursts, these early studies were limited to a small number of sources and suffered from the statistical uncertainties inherent to fitting spectral models to low signal-to-noise data. In recent years, observations with {\\it BeppoSAX}\\/ and the {\\em Rossi X-ray Timing Explorer\\/} (\\xte) have revealed a large number of Eddington-limited bursts from several sources, which have been studied in great detail. Using {\\it BeppoSAX}\\/ and \\xte\\/ data, \\cite{kuul02} recently studied the Eddington-limited bursts of 12 globular cluster sources with well known distances and showed that, with one exception, their peak fluxes were constant to within $\\simeq 15$\\% and were comparable to the Eddington limit for neutron stars with H-poor atmospheres.  In this series of articles, we use all the publicly available data obtained with \\xte\\/ to date in order to observationally test the hypothesis that Eddington-limited bursts can be used as distance indicators for neutron-star LMXBs.  In the present study we quantify, and examine the causes of, systematic variations in the peak burst flux from the source with the greatest number of bursts detected by \\xte, \\src. ",
        "conclusions": "\\label{disc}  We have studied \\burstnum\\ thermonuclear X-ray bursts from \\src, which includes \\preburst\\ exhibiting evidence for photospheric radius expansion, as observed by the {\\em RXTE}/PCA. We have shown that the radius expansion bursts exhibit a significant variation in their peak fluxes \\fpkre.  The fractional standard deviation was $\\simeq \\fstddevp$\\%, while the total variation was $\\simeq$\\netvarp\\%.  This is significantly larger than the formal measurement errors, which were typically $\\simeq 2$\\%.  This result appears inconsistent with the simple picture of radius-expansion bursts \\cite[e.g.][]{lew93}, according to which the peak flux should be nearly constant and equal to the Eddington critical flux.  However, we must also consider the possibility of a variety of systematic effects arising as a consequence of our analysis method, which may contribute (or give rise) to the measured variation.  \\subsection{Possible systematic effects} \\label{system}  While subtraction of the pre-burst emission as background has dubious theoretical justification, and may have originally been adopted for convenience, the method has been shown in several analyses to be relatively robust \\cite[e.g.][]{vpl86,kuul01}. An implicit assumption is that the persistent emission remains unchanged throught the burst. This may not be true, especially for the very energetic, radius-expansion bursts that may disrupt the inner accretion flow.  As a result, variations in the true persistent flux $F_{\\rm per}(t)$ during the burst may contribute to variation in the measured bolometric burst flux.  The persistent flux prior to the radius expansion bursts in \\src\\ was typically around $3.7\\times10^{-9}\\ \\epcs$, but for some bursts was as high as $5\\times10^{-9}\\ \\epcs$.  The net variation over all the bursts was only $3.2\\times10^{-9}\\ \\epcs$, which is insufficient to account for the \\netvar\\ net variation in the peak burst fluxes.  Even if the persistent flux was quenched completely during the burst, this would still only give a variation of at most $5\\times10^{-9}\\ \\epcs$. Thus, variations in the peristent flux cannot completely account for the observed variation in \\fpkre, but may contribute to some extent. If this is the case, we may expect a correlation between the \\fpkre\\ and the pre-burst persistent flux.  In fact, \\fpkre\\ and the persistent flux were weakly anticorrelated, with Spearman's rank correlation $\\rho=-0.32$ (estimated signifcance $1.5\\times10^{-2}$, equivalent to $2.4\\sigma$). If anything, the persistent flux variation serves to suppress slightly the true variation in \\fpkre.  The bolometric correction (equation \\ref{flux}) typically adds $\\simeq 7$\\% to the peak 2.5--20~keV flux of the radius expansion bursts as measured by the PCA.  This correction may influence our result in one of two ways.  On the one hand, systematically biased bolometric corrections may give rise to a variation in flux which is not present in the observed source fluxes (restricted to the PCA passband).  We can easily rule out this possibility, since the flux integrated just over the PCA passband exhibits identical (fractional) variation as the inferred bolometric flux. Alternatively, deviations in the spectrum from a perfect blackbody may give rise to (unobserved) variations in the flux outside the PCA passband that contribute to the true bolometric flux variation. In the latter case these unobserved bolometric flux variations may (for example) partially or completely compensate for the variation we see in the PCA passband, so that the bolometric flux variation we infer is exaggerated, or erroneous. The second effect is more difficult to discount, since (obviously) the spectral variations outside the PCA passband are not measurable with the present data.  However, the $\\simeq 7$\\% typically added to the flux in the PCA passband by the bolometric correction is significantly smaller than the $\\simeq$\\netvarp\\% observed variation in the peak fluxes. In order to compensate for the flux variation we observe, this contribution would then need to fluctuate by a factor of $\\simeq6$, which seems unlikely.  Thus, we conclude that the observed variation was not an artifact of any aspect of our analysis method and, hence, must be genuine.  \\subsection{Origin of the peak flux variation} \\label{opfv}  The large number of radius-expansion bursts observed from \\src\\ with {\\em RXTE}/PCA along with the long-term flux history accumulated by the ASM provides for the first time a plausible cause for the observed broad distribution of peak burst fluxes. The fact that (i) the peak burst fluxes were correlated with the X-ray colors of the persistent emission, (ii) they varied in a quasi-periodic manner, and (iii) the timescale and fractional amplitude of the variability were similar to those of the persistent emission, strongly suggest that the same phenomenon causes the variability in both the persistent and burst fluxes.  Since the variability of the persistent emission is not coherent, it is unlikely to be due to orbital modulation. As noted by \\cite{kong98}, its timescale (whether $\\sim 30$ or 60--70~d) is much longer than would normally be expected for the orbital period of a Roche lobe-filling low-mass X-ray binary.  Such ``super-orbital'' periodicities are observed in several similar sources \\cite[e.g.][]{white95}, and are generally attributed to variations in the accretion geometry, possibly caused by the precession of a warped accretion disk about the neutron star.  If the persistent emission is modulated at this timescale because of a slowly evolving warp in the accretion disk that is reflecting a small fraction of the X-ray luminosity of the central object to the observer, then the peak burst fluxes would also be modulated at the same timescale and with a similar amplitude. Our analysis of the radius-expansion bursts of \\src\\ strongly suggest that this is the case. Further support is provided by the highly significant correlation observed between the peak flux and the fluence for the radius-expansion bursts. If there was no additional contribution to the burst flux from disk reflection, theory predicts that the peak flux would be independent of the fluence.  It remains to be established whether, given the conditions in the \\src\\ system, the accretion disk can become warped, undergo precession at approximately the measured (quasi-) periodicities, and give rise to the observed degree of modulation of the persistent and peak radius-expansion burst fluxes.  The disk warp may arise (for example) from non-axisymmetric radiation pressure forces \\cite[e.g.][]{pringle96,mbp96,mb97}.  The conditions required for the initial warping, and steady precession thereafter, depend primarily upon the orbital separation and the efficiency of accretion \\cite[]{od01}. The orbital parameters are presently unknown for \\src, but are in principle measurable; the accretion efficiency is much more difficult to measure for this, or any other, LMXB. Thus, our present level of knowledge is not sufficient to reject the hypothesis that a precessing, warped disk (whether arising by radiation instabilities, or some other mechanism) is present in \\src. While the question of whether a disk warp can give rise to the observed modulation is comparatively more straightforward, we are still limited by the lack of measured system parameters, in particular the inclination. For a flat accretion disk \\cite{ls85} calculated an anisotropy factor of 2.8, depending upon the inclination angle. The precessing of a warped disk is likely to affect the proportion of reprocessed radiation observed during the burst in the same way that varying the inclination would. The derived anisotropy factor is more than sufficient to explain the observed modulation in the peak flux of radius-expansion bursts.  Because of these uncertainties, we can most likely adopt a relatively wide range of parameters (disk warping angle, disk albedo) which will give rise to a modulation of at least the level measured in \\src; thus, such an approach would also have no ability to rule out warped disk precession as a mechanism for the X-ray flux modulation. We note that in the archetypical precessing warped disk system Her~X-1, periodic obscuration of the neutron star by the disk gives rise to a modulation of the persistent X-ray flux of essentially 100\\% \\cite[e.g][]{slw00}, much larger than the $\\sim10$\\% measured for \\src.  Since it exhibits neither X-ray eclipses or dips, \\src\\ must have a lower inclination ($i\\la85\\arcdeg$) than Her~X-1, making obscuration by the disk less likely.  For a disk warped to the degree inferred for Her~X-1 ($20\\arcdeg$ at the outer edge), the {\\it a priori} probability for obscuration in \\src\\ is $\\sim25$\\%. Even if the inclination is not sufficiently high to permit obscuration, X-ray reflection from the disk, coupled with the variations in the projected disk area due to the precessing warp, may yet be sufficient to give rise to the observed modulation.  \\subsection{Possible anisotropy of the burst emission} \\label{aniso}  When the systematic trends in the variation of the peak burst fluxes are removed, the residual variation is only 2.8--3.2\\%, which is comparable to the typical measurement uncertainty of 2\\%. This has a very important implication for the anisotropy of radius-expansion bursts in \\src, as described by the parameter $\\xi$ in equation~(\\ref{ledd}). The small residual scatter of the peak fluxes strongly suggests that the intrinsic variation of the peak burst flux is also small, $\\simeq1$--2\\%. It seems unlikely that we observe the same face of the neutron star at the same orientation during every one of these bursts, particularly given the rapid rotation inferred from the burst oscillations \\cite[364~Hz;][]{stroh96}. Additionally, these same oscillations are almost never observed during the radius expansion episode itself, even if they are present earlier or later in the burst \\cite[]{muno02b}. We conclude that the longitudinal dependence of the burst flux during the radius expansion episodes is negligible. A latitudinal variation in flux remains plausible, particularly since the effective gravity is smaller at the neutron star equator than at the poles, and so we might expect a greater degree of expansion of the atmosphere there. However, we observe significant variation in the blackbody normalization when the peak burst flux is achieved, which suggests that the radius expansion episodes reach different peak radii. Since we might expect the degree of latitudinal anisotropy to vary with increasing radius, the effect of such a latitudinal variation of flux would be a dependence of the peak flux on the blackbody normalisation at the peak, which is not observed. Thus, we conclude that the degree of latitudinal flux anisotropy is most likely also limited by the (inferred) intrinsic variation of the peak burst fluxes.  We conclude that the burst emission during the radius expansion episode is isotropic to within $\\simeq1$--2\\%. Note that it is still possible for the burst emission at the neutron star surface to be significantly anisotropic, but that this anisotropy is smoothed out through reprocessing in the extended atmosphere present during the radius expansion episodes.     \\subsection{Consequences for distance estimates} \\label{dist}  Studies such as this provide a measure of the systematic uncertainties of the distance estimates of X-ray sources that are based solely on Eddington-limited bursts \\cite[see, e.g.,][]{vpw95}. We note that the standard deviation we measure is within the typical peak burst flux uncertainty ($\\simeq15$\\%) measured by \\cite{kuul02} for the globular cluster burst sources.  While the inferred (intrinsic) isotropy of the burst radiation (section \\S\\ref{aniso}) allows us to at least eliminate that contribution to uncertainties in distance estimates, the additional systematic error contributed by the observed scatter in the peak burst fluxes is still smaller than the usual other uncertainties due to the unknown neutron star mass and atmospheric composition. Furthermore, without a detailed understanding of the degree of reprocessing occurring in the region around the neutron star, we cannot at this time determine the intrinsic peak luminosity of the radius expansion bursts, from the broad distribution we have observed.  Nevertheless, we now calculate a probable range for the distance to \\src, given plausible values for the neutron star mass and atmospheric composition. We identify the minimum peak flux of the radius expansion bursts as the best estimate of the Eddington limit; this burst will have the smallest contribution due to reprocessed radiation, and thus will provide the best estimate of the intrinsic maximum flux. Since the peak flux is typically reached near the end of the radius contraction, we calculate the gravitational redshift parameter at the neutron star radius $R_{\\rm NS}=10$~km. We also reduce our observed fluxes by 20\\% to correct for the observed systematic flux offset measured for \\xte\\/ (see \\S\\ref{pca}), so that the inferred Eddington flux is $6.2\\times10^{-8}\\ \\epcs$. Thus, for a 1.4(2.0)~$M_\\sun$ neutron star with cosmic atmospheric abundance ($X=0.7$), the distance is 4.4(4.8)~kpc. For a pure He atmosphere the distance is 30\\% greater. These values are roughly consistent with previous estimates \\cite[]{vp78,bas84,kam89} and place the source within 12~pc of the Galactic plane, about 4~kpc from the center."
    },
    "0208/astro-ph0208187_arXiv.txt": {
        "abstract": "{\\small The Chandra AO1 HETGS observation of the micro-quasar GRS~1915+105 in the low hard state reveals (1) neutral K absorption edges from Fe, Si, Mg, and S in cold gas, and (2) highly ionized (Fe~{\\sc xxv} and Fe~{\\sc xxvi}) absorption attributed to a hot disk, disk wind, or corona. The neutral edges reveal anomalous Si and Fe abundances which we attribute to surrounding cold material in/near the environment of \\grs1915.  We also point out the exciting possibility for the first astrophysical detection of XAFS attributed to material in interstellar grains. We place constraints on the ionization parameter, temperature, and hydrogen equivalent number density of the absorber near the accretion disk based on the detection of the  H- and He-like Fe absorption.   Observed spectral changes in the ionized lines which track the light curve point to changes in both the ionizing flux and density of the absorber, supporting the presence of a flow. \\\\ {\\bf Details can be found in Lee et al., 2002, ApJ., 567, 1102 \\cite{lee02}. } } ",
        "introduction": "The {\\it Chandra} High Energy Transmission Grating (HETGS) and \\rxte Proportional Counter Array (PCA) observed \\grs1915 in the low hard state on 2000 April 24 (MJD: 51658.06654, orbital phase zero \\cite{ephemeris}) for $\\sim$~31.4~ks. The absolute absorption corrected luminosity $L_{\\rm bol} \\sim L_X \\approxgt 6.4 \\times 10^{38}\\ $~\\ergs.   The Greenbank Interferometer radio observations on 2000 April 24.54 indicate a flux of $20 \\pm 4$ mJy at 2.25 GHz, which is consistent with the presence of a steady jet.  \\begin{figure*}[h] \\includegraphics[angle=0,height=1.9in,keepaspectratio=false,width=2.6in]{lee_1.ps} \\hspace{-0.05in} \\includegraphics[angle=0,height=1.9in,keepaspectratio=false,width=2.6in]{lee_2xfig.eps}\\\\ \\caption[h]{\\footnotesize(LEFT) Photoelectric K-shell edges of S, Si and Mg.   Note the prominence of the Si edge, and possible XAFS structure.  The Fe~K edge is shown to the (RIGHT) where ionized (most likely from the accretion disk atmosphere) H-- and He--like Fe~{\\sc xxv} and {\\sc xxvi} are also seen.  Over-plotted is the best fit continuum (dashed) and identified (solid) lines. The unidentified line may be a shift in the edge due to XAFS. } \\label{fig-spec} \\end{figure*}  \\begin{figure}[b] \\parbox{8truecm} {\\psfig{file=lee_3.eps,width=8.0cm,height=4.0cm}} \\parbox{5truecm} {{\\it Figure 2.}  A possible picture of \\grs1915 near the black hole, as inferred from the \\chandra spectra. For our calculations (\\S3), we assume that the volume filling factor $\\Delta R / R$ must be small (e.g. $\\approx 0.1$)  in order that $\\xi$ not change over the region. \\label{fig-grs1915} } \\ \\vspace{-0.5truecm} \\  \\hspace{0.5truecm}     \\ \\end{figure}  \\vspace{-0.15in} ",
        "conclusions": "The \\chandra\\, HETGS and simultaneous \\rxte\\, PCA observations of \\grs1915 during the low hard state reveal : \\\\  \\noindent$\\bullet$ {\\bf Cold} material with anomalous Fe and Si abundances : There is a possibility that these abundance excesses may be related to material that is  associated with the immediate environment of \\grs1915. See also similar suggestions from IR studies (\\cite{ir1},\\cite{ir2}). Abundance anomalies (of the lighter $\\alpha$-process elements such as sulfur and oxygen) have also been reported from optical spectra of the companion stars of the other microquasars GRO~J1655--40 and V4641 Sgr (\\cite{m1},\\cite{m2},\\cite{m3}). \\\\  \\noindent$\\bullet$~{\\bf Hot} material from the accretion disk atmosphere/wind/corona is detected in the form of highly ionized H- and He-like Fe lines. \\\\ \\\\ \\noindent$\\bullet$~{\\bf Flow} : The possibility for the presence of a slow flow is suggested by variability studies.\\\\ \\\\ \\noindent$\\bullet$~{\\bf Dust} :  If confirmed, the detection of XAFS will have important implications for directly probing the structure and chemical composition of interstellar grains. \\\\"
    },
    "0208/astro-ph0208302_arXiv.txt": {
        "abstract": "We use published \\ROSAT\\ observations of the X-ray Nova V1974 Cygni 1992 to test a model for interstellar dust, consisting of a mixture of carbonaceous grains and silicate grains. The time-dependent X-ray emission from the nova is modelled as the sum of emission from a O-Ne white dwarf plus a thermal plasma, and X-ray scattering is calculated for a dust mixture with a realistic size distribution. Model results are compared with the scattered X-ray halos measured by \\ROSAT\\ at 9 different epochs, including the early period of rising X-ray emission, the ``plateau'' phase of steady emission, and the X-ray decline at late times. We find that the observed X-ray halos appear to be consistent with the halos calculated for the size distribution of Weingartner \\& Draine which reproduces the Milky Way extinction with $R_V=3.1$, provided that the reddening to the nova is $E(B-V)\\approx 0.20$, consistent with $E(B-V)\\approx 0.19$ inferred from the late-time Balmer decrement. The time delay of the scattered halo relative to the direct flux from the nova is clearly detected.  Models with smoothly-distributed dust give good overall agreement with the observed scattering halo, but tend to produce somewhat more scattering than observed at 200--300\\arcsec, and insufficient scattering at 50--100\\arcsec. While an additional population of large grains can increase the scattered intensity at 50--100\\arcsec, this could also be achieved by having $\\sim$30\\% of the dust in a cloud at a distance from us equal to $\\sim$95\\% of the distance to the nova. Such a model also improves agreement with the data at larger angles, and illustrates the sensitivity of X-ray scattering halos to the location of the dust. The observations therefore do not require a population of micron-sized dust grains.  Future observations by \\Chandra\\ and {\\it XMM-Newton} of X-ray scattering halos around extragalactic point sources can provide more stringent tests of interstellar dust models. ",
        "introduction": "}  Interstellar grains scatter X-rays through small scattering angles, and as a result distant X-ray point sources appear to be surrounded by a diffuse ``halo'' of scattered X-rays (Overbeck 1965; Martin 1970; Hayakawa 1973). The angular structure and absolute intensity of these scattered halos can be measured using imaging X-ray telescopes, thus providing a test for interstellar grain models (Catura 1983; Mauche \\& Gorenstein 1986; Mitsuda et al.\\ 1990; Mathis \\& Lee 1991; Clark et al.\\ 1994; Woo et al.\\ 1994; Mathis et al.\\ 1995; Predehl \\& Klose 1996; Smith \\& Dwek 1998; Witt et al.\\ 2001; Smith et al.\\ 2002). Given an accurate model for the dust grain size distribution and its scattering properties, observations of scattering halos can also constrain the spatial distribution of dust towards a source and the distance to a source, particularly if the emission is time variable (Tr\\\"umper \\& Sch\\\"onfelder 1973; Predehl et al.\\ 2000).   Nova V1974 Cygni 1992, a bright X-ray nova, was observed extensively by the imaging X-ray telescope on \\ROSAT, resulting in the best extant data set for studies of the X-ray scattering properties of dust (Krautter et al.\\ 1996). Mathis et al.\\ (1995) compared model calculations to the X-ray halo observed 291 days after optical maximum, and argued that the angular structure of the observed X-ray halo favored a grain model based on highly porous grains. Smith \\& Dwek (1998) disagreed with this conclusion, arguing that the halo around Nova Cygni 1992 did not require porous grains, but was in fact consistent with the scattering expected from a mixture of nonporous silicate and carbon grains. More recently, Witt, Smith, \\& Dwek (2001) reached a different conclusion, arguing that the observed X-ray halo around Nova Cygni 1992 requires that the size distribution of interstellar dust grains extend to radii $a\\geq2.0\\micron$, with $\\gtsim$40\\% of the dust mass in grains with radii $a > 0.5\\micron$.  Weingartner \\& Draine (2001) and Li \\& Draine (2001) have recently put forward a physical dust model which is in quantitative agreement with the wavelength-dependent extinction of starlight as well as the observed spectrum of infrared emission from interstellar dust. The model consists of a mixture of carbonaceous grains (including ultrasmall grains with the properties of polycyclic aromatic hydrocarbon molecules) and amorphous silicate grains. By appropriate adjustment of the size distribution, the model can reproduce the extinction in different regions of the Milky Way, and in the Large and Small Magellanic Clouds. Li \\& Draine (2002) show that the model is also consistent with the observed infrared emission from the Small Magellanic Cloud. Here we use the observed X-ray halo around Nova Cygni 1992 to test this dust grain model. Since the sightline toward Nova Cygni 1992 is presumably typical diffuse interstellar medium, we use the Weingartner \\& Draine (2001, hereafter WD01) size distribution for Milky Way dust with $R_V=3.1$.  In \\S\\ref{sec:distance} we review estimates of the distance to and the gas and dust toward Nova Cygni 1992. An empirical model for the X-ray emission from the nova is described in \\S\\ref{sec:spectrum}, with the emission modelled as the sum of emission from a hot thermal plasma plus a white dwarf photosphere with varying temperature and radius. The methodology for calculation of X-ray scattering by dust is presented in \\S\\ref{sec:scatter}, including multiple scattering, the effects of time delay, and the calculation of dust scattering cross sections.  Our results are presented in \\S\\ref{sec:results}. We test our model using observations from 9 different epochs, at radii out to 2000\\arcsec. We show that the WD01 model is in quite good agreement with the observations if the dust is assumed to follow an exponential density law and the nova is at a distance of $\\sim$2.1~kpc. The time delay of the halo relative to the nova is clearly visible at late times when the nova is in decline. We discuss the uncertainties associated with possible clumping of the dust into clouds along the line-of-sight, and show that agreement with the observed halos can be improved if $\\sim30\\%$ of the dust is concentrated in a cloud $\\sim100\\pc$ from the nova. We conclude that the WD01 dust model is consistent with the observed X-ray halo toward Nova Cygni 1992, and a population of large dust grains is not required.  The distance estimate to the nova depends somewhat on the assumed density distribution of dust, and is considered in more detail in a separate paper (Draine \\& Tan 2003). ",
        "conclusions": "}  Figure \\ref{fig:compare} shows the importance of various features of the modelling on the intensity of the X-ray halo. The curve labelled ``ref'' is the single-scattered intensity for a model where the source is radiating steadily at a single energy $h\\nu=400\\eV$, with uniform dust density between observer and source. Other curves show the effect of including multiple scattering, time delay, and a realistic nova spectrum (same point source count rate), and replacing the uniform dust distribution with an exponential distribution. \\begin{itemize} \\item Doubly-scattered photons add $\\sim$10\\% to the intensity at $\\theta\\approx600\\arcsec$, and $\\sim$20\\% at $\\theta\\gtsim1200$\\arcsec. \\item Changing from a uniform to an exponential dust distribution reduces the intensity for $\\theta\\ltsim450$\\arcsec\\ (less dust close to the source), and increases the intensity for $\\theta\\gtsim 450$\\arcsec\\ (more dust far from the source). The increase is $\\sim$10\\% for $\\theta>800$\\arcsec. \\item Replacing the monochromatic spectrum with a realistic spectrum reduces the flux at large angles; the adopted $h\\nu=400\\eV$ is probably slightly low compared to the more realistic spectrum (see Fig.\\ \\ref{fig:compare_spec}), resulting in slightly more large angle scattering than for the more realistic spectrum. \\item Since the light curve is rising at day 291, including time delay leads to a reduction in the intensity, particularly for larger halo angles for which the time delays are larger. At 1000\\arcsec, inclusion of time delay reduces the intensity by $\\sim$30\\%. \\end{itemize} Some of these effects increase, others decrease, the intensity. When all are included together, the intensity for day 291 is reduced at angles $>60$\\arcsec, with a reduction by $\\sim$30\\% at 1000\\arcsec. \\begin{figure}[h] \\begin{center} \\epsfig{ file=f16.cps, angle=270, width=5.0in} \\end{center} \\vspace*{-2.em} \\caption{ \\label{fig:compare} \\footnotesize $\\pi\\theta^2I$ for reference model: $h\\nu=400\\eV$, uniformly-distributed dust. single scattering only, time delay neglected (curve labelled ``ref''). Other curves show effect of adding multiple scattering, changing from monochromatic to realistic nova spectrum; including effects of time delay for scattered photons, for $D=2.1$kpc; and changing from a uniform density to exponential density profile with $z_{1/2}=300$pc. Curve labelled ``all'' is exponential density model with full spectrum, multiple scattering, and time delay. } \\end{figure}  For smoothly-distributed dust and a standard dust-to-gas ratio, our modelling strongly favors a column density $N_{\\rm H}^{\\rm ISM}\\approx 1.15\\times10^{21}\\cm^{-2}$ between us and the nova; this is only $\\sim$50\\% of the estimated total gas column in this direction. This means that the nova must be close enough to have $\\sim$50\\% of the column density beyond it. For the exponential distribution (eq. \\ref{eq:exp}), this places it at a distance $\\sim2.1(z_{1/2}/300\\pc)\\kpc$. Since the actual variation of gas density as a function of height is not well known, and since much of the gas and dust is likely to be contributed by discrete clouds, it is possible for the nova distance to be as large as $2.5\\kpc$ and still have $\\sim$50\\% of the gas and dust beyond the nova.  Most estimates of the reddening to the nova have been larger than the value $E(B-V)=0.20$~mag which we favor. Austin et al.\\ took the weighted mean of 5 different methods, and obtained $E(B-V)=0.36\\pm0.04$~mag. Our model with $E(B-V)=0.20$~mag has good overall agreement with the observed halo intensities, but an increase in $E(B-V)$ to 0.36 would result in halo intensities almost twice as strong as observed. If the reddening were shown to be $E(B-V)\\gtsim0.3$ toward Nova Cygni 1992, this would rule out the WD01 dust grain model which we have used here {\\it if} the dust is smoothly distributed. However, we note that the reddening estimated from the Balmer decrement at late times is consistent with our estimate.  Furthermore, we show that a plausible modification of the spatial distribution of the dust can produce good agreement between the observed and calculated halos: if $\\sim30\\%$ of the dust (i.e., $E(B-V)=0.06$) is located in a ``cloud'' within $\\sim 100\\pc$ of the nova, the calculated scattering halo is in good agreement with observation. Since such a spatial distribution is not improbable, the observed X-ray halo around Nova Cygni 1992 does not require the existence of a population of large grains. Recent observations by \\Chandra\\ of the scattered halo around GX 13+1 also do not support an additional population of large grains (Smith et al.\\ 2002).  While we have shown that the WD01 dust model is generally consistent with observations of Nova Cygni 1992, this conclusion would be overturned if the larger estimates of the reddening prove to be correct.\\footnote{% X-rays scattered by dust {\\it very} close to the nova would be lost in the point spread function. At 0.5 keV the median scattering angle $\\Theta_m\\approx 1000\\arcsec$, so most of the photons scattered by dust located at $x = 0.97$ would be at halo angles $\\theta < 30\\arcsec$. Thus dust located at $x > 0.97$ would not make an observable contribution to the X-ray halo. } Unfortunately, Nova Cygni 1992 is a less-than-ideal test of a dust model: as we have seen, there is uncertainty regarding the value of the reddening $E(B-V)$ (i.e., the total amount of dust); in addition, there is uncertainty regarding the distribution of gas and dust along the sightline to Nova Cygni 1992 -- it was plausible to consider that $\\sim$30\\% of the dust might be contributed by a cloud $\\sim$100pc from the nova.  The ideal way to study scattering by interstellar dust is to employ an X-ray source well outside the galactic plane, so that most of the dust contributing to the scattering is at $y\\ltsim 0.1$, in which case the halo angle $\\theta$ and scattering angle $\\Theta_s$ are nearly the same for single-scattering. An extragalactic source would be ideal for this purpose. In this case, uncertainties regarding the precise location of the dust in the Galactic disk are unimportant (i.e., dust at $y=0$ and dust at $y=0.001$ produce virtually identical halos). We can hope that the great sensitivity of \\Chandra\\ and \\XMM\\ may make feasible observations of such out-of-plane sources. Alternatively, a source located in the galactic plane at $15^\\circ \\ltsim |l| \\ltsim 75^\\circ$ would allow HI or CO observations, or optical absorption line spectroscopy, to locate the absorbing material using galactic rotation.  The higher angular resolution of \\Chandra\\ may allow observations at smaller halo angles, and the improved energy resolution will allow study of the energy spectrum of the scattered halo."
    },
    "0208/astro-ph0208072_arXiv.txt": {
        "abstract": "Brown dwarfs occupy the important region in the mass range between stars and planets. Their existence, ambigious until only recently, and their properties give insight into stellar and planetary formation. We present statistical results of an infrared, coronagraphic survey of young, nearby stars that includes probable companions to three young G-type stars, Gl 503.2 (G2V), HD 102982 (G3V), and Gl 577 (G5V). The companion to Gl 577 is a possible binary brown dwarf, according to evolutionary models. A dynamical determination of the components' masses will be achievable in the near future and be an excellent test of the predictive ability of the evolutionary models. ",
        "introduction": "To explore fully the similarities and differences between stellar and planetary formation, the study of brown dwarfs as companions is essential. Since the primary's age and distance have been determined, these two properties, usually uncertain for field brown dwarfs, are already known. The primary goal of this survey of 45 young (t$<$300Myr), nearby (d$<$50pc) stars was the discovery of substellar companions. Because substellar objects are hotter and brighter when younger, they can be more easily detected. Each target star was selected based on its being single and having at least two youth criteria (chromospheric activity, lithium, proper motion) indicating a young age. The final sample had a median age of 150 Myr and a median distance of 30 pc. The survey utilized the infrared coronagraph on the Near-Infrared Camera and Multi-Object Spectrometer (NICMOS) on the Hubble Space Telescope (HST). Earlier results include the brown dwarf companions TWA5B (Lowrance et al 1999; L99) and HR 7329B (Lowrance et al 2000). ",
        "conclusions": ""
    },
    "0208/astro-ph0208354_arXiv.txt": {
        "abstract": "We observed the ``micro-quasar'' \\Grs\\ four times with \\chandra.  Two HRC-I observations were made in 2000 September-October spanning an intermediate-to-hard spectral transition (identified with \\rxte).  Another HRC-I and an ACIS/HETG observation were made in 2001 March following a hard-to-soft transition to a very low flux state.  Based on the three HRC images and the HETG zero order image, the accurate position (J2000) of the \\xray\\ source is RA $=$ \\grsra, Dec $=$ \\grsdec\\ (90\\% confidence radius $=$ 0\\asec.45), consistent with the purported variable radio counterpart.  All three HRC images are consistent with \\Grs\\ being a point source, indicating that any bright jet is less than \\aprx1\\,light-month in projected length, assuming a distance of 8.5\\,kpc. ",
        "introduction": "\\Grs\\ and its sister source, \\onee, were the first objects dubbed ``micro-quasars'' \\citep{Mir92,Mir94}.  Their \\xray\\ spectra are typical of Galactic black hole candidates (BHCs), and they are apparently associated with the time variable cores of arcminute scale double-lobed radio sources, reminiscent of extra-Galactic radio sources.  This morphology, seen on a parsec scale within the Milky Way, earned them their nickname. \\Grs\\ and \\onee\\ are the brightest persistent sources in the Galactic bulge above $\\sim$50~keV \\citep{Su91}.  Their timing characteristics are typical of the black hole low/hard state \\citep{Mai99,Smi97,Hei93,Su91}, and they consistently emit near their brightest observed levels, although they vary over times of days to years.  Their \\xray\\ emission properties are readily likened to the canonical BHC, \\cyg.  In fact, together with \\cyg, they are the only known persistent, low-state BHCs, and all three sources have maximum luminosities around $3 \\times 10^{37}$ergs s$^{-1}$. Radio jets have now been observed in \\cyg, furthering the similarity \\citep{Fen00}.  \\Grs\\ and \\onee\\ are, however, quite different from the Galactic {\\em superluminal} radio sources (e.g. GRS~1915+105 and GRO~J1655-40) more typically thought of as micro-quasars.  The \\xray\\ emission from these objects is much brighter and more spectacularly variable.  Their radio jets, too, are brighter and are highly variable, being unresolved or absent except during exceptional ejection events which last only weeks. In contrast, the radio lobes of \\Grs\\ and \\onee\\ are quite stable \\citep{Marti02}.  The association of \\Grs\\ and \\onee\\ with their corresponding radio counterparts has been based primarily on relatively coarse (\\aprx10\\,\\asec) \\xray\\ positions and the very unusual (and yet nearly identical) nature of the radio sources.  Some hint of correlated radio and hard X-ray emission was found for \\onee\\ \\citep{Mirabel93}, and recently \\citet{Cui01} confirmed the \\xray/radio association for \\onee\\ by using \\chandra\\ to obtain a precise \\xray\\ position.  In this \\emph{Letter} we do the same for \\Grs, and using the fine resolution of the High Resolution Camera (HRC) place limits on emission from any arcsecond scale jets. \\begin{figure} \\plotone{f1.eps} \\caption{\\label{f:lc}The \\rxte/PCA light curve of \\Grs\\ in two energy bands. Our \\chandra\\ observations (see Table~1) are indicated by dashed vertical lines.  Observations 400163 and 400164 were made consecutively and so appear as a single line near 2001.2. The 1996-1997 flux history appears very similar to 1998 with the source remaining quite stable within a factor of \\aprx2.} \\end{figure} ",
        "conclusions": "\\citet{Gol01} recently used \\emph{XMM} data to derive a 5\\asec\\ error radius for \\Grs. This work, which is consistent with their result, reduces the area of the error region by two orders of magnitude. The coincidence of the sub-arcsecond VLA and \\chandra\\ error circles seals the association of \\Grs, the \\xray\\ source, with the variable radio source \\citep[``VLA-C'', ][]{Mar98}. From radio source counts \\citep{Condon84}, we estimate a chance probability of \\aprx$10^{-5}$ for a radio source brighter than VLA-C (0.2\\,mJy) to fall in the \\chandra\\ error circle.  \\citet{Mar98} identified two candidate counterparts to VLA-C in I and K band images. The brighter and closer candidate was found through multi-band photometry and near infrared spectroscopy most likely to be an early K giant.  Revised astrometry \\citep{Roth02} of infrared observations by \\citet{Eik01} confirm that this star (labeled `A') is consistent with VLA-C at the 3$\\sigma$ level.  The second candidate of \\citet{Mar98} is likely a main sequence F star, but is inconsistent with the astrometry of \\citet{Roth02}.  Furthermore, \\citet{Smi02} have found a low amplitude (\\aprx4\\%), $18.45\\pm 0.10$\\,dy periodic modulation of the \\Grs\\ \\xray\\ emission observed with \\rxte.  This period fits well with the orbit of a Roche lobe filling K giant companion \\citep{Roth02}, and so a more complete picture of \\Grs\\ is emerging.  With the \\xray/radio association confirmed and periodic \\xray\\ emission pointing to a K giant, \\Grs\\ appears to be a black hole with an intermediate mass giant companion.  GRS~1915$+$105 \\citep{Gre01}, GRO~J1655$-$40 \\citep{Shahbaz99}, XTE~J1550$-$564 \\citep{Orosz02}, and now \\Grs\\ -- all four of the black hole microquasars in low mass systems -- have evolved companions.  Meanwhile, jets have not been observed in the more common black hole soft \\xray\\ transients (SXTs) with main sequence companions.  This prompts us to examine the connection between giant companions and jet activity. Perhaps an evolved companion overflows its Roche lobe by a large factor, providing a higher typical accretion rate. These four objects are persistent or quasi-persistent, having very different time histories from the standard SXT ``fast rise and exponential decay'' light curve.  The SXT outbursts are driven by an accretion instability following long periods of relatively low accretion rates \\citep[see for example, ][]{Cannizzo98}.  Possibly steady, high rate accretion provides an environment conducive to jet formation.  Finally, we note that the present lack of \\xray\\ jets is completely consistent with the energetics of the radio lobes. \\citet{Ro92}, assuming a distance of 8.5\\,kpc, estimated the luminosity of the radio lobes to be \\aprx$10^{30}$\\,\\lumin and their energy content to be roughly $10^{44}$\\,ergs.  With a typical \\xray\\ luminosity of a few$\\times 10^{37}$\\,\\lumin, the present lobes could have been energized by a small fraction of the \\xray\\ power over a period as short as a few years. In fact, given the long lifetime (\\aprx$3\\times10^{6}$\\,yr) of the radio lobes, this indicates that either the duty cycle for or efficiency of feeding the jets (or both) is very low."
    },
    "0208/astro-ph0208162_arXiv.txt": {
        "abstract": "We have obtained deep imaging in the near-infrared J and K bands for 2 nearby fields in the bar of the LMC with the ESO NTT telescope, under exquisite seeing conditions. The K, J-K color-magnitude diagrams constructed from these data are of outstanding photometric quality and reveal the presence of several hundreds of red clump stars.   Using the calibration of Alves for the K-band absolute magnitude of {\\it Hipparcos}-observed red clump stars in the solar neigbourhood we derive a distance modulus to our observed LMC fields of 18.487 mag. Applying a correction for the tilt of the LMC bar with respect to the line of sight according to the geometrical model of van der Marel et al., the corresponding LMC barycenter distance is 18.501 mag. The random error on this result is +/- 0.008 mag, whereas the systematic uncertainty on this distance result due to the photometric zero point uncertainty in our LMC K-band photometry, to the uncertainty of the {\\it Hipparcos}-based absolute magnitude calibration of red clump stars in the solar neighborhood, and due to reddening uncertainties mounts up to +/- 0.048 mag.  If we adopt a K-band population correction of -0.03 mag, as done by Alves et al. 2002, to account for the difference in age and metallicity between the solar neighborhood and LMC red clump star populations, we obtain an LMC barycenter distance modulus of 18.471 mag from our data. This is in excellent agreement with the result of Alves et al., and of another very recent study of Sarajedini et al. (2002) obtained from K-band photometry. However, we emphasize that current model predictions about the uncertainties of population corrections seem to indicate that errors up to about 0.12 mag may be possible, probably in any photometric band. Therefore, work must continue to tighten the constraints on these corrections.  We also determine the mean red clump star magnitude in our LMC fields in the J band, which could be a useful alternative to the K band should future work reveal that population effect corrections for red clump stars in the J band are smaller, or more reliably determined than those for the K band. ",
        "introduction": "We have recently started on an ambitious project to improve the calibrations of a number of stellar distance indicators which can be used to determine the distances to nearby galaxies, located at $\\lesssim$ 10 Mpc. This project, named the ARAUCARIA project, is particularly focussing on the dependence of the various stellar distance indicators on galaxy environmental properties, like metallicity and age of the stellar populations. Among the objects whose usefulness for an accurate distance measurement we want to improve within the ARAUCARIA project are Cepheid variables, RR Lyrae stars, red clump stars, blue supergiants, and planetary nebulae. These stellar distance indicators span a large range in masses, typical ages, and evolutionary states, and in any galaxy at least some of these object classes can be found. In the Local Group and other nearby galaxy groups like Sculptor, there are a number of galaxies which contain stellar populations of all ages (old, intermediate, young) and in which all of our standard candles can be found together-these are our principal target galaxies for establishing the possible environmental dependences of our stellar standard candles more accurately than hitherto done. The aims of the project have been described in more detail in Gieren et al. (2001), and first results have recently been published in Pietrzy{\\'n}ski et al. (2002).  As a part of the ARAUCARIA project, we are conducting a deep near-infrared (JK) imaging survey in several of our target galaxies, including the Magellanic Clouds, and the Carina and Fornax dwarf galaxies. An immediate aim is to use red clump stars, and the tip of the red giant branch technique to determine the distances to these galaxies. In this paper, we are presenting a distance determination to the Large Magellanic Cloud from the red clump star method, applied in the pure infrared domain (from a K, J-K color-magnitude diagram). The LMC is arguably the most important galaxy in the process of establishing an accurate extragalactic distance scale, and our current lack of ability to measure extragalactic distances is best reflected in the large range of distances measured for the LMC with different techniques (e.g. Gibson, 2000). Regarding the red clump stars as a distance indicator, the method was introduced in a fundamental paper of Paczy{\\'n}ski \\& Stanek (1998), and the red clump star mean brightness was calibrated in the optical V and I bands in their paper. Important progress has recently been made by Alves (2000) who was the first to calibrate the absolute magnitudes of a large sample of {\\it Hipparcos}-observed nearby red clump stars in the K-band. An outstanding and obvious advantage of the red clump star technique is the fact that accurate distances were measured by {\\it Hipparcos} to nearly a thousand of these stars in the solar neighborhood, making the red clump star method the currently only technique having a zero point which is set from an accurate geometrical calibration (Paczy{\\'n}ski \\& Stanek 1998). While previous efforts to calibrate the red clump star absolute magnitude in the optical I-band, and apply it to other galaxies were plagued by problems with extinction corrections, and with a small, but significant dependence of $M_{\\rm I}$ on metallicity, which manifested themselves in considerable differences between red clump distances to the LMC using I-band magnitudes (e.g. Udalski et al. 1998; Udalski et al. 2000; Romaniello et al. 2000), the use of K-band magnitudes minimizes these problems, and in the case of the LMC virtually eliminates any dependence of the final K-band red clump distance result on the adopted reddening. In  the very recent work of Grocholski \\& Sarajedini (2002) who studied population effects on the K-band magnitude of red clump stars from 2MASS photometry of open clusters it was suggested that $M_{K}$ may be  less dependent on stellar age and metallicity than $M_{I}$. Therefore, it is of a profound importance to verify in detail possible population effects on the mean  K-band red clump magnitude in order to be able to  derive truly accurate distances with this method.   In the following, we describe what we believe is the as yet most accurate set of near-IR data for samples of red clump stars in 2 fields in the bar of the LMC, obtained with the {\\it New Technology Telescope} on La Silla under superb seeing conditions, analyze these data together with optical photometry of the red clump stars in these fields taken from OGLE II databases,  and derive the distance to the LMC. We thoroughly discuss the accuracy of our distance determination, including the effect of environmental corrections on our (and other) results. In forthcoming papers, we will extend this work to a number of galaxies in the Local Group, in an effort to tighten the constraints upon the appropriate corrections which have to be applied to the observed red clump star magnitudes to account for differences in the red clump star populations we are observing in these galaxies, and in the solar neighborhood. ",
        "conclusions": "Population effects on red clump star absolute magnitudes in different environments, including the solar neighborhood and the LMC, have been studied in some detail by Girardi et al. (1998), and by Girardi and Salaris (2001). In particular, under the assumption of a star formation history (SFH) and chemical evolution model for the solar neighborhood and the LMC the latter authors found from population synthesis models that the distance moduli derived from a comparison of the mean magnitude of the red clump stars in the LMC to the corresponding mean magnitude of the red clump stars in the Hipparcos sample should be corrected by -0.03, 0.2 and 0.3 mag in K, I and V, respectively, in order to compensate for the difference in metallicity and age between these two environments. A corresponding -0.03 mag correction has been applied by Alves et al. (2002) to their K-band data to allow for the population effect. Applying the same correction, our LMC barycenter distance result becomes 18.471 mag, in excellent agreement with their result.  Unfortunately, a discussion of the accuracy of such corrections was not presented in Girardi and Salaris (2001). An estimation of the uncertainties on these corrections should be based on population synthesis calculations performed for very different SFHs and chemical evolution models, and should also take into account current uncertainties on the input physics of the stellar models used, which is a task far beyond the scope of this paper. However, we can make a rough estimate of the uncertainties of populations corrections using the published simulations of Girardi et al (1998) and Girardi and Salaris (2001) for the I band. In spite of the fact that they performed their simulations for a very limited range of possible SFHs and chemical evolution models, their data show that changing these parameters one can obtain changes in the synthetic mean brightness of red clump stars  of 0.115 mag (see Table 4 in Girardi and Salaris 2001). This is in agreement with the error estimate of about 0.1 mag for the model calculation results which was earlier given in Girardi et al. (1998; their section 5). So far no similar data for the K band are available, but it seems reasonable to assume that the current uncertainty on the population correction in the K band may be of a similar order of what is extracted from the work of Girardi and coworkers for the I band. It is therefore of great importance to reduce the current uncertainty on population corrections for red clump stars, particularly in near-infrared bands, in order to make these objects truly superb standard candles. We note, for example, that Udalski (1998, 2000) derived an I-band population correction for LMC red clump stars of only 0.04 mag, based on a thorough study of red clump stars in LMC clusters, which is quite different from the 0.2 mag correction derived from the models of Girardi and Salaris and suggesting that there is still room for significant improvement in the future from both the model, and the observational side. A reduction of these uncertainties is likely to reconcile the apparent discrepancies of the LMC distance modulus obtained in the past by different workers from data in different bands. We expect that our ongoing near-infrared work on red clump stars in other Local Group galaxies which span a range in environmental properties will lead, in the near future, to such an improved empirical calibration of the dependence of red clump star magnitudes on the stellar ages and metallicities."
    },
    "0208/astro-ph0208448_arXiv.txt": {
        "abstract": "From 1991 until 1997, the 3.8m UK Infrared Telescope (UKIRT) underwent a programme of upgrades aimed at improving its intrinsic optical performance. This resulted in images with a FWHM of 0.\\hspace*{-1mm}$^{\\prime\\prime}$17 at 2.2 $\\mu$m in September 1998. To understand and maintain the improvements to the delivered image quality since the completion of the upgrades programme, we have regularly monitored the overall {\\it atmospheric} seeing, as measured by radial displacements of subaperture images (i.e. seeing--generated focus fluctuations), and the {\\it delivered} image diameters. The latter have been measured and recorded automatically since the beginning of 2001 whenever the facility imager UFTI (UKIRT Fast Track Imager) has been in use.  In this paper we report the results of these measurements. We investigate the relation between the delivered image diameter and the RMS atmospheric seeing (as measured by focus fluctuations, mentioned above). We find that the best seeing occurs in the second half of the night, generally after 2am HST and that the best seeing occurs in the summer between the months of July and September. We also find that the relationship between $Z_{rms}$ and delivered image diameter is uncertain. As a result $Z_{rms}$ frequently predicts a larger FWHM than that measured in the images.  Finally, we show that there is no correlation between near--infrared seeing measured at UKIRT and sub--mm seeing measured at the Caltech Submillimetre Observatory (CSO). ",
        "introduction": "\\label{sect:intro}  %  Between 1991 and 1997 the UK Infrared Telescope (UKIRT) was the subject of a systematic campaign of improvements, the UKIRT upgrades programme, which had the explicit goal of providing imaging performance competitive with the best ground--based facilities (Hawarden et al. 1994, 1996, 1998). By active control of the primary--mirror figure and secondary--mirror alignment (Chrysostomou et al. 1998) and tip--tilt actuation of the secondary mirror under control of a fast guider, this goal has been met (Hawarden et al. 1999, 2000). The telescope routinely delivers images with FWHM less than 0.\\hspace*{-1mm}$^{\\prime\\prime}$6. In 1998 a programme to measure the delivered seeing was undertaken. Images were taken with the Infrared Camera, IRCAM, and the median seeing was found to be 0.\\hspace*{-1mm}$^{\\prime\\prime}$45. Some images obtained in this programme in September 1998, had a delivered FWHM of 0.\\hspace*{-1mm}$^{\\prime\\prime}$17.  The new facility imager, the UKIRT Fast Track Imager (UFTI), was delivered at the end of 1998. At the beginning of 2001, a programme to regularly monitor the delivered seeing with UFTI was initiated. In this paper we present results from this programme and compare them with pure atmospheric seeing, as measured from the RMS in the focus position, $Z_{rms}$. ",
        "conclusions": "The main conclusions of this work are summarized as follows:  \\begin{itemize}  \\item The best seeing occurs in the second half of the night, generally after 2am HST.  \\item The best seeing occurs in the summer, between the months of July and September. This seems to be consistent with both 2000 and 2001, and also 1997, when seeing of 0.\\hspace*{-1mm}$^{\\prime\\prime}$17 was achieved in September.  \\item The relationship between $Z_{rms}$ and delivered image diameter is uncertain. When mean $Z_{rms}$ is plotted versus seeing bins, the relationship between the two fits the empirical model quite well. Therefore, for a large dataset, the mean value of $Z_{rms}$ agrees well with the inferred seeing, but in an individual measurement, the inferred seeing can disagree with the measured image FWHM by up to $\\sim$40\\%, due to the large scatter in $Z_{rms}$ with respect to the delivered image diameter. This impacts on how one uses $Z_{rms}$ to predict the seeing with a flexibly scheduled telescope.  \\item There is no correlation between CSO submillimetre seeing and near--infrared seeing.  \\end{itemize}"
    },
    "0208/astro-ph0208212_arXiv.txt": {
        "abstract": "We interpret  the correlation over five orders of magnitude between high frequency $\\nu_{high}$ and low  frequency $\\nu_{low}$ in  a quasi-periodic oscillations (QPO)  found  by Psaltis, Belloni \\& van der Klis (1999) for black hole (BH), neutron star (NS) systems and then extended by Mauche (2002) to white dwarf (WD) binaries. The observed correlation strongly constrains theoretical models and provides clues to understanding the nature of the QPO phenomena at large. We argue that the observed correlation is a natural consequence of the Keplerian disk flow adjustment to the innermost sub-Keplerian boundary conditions near the central object which ultimately leads to the formation of the sub-Keplerian transition layer (TL) between the adjustment radius and the innermost boundary (the star surface for NS and WD and the horizon for BH). In the frameworks of the TL model $\\nu_{high}$ is related to the  Keplerian  frequency at the outer (adjustment) radius $\\nu_{\\rm K}$ and $\\nu_{low}$ is related to the magnetoacoustic oscillation frequency $\\nu_{MA}$. Using  a relation between $\\nu_{MA}$ the magnetic and gas pressure  and  the  density and  the hydrostatic equilibrium condition in the disk we  infer a linear correlation between $\\nu_{\\rm K}$ and $\\nu_{MA}$, Identification of $\\nu_{high}$, $\\nu_{low}$ with $\\nu_{\\rm K}$, $\\nu_{MA}$ respectively,  lead us to the determination of $H/r_{out}=1.5\\times10^{-2}$ and $\\beta=0.1$ (where H is the half-width of the disk and $\\beta$ is a ratio of magnetic pressure to the gas pressure). We estimate the magnetic field strength near the TL outer radius for BHs NSs and WDs.   The fact that the observed high-low frequency correlation over five orders of magnitude is valid for BHs, NSs, and down to WDs strongly rules out  relativistic models for QPO phenomena. We come to the conclusion that the QPOs observations indicate  the  adjustment of the geometrically thin disk to sub-Keplerian motion near  the central object. This effect is a common feature for a wide class of systems, starting from  white dwarf binaries up to black hole binaries. ",
        "introduction": "\\label{sec:intro} Quasi-periodic oscillations (QPOs) in the bright white dwarf (WD), neutron star (NS)  and black hole (BH)  binaries provide invaluable information on the accretion dynamics in the innermost parts of these systems. The question is  what kind of  objective information we can extract from  observations using the fundamental principles of  fluid physics,  radiative transfer theory and oscillatory processes? In this sense the discovery of low 20-50 Hz QPOs in the luminous NS binaries (van der Klis et al. 1985), and discoveries of NS kilohertz QPOs by Strohmayer et al. (1996) and of BH hectohertz by Morgan, Remillard \\& Greiner (1997) opened a new era in the study of the dynamics near compact objects. Psaltis, Belloni \\& van der Klis (1999), hereafter PBK,  demonstrated that these NS and BH low and high frequencies follow a tight correlation. Their result suggests  that the same types of variability may occur in both neutron star and black systems over 3 orders of magnitude in frequency. These two features are ``horizontal branch oscillation'' (HBO) along  with ``low frequency noise Lorentzian'', $\\nu_{low}$ and the lower kHz QPO $\\nu_{high}$ respectively. Belloni, Psaltis \\& van der Klis (2002), hereafter BPK,  have updated PBK's correlation adding more points from Nowak (2000), Boirin et al. (2000), Homan et al. (2002), Di Salvo et al. (2001) and  Nowak et al. (2002) data.    PBK suggest that the low and high frequencies correlate in a way that seems to depend only weakly on the properties of the sources, such the mass, magnetic field, or possibly the presence of a hard surface in compact object. At large scales (over 3 orders of magnitude) the PBK-BPK correlation for NS and BH systems is fitted by a linear function $\\nu_{low}=0.081\\nu_{high}$ and it also  has a fine structure at smaller frequency scales (see Fig.1).  Recently Mauche (2002) has reported low and high frequency QPOs in the dwarf nova SS Cygni and VW Hyi (see  Wouldt \\& Warner 2002 for  for the VW Hyi observations), and {\\it has called attention to the fact that these QPOs do extend the correlation of PBK downward in frequency by more than two orders of magnitude} (see Fig.1).  In fact, Titarchuk, Lapidus \\& Muslimov (1998), hereafter TLM98 were first to put forth the possibility of dynamical adjustments of a Keplerian disk to the innermost sub-Keplerian boundary conditions to explain most observed QPOs in bright low mass X-ray binaries (LMXBs). TLM98 concluded that an isothermal sub-Keplerian transition layer between the NS surface and its last Keplerian orbit forms as a result of this adjustment. The TLM98 model is a general treatment  applicable to both NS and black hole  systems. The primary problem in both NS and BH systems is understanding how the flow changes from pure Keplerian to the sub-Keplerian as the radius decreases to small values. TLM98  suggested that the discontinuities and abrupt transitions in their solution result from derivatives of quantities such as angular velocities (weak shocks). Titarchuk \\& Osherovich (1999) and Titarchuk et al. (2001) interpreted the low noise  frequencies as viscous oscillations and expected them to provide reliable estimates of radial velocities that would be very close to the magnetoacoustic (MA) velocities. The formulation of the MA problem and derivation of MA frequencies are presented by Titarchuk, Bradshaw \\& Wood (2001), hereafter TBW. They demonstrate that the observed correlation of the low frequency on the low kHz frequency is perfectly described by the dependence of the inferred MA frequency on the Keplerian frequency.  In this {\\it Letter}, we explain   a general correlation between low frequency and high frequency oscillations in the TL. In \\S 2, we refer to  the QPO  data obtained from  NSs and BHs by PBK and BPK, from the dwarf nova SS Cygni  by Mauche (2002) and from VW Hyi by Woudt \\& Warner (2002). The formulation of the problem of low frequency oscillations and the relationship with MA oscillations in the transition layer are described in \\S 3. The derivation of the low-high frequency correlation is also  present in \\S 3. Our summary and conclusions follow in \\S 4. ",
        "conclusions": "We have presented a model which treats the radial oscillations in the transition layer surrounding a black hole,  neutron star and white dwarf.  (1) We found  that the observed high-low frequency correlation is a natural consequence of an adjustment of the Keplerian disk flow  to the innermost sub-Keplerian boundary conditions near the central object.  This ultimately leads to the formation of the sub-Keplerian transition layer (TL) between the adjustment radius and the innermost boundary (the star surface for NS and WD and the horizon for BH). (2) In the framework of the TL model $\\nu_{high}$ is related to the  Keplerian frequency at the outer (adjustment) radius $\\nu_{\\rm K}$ and $\\nu_{low}$ is related to the magnetoacoustic oscillation frequency $\\nu_{MA}$. Using  a relation between $\\nu_{MA}$ the magnetic and gas pressure  and  the  density and  the hydrostatic equilibrium condition in the disk we  infer a linear correlation between $\\nu_{\\rm K}$ and $\\nu_{MA}$ (see formula 12). (3) Identification of $\\nu_{high}$ and  $\\nu_{low}$ with $\\nu_{\\rm K}$ and  $\\nu_{MA}$ respectively,  lead us to the determination of $H/r_{out}\\sim 1.5\\times10^{-2}$. (4) We present strong arguments that the ratio $\\beta=P_M/P_g\\approx 0.1$  in all  binaries studied. (5) We estimate the magnetic field strength near the TL outer radius for BHs, NSs and WDs.  {\\it (6) The fact that the observed  correlation  of high and low frequencies holds over  five orders of magnitude shows that the QPO frequencies cannot be set by general relativistic effects, because they are essentially non-relativistic in the white dwarf case.}   We acknowledge Chris Mauche and Tomaso Belloni for kindly supplying us data shown in Figure 1 and Chris Shrader, Paul Ray  and  particularly, Chris Mauche for fruitful discussions that motivated this paper."
    },
    "0208/astro-ph0208196.txt": {
        "abstract": "We report the results of a 20 ks \\chan ACIS-S observation of the galaxy pair NGC 4485/4490.  This is an interacting system containing a late-type spiral with an enhanced star formation rate (NGC 4490), and an irregular companion that possesses a disturbed morphology.  A total of 29 discrete X-ray sources are found coincident with NGC 4490, but only one is found within NGC 4485.  The sources range in observed X-ray luminosity from $\\sim 2 \\times 10^{37}$ to $4 \\times 10^{39}$ erg s$^{-1}$.  The more luminous sources appear, on average, to be spectrally harder than the fainter sources, an effect which is attributable to increased absorption in their spectra.  Extensive diffuse X-ray emission is detected coincident with the disk of NGC 4490, and in the tidal tail of NGC 4485, which appears to be thermal in nature and hence the signature of a hot ISM in both galaxies. However, the diffuse component accounts for only $\\sim 10\\%$ of the total X-ray luminosity of the system ($2 \\times 10^{40}$ erg s$^{-1}$, 0.5 - 8 keV), which arises predominantly in a handful of the brightest discrete sources.  This diffuse emission fraction is unusually low for a galaxy pair that has many characteristics that would lead it to be classified as a starburst system, possibly as a consequence of the small gravitational potential well of the system.  The discrete source population, on the other hand, is similar to that observed in other starburst systems, possessing a flat luminosity function slope of $\\sim -0.6$ and a total of six ultraluminous X-ray sources (ULX). Five of the ULX are identified as probable black hole X-ray binary systems, and the sixth (which is coincident with a radio continuum source) is identified as an X-ray luminous supernova remnant.  The ULX all lie in star-formation regions, providing further evidence of the link between the ULX phenomenon and active star formation. Importantly, this shows that even in star forming regions, the ULX population is dominated by accreting systems.  We discuss the implications of this work for physical models of the nature of ULX, and in particular how it argues against the intermediate-mass black hole hypothesis. ",
        "introduction": "In the two years since its launch, the high spatial resolution ($\\sim 0.5''$) X-ray optics of the \\chan observatory have revolutionised the study of spatially complex X-ray sources.  One particular area to benefit greatly from this advance is the study of the X-ray properties of nearby galaxies.  This is highlighted by the recent studies of nearby starburst systems such as M82 (Matsumoto et al. 2001; Kaaret et al. 2001), NGC 253 (Strickland et al. 2000) and the Antennae (Fabbiano, Zezas \\& Murray 2001), where populations of luminous point-like sources and widespread, luminous diffuse emission components are clearly resolved.  A further advantage of the high spatial resolution offered by \\chan is the sensitivity it provides to faint point-like sources, reaching limiting fluxes at least an order of magnitude fainter than in comparable exposures with previous missions.  The result has been the detection of large numbers of X-ray sources in nearby galaxies; for example 110 X-ray sources were detected in the central regions of M101 to a limiting luminosity of $\\sim 10^{36} \\ergsec$ (Pence et al. 2001), and 47 detected within the bulge of NGC 1291 down to L$_{\\rm X} \\sim 10^{37} \\ergsec$ (Irwin et al. 2002).  This has, in turn, allowed the first detailed studies of the luminosity functions of discrete X-ray source populations over a wide range of galaxy types and luminosities.  For example, Tennant et al. (2001) derive separate luminosity functions for the bulge and disk regions of M81, showing that the disk luminosity function has a shallower slope, consistent with a younger, more luminous source population.  On a similar theme, Kilgard et al. (2002) compare the luminosity functions of starburst, disk-dominated and bulge-dominated spiral galaxies, showing that galaxies hosting enhanced star formation have flatter luminosity function slopes due, primarily, to the presence of younger, more luminous X-ray sources in starburst systems.  In this paper we focus upon a new \\chan observation of the nearby ($d = 7.8$ Mpc, Tully 1988\\footnote{We note that Tully gives separate distances of 7.8 and 9.3 Mpc for NGC 4485 and NGC 4490 respectively. This is obviously unphysical for a closely interacting system, so we assume a distance of 7.8 Mpc for both galaxies throughout this paper, as taken by Elmegreen et al. (1998).  A consistent distance of 8 Mpc is assumed by Viallefond, Allen \\& de Boer (1980), based on the suspected membership of NGC 4485/90 to the CVn {\\small II} cloud (de Vaucouleurs, de Vaucouleurs \\& Corwin 1976).}) interacting galaxy pair NGC 4485/NGC 4490.  NGC 4490 is a late-type spiral galaxy, classified as type SB(s)d with absolute magnitude $M_B = -19.55$ (de Vaucouleurs et al. 1991).  Its smaller companion NGC 4485 is an irregular galaxy (type IB(s)m) with $M_B = -17.65$.  At the quoted distance, their projected separation is a mere $\\sim 8$ kpc.  The closest encounter between the systems occurred about $4 \\times 10^8$ years ago (based on a comparison with $N$-body simulations, and the age of the stellar populations; Elmegreen et al. 1998), and was a prograde encounter that has produced tidal features in both galaxies, including a ``bridge'' of material linking a tidal tail at the south of NGC 4485 to the western arm of NGC 4490, and a faint tidal tail to the east of NGC 4490.  NGC 4490 shows considerable evidence for enhanced star formation in both radio continuum and far infra-red observations (Viallefond, Allen \\& de Boer 1980; Klein 1983; Thronson et al. 1989).  Clemens, Alexander \\& Green (1999) suggest that it is a relatively young galaxy, formed $2 \\times 10^9$ years ago, with an approximately constant star formation rate of $\\sim 5$ M$_{\\odot}$yr$^{-1}$ throughout its lifetime.  This activity predates the interaction with NGC 4485, which has only had time to significantly affect the western arm of NGC 4490.  The star formation appears to have driven a galactic-scale bi-polar outflow of \\hi gas perpendicular to the plane of NGC 4490, resulting in a \\hi envelope $\\sim 56$ kpc across ($\\equiv 25'$ on the sky) surrounding the galaxies (Clemens, Alexander \\& Green 1998).  The passage of NGC 4485 through this \\hi cloud appears to be ram-pressure stripping it of its atomic, molecular and dusty ISM components (Clemens, Alexander \\& Green 2000).  Detailed \\Ha imaging (Duval 1981; Thronson et al. 1989) shows the ongoing star formation in the system to be predominantly in the tidal arms between NGC 4485 and NGC 4490, with additional activity peaking slightly to the west of the nucleus of NGC 4490, and in a northern spiral arm.  This pair of galaxies is not particularly well studied in X-rays.  The most detailed study to date was undertaken as part of a survey of nearby spiral galaxies with the \\ro PSPC by Read, Ponman \\& Strickland (1997; hereafter RPS97).  They detect four compact X-ray sources embedded in extensive diffuse emission.  Three sources are associated with the disk of NGC 4490, including one coincident with the nucleus, and the fourth is located at the southern tip of NGC 4485.  The spectrum of the diffuse emission appears unusually hard, leading to the speculation that it is largely composed of unresolved X-ray binary systems.  Subsequently, two \\ro HRI images of the galaxies were analysed by Roberts \\& Warwick (2000; hereafter RW2000) as part of a survey of point-like X-ray sources in nearby galaxies.  They detected the same four sources in each observation, plus a fifth apparently transient source on the southern side of NGC 4490, which is only present in the first HRI observation.  Here, we present the results of a recent \\chan ACIS-S observation of the NGC 4485/90 system.  We sub-divide the paper in the following manner.  In Section 2 we give details of the observation and data reduction.  We catalogue the luminous discrete X-ray source population of the galaxies in Section 3, and then investigate the X-ray properties of the population in Section 4, which includes an examination of the properties of six ``ultraluminous X-ray sources'' (ULX) found within the system.  This is followed by the analysis of the remaining diffuse X-ray emission in Section 5.  In Section 6 we discuss the total luminosity of the galaxies and the luminosity function of the constituent sources, and compare our results to other starburst systems observed by {\\it Chandra\\/}.  We go on to discuss the implications of the discovery of so many ULX in this particular small galaxy pair for our understanding of the ULX phenomenon in Section 7 before, finally, presenting our conclusions. ",
        "conclusions": "In this paper we have analysed a 20 ks \\chan observation of the nearby, interacting galaxy pair NGC 4485/90.  Our results may be summarised thus:  \\begin{itemize} \\item We detect a total of 31 discrete X-ray sources coincident with the galaxies (including a faint bridge linking the two), ranging in observed luminosity between $\\sim 2 \\times 10^{37}$ and $4 \\times 10^{39} \\ergsec$ (0.5 - 8 keV).  Only five had been detected in previous X-ray observations of the galaxies (RPS97; RW2000).  No more than four of these detections are expected to be due to background contamination. \\item An analysis of the 0.3 -- 2 keV vs 2 -- 10 keV hardness ratios for the sources reveals a trend of spectral softening as the flux decreases. This is not a result of a new, soft source population appearing at lower fluxes (though we do identifying one candidate super-soft source), but rather the fact that the more luminous sources are more absorbed.  This absorption may be intrinsic to the sources, or may be a result of their being located deeper within, or in denser regions of, the galaxies. \\item We identify six sources with luminosities in excess of $10^{39} \\ergsec$, which are examples of the ultraluminous X-ray source phenomenon.  All of these sources are point-like and none exhibit short-term X-ray variability.  On the basis of their X-ray spectra and long-term lightcurves five of the ULX are likely to be black hole X-ray binary systems and the sixth, which is directly coincident with a radio source, is an X-ray luminous supernova remnant. \\item There is extensive diffuse X-ray emission associated with the disk of NGC 4490.  A further patch is located in the southern regions of NGC 4485, probably as a result of the ram-pressure stripping of its ISM as it passes through the giant \\hi cloud surrounding NGC 4490.  X-ray spectroscopy of the diffuse component shows it to be predominantly thermal in nature, consistent with an origin in the hot interstellar medium of the galaxies, and not in an unresolved source population. \\item The total observed X-ray luminosity of the NGC 4485/90 system is $\\sim 2 \\times 10^{40} \\ergsec$ (0.5 - 8 keV).  Over 90\\% of this is contributed by the luminous X-ray source population, with the three brightest sources contributing more than 50\\% between them.  The luminosity function of the discrete X-ray source population shows a very flat slope ($-0.57 \\pm 0.1$) similar to that seen in classical starburst systems, implying that NGC 4485/90 hosts a young, luminous X-ray source population. \\item A comparison with the classic galaxy merger starbursts in the Antennae and NGC 3256 shows that the starburst in NGC 4485/90 is relatively X-ray weak, particularly in terms of its diffuse emission component. This is especially puzzling when compared to the Antennae, which has a similar activity level.  This may be due to the type of interaction (direct merger versus tidal encounter), or the smaller gravitational potential well of NGC 4490 retaining less hot gas. \\item The high ULX content of NGC 4485/90, and the direct spatial coincidence of its ULX with star formation regions, is further convincing evidence of the relationship of many ULX with active star formation.  We note that only one of the ULX is identifiable as a supernova remnant, implying that even in active star formation regions the ULX population is dominated by accreting sources. \\item It is unlikely that the putative black hole X-ray binary ULX in NGC 4485/90 contain intermediate-mass black holes, due to their extra-nuclear positions, the age of the stellar populations in their host regions, and their apparent relationship with the active star formation regions within the galaxies.  There are a number of very plausible alternative explanations in terms of a simple high-mass X-ray binary nature for the ULX.  It now seems credible that many ULX may be nothing more exotic than ordinary high-mass X-ray binary systems viewed in an epoch of unusually high accretion rate and/or with a preferred orientation. \\end{itemize}  The most remarkable feature of NGC 4485/90 in the X-ray regime is its highly luminous discrete X-ray source population, and in particular the six ULX it hosts.  Indeed, NGC 4485/90 may present one of the best opportunities to study this phenomenon in the local universe, since other systems with similar numbers of ULX (e.g. the Antennae and NGC 3256) are over a factor 2.5 more distant, and dominated by potentially-confusing extensive hot gas components.  A better understanding of the phenomenology of this interesting class of sources may therefore rely upon future studies of the ULX in NGC 4485/90.  \\vspace{0.2cm}  {\\noindent \\bf ACKNOWLEDGMENTS}  We thank the referee, Andy Read, for his many useful comments that have improved this paper.  TPR gratefully acknowledges financial support from PPARC.  This paper uses DSS-2 data extracted from the ESO online digitised sky survey.  The archival {\\it ROSAT\\/} data were obtained from the Leicester database and archive service (LEDAS) at the University of Leicester."
    },
    "0208/astro-ph0208024_arXiv.txt": {
        "abstract": "Recent determinations of the internal composition and structure of two helium-atmosphere variable white dwarf stars, GD~358 and CBS~114, have led to conflicting implied rates for the \\cago reaction. If we assume that both stars were formed through single-star evolution, then the initial analyses of their pulsation frequencies must have differed in some systematic way. I present improved fits to the two sets of pulsation data, helping to resolve the tension between the initial results. ",
        "introduction": "When a white dwarf is being formed in the core of a red giant star during helium burning, there are two nuclear reactions that compete for the available helium nuclei: the 3$\\alpha$ reaction, which combines three helium nuclei to form carbon, and the \\cago reaction, which combines an additional helium nucleus with the carbon to form oxygen. At a given core temperature and density, the relative rates of these two reactions largely determines the C/O ratio in the resulting white dwarf star. The rate of the 3$\\alpha$ reaction is known to about 10\\% precision, but the \\cago reaction is still uncertain by about 40\\%. So, if we can measure the C/O ratio in the core of a pulsating white dwarf, it is effectively a measurement of the \\cago reaction rate. The C/O ratio is interesting by itself, since the core composition in our models affect the derived cooling ages of white dwarfs by up to a few Gyr \\cite{1}. But we can also use it to provide an independent measurement of a nuclear reaction that is important to many areas of astrophysics, from the energetics of type Ia supernovae explosions to galactic chemical evolution.  The model-fitting method that I describe below has only been applied to DB white dwarfs, since they are structurally the simplest. But in principle it can be extended to the pulsating DA stars quite easily, and with a little more work to the DOVs. Presently, the method requires that the spherical degree of the pulsation modes is known, and sufficient data exist for only two stars---though we have just finished a Whole Earth Telescope run on a third object. The first application was to GD~358, which showed 11 consecutive radial overtones with the same spherical degree during a WET run in 1990 \\cite{2,3}. The second application was only recently finished, and came from single-site data on the star CBS~114, which showed 7 independent modes that all appear to be $\\ell$=1 \\cite{4}. What we set out to do was search for a theoretical model that could reproduce, as closely as possible, the pulsation periods that we have observed in these stars. ",
        "conclusions": "By measuring the interior composition of pulsating white dwarfs, we can get precise measurements of the important \\cago reaction rate, and as our models improve we can be more and more sure that they are not only precise, but also accurate. The model-fitting tool that we have used, involving a genetic algorithm, is a very powerful way to explore large ranges of interesting physical parameters and find the globally optimal model to match the observations. And since it evaluates so many models along the way, it produces some good maps of the search space in the regions where it is most interesting. True, this method is still computationally intensive, but there is no getting around that if we want the global solution, and Linux clusters are getting cheaper and cheaper. Finally, in the future we hope to be able to say something more about the detailed shape of the C/O profile all the way from the center to the surface."
    },
    "0208/astro-ph0208268_arXiv.txt": {
        "abstract": "We show that the small scatter around the Fundamental Plane (FP) of massive elliptical galaxies can be used to derive important properties of their dark  and luminous matter. The {\\it central velocity dispersion}  $\\sigma_0$, appearing  in (e.g.) the  Fundamental Plane, is  linked to photometric, dynamical  and geometrical properties of (luminous and dark) matter. We find that,  inside the effective radius $R_e$, the matter traced by the light must largely dominate over the dark matter (DM), in order to keep the ellipticals close enough to the FP. This recalls analogous findings for spiral galaxies.  In particular we also find  that cuspy DM distributions, as predicted by numerical simulations in $\\Lambda$CDM  cosmology,  are unable to explain the very existence of the FP; in fact, according to this theory, the structural properties of dark and luminous matter are so interwoven that a curved surface is predicted in the log--space ($\\sigma_0$, $R_e$, $L$), rather than a plane.  In order to agree with the FP, CDM halos must have concentrations parameters in the range of $5-9$ (i.e.  values  significantly lower than the current predictions).  Assuming a more heuristic approach and allowing for cored DM halos, we find  that the small  intrinsic scatter of the  FP yields to {\\it i)} an average value for the dark--to--light--traced mass ratio inside the  length--scale of light $R_e$ of about $0.3$, {\\it ii)} a mass--to--light ratio of the matter traced by the light  increasing with spheroid luminosity: $M_{sph}/L_r \\propto L_r^{0.2}$ in Gunn--$r$ band, with a value of $ 5.3$ at $L_{\\ast r} \\equiv 2.7 \\times 10^{10} L_{r \\odot}$. ",
        "introduction": "In the hierarchical scenario, dark matter (DM) halos have driven, from a variety of initial conditions,  a dissipative infall of baryons and formed the galactic systems we observe today (White and Rees, 1978). Thus, we  expect that DM halos exist within and surrounding any galaxy, regardless of its luminosity and morphological type. This prediction had overwhelming confirms for disk galaxies, due to the existence of good dynamical tracers and  their intrinsic simple geometry (see Persic and Salucci, 1997). Elliptical galaxies (E's), however, are much  more complicated objects, due to their  3--dimensional shape,  stellar orbital structure and velocity dispersion anisotropy. These factors  have made ambiguous the interpretation of observational data.  A number of different mass tracers have been used to probe the gravitational potential in tenth of E's and  derive their mass distribution: integrated stellar absorption spectra,  X--ray emission from hot gas, rotating gas disks, motions of globular  clusters or satellite galaxies and, in last years, weak gravitational lensing. As result,  the presence of dark matter in E's, especially in the external regions ($\\gsim 10$ kpc), is proven (e.g. Loewenstein and White, 1999). On the other hand,  a kinematical  modeling of the inner regions (i.e. within the half--luminosity ``effective\" radius $R_e$), has been performed for only a  small number of ellipticals (e.g. van der Marel, 1991; Saglia et al. 1992, 1993; Bertin et al., 1994; Kronawitter et al., 2000; Gerhard et al., 2001); the results  point to a  tendency for moderate dark matter amounts inside $R_e$.  Since its discovery, the ``Fundamental Plane\"  (Djorgovski \\& Davis, 1987; Dressler et al., 1987) has been  one of the main tools to investigate E's properties: effective radius $R_e$, central velocity  dispersion  $\\sigma_0$ and mean effective surface brightness $I_e$ of spheroidal galaxies are linearly related in the logarithmic space and  galaxies closely cluster  on a plane, with a surprisingly low orthogonal scatter. To explain these linear relations between photometric and dynamical  quantities in log--space, most  studies on the Fundamental Plane (FP) have considered models in which the mass is distributed parallel to light. However, in presence of non--baryonic dark matter, this hypothesis is an obvious oversimplification and, at least, unjustified. Indeed, this would {\\it a priori} require either:  {\\it i)}  dark and luminous component are distributed according the same profile, thus revealing a similarity of properties and behavior which seems very unlikely  or {\\it ii)} the dark matter component is always negligible with respect to the luminous matter.  Within the above framework, in this paper  we address the following issues: \\begin{itemize} \\item  to derive the relation between the central velocity dispersion $\\sigma_0$ and the mass distribution parameters, including the effect of a dark matter halo.  In particular, we assume a spherical model  with an isotropic luminous component and a dark halo, more diffuse than the spheroid, \\item to reproduce the observed Fundamental Plane and, therefore, to constrain the mass distribution in E's, \\item   to discuss the results in the light of Cold Dark Matter predictions. \\end{itemize}   Considering elliptical galaxies as two--components systems, complementary strategies are possible. One chooses a distribution function for both components and then imposes  specific constraints from the observations. The other includes  the ordinary stellar component  (or, better, any traced by light (TBL) mass component) in a frozen spherical halo. The former approach is helpful in exploring  the self--consistency of the dynamical configuration (e.g. Ciotti, 1999). The latter, we will adopt   in this paper, has the advantage of providing a simpler connection  between observational quantities and the  parameters of the mass model.   The outline of this paper is the following:  in \\S2 we describe two--components models, whose mass distributions are shown in  \\S3. In \\S4  we derive and  discuss the velocity dispersion (line--of--sight profile and central value) predicted by the mass models we consider. In \\S5, we introduce  the data,  fit the models to the Fundamental Plane and discuss the results. Finally, conclusions are presented in \\S6. Throughout the following work, we assume, where needed, a flat $\\Lambda$CDM Universe, with $\\Omega_m=0.3$, $\\Omega_{\\Lambda}=0.7$, $h=0.7$ and $\\sigma_8 =1.0$. ",
        "conclusions": "We have shown that the very low scatter of elliptical galaxies around the Fundamental Plane can be statistically  used to put very interesting constraints on DM distribution within them. The central velocity dispersion is the key quantity we have dealt with. We have briefly reviewed its relationship with the mass distribution of both traced--by--light and dark matter. Then, we selected a sample of 221 E/S0 galaxies with $L_r\\geq 2 \\times 10^9$ $L_{\\odot}$ in 9 clusters, endowed with very good photometric and spectroscopic data. The sample defines the classical FP in the log--space ($\\sigma_0$, $R_e$, $L_r$), with the expected small scatter (0.084 in $\\log\\ R_e$, to be compared to a measurement uncertainty $\\Delta\\log R_e = \\pm 0.045$).  We tested the reference model of cuspy DM distribution, namely the NFW model, and the cored model proposed by Burkert (1995). Our analysis shows that these luminous galaxies are largely dominated within the effective radius by matter traced by light, independently of the DM distribution model, cuspy or cored. In particular, for the cuspy NFW model, we have shown that the small scatter of our sample galaxies around the Fundamental Plane severely challenges the $\\Lambda$CDM predictions. In such a theory, the structural properties of dark and luminous matter are so interwoven that in the log--space ($\\sigma_0$, $R_e$, $L$) they produce a curved surface, rather than a plane, for plausible values of the total dark--to--TBL mass ratio. We conclude that in order to keep the small scatter around the FP we have either to keep $\\Gamma_{vir}$ unacceptably low or to decrease the halo concentration well below the value currently predicted by simulations in $\\Lambda$CDM cosmology.   Considering a cored DM density distribution,  the agreement with the observed FP implies a dark--to--luminous mass fraction within the effective radius of  $\\sim 30\\%$ and a luminosity dependence  of the spheroid mass--to--light ratio in Gunn--$r$ band: $M_{sph}/L_r=(5.3\\pm 0.1) (L_r/L_{\\ast r}) ^{0.21\\pm 0.03}$. An important result is the robustness of the mass--to--light ratio of the spheroidal component we obtained, which is in good agreement with predictions by stellar evolution models.   It is also worth noticing that, besides for spiral and dwarf galaxies, a cored DM halo, with low internal (within  $2-3 R_e$) density which increases as $r^{-3}$ at larger radii, is successful to explain also the structure of {\\em elliptical} galaxies, pointing to an intriguing homogeneous scenario. Within this framework, we argue  that dark matter in E's can be investigated by  a reasonably large number of galaxies with measures  of  l.o.s. velocity dispersion at  $\\sim  R_e$. Although, so far, such observations have been severely hampered by the steep  decreasing of  the surface brigthness with radius, higher and higher sensitivity  reached by recent surveys offers a good view  to obtain a better resolution  of the two mass components, in the whole region  where baryons reside.  \\ \\   {\\bf Acknowledgments}  \\noindent The authors would like to aknowledge financial support from  ASI and from MIUR  trough COFIN. We also  thank the anonymous referee for very helpful comments."
    },
    "0208/astro-ph0208097_arXiv.txt": {
        "abstract": "{ We present mass reconstructions from weak lensing for the galaxy clusters A1835 and A2204 over $34\\arcmin \\times 34\\arcmin$ fields using data from the ESO/MPG Wide Field Imager.  Using a background galaxy population of $22<R<25.5$ we detect the gravitational shear of A1835 at $8.8\\sigma$ significance, and obtain best-fit mass profiles of $\\sigma_v=1233^{+66}_{-70}$ km/s for a singular isothermal sphere model and $r_{200}=1550 h^{-1}$ kpc, $c=2.96$ for a `universal' CDM profile. Using a color-selected background galaxy population of $22<R<25.8$ we detect the gravitational shear of A2204 at $7.2\\sigma$ significance, and obtain best-fit mass profiles of $\\sigma_v=1035^{+65}_{-71}$ km/s for a SIS model and $r_{200}=1310 h^{-1}$ km/s, $c=6.3$ for a `universal' CDM profile. The gravitational shear at distances greater than $10\\arcmin $ is significantly detected for both clusters.  The best fit weak lensing cluster masses agree well with both X-ray and dynamical mass measurements, although the central concentration of A1835 is much lower in the weak lensing mass profile than that measured by recent Chandra results.  We suggest that this lower concentration is most likely a combination of contamination of the 'background' galaxy population with cluster dwarf galaxies and the effect of a prolate or tri-axial cluster core with the major axis lying near the plane of the sky. We also detect a number of additional structures at moderate significance, some of which appear to be sub-haloes associated with the clusters. ",
        "introduction": "In recent years, cosmological N-body simulations have been used to obtain theoretical mass profiles for massive clusters of galaxies (\\citealt{NA97.6}, hereafter NFW; \\citealt{MO99.5}; \\citealt{JI00.1}).  While there is some discrepancy between the models for the mass profile at small radii ($\\la 100$ kpc), at large radii ($\\ga 1$ Mpc) they agree that the mass density should decrease as $r^{-3}$.  This is a much steeper slope at large radii than the singular isothermal sphere (hereafter SIS) model, which falls as $r^{-2}$.  Measuring the mass profiles of clusters at large radii, therefore, should provide an excellent test of the predictions from N-body simulations.  Weak gravitational lensing, in which one determines the mass of an object by measuring the shear induced in background galaxies by the gravitational potential of the object, is a powerful tool for determining the mass and mass profile of clusters of galaxies.  Because the gravitational shear depends linearly on the two-dimensional mass surface density, it can be used to measure mass profiles to much larger radii than X-ray observations, for which the emissivity scales as the square of the baryonic mass density.  Further, the resulting mass measurements have no dependence on the dynamical state of the cluster and are not influenced by shocks caused by infalling material. To date, the primary limitation on the maximum radius to which the cluster mass profiles has been measured with weak lensing ($\\sim 400-600 h^{-1}$ kpc) has been the size of the detector and the fact that the depth of exposure needed to obtain a sufficient number density of background galaxies for shear analysis made large-area mosaics impractical in telescope time.  With the advent of wide-field mosaic CCD detectors \\citep{LU98.1}, it is now possible to obtain a weak lensing signal to large radii ($\\ga 2 h^{-1}$ Mpc) on moderate redshift clusters in a few hours of telescope time. The mass profile for A1689 out to $2 h^{-1}$ Mpc was measured by \\citet[hereafter CS01]{CL01.1}. Here we present the mass profiles to similar radii for two additional massive clusters, A1835 and A2204.  In Sect.~2 we discuss the image reduction and object catalog generation.  We analyze the weak lensing signal for A1835 in Sect.~3 and for A2204 in Sect.~4.  In Sect.~5 we present out conclusions.  Unless otherwise stated we assume a $\\Omega _\\mathrm{m} = 0.3, \\Lambda = 0.7$ cosmology, $H_0 = 100 h$ km/s/Mpc, and that the mean redshift of faint galaxies is at $z_\\mathrm{bg}=1$. ",
        "conclusions": "We have detected a shear profile for the clusters A1835 and A2204 over a $1\\arcmin < r < 16\\arcmin$ range of radii.  Fitting a SIS model to these shear profiles gives best fit velocity dispersions of 1180 km/s for A1835 and 1040 km/s for A2204, in agreement with the mass measurements from X-ray observations, and significances of 8.8$\\sigma$ and 7.2$\\sigma$ from zero-mass models, respectively.  A2204 has a best fit NFW profile of $r_{200} = 1310 h^{-1} \\mathrm{kpc}, c = 6.3$ which is a better fit to the data at $87\\%$ confidence.  A1835 has a best fit NFW profile of $r_{200} = 1480 h^{-1} \\mathrm{kpc}, c = 2.86$ which is statistically indistinguishable in quality of fit from the SIS model.  For both clusters we detect the shear between $10\\arcmin$ and $16\\arcmin$ from the BCG at high significance, which suggests that one will still measure a shear signal from such clusters with even larger format cameras.  \\begin{figure} \\resizebox{\\hsize}{!}{\\includegraphics{H3633F13.ps}} \\caption{Shown above is the galaxy number density in radial bins around the BCG of each cluster, after correction for loss of observable area due to bright objects, stellar reflection rings, etc. The data for A2204 is shown for all the faint galaxies (ncs) and only the galaxies surviving the rejection of probable low-z galaxies by color (cs).  All three sets of radial bins have the same radial range per bin but are offset from each other in the figure for greater clarity.  The A1835 data points are plotted at the true geometrical mean of each bin.  The error bars give the $1\\sigma$ Poissonian noise for the bins, but are underestimates of the true error as they do not allow the galaxies to be correlated.  The depletion in galaxy number counts at small radius is caused primarily by the rejection of galaxies with near neighbors from the 'background' galaxy catalog and not by gravitational lensing. As the cluster core has the highest number density of objects in the field, the fraction of galaxies rejected by this criteria increases in this region.} \\label{bgfig} \\end{figure}  While the total mass measured for A1835 is in agreement with the Chandra X-ray observations of SA01, the concentration of mass towards the center of the cluster is much smaller.  We believe this is due primarily to contamination of the `background' galaxy catalog used to measure the shear with cluster dwarf galaxies.  This would result in the shear in the inner regions of the cluster being reduced due to the presence of a population of zero shear objects, but the shear at large radii being virtually unchanged due to the decrease in number density of the cluster dwarfs with increasing distance from the cluster.  We measured this effect in the A2204 data by comparing a catalog of galaxies which had probable cluster members rejected by color with a catalog without any color rejection.  The color-selected catalog did have a higher shear in the inner few arcminutes of the cluster radial shear profile while having essentially the same shear at larger radius.  The amount of the increase in shear by rejecting cluster dwarf galaxies in A2204 would not be sufficient to bring the X-ray and weak lensing mass concentrations of A1835 into agreement.  Using the cluster dwarf number counts of Coma \\citep{TR98.2} we find from the observed magnitude limits that, given the same cluster richness, A2204, at $z=0.152$, would have 2.5 times more cluster dwarfs in the `background' galaxy catalog than A1835, at $z=0.252$, but spread over a 2.2 times greater area.  Thus, the actual contamination of the A1835 signal should be less than that observed in A2204, but in order to reconcile the shear profile with the X-ray observations, we need the contamination to be three times larger than that in A2204.  { The galaxy density profiles in the background galaxy catalogs for the two clusters can be seen in Fig.~\\ref{bgfig}.  The rejection of low-z galaxies by color in the A2204 data results in a decrease in the number density of galaxies used for the weak lensing analysis of $\\sim 17\\%$ at large radius and $\\sim 25\\%$ at small radius.  As can be seen, while A1835 has a similar number density of galaxies to the catalog without color selection for A2204 at large radius, it has a much larger galaxy density at small radius. There are three possible explanations for this increase in galaxy density in A1835 at small radius.  The first is an increased number of cluster dwarf galaxies in the core of A1835 compared to A2204. The second is that there is either an overdensity of background galaxies near A1835 or an underdensity near A2204.  The first explanation would result in an increase in the cluster dwarf fraction at small radius for A1835, and thus a greater decrease in the measured mass concentration than is seen in the non-color selected catalog for A2204, while the opposite would be true for the second explanation.  The third possible explanation of the increased galaxy density is that the decrease in the number density of galaxies in A2204 with decreasing radius is mostly due to the rejection of galaxies from the background galaxy catalog which have near neighbors and the increased bright galaxy density at the cluster core.  A catalog of faint galaxies without this near neighbor rejection actually shows a slight rise in galaxy density at small radius in A2204, consistent with the small increase in color selected galaxies with decreasing radius discussed above.  Because A1835 is at higher redshift than A2204, however, the cluster core has a much smaller angular size.  As a result, the increase in rejection of galaxies from the background galaxy catalog should occur at a smaller angular radius in A1835 than in A2204, which is what appears to be occurring in Fig.~\\ref{bgfig}. The fact that the number density profile of galaxies without the near neighbor rejection is similar for the two clusters suggests that this third explanation is the most likely, and thus the lensing mass profile does, in fact, have a smaller concentration than that measured from the X-ray data.}  One possibility for the lower concentration in the weak lensing mass model than in the X-ray mass model would be an elliptical cluster core.  The X-ray NFW profile was measured by using a spherical 3-dimensional model which was compared with a deprojected spectrum of the Chandra data. The weak lensing profile, however, is created by first integrating the mass along the line of sight, then fitting with a circularly symmetric model. In the case of a prolate cluster core, if the major axis of the core is lying in the plane of the sky then the best-fit weak lensing profile will have a smaller concentration than that of the X-ray model due to the fact that in the weak lensing analysis the mass is projected into a plane before being circularly averaged, while in the X-ray analysis the mass is spherically averaged.  Conversely, if the major axis is lying along the line of sight, the weak lensing profile will have a higher concentration than the X-ray model.  In studies of a CDM N-body cluster, \\citet{KI01.1} found that the best fit 2-dimensional NFW concentration parameter among different projection axes could vary by $\\sim 20\\%$.  In addition to the primary cluster mass peak, several additional peaks were found in both clusters at $\\sim 3-4\\sigma$ significance using $4\\arcmin $-$7\\arcmin $ radius mass aperture statistics.  Only two of the six peaks are not spatially coincident with galaxy overdensities, but both are located at the edge of the image, and thus could be associated with structures immediately outside the field of view.  In addition, the mass reconstructions of both clusters have filamentary-like structures extending from the cluster cores.  These structures are also seen in faint galaxy distributions, and part of the structure in A1835 is spatially coincident with an X-ray source.  { Finally, we note that all of the masses quoted herein have assumed that the background galaxies have the redshift distribution given by the \\citet{FO99.1} HDF-S photometric redshift catalog.  The appropriate magnitude and color cuts were made to this catalog, and the mean lensing redshift for the sample was calculated for each cluster.}  While the mean lensing redshift of the background galaxies does not affect the significance of the shear signals, it does affect both the total mass of the clusters and the concentration parameter of the best fit NFW profile.  An overestimate of the mean background galaxy redshift would result in both the best-fit mass and concentration value for a cluster being lower than the true value. Because both A1835 and A2204 are at relatively low redshifts, the measured mass changes slowly with the mean lensing reshift.  {For the A2204 color selected catalog, the background galaxy mean lensing redshift would need to shift from the measured $\\bar{z}_\\mathrm{bg} = 1.06$ to $\\bar{z}_\\mathrm{bg}\\sim 0.6$ or $\\bar{z}_\\mathrm{bg}\\sim 4$ in order for the systematic error from the redshift estimate to equal the level of the random error from the intrinsic galaxy ellipticities and mass sheet degeneracy.  For the A1835 catalog, the background galaxy mean lensing redshift would need to shift from the measured $\\bar{z}_\\mathrm{bg} = 0.79$ to $\\bar{z}_\\mathrm{bg}\\sim 0.62$ or $\\bar{z}_\\mathrm{bg}\\sim 1.1$ for the systematic error in the assumed mean lensing redshift to equate to the random error from the background galaxy distribution.}"
    },
    "0208/astro-ph0208118_arXiv.txt": {
        "abstract": "We have observed planetary nebulae (PNe) in several early-type galaxies using new techniques on 4- to 8-meter-class telescopes. We obtain the first large data sets ($\\ga 100$ velocities each) of PN kinematics in galaxies at $\\ga 15$ Mpc, and present some preliminary dynamical results. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0208/astro-ph0208432_arXiv.txt": {
        "abstract": "{\\small We present mid-infrared spectrophotometric observations of SS\\,433 with ISOPHOT. The He{\\footnotesize I}+He{\\footnotesize II} lines in both spectra of SS\\,433 and of the Wolf-Rayet star WR\\,147, a wind-colliding WN8+BO5 binary system, closely match. The 2.5-12 $\\mu$m continuum radiation is due to an expanding wind free-free emission in an intermediate case between optically thick and optically thin regimes.  The inferred mass loss rate evaluation gives \\mbox{$\\sim 10^{-4}$ M$_\\odot$.yr$^{-1}$.} Our results are consistent with a Wolf-Rayet-like companion to the compact object in SS\\,433.  A similar study for Cygnus\\,X-3 confirms the Wolf-Rayet-like nature of its companion, although with a later WN type than previously suggested.} ",
        "introduction": "\\hspace*{0.5cm} SS\\,433 has been discovered as a source of strong H$\\alpha$ lines, and as a variable radio and X-ray source \\cite{margon84}. It is an X-ray binary with a $\\sim 13$ days period, producing bipolar relativistic (0.26\\,c) jets undergoing a precession movement in \\mbox{$\\sim$ 162.5 days,} so covering a cone with an opening angle of 19.8$^\\circ$ and an axis of $\\sim 78.8^\\circ$ with the line of sight. This precession has indeed been observed in radio, at arcsec \\cite{hjell} and milliarcsec scales \\cite{verm}. So SS\\,433 was actually the first microquasar discovered. In spite of intensive studies, no line has been associated to the companion star yet. It is generally accepted but has never been proved that SS\\,433 is a neutron star + massive star binary system. ",
        "conclusions": "\\vspace*{-0.2cm} \\hspace*{0.5cm} SS\\,433 shows a WN8-like spectrum, which continuum is due to free-free emission corresponding to a standard O or WR wind or a geometrically complex thick wind. The inferred mass loss of $\\sim 10^{-4}$ M$_\\odot$.yr$^{-1}$ is compatible with a WNL wind. So either SS\\,433 is the second Wolf-Rayet X-ray binary known in our Galaxy after Cygnus X-3, or it shows WR-like conditions, and at that point we cannot distinguish between these two cases.  \\vspace*{-0.2cm}"
    },
    "0208/astro-ph0208426_arXiv.txt": {
        "abstract": "The velocity dispersion of nearby stars in the Galactic disc are well known to increase substantially with age; this is the so-called Age-Velocity relation, and is interpreted as a ``heating'' of the disc as a function of time.  We have studied the heating of the Galactic stellar disc due to giant molecular clouds and halo black holes, via simulations of the orbits of tracer stars embedded in a patch of the local Galactic disc. We examine a range of masses and number densities of the giant molecular cloud and halo black hole perturbers. The heating of the stellar disc in the simulations is fit with a simple power law of the $\\sigma \\propto t^\\alpha$ where $\\sigma$ is the velocity dispersion of the tracer stars as a function of time, $t$. We also fit this form to the best determinations of the increase in the velocity dispersion as a function of time as derived from stars in the solar neighbourhood for which ages can be reliably assigned. Observationally, $\\alpha$ is found to lie in the range 0.3 to 0.6, i.e. it remains poorly constrained and its determination is probably still dominated by systematic errors. Better constrained observationally is the ratio of the velocity dispersions of the stars in the vertical $z$ and horizontal $x$ (i.e. toward the Galactic center) directions, being $\\sigma_z/\\sigma_x = 0.5 \\pm 0.1$.  For the heating of the stellar disc due to giant molecular clouds (GMCs) we derive a heating $\\sigma \\propto t^{0.21}$, which differs somewhat from early (analytic) studies in which $\\sigma \\propto t^{0.25}$. This confirms the well known results that there are insufficient GMCs to heat the Galactic disc appropriately. A range of dark halo black hole scenarios are verified to heat the stellar disc as $\\sigma \\propto t^{0.5}$ (as expected from analytical studies), and give $\\sigma_z/\\sigma_x$ in the range 0.5 to 0.6, which is consistent with observations. Black holes with a mass of $10^7$ \\Msun\\ are our favoured disc heaters, although they are only marginally consistent with observations. Simulations featuring a combination of giant molecular clouds and halo black holes can explain the observed heating of the stellar disc, but since other perturbing mechanisms, such as spiral arms, are yet to be included, we regard this solution as ad hoc. ",
        "introduction": "The random motions of stars near the Sun are well known to increase with stellar age, and this effect is known as the disc Age Velocity Relation (hereafter AVR). Broadly speaking, the total velocity dispersion of stars rises from $\\approx $20 \\kms\\ at the lowest measurable ages to $\\approx$ 60-80 \\kms\\ at an age of $\\approx$ 10-12 Gyr. The ``heating'' of the orbits is initially rapid but levels off after some Gyr. The heating is not uniform in each of the three velocity components; the velocity dispersion rises to a higher value in the Galactic plane than in the vertical direction.  A number of mechanisms have been proposed to explain the AVR, in particular the gravitational perturbative effect on stellar orbits by giant molecular clouds (GMCs) in the Galaxy's gas layer \\cite{spit51,spit53}, by spiral arms in the disc \\cite{bawo}, or by massive black holes in the Galactic dark halo (BHs) \\cite{lacey85}. Dramatic disc heating also takes place when satellite galaxies fall onto galactic discs \\cite{quinn93}, and such events may well leave an abrupt feature in the AVR, but if the rain of such satellites is relatively gentle they might also heat discs smoothly \\cite{vel99}.  Spitzer and Schwarzschild \\shortcite{spit51,spit53} first argued that the observed growth of the stellar velocity dispersion could be explained by encounters of disc stars with GMCs with masses of $M_{\\rm GMC} \\sim 10^{6} \\Msun$. The main difficulty with this scenario is that the observed number of GMCs modify stellar orbits in a manner inconsistent with observations. Firstly, they are too few to reproduce the observed heating, and secondly, too much heating is produced in the vertical direction relative to the disc plane \\cite{lacey84}. However, GMCs do exist so they certainly provide a mechanism for redirecting orbits out of the Galactic plane, whatever is responsible for the heating. Heating due to transient spiral arms suffers the opposite problem that the amount of vertical heating is too low compared to observations \\cite{carl87}.  This naturally leads to the idea that the observed heating could be explained by the combined effect of the transient spiral arms and giant molecular clouds. In this model spiral density waves would act as main source of heating and GMCs would delect the stellar orbits out of the Galactic plane, thus creating the required velocity dispersion in the vertical direction \\cite{carl87}.  However, these calculations are rather complicated: e.g. the relative effectiveness of spiral and GMC heating has to modelled with an empirical parameter. The rate at which spiral features heat a stellar disc depends on the potential's spatial pattern and on the time variability of the pattern. In principle, spatial stuctures of galaxies can be determined from photometry of real galaxies and temporal information about spiral stucture can be obtained from numerical simulation and dynamical theory. In practice, however, we do not have adequate information about strengths, growth rates or duty cycles of spiral features. So empirical parametrization has so far been the only possible approach \\cite{jebi,je}.  Lacey and Ostriker \\shortcite{lacey85} proposed that the Galactic dark halo might consist of massive black holes (BHs) which would heat the disc as they pass through it. The total heating produced by their proposed $2 \\times 10^6$ \\Msun\\ black holes is consistent with observations, but the velocity ellipsoid was found to be quite round, which is inconsistent with observations. Furthermore, if the dark halos of dwarf spiral galaxies were dominated by similar BHs their discs would be easily destroyed \\cite{friese95}. Finally, BHs passing through the disc might be expected to accrete and reveal themselves as high proper motion X-ray sources, and no such objects have been seen. It seems worth noting though that the central black hole in the Milky Way demonstrates that the accretion rate and emergent flux from a black hole of about the desired mass ($2.6 \\times 10^6$ \\Msun) is far from well understood \\cite{genz98}; the central black hole in the Milky Way is remarkably quiet in X-rays. Another way of detecting these putative black holes is by statistically studying large numbers of orbits of nearby stars; this may be possible with very accurate distances and kinematics which will become available for stars within some kiloparsecs from ESA's GAIA satellite.  In order to avoid problems related to the black hole scenario Carr and Lacey \\shortcite{carr87} proposed dark clusters of less massive objects. The latest generation of cosmological simulations of galaxy formation do indeed show quite clumpy dark halos around Milky Way type spirals \\cite{moore99}. The amount, size and distribution of these clumps is still very uncertain, mainly due to numerical resolution issues, but if they can survive in the inner regions of dark halos they may well be able to heat discs in a manner consistent with observations.  With this in mind we have embarked on a new study of disc heating simulations, starting with examining heating due to GMCs and BHs. The European Space Agency's Hipparcos satellite has provided a wealth of new data on the distances and kinematics of nearby stars, so that one can now construct very much improved measurements of ages and kinematics. The velocities of the stars are much improved through the excellent Hipparcos parallaxes, while the ages of stars can be much better estimated, particularly for stars near the disc main sequence turn-off, because of the greatly improved absolute magnitudes.  Ideally, one would like to carry out a self-consistent N-body simulation of an entire disc galaxy, with and without the perturbing influences due to GMCs/BHs etc, but this is not yet possible even with the fastest computers. The computation would involve both the stars (at approx 1 \\Msun) and the perturbers (for a range of masses, $10^5$-$10^7$ \\Msun), implying some $10^{11}$ stars and thousands of GMCs to millions of BHs in the simulations. This is not yet feasible in an N-body simulation. In this paper we render the problem tractable by simulating the heating in a local patch of the Galactic disc consisting typically of few tens of thousands of (massless) tracer stars, embedded within a fixed Galactic background potential, through which massive perturbers move. By using a local simulation method we can resolve very well the effects of perturbers on nearby stellar orbits, for comparison with the locally obtained observations. A very similar method has been used by Fuchs et al. \\shortcite{fuchs94} who studied disc heating due to GMCs and/or $5 \\times 10^6 \\Msun$ black holes. We are able to broadly reproduce the results of their code in this area, but we also explore a larger range of scenarios.  From our simulations, we find that GMC heating is less effective than earlier thought and further that heating due to GMCs creates a flatter velocity ellipsoid than found in earlier studies. We find that BHs are not as effective in heating the disc as earlier thought. However, very massive halo black holes ($M_{BH} = 1 \\times 10^7 \\Msun$) can heat the disc up to the observed amount, but the results of the heating rate and the ratio of the velocity dispersions are only marginally consistent with observations.  A well selected combination of BHs and GMCs can give the observed heating in every component of the velocity dispersion within the observational error limits, but while interesting this solution is probably ad hoc, as the simulations do not yet model the effects of other perturbation sources, such as spiral waves. This will be examined in future work.  This paper is organised as follows. In section 2 we review pre- and post-Hipparcos work on the observational determination of the age-velocity relation (AVR) for the local disc. We find the AVR is much less well measured than has previously been assumed. Observational issues thus still fundamentally limit our ability to discriminate between disc heating mechanisms.  In section 3 we describe briefly our simulation method and numerical accuracy. In sections 4, 5, and 6 we demonstrate how massive perturbes (GMCs and halo BHs) heat up the stellar disc. We draw our conclusions in section 7. ",
        "conclusions": "There is firm observational evidence for the heating of the stellar disc with time.  We have assembled the available post-Hipparcos samples of ages and kinematics for nearby stars and fit a heating law of the form $\\sigma(t) = \\sigma_\\circ (1 + {t}/{\\tau})^\\alpha$ to the data, where $\\sigma$ is the stellar velocity dispersion. The observational data limit $\\alpha$ to values between about $0.3$ and $0.6$, i.e. despite the recent advances in measuring the kinematics and ages of nearby stars due to the availablity of Hipparcos data, the age velocity relation in the nearby disc remains poorly determined. Fortunately, the ratio of the vertical component of the velocity dispersion to the radial component $\\sigma_z / \\sigma_x \\simeq 0.5 \\pm 0.1$ is much better constrained by available data and can be used to test our simulations.  We simulate disc heating by modelling a patch of the Galactic disc populated with tracer stars. We examine the effect of Giant Molecular Clouds (GMCs) travelling on disc orbits through the patch, and of massive black holes travelling through on plunging orbits characteristic of the dark halo.  In our simulations of GMC heating we consistently find slightly lower values for $\\alpha$ than earlier studies have found, which are based on either theoretical estimates of the diffusion rate of stellar orbits, or scaled simulations of the whole Galactic disc (with a small number of GMCs).  We find $\\alpha = 0.21 \\pm 0.02$ for the evolution of the total velocity dispersion and $\\alpha = 0.26 \\pm 0.04$ for the evolution of the vertical component of the velocity dispersion. The ratio $\\sigma_z / \\sigma_x$ of heating in the vertical and horizontal directions in the disc is found to depend on the adopted GMC density.  The observed heating can not be explained by GMCs, because even a four-fold GMC number density compared to the present day value in the Solar neighbourhood, $\\Sigma \\sim 5 \\Msun / \\parc^2$ \\cite{scosa86}, does not heat up the disc enough.  However, coincidentally we find that for the present day GMC number density the ratio $\\sigma_z / \\sigma_x \\simeq 0.5$ which is consistent within the observational limits for stars in the local disc.  We have also run simulations of the local disc heating being caused by massive ($10^6$-$10^7$ \\Msun) black holes of the dark halo penetrating the disc and perturbing the stellar orbits.  Our simulations verify analytically obtained results in the literature, that the heating index is of order $\\alpha = 0.5$. Even if the entire halo consisted of $M_{BH} = 1 \\times 10^6 \\Msun$ black holes, the resulting heating of the stellar disc would be much less than what is observed. However, if the whole halo were comprised of black holes of order $M_{BH} = 1 \\times 10^7 \\Msun$ the disc would be heated sufficiently. They would heat the disc in such a way that the ratio $\\sigma_z / \\sigma_x$ would increase from $0.5$ to $0.6$ which is still within the error limits of the observations.  We have examined how the stellar disc would be heated by the combination of GMCs and halo black holes. Adopting a GMC density consistent with the present observed value and a dark halo comprised of $M_{BH} = 1 \\times 10^6 \\Msun$ black holes would only heat the disc up to $\\sigma \\simeq 50 \\kms$ in 12 Gyr, which is inconsistent with most of the observations. On the other hand, if half of the halo were made of $M_{BH} = 1 \\times 10^7 \\Msun$ black holes, they could with the GMCs heat the disc up to $\\sigma \\simeq 75 \\kms$ in 12 Gyr, which is consistent with observations, but the heating would also push the ratio of the vertical velocity dispersion to the horizontal dispersion higher: $\\sigma_z / \\sigma_x \\simeq 0.63$, which is still barely within the observational limits. In this case the heating index was observed to be $\\alpha = 0.42$.  Taking into account the observational uncertainties in the stellar disc heating we can produce the right amount of the heating by a proper combination of massive halo black holes, $M_{BH} = 1 \\times 10^7 \\Msun$, and the right number density of GMCs in the Solar neighbourhood. Also the ratio $\\sigma_z / \\sigma_x$ and the heating index $\\alpha$ fall within observational error limits. We do not regard this as a particularly satisfactory solution, because there are other potential causes of heating of stellar orbits, such as spiral arms, which we plan to investigate further in the simulation method presented here."
    },
    "0208/astro-ph0208560_arXiv.txt": {
        "abstract": "This paper reviews the theoretical approaches and existing models of the solar wind interaction with the Local Interstellar Cloud (LIC). Models discussed take into account the multi-component nature of the solar wind and local interstellar medium. Basic results of the modeling and their possible applications to interpretation of space experiments are summarized. Open questions of global modeling of the solar wind/LIC interaction  and future perspectives are discussed. ",
        "introduction": "The Local Interstellar Cloud (LIC) is a cloud of partly ionized plasma surrounding the Solar System. The plasma component of LIC interacts with the solar wind plasma and forms the heliospheric interface (Figure 1). The heliospheric interface is a complex structure, where the solar wind and interstellar plasma, interplanetary and interstellar magnetic fields, interstellar atoms of hydrogen, galactic and anomalous cosmic rays (GCRs and ACRs) and pickup ions play prominent roles.  Although a space mission into the Local Interstellar Cloud is becoming now more realisable, there are no yet direct observations inside the heliospheric interface. Therefore, at the present time the heliospheric interface structure and local interstellar parameters can be derived only from remote experiments and measurements. Currently, backscattered solar Ly-$\\alpha$ radiation, pickup ions, anomalous cosmic rays, and Voyager measurements of distant solar wind are the major sources of information on the heliospheric interface structure and position of the termination shock \\cite{issibook_96}. kHz emission detected by Voyager can put some constraints. Recently, it was shown that study of Ly-$\\alpha$ absorptions toward nearby stars can serve as remote diagnostics of the heliospheric interface and, in particular, the hydrogen wall around the heliopause (e.g., \\cite{Linsky_Wood96} - \\cite{iwl2002}). In the foreseeable future, remote diagnostics will be also possible with images of heliospheric energetic neutrals (ENAs) \\cite{Gruntman_2001}. To reconstruct the structure of the interface and physical processes inside the interface on the basis of remote observations, a theoretical model should be employed.  \\begin{figure}[htb] \\label{fig1}  \\psfig{figure=izmodenov_fig_01.eps,height=10cm,width=8cm,angle=0} \\caption{ The heliospheric interface is the region of the solar wind interaction with LIC. The heliopause is a contact discontinuity, which separates the plasma wind from interstellar plasmas. The termination shock decelerates the supersonic solar wind. The bow shock may also exist in the interstellar medium. The heliospheric interface can be divided into four regions with significantly different plasma properties: 1) supersonic solar wind; 2) subsonic solar wind in the region between the heliopause and termination shock; 3) disturbed interstellar plasma region (or \"pile-up\" region) around the heliopause; 4) undisturbed interstellar medium. }  \\end{figure}  Theoretical studies of the heliospheric interface were performed over more than four decades after pioneering papers by Parker \\cite{parker61} and Baranov et al. \\cite{baranov71}. However, a complete theoretical model of the heliospheric interface has not been constructed yet. The difficulty in doing this is connected with the multi-component nature of both the LIC and the solar wind. The LIC consists of at least five components: plasma (electrons and protons), hydrogen atoms, interstellar magnetic field, galactic cosmic rays, and interstellar dust. The heliospheric plasma consists of original solar particles (protons, electrons, alpha particles, etc.), pickup ions and the anomalous cosmic ray component. The pickup ion component is a result of ionization of those interstellar H atoms that penetrate into the heliosphere through the heliospheric interface. A part of the pickup ions is accelerated to high energies of ACRs. ACRs may also modify the plasma flow upstream of the termination shock and in the heliosheath. Spectra of ACRs can serve as remote diagnostics of the termination shock. For a recent review on ACRs see \\cite{Fichtner_2001}.  To construct a theoretical model of the heliospheric interface, one needs to choose a specific approach for each interstellar and solar wind component. Interstellar and solar wind protons and electrons can probably be described as fluids. At the same time interstellar H atom flow requires kinetic description. For pickup ion and cosmic ray components, the kinetic approach is also required. However, for interpretations that are not directly connected to pickup ions and ACRs, a cruder model can be used.  \\begin{table}[hbt] \\centering \\caption{Number Densities and Pressures of Solar Wind Components}\\label{tab:sw_pressures} \\begin{tabular*}{\\textwidth}{@{}l@{\\extracolsep{\\fill}}ccccc} \\hline Component     &  4-5 AU & & 80 AU & \\\\ & Number Density & Pressure & Number Density & Pressure \\\\ & cm$^{-3}$ & eV/cm$^{-3}$    & cm$^{-3}$ & eV/cm$^{-3}$ \\\\ \\hline Original solar & 0.2-0.4   & 2.-4. (thermal) & ($7-14$) $\\cdot$ 10$^{-4}$ & 10$^{-3}$ - 10$^{-4}$ \\\\ wind protons &    & $\\sim 200$ (dynamic) & & $\\sim 0.5 -1. $ (dynamic) \\\\ Pickup ions & $5.1 \\cdot 10^{-4}$ & 0.5 & $\\sim 2 \\cdot 10^{-4}$ & $\\sim 0.15$ \\\\ Anomalous  &  & & &   \\\\ cosmic rays &  & & & 0.01 - 0.1 \\\\ \\hline \\end{tabular*} \\end{table}  This paper focuses on the models of the global heliospheric interface structure. Under global models I understand those models that study the whole interaction region, including the termination shock, the heliopause and possible bow shock. In this sense, this paper should not be considered as a complete review of progress in the field. Many different approaches were used to look into different aspects of the solar wind interaction with LIC connecting with pickup ion transport and acceleration, with the termination shock structure under influence of ACRs and pickup ions. For more complete overview see recent reviews \\cite{Zank_1999},\\cite{Fichtner_2001}.  The structure of the paper is the following: The next section briefly describes our current knowledge of the local interstellar and solar wind parameters. Section 3 discusses theoretical approaches to be used for the interstellar and solar wind components. Section 4 gives an overview of heliospheric interface models. Section 5 describes basic results of the Baranov-Malama model of the heliospheric interface and its future developments. In section 6 we demonstrate possible analyses of space experiments on the basis of a theoretical model of the heliospheric interface. Section 7 underlines current problems in the modeling of the global heliosphere and discusses future perspectives. ",
        "conclusions": ""
    },
    "0208/astro-ph0208083_arXiv.txt": {
        "abstract": "A recent re-analysis of Whole Earth Telescope observations of GD~358 obtained in 1990 suggests that asteroseismology of additional DBV white dwarfs can lead to independent constraints on the important, but poorly determined, $^{12}{\\rm C}(\\alpha,\\gamma)^{16}{\\rm O}$ nuclear reaction rate. Data exist for several other DBV white dwarfs, but relatively few modes are detected and there is often no multiplet structure to aid in the identification of the spherical harmonic indices ($\\ell$, $m$). I use a new grid of one million DBV models covering a broad range of masses, temperatures, and surface helium layer masses to investigate the consequences of assuming $m$=0 for global model-fitting. I find that when the spherical degree is known and the rotation period is of order 1 day, the model-fitting procedure applied to modes with unknown $m$-values will still correctly identify the {\\it families} of possible solutions, and has a high probability of identifying the same globally optimal solution found when the $m$-value is known. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0208/hep-th0208157_arXiv.txt": {
        "abstract": "We discuss the theory of dark energy based on maximally extended supergravity and suggest a possible anthropic explanation of the present value of the cosmological constant and of the observed  ratio between dark energy and energy of matter. ",
        "introduction": "After  many desperate attempts to prove that the cosmological constant must vanish, now we face an even more complicated problem. We must understand why the cosmological constant, or the slowly changing dark energy, is at least 120 orders smaller than the Planck density $M_p^4$, and, simultaneously, why its value is as large as $\\Omega_D \\rho_0$. Here $\\rho_0 \\sim 10^{-29}$ g/cm$^3 \\sim 10^{-120} M_p^4$ is the total density of matter in the universe at present, including the cosmological constant, or dark energy, and $\\Omega_D \\sim 0.7$. One of the most interesting attempts to provide such an explanation is related to anthropic principle \\cite{barrow}.  This principle for a long time was rather controversial. It was based on an implicit assumption that the universe was created many times until the final success. It was not clear who did it and why was it necessary to make it  suitable for our existence. Moreover, it would be much simpler to create proper conditions for our existence in a small vicinity of a solar system rather than in the whole universe.  These problems were resolved with the invention of inflationary cosmology. First of all, inflationary universe itself, without any external intervention, may  produce exponentially large domains with all possible laws of low-energy physics \\cite{LindeCam}.  And  if the conditions suitable for our existence are established near the solar system, inflation ensures that similar conditions appear everywhere within the observable part of the universe. In addition to considering a single inflationary universe consisting of many domains with different values of constants, one may also consider a possibility that the inflationary  universe may be born in many different quantum states  with different values of coupling constants, see e.g. \\cite{Hawking:hk,book,Vilenkin,BellLin}.   This provides a simple justification of the cosmological anthropic principle  and allows one to apply it to the cosmological constant problem.   If, for example, the cosmological constant is large and negative, $\\Lambda \\ll -10^{-29}$ g/cm$^3$, then such a universe, even if it is flat, would collapse within the time that is  smaller than the age of our universe $t \\sim 14$ billion years \\cite{barrow,linde300}. This would make our life impossible.  One may wonder whether intelligent life could  emerge within $7$ billion years or $5$ billion years, but we have no reason to believe that it could  happen  on a much shorter time scale.  \\begin{figure}[h!] \\centering\\leavevmode\\epsfysize=5.2cm \\epsfbox{negative.eps}  \\  \\caption[fig1] {Evolution of a flat $\\Lambda$CDM universe for various values of $\\Lambda$. Time is in units of the present age of the universe $t_0 \\approx 14$ billion years. The present moment is placed at $t=0$,  the big bang corresponds to $t = -1$. The upper (red) line corresponds to the flat universe with $\\Omega_{\\rm tot} = 1$,  $\\Omega_\\Lambda =  0.7$ (i.e. $\\Lambda = +0.7 \\rho_0$)  and $\\Omega_{M} =  0.3$. The next line below it (the blue line) corresponds to a flat  universe with $\\Lambda = -4.7\\rho_0$. As we see, this universe  collapses at the age of 14 billion years. The total lifetime of the universe with $\\Lambda = -18.8 \\rho_0$ (the lower, green line) is only 7 billion years. } \\label{negative} \\end{figure}  Since we are entering the age of precision cosmology, let us improve the order-of-magnitude estimates of \\cite{barrow,linde300} and obtain a  numerical constraint on negative $\\Lambda$. The investigation is  straightforward, so we will simply show the results in Fig.  \\ref{negative}. We find that the anthropic constraint on the negative cosmological constant  is a bit less stringent than  it was  anticipated.  If $7$ billion years is sufficient  for emergence of human life, then   $\\Lambda \\gtrsim -18.8 \\rho_0\\sim - 2 \\times 10^{-28}$ g/cm$^3$.   If we really need about $14$ billion years, as one can infer from  \\cite{Lineweaver:2002gd}, then the constraint is somewhat stronger: $\\Lambda \\gtrsim -4.7 \\rho_0 \\sim -5 \\times 10^{-29}$ g/cm$^3$.         However,  the present observational data suggest that $\\Lambda >0$. In this case the use of the anthropic considerations become more involved. In \\cite{linde300} it was argued that the life of our type is impossible for $\\Lambda \\gg 10^{-29}$ g/cm$^3$ because in this case the density of matter of the universe would be exponentially small due to its exponential expansion at the present stage. A more precise and rigorous constraint was obtained later by Weinberg \\cite{Weinberg87}. He pointed out that the process of galaxy formation occurs only up to the moment when the cosmological constant begins to dominate the energy density of the universe and the universe enters the stage of late-time inflation. (By galaxy formation we understand the growth of density contrast until the moment when the galaxy separates from the general cosmological expansion of the universe. After that, its density rapidly grows by a factor of $O(10^2)$, and its subsequent evolution becomes much less sensitive to the  value of $\\Lambda$.) For example, one may consider galaxies  formed  at $z \\gtrsim 4$, when the energy density of the universe was 2  orders of magnitude greater than it is now. Such galaxies would not form if  $\\Lambda \\gtrsim 10^2 \\rho_0 \\sim 10^{-27}$ g/cm$^3$.   Thus, anthropic considerations may reduce the disagreement between the theoretical expectations ($\\Lambda \\sim M_p^4$) and observational data ($\\Lambda \\sim \\rho_0 \\sim 10^{-120} M_p^4$)  from $120$ orders of magnitude to only 2 orders of magnitude! But this  is not yet a complete solution of the cosmological constant problem. Assuming that all values of the cosmological constant are equally probable, one would find himself in a universe with $\\Lambda \\sim \\rho_0$ with the probability about $1\\%$.  The next important step  was made in \\cite{Efstathiou,Vilenkin:1995nb,Weinberg96,Garriga:1999bf,Bludman:2001iz}. The authors considered not only our own galaxy, but all other galaxies that could harbor  life of our type. This would include not only the existing galaxies but also the galaxies that are being formed at the present epoch. Since the energy density al later stages of the evolution of the universe becomes smaller, even a very small cosmological constant may disrupt the late-time galaxy formation, or may prevent the growth of existing galaxies.   This allowed the authors of \\cite{Efstathiou,Vilenkin:1995nb,Weinberg96,Garriga:1999bf,Bludman:2001iz} to strengthen the constraint on the cosmological constant. According to \\cite{Weinberg96}, the probability that an astronomer in any of the universes would find a value of $\\Omega_D = \\Lambda/\\rho_0$ as small as $0.7$ ranges from $5\\%$ to $12\\%$, depending on various assumptions.  However,  our goal is not to find suitable conditions for the human life in general, but rather to explain the results of our observations. These results include the fact that for whatever reason we live in an internal part of the galaxy that probably could not be strongly affected by the existence of a cosmological constant $\\Lambda \\sim \\rho_0$.  Does it mean that we are not typical observers since we live in an atypical part of the universe where we are protected against a small cosmological constant $\\Lambda \\sim \\rho_0$? Also, galaxy formation is not a one-step process. The central part of our galaxy was formed very early, at $z \\gtrsim 20$, when the energy density in the universe was 4 orders of magnitude greater than it is now. To prevent formation of such regions one would need to have $\\Lambda \\gtrsim 10^4 \\rho_0 \\sim 10^{-25}$ g/cm$^3$. It may happen that the probability of emergence of life in such regions, or in the early formed dwarf galaxies, is very small. Moreover, one could argue that the probability of emergence of life is proportional to the fraction of matter condensed into large galaxies \\cite{Efstathiou,Vilenkin:1995nb,Weinberg96,Garriga:1999bf,Bludman:2001iz}. Even if it is so, in an eternally existing inflationary universe there should be indefinitely many  regions suitable for existence of life, so life would eventually appear in one of such places even if the probability of such event is extremely small. A more detailed investigation of this issue is in order \\cite{GarVil}.  In this respect the anthropic constraint on $\\Lambda <0$ seems to be  less ambiguous. But it is also less important since it does not seem to apply to an accelerating universe with $\\Lambda \\approx 0.7 \\rho_0$. In this paper we will show, however, that a similar constraint based on investigation of the total lifetime of a flat universe can be derived in a broad  class of theories based on N=8 supergravity that can describe the present stage of acceleration \\cite{Kallosh:2001gr,Linde:2001ae,Kallosh:2002wj,Kallosh}. This may allow us to avoid fine-tuning that is usually required to explain the observed value of $\\Omega_D$. ",
        "conclusions": "Most of the theories of dark energy have to face two problems. First of all, it is necessary to explain why does the bare cosmological constant vanish. Then one must find a dynamical mechanism imitating a small cosmological constant and explain why $\\Omega_D\\sim 0.7$ at the present stage of the evolution of the universe.  In this paper we studied cosmological consequences of the simplest toy model of dark energy based on $N=8$ supergravity. We have found that in the context of this theory one can completely resolve the cosmological constant problem, as well as the coincidence problem plaguing most models of quintessence.  Indeed, one simply cannot add a cosmological constant to this theory.   The only way to introduce something similar to the cosmological constant is to put the system close to the top of the effective potential. If the potential is very high, then  it is also very curved, $V''(0) = -2V(0)$. We have found that the universe can live long enough only if  the field $\\phi$ initially is within the Planck distance from the top, $|\\phi|\\lesssim M_p$, which sounds reasonable, and if   $V(0)$, which plays the role of $\\Lambda$ in this theory, does not exceed much the critical value $\\rho_0 \\sim 10^{-120} M_p^4$.   In our paper we made the simplest assumption that the probability to live in the universe with different $\\Lambda$ and $\\phi_0$ does not depend on their values. However, in realistic models the situation may be different. For example, as we mentioned,  $\\Lambda^{1/2}$  is related to the  4-form flux in $d=11$ supergravity, see Eq. (\\ref{lam}). This may suggest that the probability distribution should be uniform not with respect to $\\Lambda$ and $\\phi_0$ but with respect to $\\Lambda^{1/2}$ and $\\phi_0$. We studied this possibility and found that the numerical results change, but the qualitative features of the model remain the same.  The probability distribution for $\\phi_0$ also may depend on $\\phi_0$ even if $V(\\phi)$ is very flat at $\\phi <1$. First of all, it might happen that the fields $\\phi \\gg 1$ (i.e. $\\phi \\gg M_p$) are  forbidden, or the effective potential at large $\\phi$ blows up. This is often the case in $N=1$ supergravity. Secondly, interactions with other fields in the early universe may create a deep minimum capturing the field $\\phi$ at some time-dependent point $\\phi < 1$. This also often happens in $N=1$ supergravity, which constitutes one of the features related to the cosmological moduli problem. If this happens in our model, one will be able to ignore the region of $\\phi_0>1$ (the right part of Figs. \\ref{lam2}, \\ref{combo}) in the calculation of probabilities. This will increase the probability to live in an accelerating universe  with $0.5 < \\Omega_D < 0.9$.  In our estimates we assumed that the universe must live as long as 14 billion years before the human life appears. One could argue that the first stars and planets were formed long ago, so we may not need much more than about 5-7 billion years for the development of life. This would somewhat decrease our estimate for the probability to live in an accelerating universe  with $0.5 < \\Omega_D < 0.9$, but this would not alter our results qualitatively. On the other hand, most of the planets were probably formed at later stages of the evolution of the universe \\cite{Lineweaver:2002gd}, so one may argue that the probability of emergence of human life becomes much greater at $t > 14$ billion years, especially if one keeps in mind how many coincidences  have made our life possible. If one assumes that human life is extremely improbable (after all, we do not have any indications of its existence anywhere else in the universe), then one may argue that the probability of emergence of life becomes significant only if the total lifetime of the universe can be much greater than 14  billion years. This would increase our estimate for the probability to live in an accelerating universe  with $0.5 < \\Omega_D < 0.9$.  So far we did not use any considerations based on the theory of galaxy formation \\cite{Weinberg87,Efstathiou,Vilenkin:1995nb,Weinberg96,Garriga:1999bf,Bludman:2001iz}. If we do so,  the  probability of  emergence of life for $\\Lambda \\gg \\rho_0$ will be additionally suppressed, which will  increase the probability to live in an accelerating universe  with $0.5 < \\Omega_D < 0.9$.   To the best of our knowledge, only in the models based on extended supergravity the relation $|m^2| \\sim H^2$, together with the absence of freedom to add the bare cosmological constant,  is a property of the theory rather than of a particular dynamical regime. That is why the increase of $V(\\phi)$ in such models entails the increase in $|m^2|$. This, in its turn, speeds up the development of the cosmological instability, which  leads  to the anthropically unacceptable consequences.  The $N=8$ model discussed in our paper is just a toy model. In this model we were able to find a complete solution to the cosmological constant problem and to the coincidence problem (explaining why $\\Lambda \\sim \\rho_0$ and why $\\Omega_D$ noticeably differs both from $0$ and from $1$ at the present stage of the evolution of the universe). This model has important advantages over many other theories of dark energy, but to make it fully realistic one would need to construct a complete theory of all fundamental interactions including the dark energy sector described above. This is a very complicated task that is beyond the scope of the present investigation. However, most of our results are not model-specific. For example, instead of $N=8$ supergravity one could study any model with the effective potential of the type $V(\\phi) = \\Lambda(1 -\\alpha\\phi^2)$ with $\\alpha = O(1)$. Another example is provided by the simplest N=1 Pol\\'{o}nyi-type SUGRA  model with a very low scale of supersymmetry breaking and with a minimum of the effective potential at $V(\\phi) <0$. As shown in \\cite{Kallosh}, models of this type also have the crucial property $|m^2| \\sim H^2$. In fact, this property is required in most of the models of quintessence.  Therefore it would be interesting to apply our methods to the models not necessarily related to extended supergravity. A particularly interesting example is the axion quintessence model. The original model suggested in \\cite{Frieman:1995pm} has the  potential $\\Lambda [\\cos \\left(\\phi/f\\right)+C]$, and it was assumed  that $C=1$. It was however emphasized in \\cite{Frieman:1995pm} that this is just an assumption. The positive definiteness of the potential  with $C=1$ and the fact that it has a minimum at $V=0$ could be motivated, in particular, by the global supersymmetry arguments. In supergravity and M/string theory these arguments are no longer  valid and the derivation  of the value of the  parameter $C$ is not available. In \\cite{Choi:1999xn} the axion model of quintessence was studied using the arguments based on M/string theory.  The potential was given in the form $ V= \\Lambda \\cos \\left(\\phi/f\\right) $ without any constant part.  This  potential has a maximum at $\\phi = 0$, $V(0)=\\Lambda$. The universe collapses when the field $\\phi$  rolls to the minimum of its potential  $V(f\\pi)=-\\Lambda$. The curvature of the effective potential in its maximum is given by $ m^2= -{\\Lambda/f^2} = -{3H_0^2/f^2} $. For $f = M_p = 1$  one finds $ m^2= -{\\Lambda} = -{3} H_0^2, $ and for $f = M_p/\\sqrt 2$ one has $ m^2= -{2\\Lambda} = -{6} H_0^2$, exactly as in the $N=8$ supergravity. Therefore  the anthropic constraints on $\\Lambda$ based on the investigation of the collapse of the universe in this model (for $C=0$) are similar to the constraints obtained in our paper for the $N=8$ theory \\cite{KKL}. However, in this model, unlike in the models based on extended supergravity, one can easily add or subtract any value of the cosmological constant. In order to obtain useful anthropic constraints on the cosmological constant in this model one should use a combination of our approach with the usual approach based on the theory of galaxy formation \\cite{Weinberg87,Efstathiou,Vilenkin:1995nb,Weinberg96,Garriga:1999bf,Bludman:2001iz}.  In this sense, our main goal was not to replace the usual anthropic approach to the cosmological constant problem, but to suggest its possible enhancement. We find it very encouraging that  our approach may strengthen the existing anthropic constraints on the cosmological constant in the context of the theories based on extended supergravity.  One may find it hard to believe that in order to explain the results of cosmological observations one should consider theories with an unstable vacuum state. However, one should remember that exponential expansion of the universe during inflation, as well as the process of galaxy formation, is the result of the gravitational instability, so we should learn how to live with the idea that our world can be unstable. Also, we did not willingly pick up the theories with an unstable vacuum. We wanted to find the models based on M/string theory that would be capable of describing de Sitter state. All models related to M/string theory that we were able to find so far, with exception of the recently constructed model based on $N=2$ supergravity \\cite{Fre:2002pd}, lead to unstable dS vacuum. So maybe we need to take this instability seriously.  This brings us good news and bad news. The bad news  is that in all the theories we have considered in this paper, our part of the universe is going to collapse within the next 10-20 billion years or so.  The good news is that we still have a lot of time to find out whether this is really going to happen.    \\subsection*{ Acknowledgements} It is a pleasure to thank  T. Banks, M. Dine, T. Dent, J. Frieman, N. Kaloper, A. Klypin,  L. Kofman, D. Lyth,  L. Susskind, A. Vilenkin and S. Weinberg for useful discussions. This work was supported by NSF grant PHY-9870115. The work by A.L. was also supported by the Templeton Foundation grant No. 938-COS273."
    },
    "0208/astro-ph0208030_arXiv.txt": {
        "abstract": "{We report the results of a search for a point X-ray source (stellar remnant) in the southwest protrusion of the supernova remnant G\\,315.4$-$2.30 (MSH\\,14$-$6{\\it 3}, RCW\\,86) using the archival data of the {\\it Chandra X-ray Observatory}. The search was motivated by a hypothesis that G\\,315.4$-$2.30 is the result of an off-centered cavity supernova explosion of a moving massive star, which ended its evolution just near the edge of the main-sequence wind-driven bubble. This hypothesis implies that the southwest protrusion in G\\,315.4$-$2.30 is the remainder of a pre-existing bow shock-like structure created by the interaction of the supernova progenitor's wind with the interstellar medium and that the actual location of the supernova blast center is near the center of this hemispherical structure. We have discovered two point X-ray sources in the ``proper\" place. One of the sources has an optical counterpart with the photographic magnitude $13.38\\pm0.40$, while the spectrum of the source can be fitted with an optically thin plasma model. We interpret this source as a foreground active star of late spectral type. The second source has no optical counterpart to a limiting magnitude $\\sim 21$. The spectrum of this source can be fitted almost equally well with several simple models (power law: photon index $=1.87$; two-temperature blackbody: $kT_1 =0.11$ keV, $R_1 =2.34 $ km and $kT_2 =0.71$ keV, $R_2 =0.06$ km; blackbody plus power law: $kT =0.07$ keV, photon index $=2.3$). We interpret this source as a candidate stellar remnant (neutron star), while the photon index and non-thermal luminosity of the source (almost the same as those of the Vela pulsar and the recently discovered pulsar PSR J\\,0205$+$6449 in the supernova remnant 3C\\,58) suggest that it can be a young ``ordinary\" pulsar. ",
        "introduction": "\\object{G\\,315.4$-$2.30} (\\object{MSH\\,14$-$6{\\it 3}}, \\object{RCW\\,86}) is a bright (in both radio and X-ray) shell-like supernova remnant (SNR) with a peculiar protrusion to the southwest (e.g. Dickel et al. \\cite{dic01}, Vink et al. \\cite{vin97}). This protrusion encompasses a bright hemispherical optical nebula (Rodger et al. \\cite{rod60}, Smith \\cite{smi97}; see also Fig.~\\ref{arc}). The characteristic angular size of the SNR is about $40^{'}$, that at a distance to the remnant of 2.8 kpc (Rosado et al. \\cite{ros96}) corresponds to $\\simeq 32$ pc. The radius of the optical nebula is $\\simeq 2^{'}$ (or $\\simeq 1.6$ pc). A collection of observational data points to the young age (few thousand years) of the SNR (see e.g. Dickel et al. \\cite{dic01}).  Vink et al. (\\cite{vin97}) put forward the idea that the SNR G\\,315.4$-$2.30 is the result of a supernova (SN) explosion inside a pre-existing wind-driven cavity (see also Dickel et al. \\cite{dic01}, Vink et al. \\cite{vin02}) and noted that the elongated shape of the SNR resembles that of a wind-driven cavity created by a {\\it moving} massive star (see Weaver et al. \\cite{wea77} and Brighenti \\& D'Ercole \\cite{bri94} for details). On the other hand, it is believed that the origin of the southwest protrusion is due to the interaction of the SN blast wave with a density enhancement in the ambient interstellar medium. However, as was correctly noted by Dickel et al. (\\cite{dic01}), the density enhancement (e.g. a high-density cloud) should result in a concave dent in the shell, not in a protrusion. Dickel et al. (\\cite{dic01}) also suggested that this protrusion ``is perhaps the key to what is going on\".  We agree with the idea that G\\,315.4$-$2.30 is a diffuse remnant of an off-centered cavity SN explosion and supplement this idea by the suggestion that the massive SN progenitor star exploded near the edge of the main-sequence bubble. This suggestion implies that the southwest protrusion is the remains of a bow shock-like structure created in the interstellar medium by the post-main-sequence winds (see Sect.\\,4 and Gvaramadze \\cite{gva02}; cf. Wang et al. \\cite{wan93}) and that the SN exploded near the center of this hemispherical structure. Given the youth of the SNR and assuming a reasonable kick velocity for the stellar remnant, one can expect that the stellar remnant should still be within the protrusion. Motivated by these arguments, Gvaramadze (\\cite{gva02}) searched for a possible compact X-ray source to the southwest of G\\,315.4$-$2.30 using the {\\it ROSAT} archival data, but the moderate spatial resolution of the {\\it ROSAT} PSPC precluded detection of point sources against the bright background emission of the SNR's shell.  In this paper we report the discovery of two point X-ray sources near the center of the hemispherical optical nebula using the archival Advanced CCD Imaging Spectrometer (ACIS) data of the {\\it Chandra X-Ray Observatory}. We interpret one of the sources as a foreground active star of late spectral type, and the second one as a candidate stellar remnant (neutron star).  Note that Vink et al. (\\cite{vin00}) also reported the discovery of a point X-ray source in the southwest half of G\\,315.4$-$2.30, at about $7^{'}$ from the geometrical center of the SNR. We recall that Vink et al. (\\cite{vin97}) suggested that the SN blast wave in this SNR takes on the shape of the pre-existing elongated cavity (created by the stellar wind of the moving SN progenitor star). However, the initially spherical shape of the wind-driven cavity could be significantly affected by the stellar motion only if the massive star reaches the edge of the cavity and the stellar wind starts to interact directly with the ambient interstellar medium (e.g. Weaver et al. \\cite{wea77}); this implies that the SN explodes near the edge of the future (young) SNR (see Sect.\\,4). However, the source discovered by Vink et al. (\\cite{vin00}) is located too far from the edge of G\\,315.4$-$2.30. Moreover, the spectral characteristics of the source and the presence of a possible optical counterpart suggest that it is an active star rather than the stellar remnant associated with the SNR (Vink et al. \\cite{vin00}; see also Sect.\\,3.1). ",
        "conclusions": "We have analyzed the archival {\\it Chandra X-Ray Observatory} data on G\\,315.4$-$2.30 to search for a stellar remnant in the southwest corner of this SNR. The search was motivated by the hypothesis that the SNR G\\,315.4$-$2.30 is the result of an off-centered cavity SN explosion of a moving massive star, which ends its evolution just near the edge of the main-sequence wind-driven bubble. This hypothesis implies that the southwest protrusion in G\\,315.4$-$2.30 is the shocked material of a pre-existing circumstellar structure and that the actual location of the SN blast center is near the center of this structure. We have discovered two point X-ray sources in the ``proper\" place. One of the sources is interpreted as a foreground active star of late spectral type, while the second one as a candidate neutron star (perhaps a young ``ordinary\" pulsar). The follow-up observations of these sources will help us to understand their nature and thereby to test the hypothesis of the origin of the SNR G\\,315.4$-$2.30."
    },
    "0208/astro-ph0208206_arXiv.txt": {
        "abstract": "We consider the evolution of a warped disc around a Kerr black hole, under conditions such that the warp propagates in a wavelike manner. This occurs when the dimensionless effective viscosity, $\\alpha$, that damps the warp is less than the characteristic angular semi-thickness, $H/R$, of the disc.  We adopt linearized equations that are valid for warps of sufficiently small amplitude in a Newtonian disc, but also account for the apsidal and nodal precession that occur in the Kerr metric.  Through analytical and time-dependent studies, we confirm the results of Demianski \\& Ivanov, and of Ivanov \\& Illarionov, that such a disc takes on a characteristic warped shape.  The inner part of the disc is not necessarily aligned with the equator of the hole, even in the presence of dissipation.  We draw attention to the fact that this might have important implications for the directionality of jets emanating from discs around rotating black holes. ",
        "introduction": "Accretion discs are found around black holes in the centres of active galaxies and also in galactic X-ray binary stars.  These systems may form in such a way that the angular momentum vectors of the disc and of the black hole are not parallel.  As a particle in a tilted orbit around a rotating black hole undergoes Lense--Thirring precession at a rate dependent on the radius of the orbit, so a tilted disc experiences a gravitomagnetic torque that tends to twist and warp the disc.  If the disc is able to communicate a warping disturbance in a diffusive manner as a result of its pressure and viscosity, it may adopt a characteristic warped shape in which the plane of the disc undergoes a smooth transition from one plane to another in the vicinity of a certain radius (Bardeen \\& Petterson 1975; Kumar \\& Pringle 1985; Scheuer \\& Feiler 1996).  At large radius the disc is essentially flat and its plane is determined by the total angular momentum vector of the disc.  At small radius the disc is essentially flat and lies in the equatorial plane of the black hole.  Over a longer time-scale the disc and hole tend towards mutual alignment (Scheuer \\& Feiler 1996; Natarajan \\& Pringle 1998).  However, there are circumstances in which a warping disturbance propagates in a wavelike, rather than diffusive, manner.  This occurs in a Keplerian disc when the dimensionless viscosity parameter $\\alpha$ (Shakura \\& Sunyaev 1973) is smaller than the angular semi-thickness $H/R$ of the disc (Papaloizou \\& Lin 1995; Wijers \\& Pringle 1999; see the discussion by Pringle 1999); in a non-Keplerian disc the propagation is also wavelike, although it is dispersive (Ogilvie 1999).  The evolution of a disc around a rotating black hole under these conditions has received less attention.  On the basis of analytical calculations, Demianski \\& Ivanov (1997) and Ivanov \\& Illarionov (1997) found that the disc can adopt a steady warped shape in which the tilt angle is an oscillatory function of radius.  More recently, Nelson \\& Papaloizou (2000) conducted smoothed particle hydrodynamic simulations of accretion discs around rotating black holes under a variety of assumptions.  They did not find the tilt oscillations, but found that the inner part of the disc was aligned with the equatorial plane of the hole.  In this paper we re-examine the shape of a warped disc around a Kerr black hole, under conditions such that the warp propagates in a wavelike manner.  We explore the reasons for the existence of a steady wavelike solution, and discuss the expected wavelength, amplitude and phase of the oscillations in disc tilt.  Using time-dependent calculations of the linearized equations, we study the process by which such a steady solution is established.  Finally, we discuss the potentially important implications of the shape of the disc for the directionality of jets emanating from discs around black holes. ",
        "conclusions": "The result that, in low-viscosity discs around a Kerr black hole, the inner parts of the disc are not necessarily aligned with the black hole, as found by Ivanov \\& Illarionov (1997), is a general one. Indeed, depending on the disc properties, it is possible for the inner disc to be tilted at a greater angle to the hole than the outer parts of the disc. In addition, because the inner disc shape depends sensitively on the radial dependence of disc properties (such as surface density and disc thickness), a change (for example) in the accretion rate can give rise to a change in the inner disc warp, even without changing the tilt of the outer disc. These results contrast with the usual finding (e.g. Nelson \\& Papaloizou 2000) and/or assumption (e.g.  Natarajan \\& Pringle 1998) that the inner regions of the disc align with the equator of the hole. In the discussion above, we have noted above that confirmation of these results awaits a proper calculation using full general relativity, as well as an assessment of possible non-linear, dispersive and parametric effects. Nevertheless, it is evident that the results presented here could have important implications for the directions in which jets might emanate from accreting spinning black holes. Indeed, if the region responsible for direction of jet collimation is at several radii from the hole, which is likely to be the case for relativistic jets, then the Newtonian approximations applied above may be adequate to confirm the effect, even if the very inner regions of the disc are indeed aligned by the fully relativistic effects close to the hole. A lack of correlation between the inner and outer disc tilts could provide one explanation of the finding by Kinney et al. (2000) that the directions of jets from low luminosity AGN appear to be uncorrelated with the disc plane of the host spiral galaxies."
    },
    "0208/astro-ph0208495_arXiv.txt": {
        "abstract": "{Recurrent flaring events at X- and soft gamma-ray energies have been recently reported for the galactic black hole candidate Cygnus X-1. The observed fluxes during these transient outbursts are far higher than what is observed in ``normal'' episodes. Here we suggest that the origin of this radiation is non-thermal and produced by inverse Compton interactions between relativistic electrons in the jet and external photon fields, with a dominant contribution from the companion star field. The recurrent and relatively rapid variability could be explained by the precession of the jet, which results in a variable Doppler amplification. ",
        "introduction": "Cygnus X-1 is the most extensively studied black hole candidate in the Galaxy. It is a very bright X-ray binary with a compact object of $\\sim 10.1$ $M_{\\odot}$ and a companion O9.7 Iab star of $\\sim17.8$ $M_{\\odot}$ (Herrero et al. 1995), at an estimated distance of $\\sim 2$ kpc (e.g. Gierli\\'nski et al. 1999 and references therein). As in other sources of this type, the X-ray emission switches between soft and hard states, being most of the time in the latter. The spectrum in both states can be approximately represented as the sum of a blackbody plus a power law with exponential cut-off (e.g. Poutanen et al. 1997). During the soft state the blackbody component is dominant and the power law is steep, with a photon spectral index $\\Gamma\\sim 2.8$ (e.g. Frontera et al. 2001). During the hard state more energy is in the power law component, which is then harder, with photon indices $\\sim 1.6$ (e.g. Gierli\\'nski et al. 1997). The blackbody component is usually understood as emission from a cold, optically thick accretion disk, whereas the power law component is thought to be originated in an optically thin hot corona by thermal Comptonization of disk photons (Poutanen et al. 1997, Dove et al. 1997). The hot corona would fill the inner few tens of gravitational radii around the black hole. The accretion disk penetrates only marginally in the coronal region. In the hard state the thermal X-ray emission is dominated by the corona, with typical luminosities of a few times $10^{37}$ erg s$^{-1}$. In the soft states, the disk approaches to the black hole and then most of the energy dissipation occurs through it (Poutanen et al. 1997; see also Poutanen \\& Coppi 1998). Cygnus X-1 has a persistent, mildly variable, compact continuum counterpart of flat spectrum (e.g. Pooley et al. 1999). During many years, evidence for non-thermal radio jets in Cygnus X-1 was lacking, despite the efforts of the observers (e.g. Mart\\'{\\i} et al. 1996). Finally, the jet was detected by Stirling et al. (2001) at milliarcsecond resolution using VLBA observations. The jet-like feature extends up to $\\sim 15$ mas with an opening angle of less than 2 degrees. The spectum seems to be flat, and no counterjet is observed. The total radio emission at 8.4 GHz is $\\sim 11$ mJy, with variations of $\\sim2$ mJy over timescales of 2 days (Stirling et al. 2001). The average angle with the line of sight, if the jet is perpendicular to the disk, seems to be $\\sim 30^{\\circ}$ (Fender 2001). Very recently, the Interplanetary Network detected a transient soft-gamma ray event from the general direction of Cygnus X-1 (Golenetskii et al. 2002). Analysis of previous data indicates that at least other two events were observed during 1995. These latter events were also detected by BATSE instrument on the Compton Gamma-Ray Observatory, suggesting that they were originated in Cygnus X-1 (Schmidt 2002). The luminosities above 15 keV of the outbursts were in the range $1-2\\times 10^{38}$ erg s$^{-1}$, much higher than the typical thermal luminosity in the hard state. In this letter we suggest that these flaring events can be interpreted in terms of non-thermal microblazar activity (Mirabel \\& Rodr\\'{\\i}guez 1999). We study the effects of the interaction of the relativistic jet with the ambient photon fields from the accretion disk, the corona, and the companion star, and we calculate the expected non-thermal contribution to the keV-MeV spectrum. The recurrent character of the events can be explained through variable Doppler boosting originated in the precession of the jet (Kaufman-Bernad\\'o et al. 2002). In the next section we present the model, and then we discuss the implications. ",
        "conclusions": "The above outlined model incorporates the different known components of Cygnus X-1, accretion disk, corona, stellar companion, and relativistic jet, in an integrated picture where transient non-thermal outbursts are a natural and expected result. The amplitude of these outbursts can be similar to what has been recently observed in some intriguing flaring episodes in this source. Our model is different from the model recently proposed by Georganopoulos et al. (2002) not only because we incorporate the effects of precession, but also because we do not attempt to explain the bulk of X-ray emission as non-thermal {\\sl all the time}. This emission is normally dominated by thermal Comptonization in the hot corona around the black hole, except during the {\\sl microblazar} phase, and in this case we incorporate the effects of the interaction of the jet with the corona in our calculations. We emphasize that, as it is shown in Figures 3 and 4, during the transient microblazar phase the X- and soft $\\gamma$-ray spectrum will be softer than in the normal hard state, when the coronal emission dominates. This is an unavoidable consequence of the steepening produced by Compton losses in the injected electron spectrum and can be used to test our proposal, not only through new observations of Cygnus X-1, but also of other potential microblazars as LS5039 (Paredes et al. 2000). Recently, Brocksopp et al. (1999) have found multiwavelength evidence for the presence of a $142.0\\pm 7.1$ days period in Cygnus X-1. The optical and X-ray period seem to be originated in the precession of the accretion disc (Brocksopp et al. 1999), whereas the modulation at radio wavelengths is probably produced by the associated precession of the jet (see Pooley et al. 1999). The morphology of the extended radio jet, with a clear bend, is also consistent with a precession of the inner beam (Stirling et al. 2001). The periodic signal in the radio lightcurve, however, is not expected to be as strong as at high energies since the magnification factor for the synchrotron emission goes as $D^{(3+p)/2}$, whereas for the external Compton component it goes as $D^{2+p}$ (Georganopoulos et al. 2002). The time lag between the two high-energy flares observed in 1995 is $\\sim 75$ days, about a half of the value reported by Brocksopp et al. (1999), but since Cygnus X-1 is a wind-accreting system variations in the period along a span of several years are possible. Certainly, more observations on longer time spans are necessary to constrain the dynamical models. In the model presented here the duty cycle of the blazar phase is rather small, $\\sim 10\\%$. Future X-ray observations of non-thermal flares can be used for a better determination of the geometric parameters. An interesting feature of our model is that most of the gamma-rays produced within the coronal region will be absorbed by pair production. Sooner or later these pairs will annihilate producing a broad, blueshifted feature in the MeV spectrum. Details of calculation of the emerging spectrum are beyond the scope of this Letter (see Abraham et al. 2001), but it is clear that the forthcoming INTEGRAL satellite will be able to probe Cygnus X-1 spectrum and its temporal evolution at this energy range, helping to test and constrain the model here proposed. Hopefully, very soon we will be able to clarify the role played by non-thermal processes in this fascinating object."
    },
    "0208/astro-ph0208407_arXiv.txt": {
        "abstract": "{ We discovered that the WR9-type star WR 106 (HDE 313643) underwent a deep episodic fading in 2000.  The depth of the fading ($\\Delta V \\sim$ 2.9 mag) surpassed those of all known similar ``eclipse-like\" fadings in WR stars. This fading episode was likely to be produced by a line-of-sight episodic dust formation rather than a periodic enhancement of dust production in the WR-star wind during the passage of the companion star though an elliptical orbit.  The overall 2000 episode was composed of at least two distinct fadings.  These individual fadings seem to more support that the initial dust formation triggered a second dust formation, or that the two independent dust formations occurred by the same triggering mechanism rather than a stepwise dust formation. We also discuss on phenomenological similarity of the present fading with the double fading of R CrB observed in 1999--2000. ",
        "introduction": "Wolf--Rayet (WR) stars are massive, luminous stars which have blown away the hydrogen envelope, and are considered to be immediate precursors of some kinds of supernovae.  WCL-type stars are a carbon-rich, late-type subclass of WR stars [for the definition of the subclasses of WC-type stars, see e.g. \\citet{smi90}; more comprehensive information of WR stars can be found in the catalogue by \\citet{vdh01WRcatalog}. WC9-type stars are the coolest WCL-type stars which are characterized by strong C\\textsc{III} and C\\textsc{II} lines, and the weak or absent O\\textsc{V} feature \\citep{tor84WC9}. WC9-type stars have been receiving much astrophysical attention in that they are one of the most effectively dust-producing environments in stellar systems (for recent reviews, see \\cite{wil95,wil97}).  The dust-forming process in WCL-type (especially in WC9-type) stars is known to be either continuous or episodic.  The best-known continuous dust producer is a binary WR 104 (WC9+B0.5V), renowned for its ``dusty pinwheel nebula\" \\citep{tut99,tut02}.  Recently discovered large-amplitude optical variability even suggests the presence of a continuous ``dust jet\" in the direction of the rotation axis \\citep{kat02wr104}. WR 112 (WC9+OB?) has been also suspected to have a similar dusty pinwheel \\citep{mar02}.  \\begin{figure} \\begin{center} \\includegraphics[angle=0,width=4.1cm]{bright.eps} \\includegraphics[angle=0,width=4.1cm]{dim.eps} \\end{center} \\caption{Variation of WR 106 = Had~V84, recorded with photographs taken by one of the authors (KH). Each panel shows 10 arcmin square, north is up and east is left. The left and right panels were taken on 2000 Aug. 22 and 2000 Apr. 28, when the object was at 12.0 mag and 14.1 mag, respectively. Such dramatic variability of a Wolf--Rayet star is quite exceptional.} \\label{fig:image} \\end{figure}  \\begin{figure*} \\begin{center} \\includegraphics[angle=0,height=5.2cm]{lc.eps} \\end{center} \\caption{Light curve of WR 106.  The filled circles and squares represent observations by Takamizawa (Tmz) and Haseda (Had), respectively. The open triangles represent the upper limits.  The most prominent fading was recorded in early 2000.} \\label{fig:lc} \\end{figure*}  Another class of manifestation of dust production in WCL stars is episodic optical fading \\citep{veen97} or episodic infrared brightening \\citep{wil90}, which are considered to arise from temporary condensations of dust clouds.  In 2001 April, one of the authors (KH) serendipitously discovered a new variable star named Had~V84 (vsnet-alert 5856),\\footnote{ http://www.kusastro.kyoto-u.ac.jp/vsnet/alert5000/\\\\msg00856.html. } which was subsequently identified with WR 106 = HDE 313643 (Fig. \\ref{fig:image}). WR 106 is known to show a strong infrared excess \\citep{coh78IRspecphot,coh95IRASWC,kwo97IRASLRS,pit83IRphot}, which indicates substantial dust formation. We also noticed that the object was listed as No. 15357 in \\citet{fit73}, who suspected 0.13 mag $V$-band variability based an analysis of past photoelectric archival data.  The object was given a name for suspected variable star (NSV 10152), but the variability was not confirmed at that time. ",
        "conclusions": "WR 106 was studied for binarity by \\citet{wil00companion}.  The lack of evidence for a companion and the apparent lack of photometric periodicity (Fig. \\ref{fig:lc}) less favor the interpretation of a periodic enhancement of dust production in the WR-star wind during the passage of the companion star though an elliptical orbit, as has been proposed in WR 140 \\citep{wil90} and presumably WR 137 \\citep{wil01}.  The present phenomenon seems to be better understood as an ``eclipse-like\", line-of-sight dust formation as proposed by \\citet{veen97}. The depth of the present phenomenon, however, far surpasses those (up to 1.2 mag in visual wavelengths) of the previously known similar phenomena in other stars. Following the interpretation by \\citet{veen97}, the production rate of the optical depth or the dust production rate in the present episode should be at least a few times larger than in the previously recorded phenomena.  Furthermore, the observed depth severely constrains the amount of the unobscured scattered light to be less than 7 \\%.  The present phenomenon is composed of at least two distinct fadings (Fig. \\ref{fig:fading}).  \\citet{veen97} reported the presence of two-step fadings in some fadings.  \\citet{veen97} suggested several possibilities to explain the two-step fadings: (1) sudden enhancement of the dust production in response to an inflow of additional matter to the dust production area, (2) non-radial expansion of a neighboring cloud, or (3) formation of the second cloud in the shade of the first cloud.  In the present case, the close occurrence of two rare fadings suggests that they are not a chance superposition of two independent phenomena, but are more physically related. The similar observed depths and durations of the two fadings do not seem to support a stepwise formation of the dust cloud, as represented by the possibilities (1) and (3).  The present observation seems to more support that the initial dust formation somehow triggered a second dust formation in the proximity, or that the two independent dust formations occurred by the same triggering mechanism.  We also note phenomenological similarity of the present fading with the ``double fading\" of R CrB observed in 1999--2000 (Fig. \\ref{fig:rcrb}, the data are from VSNET.\\footnote{http://www.kusastro.kyoto-u.ac.jp/vsnet/.}). The fading mechanism proposed by \\citet{veen97} being analogous to the fading mechanism of R CrB stars (for a review, see \\cite{cla96rcrb}), the analogy may suggest a common underlying dust production mechanism between R CrB stars and WR 106.  Similar double fadings are also known in some [WC] stars (CPD$-$56$^{\\circ}$8032 = He3$-$1333 = V837 Ara: \\cite{pol92WC11phot}; V348 Sgr: \\cite{hec85v348sgrphot}), which are sometimes considered to be related to R CrB-type stars. It is widely believed that the dust formation in R CrB stars are associated with pulsation \\citep{cla96rcrb}. Although the large difference of the gravity and temperature between WR stars and R CrB stars may make it difficult to directly apply the R CrB-type dust formation to a WR star, a pulsation-type instability similar to that of R CrB stars in the outer WR wind may have caused a similar sequence of fadings in a WR star.  \\begin{figure} \\begin{center} \\includegraphics[angle=0,height=4.8cm]{rcrb.eps} \\end{center} \\caption{Light curve of the ``double fading\" of R CrB in 1999--2000. The data are from VSNET.} \\label{fig:rcrb} \\end{figure}  \\vskip 3mm  The authors are grateful to the observers who reported visual observations of R CrB to VSNET. This work is partly supported by a grant-in aid [13640239 (TK), 14740131 (HY)] from the Japanese Ministry of Education, Culture, Sports, Science and Technology. This research has made use of the Digitized Sky Survey producted by STScI, the ESO Skycat tool, and the VizieR catalogue access tool. This research has made use of the USNOFS Image and Catalogue Archive operated by the United States Naval Observatory, Flagstaff Station (http://www.nofs.navy.mil/data/fchpix/)."
    },
    "0208/astro-ph0208294_arXiv.txt": {
        "abstract": "We describe our method to construct line blanketed NLTE model atmospheres for hot stars. We employ the Accelerated Lambda Iteration and use statistical methods to deal with metal line blanketing. ",
        "introduction": "Stellar atmospheres are open systems and thus cannot be in thermodynamic equilibrium (TE). The ``Local Thermodynamic Equilibrium'' (LTE) is a working hypothesis which assumes TE not for the atmosphere as a whole but for small volume elements. As a consequence, the atomic population numbers are depending only on the local (electron) temperature and electron density via the Saha-Boltzmann equations. Computing models by replacing the Saha-Boltzmann equations by the rate equations (statistical equilibrium) are called non-LTE (or NLTE) models. NLTE calculations are more costly than LTE calculations, however, it is hard to predict if NLTE effects are important in a specific problem. Generally, NLTE effects are large at high temperatures and low densities, which implies intense radiation fields hence frequent radiative processes and less frequent particle collisions which tend to enforce LTE conditions.  We will restrict ourselves here to the classical model atmosphere problem, i.e.\\ the solution of the radiation transfer equations assuming hydrostatic, radiative and statistical equilibrium. The numerical problem going from LTE to realistic NLTE models has been solved only recently and is the topic of this paper. This was achieved by the development of new numerical techniques for model construction and on the availability of atomic data for many species. The replacement of the Saha-Boltzmann equations by the atomic rate equations requires a different numerical solution technique, otherwise metal opacities cannot be accounted for at all. Such techniques were developed with big success during the last decade, triggered by important papers by Cannon (1973) and Scharmer (1981). The Accelerated Lambda Iteration (ALI) is at the heart of of this development. Combined with statistical methods we are finally able to compute metal line blanketed NLTE models with a very high level of sophistication. ",
        "conclusions": "The construction of metal line blanketed models in hydrostatic and radiative equilibrium under NLTE conditions was the last and long-standing problem of classical model atmosphere theory and it is finally solved with a high degree of sophistication. The essential milestones for this development, starting from the pioneering work of Auer \\& Mihalas (1969) are:  \\begin{itemize} \\item Introduction of the Accelerated Lambda Iteration (ALI, or ``operator splitting'' methods), based upon early work by Cannon (1973) and Scharmer (1981). First ALI model atmospheres were constructed by Werner (1986). \\item Introduction of statistical approaches to treat the iron group elements in NLTE by Anderson (1989). \\item Linear formulation of the statistical equations (Rybicki \\& Hummer 1991, Hauschildt 1993). \\item Computation of atomic data by Kurucz (1991), by the Opacity Project (Seaton \\etal 1994) and the Iron Project (Hummer \\etal 1993). \\end{itemize}"
    },
    "0208/astro-ph0208541_arXiv.txt": {
        "abstract": "High-resolution CO($1\\to0$) observations of five ultraluminous infrared galaxies (ULIGs: $L_{\\rm IR} [8-1000\\mu{\\rm m}] \\gtrsim 10^{12}$ L$_\\odot$) with double nuclei are analyzed. These sources constitute a complete subset of local ULIGs expected to be in an intermediate stage of merging and selected with projected nuclear separations of $2\\farcs0-5\\farcs$4 (3--5 kpc) so they could be resolved with the Owens Valley Radio Observatory Millimeter Array.  The observed pairs include two mergers with cool far-infrared colors (25$\\micron$ to 60$\\micron$ flux density ratio $f_{\\rm 25\\mu{\\rm m}}/f_{\\rm 60\\mu{\\rm m}} < 0.2)$ from the {\\it Infrared Astronomical Satellite} (IRAS) Bright Galaxy Sample (IRAS 12112+0305 and IRAS 14348-1447) and three mergers with warm infrared dust temperatures ($f_{\\rm 25\\mu{\\rm m}}/f_{\\rm 60\\mu{\\rm m}} \\gtrsim 0.2$) selected from the IRAS Warm Galaxy Sample (IRAS 08572+3915, IRAS 13451+1232 = PKS 1345+12, and IRAS 13536+1836 = Mrk 463).  These ULIGs are further distinguished by the presence of pairs of active nuclei; among the ten nuclei, nine have Seyfert or LINER classifications and one is unclassified.  Molecular gas is detected only on the redder, more radio-luminous nucleus of the warm objects, whereas both nuclei of the cool ULIGs are detected in CO.  The inferred molecular gas masses for the detected nuclei are $0.1-1.2\\times10^{10}$ M$_\\odot$, and the undetected nuclei have molecular gas masses at least 1.2--2.8 times less than that of their CO-luminous companions.  Upper limits on the extent of the CO emitting regions of each detected nucleus range from 2--4 kpc, which is about 3-6 times smaller than the average effective CO diameter of nearby spiral galaxies. This is strong evidence that the high concentration of molecular gas is the result of tidal dissipation in ongoing mergers.  There is no correlation between the optical emission-line classification of the nuclei (i.e., Seyfert, LINER, or H II) and the presence of detectable molecular gas; however, there is a clear indication that the relative amount of molecular gas increases with the relative level of activity as measured via radio power and optical/near-infrared emission-line strength. Star formation rates are estimated to be in the range $\\sim 30-290$ M$_\\odot$ year$^{-1}$ nucleus$^{-1}$ by making assumption that the radio and infrared emission arise from supernovae and dust heating by massive stars, respectively; corresponding gas consumption timescales are $1-7\\times10^7$ years.  The nuclei detected in CO are extremely red at near-infrared wavelengths, suggestive of much dustier environments than in the companions undetected in CO.  Column density estimates are $N_{\\rm H_2} \\sim 10^{24-25}$ cm$^{-2}$, which correspond to more than 1000 magnitudes of extinction toward the nuclei at visual wavelengths.  Finally, the molecular gas mass densities and line-of-sight velocity dispersions show significant overlap with stellar densities and line-of-sight stellar velocity dispersions of local elliptical galaxies with $M_{\\rm V} < -19$ mag, including rapidly rotating ellipticals with disky isophotes and power-law light profiles as well as slowly rotating ellipticals with boxy isophotes and cores.  This provides strong evidence that the CO-rich nuclei of these ULIGs have the phase-space density of gas necessary to form the stellar cores of elliptical galaxies. ",
        "introduction": "Recent surveys have provided tantalizing evidence that galaxy mergers at high redshift may be the precursors of local massive, quiescent galaxies. The fraction of both optically-selected mergers and submillimeter/infrared- selected galaxies (the latter of which are observed locally to be mergers: e.g. Joseph \\& Wright 1985) has been shown to increase with increasing redshift (Patton et al.  1997; Smail et al. 1997; Barger et al.  1998; Hughes et al.  1998; Le F\\'{e}vre et al. 2000), and the space density of the high-redshift submillimeter population is observed to be similar to the space density of present day massive elliptical galaxies (Trentham 2001).  Further, N-body simulations have shown that the collision of disk galaxies naturally leads to the formation of massive elliptical galaxies (e.g., Barnes \\& Hernquist 1992 and references therein).  Given the likely connection between galaxy mergers and massive elliptical galaxy formation, reconstruction of the evolutionary steps -- beginning with the initial interaction of the progenitors and ending with the final merger byproduct -- will permit a determination of how the activity and dynamical state of galaxy mergers, and thus of massive galaxies, evolve as a function of evolutionary phase.  Molecular gas is a key ingredient in the merger process; N-body simulations employing a dissipational gas component (i.e., Barnes \\& Hernquist 1996; Mihos \\& Hernquist 1996) show that gas in the spiral disks of the progenitor galaxies is gravitationally torqued, resulting in either stripping of gas from the disk, or a net inflow of material to the circumnuclear regions of each progenitor. Multiwavelength observations of advanced infrared luminous mergers provide evidence that the enhanced molecular gas densities lead to prodigious star formation and, in some cases, mass accretion onto supermassive nuclear black holes occurs (see review by Sanders \\& Mirabel 1996).  The scenario presented above has been assembled from both theoretical and empirical work. However, observational gaps exist, and feedback mechanisms (i.e., star formation, supernovae winds) are not understood well enough to implement in simulations. In order to fill in the observational details regarding the role of molecular gas in the intermediate evolutionary stages of the most luminous mergers, a CO survey of ultraluminous infrared galaxies (ULIGs: $L_{\\rm IR} [8- 1000\\micron] \\gtrsim 10^{12}$L$_\\odot$) with double nuclei and $z < 0.15$ has been initiated.  These observations make use of the Owens Valley Radio Observatory (OVRO) Millimeter Array to resolve CO emission that may be associated with each nucleus.  This sample of ULIGs has been well-studied at other wavelengths (e.g., Condon et al. 1991, 1992; Murphy et al. 1996, 2001; Sanders, Scoville, \\& Soifer 1991; Sanders et al. 1988a,b; Scoville et al. 2000; Soifer et al. 2000; Solomon et al. 1997; Surace et al. 1998, Surace \\& Sanders 1999; Surace, Sanders, \\& Evans 2000; Veilleux, Sanders, \\& Kim 1999); thus correlations between the dynamical state of the merger (as determined via optical-to-infrared imaging) and the stellar and AGN activity occuring (as determined via optical-to-infrared spectroscopy and radio interferometry) can be investigated. In addition, limits can be placed on the column densities toward CO-detected nuclei and the gas mass density, which allow for a discussion of extinction in ULIGs and their possible association with stellar core formation in elliptical galaxies (e.g.  Kormendy \\& Sanders 1992).  Parts of our OVRO survey have been presented elsewhere (Evans et al. 1999; Evans, Surace, \\& Mazzarella 2000; Evans et al.  2000); galaxies discussed in these previous papers, as well as observations of related objects published by others (Sakamoto et al.  1999; Trung et al. 2001), are discussed along with new observations presented here for the first time.  This paper is divided into six Sections. The selection criteria of the galaxy sample is discussed in \\S 2. Section 3 is a summary of the new observations and data reduction. In \\S 4, the CO($1\\to0$) emission-line properties are presented, and the CO coordinates and morphologies are compared with the radio and HST optical and near-infrared morphologies. The method of calculating the molecular gas mass is briefly presented in \\S 5. Section 6 contains a discussion of the molecular properties of each nucleus in comparison with the optical/near-infrared and radio emission lines, determinations of the star formation rate and the column density, and comparisons of the gas mass density to the stellar densities of elliptical galaxy cores. Section 7 is a summary.  Throughout this paper, $H_0 = 75$ km s$^{-1}$Mpc$^{-1}$, $q_0 = 0.5$, and $\\Lambda = 0.0$ are assumed. ",
        "conclusions": "Molecular gas observations of a sample of double-nucleus ultraluminous infrared galaxies (ULIGs) were presented. This observations were motivated by the dearth of CO($1\\to0$) data for ULIGs at intermediate stages of evolution (i.e., with nuclear separations of 3--5 kpc).  The following conclusions have been reached:  \\noindent {\\it (1)} Three observed double-nucleus objects with warm far-infrared dust temperatures have detected CO emission associated only with one of the two nuclei -- the nucleus that is reddest and has the highest radio continuum emission. Two observed objects with cool far-infrared dust temperatures have detected CO emission associated with both nuclei.  Molecular gas masses for the detected nuclei are estimated to be in the range $0.1-1.2\\times10^{10}$ M$_\\odot$.  \\noindent {\\it (2)} The relative amount of molecular gas in each galaxy pair appears to be correlated with the relative levels of activity as measured by both optical/near-infrared recombination line emission and radio flux density.  The presence of LINER and Seyfert nuclei combined with the high central concentration of CO provides evidence that molecular gas is an important component in fueling AGN and vigorous, massive star formation.  \\noindent {\\it (3)} Where possible, star formation rates have been computed for each nucleus using four different techniques.  Star formation rates estimated to be 20--290 M$_\\odot$ yr$^{-1}$ nucleus$^{-1}$, with corresponding gas consumption timescales of $1-7\\times10^7$ years. The star formation rates may be considerably lower (and the gas consumption timescales considerable longer) if AGN contribute significantly to the infrared and radio emission observed in these galaxy pairs.  \\noindent {\\it (4)} The nuclei with associated molecular gas appear to have significantly redder infrared colors than their companions that lack CO detections, suggestive of dustier environments in the CO-luminous nuclei.  \\noindent {\\it (5)} Column densities for the nuclei of the ULIGs were estimated to be $\\sim 10^{24-25}$ cm$^{-2}$, which corresponds to greater than 1000 magnitudes of visual extinction.  \\noindent {\\it (6)} Where possible, molecular gas mass densities were estimated for each nucleus. A reasonable estimate for the range of molecular gas densities is $10^{2-4}$ M$_\\odot$ pc$^{-3}$. Such values, combined with measured line-of-sight CO($1\\to0$) velocity dispersions, are equivalent to the stellar mass densities and velocity dispersions of elliptical galaxies with $M_V < -19$ and indicate that the molecular gas has a sufficiently high phase-space density to form the stars in elliptical galaxy cores.  The present sample is small (5 galaxies), but the analysis presented here shows compelling results that may be solidified or refuted with a larger, more statistically significant sample of ULIGs in their intermediate phases of evolution.  A CO($1\\to0$) survey is presently underway to observe all of the intermediate stage ULIGs in the IRAS 2 Jy sample. These data will be combined with multiwavelength data in a manner similar to what has been presented for the five ULIGs in this paper. Further improvement of the present dataset will also be possible with high-resolution CO observations with the soon-to-be commissioned Smithsonian SubMillimeter Array (SMA) and the upcoming Combined Array for Research in Millimeter Astronomy (CARMA); these interferometers will make it possible to obtain CO resolutions approaching that of the 2$\\micron$ HST data, thus removing the ambiguity in the true extent of the molecular gas."
    },
    "0208/astro-ph0208588_arXiv.txt": {
        "abstract": "The results of an extensive numerical study of the orbital dynamics of small bodies ranging from micron-sized dust grains to 1 km objects subject to gas drag and also the gravitational attraction of a non-uniform gaseous nebula are presented. The results indicate that it is possible for small bodies to migrate rapidly toward the locations of the maxima of the gas density where the probabilities of collisions and coagulations are enhanced. ",
        "introduction": "It has recently been pointed out that a solar nebula massive enough to form gas-giant planets through the core-accretion model is likely gravitationally unstable (Pollack et al. 1996; Boss 2000; Inaba $\\&$ Wetherill 2002). Such an unstable nebula is not entirely undesirable. The alternative model of the giant planets formation, namely, the disk instability mechanism, suggests rapid formation of gas-giant planets followed by sedimentation of small solids at the locations of spiral arms and clumps of a gravitationally unstable disk. It is, therefore, fundamentally important to study the dynamical evolution of solids in such an environment and in particular, the implications for the collisional coagulation and growth processes.  A turbulence-free rotating gaseous nebula is at hydrostatic equilibrium when the gravitational attraction of its central star is balanced by a radial gradient in its pressure known as the pressure gradient (Figure 1). When the pressure gradient is positive, the velocity of a gas molecule is greater than its local Keplerian velocity (Eq. (1)). A solid in the gas, in this case, feels, effectively, a larger acceleration along its orbit and, consequently, the increase in its orbital angular momentum forces the solid to a larger orbit. The opposite is true when the pressure gradient is negative. \\vskip -6pt \\begin{equation} r{\\omega_{\\rm g}^2}\\,=\\,r{\\omega_{\\rm K}^2}\\,+\\, {1\\over {\\rho_{\\rm g}}}\\,{{d{P_{\\rm g}}}\\over {dr}} \\qquad,\\qquad {\\omega_{\\rm K}^2}\\,=\\,{{GM}\\over {r^3}} \\end{equation} \\vskip -1pt \\noindent In this equation, $M$ is the mass of the central star, $G$ is the gravitational constant and ${P_{\\rm g}},{\\rho_{\\rm g}}$ and $\\omega_{\\rm g}$ represent the pressure, the density and the angular velocity of the gas, respectively.  In a rotating gravitationally unstable disk, gas density enhancements appear in the forms of spiral arms and clumps. It is possible for the pressure of the gas     \\begin{figure} \\vskip -160pt \\plotone{fig1.eps} \\caption{Pressure gradient and the radial migration of solids (Whipple 1972). The velocity of the gas slightly differs from Keplerian circular (Eq.(1)).} \\end{figure} \\noindent to have a radial gradient in the vicinity of such density enhancements. In this case, solids migrate toward the location of the maximum gas density where the probabilities of their collisions and coagulations are enhanced. In this paper, we study the dynamics of solids that undergo such migrations while subject to gas drag and the gravitational attraction of the nebula. ",
        "conclusions": "In general, the rate of the migration of a solid in a turbulence-free gaseous nebula in the presence of gas drag and the gravitational force of the nebula varies with the solid's mass and size and also with the gas density and temperature. The results of this study indicate that it is, indeed, possible for solids within certain ranges of size and density to migrate rapidly to the locations of the maximum values of the gas density. Given the likelihood that the solar nebula was marginally-gravitationally unstable, the processes studied here may have enhanced the growth rates of solid planetesimals.  This work is partially supported by the NASA Origin of the Solar System Program under grant NAG5-10547 and by the NASA Astrobiology Institute under the grant NCC2-1056.      \\begin{figure} \\plotone{fig3.eps} \\vskip -5in \\caption{Rapid migration of 10 cm and 100 cm objects with densities equal to 2 and 5 g cm$^{-3}$ in an isothermal solar nebula at 1000 K.} \\end{figure}    \\vskip 4pt \\centerline{ \\epsfbox{fig4a.eps} \\epsfbox{fig4b.eps}} \\vskip -80pt {\\rightskip 0.3in Figure$\\,$4. \\hskip 13pt Radial migration of a 10 cm-sized object with a density of \\par 2 g cm$^{-3}$ for four different values of the gas temperature.\\par}"
    },
    "0208/astro-ph0208011_arXiv.txt": {
        "abstract": "Bipolar outflows are present in many disk-accreting astrophysical systems. In disk-accreting cataclysmic variables (CVs), these outflows are responsible for most of the strong features in the ultraviolet spectra of these systems. However, there have been few attempts to model these features quantitatively. Here, we describe a new, hybrid Monte Carlo/Sobolev code, which allows us to synthesize the complete spectrum of a disk-dominated, mass-losing CV.  The line profiles we calculate for C IV resemble those calculated by previous workers when an identical geometry is assumed. However, our synthetic spectra exhibit not only the well-known resonance lines of O VI, N V, Si IV and C IV, but, with an appropriate choice of mass-loss rate and wind geometry, also many lines originating from excited lower states. Many of these lines have already been seen in the far ultraviolet spectra of CVs obtained with HUT, FUSE, and HST. In order to illustrate the degree to which we are currently able to reproduce observed spectra, we finally present a preliminary fit to the Hopkins Ultraviolet Telescope spectrum of the dwarf nova Z Cam in outburst. ",
        "introduction": "Non-magnetic cataclysmic variables stars (CVs) are semi-detached binary systems in which a late-type companion star loses mass to a white dwarf (WD) primary via Roche-lobe overflow onto an accretion disk. The discovery of winds in high-state CVs -- from the presence of blue-shifted absorption troughs and P-Cygni profiles in the ultraviolet (UV) resonance lines -- dates back about two decades \\citep{heap1978,cordova1982}. However, most of the fundamental {\\em theoretical} questions concerning the nature and origin of this mass loss are only now beginning to be addressed \\citep{drew2000}.  Recent work suggests that the impact of the outflowing material on {\\em observations} of CVs may have been underappreciated, in that its presence may leave more numerous and subtle observational signatures than just the shapes of the ``classical'' UV wind lines. For example, it has been suggested that a powerful disk wind may be responsible for the peculiar behavior of the SW~Sex stars \\citep{honeycutt1986,dhillon1995}, and that the often single-peaked optical emission line profiles seen in these systems (and also in other nova-like variables [NLs]) are the direct consequence of a wind-induced velocity gradient in the line-emitting material \\citep{murray1996}. Similarly, \\cite{knigge1997} suggested that virtually all of the lines in the UV spectrum of a typical high-state CV, Z Cam, are formed in the outflow, either in the supersonic portion of the wind or in a lower velocity portion of the wind near the interface with the disk photosphere. Based on an analysis of low resolution (1140-8950 \\AA) spectra of the eclipsing novalike variable UX~UMa, \\cite{knigge1998} then suggested that continuum emission from an optically thin disk wind and/or atmosphere provided one possible explanation for the absence of a clear Balmer jump and the flatter than expected  UV continua of high-state CVs \\citep{wade1984,wade1988}. Optical spectra of at least one CV, BZ Cam, shows P-Cygni features in the hydrogen and helium lines that are most easily interpreted in terms of a highly variable wind \\citep{ringwald1998}.  In addition, the EUVE spectrum of the dwarf nova (DN) U~Gem in outburst appears to be dominated by strong wind features to such an extent that any attempt at a detailed spectral analysis {\\em must} account for the outflowing material \\citep{long1996}. Finally, there is also direct evidence that CV winds act as partially ionized absorbers of (soft) X-rays in high-state, non-magnetic systems \\citep{baskill2001}.  The first attempts to interpret UV line profiles in high-state disk-dominated CVs were quite naturally based on wind modeling theory that was being developed to interpret the UV spectra of early-type stars \\citep{mauche1987, drew1987, vitello1988}. These studies, along with a burgeoning amount of observational analysis, indicated the mass loss rates in CV winds are a substantial fraction ($\\sim$10\\%) of the accretion rate, that the acceleration length scale is substantial (20-100 $R_{wd}$), and that the winds are strongly affected by rotation, presumably reflecting an origin in the rapidly rotating inner disk.  The evident importance of rotation led to a second generation of radiative transfer codes that provided for azimuthally symmetric flows. In the CV wind code described in \\cite{SV1993}, hereafter SV93, the wind ionization structure is computed first and the radiative transfer then calculated using the Sobolev (SA) and disconnected approximations. This method is comparatively fast, but does not treat multiple scattering correctly in detail. It was applied by \\cite{vitello1993} to show that, depending on the wind geometry, wind mass loss rates of order 1-10\\% of the accretion rate could account for the C IV line profiles of RW Sex, RW Tri and V Sge . By contrast, the Monte Carlo (MC) code described by \\cite{knigge1995}, hereafter KWD95, simulates the transfer of resonance line photons through an outflow exactly, but does not perform an ionization calculation. It was used by \\cite{knigge1997b} to model high resolution time-resolved spectra of the C IV region in UX~UMa to show that a relatively dense, slowly outflowing transition region between the disk photosphere and the fast wind was necessary to understand the profiles in that system.  Both of these radiative transfer codes have limitations. Neither method solves self-consistently for the thermal balance in the wind (constant wind temperature is assumed); neither is designed to allow for the thermal creation or destruction of line photons in the wind  \\citep[but see][for an approximate treatment in the MC context]{knigge1997}. Finally, both of these codes are intended for modeling of individual spectral lines; as a result it is usually unclear whether a wind model intended to reproduce the C IV profile has any predictive power for other wind signatures.  Here, we present a new, hybrid SA/MC method designed to provide improved spectral modeling of azimuthally symmetric winds. Various aspects of our method build on the work of Lucy and collaborators \\citep{mazzali1993,lucy1999a,lucy1999b} who have to-date considered only spherically symmetric situations. In addition to combining the respective strengths of the SA and MC methods in the (conservative) transfer of resonance lines, our method solves self-consistently for the thermal and ionization balance in the outflow and accounts for the destruction and creation of both line and continuum photons in the wind. As a result, our method can be used to synthesize any part of the emergent, wind-affected spectrum, allowing detailed analyses of previously neglected wind signatures. ",
        "conclusions": "The inherent advantage of a Monte Carlo approach in simulating the spectra of cataclysmic variables is the ease with which one can model complex geometries.  In the Monte Carlo approach, axially symmetric winds are not significantly more difficult to model than spherical winds.  In the case of our code, we have now progressed to the point of spectral verisimilitude, where it is difficult to distinguish a simulation from data.  The wind prescriptions required to produce verisimilitude are not that different from those which were developed to model C IV line profiles.  On the other hand, preliminary comparisons of the models and data suggest that considerable work will be required to reproduce the spectra of some dwarf novae.  Nevertheless, we are already having some successes, such as the model shown here for Z Cam and our model of the FUSE spectrum of SS Cygni (Froning et al. 2001). We are now embarking upon an effort to fully explore the parameter space inherent in the existing models and to compare these models to a variety of HUT, FUSE and HST spectra of high state CVs. This should allow us to determine whether there is a single class of models, broadly or narrowly collimated for example, that can be used to approximate the wind geometry in the majority of CVs."
    },
    "0208/astro-ph0208361_arXiv.txt": {
        "abstract": "{ Via the magnification bias, gravitational lensing by large-scale structures causes angular cross-correlations between distant quasars and foreground galaxies on angular scales of arc minutes and above. We investigate the three-point cross-correlation between quasars and galaxy pairs measurable via the second moment of the galaxy counts around quasars and show that it reaches the level of a few per cent on angular scales near one arc minute. Combining two- and three-point correlations, a skewness parameter can be defined which is shown to be virtually independent on the shape and normalisation of the dark-matter power spectrum. If the galaxy bias is linear and deterministic, the skewness depends on the cosmic matter density parameter $\\Omega_0$ only; otherwise, it can be used to probe the linearity and stochasticity of the bias. We finally estimate the signal-to-noise ratio of a skewness determination and find that around twenty thousand distant quasars e.g.~from the Sloan Digital Sky Survey should suffice for a direct measurement of $\\Omega_0$. ",
        "introduction": "It is widely believed that structures and galaxies in the Universe formed  from gravitational growth of Gaussian primordial mass density  fluctuations dominated by dark matter. Direct support for this picture is provided by the recent weak lensing surveys of galaxies which measured  the systematic distortion of faint background-galaxy images produced by the gravitational tidal field of intervening dark-matter inhomogeneities~: the \\emph{cosmic shear} (Bacon et al. 2000 and 2002; H\\\"ammerle et al.; 2002; Hoekstra et al. 2002; Kaiser et al. 2000; Maoli et al. 2001; R\\'efr\\'egier et al 2002; Rhodes et al. 2001; Van Waerbeke et al. 2000, 2001, 2002; Wittman et al. 2000).   The shape of the cosmic shear signal as a function of angular scale remarkably follows theoretical expectations, which successfully confirms the gravitational instability scenario, even on small scales where non-linear structures dominate the lensing signal.  Further evidence for lensing is provided by the gravitational \\emph{magnification bias}. In addition to distortion, distant objets are magnified or demagnified, depending on whether the matter along their lines-of-sight is over- or underdense compared to the mean. Magnified sources are preferentially included into flux-limited samples, thus sources behind matter overdensities are somewhat over-represented. Since galaxies are biased with respect to the dark-matter distribution, it is expected that this effect induces cross-correlations between distant sources and foreground galaxies.  The existence of significant cross-correlations between distant quasars and foreground galaxies on angular scales of several arc minutes has indeed been firmly established (see Bartelmann \\& Schneider 2001 for a review) and motivated further theoretical development in order to predict how the magnification bias depends on cosmological models. Following earlier work by Bartelmann (1995) and Dolag \\& Bartelmann (1997), M\\'enard \\& Bartelmann (2002) demonstrated the high sensitivity of  angular quasar-galaxy cross-correlation function to several cosmological parameters, namely the matter density parameter, $\\Omega_0$, the normalisation and shape of the dark-matter power spectrum, $\\sigma_8$ and $\\Gamma$, and the bias parameter of the galaxies, $b$. Hence, magnification bias of quasars is equally  efficient as cosmic shear in constraining the geometry and the dark matter power spectrum of the Universe.  However, as with cosmic shear, the information provided by the quasar-galaxy correlation function alone is insufficient for independently constraining all these parameters.  Following similar motivations as Bernardeau, van Waerbeke \\& Mellier (1997) and Jain \\& Seljak (1997), we decided to explore how deviations from Gaussian statistics produced by non-linear growth of structures could modify the galaxy-quasar cross-correlation signal and eventually break some degeneracies between comological parameters. The easiest approach is to focus on the additional information that can be extracted from higher-order correlations between quasars and galaxies which are most sensitive to non-Gaussianity, namely the correlation between distant quasars and foreground galaxy \\emph{pairs}. As for the skewness of the convergence field, we can expect that some parameter dependencies disappear by normalising the three- with two-point correlations.  The paper is structured as follows. We briefly present the formalism of the quasar-galaxy correlation function in Sect.~\\ref{basics}, assuming the paradigm of gravitational instability of a Gaussian random field is valid. In Sect.~\\ref{3-point}, we then introduce the quasar-galaxy-galaxy correlator and predict some useful observational signatures. Section~\\ref{evaluation} deals with density statistics and the numerical evaluation of the triple correlator. We then define a skewness parameter in Sect.~\\ref{mesuring_omega} and demonstrate how it can be used for directly measuring $\\Omega_0$. Similarly, we show in Sect.~\\ref{mesuring_omega} how several properties of the galaxy bias can be constrained. We finally estimate the signal-to-noise ratio of the corresponding observation, and specialise it for the \\emph{Sloan Digital Sky Survey} in Sect.~\\ref{section_sn}. ",
        "conclusions": "Via the magnification bias, gravitational lensing by large-scale structures gives rise to angular cross-correlations between distant sources and foreground galaxies although the two populations are physically uncorrelated. Depending on whether the matter along the lines-of-sight towards these background sources is over- or underdense with respect to the mean, and depending on the value of the slope $\\alpha$ of the cumulative number counts of the sources, magnification effects can cause an excess or a deficit of distant sources near foreground galaxies.  These lensing-induced correlations carry information on the projected dark-matter distribution along the lines-of-sight and can thus provide constraints on cosmology. Considering distant quasars and foreground galaxies, M\\'enard \\& Bartelmann (2002) quantified these constraints and showed that given the large number of parameters involved (the matter density parameter, $\\Omega_0$, the normalisation and shape of the dark-matter power spectrum, $\\sigma_8$ and $\\Gamma$, respectively, and the bias parameter of the galaxies, $b$), the information provided by the quasar-galaxy correlation function alone is insufficient for independently constraining all of these parameters.  In this paper, we have investigated what additional information can be expected from higher-order statistics. Such statistics have similar weightings of the power spectrum along the line-of-sight and can measure non-Gaussianities of the density field due to the non-linear growth of structures. We have specifically considered correlations beween distant quasars and foreground galaxy \\emph{pairs} and showed that this three-point correlator can be related to the excess scatter of galaxies around quasars with respect to random positions, which is a straightforwardly measurable quantity. Using the assumptions that \\begin{itemize} \\item galaxies are linearly biased with respect to the underlying dark matter, \\item the dark-matter distribution is described by a CDM power spectrum, \\item and the lensing magnifications can be approximated by $\\mu=1+2\\kappa$ in the weak lensing regime, \\end{itemize} we have computed the amplitude of the expected excess scatter of galaxies in the vicinity of quasars. For a flat cosmology with $\\Omega_0=0.3$ and $\\sigma_8=0.93$, we find the amplitude of this effect to reach the per cent level at angular scales near one arc minute.  We further showed that combining second- and third-order statistics allows a pseudo-skewness parameter $S_3'$ to be defined which turns out to be weakly sensitive to the normalisation and the shape of the power spectrum. Moreover, if the linear biasing scheme is valid, this parameter is only sensitive to the matter density $\\Omega_0$; the dependences on the other parameters ($\\sigma_8$, $\\Gamma$, $\\Lambda$ and $\\bar b$) are weak or cancel out completely. Thus the skewness $S_3'$ provides a direct and independent measurement of $\\Omega_0$.  We computed the expected angular variation of $S_3'$ and showed that for a given cosmology and assuming a CDM power spectrum any departure from the predicted angular shape must be due to a non-linear and/or stochastic behaviour of the galaxy bias, on scales larger than a few arc minutes.  Finally, we estimated the signal-to-noise ratio of the expected excess scatter of galaxies near quasars, which is the main source of noise in $S_3'$. We derived the noise coming from the finite sampling of galaxies having realistic distributions. Applying our result to the Sloan Digital Sky Survey, we find that $S_3'$ should be measurable on large scales from that survey with about twenty thousand distant and bright quasars, allowing thus a direct and independent measurement of $\\Omega_0$.  On smaller angular scales less quasars are required and the parameter $S_3'$ can probe the angular range and the corresponding physical scales where the linear relation between dark matter and galaxy fluctuations may break down."
    },
    "0208/astro-ph0208469_arXiv.txt": {
        "abstract": "We observed the ROSAT-selected eclipsing dwarf nova GY Cnc (=RX J0909.8+1849) during the 2001 November outburst.  We refined the orbital period to be 0.17544251(5) d.  The fading portion of the outburst was indistinguishable from those of typical dwarf novae with similar orbital periods.  However, the signature of orbital humps (or a hot spot) was far less prominently observed in the orbital light curves and eclipse profiles than in usual dwarf novae with similar orbital periods.  The combination of low frequency of outbursts and the apparent lack of slowly rising, long outbursts in GY Cnc is difficult to reconcile within the standard framework of dwarf novae. We suspect that GY Cnc may be the first above-the-gap counterpart of unusual eclipsing dwarf novae HT Cas and IR Com. ",
        "introduction": "Cataclysmic variables (CVs) are close binary systems consisting of a white dwarf and a red-dwarf secondary transferring matter via Roche-lobe overflow.  Dwarf novae are a class of CVs showing outbursts, which are believed to be a result of instabilities in the accretion disk [see \\citet{osa96review} for a review].  Depending on orbital inclination, some dwarf novae show a various degree of eclipses.  Eclipses in such dwarf novae provide a powerful tool in studying the time-variation of the structure of accretion disks (e.g. \\cite{EclipseMapping}; \\cite{woo89ippeg}).  GY Cnc (=RX J0909.8+1849 = HS 0907+1902) is a CV identified in the course of the Hamburg/RASS identifications of ROSAT sources \\citep{bad98RASSID}.  The dwarf nova nature was suspected upon the recognition of an apparent outburst on Guide Star Catalog (GSC) \\citep{kat00gycnc}.  During the 2000 February outburst, several observers independently discovered that the object shows deep eclipses (\\cite{kat00gycnc} and references therein; \\cite{gan00gycnc}). \\citet{tho00gycnc} further studied the object spectroscopically, and obtained component masses.  \\citet{tho00gycnc} reported that a modeling with a flat, Keplerian disk did not yield a good fit to the observed profile of the quiescent H$\\alpha$ emission line. \\citet{sha00gycnc} studied the object during post-outburst quiescence, and derived orbital parameters.  \\citet{sha00gycnc} found that the bright spot (hot spot) is less conspicuous in GY Cnc compared to IP Peg, the eclipsing dwarf nova having similar orbital parameters. These observations suggest a some degree of peculiarity in GY Cnc compared to other dwarf novae above the period gap.  In spite of independent eclipse detections, the observation of the 2000 February outburst turned out to be rather fragmentary.  We present first-ever complete photometric coverage of the declining branch of the 2001 November outburst. ",
        "conclusions": "We observed the ROSAT-selected eclipsing dwarf nova GY Cnc (=RX J0909.8+1849) during the 2001 November outburst.  We refined the orbital period to be 0.17544251(5) d.  The fading portion of the outburst was indistinguishable from those of typical dwarf novae with similar orbital periods.  However, the orbital light curves and eclipse profiles show a lesser signature of orbital humps (or a hot spot) than in usual dwarf novae with similar orbital periods. We suspect that GY Cnc may be the first above-the-gap counterpart of unusual eclipsing dwarf novae HT Cas and IR Com.  \\vskip 3mm  We are grateful to many VSNET observers who have reported vital observations. This work is partly supported by a grant-in aid (13640239) from the Japanese Ministry of Education, Culture, Sports, Science and Technology. Part of this work is supported by a Research Fellowship of the Japan Society for the Promotion of Science for Young Scientists (MU)."
    },
    "0208/astro-ph0208143_arXiv.txt": {
        "abstract": "Observations of the Galactic ISM have had tremendous impact on our understanding of the physics of galactic gas and the processes of galaxy formation.  Similar observations at $z>2$ reveal the neutral baryonic content of the universe, trace the evolution of metal enrichment, shed light on process of nucleosynthesis and dust formation, and yield precise measurements of galactic velocity fields. Owing to the limitations of UV spectroscopy, however, researchers are unable to examine galactic gas at $0 < z < 2$, an epoch spanning $\\approx 80\\%$ of the current universe. To complement the multitude of ongoing programs to identify and research $z<2$ galaxies, a next generation space telescope is essential to investigate the gas which feeds and records the history of galaxy formation. ",
        "introduction": "In the next decade, numerous observational programs will identify and investigate overwhelming numbers of $z<2$ galaxies. These surveys will characterize the history of star formation, the evolution of galaxy morphology, and the assembly of large-scale structure.  Altogether, these efforts are likely to revolutionize our view of stars and galaxies at $z<2$.  In contrast to these achievements, these observations will have minimal impact on our understanding of the gas which feeds star formation. At all redshift, gas is the major baryonic component of the universe and, at $z>1$, probably the dominant baryonic component of most galaxies. The principal challenge associated with examining the physics (e.g.\\ metallicity, density, ionization state, temperature) of this gas is that the majority of diagnostics lie within the ultraviolet (UV) pass-band. To cover this epoch, one requires a next generation space telescope.  With the advent of echelle spectrographs on 10m-class optical telescopes, researchers have pursued quasar absorption line (QAL) studies at unprecedented levels in the early universe.  These observations have revolutionized our understanding of the \\lya forest (e.g.\\ Miralda-Escud\\'e et al.\\ 1996; Rauch 1998), measured the baryonic density of the universe (Burles \\& Tytler 1998; Rauch et al.\\ 1997), revealed metals in among the least overdense structures observed (Tytler et al.\\ 1995; Ellison et al.\\ 2000), provided new insight into the chemical enrichment history of the universe (Pettini et al.\\ 1997; Prochaska \\& Wolfe 2000), and traced the velocity fields of protogalaxies (Prochaska \\& Wolfe 1997). These studies have tremendous impact on our theoretical description of gas in the early universe.  For example, aside from the CMB, comparisons of CDM predictions with the \\lya forest stand as one of the greatest successes of this cosmological paradigm.  Furthermore, the majority of theorists at this workshop stressed one particular point: advances in 'gastrophysics' are essential to addressing the next theoretical frontier. Empirical constraints on 'gastrophysics' can only be made through observations of gas and QAL observations provide the most efficient avenue of investigation.  In contrast with the $z=2-5$ universe, there are very few diagnostics of the IGM or extragalactic ISM at $z<2$.  This antithesis is easily explained: the majority of physical diagnostics have observed wavelengths below 3000\\AA\\ at $z<1.5$. In particular, the \\lya transition (at 1215\\AA, the reddest H\\,I resonance line) can only be observed at $z>1.6$ with optical spectrographs.  Without coverage of this key transition, one cannot begin to address the preceding list of scientific inquiry.  An extragalactic observer like myself might be prone to nonchalantly dismiss the $z<2$ universe.  After all, this represents only 25$\\%$ of the redshift path accessible to QAL analysis. This perspective, however, is horribly skewed. The epoch spanning $z=[0,1.5]$ encompasses roughly 80$\\%$ of the current age of the universe.  If one considers the temporal evolution of any quantity (e.g.\\ chemical enrichment, galaxy clustering, luminosity functions), then to overlook this epoch is to remain ignorant of the universe.  In this brief proceeding -- written in support of a next generation UV telescope -- I will emphasize the preceding introduction through a review of one major area of QAL research: the damped \\lya systems (DLA). These QAL systems are defined to have an H\\,I surface density in excess of $10^{20.3} \\cm{-2}$ and they dominate the universal H\\,I content at all epochs following reionization.  At $z>2$, the DLA are believed to be the progenitors of present-day galaxies (Kauffmann 1996; Steinmetz 2002) and echelle observations of the sightlines penetrating these galaxies provide detailed physical measurements of the protogalactic ISM. The measurements include chemical enrichment level, star formation rate, dust content, nucleosynthetic enrichment history, as well as clues to the pressure, temperature, and ionization balance of this multi-phase medium. These observations are directly analogous to observations of the Galaxy, LMC, and SMC carried out by HST and past UV telescopes.  This is the central thesis of this proceeding: current observations reveal the physics of galactic gas locally and at $z>2$ but very rarely address the $\\approx 10$~Gyr in between.  To probe what amounts to $80\\%$ of the universe, one needs a next generation space telescope. In the following, I will highlight some of the major results of research on damped \\lya systems, primarily those related to my own research with optical echelle spectrographs.  To pursue these same research areas at $z<2$ will require similar instrumentation in the UV pass-band. ",
        "conclusions": ""
    },
    "0208/astro-ph0208375_arXiv.txt": {
        "abstract": "We report Sr, Pd and Ag abundances for a sample of metal-poor field giants and analyze a larger sample of Y, Zr, and Ba abundances.  The $\\lbrack$Y/Zr$\\rbrack$ and $\\lbrack$Pd/Ag$\\rbrack$ abundance ratios are similar to those measured for the r-process-rich stars CS 22892-052 and CS 31082-001. The $\\lbrack$Pd/Ag$\\rbrack$ ratio is larger than predicted from the solar-system r-process abundances.  The constant $\\lbrack$Y/Zr$\\rbrack$ and $\\lbrack$Sr/Y$\\rbrack$ values in the field stars places strong limits on the contributions of the weak s-process and the main s-process to the light neutron-capture elements.  Stars in the globular cluster M 15 possess lower $\\lbrack$Y/Zr$\\rbrack$ values than the field stars.  There is a large dispersion in $\\lbrack$Y/Ba$\\rbrack$.  Because the r-process is responsible for the production of the heavy elements in the early Galaxy, these dispersions require varying light-to-heavy ratios in r-process yields. ",
        "introduction": "Burbidge \\etal{} (1957) and Cameron (1957) showed that only two sets of physical conditions were necessary to explain the abundances of the heavy (A $>$ 65) elements in the solar system. Because of strong Coulomb forces, the build-up of heavier nuclei happens through neutron-capture. The first of these neutron-capture processes is the s-process, where neutron captures onto seed nuclei take place much more slowly than $\\beta -$decays.  The s-process is thought to take place in two distinct environments.  The `main' s-process occurs in low-mass AGB stars, while the `weak' s-process occurs during helium burning in massive stars. While the main s-process can make neutron-rich material up to $^{209}$Bi (Clayton \\& Rassbach 1967), the weak s-process is not predicted to make significant amounts of material with A$> 90$ (Couch, Schmiedekamp, \\& Arnett 1974).  The second neutron-capture process, the r-process, takes place when conditions are such that neutron capture rates are much higher than $\\beta-$decay rates. It produces a distinctive pattern in the abundance ratios, the most noticeable features being the so-called r-process peaks. These peaks, at A$\\sim$80, 130, and 196, are the signatures of nucleosynthesis events which reached the neutron magic numbers of 50, 82, and 126.  Despite having been identified as taking place in an environment with rapid neutron captures, the astrophysical phenomena that create the r-process remain unidentified.  The neutrino wind in Type II SN showed promise (Woosley \\& Hoffman 1992; Woosley \\etal{} 1994), but two problems arose. First, there remain questions about whether the entropy in the wind is sufficiently high to produce the r-process (Takahashi, Witti, \\& Janka 1994; Qian \\& Woosley 1996, but see Otsuki \\etal{} 2000; Wanajo \\etal{} 2001).  Also, Freiburghaus \\etal{} (1999a) found that high-entropy wind models cannot produce an r-process pattern for A$<$110 because those elements are synthesized during the low-entropy, neutron-deficient $\\alpha$-rich freezeout phase of the wind. Witti, Janka, \\& Takahashi (1994) and Woosley \\etal{} (1994) also show that the elements near N=50 are overproduced relative to the more massive nuclei in neutrino wind models. So either the neutrino-wind is not the source of any r-process products, or the conditions are such that the material with A$<$110 is either not ejected or is diluted. Models have shown that merging neutron stars may be the source of significant amounts of r-process material (e.g. Lattimer \\& Schramm 1974; Rosswog \\etal{} 2000).  Freiburghaus, Rosswog \\& Thielemann (1999b) did parameterized calculations of the nucleosynthesis in the ejecta and found that it could be a source for nuclei with A\\gtsima130.  An earlier suggestion that the r-process occurred in helium-burning regions, either cores of low-mass stars or in the helium shell in SN, was rejected because unacceptably large amounts of $^{13}$C were required to make the A$\\sim$195 peak (Cowan, Cameron \\& Truran 1985).  However, helium-burning phases can still make r-process isotopes in the range A$\\sim$80 with about half as much $^{13}$C. Truran, Cowan, \\& Fields (2001) updated the calculations for the helium-shell shock r-process and find that for normal amounts of C, it can provide interesting amounts of material with A $<$ 130.  Abundances in metal-poor stars provide insight into the r-process because they bear the marks of relatively few nucleosynthesis events and because the r-process is thought to be the sole source of many of the heavy elements in the early Galaxy (Truran 1981). However, alternative sources, such as the weak s-process, may contribute substantial amounts to the lightest of the neutron-capture elements, such as Sr, Y and Zr (e.g. Prantzos, Hashimoto \\& Nomoto 1990).  In recent years, much new information has become available on the abundances of heavy elements in metal-poor stars (e.g. Gilroy \\etal{} 1988; McWilliam \\etal{} 1995; Ryan \\etal{} 1996; Westin \\etal{} 2000; Burris \\etal{} 2000), revealing a wide diversity in some abundance ratios, such as [Sr/Ba] and a remarkable consistency in others, such as [Ba/Eu]. The information on the intermediate-mass elements such as Pd and Ag is particularly interesting, since the abundance ratios in one star, CS 22892-052, showed a larger difference between the abundances of the odd-Z and even-Z elements than seen in the solar system (Sneden \\etal{} 2000a).  Johnson (2002) (Paper I) presented the abundances of up to 17 neutron-capture elements in a sample of 22 metal-poor ([Fe/H] $< -1.7$) field red giants. In this paper we add Sr abundances for these stars, as well as Pd and Ag abundances for three stars from this sample. We then analyze the abundance patterns for all the heavy elements in the Paper I sample to learn more about the r-process in the early Galaxy. ",
        "conclusions": "Abundance ratios of neutron-capture elements in metal-poor stars show both striking similarities and large dispersions. There is real scatter in the [Y/Ba] ratios observed and differences between metal-poor stars and \\rss{} in the [Pd/Ag] ratio. The Sr-Y-Zr and the Ba$-$Yb regions show similar abundance patterns in all stars in the Paper I sample, with some deviation in the lighter elements in the red giants from M 15. The M 15 giants provide the strongest evidence that the light neutron-capture elements in metal-poor stars can be produced in the same events as the heavy neutron-capture elements, since stars with very different [Ba/Fe] have identical [Y/Ba]. Similar evidence is provided by CS 22892-052 and CS 31082-001, which have large enhancements in both Y and Ba.  Theoretical results show that the main s-process cannot produce substantial amounts of the neutron-capture elements in low metallicity stars. This result is supported by the good agreement between abundance pattern in the Ba region in our sample of stars and \\rss. The weak s-process potentially could contribute to elements with A$<90$. However, the constant abundance ratios of [Sr/Y] and [Y/Zr] precludes substantial contributions from processes other than the r-process in the early Galaxy. As a result, the abundances of the neutron-capture elements in metal-poor stars provide strong constraints on the r-process. Any model of the r-process must explain the scatter seen in [Y/Ba] and [Ba/Th]. In addition, the models need to reproduce the enhanced odd-even effect in the Pd-Ag region. The variety of phenomena proposed for the r-process shows that the r-process production need not be confined to one kind of event, which could aid in describing the dispersion seen."
    },
    "0208/astro-ph0208005_arXiv.txt": {
        "abstract": "We explore the cosmological implications of 7 deep survey fields observed at the Berkeley-Illinois-Maryland Association (BIMA) Array with 30 GHz receivers. These observations probe the Cosmic Microwave Background anisotropy on scales corresponding to $l\\sim 5500$, and an earlier analysis of these data detected no galaxy clusters. We use numerical cluster simulations and mock observations to characterize the cluster detection efficiency for each of the BIMA fields.  With these detection efficiencies we derive constraints on the cosmological parmaters $\\Omega_M$ and $\\sigma_8$, ruling out those models which overproduce galaxy clusters.  Using only these seven BIMA fields, we calculate a 2$\\sigma$ upper limit of $\\sigma_8 < 1.00\\, \\Omega_M^{-0.43 \\Omega_M-0.22}$ for flat models with $0.1\\le\\Omega_{M}\\le1$.  When the power spectrum of density fluctuations is COBE normalized, we find $\\Omega_M < 0.63$ at $95\\%$ confidence level for flat models. This constraint includes our estimate of the large uncertainties in the SZE flux--virial mass relationship as well as published uncertainties in the Hubble parameter, the COBE power spectrum normalization and the primordial power spectrum index.  In addition, we account for the effects of sample variance.  Thus, we conclude that the non--detections are to be expected in a low $\\Omega_M$ universe, given the sensitivity and solid angle of the deepest SZE survey to date. ",
        "introduction": "Cosmic microwave background  (CMB) images provide information about the presence of galaxy clusters over a wide range of redshift, because galaxy clusters leave signatures in the CMB through the so--called Sunyaev-Zel'dovich effect \\citep[SZE;][]{sunyaev70,sunyaev72}. The SZE is produced by inverse Compton interactions between the CMB photons and the hot electrons ($k_B T_e \\sim 1$\\,keV, see \\S\\ref{sec:sze}) in the intracluster medium. The intracluster medium is shocked and heated in the process of cluster formation. The net result in the SZE is a transfer of energy from the electron population to the CMB photons, leading to a distortion in the CMB spectrum. The thermal SZE signature of a cluster in 1~cm ($\\sim$ 30~GHz) observations is a reduction in the CMB brightness.  On angular scales of $\\sim 5^{'}$ and smaller the thermal SZE contribution to the CMB anisotropy can dominate that of the primary CMB anisotropy \\citep{holder99,holzapfel00a,hu01,springel01}.  Here we discuss the cosmological implications of 7 deep survey fields observed at the Berkeley--Illinois--Maryland Association (BIMA) array with the 30~GHz receivers \\citep{carlstrom96,holzapfel00a}. The observations were made during 1997 and 1998, in a compact configuration at 28.5\\,GHz, providing a Gaussian primary beam with FWHM$\\sim 6'_.6$. This survey, originally planned to probe the CMB anisotropy on arcminute scales $l\\sim 5500$ (corresponding to angular size $\\sim 2'$), provides relatively large sky coverage ($\\sim 250$ arcminute$^2$) for an arcminute-scale anisotropy experiment.  These observations have been used to place an upper limit on small scale CMB anisotropy \\citep{holzapfel00a}. Continued observations on these and additional fields has led to a detection of anisotropy with flat band power at the level $Q_{flat}=6.1^{+2.8}_{-4.8}\\mu K$, where uncertainties describe 68\\% confidence regions \\citep[][see also \\citealt{dawson02}]{dawson01}. In addition, no galaxy cluster detections were reported, which is interesting because of the sensitivity of SZE surveys to high redshift clusters.  In the currently favored cosmology, these seven fields sweep out as much comoving volume to $z\\sim2$ as a 130~deg$^2$ local survey sweeps out to $z\\sim0.1$.  In principle, cluster surveys constrain cosmological parameters through the effects those parameters have on the volume surveyed per solid angle and the evolution of cluster abundance \\citep[e.g.][]{haiman01,holder01b,mohr01}.  In this analysis we focus on the lack of clusters in these deep '97 and '98 BIMA fields.  A less sensitive, earlier survey with the Australian Telescope Compact Array also provided upper limits on the CMB anisotropy on similarly small scales \\citep{subrahmanyan00}. An analysis of the cosmological implications of that survey, which relied on the assumption that the dominant source of anisotropy on arcminute scales is the thermal SZE from clusters, concluded that a low-$\\Omega_M$ universe is preferred \\citep{majumdar00}.   An analysis of this BIMA survey allows us to rule out cosmological models which produce too many clusters.  The expected number of detected clusters in a survey is \\begin{equation} \\langle N \\rangle  = \\int d\\Omega \\int dV \\int_0^{\\infty} dM\\,f(M)\\,\\frac{dn}{dM}\\left(M,z\\right), \\label{eq:expectation} \\end{equation} where $(dn/dM)dM$ is the comoving number density of objects of mass between $M$ and $M+dM$, $f(M)$ is the detection efficiency ($0\\le f(M)\\le 1$), which takes into account the fact that only clusters with masses above some mass limits can be detected. The first integral is over the solid angle of the survey, the second integral is over the redshift, and the third integral sums over the portion of the cluster mass function to which our survey is sensitive. In addition to its dependence on mass, the detection efficiency $f(M)$ depends on the sky position $(\\alpha,\\delta)$, the redshift $z$ and cosmological parameters such as $H_0$, $\\Omega_M$ and $\\Omega_\\Lambda$.   Cosmological constraints come through comparison of the observed and the expected number of clusters.  In the BIMA survey no clusters were detected; assuming Poisson statistics, the probability $P$ of observing $0$ clusters is $ P(0|\\langle N\\rangle) = e^{-\\langle N\\rangle}$.  Using this approach we explore the parameter space spanned by $\\Omega_M$ and $\\sigma_8$, checking the consistency of these cosmological models.  We also consider the uncertainties in the cluster detection efficiency $f(M)$, Hubble parameter, COBE power spectrum normalization, power spectrum index and the effects of sample variance.  This paper is organized as follows: in \\S\\ref{sec:masssens} we present a general description of interferometric SZE observations, describe the BIMA survey data, and determine the survey detection efficiency via mock observations of hydrodynamic cluster simulations. We discuss the ingredients needed for the calculations of cluster abundance in \\S\\ref{sec:yields}. The cosmological constraints and a discussion on effects of systematic errors appear in \\S\\ref{sec:res}. We present a summary and discussion on prospects of future interferometry SZE observations in \\S\\ref{sec:summary}. We discuss the $H_0$ scaling of interferometric SZE observations in the Appendix. ",
        "conclusions": "\\label{sec:summary}  We describe a systematic analysis of an interferometric SZE survey carried out at BIMA with the Carlstrom--Joy 30~GHz receivers. The data were taken from observations of the CMB anisotropy on arcminute scale ($l \\sim 5500$), in which no anisotropy and no galaxy clusters were detected \\citep{holzapfel00a,holzapfel00b}.   We use these data to study the allowed range of cosmological parameters $\\sigma_8$ and $\\Omega_M$. In our analysis, we consider models that are flat with a cosmological constant ($\\Omega_M+\\Omega_\\Lambda=1$) and models that are open with no cosmological constant.  Our analysis hinges on our ability to calibrate the survey sensitivity or galaxy cluster selection function.  To this end, we use mock interferometric observations of hydrodynamical cluster simulations to determine the selection function for each of the seven fields and as a function of redshift and position within that field.  Relying on hydrodynamical simulations to calibrate the survey sensitivity introduces significant uncertainties, which can be reduced once empirical calibration using large surveys that include detailed followup observations are available.  For this analysis we estimate the scale of the uncertainties that come with using hydrodynamical cluster simulations and include those uncertainties in determining our constraints on the cosmological parameter \\Ome.  With BIMA survey data alone, we obtain a 95\\% confidence upper limit on $\\sigma_8(\\Omega_M)$: $\\sigma_8 < 1.00\\, \\Omega_M^{-0.43 \\Omega_M-0.22}$ for flat, and $\\sigma_8 < 1.01\\, \\Omega_M^{-0.40 \\Omega_M-0.14}$ for open models. Combining the BIMA survey with external constraints on the power spectrum allows us to constrain \\Om alone.  In combination with the local abundance of high temperature X--ray clusters, the BIMA data lead only to a trivially weak lower limit: $\\Omega_M>\\Omega_B$. In combination with the COBE normalization of the power spectrum, the data provide an upper limit on $\\Omega_M$: $\\Omega_M<0.63$ for flat models, and $\\Omega_M<0.73$ for open models at 95\\% confidence.   Included in the constraints on \\Ome are 40\\% uncertainties in the cluster visibility--mass relation as well as published uncertainties on the Hubble parameter and power spectrum normalization $\\delta_H$ and spectral index $n$.  We also include the effects of sample variance.  These constraints are valid only if these BIMA fields represent an unbiased sample of the universe.   The field selection in this survey is somewhat suspect, with four of the seven fields specifically chosen to exclude bright radio, X--ray and optical point sources.  A firm interpretation of our results is that -- if the fields are an unbiased clusters survey -- we have shown that within the context of the cosmological models favored today, we expect no clusters to have been detected.  If the survey was biased against finding clusters, then our constraints on \\Om weaken, and it is even less surprising that no clusters were detected.  Of course, if the survey were biased toward finding clusters, then finding none would imply tighter constraints on \\Ome.  We eagerly await planned SZE surveys of larger and perhaps better chosen fractions of the universe.  In previous work, \\citet{majumdar00} studied constraints on $\\Omega_M$ by comparing simulated SZE maps with the upper limit on the arcminute--scale anisotropy in CMB obtained by \\citet{subrahmanyan00} using the Australia Telescope Compact Array (ATCA) at $8.7$ GHz. Their 95\\% confidence upper limit is $\\Omega_M<0.8$ for open models. This result is clearly consistent with ours, but there are important differences in the analysis and the data.  First, the ATCA survey has lower sensitivity than the BIMA survey, yielding a 95\\% confidence limit of $Q_{flat} < 25\\,\\mu$K compared to the BIMA survey constraints of $Q_{flat} < 14.1\\,\\mu$K.  Second, their focus was on reproducing the statistical properties of the noise in the sky maps, whereas we have on the non--detection of clusters. Third, our modeling technique is more sophisticated than theirs through its use of hydrodynamical cluster simulations that incorporate the effects of cluster morphology and internal structure and the evolution of these properties with redshift. And fourth, we model the redshift evolution of the cluster abundance using the results from the latest N--body studies of structure formation \\citep{jenkins01}, whereas they modeled cluster abundance using the original Press-Schechter formalism.  We believe that the techniques we outline in this manuscript provide a viable approach for analyzing the forthcoming interferometric SZE surveys that will use much higher sensitivity instruments to survey larger portions of the sky.  In short, the principle improvement of our analysis over previously existing analyses and estimates of survey yields include (1) the direct accounting for the uncertainty in the cluster detectability (specifically, we model the scatter in the detection significance--virial mass relation and we include the uncertainty in the normalization of the SZE visibility--virial mass relation), (2) the modeling of the survey sensitivity as a function of position on the sky, (3) the inclusion of uncertainties on other cosmological parameters of interest, and (4) an accounting for the effects of sample variance on our \\Om constraint.  In addition, we have examined the $h$ dependence of the survey limiting mass (see Appendix), showing that for this interferometric survey, which includes sensitivity to cluster shape {\\it and} flux, the $h$ dependence departs from the flux limited survey expectation $M_{lim}\\propto h^{-8/5}$ to a steeper dependence of $M_{lim}\\propto h^{-11/6}$.  Similar departures would be expected for other interferometric surveys.  Large solid angle, high redshift SZE cluster surveys can in principle provide a wealth of information on structure evolution and cosmology. Because of its redshift--independent nature, the SZE is a powerful tool in studies of the high redshift universe. An interferometric SZE survey has advantages such as superb angular resolution and the ability to separate point source contamination from the extended clusters \\citep{holder00,kneissl01}. Future SZE surveys will have much higher sensitivity than the BIMA survey analyzed here, and will be carried out with the Sunyaev-Zel'dovich effect Array \\citep[SZA,][] {holder01b}, the Arcminute Microkelvin Imager \\citep[AMI,][] {kneissl01}, and the Arcminute Mm-wave interferometer for Background Anisotropy \\citep[AMiBA,][]{lo00}.   These surveys are expected to detect hundreds of clusters, and these large, high redshift cluster samples will be invaluable to cosmological studies, enabling tests of the mix of dark energy and dark matter in our universe and measurements of the nature of this dark energy\\citep[e.g.][]{barbosa96,haiman01,holder01b,fan01,mohr01,weller01,hu02}."
    },
    "0208/astro-ph0208233_arXiv.txt": {
        "abstract": "SuperWASP is a fully robotic, ultra-wide angle survey for planetary transits. Currently under construction, it will consist of 5 cameras, each monitoring a $9.5^{\\circ} \\times 9.5^{\\circ}$ field of view.  The {\\em Torus} mount and enclosure will be fully automated and linked to a built-in weather station. We aim to begin observations at the beginning of 2003. ",
        "introduction": "The observations of the planetary transits of HD~209458 by Charbonneau et al. (2000) and Henry et al. (2000) highlighted the role which can be played by small aperture/ultra-wide field surveys in detecting transits by hot Jupiter-type planets.  These surveys have the great advantage of requiring only relatively inexpensive, off-the-shelf equipment, which is then dedicated to the project.  The wide field of view allows many thousands of $\\sim$7--13\\,mag stars to be photometrically monitored simultaneously.  As radial velocity surveys indicate $\\sim1$\\% of these stars have hot Jupiters, we should, in theory, be able to discover statistically significant numbers of these planets within a reasonably short timescale.  Such a sample is necessary to answer questions about the formation and evolution of these planets, and how it is related to factors such as stellar metallicity, age, type, etc.  The magnitude range matches that of the radial velocity programs, allowing detailed follow-up observations.  The {\\em SuperWASP} project has developed from our experience in building and operating the prototype {\\em WASP0}.  This instrument is the subject of separate papers by Kane et al. (this volume) and Street et al. (2000).  Having proven that we can achieve the required high-precision photometry from this single, manually-operated camera, we are now building a fully robotic, multi-camera instrument supported by a custom-built mount.  {\\em SuperWASP}, which is primarily funded by Queen's University, will initially be able to monitor 5 separate fields simultaneously, with the potential for up to 10. This will provide precise photometry on $\\sim$25,000 -- 50,000 stars at a time, data which will form an important resource for bright star astronomy.  With this in mind, we will be making use of the data to search for a number of phenomena, including Near Earth Asteroids and optical transients in addition to the primary goal of planet-hunting.  Here we present our science goals followed by the equipment design, a discussion of the data we expect to gather and our plans for its analysis and dissemination. ",
        "conclusions": "\\protect\\label{sec:concs}  We present our plans for a multi-camera, robotic search for planetary transits.  {\\em SuperWASP} will have a greater field of view than other similar projects, and a higher expected yield of planetary transits.  It will provide a very large catalogue of densely sampled lightcurves of hundreds of thousands of stars which could result in more than 100 planets being detected per year.  We aim to make this data publically available.  {\\em SuperWASP} is currently under construction, and is on schedule to begin observations from La Palma in early 2003."
    },
    "0208/astro-ph0208555_arXiv.txt": {
        "abstract": "Convection-dominated accretion flows (CDAF) are expected to have a shallower density profile and a higher radiation efficiency as compared to advection-dominated accretion flows (ADAF). Both solutions have been developed to account for the observed properties of the low luminosity, high temperature X-ray sources believed to involve accretion onto massive black holes. Self-similar CDAFs also have steeper poloidal density gradients and temperatures close to the virial temperature. All these characteristics make CDAFs more capable of producing polar outflows driven by Compton heating as compared to other classical accretion disks. We investigate the conditions for producing such outflows in CDAFs and look for the mass accretion rate, or, equally, the luminosity of CDAFs for which such outflows will exist. When the electron temperature saturates around $10^{11}\\K$ at the inner region, polar outflows are probable for $8\\times10^{-7} \\lesssim L/L_E \\lesssim 4\\times10^{-5}$, where \\(L_E\\) is the Eddington luminosity. Outflows are well collimated with small opening angles. The luminosity range for which outflow solutions exist is narrower for lower electron temperature flows and disappears completely for electron temperature $\\lesssim 6\\times 10^9\\K$. When the magnetic field is present, we find that outflows are possible if the magnetic field is less than from 10\\% to 1\\% of the equipartition field. We also find that outflows are more likely to be produced when the viscosity parameter \\(\\alpha\\) is small. The tendency for jet-like collimated outflows for these solutions is presumably astrophysically relevant given the high frequency of jets from AGNs. ",
        "introduction": "Advection-dominated accretion flows (ADAF) nicely complement the classic thin disk accretion flows (Shakura \\& Sunyaev 1973; Ichimaru 1977; Rees et al. 1982; Narayan \\& Yi 1994, 1995; Abramowicz et al. 1995), and have been successfully applied to variety of cosmic objects, from galactic X-ray binaries to diffuse X-ray background (see Narayan, Mahadevan, \\& Quataert 1999 for review). However, analytic studies of ADAFs indicated the convectively unstable nature of ADAFs (Narayan \\& Yi 1994, 1995a; see Begelman \\& Meier 1982 for radiation-dominated ADAF), which has subsequently been proved in a series of numerical simulations (Igumenshchev, Chen, \\& Abramowicz 1996; Igumenshchev \\& Abramowicz 1999, 2000; Stone, Pringle, \\& Begelman 1999; Igumenshchev, Abramowicz, \\& Narayan 2000). Especially, the numerical studies of ADAFs by Igumenshchev \\& Abramowicz (1999, 2000) show that an ADAF becomes a convection-dominated accretion flow (CDAF) whenever the viscosity parameter $\\alpha \\lesssim 0.1$. Further analyses of self-similar CDAF solutions show their unique properties (Narayan, Igumenshchev \\& Abramowicz 2000, hereafter NIA; Quataert \\& Gruzinov 2000, hereafter QG): the density varies as $\\rho \\propto R^{-1/2}$ ($R$ is the radius), much flatter than the usual $R^{-3/2}$ in an ADAF or in spherical accretion. Correspondingly, the mean radial velocity varies as $v \\propto R^{-3/2}$, compared to $R^{-1/2}$ in ADAF or in spherical flow. Energy generated at the inner part of the flow is transported to the outer part by convection. A CDAF has perhaps as much resemblance to the rotating stellar envelope of a massive star as to the usual accretion flow.  The other aspect of the two-dimensional nature of ADAF or CDAF solutions, often neglected, is the interaction between the outgoing radiation produced at smaller radii with the inflowing gas in the outer part of the flow. In optically thick stars this interaction plays a vital role in establishing the equilibrium states. The radiative interaction also plays a very important role in pure spherical accretion flows (Ostriker et al. 1976; Cowie, Ostriker, \\& Stark 1978; Wandel, Yahil, \\& Milgrom 1984; Park 1990a, 1990b; Nobili, Turolla, \\& Zampieri 1991; Zampieri, Miller, \\& Turolla 1996; Ciotti \\& Ostriker 1997, 2001). Park \\& Ostriker (1999, 2001) studied the same radiative interaction in the context of the ADAF solution and found that a polar outflow can be generated through Compton heating of electrons by high-energy photons emitted by the inner, hot part of the flow. The winds generated by the processes in the papers listed above are not momentum driven. Rather, they are caused by overheating of the gas in the slowly moving, low density polar regions. Outflows may also be generated from ADAF by other hydrodynamic processes (Narayan \\& Yi 1995; Xu \\& Chen 1997; Blandford \\& Begelman 1999).  In this work, we study the conditions for CDAFs to develop radiation driven outflows. We adopt the self-similar CDAF solution as the background flow structure (NIA; QG). The treatment in this paper is two dimensional, adopting the angular profile of NIA and QG, except that the radiation field is simplified and treated as spherically symmetric. ",
        "conclusions": "Hot accretion flows like ADAFs have a number of physical characteristics that compliment the classic low-temperature disk flows. In previous work, we have explored the consequences of the high temperature and the two-dimensional density structure of these flows, and have found that ADAFs may be able to produce radiatively driven outflows. We subsequently have noted that self-similar CDAFs have even more suitable properties for producing outflows as compared to ADAFs: steeper poloidal density gradients and higher radiation efficiencies. In this paper, we have studied the conditions for self-similar two-dimensional CDAFs (NIA; QG) to develop radiatively heated polar outflows.  1. We have found that CDAFs produce enough luminosity and photon energy to drive polar outflows via Compton heating for a reasonable range of mass accretion rate, or, equally, luminosity, as long as the magnetic field is less than from 10\\% to 1\\% of the equipartition field. When the electron temperature saturates around $10^{11}\\K$ at the inner region, polar outflows are possible for $8\\times10^{-7} \\lesssim L/L_E \\lesssim 4\\times10^{-5}$ for radiation efficiency of $10^{-2}$, where $L_E$ is the Eddington luminosity. The luminosity range for which outflow exists is narrower for lower electron temperature flows and disappears completely for electron temperature $\\lesssim 6\\times 10^9\\K$  2. In most cases, outflows are well collimated along the rotation axis, with an opening angle typically in the range $\\lesssim 10^\\circ$.  3. If we, instead of taking efficiency as constant, assume that it is proportional to the mass accretion rate $\\dot m$, the solutions are qualitatively the same but are more self similar, i.e., opening angle and outer boundary (in Schwarzschild units) depend less strongly on $\\dot m$.  4. Outflow is more probable for small viscosity parameter \\(\\alpha\\).  The treatment in this work is not completely satisfactory in the sense that the dynamics and the temperature profiles of CDAFs are not self-consistently solved. However, it was not our intention to solve fully three-dimensional, self-consistent gloabl CDAFs with proper consideration for all gas, radiative, and magnetic processes, which will be eventually needed to fully understand CDAFs. Rather, our goal has been to show that even in the framework of simple self-similar solutions, radiatively heated polar outflows appear as natural consequences of the physical characteristics of CDAFs."
    },
    "0208/astro-ph0208249_arXiv.txt": {
        "abstract": "We report on photometry results of the equatorial globular clusters (GCs) M10 and M12. These two clusters are part of our sample of GCs which we are probing for the existence of photometrically varying eclipsing binary stars. During the search for binaries in M10 and M12, we discovered the signature of differential reddening across the fields of the clusters. The effect is stronger for M10 than for M12. Using our previously described dereddening technique, we create differential extinction maps for the clusters which dramatically improve the appearance of the color-magnitude diagrams (CMDs). Comparison of our maps with the dust emissivity maps of Schlegel, Finkbeiner, \\& Davis (SFD) shows good agreement in terms of spatial extinction features. Several methods of adding an $E_{V-I}$ zero point to our differential maps are presented of which isochrone fitting proved to be the most successful. Our $E_{V-I}$ values fall within the range of widely varying literature values. More specifically, our reddening zero point estimate for M12 agrees well with the SFD estimate, whereas the one for M10 falls below the SFD value. Our search for variable stars in the clusters produced a total of five variables: three in M10 and two in M12. The M10 variables include a binary system of the W Ursa Majoris (W UMa) type, a background RR Lyrae star, and an SX Phoenicis pulsator, none of which is physically associated with M10. M12's variables are two W UMa binaries, one of which is most likely a member of the cluster. We present the phased photometry lightcurves for the variable stars, estimate their distances, and show their locations in the fields and the CMDs of the GCs. ",
        "introduction": "Variable stars have historically served as important tools and ``laboratories'' in our understanding of star formation, the formation of stellar clusters, and the calibration of distance determination methods. In particular, the study of eclipsing binary stars (EBs) in a globular clusters (GC) may be used to obtain a value for the cluster's distance and a constraint concerning turnoff masses of the GC stars \\citep{paczynski96}.  Simply detecting EBs in the fields of GCs and confirming cluster membership is a straightforward - though data-intensive - task. These systems expand the relatively meager sample of EBs which are currently confirmed GC members \\citep[see for example][and references therein] {mateo1996,mcvean97,rubenstein97,rucinski2000,clement01}. A statistical evaluation of the number of known member EBs in GCs can help in the determination of physical quantities such as the binary frequency in GCs as a parameter in the study of dynamical evolution of GCs \\citep{hut92}.  Member stars of the clusters interact with each other, primarily toward the core of the GC where the stellar density is much higher than in the GC halo.  Due to consequent redistribution of stellar kinetic energies and orbits, the core stars gradually diffuse into the halo which, as a result, grows in size. At the same time, the cluster core itself shrinks, and its density will theoretically reach infinitely high values in a finite period of time \\citep[10-20 $t_{relax}$, according to simulations, see][]{bt87}, a phenomenon known as core collapse.  Binary stars will most likely be located toward the GC center due to the fact that a) they are more likely to form in regions of high stellar density, and b) they will sink toward the core due to their high masses after a few cluster relaxation times. Since the binding energy of one individual hard binary\\footnote{Hard binaries are systems whose binding energies are higher than the kinetic energies of the system itself and of the field star with which it might interact. Consequently, hard binaries tend to not be disrupted by encounters with other stars, but actually turn part of their kinetic energy into additional binding energy of the system.} can be as high as a few to ten percent of the binding energy of the entire GC, they may act as an energy source (similar to the nuclear reaction inside the centers of stars) to halt core collapse. A binary fraction as low as 10\\% in the cluster core will suffice to cool the the central region of the GC by reversing the outward flow of energy toward the halo \\citep{bt87,hut92} and arrest core collapse.  An additional use of simply detecting member-EBs in GC lies in the calibration of absolute magnitudes of and corresponding distances to W Ursa Majoris (W UMa) binaries \\citep[see also Section 4.3]{rucinski1994,rucinski1995,rucinski2000}.  The simultaneous analysis of photometric and spectroscopic data for individual EB systems can moreover provide a direct estimate of the distance to the system \\citep{andersen91,paczynski96}, and thus, if the EB is a GC member, to the GC itself. The main sources of error in this distance determination are a) the relation between surface brightness and effective temperature of the binary and b) the precise determination of the interstellar reddening along the line of sight to the EB which can vary substantially across the field of view of the GC. The distance determination method itself, however, is free of intermediate calibration steps and can provide direct distances out to tens of kpc. In turn, the knowledge of the distances to GCs can then be used to calibrate a variety of other methods, such as the relation between luminosity and metallicity for RR Lyrae stars. The very same analysis can, in principle, be used to obtain the Population II masses of the individual components of the EB system \\citep{paczynski96} to provide a fundamental check of stellar models at low metallicities.  We are currently undertaking a survey of approximately 10 Galactic GCs with the aim of identifying photometrically variable EBs around or below the main-sequence turnoff (MSTO). Our observing strategy, aimed at detecting binaries in the period range of approximately 0.1 to 5 days \\citep{hut92} consists of repeated observations of a set of GCs during each night of an observing run. Multiple runs are helpful in detecting variables with a period of close to one day (or to a multiple thereof).  M10 (NGC 6254) and M12 (NGC 6218) are two equatorial clusters in our sample, located at $\\alpha_{2000} = 16^{h} 57^{m} 08.9^{s}$ and $\\delta_{2000} = -4^{\\circ} 05^{'} 58^{''}$ ($l = 15.14^{\\circ}$ ; $b = 23.08^{\\circ}$), and at $\\alpha_{2000} = 16^{h} 47^{m} 14.5^{s}$ and $\\delta_{2000} = -1^{\\circ} 56^{'} 52^{''}$ ($l = 15.72^{\\circ}$ ; $b = 26.31^{\\circ}$), respectively \\citep{harris1996}. They are nearby (M10: 4.4 kpc and M12: 4.9 kpc) which makes them attractive targets for monitoring studies. Both clusters have been probed for the existence of variable stars in the past \\citep[for a summary, see][]{clement01}. Previous studies searched for luminous variables on the red giant and/or horizontal branch of the clusters and therefore do not overlap with the magnitude range covered in this work.  Details on our photometry observations and basic data reductions are given in Section 2. We discuss how we correct for interstellar extinction along the line of sight to M10 and M12 in Section 3. Section 4 contains the description and the results of our search for variable stars in the clusters including the phased lightcurves for the variables in the cluster fields and our estimates concerning distances and cluster membership. The methods used for Sections 3 and 4 are outlined in detail in two previous publications, namely \\citet[BM01 hereafter]{BM01} and \\citet[BM02 hereafter]{BM02} , respectively. Finally, we summarize and conclude with Section 5. ",
        "conclusions": "The GCs M10 and M12 were monitored for the existence of photometrically variable stars in the magnitude range of approximately $16.5<V<20$ and for periods between roughly 0.05 and 5 days. Observations were conducted over a span of about four years with only relatively few epochs per observing run due to the equatorial locations of the clusters.  During the search for variables, we noticed strong differential reddening effects across the field of M10 (Figure \\ref{M10_rawcmd}) and a similar, but weaker signature across M12 (Figure \\ref{M12_rawcmd}). We correct for this differential reddening by creating extinction maps (Figures \\ref{M10_extmap_lowres} and \\ref{M12_extmap_lowres}). The features visible in these maps are similar to the ones detected by SFD in their dust emissivity maps of the same regions (Figures \\ref{M10_sfd_lowres} and \\ref{M12_sfd_lowres}). Applications of our differential extinction maps to our data significantly improves the appearance of the CMDs of the GCs (Figures \\ref{M10_dered_res3} and \\ref{M12_dered_res3}), especially in the case of M10.  The $E_{V-I}$ zero points for M10 and M12, which need to be added to the differential $E_{V-I}$ values in the grids of Figures \\ref{M10_grid_lowres_trans} and \\ref{M12_grid_lowres_trans}, are 0.23 mag and 0.24 mag, respectively, determined by fitting VDB isochrones to the differentially dereddened data. Since previous reddening estimates for these two GCs vary, especially for M10, our results agree with some literature values but not with others. Specifically, our reddening estimate for M12 agrees well with the SFD maps, whereas our value for M10 falls below the one quoted in SFD ($E_{V-I} \\sim 0.389$), but produces by far the best VDB isochrone fit.  Reddening studies like the one presented here may be useful in determining properties of the interstellar medium such as a dependence of $R_{V}$ upon position in the field of view. In addition, they may give insight into the properties and the distribution of the dust along the line of sight. One may argue, for instance, that M10 and M12, since they are so close to each other in the sky, lie behind a common, larger scale ``layer\" of dust that accounts for the practically identical reddening zero point. M10 seems to fall behind another dust ridge \\citet[suggested by, e.g.,][]{kbk96} causing the additional differential reddening which varies on arcmin scales by tenths of a magnitude. Furthermore, reddening maps such as the ones presented here and in BM01 can give a sense of how small a scale differential extinction in general may vary on, and by how much, perhaps as a function of position in the sky or Galactic latitude.  We hope to shed some light on this matter with a similar analysis of the remaining GCs in our sample.  Our search for variable stars in M10 and M12 produced five candidates. In M10, we detected a W UMa binary, an RR Lyrae, and an SX Phe field variable. None of these three systems appears to be physically associated with M10, given their locations in the CMD (Figure \\ref{M10_isocmd}) and their estimated distances (Table 7 and Section 4.3.3). In M12, we detected two W UMa binary systems, one of which (M12-V1) is most likely a cluster member.  Again, we base this estimate on CMD location (Figure \\ref{M12_isocmd}) and calculated distance moduli (Table 7).  As in the case of NGC 3201 (BM02), we detect a number of foreground and background variables but only one (likely) GC member binary. Both M10 and M12 are at higher Galactic latitude ($\\sim 25^{\\circ}$) than NGC 3201 ($\\sim 8.5^{\\circ}$), so the disk contamination is not as significant. The number of total stars which were monitored was only slightly different (27000 for NGC 3201 and 34000 for M10 and M12 combined), but in the case of NGC 3201, we found 14 variables (11 binaries, 1 member), whereas in this work we find 5 variables (3 binaries, 1 probable member).  In both studies, cluster-member stars outnumber non-member stars, as may easily be seen in the CMDs of the clusters. Yet the number of non-member binaries we find is much higher than the number of member binaries\\footnote{It should be noted that we are not sensitive to cluster-member pulsating variables since almost all of them are saturated in the data we analyze for the existence of binaries. A comparison between GC member and non-member pulsating variables is there not meaningful.}. It will be interesting to see whether this trend continues in the rest of the GCs in our sample which will be analyzed using the same methods as outlined in this work, BM01, and BM02, or whether NGC 3201, M10, and M12 are merely ``special cases\".  Finally, we have shown that we are able to detect binaries of very low amplitude ($\\sim 0.07$ mag; BM02) and duty cycle (0.1; BM02) and of short ($\\sim 0.13$ days\\footnote{M10-V3 actually has a period of just above 0.06 days, but only has one maximum per period. Were it a binary star, it would have two maxima, and its period would consequently be around 0.13 days}; this work) and long periods (2.85 days; BM02), even when using very different observing techniques for the clusters. Using our existing 1m-class telescope photometry data set of the rest of the GCs in our sample, we should be able to identify binaries and other variables in these clusters (provided they exist) and determine their periods. With the spectroscopic capabilities of modern 8m-class telescopes for follow-up observations, we can determine cluster membership for these variables in the near future."
    },
    "0208/astro-ph0208413_arXiv.txt": {
        "abstract": "{We describe how causality and energy-momentum conservation constrain the superhorizon behaviour of perturbation variables in a general FRW spacetime. The effect of intrinsic curvature upon the horizon scale is discussed and `white noise' results for the power spectra of generic `causally-seeded' superhorizon perturbations are obtained.  We present a detailed derivation of the superhorizon energy density power spectrum for curved universes, and provide elegant mathematical arguments for a curved universe analogue of the Bessel function addition theorem.  This yields physical insight into both the familiar $k^4$ result for flat space and  the  $K \\not= 0$ case: curvature effects are seen to be evident from as early as half the curvature scale. We consider the implications of these results for the physically and numerically significant issue of setting consistent initial conditions.} ",
        "introduction": "\\label{initial}  The standard flat universe inflationary plus dark energy paradigm appears to be remarkable accord with recent CMB experiments \\cite{Jaffe}.  Nevertheless, the confrontation with observation remains indecisive, not only because of the significant experimental uncertainties, but also because good quantitative accuracy has not yet been achieved for competing paradigms such as actively sourced  theories, which have so far only been extensively studied for cosmic defects in flat FRW ($K = 0$) backgrounds \\cite{ACSSV,pst,pst2,durrer}. Moreover, even for flat universes,  a subsidiary role for defect networks complementing the inflationary power spectrum cannot be excluded \\cite{bprs}. The possibilty of searching for distinct defect signatures in the CMB, such as nonGaussianity \\cite{CMBstrings} or in the polarisation spectra \\cite{SZ} may go some way toward resolving these remaining uncertainties.  However, in order to have confidence in cosmological parameter estimation with a purely inflationary paradigm, it will be necessary to constrain the alternative models, including both `primordial' and `active' perturbation sources, as well as the effects of vector and tensor modes and  $K \\not = 0$ backgrounds.  Here the combination of intrinsic curvature and defect sources is particularly interesting,  motivating a study of the superhorizon behaviour and the  choice of consistent initial conditions for such perturbations.  Of particular interest is the possibility of a testable signature of superhorizon fall-off in `causal' or `active' models in the CMB polarisation power spectrum.  The Cauchy problem for cosmological perturbations over a $K = 0$ FRW background has been extensively discussed in the literature. Using energy-momentum conservation, and assuming that the early universe was purely homogeneous and isotropic, one can show \\cite{pa,CS,peeb,vs} that the power spectrum of the stresses should be that of white noise ($\\propto k^0$), while that of the density should be suppressed as $k^4$, for wavelengths much larger than the horizon; here, $k$ is the comoving wavenumber so this the superhorizon condition is $k \\ll aH$ where $H={\\dot a/a}$ and $a$ is the scale factor. Assuming that matter undergoes random displacements of length $d_c$ at each point in space, while satisfying energy-momentum conservation, one can estimate that the density power spectrum turns over to $k^4$ behaviour at $k_c \\sim 2 d_c^{-1}$ \\cite{pa}. This result may be used \\cite{vs,pst} to simplify the calculations to determine consistent initial conditions for numerical analyses. One assumes that all perturbations of relevance today come from far outside the initial horizon ($k \\tau_{\\rm i} \\ll 1$) so that one may argue that the strong $k^4$ suppression allows us to set all these variables to zero initially.  For the general $K \\not= 0$ FRW cosmologies, these issues remain unaddressed, although there do exist several analyses of the effect of curvature on the CMB assuming primordial inflationary perturbations described by power law ansatzes. Some examples include: the generalisation of the Sachs-Wolfe effect to include curvature effects \\cite{peeb}; refs. \\cite{KamiSper} and \\cite{Wilson} treat the open universe case, respectively using a gauge invariant and a synchronous gauge formalism --- see also  ref. \\cite{as} --- and ref. \\cite{WhitScot} investigated the closed universe case, again using a gauge invariant approach. However, the literature does not contain (to our knowledge) any direct analysis of the superhorizon effects of energy-momentum conservation on active and causal seed sources in intrinsically curved FRW spacetimes.  In this paper we shall therefore describe the manner in which causality and energy-momentum conservation constrain the superhorizon behaviour of the perturbation variables in the general FRW  spacetime, in curvilinear coordinates.  As discussed in our previous paper --- ref. \\cite{Amery3} --- we may describe the symmetry properties and stress energy conservation laws of these spacetimes in two equivalent languages corresponding to integral and differential conservation laws. The language of conserved vector densities $\\hat{I}^\\mu$ ({\\it c.f.} Katz \\et \\cite{kblb-1}) uses the conformal geometry of the background spacetime to naturally construct integral constraint equations, representing strong conservation laws defined over some region of a spacelike hypersurface.  These reduce to Noether conservation laws when constructed using the  Killing geometry of the background.  Traschen \\cite{tras1} obtained such integral constraints using as a background manifold a maximally symmetric de Sitter spacetime in quasi--Minkowskian coordinates, and these were related to the vector densities $\\hat{I}^\\mu_{\\bf \\xi}$ by Dereulle \\et \\cite{dku-1}.   However, in order to make use of these integral constraints, we need to know their behaviour on the boundary of the integration volume.  In the case of bounded (or ``local'') perturbations, the boundary terms are assumed to vanish, that is, these previous analyses consider perturbations whose stress energy vanishes beyond the horizon. This is a stronger assumption than required by the physics, since only the two-point correlators between these quantities have to vanish in order to satisfy causality. In other words, a physical assumption has been made that is perhaps better suited to the analysis of an explosive event rather than more general causal perturbations, such as those seeded by topological defects, which may persist beyond the horizon ({\\it e.g.} cosmic strings).  Moreover, the choice of background relative to which energy-momentum is formally defined  is that of a de Sitter universe, rather than the FRW background convenient for most simulations.  Hence, we turn to the equivalent language of energy-momentum pseudo-tensors \\cite{Amery3}, outlined in \\S \\ref{SEC-em}.  Again, the relative simplicity of the flat space case is complicated by the introduction  of curvature.  In \\S \\ref{int-k4sec} we define the two--point correlator, and discuss causality, while the transformed correlator (with respect to the eigenfunctions of the Laplacian) is described in \\S \\ref{Tcor-k4sec}. The effect of intrinsic curvature upon the horizon scale is dealt with in \\S \\ref{HS-k4sec};  and in \\S \\ref{wn-k4sec} we obtain results for a general FRW spacetime analogous to the white noise power spectrum of two arbitrary functions in flat space. Then, in \\S \\ref{SEC-SHiso},  we consider in more detail the superhorizon behaviour of the pseudo-energy. This not only yields the familiar $k^4$ behaviour for the isotropic ($00$-)mode on superhorizon scales for $K = 0$, but is also well suited to study the $K \\not= 0$ case.  We regain the flat space results in the local limit and discuss the effects of intrinsic curvature on the isotropic mode in a Helmholtz decomposition of the density perturbation associated with causal sources, for both open and closed cosmologies. In \\S \\ref{INITCON} we comment on the implications of these results for the choice of consistent initial conditions. We work in the synchronous gauge because of its ubiquity in numerical simulations and the greater physical transparency offered by this gauge choice. ",
        "conclusions": "\\label{INITCON}  It would be of great physical interest if the consequences of this superhorizon behaviour were within the range of observational searches because it would test  `active' theories for structure formation, and possibly probe the curvature of the universe.  The cosmic microwave sky may provide the best prospect, however, for the CMB temperature spectrum the superhorizon `fall-off' expected on the very largest scales will be confused by cosmic variance.  The most likely signature would be through the CMB polarization spectrum because it represents a primordial snapshot of the universe at recombination which includes about $10^5$ different horizon volumes.  The lack of spatial correlations on scales corresponding to the horizon size at that time (about $2^\\circ$), and the fall-off in their transformed counterparts, should be observationally testable. Previous analysis of this possibility \\cite{SZ}, however, has only considered scalar modes, so we comment on the possible impact of vector and tensor modes here.  We leave discussion of the observational imprint of this superhorizon behaviour for a future publication \\cite{Amery5}, concentrating instead on the consistent implementation of initial conditions for numerical simulations.  \\subsection{SVT decomposition and superhorizon coherence} \\label{SHCOH}  The decomposition of a tensor perturbation into its scalar, vector and tensor parts is not without its subtleties \\cite{vs}, since the projection operators yielding the scalar, vector and tensor parts are not, in general, analytic. The spatial distribution of the separate geometric parts may therefore be significantly different from the physical distribution of the entire object, so that the decomposed parts of a causal tensor object can manifest acausal features. The transformed variables will then in turn exhibit non-analytic behaviour \\cite{durrer}, complicating the analysis of their superhorizon behaviour.  However, causality is merely hidden, and, since the arguments presented in this paper for the (scalar) energy density perturbation are (trivially) phrased in terms of full tensors, they  escape this difficulty.  On the other hand, it is common to employ the assumption of coherence, in which case  one regains analyticity \\cite{durrer}, all the decomposed parts of the perturbation are causal, and one may obtain the power spectra by simple power counting.  This allows us to write down the auto-correlators for both the scalar and the vector parts of  the pseudo-tensor $\\tau^\\mu_{\\;\\; \\nu}$, under the influence of energy-momentum conservation.  If we decompose the pseudo-tensor $\\tau^\\mu_{\\;\\; \\nu}$ as in (\\ref{met-decomp}), we may write the energy-momentum equations (\\ref{tau-cons}) as \\begin{eqnarray} \\tau_{IV} &=& \\frac{\\dot{\\tau}_S}{{k}} \\; , \\hspace{20mm} k \\dot{\\tau}_{IV} = 2 {k}^2 \\left[ \\tau_L + 2 \\left( \\frac{K}{{k}^2} - \\frac{1}{3} \\right) \\tau_T \\right]  = \\ddot{\\tau}_S \\; , \\nonumber \\\\ \\tau^{(\\pm 1)}_{IT} &=& - 2 {k} [ {k}^2 - 2K ]^{-1} \\dot{\\tau}^{(\\pm 1)}_V = \\frac{\\ddot{h}^{(\\pm 1)}_V}{\\kappa} \\; , \\label{T-DEF2-3} \\end{eqnarray} where we have used the first equation to obtain the final equality in the second.  Assuming coherence, we find  that the scalar ($\\tau_S$) and vector ($\\tau_{IV}$, $\\tau_V$) degrees of freedom have auto-correlators that go as \\begin{eqnarray} \\langle |\\tau_S|^2 \\rangle (\\beta) &~~\\propto~~& \\tilde{k}^2 \\langle |\\tau_{IV}|^2  \\rangle (\\beta) \\; , \\nonumber \\\\ \\langle |\\tau_{IV}|^2 \\rangle (\\beta) &~~\\propto~~& \\tilde{k}^2 \\left( 4 \\langle |\\tau_{L}|^2 \\rangle (\\beta) - \\frac{16}{3} \\langle \\tau_L \\tau_T^* \\rangle (\\beta) + \\frac{16}{9} \\langle |\\tau_T|^2 \\rangle (\\beta) \\right) \\nonumber \\\\ &~~~~~~+& 16 K \\left( \\langle | \\tau_L|^2 \\rangle - \\frac{2}{3} \\langle | \\tau_T|^2\\rangle (\\beta) \\right) + 16 \\frac{K^2}{\\tilde{k}^2} \\langle | \\tau_T |^2\\rangle (\\beta) \\; , \\nonumber \\\\ \\langle |\\tau^{(\\pm 1)}_V |^2 \\rangle (\\beta) &~~\\propto~~& \\left( \\frac{1}{4} \\tilde{k}^2 - K + \\frac{K^2}{\\tilde{k}^2} \\right) \\langle |\\tau^{(\\pm 1)}_{IT} |^2 \\rangle (\\beta) \\; , \\label{ApCoSh} \\end{eqnarray} on superhorizon scales. Moreover, the correlators for $\\tau_L$, $\\tau_T$, and $\\tau_{IT}$  have white noise superhorizon behaviour (refer to \\S \\ref{wn-k4sec} and \\cite{durrer}) so we see that in the local limit ($\\chi << 1$, $\\beta >> 1$) we regain the flat space results, with the scalar power spectrum $\\propto \\tilde{k}^4$, while those of the vectors is proportional to $\\tilde{k}^2$.  In the supercurvature limit, they all tend to a constant. Note that the correlators in (\\ref{ApCoSh}) do not diverge as $\\beta \\longrightarrow 0$, since $\\tilde{k}$ tends to $1$ in this limit.  The results (\\ref{ApCoSh}) are more general than one might suppose upon first inspection, as an arbitrary decoherent source may be written as a `decoherent sum' of coherent sources, each of which will then be amenable to the above arguments \\cite{turok1}. Moreover, we believe that it is plausible to expect approximate coherence on the very largest scales, since the eigenfunctions used to diagonalise the `decoherent sum' are likely to either all tend to white noise on these scales, or to be dominated by just one element in the sum --- see also ref. \\cite{Amery5}.  This notion  has been ventured before as an hypothesis \\cite{DS} for scaling sources, and is supported by both numerical results \\cite{KD}, and by the observation that, for causal sources, modes with wavelengths much larger than the horizon cannot be significantly out of phase with one another.  We shall consider this point in more detail elsewhere \\cite{Amery5}.   \\subsection{Matching conditions and initial conditions}  Finally, we wish to propose a set of initial conditions which would find useful application in numerical simulations of causal perturbation models, such as cosmic defects.  We shall assume, as is usual, that we are dealing primarily with the evolution of transformed quantities. By considered an instantaneous phase transition early in the universe as a first approximation to a model for cosmic defects ``switching on'',  and employing a gauge in which constant energy and constant time surfaces coincide,  matching conditions can be used to show that there exists an entire class of objects which are  continuous across the transition  and are related by gauge transformation to our pseudo-tensor components.  In a previous paper (ref. \\cite{Amery3}), we used the notion of compensation together with a particular gauge specification to remove this  redundancy  such that the  $\\tau_S$ (pseudo-energy) and $\\tau^{(\\pm 1)}_V$ (divergenceless vector) components of our generalized pseudo-tensor have this property: \\begin{eqnarray} [\\tau_S ]_{\\pm} = 0 \\; , \\hspace{15mm} [ \\tau_V^{\\pm 1}]_{\\pm} = 0 \\,. \\nonumber \\end{eqnarray} While we emphasise that this matching depends on a special gauge choice, it is one which should be widely applicable in many physical situations, for example, a defect-forming transition \\cite{Amery3}.  For a universe which was unperturbed (and hence homogeneous and isotropic) prior to the transition, we may then take  $\\tau_S = 0 = \\tau^{(\\pm 1)}_V$  as natural initial conditions.  This result is consistent on all scales and is especially appropriate for suppressed superhorizon modes. The results of \\S \\ref{SEC-SHiso} confirm this conclusion for the density perturbation tracked by $\\tau_S$, and the previous section argues that this conclusion may be extended also to the `induced vector' mode $\\tau_{IV}$ and the true vector modes $\\tau_V^{\\pm 1}$. Hence, if one wishes to set initial conditions for perturbations initially outside the horizon, one may consistently  set both the pseudo-energy $\\tau^0_{\\;\\;0}$ and the pseudo-momentum $\\tau^0_{\\;\\;i}$ to zero.  If one wishes to include long wavelength modes that are affected by curvature, then one can proceed in a similar fashion.  Although the transformed correlators become constant on these scales, at early times they are sufficiently outside the horizon for their amplitude to be strongly suppressed.   Therefore, the matching conditions we propose as initial conditions will also be applicable and self-consistent for a curved universe."
    },
    "0208/astro-ph0208139_arXiv.txt": {
        "abstract": "We develop a useful formula for power spectrum analysis for high and intermediate redshift galaxy samples, as an extension of the work by Feldman, Kaiser \\& Peacock (1994). An optimal weight factor, which minimizes the errors of the power spectrum estimator, is obtained so that the light-cone effect and redshift-space distortions are incorporated. Using this formula, we assess the feasibility of the power spectrum analysis with the luminous red galaxy (LRG) sample in the Sloan Digital Sky Survey as a probe of the equation of state of the dark energy. Fisher matrix analysis shows that the LRG sample can be sensitive to the equation of state around redshift z=0.13. It is also demonstrated that the LRG sample can constrain the equation of state with ($1$-sigma) error of $10 \\%$ level, if other fundamental cosmological parameters are well determined independently. For the useful constraint, we point out the importance of modeling the bias taking the luminosity dependence into account. We also discuss the optimized strategy to constrain the equation of state using power spectrum analysis. For a sample with fixed total number of objects, it is most advantageous to have the sample with the mean number density $10^{-4}~h^3{\\rm Mpc}^{-3}$ in the range of the redshift $0.4 \\simlt z\\simlt 1$. ",
        "introduction": "The clustering of the cosmological objects like galaxies, clusters of galaxies and QSOs, is the fundamental to probe the Universe because it directly reflects properties of dark components and the primordial density fluctuations. The power spectrum is a simple but very useful tool to characterize their spatial distribution. Actually, useful constraints on the cosmological parameters are obtained from the power spectrum analyses of the Two Degree Field (2df) Galaxy Redshift Survey (Percival et~al. 2001) and the 2df QSO Redshift Survey (Hoyle et~al. 2002). In such power spectrum analyses, many authors base their methods on the seminal paper by Feldman, Kaiser \\& Peacock (1994, Hereafter FKP).  For redshift surveys such as the 2df survey and the Sloan Digital Sky Survey (SDSS), however, several observational effects on the power spectrum analysis can be very important. Because cosmological observations are feasible only on the light-cone hypersurface defined by the current observer, the effect of the redshift evolution of the luminosity function, the clustering amplitude, and the bias, contaminates the observational data. We call this the light-cone effect (Matarrese et al. 1997; Matsubara, Suto \\& Szapudi 1997; de Laix \\& Starkman 1998; Yamamoto \\& Suto 1999). On the other hand it is well known that the distribution of the sources in redshift sapce is different from that in the real space due to the redshift-space distortions. The linear redshift distortion is the effect of the bulk motion of the sources within the linear theory of density perturbation (Kaiser 1987; Hamilton 1998 and references therein). The finger of Got effect is the redshift distortion due to the random motion of the sources in the nonlinear regime (Mo, Jing \\& Boerner 1997; Magira, Jing \\& Suto 2000). The geometric distortion is the effect caused by a choice of the distance-redshift relation to plot a map of the sources (Ballinger, Peacock \\& Heavens 1996; Matsubara \\& Suto 1996).  The LRG sample of the SDSS spectroscopic survey will provide a sample of $10^5$ intrinsically luminous early-type galaxies to $z\\sim0.5$ (Eisenstein et~al. 2001). In the analysis of clustering statistics of such a sample, the light-cone effect and the redshift-space distortions can be substantial. Fortunately these observational effects have been well investigated and we can model the power spectrum incorporating them (see section 3.1). The primary purpose of the present paper is to extend the formulas in FKP in order to incorporate these observational effects and to derive a generalized expression of the optimal weight factor in the power spectrum analysis (cf. Tegmark 1995).  The results of the 2df galaxy survey (Peacock et~al. 2001), weak lensing surveys (Refregier et~al. 2002; Bacon et~al. 2002; and references therein), cosmic microwave background anisotropy measurements (e.g., Ruhl et~al. 2002) and type Ia supernovae measurements (Perlmutter et~al. 1999; Riess et al. 1998), support the concordance model (Wang et~al. 2000): A spatially flat universe dominated by the dark energy component and, with respect to structure formation, cold dark matter with the primordial density fluctuations predicted in the inflationary scenario. The very recent result by the Wilkinson Microwave Anisotropy Probe (WMAP) strongly supports the concordance model (Spergel et~al. 2003). Now the mystery of the dark energy component has become one of the most important issues in cosmology, which has led to recent activity in investigating the dark energy (See e.g. Peebles \\& Ratra 2003, for a recent review, and references therein). The dark energy can be characterized by its equation of state $w_X=p_X/\\rho_X$, where $p_X$ is the pressure and $\\rho_X$ is the energy density. For the cosmological constant, $w_X=-1$. However, if the dark energy originates from the vacuum energy of a variable scalar field, like the quintessence model, $w_X$ can take values $w_X>-1$, and can in general be a function of redshift. Thus constraints on the equation of state is quite important in considering the origin of the dark energy. Then various strategies for probing the equation of state have been investigated (e.g. Newman \\& Davis 2000, Saini, et~al. 2000, Wang, et~al. 2000, Chiba \\& Nakamura 2000, Huterer \\& Turner 2001 , Yamamoto \\& Nishioka 2001, Kujat et~al. 2002, and references therein).  Second purpose of this paper is to assess the feasibility of measuring the equation of state using the power spectrum analysis. \\footnote{After completing this work, a similar investigation by Matsubara \\& Szalay (2003) has been announced.} Recently, Matsubara \\& Szalay have discussed the usefulness of the LRG sample in SDSS for measuring the cosmological parameters (2002). Their method is based on maximum likelyhood analysis in redshift space. They demonstrated the usefulness of various SDSS samples to constrain cosmological parameters by estimating the Fisher matrix element. Motivated by their work, we assess the feasibility of the method with the power spectrum analysis of the LRG sample in SDSS to probe the equation of state based on the Fisher matrix formalism for power spectrum analysis (Tegmark 1997, Tegmark et~al. 1998a). The advantage of our method is that the formulae are simple and analytic, which allows us to evaluate the formulae easily and to understand meaning of results clearly.  This paper is organized as follows: In section 2, we derive formulae for the power spectrum analysis taking the light-cone effect and the redshift-space distortions into account. In section 3, the Fisher matrix element is evaluated using an approximate method. There we focus on the matrix elements relevant to a measurement of dark energy, especially the equation of state. In section 4, we discuss optimizing the strategy, for a sample with a fixed total number of objects, using  the power spectrum to probe the equation of state. Section 5 is devoted to summary and conclusions. Throughout this paper we use a system of units in which the velocity of light $c$ equals $1$.   \\def\\bfs{{\\bf s}} \\def\\bfk{{\\bf k}} \\def\\calF{{\\cal F}} \\def\\calP{{\\cal P}} \\def\\DeltaV{{V}} \\def\\bfF{{F}} ",
        "conclusions": "In summary we derived a rigorous optimal weighting scheme for power spectrum analysis, which is useful for samples in which the light-cone effect and the redshift-space distortions are substantial. Our result is a simple extension of the work by FKP, and we obtained a generalized optimal weight factor which minimizes the errors of the power spectrum estimator. As an application of our formula, we investigated the capability of the LRG sample in SDSS to constrain the equation of state parameters by evaluating Fisher matrix elements. Even if the transfer function for the matter power spectrum does not depend on $w_X$, the power spectrum analysis of the redshift-space sample can constrain $w_X$ due to the geometric distortion. To incorporate uncertainties of redshift evolution of the clustering bias, we considered the marginalized probability function by integrating over the parameters of the bias models. This analysis shows that the LRG sample in SDSS has a serviceable potential for constraining the equation of state around $z=0.13$ with $1$ sigma errors at the $10$ \\% level, if other fundamental parameters are well determined in an independent fashion. We also showed that this conclusion is not altered in the case of a bias model incorporating the redshift-evolution due to selection effect depending on luminosity. However, even in our realistic treatment of the bias, we made simplifications: Uncertainties including the stochasticity and the nonlinearity in modeling the bias are not considered in our investigation. Then, tests on the bias properties will be required for more definite conclusions.  In the present paper, we assumed $10^4$ deg${}^2$ as the complete SDSS survey area. When the planed survey area are not achieved, the capability to constrain the equation of state reduces. As the Fisher matrix element is in proportion to the survey area $\\Delta \\Omega$, then the statistical error $\\Delta \\bar w$ increases in proportion to $\\sqrt{\\Delta \\Omega}$. For example, when we assume $5\\times 10^3$ deg${}^2$ and $7.5\\times 10^3$ deg${}^2$ as the final SDSS survey area, $\\Delta \\bar w$ increases by $30$ \\% and $15$ \\%, respectively, as long as the inhomogeneity of the incomplete survey area does not cause additional systematic errors.  In section 4, we considered the optimized sample to constrain $w_X$ using power spectrum analysis. We found that it is most advantageous to have the sample with the comoving number density $\\bar n\\simeq 10^{-4}~h^3{\\rm Mpc}^{-3}$ in the range of redshift $0.4\\simlt z \\simlt 1$. For such a sample, the efficiency per object to constrain $w_X$ is optimized. Information from anisotropic power spectrum would improve the capability of constraining the parameters, as demonstrated in the 2df QSO sample (Outram et~al. 2001)."
    },
    "0208/astro-ph0208305_arXiv.txt": {
        "abstract": "We present first results from the Survey for Transiting Extrasolar Planets in Stellar Systems (STEPSS). Our goal is to assess the frequency of close-in extrasolar planets around main-sequence stars in several open clusters.  By concentrating on main-sequence stars in clusters of known (and varied) age, metallicity, and stellar density, we will gain insight into how these various properties affect planet formation, migration, and survival.  We show preliminary results from our 19 night photometric campaign of the old, solar metallicity cluster NGC 1245.  Taking into account the photometric precision, observational window function, transit probability, and total number of stars monitored, we estimate that we should be able to probe planetary companion fractions of $<1\\%$ for separations of $a<0.03$ AU.  If $1\\%$ of the stars in the cluster have Jupiter-sized companions evenly distributed in $\\log{a}$ between 0.03 and 0.3 AU, we expect to find $\\sim 2$ transits.  A preliminary search of our light curve data has revealed a transit with a depth $\\sim 4\\%$. Based on its shape, it is likely to be a grazing binary eclipse rather than a planetary transit, emphasizing the need for high temporal resolution in transit surveys. ",
        "introduction": "The recent detections of candidate low-mass companions to local disk stars via transits by several groups has verified the possibility of discovering statistically significant numbers of extrasolar planets through ground based transit surveys (Udalski et~al.\\ 2002, Mallen-Ornelas et~al.\\ 2003).  Although such surveys will provide valuable information about the types of planetary systems that exist, it will be difficult to piece together the dynamical, chemical, and evolutionary history of the parent stars.  The parent stars' physical properties are particularly important since radial velocity surveys for low-mass companions have revealed the trend of the fraction of stars with planets increases as a function of the metallicity of the parent star.  The interpretation of this trend is unclear and more information is needed.  Simple stellar systems, such as globular clusters and open clusters, are an excellent laboratory for transit surveys.  Such fields provide $\\sim 10^{3-5}$ stars of the same age and metallicity.  Interpretation of the null result for the metal-poor globular cluster 47 Tuc (Gilliland et~al.\\ 2000) has been complicated by the possible effects of the dense stellar environment on planet formation and survival.  Therefore, we are concentrating on metal-rich, sparser open clusters. By observing main-sequence stars in 4-5 open clusters of known (and varied) age, metallicity, and stellar density, we will gain insight into how these various properties affect planet formation, migration, and survival. Our primary instrumentation is the MDM 8192x8192 4x2 Mosaic CCD imager.  In combination with the MDM 2.4m telescope it yields a 25x25 arcmin$^{2}$ field of view with 0.18\"/pixel.  The Ohio State University has access to a 25\\% share of the MDM facility; in 2001 we were granted 19 nights on the 2.4m, we have been granted $\\sim 60$ nights during Fall 2002, and we expect to obtain 20-40 nights for each cluster in the future with the same instrumental setup.   \\begin{figure} \\plottwo{errors.eps}{scatter.eps} \\caption{(a) RMS scatter over the entire run as a function of $I$ magnitude for all stars on a single CCD chip.  (b) The ratio of the transit depth due to a 1 $\\rjup$ planet to the RMS scatter. } \\end{figure}  \\begin{figure} \\plotfiddle{var.3.207.eps}{4.0cm}{0}{35}{35}{-120}{-90} \\caption{light curves for a transit candidate observed on two separate nights. Unfortunately, this is likely to be a grazing binary eclipse. } \\end{figure} ",
        "conclusions": ""
    },
    "0208/astro-ph0208133_arXiv.txt": {
        "abstract": "We examine constraints from Big Bang nucleosynthesis on type II Randall-Sundrum brane cosmologies with both a dark radiation component and a quadratic term that depends on the 5-dimensional Planck mass, $M_5$.  Using limits on the abundances of deuterium and helium-4, we calculate the allowed region in the $M_5$--dark radiation plane and derive the precise BBN bound on $M_5$ alone with no dark radiation:  $M_5 > 13$ TeV. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0208/astro-ph0208419_arXiv.txt": {
        "abstract": "We develop a method based on the collisionless Boltzmann equation to calculate the gravitational clustering of relic neutrinos in realistic cosmological models dominated by cold dark matter (CDM) and the cosmological constant.  This method can be used to estimate the phase-space distribution of any light particles in CDM halos.  We find that neutrinos with masses $\\agt$ 0.3 eV cluster appreciably in dark matter halos above the galactic size.  The resulting neutrino overdensity above the cosmic mean neutrino density increases with both the neutrino mass and the halo mass, ranging from $\\sim 10$ for 0.3 eV neutrinos in $\\sim 10^{13} M_\\odot$ halos to $\\sim 1500$ for 1.8 eV neutrinos in $\\sim 10^{15} M_\\odot$ halos.  We examine the implications of neutrino clustering for the Z-burst model of ultra high energy cosmic rays (UHECR), which interprets the observed events at $E > 4\\times 10^{19}$ eV as decay products of Z-bosons from the resonant scattering between relic and high energy neutrinos and anti-neutrinos.  We estimate the UHECR energy spectrum for various neutrino masses towards five of the most massive clusters in the local universe (within 100 Mpc): Virgo, Perseus-Pisces, Hydra, Centaurus, and Coma.  The UHECR flux in the Z-burst model is expected to be significantly higher towards these clusters if $m_\\nu\\agt 0.3$ eV and nearly isotropic otherwise. ",
        "introduction": "The nature of cosmic rays above the Greisen-Zatsepin-Kuzmin (GZK) cutoff \\cite{GZK} at $\\sim$ 4 $\\times$ 10$^{19}$ eV is an unsolved problem in ultra high energy cosmic ray (UHECR) physics \\cite{review}. These events have been reported by the Akemo Giant Air Shower Array (AGASA) \\cite{AG}, Fly's Eye \\cite{FE}, Havera Park \\cite{HP}, HiRes \\cite{HR}, and Yakutsk \\cite{YK} collaborations.  Interactions with the cosmic background photons $\\gamma_{cmb}$ via photoproduction of pions ($p \\gamma_{cmb} \\rightarrow p + N \\pi, n\\gamma_{cmb} \\rightarrow n + N\\pi$), photopair production ($p \\gamma_{cmb} \\rightarrow p e^{+}e^{-},\\gamma \\gamma_{cmb} \\rightarrow e^{+}e^{-}$), and inverse Compton scattering ($e^{\\pm} \\gamma_{cmb} \\rightarrow e^{\\pm} \\gamma$, $p \\gamma_{cmb} \\rightarrow p\\gamma$) at high energies \\cite{L94} constrain a $\\sim 10^{20}$ eV cosmic ray to a few Mpc for the characteristic lengths of either charged cosmic rays or neutrons and photons.  More specifically, the attenuation length of protons above the GZK cutoff is $\\sim$ 50 Mpc.  The lack of known processes to accelerate cosmic rays in small Galactic objects makes the Galactic origin of these ultra high energy particles unfeasible \\cite{FS95}.  Novel powerful acceleration mechanisms for light nuclei are required if these energetic particles are produced in nearby galaxies \\cite{A02}.  Exotic particles and dynamics have also been suggested, but these come with their own difficulties \\cite{review}.  One proposed explanation for the UHECRs is the Z-burst model, which tries to solve the puzzle without invoking new physics beyond the standard model of particle physics except for neutrino masses. Several recent experiments \\cite{SK601,MAC01,SN01,TR01} have found evidence for non-zero neutrino mass.  The Z-burst model hinges on the fact that ultra high energy neutrinos (and anti-neutrinos) produced at cosmological distances can reach the GZK zone unattenuated.  Their resonant annihilation on the relic anti-neutrinos (and neutrinos) produces Z bosons, about 70\\% of which decay into hadrons within $\\sim 10^{-25}$ sec.  The final state has fifteen pions and 1.35 baryon-antibaryon pairs on average \\cite{FMS99}, where the fifteen pions decay into thirty high energy photons.  The Z boson is highly boosted ($\\sim 10^{10}$) \\cite{W99}, resulting in a highly collimated beam with a half angle of $\\sim 10^{-10}$.  This and the fact that the effect of magnetic fields at such high energies is negligible \\cite{GMT01} ensure a high probability for the protons and photons to reach the observer if the Z-burst occurs in the direction of the Earth.  The Z-burst model has been discussed in detail in many papers \\cite{W82,R93,YSL98,GK,CDF01,FKR01,FKR02}.  The resulting cosmic ray flux has been shown to depend strongly on the density of the relic neutrinos \\cite{YSL98,FMS99,W99,F02,FGSD01}, but the neutrino density in these calculations has been taken to be either the constant relic density from the big bang or some ad hoc value.  In this paper we perform a detailed calculation of the neutrino clustering in the local universe using realistic cosmological models and apply the results to the Z-burst model for UHECRs.  Since the current constraints from cosmological observations and laboratory experiments indicate that the neutrino masses are small ($\\lo 2$ eV; see Sec~II) and the CDM dominates the dark matter density ($\\Omega_{\\rm cdm} \\gg \\Omega_\\nu$), we do not expect the clustering of neutrinos to affect significantly that of the CDM.  As a result, it is not essential to use full scale, time consuming $N$-body simulations.  Instead, we solve the collisionless Boltzmann equation for the neutrino phase space distribution in a background potential given by the universal profile of CDM halos reported in recent high resolution simulations \\cite{NFW96}.  The Boltzmann equation is then linear in the neutrino density contrast and has tractable integral solutions.  The advantage of this method over the conventional $N$-body simulations is that we can obtain the neutrino density profile much below the resolution scale ($\\sim 50$ kpc) of large cosmological simulations by using as an input the CDM potential determined from much higher resolution simulations of individual halos.  Moreover, the computation time required for our approach is negligible compared with numerical simulations, thereby allowing us to explore a large parameter space of neutrino masses and dark matter halo masses.  In Sec~II the relevant Boltzmann equation and the integral solutions are derived.  In Sec~III results for the clustering of neutrinos for different neutrino masses and CDM halos are presented and compared with physical arguments based on neutrino free streaming and the Tremaine-Gunn constraint \\cite{TG79}.  The resulting neutrino density profiles are also compared with earlier $N$-body simulations \\cite{KKPH96}, which show good agreement.  In Sec~IV the neutrino overdensity calculation is incorporated in the Z-burst model for UHECRs, where we make realistic predictions for the UHECR energy spectrum for different neutrino masses.  We estimate the level of anisotropy in the UHECR flux by examining lines of sight towards five of the most massive clusters (Virgo, Perseus-Pisces, Hydra, Centaurus, and Coma) in the local universe (within 100 Mpc) where neutrinos are likely to be most clustered. ",
        "conclusions": "We have introduced and tested a method based on the collisionless Boltzmann equation to calculate the gravitational clustering of massive neutrinos in CDM halos for realistic cosmological models. This method is valid for currently favored models with $\\Omega_{\\rm cdm} \\gg \\Omega_\\nu$ in which the clustering of neutrinos is mostly determined by the existing CDM halos while the clustering of the CDM is little affected by the neutrinos.  One can then obtain the neutrino phase space distribution by solving the collisionless Boltzmann equation in a background potential given by the universal profile of CDM halos from high resolution simulations.  The resulting Boltzmann equation is linear in the neutrino density contrast and has tractable intergral solutions that require negligible computational time in comparison with N-body simulations.  This method has enabled us to obtain specific predictions for the neutrino overdensity as a function of halo radius, halo mass, and neutrino mass for a wide range of parameters.  Our calculation shows that neutrinos with masses $\\agt$ 0.3 eV can cluster appreciably in CDM potential wells with masses $\\agt 10^{13} M_{\\odot}$.  The predicted neutrino overdensity increases with both the neutrino mass and the halo mass, ranging from $\\sim 10$ for 0.3 eV neutrinos in $\\sim 10^{13} M_\\odot$ halos to $\\sim 1500$ for 1.8 eV neutrinos in $\\sim 10^{15} M_\\odot$. Specific predictions are plotted in Figs.~2 and 3.  Neutrino clustering has a strong impact on the Z-burst model that has been proposed as a possible explanation for the UHECR events.  The predicted UHECR spectrum shown in Figs.~5 and 6 depends sensitively on the neutrino mass and overdensity, showing distinct spectral features towards nearby galaxy clusters if $m_\\nu\\agt 0.3$~eV.  To illustrate the effects of neutrino mass and overdensity on the UHECR spectrum, we have chosen to normalize the flux in Figs.~5 and 6 with the same value (i.e. $F_{\\nu_{i}}(E_{\\nu_{i}}^{res})= 1.7 \\times 10^{-35}$ (eV m$^2$ s sr)$^{-1}$ for each flavor for the three degenerate mass models and three times higher for the one massive species model) instead of adjusting it by fitting individual spectrum to existing data.  We have nonetheless included current data from the AGASA \\cite{AG} and HiRes \\cite{HR} experiments in Figures~5 and 6 for comparison.  More events are needed to discriminate the different models and the directional dependence.  The large increase in flux towards Virgo is an interesting signature of the Z-burst model for upcoming experiments such as Auger \\cite{AUG} and OWL \\cite{OWL99} that will provide an angular resolution of $\\sim$ 1$^{\\circ}$. Experimental limits on the anisotropy would in turn imply small neutrino inhomogeneities in the Z-burst model and can be used to place upper bounds on the neutrino mass.  A useful constraint on the Z-burst model is provided by the Energetic Gamma Ray Experiment Telescope (EGRET) measurement of the GeV $\\gamma$-ray background flux, which must not be exceeded by the high energy photons produced in the Z-burst models once they cascade down to the GeV energy range.  The result depends on the assumed redshift evolution of the sources that produce the incident high energy neutrinos, and on whether the sources themselves produce photons. The normalization of the neutrino flux cited in the previous paragraph rules out sources emitting a comparable flux in $\\gamma$-rays because it leads to a conflict with the existing EGRET limits for the GeV $\\gamma$-rays. For pure neutrino sources, calculations based on particle transport codes show that for neutrino masses of 0.1, 0.5, and 1 eV (ignoring neutrino clustering), the EGRET bound is met for $\\alpha\\lo -3$, $\\lo 0$, and $\\lo 3$, respectively, where the source number density evolves as $(1+z)^\\alpha$ \\cite{FKR02,KKSS02}. When neutrino clustering is taken into account, our results from Fig.~2 show that the bound above for $m_\\nu \\lo 0.3$ eV should be unaffected since they do not cluster appreciably in the Local Group.  For larger neutrinos masses, however, we expect a less stringent bound on the source evolution due to local neutrino clustering.  To derive quantitative constraints would require detailed transport calculations.  The implications of the neutrino clustering results presented in this paper extend beyond the problem of the UHECR spectrum.  For UHECR, upcoming experimental results may converge on a spectrum that is consistent with the GZK cutoff and would therefore not require models such as the Z-burst.  It is also likely that the Z-burst model is not the correct explanation for the UHECR events.  However, the neutrino-anti-neutrino resonance scattering process remains one of few ways to detect the relic neutrinos, as first suggested in Ref.~\\cite{W82}.  This paper has addressed neutrino clustering, a major uncertainty in all studies concerning relic neutrinos."
    },
    "0208/astro-ph0208243_arXiv.txt": {
        "abstract": "We have measured the cosmic ray spectrum above $10^{17.2}$~eV using the two air fluorescence detectors of the High Resolution Fly's Eye observatory operating in monocular mode.  We describe the detector, photo-tube and atmospheric calibrations, as well as the analysis techniques for the two detectors.  We fit the spectrum to a model consisting of galactic and extra-galactic sources. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0208/astro-ph0208239_arXiv.txt": {
        "abstract": "The recent detection of a 3-hr X-ray flare by the Chandra Observatory has raised the possibility of enhanced emission over a broad range of wavelengths from Sgr A*, the suspected 2.6 x 10$^{6}$ \\msun black hole at the Galactic Center, during a flaring event. We have, therefore, reconstructed 3-hr data sets from 2\\micron ~speckle and adaptive optics images ($\\theta_{core}$ = 50 - 100 mas) obtained with the W. M. Keck 10-m telescopes between 1995 and 2001. In 25 separate observations, no evidence of any significant excess emission associated with Sgr A* was detected. The lowest of our detection limits gives an observed limit for the quiescent state of Sgr A* of 0.09$\\pm$0.005 mJy, or, equivalently, a dereddened value of 2.0$\\pm$0.1 mJy, which is a factor of 2 lower than the best previously published quiescent value. Under the assumption that there are random 3-hr flares producing both enhanced X-ray and near-infrared emission, our highest limit constrains the variable state of Sgr A* to $\\lesssim$0.8 mJy (observed) or 19 mJy (dereddened).  These results suggest that the model favored by \\citet{mark01}, in which the flare is produced through local heating of relativistic particles surrounding Sgr A* (e.g., a sudden magnetic reconnection event), is unlikely, because it predicts peak 2\\micron ~emission of $\\sim$300 mJy, well above our detection limit. ",
        "introduction": "\\label{intro}  The variability of Sagittarius (Sgr) A* at X-ray wavelengths \\citep{bag01a} has bolstered the case for associating this source with the suspected 2.6 x 10$^{6}$ \\msun black hole at the center of our Galaxy \\citep{eckart97, genz97, genz00, ghez98, ghez00}. In two Chandra observations separated by almost a year and having a total of 76 ksec of exposure time, Sgr A* was detected at X-ray wavelengths for the first time and was also seen to flare in intensity over a time scale of 3 hours \\citep{bag01a,bag01b}. While the flare's short duration implied a small region of origin, $\\lesssim$400 R$_{s}$ (where $R_{s}$ is the Schwarzschild radius = 2GM$_{\\bullet}$/c$^{2}$), its large amplitude, a factor of 50, has raised the possibility of detecting corresponding intensity enhancements at wavelengths outside the X-ray regime.  Existing models for Sgr A*'s flared state make very disparate predictions for the emission at wavelengths between the X-ray and radio regimes \\citep{mark01,liu02}. The wide differences between these models are a result of assuming different geometries (disk vs. jet) and emission mechanisms for the flaring process (e.g., enhanced accretion rates vs. magnetic reconnection). In some models, the predicted emission in the flared state, at infrared (IR) wavelengths, dramatically exceeds that of existing detection limits \\citep{genz99, stol99, morris01}. For example, the preferred model of \\citet{mark01} predicts an observed 2\\micron ~flux density of $\\sim$13 mJy, or a dereddened flux density of $\\sim$300 mJy, during the flared state. Unlike the situation at radio wavelengths, where long-term monitoring campaigns have been used to constrain the flared state of Sgr A* \\citep{zhao01}, the limited time coverage and spatial resolution of published IR experiments prevent meaningful constraints on the flared state's IR emission from being inferred\\footnote{While several papers have reported the possible detection of a variable near-infrared source coincident with Sgr A* \\citep{herbst93,close95,genz97}, subsequent high resolution observations have identified this emission to be from high proper motion sources \\citep{eckart95, eckart97, ghez98, ghez02}.} and, thus, the reported limits are assumed to be associated with Sgr A*'s quiescent state.  The W. M. Keck Observatory dynamical study of stars in the central stellar cluster \\citep{ghez98,ghez00,suvi02} provides a rich source of high angular resolution 2\\micron ~data between 1995 and 2001. In this paper, we present 2\\micron ~flux density limits from maps that were each composed of data from a single night. The elapsed time of 3-4 hours in each map is approximately the same as the time scale of the observed X-ray flare, making this data set ideally suited for possibly detecting a flare of this type. Given the quantity of these observations, our non-detections establish a robust upper limit for the flared state's 2\\micron ~emission intensity. ",
        "conclusions": "This paper summarizes a search for a near-infrared counterpart to Sgr A* in the flared state. From the length of our observations, this search was sensitive to variability on time scales of 3 hours. No such counterpart was detected. However, by identifying and removing all the stars in the crowded inner $\\sim$0\\farcs6$\\times$0\\farcs6 of the Galactic Center, an upper limit for the emission from Sgr A* has been inferred for each observation epoch. These limits constrain the quiescent emission from Sgr A* to $\\lesssim$0.09 mJy (2.0 mJy, dereddened) and the variable component to $\\lesssim$0.8 mJy (19 mJy, dereddened) at the 2$\\sigma$ confidence level.  More X-ray data will improve the estimated flaring duty cycle and is likely to increase the confidence level of our limits for the near-infrared component of the flared emission."
    },
    "0208/astro-ph0208525_arXiv.txt": {
        "abstract": "Lithium abundances in the atmospheres of the super Li-rich C-giants WZ Cas and WX Cyg are derived by the spectral synthesis technique {\\bf using} the Li I resonance line at \\la670.8 nm and {\\bf three subordinate} lines at \\lala 812.6, 610.4 and 497.2 nm. The differences between the Li abundances derived from the \\la670.8 nm line and the \\lala 497.2, 812.6 nm lines do not exceed $\\pm$ 0.5 dex. The lithium line at \\la610.4 nm provides typically lower abundances than the resonance line (by $\\approx$ 1 dex). The mean LTE and NLTE Li abundances from three Li I lines (excluding \\la610.4 nm) are 4.7, 4.9 for WZ Cas, and  4.6, 4.8 for WX Cyg, respectively. ",
        "introduction": "{\\bf The knowledge of lithium} abundances in stellar atmospheres {\\bf is important because it provides} information about physical processes in stars and stellar evolution. {\\bf Some asymptotic giant branch (AGB) carbon stars (C/O$>$1, by number) are observed to be lithium rich with respect to the solar value of \\nli = 1.2 (in scale log N(H)=12). WZ Cas and WX Cyg are well known super Li rich (SLR, \\nli $\\geq$ 4) giant carbon stars.} {\\bf AGB SLR} stars{\\bf,} with {\\bf their strong} mass loss{\\bf,} could be the main suppliers of lithium in {\\bf the} Galaxy (cf. Abia et al. 1993, Wallerstein \\& Knapp 1998). {\\bf Furthermore, the determination of Li abundances in the atmospheres of these stars has also cosmological interest.}  Currently, the overwhelming majority of determinations of the lithium abundances {\\bf in carbon} stars are based on the analysis of the resonance Li I line \\la670.8 {\\bf nm only}. However, spectra of {\\bf Li rich} stars contain {\\bf several Li lines that are possible probes of the Li abundance.} {\\bf In Yakovina} \\& Pavlenko (2001), six subordinate Li I lines in the region \\lala 400-820 nm were analysed. {\\bf The present paper examines which of them can be used as Li abundance indicators in carbon stars, and the ranges of Li abundances over which they work.} This {\\bf paper} is an extension of {\\bf that} by Abia et al. (1999), in which the formation of the Li I lines in atmospheres of SLR AGB stars {\\bf was} analysed.  We use {\\bf echelle-}spectra of WZ Cas (C9.2J) and WX Cyg (C8.2eJ) obtained in 1997--1999 with the 4.2 m WHT (Roque de los Muchachos observatory) and 2.2 m telescope (Calar Alto observatory; see Abia et al. (1999) for more details). {\\bf The spectral resolution was $\\lambda$/$\\Delta\\lambda$ $\\approx$ 50000 and 35000, respectively.} ",
        "conclusions": "Our results show that the differences in {\\bf the} lithium abundances {\\bf estimated} from subordinate lines Li I \\lala 497.2 and 812.6 nm and from {\\bf the} resonance line Li I \\la670.8 nm are of the same order {\\bf: $\\pm$ 0.5 dex. This is similar to} the possible errors in \\nli due to uncertainties in the atmosphere parameters, model atmospheres and continuum level (see Denn et al. 1991, Abia et al. 1991, Plez et al. 1993, Abia et al. 1999).  The underestimations of {\\bf the Li abundance from the $\\lambda$ 610.4 nm line might} be explained by the bad fits to the observed spectra in this region. On the other hand, as it was already mentioned, the saturated Li I line \\la610.4 nm shows a rather weak dependence on log N(Li).  Finally, we conclude that the resonance \\la670.8 nm and subordinate \\la497.2 nm Li I lines are the best lithium abundance indicators in atmospheres of SLR carbon stars.  Our lithium abundances for WZ Cas and WX Cyg (Table 3) agree well with results of Abia et al. (1991) based on the resonance Li I line. {\\bf Their Li estimations} are higher by 0.2--0.3 dex, probably due to the incompleteness of the line lists used. Lithium abundances {\\bf by} Abia et al. (1999) agree well with our results for WZ Cas. For WX Cyg our estimations are higher by 0.7--0.8 dex. We believe that this difference is due to the different location of continuum level."
    },
    "0208/astro-ph0208149_arXiv.txt": {
        "abstract": "Planetary nebulae are now well established as probes of galaxy dynamics and as standard candles in distance determinations.  Motivated by the need to improve the efficiency of planetary nebulae searches and the speed with which their radial velocities are determined, a dedicated instrument - the Planetary Nebulae Spectrograph or \\ePNS\\ - has been designed and commissioned at the 4.2m William Herschel Telescope.  The high optical efficiency of the spectrograph results in the detection of typically $\\sim 150$ \\PNe\\ in galaxies at the distance of the Virgo cluster in one night of observations.  In the {\\em same observation} the radial velocities are obtained with an accuracy of $\\sim 20$~\\kms \\\\ {\\bf note that due to archival restrictions the figures have been strongly compressed - please contact any of the authors for a better preprint} ",
        "introduction": "\\label{intro}  The study of the internal dynamics of galaxies provides some of the best observational clues to their formation history and mass distribution, including dark matter, but much of the interesting information is inaccessible with conventional techniques.  Stellar kinematics, the most important diagnostic, have generally been determined using absorption-line spectra of the integrated light, the surface brightness of which is such that only the inner parts of the galaxy can be observed in a reasonable amount of telescope time.  In practical terms, it is hard to measure the integrated stellar spectra beyond $1-2 R_e$ (the ``effective radius'', which contains half the galaxy's projected light). This is a serious limitation, because it is at larger radii that the gravitation of the dark matter halo is likely to dominate, and that the imprint of the galaxy's origins are likely to be found in its stellar orbits.  Alternative diagnostics of the gravitational potential include measurement of the 21cm-wavelength emission from neutral hydrogen, observations of \\Htwo\\ regions, and the observation of the motions of globular clusters and planetary nebulae.  But neutral hydrogen and \\Htwo\\ regions are effectively absent from early- type galaxies, and globular clusters, not surprisingly, have been found to be kinematically distinct from the stellar population which is of primary interest.  Planetary nebulae (\\PNe) are part of the post-main-sequence evolution of most stars with masses in the range 0.8--8 M$_{\\odot}$. Taking into account the duration of the PN phase itself, even in a galaxy with continuing star formation most {\\em observed} \\PN\\ will have progenitors in the range 1.5--2~M$_{\\odot}$, corresponding to a mean age of $\\sim 1.5$~Gyr. In the case of early-type galaxies the \\PN\\ are drawn from the same old population that comprises most of the galaxy light.  \\PNe\\ are sufficiently bright to be detected in quite distant galaxies and their radial velocities are readily measured by a variety of techniques. Moreover, because they are easier to detect at large galactocentric radius where the background continuum is fainter, they represent the crucially important complement to absorption line studies. These properties make \\PNe\\ an ideal kinematic tracer for the outer parts of such galaxies, allowing the measurement of stellar kinematics to be extended out to typically $4-5 R_e$.  A common and well-tested technique for obtaining the radial velocities of \\PN\\ consists of a narrow-band imaging survey of [OIII] emission to identify candidates, followed by a spectroscopic campaign to obtain their spectra. This approach has in the past commonly resulted in low yields (see \\S\\ref{mos}).   We have developed an alternative single-stage method, which uses narrow-band slitless spectroscopy.  By obtaining two sets of data with the spectra dispersed in different directions (a technique we call ``counter-dispersed imaging'' or CDI), one can identify \\PN\\ {\\em and} measure their radial velocities in a single observation. The use of CDI by modifying existing instrumentation has been so successful that we decided to design and build a custom-made CDI spectrograph, with an overall efficiency improvement of about a factor of ten over a typical general-purpose spectrograph.  The organisation of this paper is as follows. In \\S\\ref{obs} we review the important observational characteristics of \\PN\\ and discuss the implications of their luminosity function for  the number of \\PN\\ which can be detected.   In \\S\\ref{ot} we introduce CDI and compare it with more traditional techniques. The PN.Spectrograph, which is the main subject of this paper, is presented in some detail in \\S\\ref{project}. As well as a complete technical discussion, and a short history of the project, this section includes some of the first images  obtained with the instrument, following its recent successful commissioning. Further technical information is presented in three appendices. ",
        "conclusions": "\\label{conclusion} We have constructed an instrument which can efficiently detect extragalactic \\PN\\ and measure their velocities.  Using a novel application of slitless spectroscopy, the data can be obtained in a single observation with a single instrument, rather than the more complex procedures previously necessary.  We have presented design information for the highly optimised, dedicated instrument now in use at La Palma, and first results have been presented.  In effect the telescope efficiency can be nearly doubled by this approach, opening up the prospect of routine observations of galaxy kinematics at a distance of up to 25~Mpc with 4m-class telescopes. The technique is applicable to larger telescopes."
    },
    "0208/astro-ph0208463_arXiv.txt": {
        "abstract": "We have analyzed a \\textit{Chandra}/HETGS spectrum of the Galactic black hole Cygnus~X-1, obtained at a source flux which is approximately twice that commonly observed in its persistent low-intensity, spectrally-hard state.  We find a myriad of absorption lines in the spectrum, including Ly-$\\alpha$ lines and helium-like resonance lines from Ne, Na, Mg, and Si.  We calculate a flux-weighted mean red-shift of $\\simeq 100$~km/s and a flux-weighted mean velocity width of $\\simeq 800$~km/s (FWHM) for lines from these elements.  We also detect a number of transitions from Fe~XVIII--XXIV and Ni~XIX--XX in absorption; however, the identification of these lines is less certain and a greater range of shifts and breadth is measured.  Our observation occurred at a binary phase of $\\phi \\simeq 0.76$; the lines observed are consistent with absorption in an ionized region of the supergiant O9.7~Iab companion wind.  The spectrum is extremely complicated in that a range of temperatures and densities are implied. Prior \\textit{Chandra}/HETGS spectra of Cygnus~X-1 were obtained in a similar transition state (at $\\phi \\simeq 0.93$) and in the low/hard state (at $\\phi \\simeq 0.84$).  Considered together, these spectra provide evidence for a companion wind that is focused as it flows onto the black hole primary in this system. ",
        "introduction": "Cygnus~X-1 is the only known Galactic black hole which has displayed persistent X-ray activity since the dawn of X-ray astronomy.  It is also a rare black hole binary in that it has a high-mass companion star (HDE~226868 --- an O9.7~Iab supergiant; Gies \\& Bolton 1982, 1986).  The mass of the black hole primary is most likely $M_{1} \\simeq 10.1~M_{\\odot}$, and the mass of the companion secondary is most likely $M_{2} \\simeq 17.8~M_{\\odot}$ (Herrero et al. 1996).  At present, there are 14 Galactic systems with low-mass companions ($M_{2} \\sim 1~M_{\\odot}$) for which optical radial velocity curves imply a primary with $M_{1} \\geq 3~M_{\\odot}$ (the theoretical upper-limit for a neutron star mass; for a list of dynamically-constrained low-mass black hole systems see http://www.astro.uu.nl/$\\sim$orosz/).  These systems are transients, undergoing outbursts in which the X-ray luminosity may change by factors of $10^{6}$ or more on scales ranging from days to months.  It is logical, then, to associate the persistent nature of Cygnus~X-1 with the fact that it is a high-mass X-ray binary (HMXB).  While the companion wind is very likely to play an important accretion role in Cygnus~X-1 --- perhaps through a ``focused wind'' scenario (Friend \\& Castor 1982, Gies \\& Bolton 1986; see below) --- numerous detections of the disk's thermal spectrum, strong and broad Fe~K$\\alpha$ emission lines, and strong disk reflection in Cygnus~X-1 clearly demonstrate that standard disk accretion is also important in this system.  The relative importance of the wind and disk in driving the accretion process and X-ray ``states'' in Cygnus~X-1 is still uncertain (for a review of states, see Done et al. 2002; for a critical discussion see Homan et al. 2001).  Winds from isolated massive stars may be approximated as being spherically symmetric.  Friend \\& Castor (1982) explored the geometry of winds from massive stars in the presence of a compact object.  Due to the compact object's gravity, continuum radiation pressure, and centrifugal forces from binary orbital motion, it was found that winds are likely to be strongly asymmetric if the mass donor is close to filling its critical Roche surface.  The region of highest mass flux is likely to be along the axis connecting the compact object and mass donor components.  Optical spectroscopy of HDE~226868 (Cygnus X-1) later revealed that modulations of the He II $\\lambda$4686 emission line could be described by this geometry, and that the bulk of this emission may come from a cone ($\\theta < 20^{\\circ}$) along the connecting axis (Gies \\& Bolton 1986).  The sensitivity and resolution of previous X-ray observatories has been insufficient to directly probe this focused wind geometry. However, matters have improved greatly with the \\textit{Chandra} High Energy Transmission Grating Spectrometer (HETGS).  Marshall et al. (2001a) discussed the detection of several highly ionized absorption lines in a 15~ksec \\textit{Chandra}/HETGS observation made in the persistent low-luminosity, hard spectrum state (at a binary phase of $\\phi \\simeq 0.84$, defining $\\phi=0$ as the point at which the companion is closest along our line of sight and based on the ephemeris of La~Sala et al. 1998) commonly observed in Cygnus X-1.  It is natural to associate the observed absorption features with the companion wind as the same lines have been observed in emission in a number of HMXBs with neutron star primaries.  Schulz et al. (2002) report the detection of numerous emission and absorption features, and P-Cygni-type line profiles, in a prior 15~ksec \\textit{Chandra}/HETGS observation of Cygnus X-1 (at $\\phi \\simeq 0.93$) made in a state wherein the flux was approximately twice that reported by Marshall et al. (2001a).  We observed Cygnus~X-1 with the \\textit{Chandra}/HETGS for 32.1~ksec on January 4, 2001, at a flux which was approximately twice that commonly observed in the ``low/hard'' spectral state (we likely observed the source in an ``intermediate'' state).  The observation occurred at a binary phase of $\\phi \\simeq 0.76$ in the 5.6-day orbital period.  The broad-band spectrum obtained in this observation is discussed by Miller et al. (2002).  A composite Fe~K$\\alpha$ emission line was revealed --- the first such composite line clearly resolved in a binary black hole system.  The broad component may be shaped by Doppler shifts and strong gravitational effects at the inner edge of an accretion disk extending close to the marginally stable circular orbit, while the neutral, narrow component may be produced in the outer accretion disk.  Herein, we report the results of our analysis of the time-averaged high resolution spectrum in the 6--24~\\AA~ (0.5--2.0~keV) band.  While some of the lines we detect were also found by Marshall et al. (2001a) and Schulz et al. (2002), in this paper we present the first attempt to systematically fit a high-resolution spectrum of Cygnus X-1 and to measure the parameters of the strongest absorption and emission lines in this band.  Moreover, it is clear that the spectrum we have observed differs considerably from those reported earlier, likely indicating that the appearance of the wind changes with orbital phase and the intensity of the source. ",
        "conclusions": "We have performed the first systematic fits to the line-rich low-energy spectrum of Cygnus X-1.  The temperatures, velocities, and densities implied by the weak, narrow lines in the spectrum are consistent with absorption in an ionized portion of the companion wind.  The spectrum we have observed from Cygnus X-1 is dominated by absorption lines, however, which stands in strong contrast to the spectra observed from other HMXBs.  \\textit{ASCA} and \\textit{Chandra} observations of Vela X-1 (Sako et al. 1999, Schulz et al. 2002b) and Cen X-3 and (Wojdowski, Liedahl, \\& Sako 2001) reveal emission line spectra due to recombination in a photoionized companion wind (the emission lines are most prominent in eclipse, but persist outside of eclipse).  A recent \\textit{Chandra} observation of Cygnus X-3 revealed many of the same strong emission lines independent of orbital phase (Paerels et al. 2000).  While the spectrum of Cen X-3 can be modeled in terms of a spherically-symmetric wind centered on the neutron star (Wojdowski, Liedahl, \\& Sako 2001), the absorption spectrum of Cygnus X-1 may require dense material preferentially along the line of sight.  A relative lack of material outside of the line of sight is also required to explain the absence of resonance emission lines excited by photoionization from the central accretion engine.  The inclination of Cygnus X-1 is rather low ($\\theta \\simeq 35^{\\circ}$; Gies \\& Bolton 1986), so it is unlikely that the absorption region can be associated with the outer accretion disk or a warm (few keV) disk atmosphere.  As optical observations of Cygnus X-1 (Gies \\& Bolton 1986) have already provided evidence for a focused wind geometry, we suggest that such a geometry might also explain the X-ray spectrum we have observed.  In the discussion below, we first examine the evidence for a focused wind from this and other recent \\textit{Chandra} observations of Cygnus X-1.  We then address the implications of the neutral absorption features and the impact of this observation on our understanding of Cygnus X-1.  \\subsection{Evidence for a Focused Wind Geometry} Kallman \\& Bautista (2001) have calculated ionization fractions and ionization parameters for a gas with properties consistent with stellar winds, for an incident spectrum similar to that measured here (Miller et al. 2002).  The ionization parameters consistent with the observed absorption spectrum (${\\rm log}(\\xi) = 2-3$) imply distances between the source and absorbing gas ($5-9 \\times 10^{10}$~cm) that are far less than the distance between the black hole and the companion surface ($\\sim 1.4 \\times 10^{12}$~cm; LaSala et al.\\ 1998). Moreover, the estimates we obtain for the equivalent neutral hydrogen column densities for various elements generally lie well below expectations for a spherically symmetric wind (assuming that our line of sight through the wind is comparable to the separation between the black hole and companion surface).  An examination of the velocities implied by the absorption lines also supports a focused wind geometry.  For the Ly-$\\alpha$ and helium-like resonance lines from Ne, Mg, Na, and Si, we measure a flux-weighted mean red-shift of $\\simeq$100~km/s and flux-weighted mean FWHM of $\\simeq$800~km/s (again, our observation occurred at $\\phi \\simeq 0.76$).  Marshall et al. (2001a) report a number of absorption lines with a mean red-shift of $\\sim$450~km/s and a typical FWHM of $\\sim$300~km/s at $\\phi \\simeq 0.84$.  This is consistent with the expectation that the a focused flow should be largely transverse to the line of sight at intermediate phase points, and have a greater component along the line of sight at other phases.  Schulz et al. (2002) report marginal evidence for ionized Fe transitions with P-Cygni-type line profiles at $\\phi = 0.93$.  This may indicate that the wind from the companion surface opposite to the X-ray source may have a significant radial component.  The absence of such line profiles at more intermediate phases (this work; Marshall et al. 2001a) suggests that the wind does not have a radial component of typical strength at phases away from conjunction.  We suggest that when considered together, these spectra may provide the first direct evidence in X-rays for a focused wind geometry in Cygnus X-1 (see Figure 3 for a possible geometry).  The column densities measured from absorption lines are smaller than expected for a spherically symmetric wind geometry.  At an intermediate phase point the flow is mostly transverse to the line of sight; at points closer to superior conjunction, the flow has smaller transverse velocities and higher velocities parallel along the line of sight.  Finally, very close to superior conjunction, the wind from the face of the companion opposite to the X-ray source may have a radial component; indeed P-Cygni profiles may have been observed by Schulz et al. (2002a).  A caveat is that the observation reported on by Marshall et al. (2001a) occurred in the low/hard state typically observed in Cygnus X-1, while this observation and that discussed by Schulz et al. (2002a) were made during transitional states with higher fluxes. At present, it is not clear how the source state and wind are related. Moreover, all three observations were separated by several months, and the nature of the wind may vary with the long-term period in this system (e.g., Brocksopp et al. 1999).  The observation of narrow, blue-shifted absorption lines from low-Z elements near to inferior conjunction and broadened lines near $\\phi = 0.25$ would strengthen the evidence for a focused wind accretion geometry.  Future observations with the \\textit{Chandra}/HETGS (or \\textit{XMM-Newton}/RGS) achieving sensitivities similar to those observations considered here will be required for this purpose.  \\subsection{Neutral Absorption Edges, and the Atomic O $1s-2p$ Absorption Line}  We measure absorption edge depths from neutral atoms in the interstellar medium that are consistent with those reported by Schulz et al. (2002).  This suggests that any cold absorbing gas intrinsic to Cygnus X-1 does not vary with orbital phase.  Although the atomic O $1s-2p$ line is saturated (and its profile therefore distorted) in both observations, the width we measure (FWHM$ = 0.18 \\pm 0.04$\\AA) is roughly twice the width measured previously.  The width of the O $1s-2p$ lines measured in observations of Cygnus X-1 are factors of 10--20 higher than the width of the same line measured in a \\textit{Chandra}/LETGS spectrum of the neutron star 4U~0614$+$091 (Paerels et al. 2001).  Although the lines in Cygnus X-1 are saturated, this disparity is likely larger than can be attributed solely to distortion of the line Cygnus X-1 profile.  It is possible that the variation can be accounted for by different turbulent velocities in the neutral oxygen intrinsic to each system.  However, due to the ionizing X-ray flux from accretion onto the compact object in these systems, little neutral oxygen is expected locally.  We suggest that the atomic O $1s-2p$ line, then, may serve as a probe of turbulent velocities in the ISM along different lines of sight. Studies of the Galactic H~I power spectrum which suggest that column density variations due to cold gas may be dominated by localized velocity field fluctuations (Dickey et al. 2001).  At present, this is a relatively unexplored regime in X-ray astronomy but well within the reach of current observatories.  \\subsection{Understanding the Accretion Geometry in Cygnus X-1} Future analysis and observations will achieve a better understanding of the degree to which the wind in Cygnus X-1 may be focused, and its temperature and density characteristics.  In the near future, then, it may be possible to constrain the mass accretion rate onto the black hole via focused wind accretion.  Comparisons to the observed X-ray luminosity will then indirectly provide a constraint on the mass delivered to the black hole via disk accretion.  Constraints of this kind are important for a number of reasons. Cygnus X-1 often serves as a standard for testing models for accretion flow geometries and state transitions based on the mass accretion rate (such as advection-dominated accretion flow models, or ``ADAFs''; see, e.g., Esin et al. 1998).  This source is also a testbed for X-ray ``reflection'' models (see, e.g., Ross, Fabian, \\& Young 1999), which suggest that the innermost accretion flow geometry in Cygnus X-1 may be very similar to that in some AGN.  In both cases, a central source of hard X-rays (a ``corona'') is important, as is an accretion disk (some reflection models suggest that an ionized transition layer may lie on top of the accretion disk).  Neither family of models has considered the role of the companion wind, or the nature of disk-wind or corona-wind interactions.  Although the timing properties of Cygnus X-1 are well-studied (see, e.g., Pottschmidt et al. 2002), quasi-periodic oscillations (QPOs) have never been observed at high frequencies (``high'' is somewhat arbitrary, $\\nu > 30~{\\rm Hz}$ is a reasonable distinction).  Such QPOs have been very useful in constraining black hole spin in systems such as GRO~J1655$-$40 (Strohmayer 2001); in contrast spin can only be investigated in Cygnus X-1 via spectroscopy (e.g., Miller et al. 2002).  It is possible that the companion wind in Cygnus X-1 acts to dampen high-frequency signals preferentially; when a better understanding of the wind geometry and density is achieved this possibility can be addressed.  At present, Cygnus X-1 is the only Galactic HMXB for which a black hole primary is required.  LS~5039 may provide an interesting comparison: the companion is an O6.5 V(f) supergiant (which does not fill its critical Roche radius), the orbital period is 4.1 days, and its distance is likely $\\sim$3.1~kpc (McSwain \\& Gies 2002).  Our preliminary analysis of archival \\textit{ASCA} data suggests that LS~5039 is very weak in X-rays relative to Cygnus X-1: $L_{X} < 2 \\times 10^{34}~ {\\rm erg}~ {\\rm s}^{-1}$ (0.5--10.0 keV).  This may underscore the importance of disk accretion via Roche-lobe overflow in Cygnus X-1.  At present, the nature of the compact object in LS~5039 is not known.  Even if the primary in LS~5039 is a neutron star, this system bears important similarities to Cygnus X-1, and we look forward to spectroscopy of this source with \\textit{Chandra} and \\textit{XMM-Newton}."
    },
    "0208/astro-ph0208180_arXiv.txt": {
        "abstract": "{ We present three-frequency VLA observations of the flocculent spiral galaxy NGC~4414 made in order to study the magnetic field structure in absence of strong density wave flows. NGC 4414 shows a regular spiral pattern of observed polarization B-vectors with a radial component comparable in strength to the azimuthal one. The average pitch angle of the magnetic field is about 20$\\degr$, similar to galaxies with a well-defined spiral pattern. This provides support for field generation by a turbulent dynamo without significant ``contamination'' from streaming motions in spiral arms. While the stellar light is very axisymmetric, the magnetic field structure shows a clear asymmetry with a stronger regular field and a smaller magnetic pitch angle in the northern disk. Extremely strong Faraday rotation is measured in the southern part of the disk, becoming Faraday thick at 6\\,cm. The distribution of Faraday rotation suggests a mixture of axisymmetric and higher-mode magnetic fields. The strong Faraday effects in the southern region suggest a much thicker magnetoionic disk and a higher content of diffuse ionized gas than in the northern disk portion. An elongation of the 20\\,cm total power emission is also seen towards the South. Although NGC 4414 is currently an isolated spiral, the asymmetries in the polarized radio emission may be sensitive tracers of previous encounters, including weak interactions which would chiefly affect the diffuse gas component without generating obvious long-term perturbations in the optical structure. ",
        "introduction": "The relative role of small-scale velocity perturbations and galaxy-scale gas flows in determining the evolution and structure of galactic magnetic fields is one of the most hotly debated questions in studies of the evolution and structure of galactic magnetic fields (e.g. Zweibel~\\cite{zwi96}, Zweibel \\&~Heiles~\\cite{zwi97}, Beck et al.~\\cite{bec96}). Large-scale galactic magnetic fields are known to display a coherent spiral-like pattern of polarization B-vectors (see Beck et al.~\\cite{bec96} for a review), indicating a substantial radial component capable of resisting the shear due to differential rotation. The spiral magnetic field pattern could be produced by the dynamo mechanism (Wielebinski \\&~Krause~\\cite{wie93}, Beck et al.~\\cite{bec96}) in which the small-scale motions are constantly feeding the large-scale poloidal (i.e. radial and vertical) field. The radial field could also be produced by continuous field stretching by large-scale flows due to density waves or bars (e.g. Otmianowska-Mazur \\&~Chiba~\\cite{otm95}). In choosing a flocculent spiral, where flocculent means without large-scale spiral structure, density waves and bars should play no significant role in determining the magnetic field orientation.  In well-studied nearby spiral galaxies it is extremely difficult to discern the effects of the turbulent dynamo from those due to processes in spiral arms. These objects have well-developed grand-design patterns with strong density wave compression and/or a high concentration of star formation in spiral arms. The first process may strongly modify the magnetic field by effects of compression and gas flows along the arms (Otmianowska-Mazur \\&~Chiba~\\cite{otm95}). In some galaxies like M83 spiral-like compression regions (as traced by aligned dust filaments) fill the whole interarm space. Strong star formation in spiral arms acts destructively on regular fields (Beck et al.~\\cite{bec96}). Differences in turbulent activity between the arms and the interarm region may give rise to strong concentration of regular magnetic fields between the stellar arms, thereby strongly influencing the global field structure (Rohde \\&~Elstner~\\cite{roh98}).  Knapik et al.~(\\cite{kna00}) first reported that flocculent galaxies may possess regular fields with a strong radial component. Their study was made at low resolution and only at one frequency. In this paper we present a high-resolution multifrequency polarization study of the flocculent galaxy NGC~4414 (Thornley \\&~Mundy~\\cite{tho97}). The velocity field shows no evidence for non-circular gas flows (Braine et al.~\\cite{bra93}, Sakamoto~\\cite{sak96}, Thornley \\&~Mundy~\\cite{tho97}). NGC~4414 has a high surface gas density and is forming stars fairly intensively, much like in our own galaxy, but is in no way a starburst. All this ensures good conditions for building up a global magnetic field by turbulent processes in a way unaffected by large-scale gas flows and compressions.  \\begin{table}[th] \\caption {Basic properties of NGC~4414} \\begin{flushleft} \\begin{tabular}{lll} \\hline Other names & PGC~40692 & Reference\\\\ & UGC~7539 & \\\\ R.A.$_{1950}$ &$\\rm 12^h23^m57\\fs 8$& de Vaucouleurs et al.~(\\cite{vau91})\\\\ Decl.$_{1950}$ & $31\\degr 29\\arcmin 58\\farcs 0$ & de Vaucouleurs et al.~(\\cite{vau91})\\\\ Inclination & 55$\\degr$ & LEDA\\\\ Position Angle & 155$\\degr$ & LEDA\\\\ Distance & 19.2 Mpc & Thim~(\\cite{thi00})\\\\ Morphol. Type & Sc & LEDA\\\\ \\hline \\end{tabular} \\end{flushleft} \\end{table} ",
        "conclusions": "We have performed a three-frequency VLA study of the flocculent galaxy NGC~4414 known to have extremely weak traces of optical spiral structure and no evidence for non-axisymmetric gas flows. The data were analyzed together with CO(1--0) and H$\\alpha$ maps, yielding the following results:  \\begin{itemize}  \\item[-] The galaxy shows a bright total power disk with a central depression corresponding to the minimum in H$\\alpha$ and CO brightness, which may serve as an additional argument for the absence of radial gas flows.  \\item[-] The correlation of the nonthermal brightness with the CO and H$\\alpha$ emission also holds in a galaxy without strong spiral arms and thus it is not due to an accumulation of cold gas and star formation in density-wave compression regions.  \\item[-] Despite the lack of spiral arms and of non-azimuthal gas flows the galaxy shows a clear magnetic spiral pattern with a significant radial component, resembling that in grand-design galaxies. Dynamo action is the most likely source of a significant radial magnetic field component. Admixtures of bisymmetric field and possibly higher modes are likely to exist even in a flocculent galaxy.  \\item[-] There is some correspondence between the optical armlets discussed by Thornley \\& Mundy (\\cite{tho97}) and local magnetic field orientations. The optical structure seems to follow the general global asymmetry of magnetic pitch angles, while the magnetic lines seem to be aligned with only some segments of individual armlets.  \\item[-] NGC~4414 shows a strongly asymmetric distribution of Faraday rotation. Our observations are suggestive of a much thicker magnetoionic disk and an increased relative content of diffuse ionized gas (compared to classical \\ion{H}{ii} regions) in the southern disk. These asymmetries, together with those in the magnetic pitch angles and total power asymmetries suggest some past interaction (e.g. merging with dwarf galaxies or more likely accreting gas clouds) in the history of NGC~4414.  \\end{itemize}  In this work we show that the regular spiral magnetic pattern observed in all nearby galaxies does not require strong density wave action. We propose that the magnetic pattern in NGC~4414 is due to the dynamo process free from contamination by flows occurring in grand-design spiral arms. On the other hand, the picture of the magnetic field in NGC~4414 is still far from being clear and no good description of the three-dimensional magnetic field geometry is yet available, except for some rough and very qualitative guesses. We suggest that observing the galactic magnetic field may provide a clue to the properties of rarefied ionized gas, contributing greatly to the Faraday rotation but little to the H$\\alpha$ emission. This gas phase may be more sensitive to external interactions than either the molecular gas or \\ion{H}{ii} regions associated with recent star formation. Thus, we propose to use the magnetic fields as an indicator of past external interactions working even in cases when the optical information does not indicate significant perturbations. We note finally that a more quantitative magnetic field description in NGC~4414 may come from extensive modeling using more sophisticated multi-component field topologies and realistic distributions of thermal gas and relativistic electrons. A deep study of its environment in \\ion{H}{i} line and X-ray is highly desirable. Such a study is planned for the near future."
    },
    "0208/astro-ph0208194_arXiv.txt": {
        "abstract": "We have developed an atmospheric monitoring system for the Telescope Array experiment at Akeno Observatory. It consists of a Nd:YAG laser with an alt-azimuth shooting system and a small light receiver. This system is installed inside an air conditioned weather-proof dome. All parts, including the dome, laser, shooter, receiver, and optical devices are fully controlled by a personal computer utilizing the Linux operating system. It is now operated as a back-scattering LIDAR System. For the Telescope Array experiment, to estimate energy reliably and to obtain the correct shower development profile, the light transmittance in the atmosphere needs to be calibrated with high accuracy. Based on observational results using this monitoring system, we consider this LIDAR to be a very powerful technique for Telescope Array experiments. The details of this system and its atmospheric monitoring technique will be discussed. ",
        "introduction": "\\label{Introduction}  \\begin{figure}[htbp] \\centering \\leavevmode \\epsfxsize=14.5cm \\epsfbox{sim2.eps} \\caption{Results of Monte-Carlo simulation for the Telescope Array. Left panel indicates the energy resolution of the Telescope Array for $10^{20}$ eV cosmic rays. The energy resolution is better than 6\\% in 1 $\\sigma$ as long as we know the transmittance of the atmosphere. Right panel shows the error of estimated energy for $10^{20}$ eV cosmic rays as a function of incorrectness of atmospheric correction of attenuation length. The values used for the simulation were $L_M=20$km and $H_M=1.2$km. } \\label{fig:ta_sim} \\end{figure}  The air fluorescence technique for air shower observation has many advantages, (for example, direct measurement of shower longitudinal development, and stereo geometrical reconstruction with multiple eyes). These advantages will bring technical breakthroughs in EHE Cosmic Ray physics.  On the other hand, this fluorescence technique has a serious problem. In this technique the light yielded from air showers after the transmission is measured in atmosphere at a 10 $\\sim$ 60 km distance. Atmosphere can be considered a part of the detector. Without any monitoring, the atmospheric conditions may cause large uncertainty in the experiment and result in serious systematic error. Therefore, the calibration of the light transmittance in the atmosphere is essential for the air fluorescence experiment.  Charged particles in the air shower excite air molecules and yield fluorescence light of between 330nm and 400nm. This fluorescence light is also scattered by molecules and aerosols in the air before it reaches the detector. The scattering process caused by molecule is called Rayleigh scattering. The attenuation length of this scattering process $X_r$ is 2974 $g/cm^2$ for a 400nm wave length. Amount of scattered photons with wave length of $\\lambda$ can thus be calculated by following equation \\begin{equation} \\frac{dN_r}{dl} = -\\rho \\frac{N_r}{X_r}(\\frac{400nm}{\\lambda})^4 \\end{equation} where $\\rho$ is the atmospheric density. In this equation, the number of scattered photons is proportional to $\\lambda^{-4}$. This scattering process seriously effects the observation of fluorescence light in the air.  For example, when we consider the typical measurement of fluorescence light from an air shower at a distance of 30 km from detector, with a detector altitude of 1.5km and an elevation angle of 30 degrees, the fluorescence light has to pass through about 2200 $g/cm^2$ in the air and 70\\% of the photons will be scattered by this Rayleigh process.  There is another scattering process, Mie scattering. There are many kinds of scattering materials in the atmosphere, such as, water vapors, mists, clouds, dusts, blown small sands, artificial fumes, exhaust gas from cars, and smoke from forest fires. Furthermore, 'Kosa' from the Gobi desert in China is spread over a wide area of the northern hemisphere. These atmospheric aerosol's densities and their components change significantly with location and time. The scattering of photons by these aerosols is called Mie Scattering. In the Utah desert, the typical scale height of Mie scattering $H_M$ is 1.2 km and horizontal attenuation length $L_M$ is 20 km. The amount of scattering photons can be expressed by the following equation \\begin{equation} \\frac{dN_r}{dl} = -\\frac{N_r}{L_M} \\times exp(-\\frac{h}{H_M}) \\end{equation} Therefore, under the previously described conditions, approximately 27\\% of the photons contained in the fluorescence light from the air shower are scattered by the Mie process. When we take into account Rayleigh and Mie scattering, about 22\\% of the photons can reach our detector.  The Telescope Array (TA) will observe the air showers which flash far from the detector, for example, at more than 50km. Figure \\ref{fig:ta_sim} shows the results of Monte-Carlo simulation for TA.  According to this simulation study, the energy resolution of TA is better than 6\\% if atmospheric transmittance is accurately taken into account. However, if we apply the 20\\% shifted value in the attenuation length, the estimated cosmic ray energy has a systematic error of nearly 10\\%. To estimate the primary composition, TA will use $X_{max}$ which is the maximum position in the longitudinal shower development. The intrinsic resolution of $X_{max}$ is estimated to be 20g in the TA experiment using Monte Carlo simulation. If the assumed attenuation length differs by 20\\%, we estimate $X_{max}$ with a 10g systematic error. The correction of atmospheric transmittance affects not only energy determination but also air shower reconstruction. To realize this high resolution power of TA, we need to measure the attenuation length of the atmosphere with an accuracy of a few percent or better.  For the purpose of the atmospheric monitoring, TA and High Resolution Fly's Eyes are  developing various methods which use a laser or flashers\\cite{law99}\\cite{og1}\\cite{og2}\\cite{og3}\\cite{og4}. One of well-known technique for atmospheric monitoring is LIDAR which has been developed for environmental science. Using this technique, atmospheric conditions can be monitored remotely by measuring back-scattered light from a pulsed laser beam.  We are developing a steerable LIDAR system to study the atmospheric monitoring method in Akeno Observatory. In the present paper, we will describe the details of this system and discuss how to accurately measure transmittance in the atmosphere. ",
        "conclusions": "\\label{Summary}  We have developed a steerable LIDAR system for atmospheric monitoring by the Telescope Array. The system consists of a 5 mJ pulse laser and 16 cm diameter mirror. Using this system, a technique for atmospheric monitoring was developed.  First, the extinction coefficient $\\alpha$ at the level of the detector is measured. Then $\\alpha$ is estimated at all directions using Klett's method. The transmittance of the night sky to a distance of more than 10 km can be measured successfully. The statistical error in this analysis is less than 2\\% under a clear sky. It takes approximately 20 minutes to measure one azimuthal direction.  The systematic error which is caused by this method of analysis is estimated to be approximately  0.5\\% in a typical atmospheric model. The detector constants do not contribute to the systematic errors in this analysis except for the linearity of the PMT. The most significant systematic error is caused by uncertainty of the critical value and variation of the aerosol type. Some of these problems can be solved if we use a larger mirror or a higher intensity laser.  Based on these observational results, we are considering future plans regarding atmospheric monitoring for the Telescope Array. The signal to noise ratio is in proportion to beam intensity, mirror diameter, square root of observation time, and transmittance. $5mJ \\times 16 cm \\times \\sqrt{20 min} = 358[mJ \\cdot cm \\cdot min^{1/2}]$ is required to measure to a distance of 10km under a clear sky (in this case, the transmittance is about 60\\% in the vertical direction). To measure more than 50 km, we need $358 \\times 5^2 / 0.6^5 = 115000 [mJ \\cdot cm \\cdot min^{1/2}]$. This means that a 270 mJ laser is necessary for a 3 m diameter mirror if we want to measure one azimuthal direction in 2 minutes. This value can be reduced by a narrow band filter, small field of view, and high altitude location since the main component of the noise is night sky background light. For a telescope with a 1 degree field of view and F=1, the observed night sky background is about 5 MHz in the ultraviolet region.  For the operation of the Telescope Array detector, independent and redundant measurements of the atmosphere should be done for the sake of verification and cross-checking. In fact, several measurements being performed by the Telescope Array and the HiRes groups \\cite{he1.71}\\cite{he1.712}\\cite{he1.72}\\cite{he1.73}\\cite{he1.74}\\cite{he1.75}\\cite{he1.76}. For example, These include the measurement of the large angle scattered light of a polarized laser beam and radio-controlled xenon flashers using the HiRes detector, monitoring the transmittance of the air around the detector using the Nepherometer, measurement of the zenith angle dependence of star light and so on. By combining these measurements and the LIDAR method, more reliable measurements can be performed in the future.        \\begin{ack}   This work is supported by grants in aid \\#12304012 and \\#11691117 for scientific research from the JSPS(Japan Society for the Promotion of Science). The authors thank Dr. Lawrence R. Wiencke, Dr. Michael E. Roberts, and Prof. John A.J.Matthews for their thoughtful discussions.  \\end{ack}"
    },
    "0208/astro-ph0208477_arXiv.txt": {
        "abstract": "We present the first results from 60ks of observations of Arp 220 using the ACIS-S instrument on Chandra. We report the detection of several sources near the galaxy's nucleus, including a point source with a hard spectrum that is coincident with the western radio nucleus B. This point source is mildly absorbed ($N_H\\sim 3\\times 10^{22}\\mbox{cm$^{-2}$}$) and has an estimated luminosity of $4\\times 10^{40}$ erg/s. In addition, a fainter source may coincide with the eastern nucleus A. Extended hard X-ray emission in the vicinity raises the total estimated nuclear 2-10 keV X-ray luminosity to $1.2\\times 10^{41}$ erg/s, but we cannot rule out a hidden AGN behind columns exceeding $5\\times 10^{24}\\mbox{cm$^{-2}$}.$ We also detect a peak of soft X-ray emission to the west of the nucleus, and a hard point source 2.5 kpc from the nucleus with a luminosity of $6\\times10^{39}$ erg/s. ",
        "introduction": "Ultraluminous Infrared Galaxies (ULIRGs) have quasar-like bolometric luminosities ($>10^{12} L_{\\odot}$) dominated by the far-infrared (8--1000$\\mu$m) part of the spectrum (Sanders \\& Mirabel, 1996). Almost all ULIRGs are interacting or merging galaxies (Clements et al. 1996), possibly linking them to the transformation of disk galaxies into ellipticals (eg. Wright et al, 1990; Baker \\& Clements, 1997). The prodigious luminosity of ULIRGs is thought to be powered by a massive starburst, a dust--buried AGN or some combination of the two. Despite a decade of work we still have not been able to decide between these paradigms. Various scenarios have also been suggested linking the evolution of quasars with ULIRGs (eg. Sanders et al., 1988). These suggest that part of the luminosity we see from some ULIRGs originates in a dust--obscured AGN which later destroys or expels the enshrouding material. Meanwhile, studies of the X-ray background (Mushotzky et al, 2000) suggest that dust--enshrouded AGN make a substantial contribution to its hard component.  Such objects may also be linked (Trentham \\& Blain, 2001; Almaini et al., 1999) to the recently discovered Cosmic Infrared Background (Puget et al. 1996; Fixsen et al., 1998) and the objects that contribute to it (Puget et al. 1999; Sanders 2000 and references therein).  As the most obscured objects in the local universe, and as strong candidates for making the CIB, ULIRGs are ideal local laboratories for studying many of these issues.  Arp 220 is the nearest ULIRG, having an 8-1000$\\mu$m luminosity of $\\sim 1.2 \\times 10^{12}L_{\\odot}$ and a redshift of $z=0.018$. As such it is an ideal target for ULIRG studies. The consensus since ISO is that Arp 220 is powered by a massive burst of star formation rather than an AGN (Sturm et al 1996), but the possibility of a heavily obscured AGN powering the bulk of its emission remains (Haas et al 2001). The evolutionary scenario linking ULIRGs to AGN also allows the possibility that a weak, but growing, AGN may lie at the centre of Arp 220. While this may not be energetically significant at the present time, it may grow to prominence at later stages in the object's evolution. The plausibility of such a scenario has been investigated by Taniguchi et al. (1999), who show that it is quite possible for a massive black hole ($\\sim 10^6 M_{\\odot}$) to grow to $\\sim 10^8M_{\\odot}$ during the course of a galaxy merger, and thus to be capable of powering a quasar.  Signs of AGN activity can be sought with X-ray observations. The current data for Arp 220 includes soft X-ray images from ROSAT (Heckman et al. 1996). These show extended X-ray emission associated with the H$\\alpha$ nebula (Arribas, Colina \\& Clements 2001), which are thought to be produced by a superwind. However the overall soft X-ray luminosity is small relative to the far-IR luminosity when compared to other starbursts, and might allow room for some AGN contribution (Iwasawa, 1999). At higher energies, where an AGN would be more prominent, data is available from HEAO-1 (Rieke, 1988), CGRO (Dermer et al., 1997), ASCA (Iwasawa 1999), and BeppoSAX (Iwasawa et al. 2001). These rule out the possibility of an unobscured energetically significant AGN in Arp 220. The possibility remains, however, of a Compton thick AGN, with an obscuring column in excess of 10$^{25}$cm$^{-2}$, or of a weaker lower luminosity AGN that will grow into a quasar.  We have thus undertaken Chandra X-ray observations of Arp 220 aimed at detecting a weak or obscured AGN in its nucleus, and to study the extended superwind emission in detail. This paper presents the first results from our study of the nuclear regions. Our results on the superwind can be found elsewhere (McDowell et al. 2002, Paper II). We assume a distance of 76 Mpc (Kim \\& Sanders 1998) to Arp 220 throughout this paper. ",
        "conclusions": "We have shown that there is a source of hard, power-law-like, X-ray emission in the nuclear regions of Arp 220. This source is extended EW, consistent with the emission coming from both the radio/IR nuclei. The central concentration of hard X-ray emission in Arp 220 is in contrast to other interacting galaxies, where hard emission comes from clumps distributed across much larger physical distances. This difference may be associated with the merger in Arp 220 being older than in the Antennae or NGC3256, and that compact objects have sunk to its core (Tremaine, Ostriker \\& Spitzer, 1975) but may also be associated with Arp 220's greater luminosity. The origin of the hard emission is unclear.  Its spectrum is unlikely to be produced by young supernovae, but inverse Compton emission, albeit of very low efficiency, accretion onto ultra-luminous X-ray binaries or onto an AGN are all possible. If there is an AGN contribution, it has too low a luminosity for it to play a significant role in the energetics of the object, but the presence of even a weak AGN in Arp 220 would support a connection between ULIRGs and quasars (Sanders et al. 1988). If it is not an AGN, we cannot rule out the presence of a true AGN behind a Compton screen of column $\\sim10^{25}\\mbox{cm$^{-2}$}$.  If XRBs are responsible for the emission then a large number of conventional XRBs are needed, or a smaller number of ultraluminous ULX sources.  In either case, the concentration of these objects in the very centre of Arp 220 might indicate that these objects will later merge together to form a supermassive black hole and AGN (Taniguchi et al. 1999). If there is a weak AGN already in the nuclear regions, then this process may already have begun."
    },
    "0208/astro-ph0208531_arXiv.txt": {
        "abstract": "{ Using the IRAM 30\\,m telescope two molecular structures have been  detected which cover very small areas, $FWHM\\le1'$. The clouds have velocities of $v_{\\rm lsr}\\approx 5$\\,\\kms\\ and linewidth of $\\Delta v\\approx0.8$\\,\\kms; thus they belong most likely to the Milky Way. Applying standard conversion factors one finds that even at the upper distance limit of 2200\\,pc the structures are low mass objects ($M=(1-6)\\times10^{-4}\\, \\large({d\\over100{\\rm pc}}\\large)^2$\\mo) which are not gravitationally virialized. \\ion{H}{I} 21cm line data towards the clouds show no prominent HI clouds. The total \\ion{H}{I} column densities for both structures are below $N($HI$)\\le2.1\\times10^{20}$\\,cm$^{-2}$, corresponding to $A_{\\rm V}\\le0.2$\\,mag, assuming a standard gas-to-dust ratio. IRAS 100\\mue\\ data towards the structures show also only low emission, consistent with low extinction. Unless there is unseen cold dust associated with the structures this shielding is too low for the structures to survive the interstellar radiation field for a long time. The detection of 2 such structures in a rather limited sample of observations suggests that they could be a rather common feature in the interstellar medium, however, so far not recognized as such due to the weakness of their lines and their small extent. ",
        "introduction": "H$_2$ is the most abundant molecule in the universe. Due to its missing dipole moment and its high rotational constant it can only be observed directly in regions with high temperatures. The amount of molecular gas is however reasonably well estimated from observations of the second most abundant molecule, CO, if the metallicity of the gas is about solar (e.g. Dame et al. \\cite{dame:etal2001}).  There has been much debate on how much molecular gas can be hidden from the observer in form of small, cold molecular clouds (Pfenniger \\& Combes \\cite{pfenniger:combes1994}; Gerhard \\& Silk \\cite{gerhard:silk1996}; Walker \\& Wardle \\cite{walker:wardle1998}). Pfenniger \\& Combes argue that the molecular mass can be underestimated in the outer galaxy by as much as a factor 10, if this gas has a fractal structure. Thus, in this form molecular gas could account for all of the missing baryonic dark matter in our galaxy. These so-called ``clumpuscules'' (Pfenniger \\& Combes \\cite{pfenniger:combes1994}) i) must be  small to escape detection, ii) must have high-column densities to survive destruction and iii) must be dense to allow formation of H$_2$ within a reasonable amount of time. Thus the question on baryonic dark matter in form of small molecular clouds is directly linked to the question on the formation and destruction of H$_2$ and CO.  In this paper the detection of a previously unobserved component in the interstellar medium is described: molecular structures covering a very small area on the sky located in an environment with very low hydrogen column densities. These objects are ideal targets to study under which conditions CO and H$_2$ can survive in the interstellar medium. The detection was made serendipitously during a search for molecular gas in the tidal arms around the M\\,81 group of interacting galaxies; the results of this search will be described elsewhere.  \\begin{figure*} \\rotatebox{-90}{\\resizebox{9.5cm}{!}{\\includegraphics{ef251_f1.eps}}} \\caption{The small-area molecular structures and their environment. {\\it On the right:} Surrounding of the location were the two SAMS have been detected. The grey scale is taken from the IRAS survey at 100\\mue; the intensity scale is displayed on the right of the box. Contours represent the CO (1$\\to$0) observations with the 1.2m CfA telescope observed by de Vries et al. (\\cite{devries:etal1987}); they are every 0.8\\,K \\kms\\ starting at 0.8\\,K \\kms. The point source near SAMS1 is  the galaxy NGC\\,3077, the one near cloud SAMS2 is M\\,81 and the source below is M\\,82. Arrows point to the centres of the structures. {\\it On the left:} Integrated CO (1$\\to$0) intensity maps of the structures. The lowest contour (dashed) is at 0.06\\,K\\,\\kms\\ ($3\\sigma$), step size  between following contours is 0.06\\,K\\,\\kms. Observed positions are marked by open squares.} \\label{overview} \\end{figure*}  \\begin{table*} \\caption{Parameters for the molecular structures} \\begin{tabular}{l l l l l l l l l l} \\noalign{\\hrule} \\noalign{\\medskip} Number & $l$ & $b$ & transition & $T_{\\rm A}^*$ & $rms$ & $v_{\\rm lsr}$ & $\\Delta v$ & $FWHM$ & $M$(H$_2$)$^{(2)}$\\\\ &  (deg) & (deg)&         & (K)     &  (K)  & (\\kms)    & (\\kms)     & ($''$) & $10^{-4}\\, \\large({d\\over100{\\rm pc}}\\large)^2$\\mo\\\\ \\noalign{\\medskip} \\noalign{\\hrule} \\noalign{\\medskip} \\#1 & 141.9275 &     41.6970 & CO (1$\\to$0) & 0.142 & 0.008 & $+3.92\\pm0.02$ & $0.75\\pm0.04$ & 55 & 2\\\\ \\#1 & & & CO (2$\\to$1) & 0.034 & 0.009 & $+3.91\\pm0.08$ & $0.87\\pm0.15$ & $-$\\\\ \\#2a & 141.7154 &    40.8557 & CO (1$\\to$0) & 0.651 & 0.014 & $+5.38\\pm0.01$ & $0.90\\pm0.02$ & $60$ & 6\\\\ \\#2a & & & CO (2$\\to$1) & $0.24$ & 0.02 & $+5.35\\pm0.02$ & $0.85\\pm0.04$ & $-$\\\\ \\#2b & 141.7078 & 40.8446 & CO (1$\\to$0) & 0.32 & 0.04 & $+5.00\\pm0.03$ & $0.88\\pm0.09$ & $25^{(1)}$ & 1\\\\ \\#2b & & & CO (2$\\to$1) & $0.37$ & 0.08 & $+4.88\\pm0.04$ & $0.50\\pm0.06$ & $-$\\\\ \\noalign{\\medskip} \\noalign{\\hrule} \\noalign{\\medskip} \\noalign{Remarks: Values for $T_{\\rm A}^*$, $rms$, $v_{\\rm lsr}$, and $\\Delta v$ are derived from cloud averaged spectra. 1) Unresolved structure. 2) Mass is determined assuming $X_{\\rm CO}=0.6\\times10^{20}$cm$^{-2}$(K km/s)$^{-1}$, $d$ is the distance to the cloud.} \\end{tabular} \\label{obstable} \\end{table*} ",
        "conclusions": "In this Letter I have described the detection of a previously unobserved component of the interstellar medium: small-area molecular structures embedded in a very low column density environment. Aiming at the detection of molecular clouds in extragalactic tidal arms, the observations were not set up to detect local clouds, thus they are unbiased. Based on their location in a low column density environment with insufficient shielding against the interstellar radiation field, the clouds should not exist for very long. Whether sufficient shielding is provided by very cold dust remains to be tested by bolometer observations.  Whether these two structures were picked up just by chance before they are dissolved or whether their detection indicates that H$_2$ and CO can survive with less shielding than previously adopted is an open question. Because two structures were discovered in a rather limited sample of observations, one could  speculate that such structures might be frequent in the interstellar medium. This explanation is also supported by the detection of H$_2$ in many lines of sight through diffuse galactic clouds in the spectra of  distant quasars (e.g. Richter et al. \\cite{richter:etal2001}). Due to their weakness and compactness these structures will however be missed by most observations.   Whether or not  SAMS are related to the tiny-scale atomic structures (TSAS, Heiles \\cite{heiles1997}) is unclear. TSAS have been found in \\ion{H}{I} 21cm line observations. They are thought to be predominantly made of atomic gas and have sizes of about 25\\,AU (Diamond et al. \\cite{diamond:etal1989}).  SAMS described here are molecular and have unknown sizes, due to the unknown distance. If we adopt for the moment that they have the same sizes as the TSAS, they would be as close as 0.5\\,pc. At that distance they would only live for a few hundred years. Whether SAMS could be related to the sodium absorption line clouds which were found to be non-coincident with \\ion{H}{I} clouds (Lilienthal \\& Wennmacher \\cite{lilienthal:wennmacher1990}) remains to be tested by sensitive CO observations. If so, these observations will provide the crucial distance information for SAMS.  To estimate the frequency of SAMS, it will be worthwhile to check existing CO observations made towards extragalactic objects for the presence of similar absorption- or emission-like features. Clearly, more information on SAMS are needed before one could draw more firm conclusions on their nature.  \\begin{acknowledgement} I thank J\\\"urgen Ott for providing Effelsberg HI data, Fabian Walter for providing the VLA data,  and Axel Wei\\ss\\ for doing the zero-spacing correction. I thank Ulrich Mebold, Fabian Walter and Peter Kalberla for critically reading the manuscript. \\end{acknowledgement}"
    },
    "0208/hep-ph0208248_arXiv.txt": {
        "abstract": "Protons accelerated in the cores of active galactic nuclei can effectively produce neutrinos only if the soft radiation background in the core is sufficiently high.  We find restrictions on the spectral properties and luminosity of blazars under which they can be strong neutrino sources. We analyze the possibility that neutrino flux is highly beamed along the rotation axis of the central black hole. The enhancement of neutrino flux compared to GeV $\\gamma$-ray flux from a given source makes the detection of neutrino point sources more probable. At the same time the smaller open angle reduces the number of possible neutrino-loud blazars compared to the number of $\\gamma$-ray loud ones. We present the table of 15 blazars which are the most likely candidates for the detection by future neutrino telescopes. ",
        "introduction": "Neutrino telescopes which already operate or under construction will presumably be able to detect point sources of neutrinos with energies  up to $\\sim10^{17}\\,$eV by looking for the showers and/or tracks from charged leptons produced by charged current reactions of neutrinos in ice, in the case of AMANDA~\\cite{amanda,amanda_limit} and its next generation version ICECUBE~\\cite{icecube}, in water, in the case of BAIKAL \\cite{baikal,baikal_limit}, ANTARES~\\cite{antares}, and NESTOR ~\\cite{nestor} (for recent reviews of neutrino telescopes see Ref.~\\cite{nu_tele}).  From the other side future Ultra-High Energy Cosmic Ray (UHECR) experiments like the Pierre Auger Observatory~\\cite{auger,auger_nu} will be able to detect neutrinos with energies above $10^{17}\\,$eV. At the highest energies above $\\sim10^{19}\\,$eV the  telescope array~\\cite{ta,ta_nu} or space based observatories such as EUSO~\\cite{euso} and OWL~\\cite{owl,owl_nu} will also measure the neutrino flux. The flux from point-like sources at those energies can be a combination of  the direct flux from a source and the secondary neutrino flux produced by Ultra-High Energy (UHE) protons emitted by the source in interactions with cosmic microwave background radiation (CMBR) photons. If neutrino flux at high energies $E>10^{17}$ eV is large enough to give more then 1-3 events per km$^2$ per year, future km$^2$ neutrino telescopes like ICECUBE will be able to detect those neutrinos coming from above \\cite{halzen-alvarez}.  At the intermediate energies between $\\sim10^{15}\\,$eV and $\\sim10^{19}\\,$eV there are plans to construct telescopes to detect fluorescence and \\v{C}erenkov light from near-horizontal showers produced in mountain targets by neutrinos ~\\cite{fargion,mount}. The alternative of detecting neutrinos by triggering onto the radio pulses from neutrino-induced air showers is also currently investigated~\\cite{radhep}. Two implementations of this technique, RICE, a small array of radio antennas in the South pole ice~\\cite{rice}, and the Goldstone Lunar Ultra-high energy neutrino Experiment (GLUE)~\\cite{glue}, have so far produced neutrino flux upper limits. Acoustic detection of neutrino induced interactions is also being considered~\\cite{acoustic}.   The simplest way to produce neutrinos in astrophysical objects is to accelerate protons and then collide them with soft photon background with energy above  photo-pion production threshold. The produced pions will decay in photons, electrons, positrons  and neutrinos. If protons are captured within the source, the estimate on neutrino flux from a given source can be obtained from the detected $\\gamma$-ray flux, since the energy deposit in neutrinos in pion decays is of the same order as the energy in photons. If the sources are transparent for the primary protons than a limit on diffuse neutrino flux from all the possible sources can be obtained from the detected high-energy proton flux. This idea was first suggested in \\cite{halzen}. For a particular case of $E^{-2}$ proton spectrum coming from AGN's the calculation was done in \\cite{wb}. The same calculation for $E^{-1}$ proton flux was made in \\cite{mpr}. In \\cite{diffuse} the dependence of neutrino flux from proton spectrum, cosmological parameters and distribution of sources was investigated in details.  In particular it was shown that in many cases neutrino flux can exceed value calculated in \\cite{wb} or even value of \\cite{mpr} and only bound on diffuse neutrino flux come from EGRET measurement.     The Universe is not transparent for photons with energies above 100 GeV. The highest energy photons from astrophysical objects (nearby TeV blazars) seen so far had energies $E \\sim 10^{13}$ eV. No direct  information about emission  of $E > 10^{13}$ eV particles is available now. At the same time  it is well established that photon emission from blazars (active galactic nuclei (AGN), which we see almost face on) in the MeV-TeV energy range is highly anisotropic. Typical estimates of the $\\gamma$ factors of the emitting plasma, $\\gamma\\sim 10$, imply that in the $10^{6-13}$ eV band almost all $\\gamma$-ray flux  is radiated in a cone with the opening angle $\\theta\\sim 1/\\gamma\\sim 5^\\circ$. Particles (photons, neutrinos) in the higher energy range $E> 10^{13}$ eV can be emitted in an even narrower cone. This fact favors blazars  as promising neutrino sources.  Recent X-ray observations of large-scale jets in AGN can shed some light on the issue of particle acceleration to the energies much above TeV in the AGN cores. Indeed, in order to explain X-ray synchrotron emission on very large scales of order of 100 kpc away from the AGN core one needs to suppose that multi-TeV electrons are continuously produced over the whole jet length. A model which naturally explains this continuous production of multi-TeV electrons was recently proposed in \\cite{gamma_jet} (see Fig. \\ref{fig:cartoon} and \\ref{fig:blazar_gev}). The idea is that $\\gamma$-rays with energies $10^{14-16}$ eV emitted from the  AGN core  produce $e^+e^-$ pairs in interactions with the CMBR photons  at the distance scale  10-100 kpc away from the core. Thus, within this model the fact that jets with the lengths about 10-100 kpc are commonly observed in AGNs  enables to conclude that (1) particles with energies $ E \\ge 10^{14-16}$ eV are produced in the AGN cores and (2) these particles are normally emitted in a cone with opening angle $\\theta\\sim 1^\\circ$. The diffuse neutrino flux in this model was calculated in \\cite{diffuse}. In this paper we  discuss which blazars will be the most promising neutrino sources if neutrinos are produced in the AGN cores, as in the model \\cite{gamma_jet}. AGN which can be significant point sources of neutrinos were analyzed in \\cite{eichler,stecker,atoyan,halzen}. In particular, high neutrino fluxes were conjectured to come from brightest quasars like 3C 273 \\cite{eichler,stecker} or TeV blazars, like Mkn 421 \\cite{halzen}. Enhancement of neutrino flux during the flaring activity in 3C 279 was considered in \\cite{atoyan}. Predictions of the model \\cite{gamma_jet} are quite different. In particular, none of the three above cited blazars enters our list of most probable neutrino sources.    In Section  \\ref{sec:two} we will discuss a mechanism of neutrino production and derive a bound on the magnitude and redshift of blazars which can be neutrino-loud. In the third section we discuss the neutrino flux from TeV gamma-ray sources. In  Section \\ref{sec:four} we will derive the neutrino fluxes from 15 blazars which are the most promising neutrino sources. In Section \\ref{sec:five}  we will discuss the secondary neutrino fluxes from UHECR sources. ",
        "conclusions": "In this work we have considered conditions under which $\\gamma$-ray loud blazars can be significant neutrino sources. High energy neutrinos are produced in the photo-pion interactions of protons accelerated in the AGN cores with soft photon background. We have derived the bound Eq.~(\\ref{bound}) on the redshift and V-magnitude of the candidate neutrino-loud blazars. From 27 GeV-TeV $\\gamma$-ray loud blazars we selected 15 most favorite candidates which satisfy the criteria {\\bf 1-3} listed in Section \\ref{sec:four}.  An estimate of the neutrino flux from a given object can be obtained from the observed $\\gamma$-ray flux if we assume that the main contribution to the GeV-TeV luminosity of a blazar comes from the electromagnetic cascade initiated by $10^{14-16}$ eV photons which are produced together with neutrinos in photo-pion reactions. It is important to note that the neutrino flux from a given source  can be much higher than the $\\gamma$-ray flux detected by EGRET or other $\\gamma$-ray telescopes due to the variety of reasons listed in Section \\ref{sec:four}.  Because the optical depth for TeV photons in the AGN core is three orders of magnitude smaller than the  optical depth for protons in the same photon background we conclude that confirmed TeV-loud blazars can not be  sources of significant neutrino flux. (At the same time, this argument does not exclude the possibility that these objects can be  neutrino sources from UHECR protons which produce neutrinos in interactions with CMBR photons.)  We have considered the  model in which the neutrino flux is highly beamed in the directions of the large scale jets emitted by the AGN. This model has several experimentally testable predictions. First, GeV-loud sources  in which the large scale jets are not seen face on (like 3C 279) can not be neutrino sources. Next, the neutrino flux from a given source can be much larger than the observed gamma-ray flux due to a smaller opening angle for neutrinos as compared to GeV gamma-rays.  We have also found that if the optical depth for protons is large enough and protons are accelerated up to the highest energies, $10^{20}-10^{21}$ eV, the same sources from Table \\ref{tab:sources} can be seen both by future UHECR detectors  and by neutrino telescopes, see  Fig.~\\ref{fig:detect_energy}.  We conclude that the next generation  of neutrino telescopes and UHECR detectors will have a good chance to see  point-like neutrino sources."
    },
    "0208/astro-ph0208582_arXiv.txt": {
        "abstract": "{We report the detection of hard X-ray emission ($>2$~keV) from a number of point sources associated with the very young massive star-forming region IRAS~19410+2336. The X-ray emission is detected from several sources located around the central and most deeply embedded mm continuum source, which remains undetected in the X-ray regime. All X-ray sources have K-band counterparts, and those likely belonging to the evolving massive cluster show near-infrared colors in the 2MASS data indicative of pre-main-sequence stages. The X-ray luminosities around $10^{31}$~erg\\,s$^{-1}$ are at the upper end of luminosities known for low-mass pre-main-sequence sources, and mass estimates based on the infrared data indicate that at least some of the X-ray detected sources are intermediate-mass objects. Therefore, we conclude that the X-ray emission is due to intermediate-mass pre-main-sequence Herbig Ae/Be stars or their precursors. The emission process is possibly due to magnetic star-disk interaction as proposed for their low-mass counterparts. ",
        "introduction": "\\label{introxray}  In the past, star formation research in the X-ray regime has focused strongly on low-mass objects, e.g., T Tauri stars. These objects emit mainly in the soft range of the X-ray spectrum ($<2$\\,keV) with typical X-ray luminosities between $10^{28}-10^{30}$~erg\\,s$^{-1}$. The satellite observatory ROSAT was an ideal instrument to study such stars within a few 100~pc distance from the Sun, and the observed emission can be explained by enhanced solar-type magnetic activity \\citep{feigelson 1999}. In the last few years, a rising number of Class I protostars, which are still deeply embedded within their natal molecular cores ($A_{\\rm{V}} \\approx 10-100$~mag), have been detected in the hard X-ray regime between 2 and 10 keV with X-ray luminosities higher than $10^{30}$~erg\\,s$^{-1}$. These detections were mostly made with the X-ray satellites ASCA and, most recently, with {\\it Chandra}. Magnetic star--disk interactions are thought to be the most likely explanation for the hard X-ray emission \\citep{hayashi 1996,montmerle 2000}. Furthermore, X-ray variability is observed in all types of low-mass pre-main-sequence objects. For an excellent recent review on these topics see \\citet{feigelson 1999}. Recently, \\citet{tsuboi 2001} and \\citet{tsujimoto 2002} reported the first tentative detections of deeply embedded Class 0 protostellar candidates in OMC3 by {\\it Chandra}.  In comparison with the low-mass regime, X-ray observations of massive star-forming regions have been rare. Due to the high visual extinction within such regions ($A_{\\rm{v}}$ up to a few 100 or even 1000), soft X-ray emission is completely absorbed by the gas along the line of sight. \\citet{hofner 1997} detected with ASCA for the first time hard X-ray emission in the massive star-forming region W3. Because of the low angular resolution of ASCA ($>1'$) they could not determine whether the emission is caused by the superposition of many point sources, e.g., protostellar clusters, or whether it is due to a wind-shocked cavity resulting from strong stellar winds interacting with the surrounding medium.  Recently, \\citet{churchwell 2001} reported that {\\it Chandra} data of the same region with a spatial resolution of $0.5''$ resolve the emission into many individual sources distributed over the entire W3 complex. Similarly, \\citet{garmire 2000} and \\citet{feigelson 2002} reported about 1000 X-ray emitting pre-main-sequence stars between 0.05 and 50 M$_{\\odot}$ in the Orion Nebula. Additionally, \\citet{zinnecker 1994} found X-ray emission with the ROSAT satellite in the soft X-ray band associated with several intermediate-mass Herbig Ae/Be pre-main-sequence stars. The derived X-ray luminosities for the Herbig Ae/Be stars are ranging between $10^{30}$~erg\\,s$^{-1}$ and $10^{32}$~erg\\,s$^{-1}$. \\citet{preibisch 1995} speculated that the emission might originate from coronal activity due to shear dynamo action. A study of the more distant molecular clouds Monoceros and Rosette also found indirect evidence for X-ray emission from intermediate-mass pre-main-sequence sources \\citep{gregorio 1998}. An example of hard X-ray emission from a Herbig Be star is MWC 297 \\citep{hamaguchi 2000}. In a recent X-ray study of the Monoceros R2 molecular cloud, \\citet{kohno 2002} detected X-ray emission for stars of all masses, in particular hard X-ray emission from high-mass pre-main-sequence or Zero-age-main-sequence stars. Based on these results, X-ray emission seems likely to be an ubiquitous phenomenon in the protostellar evolution of stars of all masses.  So far, X-ray studies in high-mass star-forming regions focused on more evolved massive star formation sites, while the very deeply embedded phase --~and thus the youngest stage of stellar evolution~-- has not been detected at all. Results obtained in recent years by our group \\citep{sridha, beuther 2002a, beuther 2002b, beuther 2002d} and other groups studying massive star formation (e.g., \\citealt{cesaroni 1997}, \\citealt{zhang 2002}, \\citealt{tan 2002}, \\citealt{yorke 2002}) support the hypothesis that massive stars form via disk accretion in a similar fashion as low-mass stars. Therefore, high-mass star-forming cores are promising candidate regions where in rather small spatial areas a number of sources could be hard X-ray emitters via the physical process of star-disk interactions \\citep{hayashi 1996,montmerle 2000}. Necessary observational requirements are first of all sensitivity in the hard X-ray regime ($>2$~keV), because only hard X-ray photons can penetrate high gas column densities. Additionally, high angular resolution is needed to resolve different sub-sources of the forming cluster. While ROSAT was not sensitive to hard X-ray photons, the spatial resolution of ASCA was not sufficient to study massive star-forming regions in detail at their typical distances of a few kpc.  The new-generation X-ray satellite telescopes {\\it Chandra} and XMM-Newton comprise both features, being sensitive up to 10~keV, and having a spatial resolution of $0.5''$ ({\\it Chandra}) and $15''$ (XMM-Newton), respectively. Especially {\\it Chandra} is able to resolve many different sub-sources as impressively demonstrated in Orion by \\citet{garmire 2000} and \\citet{feigelson 2002}.  Here, we present a {\\it Chandra} X-ray study of the very young, massive and deeply embedded star-forming cluster IRAS 19410+2336.  IRAS 19410+2336 is part of a large sample of 69 high-mass protostellar candidates which has been studied extensively in a series of papers during the last years \\citep{sridha,beuther 2002a,beuther 2002b,beuther 2002c,beuther 2002d}. We assume the source to be located at its near kinematic distance of 2.1~kpc \\citep{sridha}, because the derived outflow parameters are unreasonably high for the far kinematic distance \\citep{beuther 2002b}. At the distance of 2.1~kpc, its infrared derived luminosity is $10^4$~L$_\\odot$, and one observes two adjacent star-forming cores with masses of 840~M$_{\\odot}$ and 190~M$_{\\odot}$ \\citep{sridha,beuther 2002a}. Each core drives a massive bipolar outflow in east-west direction \\citep{beuther 2002b}. At the center of the southern massive core, a very compact and weak ($\\sim1$\\,mJy) cm wavelength source is detected, which coincides with H$_2$O and Class {\\sc ii} CH$_3$OH maser emission \\citep{beuther 2002c}. Figure \\ref{xrayimage}(top) gives an overview of the region of interest with the 1.2~mm dust continuum data \\citep{beuther 2002a} superposed on an infrared K-band image (\\S \\ref{kband}). As the source is located in the Galactic plane, confusion due to foreground and background sources is expected and has to be disentangled by the different observations.  We focus on the X-ray emission of this region and correlate the detected X-ray sources with high-resolution images in the mm and near-infrared regime. \\S 2 describes the different observations we performed (X-ray, near-infrared and mm data), and in \\S 3 we derive the main physical parameters of this star-forming region from our data (source detections, spectra, X-ray luminosities, plasma temperatures and masses). Finally, \\S 4 compiles our conclusions and puts the observational findings into a more general framework of high-mass star formation.  \\begin{figure}[ht] \\includegraphics[angle=-90,width=8.7cm]{f1.eps} \\includegraphics[angle=-90,width=8.7cm]{f2.eps} \\caption[K-band, mm and Xray maps]{\\footnotesize {\\bf top:} The contours present the large-scale 1.2~mm dust continuum emission from $10\\%$ to $90\\%$ (steps of $10\\%$) of the peak flux \\citep{beuther 2002a}. The grey-scale shows the K-band image. {\\bf bottom:} The contours show the PdBI 2.6~mm continuum dust cores (black: levels 3.6(1.2=$1\\sigma$)8.4~mJy/beam, white: 9.6(2.4)20~mJy/beam) superposed on the K-band image. Circles and stars mark the X-ray sources presented in this paper. Sources with circles are most likely associated with the massive star-forming region, whereas the asterisks represent those that might be foreground objects. The box outlines the region presented in Fig. \\ref{zoomxray}. \\label{xrayimage}} \\end{figure} ",
        "conclusions": "Hard X-ray emission from a number of point sources is detected in a young, massive and embedded star-forming cluster in a very early stage of evolution. We did not detect X-ray emission from the most massive and central object (upper limit $L_{\\rm{X}}<9 \\times 10^{34}$~erg\\,s$^{-1}$ for k$T=3\\,$keV) but from a few sources in its vicinity. Typical X-ray properties of high-mass main-sequence stars ($M_\\star \\geq 8\\,M_\\odot$), where the emission originates from internal shocks in the radiation-driven stellar winds, are $L_{\\rm X}/L_{\\rm bol} \\sim 10^{-7}$, soft X-ray spectra with typical temperatures of k$T \\la 0.5$~keV and very little variability \\citep{berghoefer 1997}. The X-ray properties we find for X1 to X7 are clearly different, they show hard X-ray emission (k$T \\geq 3$~keV) and most of them also near-infrared excess, very similar to those observed for extremely young embedded objects (Class I protostars), which have typical plasma temperatures of k$T \\sim 5-10$~keV \\citep{feigelson 1999,imanishi 2001}.  Combining infrared data with pre-main-sequence evolutionary tracks \\citep{palla 1999}, it is possible to estimate the approximate masses of some of the hard X-ray sources. Those estimates indicate that they are in the intermediate-mass regime of Herbig Ae/Be objects. Taking additionally into account the harder X-ray spectra compared with other Herbig Ae/Be studies, it is likely that the X-ray sources in IRAS~19410+2336 are even precursors of Herbig Ae/Be stars. The emitted X-ray photons with energies mostly above 2~keV indicate plasma temperatures $>10^7$~K and X-ray luminosities around a few times $10^{31}$~erg\\,s$^{-1}$. The latter values are well within the regime of Class I low-mass protostars \\citep{feigelson 1999}, but they are also consistent with the results obtained for Herbig Ae/Be stars \\citep{zinnecker 1994}. Thus, some of the objects are probably very young intermediate-mass pre-main-sequence sources, whereas other sources could also be low-mass Class I or T Tauri stars. The emission of one of the sources is consistent with X-ray variability.  In spite of the observation of hard X-ray emission in the weak-lined T Tauri star V773 \\citep{tsuboi 1998}, where the disk has already been dissipated to a large degree, it is unlikely that the hard X-ray spectra observed in younger class I sources are due to enhanced solar-type magnetic activity. Therefore, it is proposed that the hard X-ray emission, which is more often observed in class I sources than in weak-lined T Tauri stars, is produced by magnetic reconnection effects between the protostars and their accretion disks \\citep{hayashi 1996,feigelson 1999,montmerle 2000}. As the X-ray spectra of the intermediate-mass objects in IRAS~19410+2336 exhibit very similar signatures to such low-mass sources, our results are consistent with disks being present in intermediate-mass star formation as well.  For a better understanding of the nature of the underlying X-ray powering sources much work has to be done in the future. Deeper X-ray and near-infrared images will help to set stronger constraints on the physical properties of the sources: it will be necessary to obtain sensitive X-ray spectra to determine better the absorbing $N_{\\rm{H}}$ column densities and plasma parameters. It is also of great interest to further investigate the properties of the central and deepest embedded object, which means lowering the detection limits. Furthermore, the variability of the X-ray sources in very young massive star-forming regions is not known so far. Therefore, several approaches should be followed in the years coming: deep {\\it Chandra} observations of the source of interest will disclose variabilities and faint emission of the central object. Additionally, a sample of similar sources has to be identified, because only a statistical analysis of several young high-mass star-forming regions can build a solid picture of the relevant physical processes. As high spatial resolution is essential for many of this studies, {\\it Chandra} is a very promising choice. But considering the higher sensitivity of XMM-Newton, it might be possible to study grating X-ray spectra of the brightest sources of the sample of clusters studied then. On the near-infrared side, we suggest to get deeper images in the J, H and K bands to improve the mass estimates of the X-ray emitting sources, and near-infrared spectroscopy might help classifying the types of stars \\citep{hanson 2002}.  To summarize, X-ray studies of young massive star-forming regions are just in its infancy, and the next years with the space telescopes {\\it Chandra} and XMM-Newton will bring many new insights in that research area. We also like to stress that multi-frequency studies over a wide range of bands are extremely promising approaches for the understanding of the physical processes forming massive stars."
    },
    "0208/astro-ph0208061_arXiv.txt": {
        "abstract": " ",
        "introduction": "Newton's law of gravity is routinely used to describe galaxies, even though its validity has been fully verified only within the solar system, in regimes of acceleration orders of magnitude stronger than the ones typical of galaxies. Though we have plenty of reasons for trusting Newton's law also in these weak regimes, there are strong observational evidence that all spacecrafts in the periphery of the solar system are experiencing an anomalous, unexplained acceleration toward the sun (Anderson et al. 1998). Moreover, the modified Newtonian dynamics (MOND; Milgrom 1983, Sanders \\& McGaugh 2002), which  posit a breakdown of Newton's law of gravity below few times $a_0 \\sim 10^{-8}$ cm s$^{-2}$, succeeds in explaining many properties of galaxies and other astrophysical phenomena without invoking non-baryonic dark matter (DM).  Because of these empirical evidence, we decided to perform an experiment to test Newton's law of gravity. We focused on globular clusters (GC) because they are the smallest virialized structure believed to be DM free. This ensures GC's internal dynamics should follow precisely the prediction of newton's law for any acceleration, in particular below $a_0$. In the case a discrepancy would be found, then DM can not be invoked to explain it, and Newton's law would be falsified. ",
        "conclusions": ""
    },
    "0208/astro-ph0208257_arXiv.txt": {
        "abstract": "We present predictions for numerous statistics related to the presence of voids in the distribution of galaxies in a cold dark matter model of structure formation using a semi-analytic model of galaxy formation. Our study is able to probe galaxies with masses as low as $10^9h^{-1}M_\\odot$ corresponding to absolute magnitudes of $M_{\\rm b_J}-5\\log h=-18.1$ and $M_{\\rm r}-5\\log h=-18.7$. We quantify the void and underdense probability functions, distributions of nearest neighbour distances and void sizes and compute the density profiles of voids. These results are contrasted with the expectations for dark matter (and the difference examined in terms of the galaxy/dark matter biasing relation) and are compared to analytic predictions and observational data where available. The predicted void probability functions are consistent with those measured from the Center for Astrophysics redshift surveys given the rather large uncertainties in this relatively small (for studies of voids) observational sample. The size of the observational sample is too small to probe the bias between galaxies and dark matter that we predict. We also examine the predicted properties of galaxies living within voids and contrast these with the general galaxy population. Our predictions are aimed at forthcoming large galaxy redshift surveys which should for the first time provide statistically accurate measures of the void population.\\\\  \\noindent {\\bf Key words:} galaxies: statistics, cosmology: theory, dark matter, large-scale structure of Universe ",
        "introduction": "One of the most striking features of galaxy redshift surveys is the presence of large regions of space that are nearly devoid of galaxies. These ``voids'' are thought to form from the most underdense regions of the initial density field, although other suggestions exist such as cosmic explosions \\cite{ocow81} or first order phase transitions \\cite{amendola99}. While voids have been seen in redshift surveys since the late 1970's \\cite{greg78,kirshner81,geller89}, due to their large size and low number density it is only with the advent of the Two-degree Field Galaxy Redshift Survey (2dFGRS) and Sloan Digital Sky Survey (SDSS) that the statistical properties of voids will be quantified in a meaningful way.  It is well known that the cold dark matter (CDM) cosmogony produces voids in the distribution of dark matter (i.e. highly underdense regions which are not, however, completely empty of dark matter), and the properties of these voids are well studied \\cite{einasto91,ghigna94,little94,vogeley91,vogeley94,ghigna96,kns97,muller00,armu,sheth02}. The properties of voids, and the galaxies which live within them have been proposed as a strong test of the CDM scenario \\cite{peebles01}. However, there have been very few theoretical studies of voids in the \\emph{galaxy} distribution expected in CDM (with the notable exception of \\pcite{mathis02}), primarily due to the lack of a detailed, physical model for galaxy formation in the past. It is known both theoretically and observationally that at least some galaxies are biased tracers of the dark matter (Davis \\& Geller 1976; for recent results see Hoyle et al. 1999, Benson et al. 2000a), even though $L_*$ galaxies may be close to unbiased, as shown in the 2dFGRS by \\scite{lverde}. Recent studies of the dependence of clustering amplitude on galaxy luminosity and morphology in the SDSS \\cite{zehavi02,zehavi03} and the 2dFGRS \\cite{norberg01a,norberg01} show a substantial increase with luminosity of the correlation function amplitude which is only partly induced by the variations in the morphological mix of galaxies. Consequently, we cannot expect voids in the distributions of galaxies and dark matter to have the same statistical properties. In fact we will show that they can be quite different.  In this work we aim to remedy this deficiency by presenting predictions for the simplest and most useful statistical quantifiers of galaxy voids in a CDM universe using a combination of N-body simulations of dark matter and semi-analytical modelling of galaxy formation. This technique has been demonstrated to naturally explain several aspects of galaxy clustering --- such as the near power-law shape of the galaxy correlation function and the dependence of galaxy clustering on luminosity \\cite{kauffmann99,cluster1,cluster3}. Such an approach is currently the only means to make physically realistic predictions for galaxy voids (the only competitive method for modelling galaxy formation --- direct hydrodynamical simulation --- cannot currently be applied to sufficiently large volumes of the Universe with the desired resolution). We will make predictions specifically aimed at the 2dFGRS and SDSS surveys, and will contrast our results with previous analytical and numerical studies.  The remainder of this paper is arranged as follows. In \\S\\ref{sec:analysis} we describe the construction of galaxy catalogues and the details of the analysis which we apply to them. In \\S\\ref{sec:statistics}, \\S\\ref{sec:props} and \\S\\ref{sec:vgals} we present results for a variety of statistical properties of voids and the galaxies within them. Finally, in \\S\\ref{sec:discuss} we present our conclusions. ",
        "conclusions": "\\label{sec:discuss}  We have presented detailed theoretical predictions for a wide range of statistics related to voids in the distribution of galaxies, and for properties of galaxies living within those voids. These predictions have been specifically aimed at the 2dFGRS and SDSS, which, due to their large size, should provide accurate measures of these statistics. We have focussed on simple selection criteria, namely simple cuts in luminosity, which we believe will permit the most robust comparison between theory and observations.  The properties of voids and void galaxies are potentially a strong constraint on models of structure and galaxy formation. However, we have shown that many of the properties of galaxy and dark matter voids differ significantly, indicating (as we have shown) that galaxy bias as well as gravitational instability is important for the formation of galaxy voids. Although our understanding of galaxy bias has progressed greatly in recent years we cannot be certain that our understanding is complete. Where suitable data already exist (specifically the void and underdense probability functions analysis applied to the CfA surveys by \\pcite{vogeley94}) we find that the model is consistent with the data, but that the current observational sample is too small to probe the signal of bias expected from our model. This situation should be rectified when this analysis is repeated on larger redshift surveys.  As has been shown in previous works on galaxy clustering using semi-analytic and N-body techniques, the results are rather robust to changes in model parameters \\cite{cluster1}. If we compute void statistics such as the VPF, nearest neighbour distributions or void size distributions at a fixed number density then our predictions are unaffected by changes in most model parameters (e.g. the strength of supernovae feedback, the star formation timescales in galaxies etc.). This results from the fact that the main effect of changing these parameters is to change galaxy luminosities without significantly changing the ranking of galaxy luminosity.  Predictions are changed by parameters in the model which do change the ranking of luminosities. For example, significantly increasing or decreasing the rates of galaxy mergers can make strong differences in some of the statistics considered here. (Note however, that with the improved merging model of \\scite{benson02} we no longer have the freedom to adjust merger rates in our model.) The properties of void galaxies (e.g. Fig.~\\ref{fig:vgalsGIF}) \\emph{are} affected by changes in model parameters. As the differences between void and wall galaxies seen in Fig.~\\ref{fig:vgalsGIF} are so small we choose not to explore the dependencies of these properties on model parameters in this work.  The models employed in our analysis include the effects of ``photoionization suppression'' as described by \\scite{benson02}. This feedback mechanism has the potential to strongly alter the properties of voids in the galaxy distribution. Halos in voids are of lower mass on average than in higher density regions. Since photoionization suppression acts most effectively on low-mass halos it will cause greatest suppression of galaxy formation in voids. (Furthermore, although not included in our present modelling, reionization of the Universe may begin in voids, enhancing the suppression in these regions further.) For the lowest mass galaxies resolved in our current N-body simulations the effects of photoionization suppression are negligible (e.g. the VPF and nearest neighbour distributions for galaxy samples of fixed number density are indistinguishable between models with and without photoionization suppression). We may expect however strong differences to show up in higher resolution simulations."
    },
    "0208/astro-ph0208127_arXiv.txt": {
        "abstract": "A pulsar model is proposed which involves the entire magnetosphere in the production of the observed coherent radio emission. The observationally-inferred regularity of peaks in the pulsar profiles of `slow' pulsars (Rankin 1990,1993a), is shown to suggest that inner and outer cones of emission near the polar cap interact with and `mirror' two rings in the outer magnetosphere: one where the null line intersects the light-cylinder, and another where it intersects the boundary of the corotating dead zone.  The observed dependency of conal type on period is shown to follow naturally from the assumption that cones only form when the mirror intersection points lie between two fixed heights from the surface, suggesting that a feedback system exists between the surface and the mirror points, accomplished by a flow of charges of opposite sign in either direction. In their flow to and from the mirror points, the particles execute an azimuthal drift around the magnetic pole, thereby creating a ring of discrete `emission nodes' close to the surface. Motion of the nodes is observed as subpulse `drift', which is interpreted here as a small residual component of the real particle drift. The nodes can move in either direction or even remain stationary, and can differ in the inner and outer cones. A precise fit is found for the drifting subpulses of PSR0943+10. Azimuthal interactions between different regions of the magnetosphere depend on the angle between the magnetic and rotation axes and influence the conal type, as observed. The model sees `slow' pulsars as being at the end of an evolutionary development where the outer gap region no longer produces pair cascades, but is still the intermittent source of low-energy pairs in a magnetosphere-wide feedback system. ",
        "introduction": "In 1975 Ruderman $\\&$ Sutherland (henceforth RS) proposed a model for pulsar emission which postulated a gap region located immediately above the neutron star surface and, as a result of the inability of the electric field to remove ions from the surface, subject to an intense potential difference of around $10^{14}eV$. In this gap pair-production in the intense magnetic field generated electrons which heated the surface and created localised discharging regions. These `sparks' drifted around the magnetic pole and ejected energetic particles into the magnetosphere where a bunching mechanism, first proposed by Sturrock (1971), created coherent radiation following a two-stream instability. The model has been highly influential, since it provides a quantitative framework within which theorists and observers can work, and one of its underlying tenets, that the sources of pulsar radio emission are plasma columns circulating just above the polar cap, is now widely accepted.  Nevertheless, over the years the polar gap model has been questioned on theoretical grounds, largely through difficulties with the neutron star surface binding energy (e.g. Jones 1985, 1986, Abraham $\\&$ Shapiro, 1991, Neuhauser et al, 1987) and with the high plasma densities required in the emisson regions (Lesch et al, 1998).  On observational grounds too it is not easy to reconcile the model with the complexity of subpulse behaviour, such as stationary or counter-drifting subpulses (e.g. Biggs et al. 1985, Nowakowski 1991), mode changing and nulling. Evidence that the emission profiles generally take the form of nested cones (Rankin 1983, Lyne $\\&$ Manchester 1988, Rankin 1990(RI), 1993a(RII), 1993b) has presented a further challenge to the model, and only by appealing to surface multipoles (e.g. Gil et al, 2002a,b) does it seem possible to confine the `sparks' either radially or azimuthally. More recently, the precisely drifting emission columns of PSR0943+10, analysed in great detail by Deshpande $\\&$ Rankin (1999, 2001 (henceforth DR)), convincingly confirm the picture of circulating plasma columns. However the observed circulation rate is only reconcilable with the RS model by adopting a potential difference across the cap which is significantly lower than that predicted by the RS model, and - if generated by surface magnetic features - the columns would seem to require multipoles with a high degree of regularity (Asseo $\\&$ Khechinasvili 2001, Gil et al 2002a,b).  In response to these difficulties, some authors have analysed magnetospheres based on free electron flows from the polar cap and gradual acceleration into the upper magnetosphere (`inner accelerator' models) (e.g. Arons $\\&$ Scharlemann 1979, Mestel $\\&$ Shibata 1994, Hibschmann $\\&$ Arons 2001a,b (henceforth HAa,b), Harding $\\&$ Muslimov 2002 (HM), Harding et al 2002 (HMZ), Harding $\\&$ Muslimov 2003).  Such models are often part of wider attempts at creating a self-consistent global magnetosphere (Mestel et al 1985, Shibata 1995, Mestel 1999 and references therein). Although arguably more consistent with the physics of the neutron star surface - and with the overall current balance and torque transfer requirements - these models lack the predictive power of the RS model when faced with detailed radio observations. Possibly the belief prevails that radio emission, undoubtedly originating just a few tens of stellar radio above the surface and energetically weak compared to the spin-down energy loss, has little to say about global conditions. Furthermore the sheer complexity of highly time-dependent subpulse phenomena acts as a great deterrent to theoreticians seeking to define a steady state condition.  In trying to bring the large-scale magnetosphere models closer to observational testing, and in a deliberate attempt to explore the link between the the observed polar cap activity and the outer magnetosphere, we have taken a fresh look at the radio profiles analysed in RI and RII. Pulsar integrated profiles are suitable to this purpose, since they are one feature of pulsar observations which displays great stability: a pulsar's profile is its invariant signature. Rankin's conal classification of the profile forms is here interpreted essentially from a geometric standpoint without prejudice for or against any particular formation process. In essence, we seek to use the profiles as diagnostics of the magnetosphere's structure.  This paper argues that if regular conal structures exist in all or most pulsar profiles - as claimed in RI and RII - then explanations in terms of complex polar magnetic field topology are difficult to support, since they would require similar magnetic topologies from pulsar to pulsar.  However it is suggested here that conal emission may be the natural geometric result of interactions between the polar cap and outer regions of the outer magnetosphere in the purely dipolar environment of `slow` older pulsars, provided that the once prolific outergap production of pairs of a pulsar in its early life is now weak and intermittent. This proposed evolutionary link between the younger faster pulsars (such as the Crab pulsar), whose principle emission is in the gamma-ray band, and the older weaker radio pulsars is seen as a major and novel feature of the model here.  First, in section (2), it is argued that the fixed ratio of the cone radii is consistent with the assumption that emission occurs preferentially on two critical sets of field-lines, one bounding the corotating dead zone (defined by the last field-line to close within the light-cylinder), the other passing through the intersection of the null line (defined as the boundary between regions of net charge density of opposing sign) and the light cylinder. In section (3) it is shown that the observed period dependence of conal types can be simply explained if emission is possible only if a `mirror' point (i.e. an intersection of the null line with the critical field-lines) lies between two fixed altitudes. In section (4) it is shown how the inevitable drift of the inflowing and outflowing particles about the magnetic axis causes `emission nodes' to form above the surface, and, in section (5), that this leads to their precession around the magnetic pole, generating the well-known phenomenon of `drifting subpulses' and hence the double cone structure. In section (6) the drift model is applied to the pulsar PSR0943+10 and precise fits to both the B-mode and the Q-mode are found. In section (7) the global nature of the pulsar phenomenon is stressed: azimuthal interactions between critical regions of the outer magnetosphere are shown to explain the observed dependency of conal formation on the angle of inclination. Finally, in section (8), the physical requirements of the underlying feedback system are discussed in the light of current physical ideas.  The sketched model which emerges from this analysis contains many elements of the RS model, but on a scale which involves the entire magnetosphere: particles (presumed here to be electrons, although the system is sign-reversible) are accelerated from low Lorentz factors close to the polar cap to achieve high $\\gamma$ near the outer gap, which stretches from the corotating dead zone of the magnetosphere to the light cylinder. Pair creation occurs in these regions, although certainly not in the profusion of the pair cascades supposed in Crab-like pulsars (Cheng et al, 1986, Romani $\\&$ Yadiaroglu, 1995). Rather the production is likely to be intermittent, weak (ie of low multiplicity with low Lorentz factors) and azimuthally-dependent, with most of the particles produced forming a wind beyond the light cylinder, but with a small but essential fraction of the positrons returning to the surface, accelerated and funnelled by the increasingly tight bundle of magnetic field-lines above the poles (Michel 1992).  Unlike the case of fast pulsars (Cheng et al, 1986), the downward flow of particles is inadequate to screen the polar potential.  Shortly before reaching the surface, the positrons emit sufficiently energetic radiation to cause a bunched avalanche of pairs which bombard the surface with coherent radiation. This is then reflected and contributes to the core component of the pulsar profile. Residual electrons formed by this process (maybe augmented by electrons emitted from the heated spot on the surface) are then accelerated back to the outer gap. This process is then repeated, and is reinforced if the combined and equal drifts of the electrons and positrons around the pole maintain the hot spots at locations which are either fixed in azimuth or drift slowly backwards or forwards. The outflowing electrons are somehow stratified by the inflowing layers of pairs, and radiate curvature radiation parallel to the field lines and at about 200 km (for 1GHz) above the surface. The holistic nature of the model means that the radio emission can simply be seen as an image of, and as driven by, the activities of the outer gap.  The emission, particle flow and pair-creation processes are not worked out in detail here, and in many cases have been cannibalised from existing models (especially Mestel et al (1985), Michel (1992), Shibata (1994), HAa,b, HM, HMZ, Hirotani $\\&$ Shibata 1999, 2001 (HSa,b)). Some features lack, as yet, a proper theoretical investigation: for example, the mechanism and level of weak low-energy pair production in outergaps has never been explored, since hitherto it has been assumed that outergaps only play a significant role in the emission of fast, young pulsars. On the other hand, the recent insight (see HAa,b, HM, HMZ, above) that in older pulsars, whose pair production at the polar cap relies on Inverse Compton Scattering, there will be a residual potential to accelerate particles up towards the outergap (and by implication could accelerate returning particles of an opposite charge) gives support to a fundamental aspect the model.  However we strongly stress througout the paper that the direction and purpose here is to establish the broad characteristics of a workable model, based on observational rather than theoretical grounds. In doing so we clarify not only what features such a model needs, but also what is not needed: there is no polar gap, no pair-creation in the outflow before the outer gap is reached, no self-stratification of the outflow to generate the coherence of radio emission, and no multipoles. ",
        "conclusions": "In this paper an attempt has been made to create a feedback model for the pulsar magnetosphere which can be directly related to observations. It sees the magnetosphere as an integral whole, where the polar cap and the outergap are mirrors of each other. It is suggested that the inner and outer cones found in integrated profiles reflect the two intersections of the Holloway null surface with the light cylinder and the corotating dead zones respectively, and the core components of profiles quite literally reflect the feedback from these intersections. This simple interpretation of the profiles can reproduce with some accuracy the observed dependence of profile features on period and angle of inclination. The observed regularity of profile features from pulsar to pulsar is difficult to explain by the arbitrariness of complex surface magnetic fields.  An important feature of the model is that for the first time it suggests an natural evolutionary link between old slower radio pulsars and young Crab-like pulsars whose predominantly high-energy X-ray/$\\gamma$-ray emission is thought to emanate from the outergap regions of an inclined rotating dipole (Romani $\\&$ Yadigaroglu 1995). As a pulsar ages, cascades in the outergap accelerated regions fade (Ruderman $\\&$ Cheng 1988), but a meagre yet critical pair production still lingers, which, by interacting with polar cap production, creates the coherent radio emission we observe in the complex radio profiles.  The drift of particles in the entire magnetosphere often gives rise to discrete rings of emission nodes near the polar cap. They can move in either direction, and will have differing behaviour in different rings. The particle flow and the observed motion of the nodes may be chaotic, or adopt a `resonant' state where the particle circulation is harmonically coupled to the star's rotation rate. PSR0943+10 is shown to be in such a state, and it is argued that conditions in other stars may permit the star to cycle through several resonant states of stepwise increasing charge screening along open field-lines. In general, the observed subpulse drift patterns are only the residual drift of the particle circulation, and hence are sensitive to small variations in the ambient electrical potential.  The angle of inclination plays an important role in determining whether a pulsar successfully creates emission nodes and hence observable coherent radiation. It changes the position of the null surfaces and thereby fixes the degree to which the magnetosphere is interconnected by means of rings of azimuthally drifting particles around the light-cylinder. The nature of the observed dependency of inclination on conal type suggests that the flow of particles between the core region of the magnetic polar cap and the light-cylinder is an integral part of the system, and this can be shown to further imply that the two magnetic poles will generally be linked in a single system.  A further constraint on the formation of cones, suggested by the observed dependency of conal type on the rotational period and made plausible by the interpretation of this model, is that emission nodes, and their resulting integrated emission cones, can only form if the mirror points lie between fixed altitudes of approximately 20,000km and 70,000km.  If core emission is caused by radiation reflected from the neutron star`s surface, then it is a prediction of the model that on close inspection the core component will have a structure of miniature cones (an effect possibly easier to detect in younger pulsars).  However the model leaves many theoretical questions unanswered. How exactly can the pair creation process be made to work in the outer gap (HSa,b), and how many positrons can really be available for the backflow to the surface? To what degree is the proposed slow acceleration of particles between the polar cap and the outergap (HAa,b, HM, HMZ) affected by a returning charge flow of opposite sign and with azimuthal dependence? Can an emission process be made to work as slow electrons pass upwards through the descending pair avalanche? A more radical question is whether we need particles from the surface at all to make the radio emission. Then the model could apply to pulsar and antipulsar alike. Finally we could speculate, as in the global model of MRWW, that the outer ring represents a net flow of negative charge to the surface, surrounding and balancing the positive flow within the inner cone, and thereby solving the long-standing current balance problem, although in sufficiently inclined pulsars non-equatorial rings could close the current in the azimuthal plane (see Section 7). This model has many features needing theoretical attention, but its aim is to forge a closer link between observation and theory, and the author will be pleased if something of this is achieved."
    },
    "0208/astro-ph0208311_arXiv.txt": {
        "abstract": " ",
        "introduction": "Gamma-ray bursts (GRBs) may be the most enigmatic objects in present-day astronomy. Among their peculiar properties, highly aperiodic time variability is undoubtfully one important aspect. Among various timing analyses, we pay special attention to one by Li and Fenimore (1996), who found log-normal distributions for the peak fluence and peak time intervals of GRBs (for careful and detailed analyses, see Nakar, Piran 2002; Quilligan et al. 2002). Although its physical meaning is not obvious, it is expected that this feature may contain an important clue to understanding the central engine of GRBs.  Apparently similar, highly aperiodic intensity fluctuations are known in X-rays from Galactic black hole candidates (GBHCs) in the hard/low state, though, in the other states, similar, highly aperiodic variations have not been observed, except for rather periodic flares known as quasi-periodic oscillations (e.g., van der Klis 1995). One may thus ask if the variability properties of GBHCs in the hard state share common characters with GRBs.  An affirmative answer is actually anticipated in a sense,  since the central engines of GRBs are often discussed in terms of accretion models (e.g., Narayan et al. 1992, 2001). The basic idea underlying these models is that, whatever the origin might be (black hole--neutron star mergers, black hole--stellar core mergers, and so on), the final configuration may likely be a stellar-mass black hole surrounded by a massive accretion disk or torus with a mass ranging between 0.01--1.0 $M_\\odot$. If so, it is natural to expect some similarities to exist in gamma-ray and X-ray variabilities of GRBs and GBHCs. This expectation becomes even more strengthened by the recent discovery of log-normal distributions in the blazar variability by D. Yonetoku and T. Murakami (in preparation).  However, negative results have already been reported based on Ginga observations of the prime GBHC, Cygnus X-1 (Negoro et al. 1995, hereafter N95). The time intervals between adjacent flare-like events (called X-ray shots), $\\Delta t$, basically follow exponential distributions with no peak, as expected from random (Poisson) distributions. They also found that the peak intensities were exponentially distributed.  Yet, the entire story is not over. At the same time, they discovered an interesting suppression of the shot occurrence at short time intervals ($\\Delta t <$ 5--8 s). The duration of the suppression, ``waiting time'', is longer for shots with a larger peak intensity. Using RXTE data, Focke (1998) also found that a peak interval distribution could not be represented by a simple exponential function, but by a broken exponential function. In this Letter, we put forward this discovery by performing new analyses of shots with large peaks using the same Ginga data of Cyg X-1 with N95, and discuss its physical meaning. ",
        "conclusions": "In the present study we re-examined the temporal varibility of Cyg X-1 and found for the first time a similarity to the variability of GRBs. The similarity and differences should contain important physics to probe the central engine of GRBs.  The most intriguing difference is that the Cyg X-1 shots, {\\it including small ones}, have  smooth (exponential) peak-interval and peak-intensity distributions, and GRBs and blazar flares, on the other hand, have log-normal distributions, indicating the presence of a typical timescale and size of energy. Note that a smooth (power-law) peak-intensity distribution is known to exist in solar flares (Dennis 1985), and is also found in 3D MHD flow simulation data (Kawaguchi et al. 2000; Agol et al. 2001). Such a power-law distribution is claimed to be ubiquitous in nature (Bak 1996), and is often discussed in terms of the dynamics of diffusion systems with interacting degrees of freedom (self-organized criticality, see Bak et al. 1988; Mineshige et al. 1994; arguments about the exponential and power-law distributions, see Takeuchi et al. 1995). It can be conjectured that MHD turbulence created by various MHD processes in accretion disks exhibit spatial fractal patterns which could be responsible for the smooth distributions of flare amplitudes (Kawaguchi et al. 2000).  On the other hand, the typical peak interval duration, 7--8 s, found in the (log-normal) peak interval distributions for {\\it large} shots, seems to correspond to the timescales of filling/refilling energy in the innermost part of the accretion flow before/after the shot occurrence (e.g., N95). What happens if a large amount of (sudden) accretion takes place in the inner part of the accretion disk? VLBI observations of the extragalactic jets (Junor et al. 1999) and recent MHD jet simulations (Kudoh et al. 2002) indicate that jets originate from a compact region with a size of several tens to hundreds of Schwarzschild radii. It is of great importance to note that radio flares are observed only after large X-ray flares in the Galactic microquasar, GRS 1915$-$105 (Mirabel et al. 1998). Furthermore, recent discovery of short-term temporal correlation between X-rays and optical in the BHC, XTE J1118+480, strongly suggests that the outflow (or jet) follows the shot (Kanbach et al. 2001).  It may be that jets can be produced only when large enough disturbances have been added to the innermost part of the accretion flow.  To summarize, we newly discovered a similar behavior in Cyg X-1 to that of GRBs and blazars, if we only selected large shots from the variability light curves of Cyg X-1. There could also be large- and small-amplitude variations in accretion disks at the centers of GRBs, as in GBHCs, but only large variations (which satisfy a certain criterion) can produce observable bursts through jets in GRBs.  \\bigskip  This work was initiated through a discussion in the domestic GRBs workshop held at the Yukawa institute. HN thanks the Yukawa institute for the hospitality to write a part of this paper. This work was supported in part by the Grants-in Aid of the Ministry of Education, Culture, Sports, Science and Technology of Japan (13640238, SM)."
    },
    "0208/astro-ph0208383_arXiv.txt": {
        "abstract": "We have identified the 1.4~GHz radio source FIRST J102347.6+003841 (hereafter FIRST J1023+0038) with a previously unknown 17th-mag Galactic cataclysmic variable (CV)\\null.  The optical spectrum resembles that of a magnetic (AM~Herculis- or DQ~Herculis-type) CV\\null.  Five nights of optical CCD photometry showed variations on timescales of minutes to hours, along with rapid flickering.  A re-examination of the FIRST radio survey data reveals that the radio detection was based on a single 6.6~mJy flare; on two other occasions the source was below the $\\sim$1~mJy survey limit.  Several other magnetic CVs are known to be variable radio sources, suggesting that FIRST J1023+0038 is a new member of this class (and the first CV to be discovered on the basis of radio emission).  However, FIRST J1023+0038 is several optical magnitudes fainter than the other radio-detected magnetic CVs.  It remains unclear whether the source simply had a very rare and extraordinarily intense radio flare at the time of the FIRST observation, or is really an unusually radio-luminous CV; thus further observations are urged. ",
        "introduction": "Most of the known cataclysmic variables (CVs) were originally discovered because of their dramatic optical variability, exemplified by the outbursts of dwarf novae and classical novae. More recently, however, ultraviolet and X-ray sky surveys have led to discoveries of many new CVs, including members of subclasses that have less spectacular optical light variations (e.g., the novalike variables).  Follow-up investigations at radio wavelengths have led to detections of a few well-known, mostly nearby, CVs as radio sources.  In this paper we report what we believe to be the first case in which the sequence of discovery has been reversed: we have detected a new CV during a radio survey, with the subsequent confirmation having been made in the optical band. ",
        "conclusions": "The first CV to be detected at radio wavelengths was the prototypical magnetic CV AM Herculis, whose emission was discovered at the VLA by Chanmugam \\& Dulk (1982).  They detected a flux density of 0.67~mJy at 4.9~GHz, and suggested that the emission is from mildly relativistic electrons trapped in the magnetosphere of the white dwarf.  Subsequent observations by Dulk,  Bastian, \\& Chanmugam (1983) led to detection of a strong flare from AM~Her, which lasted about 10~min and reached a peak of 9.7~mJy at 4.9~GHz; Dulk et al.\\ were also able to set an upper limit on the quiescent flux density of AM~Her at 1.4~GHz of 0.24~mJy.  The magnetic DQ Her-type system AE Aqr has also been detected as a variable radio source, with the initial discovery having been made at the VLA by Bookbinder \\& Lamb (1987); they found a flux density at 1.4~GHz varying from 3 to 5 mJy.  Other radio detections of AM Her- and DQ Her-type CVs have been reported by Pavelin, Spencer, \\& Davis (1994) and references therein. By contrast, non-magnetic CVs in general are not detectable radio sources (Nelson \\& Spencer 1988), strongly suggesting that a highly magnetic white dwarf is a key element in the production of radio emission.  For further information, the reader is directed to the summary of radio emission from CVs by Mason, Fisher, \\& Chanmugam (1996).  We thus strongly suspect that FIRST J1023+0038 is a new magnetic CV, either of the DQ~Her or AM~Her variety. Our detection of a 6.6~mJy flare at 1.4~GHz is remarkable because, with a quiescent optical magnitude near 17.5, FIRST J1023+0038 is so optically inconspicuous. (By comparison, AE~Aqr is an 11th-mag object, and AM Her lies generally around 12th-13th mag, although with occasional drops to ``low'' states below 15th~mag). It remains to be seen whether the flare that led to our serendipitous discovery of FIRST J1023+0038 was simply a very rare and unusually energetic one, or whether this object is indeed much more radio luminous than the typical magnetic CVs.  We urge follow-up observations at radio, optical, and X-ray wavelengths. Optical spectroscopy should reveal the orbital period, and might provide direct evidence for a strong magnetic field.  Polarimetry would determine whether the object is a highly magnetic AM~Her system.  A search of archival plate collections, and future long-term photometric monitoring, might reveal either dwarf-nova outbursts, or AM~Her-like low states, although the object is rather faint."
    },
    "0208/astro-ph0208456_arXiv.txt": {
        "abstract": "We review the detectability of gravitational waves  generated by oscillations excited during a phase transition from hadronic matter to deconfined quark-gluon matter in the core of a neutron star. Neutron star properties were computed using a Boguta and Bodmer's based model and the MIT bag model. The maximum energy available to excite mechanical oscillations into the star is estimated by energy difference between the configurations with and without a quark-gluon matter core. On basis of the planned sensitivity of present laser interferometers (VIRGO or LIGO I) and those of the next generation (LIGO II), the maximum volume to be probed by these experiments is determined. These results are used as an indication of the potential detectability of neutron stars as sources of gravitational waves. Our results indicate that the maximum distance probed by the detectors of the first generation is well beyond M31, whereas the second generation detectors will probably see phase transition events at distances two times longer, but certainly not yet attaining the Virgo cluster. ",
        "introduction": "The present model for the nuclear matter was already discussed by Marranghello et al.\\cite{marranghello01}, who have investigated the effects of a finite temperature in the equation of state. Here, the same description for dense matter is adopted but only models with T=0 will be considered. Moreover, the coupling constants are slightly modified with respect to those considered in that work, in order to allow the maximum mass of the configuration to be compatible with recent results on the binary X-ray system Vela X-1\\cite{barziv01}. For the sake of completeness, we recall here the main points of the model.  The lagrangian density describing the nuclear matter is  \\begin{eqnarray} {\\cal L}    &=& \\sum\\limits_{B}   \\bar{\\psi}_{B} [( i\\gamma_\\mu (\\partial^\\mu- g_{\\omega B} \\omega^{\\mu}) - (M_B-g_{\\sigma B} \\sigma) ]\\psi_B \\nonumber \\\\ && - \\sum\\limits_{B}   \\bar{\\psi}_{B} [ \\frac12 g_{\\varrho B} \\mbox{\\boldmath$\\tau$} \\cdot \\mbox{\\boldmath$\\varrho$}^\\mu] \\psi_B +\\frac{bM}{3}\\sigma^3+\\frac{c}{4}\\sigma^4 \\nonumber \\\\ &&+\\frac12(\\partial_\\mu \\sigma \\partial^\\mu \\sigma   - {m_\\sigma^2} \\sigma^2)  - \\frac14  \\omega_{\\mu \\nu}  \\omega^{\\mu \\nu}   + \\frac12 {m_\\omega^2} \\omega_\\mu \\omega^\\mu  \\nonumber \\\\ && -   \\frac14 \\mbox{\\boldmath$\\varrho$}_{\\mu \\nu} \\cdot \\mbox{\\boldmath$\\varrho$}^{\\mu \\nu} +  \\frac12m_\\varrho^2 \\mbox{\\boldmath$\\varrho$}_\\mu \\cdot  \\mbox{\\boldmath$\\varrho$}^\\mu \\nonumber \\\\ &&+\\sum\\limits_{l}   \\bar{\\psi}_{l} [i \\gamma_\\mu \\partial^\\mu   - M_l] \\psi_l \\,\\, . \\end{eqnarray} This equation represents nuclear matter as composed by a mixture of the fundamental baryon octet (p, n, $\\Lambda$, $\\Sigma^+$, $\\Sigma^0$, $\\Sigma^-$, $\\Xi^-$, $\\Xi^0$) coupled to three mesons ($\\sigma, \\omega, \\varrho$) and leptons (for the details see [15]). The scalar and vector coupling constants, g$_{\\sigma}$, g$_{\\omega}$ and the coefficients b, c were determined by imposing that the model bulk properties should be able to reproduce the binding energy E$_b$ (= -16.3 MeV), the compression modulus K (= 240 MeV) and the nucleon effective mass $M^* = M -g_{\\sigma} \\bar{\\sigma}$ (= 732 MeV) at the saturation density $\\rho_0$ (=0.153 fm$^{-3}$). Additionally, the isovector coupling constant g$_{\\varrho}$ is determined from the coefficient for the symmetry energy in nuclear matter, a$_4$ (= 32.5 MeV). We have used the universal hyperon-nucleon coupling ratios $\\chi_i = g_{Hi}/g_i$, with $i = \\sigma, \\varrho, \\omega$. The resulting coefficients used in our computations, are given in table 1. Figure 1 shows the equation of state derived from our model and in figure 2 it is shown the energy density profile inside the star for two configurations having gravitational masses equal to 1.2 and 1.6 M$_{\\odot}$ respectively.  \\begin{table} \\caption{\\label{tab:table1} } \\begin{ruledtabular} \\begin{tabular}{lcrlc} $(g_{\\sigma}/m_{\\sigma})^2$&$(g_{\\omega}/m_{\\omega})^2$&$(g_{\\varrho}/m_{\\varrho})^2$&b$(\\times 100)$&c$(\\times 100)$ \\\\ \\hline 9.927&4.820&4.791&0.8659&-0.2421\\\\  \\end{tabular} \\end{ruledtabular} \\end{table}  \\begin{figure}[htb]% \\vspace*{10pt} % \\vspace*{1.4truein}             % \\vspace*{10pt} \\parbox[h]{4.5cm}{ \\special{psfile=prd3.ps angle=-90 hoffset=-130 voffset=160 hscale=35 vscale=35}} \\vspace{20pt} \\caption{Equation of state for hadronic matter (solid line) and for the quark-gluon matter (dotted line). \\label{1}} \\end{figure}  \\begin{figure}[htb]% \\vspace*{10pt} % \\vspace*{1.4truein}             % \\vspace*{10pt} \\parbox[h]{4.5cm}{ \\special{psfile=prd2.ps angle=-90 hoffset=-130 voffset=160 hscale=35 vscale=35}} \\vspace{20pt} \\caption{Neutron star energy density-radius relation for M=1.6M$_\\odot$ (solid line) and M=1.2M$_\\odot$ (dotted line) on which the quark-gluon core extends up to 8.6 km and 7 km, respectively. \\label{3}} \\end{figure}  \\begin{figure}[htb]% \\vspace*{10pt} % \\vspace*{1.4truein}             % \\vspace*{10pt} \\parbox[h]{4.5cm}{ \\special{psfile=prd4.ps angle=-90 hoffset=-130 voffset=160 hscale=35 vscale=35}} \\vspace{20pt} \\caption{Neutron star mass as a function of central energy density for hybrid star with constant pressure transition (solid line), hybrid (dotted line) and pure hadronis star (dashed line). configurations \\label{2}} \\end{figure}  Hybrid models including a quark-gluon core were also computed. The quark matter was described by the MIT Lagrangian\\cite{grand75} and the physical conditions at the deconfinement transition were estimated from the Gibbs criteria, namely, by the equality of the chemical potential and pressure of both phases, under conservation of the baryon number and electrical charge. The physical parameters at the transition point are given in table 2, for a bag constant $B^{1/4}=180 MeV$ and a strange quark mass m$_s$ = 150 MeV. The gravitational mass of the star as a function of the central energy density is shown in figure 3 for the case of a pure hadronic configuration (dashed line), a hybrid star with a quark core (solid line) and, for comparison, a case where a mixed phase exist in the center, compute in the same way as ref. \\cite{glendenning92}.  It should be emphasized that our equation of state is quite steep (${{dlogP}\\over{dlog{\\varrho}}} \\approx 2.6$) near saturation and, as a consequence, the deconfinement transition occurs at densities just above the saturation value ($\\rho \\sim 1.7\\rho_0$), producing hybrid stars with very extended quark-gluon cores. Here we take the opportunity to reiterate that, from the actual status of theoretical predictions to the EOS, with so many parameters to be adjusted, even in the quark phase and specially in the hadron phase, we believe that only qualitative results can be obtained with some insights on the quantitative results.  \\begin{table}[htb] \\begin{center} \\caption{Physical conditions at the  phase transition: energy densities and pressure are given in GeV fm$^{-3}$ } \\vspace{0.5cm} \\begin{tabular}{|lll|} \\hline $\\epsilon_H$ & $\\epsilon_q$ &P \\\\ \\hline 0.236 & 0.349 & 0.0163\\\\ \\hline \\end{tabular} \\end{center} \\end{table}    The maximum stable mass of a pure hadronic configuration is about M$_{max} \\approx$ 2.1 M$_{\\odot}$ while for hybrid stars this limit is reduced to 1.73 M$_{\\odot}$. Thus, the present calculations exclude the possibility that Vela X-1 has a quark core. Notice that our models obey the Seidov criterium\\cite{seidov71}, namely, that the hybrid star will be stable only if the energy jump across the transition surface satisfies the condition \\begin{equation} {{\\epsilon_q}\\over{\\epsilon_H}} < {{3}\\over{2}}(1 + {{P}\\over{\\epsilon_H}}) \\end{equation} where  P, e$_q$, e$_H$ are respectively the pressure, the energy density of  quarks and hadrons at the transition point.  One of the goals of this work is the determination of an upper limit for the energy able to excite the different oscillation modes of the star, when a modification in its internal structure occurs. In this sense, the simplest approach is to compute the energy difference between two configurations having the {\\it same baryonic number}; the first constituted of pure hadrons (H), the second having a core of deconfined matter (HQ), and use such a difference as an indication of the maximum available energy. Table 3 gives the parameters for five models defined by a given baryonic number\\cite{Weinberg} \\begin{equation} N = \\int_0^R 4 \\pi r^2 \\left[ 1 - \\frac{2 G m(r)}{r^2} \\right]^{-1/2} \\rho_B(r) dr \\, , \\end{equation} or the baryonic mass of the star ($M_{bar} = N M_B$); the expected gravitational mass of both configurations (pure hadron and hybrid), \\begin{equation} M_g = m(R) =  \\int_0^R 4 \\pi r^2 \\epsilon(r) dr \\, ; \\end{equation} expected radii, masses of the quark-gluon core and the energy difference obtained from the passage of configuration H to HQ, the binding energy. \\begin{equation} E_g = M_g - M_{bar} \\, . \\end{equation}   \\begin{table}[htb] \\begin{center} \\caption{Parameters of the stellar models: masses are in solar units and radii in km; H and HQ mean pure hadron and hybrid configurations respectively} \\vspace{0.5cm} \\begin{tabular}{|lllllll|} \\hline M$_{bar}$ & M$_g$ (H)&M$_g$ (HQ)&M$_g$ (core)& R (H)& R(HQ)& log (E) erg\\\\ \\hline 1.0749&1.0003&0.9625&0.1150&12.82&12.47&52.83\\\\ 1.3202&1.2009&1.0953&0.3608&13.04&12.16&53.28\\\\ 1.5724&1.4008&1.2468&0.6013&13.13&11.98&53.44\\\\ 1.8342&1.6006&1.4164&0.8626&13.01&11.82&53.52\\\\ 2.1045&1.8002&1.6141&1.1766&12.69&11.50&53.52\\\\ \\hline \\end{tabular} \\end{center} \\end{table} ",
        "conclusions": "The structure of pure hadronic configurations and that of hybrid stars having a quark core, with the same baryonic number, were computed using a new equation of state\\cite{marranghello01}. The maximum stable mass for pure hadronic configurations satisfies the requirement of being higher than the mass of Vela X-1 ( M = 1.86$\\pm$0.16 M$_{\\odot}$), determined recently\\cite{barziv01} from a new study of its orbital motion.  As one should expect,  the masses of quark cores increase with the baryon number of the configuration, as well as the difference between the binding energies between hybrid and single phase objects. On the basis of this result and from the point of view of the GW emission, it would be natural to expect that phase transitions occurring in more massive stars would be more easy to detect. In fact, this is not the case because, in the one hand the damping timescale due to the emission of GWs is inversely proportional to the mass and hence massive stars have low oscillation quality factors and, on the other hand the detectors have sensitivities optimized at frequencies around 200 Hz. In comparison with non-radial oscillations (for instance m= $\\ell$ = 2 modes), these have comparable frequencies but damping timescales one order of magnitude higher than those resulting from radial modes coupled to rotation. In this case, higher quality factors Q may be obtained but most of the mechanical energy will be dissipated into heat. In order to be an efficient source of GWs by the mechanism here considered, the star must be a fast rotator, i.e., to have rotation periods of the order of few milliseconds. Even in the most favorable case (model 2, corresponding to a baryonic mass equal to 1.32 M$_{\\odot}$), if the rotation period is 4.0 ms, only 1\\% of the available energy will be emitted as GWs.  Inspection of table 4 indicates that the maximum distance probed by detectors of first generation (VIRGO, LIGO I) is about 6.4 Mpc, well beyond M31, whereas the second generation (LIGO II) will probably see phase transition events at distances two times longer, but certainly not yet attaining the Virgo cluster. The small probed volume and the rapid rotation required for this mechanism be efficient imply in a low event rate, imposing  severe limitations on the detectability of such a signal."
    },
    "0208/astro-ph0208510_arXiv.txt": {
        "abstract": "This paper considers some simple surface brightness (SB) estimates for galaxies in the Automated Plate Measuring Machine (APM) catalogue in order to derive homogeneous SB data for a very large sample of faint galaxies. The isophotal magnitude and area are used to estimate the central surface brightness and total magnitude based on the assumption of an exponential SB profile. The surface brightness measurements are corrected for field effects on each UK Schmidt  plate and the zero-point of each plate is adjusted to give a uniform sample of SB and total magnitude estimates over the whole survey. Results are obtained for 2.4 million galaxies with blue photographic magnitudes brighter than b$_J$ = 20.5 covering 4300 deg$^2$ in the region of the south galactic cap. Almost all galaxies in our sample have central surface brightness in the range 20 to 24 b$_J$ mag arcsec$^{-2}$. The SB measurements we obtain are compared to previous SB measurements and we find an acceptable level of error of $\\pm 0.2$ b$_J$ mag arcsec$^{-2}$. The distribution of SB profiles is considered for different galaxy morphologies for the bright APM galaxies. We find that early-type galaxies have more centrally concentrated profiles. ",
        "introduction": "\\footnotetext{E-mail: zyshao@@center.shao.ac.cn}  \\label{sec_intro}  Surface brightness (SB) is one of the fundamental parameters describing a galaxy. It plays an important role in many diverse aspects of extragalactic astronomy and cosmology, from identifying the whole family of galaxy populations to modelling their different distributions and motions in the Universe. These aspects are generally related to the formation and evolution of galaxies, as well as to the nature of large scale structures in the Universe.  Unfortunately, accurate measurements of the SB of galaxies rely critically on high quality observations with perfect sky conditions and adequate telescope time, so it is not easy to get high quality SB data for galaxies, especially for low surface brightness galaxies (LSBG). Large, homogeneous data samples are even more difficult to construct, and so uniform SB data for galaxies is very scarce compared to other parameters, such as position, colour, redshift and morphology. Hitherto relatively few sets of SB measurements have been available; the total number of such galaxies is only a few thousand, and the samples have either concentrated on LSBGs or been limited only to bright galaxies (e.g.  de Jong \\& van der Kruit 1994; de Jong 1996; Jansen et al. 2000; Heraudeau \\& Simien 1996; Lauberts \\& Valentijn  1989; Impey et al. 1996; Morshidi-Esslinger et al. 1999, hereafter MDS).  On the other hand it is not  difficult to make a rough estimate of a galaxy's SB using  parameters related to both the luminosity and size of a galaxy.  These two kinds of basic parameters, such as magnitude and effective radius of galaxies, are common in many catalogues.  One can consequently obtain a larger sample of SB data of galaxies at the expense of  some accuracy for individual objects. The statistical analysis of a large number of galaxies will not sensitively depend on the accuracy of the SB measurement for each individual galaxy because reasonable errors in SB can be allowed for in the analysis.  The APM catalogue includes both the magnitude and isophotal size for each galaxy so these can be used to estimate the SB of galaxies. Based on this strategy, some work has already used SB estimates from the APM survey. C\\^{o}t\\'e et al.  (1999) used the mean isophotal SB of APM galaxies to select the candidates for an HI survey of LSBGs. More recently Cross et al.  (2001) calculated the effective SB of galaxies in the 2dF Galaxy Redshift Survey and estimated their bivariate brightness distribution, as well as the number and luminosity density of galaxies.  In this paper, we consider a variety of SB estimates, and present a method for determining the central surface brightness, $\\mu_0$, of galaxies in the APM Galaxy Survey.  We apply a field correction across each survey plate, and use a matching procedure over the whole survey area to yield a homogeneous SB sample of over 2 million galaxies with a wide magnitude range over a large volume of space. The results will be used in future papers to study the distribution of galaxies as a function of surface brightness, in particular their clustering properties. We describe the basic data in Section~\\ref{sec_apmcat}. The principles behind the method are described in Section~\\ref{sec measurement}, together with an introduction of the raw APM parameters that relate to the SB of galaxies. The field correction and matching process are briefly outlined in Section~\\ref{sec_corrections}, where we also discuss the uncertainty of our SB estimates and compare to other SB measurements. In Section~\\ref{sec_discussion} we consider the distributions of $\\mu_0$ and the SB profiles of bright APM galaxies. Finally we summarize our results in Section~\\ref{sec_summary}. ",
        "conclusions": "\\label{sec_discussion}  \\subsection{The Distribution of Surface Brightnesses} \\label{sec_sb}  Figure~\\ref{fig_mu0hist} shows a histogram of the frequency distribution of $\\mu_0$. The distribution is centred on $\\mu_0 = 22.2 $ mag arcsec$^{-2}$, and extends to $\\mu_0 \\simeq 20$ mag arcsec$^{-2}$ at the high SB end, and to $\\mu_0 \\simeq 24.5$ mag arcsec$^{-2}$ at the low SB end.  The star-galaxy separation criteria (Maddox et al. 1990a) mean that some very high SB images will be rejected from the galaxy sample because their profile is so similar to stars. However, the rapid drop in the apparent number of high SB galaxies will be due to their small isophotal areas and a real drop in the space density of high SB galaxies (as shown by Cross et al. 2001).  The lack of very low SB galaxies is less clearly a reflection of the true distribution of galaxy SBs, since a galaxy must have at least 4 square arcseconds brighter than 25 mag arcsec$^{-2}$ to be selected in the APM galaxy sample. The distribution is further constrained by the 20.5 mag limit on the Gaussian profile magnitudes which will also bias against low SB galaxies. Cross et al. (2001) have used the 2dF Galaxy Redshift Survey to estimate the distribution of intrinsic surface brightnesses, and conclude that, although the APM selection criteria do exclude low SB galaxies, the missing galaxies would not contribute significantly to the overall luminosity density. We will consider how our selection criteria affect the observed distribution in a future paper.   It is clear from Figure~\\ref{fig_mu0hist} that the $\\mu_0$ subsamples plotted in Figure~\\ref{fig_skymu}(a) and \\ref{fig_skymu}(c) represent the galaxies at the extremes of SB in the APM sample. This means that the distribution of galaxies in the sub-samples is very sensitive to any residual errors in the uniformity of the $\\mu_0$ measurements, and so the apparent uniformity of the galaxies in Figure~\\ref{fig_skymu}(a) and \\ref{fig_skymu}(c) show that our corrections have given us a reliable and uniform set of $\\mu_0$ measurements.  \\begin{figure} \\centerline{\\hbox{\\psfig{figure=mu0hist4.ps,width=0.5\\textwidth,angle=-90}}} \\caption{The frequency distribution of the apparent SB, $\\mu_0$, for all galaxies in the survey. \\label{fig_mu0hist} } \\end{figure}  \\subsection{The profiles of APM bright galaxies} \\label{sec_profiles}  In this Section, we use galaxies from the APM Bright Galaxy Catalogue (Loveday 1996; hereafter APMBGC) to investigate how the profile of a galaxy depends on its morphological type, and test if the assumption of an exponential profile is valid. The galaxies in the APMBGC have $b_J < 16.44$ and have a relatively large angular size ($r_{\\rm iso}$) and large $\\sigma^{2}$. Although $\\sigma^{2}$ is not very sensitive to changes in the profile of galaxies, and the observational error is quite large, this parameter does contain useful information about the SB profile. Furthermore, the APMBGC galaxies have been morphologically classified by visual inspection of the UKS plates and we have analyzed the distribution of $\\sigma^{2}$ separately for both early and late type galaxies.  We calculated the ($p_{0}$, $r_{0}$) of each galaxy, assuming an exponential SB profile as in Section~\\ref{sec measurement}. Since we also know the apparent ellipticity $e$ for each galaxy (from the density weighted second moments), and the values of the detection threshold and saturation of the plate, we can simulate the observed APM density for each pixel. Any other observational parameters can then be estimated from this exponential model image. In particular, we have estimated the value of $\\sigma^{2}$, and compare this to the actual observed value. Obviously, if the original image is exactly an exponential profile, and there are no other observational errors, then the observed value, $\\sigma_{\\rm obs}$ will be exactly equal to that from the exponential model $\\sigma_{\\rm exp}$, so the ratio $\\sigma_{\\rm obs}/\\sigma_{\\rm exp}$ should be exactly unity. To test the sensitivity of $\\sigma$ to variations in the SB profile, we created simulated images that follow the $r^{1/4}$ law profile (eq.~\\ref{eq_pr4}), and analysed them using the assumption of an exponential profile. In this case we found that $\\sigma_{\\rm obs}/\\sigma_{\\rm exp} \\simeq 0.9$, with a small variation according to their real central surface brightness and apparent ellipticity. The same analysis for Gaussian profiles gave a ratio of $\\sigma_{\\rm obs}/\\sigma_{\\rm exp} \\simeq 1.1$.  \\begin{figure}  \\centerline{\\hbox{\\psfig{figure=bright-sgma.ps,width=0.5\\textwidth,angle=-90}}} \\caption{The frequency distribution of $\\sigma_{\\rm obs}/\\sigma_{\\rm exp}$ values for APM bright galaxies. The dotted line shows E and S0 galaxies; the solid line shows Spiral and Irr . An exponential profile would have $\\sigma_{\\rm obs}/\\sigma_{\\rm exp}=1.0$, an $r^{1/4}$ law profile would have $\\sigma_{\\rm obs}/\\sigma_{\\rm exp}=0.9$, and a Gaussian profile would have $\\sigma_{\\rm obs}/\\sigma_{\\rm exp}=1.1$. \\label{fig_sig2} }  \\end{figure}   In Figure~\\ref{fig_sig2}, we plot the frequency distribution of $\\sigma_{\\rm obs}/\\sigma_{\\rm exp}$ for early and late galaxies from the APMBGC.  There is a large dispersion in $\\sigma_{\\rm obs}/\\sigma_{\\rm exp}$ , even for only late type galaxies. Nevertheless, the peak of the distribution is very close to 1.0, which implies that the exponential profile is a reasonable approximation for most galaxies  (77\\% of the sample has $0.9< \\sigma_{\\rm obs}/\\sigma_{\\rm exp} < 1.1$).  For early type galaxies (E and S0), the dispersion is larger than for late type galaxies, and most of them have smaller $\\sigma_{\\rm obs}/\\sigma_{\\rm exp}$ values that are more consistent with an $r^{1/4}$ law profile.  This shows that, although the APM measurements span a rather limited range of SB, the $\\sigma$ parameter can distinguish between early and late-type galaxies.  For fainter APM galaxies, the observational uncertainty on $\\sigma^{2}$ will be much larger, and the small differences between the $r^{1/4}$, exponential and Gaussian assumptions mean that $\\sigma^{2}$ is not effective in constraining the SB profile.    We have described a simple method to estimate the central surface brightness, $\\mu_0$, of galaxies, allowing for saturation and isophotal effects, and applied it to all the galaxies in the APM galaxy survey. We have corrected the $\\mu_0$ estimates for the effects of vignetting and differential desensitization over the field of each UKS plate. We have also used the plate overlap areas to match the individual plate measurements, and so provide homogeneous $\\mu_0$ estimates for all the APM galaxies.  Internal and external checks of the uncertainties in $\\mu_0$ suggest that the uncertainty for an individual measurement is about 0.16 mag.  For very bright APM galaxies, the second radial moment of the SB profile, $\\sigma^2$, shows that late type galaxies are consistent with an exponential profile, whereas early type galaxies have a higher mean SB, consistent with an $r^{1/4}$ law.  This uniform set of SB estimates for the APM galaxies will be used in later papers to select samples of high and low SB galaxies and analyze differences between their distributions.\\\\   \\bigskip"
    },
    "0208/astro-ph0208276_arXiv.txt": {
        "abstract": "We describe first results of a spectroscopic probe of selected fields from the Grid Giant Star Survey.  Multifiber spectroscopy of several hundred stars in a strip of eleven  fields along $\\delta \\approx -17^{\\circ}$, in the range $ 12 \\lesssim \\alpha \\lesssim 17$ hours, reveals a group of 8 giants that have kinematical characteristics differing from the main field population, but that as a group maintain coherent, smoothly varying distances and radial velocities with position across the fields.  Moreover, these stars have roughly the same abundance, according to their MgH+Mgb absorption line strengths.  Photometric parallaxes place these stars in a semi-loop structure, arcing in a contiguous distribution between 5.7 and 7.9 kpc from the Galactic center.  The spatial, kinematical, and abundance coherence of these stars suggests that they are part of a diffuse stream of tidal debris, and one roughly consistent with a wrapped, leading tidal arm of the Sagittarius dwarf spheroidal galaxy. ",
        "introduction": "There has been a long and somewhat checkered history of observations attempting to isolate substructure in the Galactic halo \\citep[e.g.,][]{E79,SC87,C91,C93,MMH96}. While observations in the last decade have confirmed the existence of a few coherent groups of halo stars \\citep{C93,MMH96,H99}, more recent large-area CCD surveys are beginning to reveal the ubiquity of these structures \\citep{V01,DP01,N02}. Given the central importance of these probable relics of satellite accretion to the \\citet{SZ} model of the formation of the Galactic stellar halo, their implications for the formations of massive halos via the Cold Dark Matter hypothesis \\citep{N97}, and their usefulness for measuring the Galactic potential \\citep{jzsh99}, there is motivation to identify and catalogue Galactic halo streams and moving groups.  We present in this paper first results from the Grid Giant Star Survey (GGSS), in which we identify a new, coherent moving group of Galactic halo stars.  The GGSS probes 1302 evenly spaced, $\\sim 0.4 - 0.6$ deg$^{2}$ areas of the sky to (a) supply several thousand bright ($V \\lesssim 12.5$), sub-solar metallicity K giants for the Astrometric Grid of the Space Interferometry Mission \\citep[see][]{conf}, and (b) explore problems related to the structure and kinematics of the Milky Way. The photometric technique for discriminating late type giants from dwarf stars, which uses the Washington $M,T_2$ and $DDO51$ filters, is described by \\citet{paperI}, and is being used by several groups \\citep{mor00,gei02,paperII} to explore Galactic structure problems.  Further discussion of the reliability of the photometric selection of giant stars is given in \\citet{rebuttal}.  It bears reiteration here that follow up high resolution spectroscopy -R=50,000; 5000-5900A$^{\\circ}$ coverage; S/N$\\approx$25 observations undertaken to monitor the spectroscopic stability of GGSS candidates and search for possible companions-  of stars selected by the GGSS as metal-poor giants in a similar magnitude range as the stars discussed here has revealed 100\\% of candidates observed so far to be, in fact, metal-poor giants (Verne Smith, private communication). While preliminary inspection of $\\approx$300 candidates suggests that they are all giants, detailed cross correlation studies with model spectra of a randomly selected sub-sample of 38 stars confirms that they are indeed so with $-1.0 \\lesssim Log(g) \\lesssim 3.1$ and $T_{eff} \\lesssim 5500 K$ for every candidate.   GGSS photometry for $\\delta < +20^{\\circ}$ fields was obtained on the Swope 1-m telescope at Las Campanas Observatory, and spans $9 \\lesssim V \\lesssim 17$.  Pipeline reductions convert the three-filter data to photometric abundances and parallaxes for likely giants according to the prescription in \\citet{paperI}.  A more detailed discussion will be given elsewhere.  We describe here first results from a spectroscopic probe of the GGSS giant star sample from among the earliest fields processed through the photometric pipeline.  Samples of candidate giant stars in 11 fields along $\\delta\\approx-17^{\\circ}$ (J2000) and $ 12^{h}< \\alpha<17^{h}$ were observed with the Blanco 4-m telescope + Hydra multifiber system on UT 28-30 Mar 2000.  We limited ourselves to $M < 15$ giants for this initial probe of GGSS fields because of limited available observing time, obtained as brief ($< 1$ hour) hour angle ``fillers\" for other projects. The data were obtained using the Schmidt camera and 2$\\times$4k SITe CCD in conjunction with the KPGLD grating. When centered on the infrared calcium triplet region, this setup delivers radial velocity precision of $ \\sim 5$ km s$^{-1}$ (verified by repeat observations of test stars). ",
        "conclusions": ""
    },
    "0208/astro-ph0208089_arXiv.txt": {
        "abstract": "Based on simultaneous observations of solar flares in hard and soft X-rays we studied several aspects of the Neupert effect. About half of 1114 analyzed  events show a timing behavior consistent with the \\mbox{Neupert} effect. For these events, a high correlation between the soft \\mbox{X-ray} peak flux and the hard \\mbox{X-ray} fluence is obtained, being indicative of electron-beam-driven evaporation. However, for about one fourth of the events there is strong evidence for an additional heating agent other than electron beams. We discuss the relevance of these findings with respect to Parker's idea of coronal heating by nanoflares. ",
        "introduction": "The Neupert effect is the name given to the observational finding that the rising part of the soft \\mbox{X-ray} (SXR) light curve often resembles the time integral of the hard X-ray (HXR) or microwave emission (\\mbox{Neupert}, 1968; Dennis \\& Zarro, 1993). The physical relevance of the Neupert effect basically arises from the fact that it is interpreted as a causal connection between the thermal and nonthermal flare emissions, which can be naturally explained within the nonthermal thick-target model (Brown, 1971). In this model, the flare energy is released primarily in the form of nonthermal electrons, and hard X-rays are produced via electron-ion bremsstrahlung when the electron beams impinge on the lower corona, transition region and chromosphere. The model assumes that only a small fraction of the electron beam energy is lost through radiation; most of the loss is due to Coulomb collisions that serve to heat the ambient plasma. As a consequence of the rapid energy deposition a strong pressure imbalance develops between the dense, heated chromosphere and the tenuous corona. The high pressure gradients cause the heated plasma to convect into the corona in a process known as chromospheric evaporation (Antonucci et al.\\ 1984; Fisher et al.\\ 1985), where it gives rise to enhanced SXR emission via thermal bremsstrahlung.  In this case, the hard X-ray flux is linked to the instantaneous rate of energy supplied by electron beams, whereas the soft X-ray flux is related to the accumulated energy deposited by the same electrons up to that time, and we can expect to see the Neupert effect. Any deviation from the Neupert effect, in principle, suggests that the hot SXR emitting plasma is not heated exclusively by thermalization of the accelerated electrons that are responsible for the HXR emission. Therefore, investigations of the Neupert effect provide insight into the role of nonthermal electrons for the flare energetics.  The Neupert effect can be expressed as % \\begin{equation} F_{\\rm P,SXR} = k \\cdot {\\cal F}_{\\rm HXR}  , \\label{EqNeup} \\end{equation} with $F_{\\rm P,SXR}$ the SXR peak flux and ${\\cal F}_{\\rm HXR}$ the HXR fluence, i.e.\\ the HXR flux integrated over the event duration. The coefficient~$k$ depends on several factors, as, e.g., the magnetic field geometry and the viewing angle, and thus may vary from flare to flare (Lee et al., 1995). However, if $k$ does not depend systematically on the flare intensity, then the SXR peak flux and the HXR fluence are linearly related. ",
        "conclusions": "24\\% of the events have $\\Delta t < 0$, i.e.\\ the SXR maximum occurs before the end of the HXR emission. These events are preferentially of long duration. Li et al.\\ (1993) have calculated time profiles of soft and hard X-ray emission from a thick-target electron-heated model, finding that, in general, the time derivative of the SXR emission corresponds to the time profile of the HXR emission, as stated by the Neupert effect. However, for gradual events they obtained that this relationship breaks down during the decay phase of the HXR event, in that the maximum of the SXR emission occurs before the end of the HXR emission. This phenomenon can be explained by the fact that the SXR emission starts to decrease if the evaporation-driven energy supply cannot overcome the instantaneous cooling of the hot plasma, which is likely to happen in gradual flares. Considering our observational findings together with the results from simulations by Li et al.\\ (1993), presumably most of the events with $\\Delta t < 0$ are consistent with the electron-beam-driven evaporation model. In particular, the high correlation between $F_{\\rm P,SXR}$ and ${\\cal F}_{\\rm HXR}$, $r \\approx 0.8$, supports such interpretation.  56\\% of the events have $\\Delta t > 0$; these events are preferentially of short and weak HXR emission. In principle, the fact that the SXR emission is still increasing although the HXR emission, i.e.\\ the electron input, has already stopped indicates that an additional agent besides the HXR emitting electrons is contributing to the energy input and prolonging the heating and/or evaporation. However, McTiernan et al.\\ (1999) have shown that the SXR time profile depends on the temperature response of the used detector: An increase of the SXR emission of low-temperature flare plasma after the HXR end may arise due to cooling of high-temperature plasma. Thus, we cannot attribute all events with $\\Delta t > 0$ as inconsistent with the electron-beam-driven chromospheric evaporation model. Instead, we consider as inconsistent only flares which show strong deviations from $\\Delta t = 0$, i.e.\\ the events belonging to set~2.  This means that for at least one fourth of the analyzed events an additional heating agent besides nonthermal electrons is suggested. A probable scenario is that energy is transported from the primary energy release site via thermal conduction fronts, which initiate chromospheric evaporation but do not produce hard \\mbox{X-rays}. The finding that for a considerable fraction of flares, preferentially weak ones, an additional heating agent other than electron beams is suggested, is not only relevant for the flare energetics but also for Parker's idea of coronal heating by nanoflares (Parker, 1988).  Hudson (1991) pointed out that if the corona is heated by flare-like events of different sizes, then the flare energy distribution must have a power-law slope $\\alpha > 2$. If the SXR flux does not vary systematically with temperature and density, then the SXR peak flux is linearly related to the maximum thermal energy of the flare plasma (see Lee et al., 1995; Veronig et al., 2002a). On the other hand, HXR fluence distributions can be considered as representative for the energy contained in nonthermal electrons. In Figure~\\ref{Fig_distr}, we show the SXR peak flux and the HXR fluence distributions derived from 1114 corresponding SXR/HXR flares, finding $\\alpha = 1.95$\\,$\\pm$\\,0.13 and $\\alpha = 1.58$\\,$\\pm$\\,0.13, respectively. The discrepancy between the slopes of the HXR fluence and the SXR peak flux distributions was already pointed out and discussed in Lee et al.\\ (1993, 1995) and Veronig et al.\\ (2002b). The present analysis provides an explanation for this difference in power-law slopes: The relationship between the SXR peak flux and the HXR fluence is not linear, % whereby the deviations from a linear correlation are strongest for weak flares (cf.~Figure~\\ref{Fig_FluencePeak}).  The soft X-ray flare emission increases due to energy supply by electron beams as well as any other heating agent, whereas the hard X-ray emission contains only information on the energy provided by electrons. Together with our finding that particularly in weak events an additional heating agent besides electron beams is suggested, this strongly suggests that soft X-ray peak flux distributions are a more meaningful indicator of flare energy distributions than hard X-ray fluence distributions. Furthermore, we have shown that weak flares have different characteristics than large flares, in the sense that electrons are less important for their energetics. Therefore, it is possible that the power-law slope of nanoflare frequency distributions differs from that derived for observed flares, which is close to the critical value of~$2$. In this respect it should be worthwhile to investigate flare frequency distributions of SXR flares without HXR counterparts, i.e.\\ without detectable particle acceleration, since these flares possibly provide the link to the smallest flare-like energy release events. %"
    },
    "0208/astro-ph0208040_arXiv.txt": {
        "abstract": "We  present  the  first detection of $^{12}$CO J=2$\\rightarrow$1 and $^{12}$CO J=1$\\rightarrow$0   emission from the LBV AG Carinae.  We show that AG Carinae resides in a region which is very rich in molecular gas with complex motions. We find evidence of a slow outflow of molecular gas, expanding at $\\simeq$ 7 km s$^{-1}$. This emission appears  spatially unresolved. We argue that it is spatially localised, rather than extended,  and possibly associated with the immediate circumstellar region of AG Carinae. Both detected CO lines are characterised by a pseudo-gaussian profile  of FWHM $\\simeq$ 15 km s$^{-1}$, indicating  a slowly expanding  region of molecular gas in close proximity to the hot central star. We have explored two possible scenarios to explain the observed profile: a circumstellar envelope, similar to carbon stars, or a circumstellar disk.  The option of the circumstellar disk is preferable  because: 1) it is consistent with additional independent indications for the existence of wind asymmetries in close proximity to the  central star,  found from spectropolarimetry and analysis of the UV and optical line profiles, and 2) it provides  the conditions of density and shielding necessary for the survival of the CO molecules in  proximity to such a hot star (T$_{eff}$ $\\simeq$ 14000 K - 20000 K). In the assumption that the CO emission originated when AG Carinae was in an evolved state, we derive a lower limit to the CO-mass of 6.5 $\\times$ 10$^{-3}$ M$_{\\odot}$. We also estimate that the CO fraction is $\\simeq$ 2.3 $\\times$ 10$^{-3}$ of the total mass of molecular gas, which then would amount to 2.8 M$_{\\odot}$. This is smaller, but still comparable with the mass of ionized gas present in the circumstellar environment (4.2 M$_{\\odot}$), with the implication that the molecular gas fraction can contribute significantly to the overall  mass lost from the central star in its post main sequence evolution. ",
        "introduction": "Massive stars spend their H-burning phase as O stars and become,  after a  few million  years, Wolf-Rayet (WR) stars undergoing core He-burning. Mass loss plays a key role in driving the stellar evolution from the main sequence to the WR  stage, especially during  the intermediate phase of Luminous Blue Variable (LBV).  Luminous Blue Variables  have been recognized to be at the point of core H-burning  exhaustion (Pasquali et al. 1997). In the H-R Diagram (HRD) they are located  close to the Humphreys-Davidson  (HD) boundary limit (Humphreys \\& Davidson 1979),  which represents an empirical instability limit to the evolution of stars more  massive than 40 M$_{\\odot}$. They  display large and irregular  spectro-photometric variability by transiting between hot visual minima phases (T$_{eff}$ $\\sim$  20000 - 30000K) to cooler visual maxima phases (T$_{eff}$ $<$ 10000K). Their  variability is also characterized by violent eruptions, with visual brightness increases  of 3 magnitudes or more, and extreme mass loss (up to several solar masses of  material ejected in the surrounding medium).  In between such dramatic outbursts, LBVs  still lose mass at  high rates -- typically 10$^{-5}$ - 10$^{-4}$ M$_{\\odot}$/yr (e.g. Leitherer et al. 1994).  The relics of these eruptions are spectacular circumstellar nebulae, such as the one  observed around Eta Carinae. A systematic investigation of LBVs in the Galaxy and in the Large Magellanic Cloud (Nota  et al. 1995) has shown that the majority of the known LBVs is surrounded by an ejected nebula, whose kinematical and chemical properties may be used to constrain the mass loss history of the central star. Their morphology  suggests they are shaped by wind interactions. The nebular dynamics indicates that the ejection timescale is comparable to the evolutionary lifetime of the LBV phase. Their chemical composition has been used to infer the evolutionary status of the central star at the moment of the ejection (Lamers et al. 2001). However, these evolutionary diagnostics are mostly derived from optical data. So far, LBVs have been little studied at infrared  or submillimeter wavelengths  although  these  spectral regions  provide a number of diagnostics probes of the nebular physical conditions which are not available at optical or UV wavelengths. A great wealth of information  can be derived from  such observations  on the presence of dust or molecular gas in the close surrounding  of these hot stars, to complement the optical data and to help determine the total amount of mass lost during the LBV phase. This piece of information, presently incomplete, is crucial to constrain the theoretical models which currently fail to reproduce the transition between O and WR stars (Langer et al. 1994).  In the past few years, near-IR imaging and spectroscopy both from the ground and more recently with ISO (Waters et al. 1998a,b; Voors et al. 1997; Morris et al. 2000) have greatly expanded our understanding of the properties, formation and survival mechanisms of dust in  hot stellar environments (T$_{eff}$ $\\simeq$ 20000 - 40000 K).  For the first time  it has been possible to study the spatial distribution of the dust, to quantify the nebular dust mass with some accuracy, and to investigate the dust - gas correlation. In addition,  the spectroscopic investigation of the dust features has contributed to better characterise the dust properties and  has provided additional independent evidence that these stars were cool and extended (mimicing a Red Supergiant - RSG) during the ejection of the nebula (Lamers et al. 2001).   Presence of molecular gas around  hot, evolved stars had  unexpectedly been found  in 1988 by McGregor et al. in a survey of blue supergiants in the LMC. McGregor et al. unambiguously detected  the first overtone band heads of CO at 2.3 $\\mu$ in six objects. Given the very low CO dissociation energy (11.1 eV), the presence of CO emission in close proximity to O and B supergiants characterised by fast stellar wind (V$_{inf}$ $\\simeq$ 200-400 km s$^{-1}$) and strong radiation fields has been difficult to explain. One possible explanation is that the CO emission is spatially extended, and far from the star. If the star has already experienced a RSG phase,  the CO emission could be a tracer of the previous cool and dense wind which has been compressed and excited by the interaction with the present hotter stellar wind. Alternatively, the CO emission may be  localised  and close to the star. McGregor et al. (1988) had tentatively identified the CO emitting region with  a circumstellar disk which is able to shield the CO from the stellar radiation and ionized wind. Localised emission can also be explained, as  is observed in the case of cooler giants and supergiants (such as S and C stars), by the survival of the  CO  in  an outer layer of the stellar envelope and its excitation by the stellar radiation field (Knapp \\& Morris 1985; Olofsson et al. 1988; Heske 1990). This explanation is only acceptable if the star mimiced a RSG during the ejection of the nebula. Unfortunately, the observations by McGregor did not  have the spatial resolution necessary to discriminate among the possible scenarios.  Among the luminous supergiants which McGregor et al. (1988) observed, there was the galactic LBV AG Carinae. AG Carinae is one of the most luminous LBVs, which is located in the  HRD   very close to the HD limit. Its distance is $6 \\pm 1$ kpc and its luminosity is $1.7~10^6$ L$_{\\odot}$ (Humpreys et al. 1989). The effective temperature is variable between about 9000 K and 25000 K, and so the radius changes from 70 to 500 R$_{\\odot}$ (Lamers 1986; Voors et al. 2000). It is surrounded by a very bright, extended nebula which has been known since the 1950's (Thackeray 1950). The nebula is composed of ionized gas and dust. When imaged in the light of H$\\alpha$, it exhibits an elliptical shell-like morphology (30$''$ $\\times$ 40$''$ in size). The nebular lines indicate that the nebula is expanding at an average velocity of 70 km s$^{-1}$ (Smith  1991). In the light of the line free continuum, which displays the stellar radiation scattered by dust, the AG Carinae nebula appears quite different (Paresce \\& Nota 1989): the dust grains are distributed in a jet-like feature which  HST has  resolved into a myriad of clumps and filaments with an overall bipolar structure (Nota et al. 1996). Recent ISO observations indicate that the dust grains are mostly crystalline olivine with an emission peak at 33.8 $\\mu$m (Waters et al. 1998). McGregor et al. (1988) did not detect any CO overtone emission in AG Carinae.  They found little or no emission from hot dust in the near-infrared, but surprisingly detected strong far-infrared emission from cool dust, and  inferred the action of a strong stellar wind, which possibly removed  the dust from the inner region. We decided to revisit the question of the presence of molecular gas in the circumstellar environment of AG Carinae, for a number of reasons:   AG Carinae is the ideal target for a neutral material study, given its proximity and  the large spatial extension of the nebula. The nebula is rich in  gas and dust and displays a strong far-infrared eccess. We have therefore decided to start an investigation aimed at detecting and characterising the properties of the neutral material around AG Carinae, with the  objective to: 1) detect CO emission from molecular gas, 2)   understand whether the emission is localised or spatially extended, and 3) determine the amount of molecular gas in the circumstellar environment.  We have carried out this program  with the SEST telescope in the submillimeter range, studying the transitions  $^{12}$CO J=$2\\rightarrow1$ and $^{12}$CO J=$1\\rightarrow0$  and obtaining CO emission maps of the entire nebula. The observation strategy and data reduction procedures are described in Sect. 2, and the results of our investigation are presented in Sect. 3. Discussion and conclusions follow in Sect. 4 and 5, respectively. ",
        "conclusions": "We have presented  the  first detection of $^{12}$CO J=2$\\rightarrow$1 and $^{12}$CO  J=1$\\rightarrow$0   emission from the LBV AG Carinae. The emission is spatially unresolved.  We believe it to be spatially localised, rather than extended,  and most likely associated with the immediate circumstellar region of AG Carinae, i.e. inside the optical nebula. Both the detected CO lines are  characterised by a pseudo-gaussian profile of FWHM $\\simeq$ 15 km s$^{-1}$, indicating  a slowly expanding region of molecular gas in close proximity to the hot central star. We have explored two possible scenarios to explain the observed profile: a circumstellar envelope, similar to carbon stars, or a circumstellar disk. We believe that the option of the circumstellar disk is preferable to provide the shielding necessary for the CO molecules to survive.   What is the origin of the CO emission? The outflow velocity   ($\\simeq$ 7 km s$^{-1}$) is atypically low when compared to the wind velocities  of LBVs (V$_{\\inf}$ $> $ 100-200 km s$^{-1}$). As we already mentioned, the CO velocity is also much slower than the expansion velocity of the optical nebula ($\\simeq$ 70 km s$^{-1}$), with whom it is unlikely associated. In terms of absolute values,  slow winds are usually found  in red supergiants (RSG). The question whether AG Carinae spent  any time as  a RSG is still open and argued on the basis of independent observational facts.  Smith et al. (1988) and Waters et al. (1998) argued that the abundances of the nebula and the composition of the dust would be in agreement with the ejection of the nebula as a RSG. Alternatively, Lamers et al.  (2001) have shown  that the abundances, the velocity, the  dynamical age and the morphology of the nebula are all consistent with the nebula being ejected  immediately after the main sequence phase of a rapidly rotating star. In that case the star cannot have been a RSG in the evolutionary sense (i.e. a massive star with a very extended convective  envelope), but it may have resembled a RSG during an outburst phase, when the wind was  optically thick, the effective radius very large and the outflow velocity very low.   We consider several possible origins for the CO emission: \\begin{itemize} \\item In the scenario proposed by Smith et al. (1998) which includes a RSG phase, or by Lamers et al. (2001) which includes a brief RSG-like phase, the molecular outflow could have originated during the RSG phase and survived protected by the disk likely formed when the star, evolving as an LBV to the blue side of the HRD, developed its fast wind. This could easily explain the low expansion velocity. \\item Alternatively, the CO may originate at the outer regions of a circumstellar disk, similar to the case of the B[e] stars. In this case the low velocity is difficult to explain. In normal B[e] stars the outflow velocity is of order of 100 km s$^{-1}$, even at large distances as shown by the forbidden lines (Zickgraf 1986). This is much larger than the value of 7 km s$^{-1}$ that we observe. \\item Or, the CO emission may originate at the interaction-zone at the edge of the IS bubble that was blown by the star. The maximum size of the emitting region ($<$ 1 arcmin) corresponds to a radius of about 1 pc. The amount of mass lost by the star in its main sequence phase is about 10 M$_{\\odot}$. The observed amount of molecular gas is $\\simeq$ 3 M$_{\\odot}$, which is only about 1/4 of the total mass of the shocked gas at the edge of the bubble. The main problem with this interpretation is  the small radius of the unresolved CO-emitting region (diameter $<$ 30$''$ or 0.9 pc) compared with the predicted much larger  size (several tens of parsecs) of the bubble blown by a spherical wind from a massive star with a velocity of 1000 km s$^{-1}$ during 4 $\\times$ 10$^6$ years (see e.g. Lamers \\& Cassinelli 1999, p. 369). However, if the wind were aspherical, e.g. concentrated in a disk due to the rapid rotation, the expansion speed of the interaction region in the equatorial plane would be much smaller because the bubble can more easily expand in the other direction. This might result in a high density and very low velocity {\\it waist} of the bipolar interstellar bubble, from which the observed CO emission might originate. \\end{itemize}  We estimate a lower limit to the mass of the molecular region of 2.8 M$_{\\odot}$. This is smaller, but still comparable with the mass of ionized gas present in the circumstellar environment (4.2 M$_{\\odot}$). The  implication  is that the molecular gas fraction can contribute significantly to the overall  mass lost from the central star in its post main sequence evolution, and therefore one should be careful when assuming that the ionized gas mass well approximates, at least in the case of AG Carinae, the total mass lost."
    },
    "0208/astro-ph0208330_arXiv.txt": {
        "abstract": "{In view of the extensive evidence of tight inter-relationships between spheroidal galaxies (and galactic bulges) with massive black holes hosted at their centers, a consistent model must deal jointly with the evolution of the two components. We describe one such model, which successfully accounts for the local luminosity function of spheroidal galaxies, for their photometric and chemical properties, for deep galaxy counts in different wavebands, including those in the (sub)-mm region which proved to be critical for current semi-analytic models stemming from the standard hierarchical clustering picture, for clustering properties of SCUBA galaxies, of EROs, and of LBGs, as well as for the local mass function of massive black holes and for quasar evolution. Predictions that can be tested by surveys carried out by SIRTF are presented.}  \\addkeyword{dust,extinction} \\addkeyword{galaxies: formation} \\addkeyword{quasars: general} \\addkeyword{Infrared: galaxies} \\addkeyword{Cosmology: theory}  \\begin{document} ",
        "introduction": "\\label{sec:intro} The hierarchical clustering model with a scale invariant spectrum of density perturbations in a Cold Dark Matter (CDM) dominated universe has proven to be remarkably successful in matching the observed large-scale structure as well as a broad variety of properties of galaxies of the different morphological types (Granato et al. 2000 and references therein). However, serious shortcomings of this scenario have also become evident in recent years.  The critical point can be traced back to the relatively large amount of power on small scales predicted by this model which would imply far more dwarf galaxies or substructure clumps within galactic and cluster mass halos than are observed (the so-called ``small-scale crisis\"), unless star formation in small objects is strongly suppressed (or the small scale power is reduced by modifying the standard model). At the other extreme of the galaxy mass function we have another strong discrepancy with model predictions, that we might call ``the massive galaxy crisis''. Even the best semi-analytic models hinging upon the standard picture for structure formation in the framework of the hierarchical clustering paradigm, are stubbornly unable to account for the (sub)-mm (SCUBA and MAMBO, Figure~\\ref{fig1}) counts of galaxies, most of which are probably massive objects undergoing a very intense starburst (with star formation rates $\\sim 1000\\,\\hbox{M}_\\odot\\,\\hbox{yr}^{-1}$) at $z>2$. Recent optical data confirm that most massive ellipticals were already in place and (almost) passively evolving up to $z\\simeq 1$--1.5. These data are more consistent with the traditional ``monolithic'' approach whereby giant ellipticals formed most of their stars in a single gigantic starburst at substantial redshifts, and underwent essentially passive evolution thereafter. In the canonical hierarchical clustering paradigm the smallest objects collapse first and most star formation occurs, at relatively low rates, within relatively small proto-galaxies, that later merged to form larger galaxies. Thus the expected number of galaxies with very intense star formation is far less than detected in SCUBA and MAMBO surveys and the surface density of massive evolved ellipticals at $z\\gtrsim 1$ is also smaller than observed. The ``monolithic'' approach, however, is inadequate to the extent that it cannot be fitted in a consistent scenario for structure formation from primordial density fluctuations. \\begin{figure*} \\includegraphics[width=\\textwidth,height=7cm]{ggranato_fig1.eps} \\caption{Integral source counts at $850\\,\\mu$m and $1.4$ mm predicted by the model by Granato et al.\\ (2001) compared with observations. The dotted, dashed, dot-dashed lines show the contributions of starburst, spiral, and forming elliptical galaxies respectively. The long-dashed (at 1.4 mm only) gives the counts of radio sources (Toffolatti et al.\\ 1998).} \\label{fig1} \\end{figure*} ",
        "conclusions": ""
    },
    "0208/astro-ph0208106_arXiv.txt": {
        "abstract": "The formation and evolution of galaxies is one of the great outstanding problems of astrophysics. Within the broad context of hierachical structure formation, we have only a crude picture of how galaxies like our own came into existence.  A detailed physical picture where individual stellar populations can be associated with (tagged to) elements of the protocloud is far beyond our current understanding.  Important clues have begun to emerge from both the Galaxy (near-field cosmology) and the high redshift universe (far-field cosmology).  Here we focus on the fossil evidence provided by the Galaxy.  Detailed studies of the Galaxy lie at the core of understanding the complex processes involved in baryon dissipation. This is a necessary first step towards achieving a successful theory of galaxy formation.  \\vskip 0.5truein {\\it Key Words: Cosmology, Local Group, Stellar Populations, Stellar Kinematics} \\vspace{.4in} ",
        "introduction": "\\subsection{The New Galaxy}  Weinberg (1977) observed that ``the theory of the formation of galaxies is one of the great outstanding problems of astrophysics, a problem that today seems far from solution.'' Although the past two decades have seen considerable progress, Weinberg's assessment remains largely true.  Eggen, Lynden-Bell and Sandage (1962; ELS) were the first to show that it is possible to study galactic archaeology using stellar abundances and stellar dynamics; this is probably the most influential paper on the subject of galaxy formation.  ELS studied the motions of high velocity stars and discovered that, as the metal abundance decreases, the orbit energies and eccentricities of the stars increased while their orbital angular momenta decreased.  They inferred that the metal-poor stars reside in a halo that was created during the rapid collapse of a relatively uniform, isolated protogalactic cloud shortly after it decoupled from the universal expansion.  ELS is widely viewed as advocating a smooth monolithic collapse of the protocloud with a timescale of order $10^8$ years. But Sandage (1990) stresses that this is an over-interpretation; a smooth collapse was not one of the inferences they drew from the stellar kinematics.  In 1977, the ELS picture was challenged by Searle (see also Searle \\& Zinn 1978) who noted that Galactic globular clusters have a wide range of metal abundances essentially independent of radius from the Galactic Center. They suggested that this could be explained by a halo built up over an extended period from {\\it independent} fragments with masses of $\\sim 10^8$ M$_\\odot$.  In contrast, in the ELS picture, the halo formed in a rapid free-fall collapse. But halo field stars, as well as globular clusters, are now believed to show an age spread of $2-3$ Ga (Marquez \\& Schuster 1994);  for an alternative view, see Sandage \\& Cacciari (1990). The current paradigm, that the observations argue for a halo that has built up over a long period from infalling debris, has developed after many years of intense debate.  This debate parallelled the changes that were taking place in theoretical studies of cosmology (\\eg\\ Peebles 1971; Press \\& Schecter 1974).  The ideas of galaxy formation via hierarchical aggregation of smaller elements from the early universe fit in readily with the Searle \\& Zinn view of the formation of the galactic halo from small fragments. The possibility of identifying debris from these small fragments was already around in Eggen's early studies of moving groups, and this is now an active field of research in theoretical and observational stellar dynamics. It offers the possibility to reconstruct at least some properties of the protogalaxy and so to improve our basic understanding of the galaxy formation process.  We can extend this approach to other components of the Galaxy.  We will argue the importance of understanding the formation of the galactic disk, because this is where most of the baryons reside. Although much of the information about the properties of the protogalactic baryons has been lost in the dissipation that led to the galactic disk, a similar dynamical probing of the early properties of the disk can illuminate the formation of the disk, at least back to the epoch of last significant dissipation. It is also clear that we do not need to restrict this probing to stellar dynamical techniques. A vast amount of fossil information is locked up in the detailed stellar distribution of chemical elements in the various components of the Galaxy, and we will discuss the opportunities that this offers.  We are coming into a new era of galactic investigation, in which one can study the fossil remnants of the early days of the Galaxy in a broader and more focussed way, not only in the halo but throughout the major luminous components of the Galaxy. This is what we mean by {\\it The New Galaxy}.  The goal of these studies is to reconstruct as much as possible of the early galactic history. We will review what has been achieved so far, and point to some of the ways forward.    \\subsection{Near-field and far-field cosmology}  What do we mean by the reconstruction of early galactic history? We seek a detailed physical understanding of the sequence of events which led to the Milky Way. Ideally, we would want to tag (i.e. associate) components of the Galaxy to elements of the protocloud -- the baryon reservoir which fueled the stars in the Galaxy.  From theory, our prevailing view of structure formation relies on a hierarchical process driven by the gravitational forces of the large-scale distribution of cold, dark matter (CDM). The CDM paradigm provides simple models of galaxy formation within a cosmological context (Peebles 1974; White \\& Rees 1978; Blumenthal\\etal\\ 1984). N-body and semi-analytic simulations of the growth of structures in the early universe have been successful at reproducing some of the properties of galaxies. Current models include gas pressure, metal production, radiative cooling and heating, and prescriptions for star formation.  The number density, properties and spatial distribution of dark matter halos are well understood within CDM (Sheth \\& Tormen 1999; Jenkins\\etal\\ 2001). However, computer codes are far from producing realistic simulations of how baryons produce observable galaxies within a complex hierarchy of dark matter. This a necessary first step towards a viable theory or a working model of galaxy formation.  In this review, our approach is anchored to observations of the Galaxy, interpreted within the broad scope of the CDM hierarchy. Many of the observables in the Galaxy relate to events which occurred long ago, at high redshift. Fig.~\\ref{fig1} shows the relationship between look-back time and redshift in the context of the $\\Lambda$CDM model: the redshift range (z $\\lta$ 6) of discrete sources in contemporary observational cosmology matches closely the known ages of the oldest components in the Galaxy. The Galaxy (near-field cosmology) provides a link to the distant universe (far-field cosmology).  Before we embark on a detailed overview of the relevant data, we give a descriptive working picture of the sequence of events involved in galaxy formation. For continuity, the relevant references are given in the main body of the review where these issues are discussed in more detail.  \\subsection{A working model of galaxy formation}  Shortly after the Big Bang, cold dark matter began to drive baryons towards local density enhancements. The first stars formed after the collapse of the first primordial molecular clouds; these stars produced the epoch of reionization. The earliest recognizable protocloud may have begun to assemble at about this time.  Within the context of CDM, the dark halo of the Galaxy assembled first, although it is likely that its growth continues to the present time. In some galaxies, the first episodes of gas accretion established the stellar bulge, the central black hole, the first halo stars and the globular clusters. In the Galaxy and similar systems, the small stellar bulge may have formed later from stars in the inner disk.  The early stages of the Galaxy's evolution were marked by violent gas dynamics and accretion events, leading to the high internal densities of the first globular clusters, and perhaps to the well-known `black hole mass -- stellar bulge dispersion' relation. The stellar bulge and massive black hole may have grown up together during this active time. We associate this era with the `Golden Age', the phase before $z \\sim 1$ when star formation activity and accretion disk activity were at their peak.  At that time, there was a strong metal gradient from the bulge to the outer halo. The metal enrichment was rapid in the core of the Galaxy such that, by $z \\sim 1$, the mean metallicities were as high as [Fe/H] $\\sim$ -1 or even higher. In these terms, we can understand why the inner stellar bulge that we observe today is both old and moderately metal rich. The first halo stars ([Fe/H] $\\approx$ -5 to -2.5) formed over a more extended volume and presumably date back to the earliest phase of the protocloud. The first globular clusters formed over a similar volume from violent gas interactions ([Fe/H] $\\approx$ -2.5 to -1.5). We believe now that many of the halo stars and globulars are remnants of early satellite galaxies which experienced {\\it independent} chemical evolution before being accreted by the Galaxy.  The spread in [Fe/H], and the relative distribution of the chemical elements, is a major diagnostic of the evolution of each galactic component. If the initial mass function is constant, the mean abundances of the different components give a rough indication of the number of SN~II enrichments which preceded their formation, although we note that as time passes, an increasing fraction of Fe is produced by SN~Ia events. For a given parcel of gas in a closed system, only a few SN~II events are required to reach [Fe/H] $\\approx$ -3, 30 to 100 events to get to [Fe/H] $\\approx$ -1.5, and maybe a thousand events to reach solar metallicities.  We wish to stress that [Fe/H] is not a clock:  rather it is a measure of supernova occurrences and the depth of the different potential wells that a given parcel of gas has explored.  During the latter stages of the Golden Age, most of the baryons began to settle to a disk for the first time. Two key observations emphasize what we consider to be the mystery of the main epoch of baryon dissipation. First, there are no stars with [Fe/H] $<$ -2.2 which rotate with the disk.  Secondly, despite all the activity associated with the Golden Age, at least 80\\% of the baryons appear to have settled gradually to the disk over many Ga; this fraction could be as high as 95\\% if the bulge formed after the disk.   About 10\\% of the baryons reside in a `thick disk' which has [Fe/H] $\\approx$ -2.2 to -0.5, compared to the younger thin disk with [Fe/H] $\\approx$ -0.5 to +0.3. It is striking how the globular clusters and the thick disk have similar abundance ranges, although the detailed abundance distributions are different. There is also a similarity in age:  globular clusters show an age range of 12 to 14 Ga, and the thick disk appears to be at least 12 Ga old. Both the thick disk and globulars apparently date back to the epoch of baryon dissipation during z $\\sim$ 1$-$5.  Fig.~\\ref{fig2} summarises our present understanding of the complex age$-$metallicity distribution for the various components of the Galaxy.  It is a mystery that the thick disk and the globulars should have formed so early {\\it and} over such a large volume from material which was already enriched to [Fe/H] $\\sim$ -2. Could powerful winds from the central starburst in the evolving core have distributed metals throughout the inner protocloud at about that time?  Finally, we emphasize again that 90\\% of the disk baryons have settled quiescently to the thin disk since z $\\sim$ 1.   \\subsection{Timescales and fossils}   The oldest stars in our Galaxy are of an age similar to the look-back time of the most distant galaxies in the Hubble Deep Field (Fig.~\\ref{fig1}).  For the galaxies, the cosmological redshift measured from galaxy spectra presently takes us to within 5\\% of the origin of cosmic time.  For the stars, their upper atmospheres provide fossil evidence of the available metals at the time of formation.  The old Galactic stars and the distant galaxies provide a record of conditions at early times in cosmic history, and both harbor clues to the sequence of events which led to the formation of galaxies like the Milky Way.    The key timescale provided by far-field cosmology is the look-back time with the prospect of seeing galaxies at an earlier stage in their evolution. However, this does {\\it not} imply that these high-redshift objects are unevolved. We know that the stellar cores of galaxies at the highest redshifts ($z\\sim 5$) observed to date exhibit solar metallicities, and therefore appear to have undergone many cycles of star formation (Hamann \\& Ferland 1999). Much of the light we detect from the early universe probably arises from the chemically and dynamically evolved cores of galaxies.   Near-field cosmology provides a dynamical timescale, $\\tau_D \\sim (G\\rho)^{-\\frac{1}{2}}$, where $\\rho$ is the mean density of the medium.  The dynamical timescale at a radial distance of 100 kpc is of order several Ga, so the mixing times are very long. Therefore, on larger scales, we can expect to find dynamical and chemical traces of past events, even where small dynamical systems have long since merged with the Galaxy.  We note that the CDM hierarchy reflects a wide range of dynamical timescales, such that different parts of the hierarchy may reveal galaxies in different stages of evolution.  In this sense, the hierarchy relates the large-scale density to the morphology and evolution of its individual galaxies; this is the so-called `morphology-density relation' (Dressler 1980; Hermit\\etal\\ 1996; Norberg\\etal\\ 2001). Over a large enough ensemble of galaxies, taken from different regions of the hierarchy, we expect different light-weighted age distributions because one part of the hierarchy is more evolved than another.  In other words, the evolution of small-scale structure (individual galaxies) must at some level relate to the environment on scales of 10 Mpc or more.    The near field also provides important evolutionary timescales for individual stars and groups of stars (see ``Stellar age dating'' below). Individual stars can be dated with astero-seismology (Christensen-Dalsgaard 1986; Gough 2001) and nucleo-cosmochronology (Fowler \\& Hoyle 1960; Cowan\\etal\\ 1997). Strictly speaking, nucleo-cosmochronology dates the elements rather than the stars. Coeval groups of stars can be aged from the main-sequence turn-off or from the He-burning stars in older populations (Chaboyer 1998).  Furthermore, the faint end cut-off of the white dwarf luminosity function provides an important age constraint for older populations (Oswalt\\etal\\ 1996). Presently, the aging methods are model dependent.      \\subsection{Goals of near-field cosmology}  We believe that the major goal of near-field cosmology is to tag individual stars with elements of the protocloud. Some integrals of motion are likely to be preserved while others are scrambled by dissipation and violent relaxation. We suspect that complete tagging is impossible. However, some stars today may have some integrals of motion which relate to the protocloud at the epoch of last dissipation (see ``Zero order signatures -- information preserved since dark matter virialized'' below).  As we review, different parts of the Galaxy have experienced dissipation and phase mixing to varying degrees. The disk, in contrast to the stellar halo, is a highly dissipated structure. The bulge may be only partly dissipated. To what extent can we unravel the events that produced the Galaxy as we see it today?  Could some of the residual inhomogeneities from prehistory have escaped the dissipative process at an early stage?  Far field cosmology currently takes us back to the `epoch of last scattering' as seen in the microwave background. Cosmologists would like to think that some vestige of information has survived from earlier times (cf. Peebles, Seager \\& Hu 2000). In the same spirit, we can hope that fossils remain from the `epoch of last dissipation', \\ie\\ the main epoch of baryon dissipation that occurred as the disk was being assembled.  To make a comprehensive inventory of surviving inhomogeneities would require a vast catalog of stellar properties that is presently out of reach (Bland-Hawthorn 2002).  The Gaia space astrometry mission (Perryman\\etal\\ 2001), set to launch at the end of the decade, will acquire detailed phase space coordinates for about one billion stars, within a sphere of diameter 20 kpc (the Gaiasphere).  In ``The Gaiasphere and the limits of knowledge'' below, we look forward to a time when all stars within the Gaiasphere have complete chemical abundance measurements (including all heavy metals).  Even with such a vast increase in information, there may exist fundamental $-$ but unproven $-$ limits to unravelling the observed complexity.  The huge increase in data rates from ground-based and space-based observatories has led to an explosion of information.  Much of this information from the near field is often dismissed as `weather' or unimportant detail.  But in fact fundamental clues are already beginning to emerge.  In what is now a famous discovery, a large photometric and kinematic survey of bulge stars revealed the presence of the disrupting Sgr dwarf galaxy (Ibata \\etal\\ 1994), now seen over a large region of sky and in a variety of populations (see ``Structures in phase space'' below).  Perhaps the most important example arises from the chemical signatures seen in echelle spectroscopy of bulge, thick disk and halo stars. In ``Epilogue: challenges for the future'', we envisage a time when the analysis of thousands of spectral lines for a vast number of stars will reveal crucial insights into the sequence of events early in the formation of the Galaxy.  In this review, we discuss fossil signatures in the Galaxy. A key aspect of fossil studies is a reliable time sequence. In ``Stellar age dating,'' we discuss methods for age-dating individual stars and coeval groups of stars.  In  ``Structure of the Galaxy,'' we describe the main components of the Galaxy.  In ``Signatures of galaxy formation,'' we divide the fossil signatures of galaxy formation into three parts: zero order signatures that preserve information since dark matter virialized; first order signatures that preserve information since the main epoch of baryon dissipation; second order signatures that arise from major processes involved in subsequent evolution. In ``The Gaiasphere and the limits of knowledge,'' we look forward to a time when it is possible to measure ages, phase space coordinates and chemical properties for a vast number of stars in the Galaxy.  Even then, what are the prospects for unravelling the sequence of events that gave rise to the Milky Way? We conclude with some experimental challenges for the future. ",
        "conclusions": ""
    },
    "0208/astro-ph0208438_arXiv.txt": {
        "abstract": "We report the discovery of a population of Wolf-Rayet stars in the young Galactic open cluster Westerlund\\,1. In an incomplete shallow spectroscopic survey, we find six nitrogen-rich (WN) and five carbon-rich (WC) WR stars. We also confirm the presence of a large population of yellow supergiants, some of which are candidate hypergiants. Given this population, Westerlund\\,1 is likely to be one of the more massive young clusters in the Local Group. ",
        "introduction": "\\vspace*{-2mm}  The highly reddened young open cluster Westerlund\\,1 (henceforth Wd\\,1) was reported by Westerlund (1987) to contain a number of both early and late-type supergiants and some other massive transitional objects.  Exact determination of its parameters was not possible, but Westerlund (1987) estimated $d$\\,$\\simeq$\\,5\\,kpc and an extinction $A_V$\\,$\\simeq$\\,11\\,mag. Recently, radio continuum observations of Wd\\,1 revealed that a number of cluster members appeared to be associated with very bright radio sources (Clark \\ea 1998; Dougherty, Waters \\& Clark, in preparation).  In view of this result, we carried out a spectroscopic survey of the brighter cluster members using the Boller \\& Chivens spectrograph on the ESO 1.52-m telescope at La Silla Observatory, Chile.  Low resolution spectroscopy over the $\\sim$\\,$\\lambda\\lambda$\\,6\\,000\\,-\\,11\\,000\\AA\\ range was obtained on the nights of June 24\\,-\\,26, 2001.  Since the field is very crowded, in addition to the relatively bright objects that were originally targeted, a large number of fainter stars was observed. Inspection of the spectra revealed the presence of a population of faint objects with strong emission lines.   \\vspace*{-2mm} ",
        "conclusions": "\\vspace*{-2mm}  We have detected eleven WR stars in Westerlund\\,1, the largest number of WR stars known in any Galactic cluster, with the possible exception of the Arches cluster (Blum \\ea 2001).  Moreover, our survey is very incomplete: only about $25$\\% of the stars at the apparent magnitude typical of the WR stars detected have been observed.  There is an obvious lack of WR detections in the central region of the cluster (where the most luminous supergiants are located), which strongly suggests that our sample is also affected by observational effects.  We can conservatively assume that the actual WR population of Wd\\,1 is easily twice as large.  Wd\\,1 appears to be unique among Galactic clusters in both the large number and variety of massive post-main sequence (PMS) objects.  Since published determinations of distance and reddening to the cluster are inaccurate and inconsistent (Westerlund 1987; Piatti, Bica \\& Clari\\'a 1998), and the field is affected by strong and probably variable reddening, we are still unable to provide accurate values for the intrinsic luminosity of members from which good estimates of the cluster age can be derived.  Several lines of argument, however, suggest that the cluster is potentially extremely massive.  On one side, the lack of an identifiable MS turnoff in available photometric data prevents us from determining cluster parameters, but at the same time provides us with a clue to the cluster size, suggesting that all observed members down to $V$\\,$\\simeq$\\,18 (in excess of 90) are evolved stars with intrinsic magnitudes in the M$_V$\\,$\\approx$\\,--\\,6 to --\\,10 range.  On the other hand, the large number of transitional objects observed in short-lived phases is suggestive of a very large population from which they are evolving.  In particular, the number of yellow hypergiants in Wd\\,1 appears comparable to that in the rest of the Galaxy.  Given the short duration of the hypergiant phase, a population of several hundred O-type stars seems to be suggested.  The presence of a large number of yellow supergiants (and a few very luminous red supergiants) indicates that Wd\\,1 is slightly older than the very massive Arches Cluster.  Assuming that the progenitors of the yellow hypergiants had ZAMS masses of $\\sim$\\,40\\,-\\,50\\,M$_\\odot$ (which can be considered an educated guess), the age of the cluster would be $\\sim$\\,4\\,-\\,5\\,Myr, compatible with the presence of a population of WC stars descending from more massive progenitors.  The alternative of considering that the progenitors of the yellow supergiants have lower ($\\sim$\\,30\\,M$_\\odot$) masses and taking an older age of 8\\,-\\,10\\,Myr in order to consider that the red stars are normal M-type supergiants seems to conflict with the large population of WR stars if we consider that they are mainly single stars.  A large population of WR stars at an age of $\\sim 8$ Myr could, however, be possible if most of them are part of binary systems.  With the much extended data set available after our 2002 observing campaign, we expect to be able to provide accurate determinations for the distance and extinction to Wd\\,1.  The lack of radio detections for most WR stars (Dougherty \\ea in preparation) suggests that the distance to Wd\\,1 has to be $\\ga$\\,2\\,kpc.  It cannot be, however, much larger than the 5\\,kpc advanced by Westerlund (1987), specially since the results of Piatti \\ea (1998) suggest that the interstellar absorption (which has a very important component local to the cluster region) may be higher than estimated by Westerlund (1987).  The cluster is then likely to be located in the Crux Arm, at a distance of 4\\,-\\,5\\,kpc.  In any case, with a true distance modulus $DM$\\,$<$\\,14\\,mag, Wd\\,1 offers the unrivalled opportunity of observing its whole stellar population using existing instrumentation (at least, in the near-IR).  It therefore represents an ideal laboratory for the study of the impact of the presence of a large population of massive stars on its environment and, specifically, on the formation of lower mass stars.  \\vspace*{+2mm}"
    },
    "0208/astro-ph0208112_arXiv.txt": {
        "abstract": "The concept of hyperbolic flux tubes (HFTs) is a generalization of the concept of separator field lines for coronal magnetic fields with a trivial magnetic topology. An effective mechanism of a current layer formation in HFTs is proposed. This mechanism is called magnetic pinching and it is caused by large-scale shearing motions applied to the photospheric feet of HFTs in a way as if trying to twist the HFT\\@. It is shown that in the middle of an HFT such motions produce a hyperbolic flow that causes an exponentially fast growth of the current density in a thin force-free current layer. The magnetic energy associated with the current layer that is built up over a few hours is sufficient for a large flare. Other implications of HFT pinching for solar flares are discussed as well. ",
        "introduction": "Over the last decades it has become clear from both observational and theoretical points of view that current sheet formation in the corona is one of the key processes for solar physics \\citep{prk94,prfrb00}. In particular, it provides a temporary deposit around the current sheets of free magnetic energy for solar flares and creates favorable conditions for subsequent rapid conversion of this energy into other forms. The corresponding local growth of current density in the current sheet formation process may stimulate the onset of plasma instabilities and the development of anomalous resistivity. Turbulent dissipation of the current layer due to such instabilities is thought to be a reason of thermal and supra-thermal processes in solar flares \\citep{som92}.  Typically, the coronal magnetic field is frozen into plasma and its Maxwellian stresses are large enough to dominate over other forces in active regions. This means that in such plasmo-magnetic configurations the magnetic pressure and tension approximately balance each other. Also the transit time of perturbations through the corona is much less than the characteristic time of photospheric motions. Therefore, the configurations have to evolve through a sequence of nearly force-free equilibria in response to the time variation of the photospheric boundary conditions. Due to the frozen-in condition the topological structure of the magnetic field in the corona is not changed by horizontal photospheric motions. However, it may be changed by a vertical injection of a new magnetic flux through the photosphere. In the generic case this may lead to the appearance of magnetic null points in the corona. The magnetic forces in the vicinity of the nulls are too weak to withstand to large variations of the ambient magnetic stress. Therefore, the corresponding neighborhoods of the nulls generally collapse in evolving fields, producing topologically accessible current singularities \\citep{prtit96}.  Similar processes have to occur in the vicinity of the so-called bald patches \\citep{titea93}, which are those segments of photospheric polarity inversion lines (IL), where the field lines touch the photosphere \\citep{shf86}. In comparison with the case of the nulls the physics of the formation of the current singularities is a bit different here: the current sheets are formed due to an ``attachment'' of the touching field lines to the very heavy photospheric material \\citep{low87, alam89, vea91}.  Thus, the presence of the topological features such as null points and bald patches is a {\\it condition\\/} for the formation  of current singularities in coronal configurations. An analysis of magnetic field structures in solar flares shows, however, that not all the observed event can be explained by the presence of such topological features and that therefore this class of features has to be extended \\citep{dem97}. A possible extension can be found by using the concept of field line connectivity. Indeed, in the framework of this approach the nulls and bald patches, as well as the separatrix field lines emanating from them, can be detected by discontinuous jumps of the field line connectivity. The desired extension then can be found by weakening the condition that the jump in connectivity is discontinuous to requiring only that the field line connectivity exhibits a ``large'' spatial variation \\citep{shf86, lngcstr94}. The flux tubes which exhibit such a behavior are called quasi-separatrix layers (QSLs) \\citep{prdem95}. It should be emphasized that QSLs are geometrical objects rather than topological ones, since they can be removed by suitable continuous deformations of the magnetic field \\citep{tithrn02}. The genuine separatrix lines and surfaces are degenerate or limiting cases of QSLs \\citep{titea02}.  For determining QSLs in a given magnetic field a measure of magnetic connectivity is required. The proper measure is {\\it the degree of squashing\\/} of elemental flux tubes, which connect opposite photospheric polarities and have infinitesimal cross-sections \\citep{titea99}. A QSL is then defined as a flux tube with abnormally large values of the squashing degree. In application to quadrupole magnetic configurations formed by two bipolar groups of sunspots this criterion reveals a special geometrical feature called {\\it hyperbolic flux tube\\/} (HFT) \\citep{titea02}. HFTs can be understood as a combination of two intersecting QSLs. In the limiting case, where the sunspot flux is concentrated in point-like sources, the HFT collapses into two separatrix surfaces intersecting along a separator field line. The separator is a field line connecting two null points which appear in this limiting case and it is a favorable site for current sheet formation caused by displacements of the sunspot positions \\citep{swt69, gs88, lf90,lnc02}. Therefore it is natural to expect that HFTs are also preferred sites for current sheet formation. Here we will call such a process of current sheet formation in an HFT {\\it  magnetic pinching\\/}, by analogy with a similar process studied in axisymmetric laboratory plasmas (the corresponding similarity will be further clarified below).  It is the purpose of this series of papers to understand and quantify the basic properties of the process of current sheet formation in HFTs. In the present paper we present basic theoretical calculations and estimates relevant to this problem in the following way. In \\S~\\ref{s:form} the problem and an idea for its solution are formulated. In \\S~\\ref{s:S2} a simplified quadrupole configuration used to model an HFT is described. In \\S~\\ref{s:km} a kinematic model of the HFT pinching process is developed. In \\S~\\ref{s:ffp} this model is improved by incorporating an approximate form of magnetic force balance. In \\S~\\ref{s:impl} the implications of the obtained results for solar flares are discussed. Section~\\ref{s:cs} presents the conclusions of this work. The results of dynamic and quasi-static simulations of the HFT pinching process will be described in forthcoming papers of this series. ",
        "conclusions": "\\label{s:cs}  We have found that the fundamental condition for the formation of layers of particularly strong current density (pinching) inside hyperbolic flux tubes (HFTs) are photospheric shearing motions which twist the HFT around its axis. Photospheric motions which merely turn the HFT have a much weaker effect. We plan to study in more detail in the following papers of this series how combinations of these two extreme types of photospheric motions influence the HFT pinching. Also the obtained results must be valid for configurations with separator field lines and their associated separatrix surfaces, since such structures are a limiting case of HFTs.  Due to  the special elastic properties of the HFT the photospheric shearing motions easily propagate into the corona and meet each other in the middle of the HFT\\@. If the shearing motions are applied to the HFT feet in a way as if trying to cause a torsion of the HFT, their superposition in the middle of the HFT forms a  hyperbolic flow pattern which is inclined by $45^{\\circ}$ with respect to the hyperbolic structure of the transverse magnetic field inside the HFT\\@. Such an arrangement of the transverse velocity and magnetic fields provides an exponentially growing current density ordered in a layer-like structure along the HFT\\@. The width of the current layer is of the order of the characteristic size of the shearing motions, while its thickness decreases exponentially fast in time.  There are basically two physical effects which are responsible for the process of HFT pinching. First, the hyperbolic component of the flow is incompressible and so does not change the longitudinal magnetic field in the HFT\\@. However, it causes a large squashing of plasma elements along the forming layer and this squashing increases the transverse magnetic field near the layer enormously in comparison with the initial state. Therefore, the strength of the resulting transverse field can easily approach that of the longitudinal field and even exceed it, depending on the initial conditions and the duration of the flow. The increase of the transverse field causes a second effect -- the compression of the  longitudinal field and plasma across the layer, which leads to an additional increase of the current density inside the layer.  The exponential time dependence of the basic physical quantities inside the current layer  can also be translated into a similar dependence on sunspot displacements. For a quadrupole configuration with an HFT  we can estimate that the displacements of the spots over distances comparable with the distances between them are sufficient to build up magnetic energy inside the pinched HFT as is required for a large solar flare. The large current density and gradient of magnetic shear inside the layer provide favorable conditions for the onset of the tearing instability in the layer and its transition to a turbulent state with a large rate of reconnection and magnetic energy release."
    },
    "0208/astro-ph0208324_arXiv.txt": {
        "abstract": "The Owens Valley Radio Observatory (OVRO) millimeter array was used to make observations of the CS(2-1) line (at 97.981 GHz) arising from the G0.07+0.04 region of the ``$-$30 \\kms'' molecular cloud near the Galactic center with a spatial resolution of \\ab8\\arcsec. The ionized edges of this cloud forms the Arched Filament HII regions which are ionized by the adjacent hes stellar cluster. The OVRO data were combined with single-dish data obtained at the 30-m IRAM telescope by Serabyn \\& \\gusten~(1987). A comparison of this CS(2-1) data and the H92$\\alpha$~recombination line data of Lang, Goss \\& Morris (2001) reveals that the ionized and molecular gas are physically related, but that their velocities in this region differ by up to 35 \\kms. This difference in velocity can be understood if the gas that gave rise to the G0.07+0.04 HII region has been fully ionized. An overall comparison of the molecular and ionized gas across the entire $-$30 \\kms~cloud based on the single dish CS(2-1) data and the H92$\\alpha$~line data illustrates that such differences in velocity between the ionized and molecular gas are common and that the geometrical arrangement of these components is complicated. Much of the ionized gas resides on the near side (to the observer) of the molecular cloud; however, in several regions, some molecular material must lie in front of the HII region. The Arches stellar cluster therefore appears to be located in the midst of the molecular clouds such that some of the near-side cloud surfaces along our line of sight have not been exposed to the ionizing radiation. ",
        "introduction": "The interplay between components of the remarkable Galactic Center Radio Arc remains one of the outstanding issues in understanding the interstellar medium in this region. The Radio Arc, which lies \\ab30 pc in projection from the center of the Galaxy, SgrA$^*$, was first revealed in detail with the Very Large Array (VLA) of the National Radio Astronomy Observatory\\footnotemark\\footnotetext{The National Radio Astronomy Observatory is a facility of the National Science Foundation, operated under a cooperative agreement with Associated Universities, Inc.} over seventeen years ago (Yusef-Zadeh et al. 1984). The Arc consists of both thermal and nonthermal structures apparently interacting with each other, but the nature of the physical connections is not well understood. The prominent non-thermal filaments (NTFs) oriented perpendicular to the Galactic plane define the striking linear morphology of the Radio Arc. Eight systems of similar NTFs have been discovered within 250 pc of the Galactic center: typically, they extend for up to 60 pc in length, but are very narrow structures ($<$0.1 pc). The NTFs also show strong linear polarization, have intrinsic magnetic field orientations aligned along their long axis, and have magnetic field strengths estimated at 0.1-1 mG (Yusef-Zadeh \\& Morris 1987a; Tsuboi et al. 1986; Yusef-Zadeh, Wardle, \\& Parastaran 1997; Lang et al. 1999a,b). The NTFs are thought to trace a large-scale polodial magnetic field configuration in the inner Galaxy (Morris 1994). However, the origin of the relativistic particles in these synchrotron NTFs and the mechanism for particle acceleration remain unclear.  In addition, two concentrations of ionized and molecular gas appear to be associated with the Radio Arc NTFs: (1) the Sickle (G0.18$-$0.04), located at the center of the Radio Arc and (2) the Arched Filaments, which define its western edge (Pauls et al. 1976, 1980; Yusef-Zadeh 1986; Serabyn \\& \\gusten~1987 (hereafter, SG87), 1991). VLA observations reveal these sources to be peculiar HII regions with extremely filamentary morphology as well as complex velocity structure, although the coherence scale for the thermal filaments is much smaller than that of the NTFs (Yusef-Zadeh \\& Morris 1987b; Yusef-Zadeh, Morris, \\& van Gorkom 1987; Lang, Goss \\& Wood 1997; Lang, Goss \\& Morris 2001 (hereafter LGM01)). Conventional photoionization was initially dismissed because of the presumed unlikely arrangement of stars required along these HII structures, and models for heating based on MHD and shock mechanisms were considered (Morris \\& Yusef-Zadeh 1989; Serabyn \\& \\gusten~1991).  Recently, however, high resolution infrared observations of this region have significantly advanced our understanding of the thermal radio structures in the Radio Arc. Two extraordinary clusters of young stars (the Quintuplet and Arches clusters) have been discovered, richly populated with O-stars and large numbers of Wolf Rayet and other highly evolved stellar types (Nagata et al. 1995; Cotera et al. 1996; Figer et al. 1995; Serabyn et al. 1998; Figer et al. 1999a,b). Both radio recombination line and far-infrared observations have shown that the Quintuplet, located at the center of curvature of the Sickle HII region, can provide adequate ionization of the ionized gas in the Sickle (Lang et al. 1997; Simpson et al. 1997). However, ionization of the Arched Filaments has been more difficult to understand in terms of a single cluster, in part due to the large extent and unusual morphology: the Arched Filaments cover 22 $\\times$ 16 pc with an areal filling factor of at most 10\\%. The uniformity of ionization properties over such a large region, as derived from Kuiper Airborne Observatory (KAO) far-infrared observations, made the Arched Filaments even more puzzling, as it would be very difficult to distribute stars near the cloud surface in such a way as to account for that uniformity (Colgan et al. 1996).  The recent recombination line study of LGM01, coupled with the recognition that the Arches stellar cluster generates a substantial ionizing flux ($\\sim$2 x 10$^{52}$ s$^{-1}$), has revealed that this cluster adequately accounts for the ionization of the thermal Arched Filaments. The uniformity of physical conditions in the ionized gas is likely to be due to the large distance at which the cluster is located from the Arched Filaments (as much as 20 pc along the line of sight). Several outstanding questions remain in understanding this complex, most notably, the physical arrangement of the ionized, molecular and stellar components. Detailed comparisons of the distribution and kinematics of the ionized and molecular gas may offer some insight into the interactions of these components.  Furthermore, the nature of the interaction between the thermal structures and the magnetic filaments is unclear. Serabyn \\& \\gusten~(1991) first noted that the molecular material near the Sickle HII region may be physically associated with the linear NTFs in the Radio Arc, and that the electrons in the ionized gas may be accelerated to relativistic energies via magnetic reconnection. Follow-up interferometric observations of the CS(2-1) line emission in this cloud showed that the molecular gas is distributed in discrete clumps corresponding well to positions where both NTFs and ionized gas are present, and where the NTFs undergo striking changes in brightness and continuity (SM94). These authors propose that magnetic field reconnection occurs between the cloud field and the field in the magnetic NTFs, thereby causing the acceleration of some of the particles to relativistic energies. The relativistic particles then are constrained to move along the NTFs as they emit synchrotron radiation. This model relies on the presence of three elements: (1) a molecular cloud with a surface moving at a relatively large velocity compared to the intercloud medium; (2) a turbulent, ionized surface on this cloud to provide electrons and sufficient mixing between the cloud and intercloud medium; and (3) a magnetic field in the partially ionized molecular cloud which has a different orientation than the ambient magnetic field. In fact, far-infrared polarization observations of the gas underlying the Sickle have confirmed that these clouds possess internal magnetic fields aligned along the Galactic plane and therefore perpendicular to the NTFs (Morris \\& Serabyn 1996). In the region of the Arched Filaments, however, the relationship between the molecular material and the Northern Thread NTF has not been explored, therefore providing an additional site to test the model of SM94.  In order to investigate (1) the arrangement of the ionized, molecular and stellar components in the Arched Filament complex, and (2) the nature of the intersection of the ionized and molecular gas with the Northern Thread NTF, we carried out high-resolution millimeter observations with the OVRO millimeter array of the molecular gas in a portion of the Arched Filaments known as G0.07+0.04. This portion of the Arched Filament H II complex is intersected by the Northern Thread NTF (Lang et al. 1999). The three elements required by the model of SG94 (molecular gas, ionized gas, and an NTF) are thus present. Figure 1 shows a diagram of the Arched Filaments with the field of view of the OVRO observations and the prominent sources labelled. In addition, we present a careful comparison of the morphology and velocity structure of the ionized and molecular gas across the Arched Filament region using the H92$\\alpha$ line observations of LGM01 and previously published single-dish molecular line observations of SG87 to provide constraints on the geometrical configuration of the components. Details of the OVRO observations, the data reduction, and the combination of the OVRO and 30-m data of SG87 are summarized in $\\S$2; the 3.4 mm continuum and the CS (2-1) line results are presented in $\\S$3, and $\\S$4 provides a discussion of the results. We assume throughout a distance to the Galactic center of 8.0 kpc (Reid 1993). ",
        "conclusions": "Detections of the 3.4 mm continuum and the CS(2-1) line arising from a region of the $-$30 \\kms~molecular cloud at the Galactic center were made using the Owens Valley millimeter array. These data were combined with the corresponding total power data obtained at the 30-m IRAM telescope by SG87. The following conclusions are made:  (1) Continuum emission at 3.4 mm was detected in the G0.07+0.04 region, coincident with the peak of 3.6 cm continuum emission. The spectrum between 3.4 mm and 3.6 cm appears to be flat, consistent with a detection at 3.4 mm of the free-free emission arising from the H II region and with a slight contribution from dust.  (2) The combined OVRO+30-m integrated CS(2-1) emission image shows that overall the molecular gas in the G0.07+0.04 region is distributed in a compact clump at the intersection of the ionized filament and Northern Thread NTF. However, there is a pronounced decrease in the CS(2-1) emission along the exact trace of the Northern Thread. This might indicate that the NTF and molecular gas are interacting, although the nature of the interaction is not clear.  (4) A comparison between the OVRO+30-m CS observations and the H92$\\alpha$ data of LGM01 in the region of G0.07+0.04 shows that the molecular and ionized gas are physically related, although the velocities are separated in some places by up to 35 \\kms, indicating that one component of the molecular gas in this direction may have been fully ionized.  (5) A larger-scale comparison between the 30-m CS observations and the H92$\\alpha$ data from LGM01 also indicates that the molecular and ionized gas are closely associated across the entire region of the Arched Filaments, as was first discussed by SG87. The velocities, velocity gradients and morphology suggest that the ionized gas is physically related to this molecular cloud.  (6) The geometrical arrangement of the stellar and gaseous components in the Arched Filament region is complex. Over much of the eastern Arched Filaments (E1 and E2) the ionized gas appears to lie on the near edge of the molecular gas, yet in the western filaments (W2 in particular), some molecular material must lie in front of the ionized gas, consistent with HI absorption results (Lasenby et al. 1989). The narrow and curved morphology of the Arched Filaments therefore suggests that the molecular cloud has a finger-like distribution of molecular material, the edges of which are ionized. The cluster is likely to be embedded within the distribution of molecular gas such that some of the cloud surfaces along our line of sight have not been exposed to the ionizing radiation."
    },
    "0208/astro-ph0208054_arXiv.txt": {
        "abstract": "{ The 352~MHz Westerbork In the Southern Hemisphere (WISH) survey is the southern extension of the WENSS, covering 1.60~sr between $-9\\degr < \\delta < -26\\degr$ to a limiting flux density of $\\sim$18~mJy ($5\\sigma$). Due to the very low elevation of the observations, the survey has a much lower resolution in declination than in right ascension ($54\\arcsec \\times 54\\arcsec {\\rm cosec}\\delta$). A correlation with the 1.4~GHz NVSS shows that the positional accuracy is less constrained in declination than in right ascension, but there is no significant systematic error. We present a source list containing 73570 sources. We correlate this WISH catalogue with the NVSS to construct a sample of faint Ultra Steep Spectrum (USS) sources, which is accessible for follow-up studies with large optical telescopes in the southern hemisphere. This sample is aimed at increasing the number of known high redshift radio galaxies to allow detailed follow-up studies of these massive galaxies and their environments in the early Universe. ",
        "introduction": "Powerful radio sources provide excellent targets to probe the formation and evolution of galaxies out to cosmological distances. The Hubble $K-z$ diagram of radio and near$-$IR selected galaxies shows that at $z \\simgt 1$, the host galaxies of powerful radio sources are $>$2 magnitudes brighter than HDF field galaxies \\citep{deb02}. Because there are strong arguments that this $K-$band emission is due to starlight, and not due to direct or scattered AGN contributions, high redshift radio galaxies (HzRGs) are among the most massive galaxies known at high redshift. This is consistent with the observations at low redshifts ($z \\simlt 1$), where radio galaxies are uniquely identified with massive ellipticals \\citep[\\eg][]{bes98,mcl00}. Because HzRGs pinpoint over-dense regions in the early Universe, they have also been successfully used as tracers of proto-clusters at very high redshifts \\citep[\\eg][]{lef96,pas96,pen00,ven02}. \\begin{figure*}[ht] \\centering \\includegraphics[width=9cm,angle=-90]{MS2836f1.ps} \\caption{Sample $u-v$ coverage plot of the field centered at $\\alpha=1^h28^m, \\delta=-20^{\\circ}$. The $v-$coverage is limited due to the low declination. Note the excellent radial coverage due the the bandwidth synthesis technique.} \\label{sampleUV} \\end{figure*}  The first redshift surveys of radio sources targeted only the brightest objects in the sky \\citep[3CR][]{ben62,spi85}. Present-day surveys reach flux densities several orders of magnitude fainter than the 3CR \\citep[\\eg][]{mcc96,lac99,raw01,wil02}. However, because the optical spectroscopy of the host galaxies requires substantial integration times on 3$-$10m class telescopes, these flux density limited surveys necessarily need to be limited in sky area. This excludes the objects with the lowest space density, such as the most distant luminous radio galaxies \\citep[\\eg][]{blu98,jar01b}. To find such objects, additional 'high redshift filters' need to be applied to the samples of radio surveys, at the expense of completeness. The most efficient filter is the selection of sources with ultra steep radio spectra (USS; $\\alpha \\lesssim -1; S \\propto \\nu^{\\alpha}$). The success of this USS technique is mainly based on the $k-$correction of the generally concave radio spectrum of powerful radio galaxies. Several such USS samples have shown to be much more efficient in finding $z>2$ radio galaxies than the complete samples \\citep[\\eg][]{rot97,ste99,deb01,jar01a}.  It is now possible to define large, well defined samples of USS sources using a new generation of large area radio surveys: the Westerbork Northern Sky Survey \\citep[WENSS; 325~MHz][]{ren97}, the Texas survey \\citep[365~MHz]{dou96}, The Sydney University Molonglo Sky Survey \\citep[SUMSS][]{boc99}, the NRAO VLA Sky Survey \\citep[NVSS; 1.4~GHz][]{con98}, and the Faint Images of the Radio Sky at Twenty centimeters \\citep[FIRST; 1.4~GHz][]{bec95}. \\citet{deb00} have used these surveys to define a sample of 669 USS sources covering the entire sky outside the Galactic plane. However, their samples necessarily favour the northern hemisphere, because the WENSS survey, which is an order of magnitude deeper than the Texas survey, covers only the sky at $\\delta > +29$\\degr. In this paper, we introduce the Westerbork In the Southern Hemisphere (WISH) survey, the southern extension of the WENSS. We use WISH in combination with NVSS to define a fainter sample of USS sample in the $-9\\degr< \\delta < -26\\degr$ region, in analogy with the northern WENSS$-$NVSS sample of \\citet{deb00}. The construction of such a southern hemisphere sample is especially timely due to the advent of several 8m class telescopes in the southern hemisphere, which can be used for the optical/near$-$IR identification and spectroscopy of the host galaxies.  The layout of this paper is as follows. In \\S 2, we introduce the WISH survey, and compare the data products with the WENSS. In \\S 3, we define the WISH$-$NVSS USS sample. \\S 4 compares this new sample with previous samples and \\S 5 concludes with an overview of the planned observations of this sample. ",
        "conclusions": "The WISH survey is the deepest low-frequency survey covering roughly a quarter of the area between the WENSS and SUMSS surveys ($-30\\degr<DEC<+28\\degr$). Because it has been observed at very low elevations, it has relatively poor declination resolution compared with the NVSS. It is sufficiently deep to provide spectral index information for $\\sim$42000 sources in common with the NVSS. This provides a unique sample of faint USS sources which can be observed with southern hemisphere telescopes. We have obtained VLA and/or ATCA snapshot observations of a first batch of 69 sources from this sample to provide more accurate positions and morphological information needed for the identification of the host galaxies. A campaign of $K-$band identifications with CTIO, and optical spectroscopy with the VLT has already identified several new $z>3$ radio galaxies from this sample (de Vries \\etal, in preparation)."
    },
    "0208/astro-ph0208262_arXiv.txt": {
        "abstract": "Spectra taken with the Space Telescope Imaging Spectrograph (STIS) allow accurate location and extraction of the nuclear spectrum of NGC 4151, with minimal contamination by extended line emission and circumnuclear starlight. Spectra since 1997 show that the P Cygni Balmer and He I absorption seen previously in low nuclear states, is present in higher states, with outflow velocity that changes with the nuclear flux. The phenomenon is discussed in terms of some of the absorbers seen in the UV resonance lines, and outflows from the central source and surrounding torus. ",
        "introduction": "NGC 4151 is the brightest Seyfert 1 type galaxy, and has been studied in considerable detail at all wavelengths. In reference to the nuclear region, the Hubble Space Telescope has been instrumental in resolving the innermost few arcseconds, and revealing the spatial and velocity structure of the narrow emission line gas (see e.g. Hutchings et al 1998, Kaiser et al 2000, Nelson et al 2000, Crenshaw et al 2000). The detailed picture that emerged, for NGC 4151 and other Seyferts, is of a hollow biconical outflow of narrow-line clouds. In the case of NGC 4151, the high velocity radio jets lie along the cone axis, and our line of sight lies close to the edge of the approaching cone. This scenario was put forward earlier by, for example, Pedlar et al (1993) and Boksenberg et al (1995).  NGC 4151 is also known to show flux variations over a factor of ten or more, and has been the subject of echo-mapping observational campaigns. These have shown the inner broad emission line region to have an extent of several light days (see overview by Peterson et al 1998). The brightness and spatial extent of the narrow emission lines has made it difficult to isolate the nuclear spectrum and emission lines. It was noted originally by Anderson and Kraft (1969) that there are shortward shifted absorptions in H$\\gamma$ and the metastable He I $\\lambda$3888 line. Anderson (1974) followed up with further data and a discussion that suggested a connection between the continuum flux and the absorption strength. This has been poorly documented since, but Sergeev, Pronik, and Sergeeva (2001) give a summary of observations over 11 years that show the absorptions are present in a nuclear low state in 1999. No systematic study has been made of the absorptions, perhaps because ground-based observing conditions cause a large range of contamination by the circumnuclear flux, both line and continuum.  The long slit (or slitless) spectroscopic capability of STIS, along with the spatial resolution of the HST, has made it possible to obtain and study the nuclear spectrum consistently and cleanly. There is extended narrow line emission with many velocity components, even within the central arcsec, which can affect the overall line profiles, if included. In this paper we discuss the series of visible range nuclear spectra from STIS, that fortuitously cover a wide range of nuclear flux variations. We are particularly interested in the outflow absorption that is seen in the strong Balmer and the metastable He I line.  Outflow is also seen in higher velocity emission line clouds near the nucleus (Hutchings et al 1999), multiple shifted absorption lines in C IV and other UV resonance lines (Weymann et al 1997, Crenshaw et al 2000, Kriss et al 2002), and warm absorbers seen in X-ray data (see e.g. Schurch and Warwick 2002). A full picture of the different outflows has yet to emerge, and this paper adds further information to the inventory. ",
        "conclusions": ""
    },
    "0208/astro-ph0208504_arXiv.txt": {
        "abstract": "We  investigate the properties of astrophysical electromagnetic  cascades in matter,  photon gas   and  magnetic fields, and  discuss  similarities and differences between characteristics of electron-photon showers developed in these 3 substances.  We apply  the same computational technique based on  solution of the  adjoint cascade equations to all 3 types of cascades, and  present precise numerical calculations of  cascade curves  and broad-band energy spectra of secondary electrons  and  photons  at different penetration lengths. ",
        "introduction": "Relativistic electrons --  directly accelerated,  or being secondary products of various hadronic processes -- may result in copious $\\gamma$-ray production  caused by  interactions with ambient targets in forms of {\\em gas (plasma),  radiation} and {\\em magnetic fields}. In  different astrophysical environments $\\gamma$-ray production may proceed  with high efficiency through {\\em bremsstrahlung},  {\\em inverse Compton scattering} and {\\em synchrotron (and/or  curvature) radiation},   respectively. Generally,  $\\gamma$-ray  production is  effective  when the cooling time that characterizes   the  rate of the process does  not  significantly  exceed {\\em (i)} the  source (accelerator) age {\\em (ii)} the  characteristic time of  non-radiative losses  caused by   adiabatic expansion of the source or  particle escape  and   {\\em (iii)} the cooling  time of  competing  radiation mechanisms  that result in  low-energy photons  {\\em outside}  the $\\gamma$-ray domain. As long as the charged particles are effectively confined in the $\\gamma$-ray production region, at some circumstances these  condition   could be  fulfilled  even in environments  with a relatively low  gas and photon densities  or weak magnetic field. More specifically,   the $\\gamma$-ray production efficiency could be close to 1 even when  $t_{\\rm rad} \\gg R/c$ ($R$ is  the characteristic linear size of the production region, $c$ is the speed of light). In  such  cases the secondary  $\\gamma$-rays  escape the  source  without significant internal absorption. Each of the above mentioned $\\gamma$-ray {\\em production} mechanisms has its major ``counterpart'' - $\\gamma$-ray {\\em absorption} mechanism of same electromagnetic origin resulting in  electron-positron pair production in matter (the counterpart of bremsstrahlung), in photon gas (the counterpart of inverse Compton scattering), and in magnetic field (the counterpart of synchrotron radiation). The  $\\gamma$-ray production  mechanisms and their absorption counterparts have similar cross-sections, therefore the condition  for radiation $t_{\\rm rad}  \\geq  R/c$ generally implies small optical depth for the corresponding $\\gamma$-ray absorption mechanism, $\\tau_{\\rm abs} \\leq  1$. But in many astrophysical  scenarios, in particular in compact galactic and extragalactic objects  with  favorable conditions for particle acceleration,  the radiation processes proceed so fast that $t_{\\rm rad}  \\geq  R/c$. At these conditions the internal $\\gamma$-ray absorption  becomes unavoidable. If  the $\\gamma$-ray production and absorption  processes occur  in  relativistic  regime,  namely   when (i) $E_{\\rm \\gamma, e} \\geq  10^3  m_e c^2$ in the hydrogen gas, (ii) $E_{\\rm \\gamma,e} \\epsilon \\gg m_{\\rm e}^2c^4$ in photon gas  (often called Klein-Nishina regime; $\\epsilon$ is the average energy of the target photons), or (iii) $(E_{\\rm \\gamma,e} /m_{\\rm e}c^2)(H/H_{\\rm cr}) \\gg 1$ in the magnetic field (often called quantum regime; $H_{\\rm cr}\\simeq 4.4 \\times 10^{13} \\ \\rm G$ is the so-called critical strength of the  magnetic field), the problem cannot be reduced to a simple  absorption effect. In this regime,  the secondary electrons produce new generation of high energy $\\gamma$-rays, these  photons again produce electron-positron pairs, so the electromagnetic cascade develops.  The  characteristics of electromagnetic cascades  in matter have been comprehensively studied,  basically  in the context of interactions of cosmic rays with the Earth's atmosphere (see e.g.  \\cite{matter})  as well as  for calculations of performance of detectors of high energy  particles   (e.g. \\cite{EGS4}).  The theory of electromagnetic  cascades in matter can be applied to some   sources  of high energy cosmic radiation, in particular to the  ``hidden source''   scenarios like massive black holes in centers of AGN or young pulsars  inside the dense shells of recent supernovae explosions  (see e.g. \\cite{Berez_book}). Within another,   so-called ``beam dump'' models (see e.g. Ref. \\cite{Berez_book,Halzen}) applied to X-ray binaries,  protons accelerated by the compact object (a neutron star or black hole), hit the atmosphere of the normal companion star, and thus  result in   production of high energy neutrinos and  $\\gamma$-rays \\cite{Ber_Vest}. In such objects,  the thickness of the surrounding  gas can significantly exceed  $100 \\ \\rm g/cm^2$, thus the  protons    produced  in the central source would  initiate (through production of high energy $\\gamma$-rays and electrons) electromagnetic showers. These  sources  perhaps represent the ``best hope'' of neutrino astronomy,  but they are generally considered as less  attractive  targets for gamma-ray astronomy.  However,  the $\\gamma$-ray emission  in  these objects is not  fully suppressed. The recycled radiation  with spectral features  determined by  the  thickness (``grammage'')  of the gas shell, should be seen in $\\gamma$-rays in any case, unless the synchrotron radiation of secondary electrons dominates over the bremsstrahlung losses and  channels the main  fraction of the nonthermal energy into the sub gamma-ray domains.  The development of electromagnetic cascades in photon gas and magnetic fields is  a  more common phenomenon in astrophysics. In photon fields  such cascades can be created on  almost {\\em all} astronomical scales,  from  compact objects like accreting black holes \\cite{AVK,Zdz,Svenson,CopBl,mastiprot}, fireballs in gamma-ray  bursts \\cite{GRBs} and sub-pc jets of  blazars \\cite{blazars} to large-scale (up to $\\geq 100$ kpc) AGN  jets  \\cite{bierman,gamma_jets} and  $\\geq 1$ Mpc size clusters of galaxies \\cite{psynch}. Electromagnetic cascades in the intergalactic medium lead to formation  of huge ($\\geq 10$ Mpc) nonthermal structures like hypothetical electron-positron pair halos  \\cite{halos}. Finally, there is little doubt that the entire Universe is a scene of  continuous  creation and development of  electromagnetic cascades. All  \\grs of energy $\\geq$ a few GeV emitted by  astrophysical objects   have a similar  fate. Sooner or later they terminate  on Hubble scales due to interactions with the diffuse extragalactic background. Since the energy density of 2.7 K CMBR significantly exceeds the energy density of intergalactic magnetic fields, these interactions initiate electron-photon cascades  \\cite{intergalactic}. The superposition of contributions of \\grs from these cascades should constitute a significant fraction of the observed diffuse extragalactic background.   Bonometto and  Rees \\cite{Bo_Rees} perhaps where the first who realized the astrophysical importance  of  development of electron-photon cascades  supported by $\\gamma$-$\\gamma$ pair-production and  inverse Compton scattering in dense photon fields. When the so called compactness parameter  \\cite{Guilbert} $l=(L/R) (\\sigma_{\\rm T}/m_{\\rm e}c^3)$  (L is the luminosity and R is the radius  of the source) is less than 10, then the cascade is developed  in the linear regime,  i.e. when  the soft radiation produced by cascade electrons do not have a significant feedback effect on the cascade development. In many cases, including the cascade development in compact objects, this approximation works quite well.  The first quantitative study of characteristic of linear cascades in photon fields has been performed using  the method of Monte Carlo simulations  \\cite{Ah_vard_kir}. Generally,  the kinetic equations for the cascade particles can  be  solved only numerically. However, with some simplifications it is possible to derive useful analytical approximations \\cite{Zdz,Svenson} which help to understand the features of  the  steady-state solutions for  cascades in photon fields.  The cascade development in the magnetic field is a key element to understand the physics of pulsar magnetospheres \\cite{sturrock,HardBaring}, therefore it is  generally treated as a process associated with very   strong magnetic fields. However, such cascades  could be triggered in many other (at first glance unusual) sites like the Earth's geomagnetic field \\cite{gonchar,Anguelov,Plyah}, accretion disks of massive black holes \\cite{Bednarek}, etc.   In general, the  pair cascades in  magnetic fields  are  effective when  the product of the particle (photon  or electron) energy and the strength of the field becomes close  to  the ``quantum threshold'' of about $H_{\\rm crit} m_{\\rm e}c^2 \\simeq 2 \\times 10^7 \\ \\rm TeV \\cdot G$, unless we assume a  specific,  regular field configuration. An approximate approach, similar to the so-called approximation A of cascade development in  matter \\cite{matter}, has been recently applied  by Akhiezer et al. \\cite{Akhiezer}. Although this theory quite satisfactorily describes  the basic features of photon-electron showers, it does not provide  an adequate accuracy for quantitative  description of the cascade characteristics \\cite{Anguelov}. Note that in both studies approximate  cross-sections for magnetic bremsstrahlung and magnetic pair-production have  been used, thus reducing the validity  of the results  to the limit $E H \\gg 10^7  \\ \\rm TeV \\cdot G$ ($E$ is the minimum energy of secondary particles being under consideration).  As long as we are interested in the one-dimensional cascade development (which seems to be quite sufficient for many astrophysical purposes),  all 3 types of cascades can be described by the same  integro-differential equations like the ones derived by Landau and Rumer \\cite{Landau},  but  in each case specifying   the cross-sections of relevant  interaction processes. The solution of these equations in a broad range of energies is however not a trivial  task. In this paper we present the results of our recent study of cascade characteristics in 3 substances - matter, photon gas and magnetic field - with emphasis on the analysis of similarities and differences between these 3 types of cascades. For quantitative studies of these characteristics we have chosen the so-called   technique of adjoint cascade equations. Although  this  work has been initially motivated by methodological and pedagogical  objectives, some results  are rather original and  may present practical interest in certain areas of high energy astrophysics. ",
        "conclusions": "In this study,   the technique of adjoint cascade equations has been applied to  investigate properties of electron-photon  cascades in  hydrogen gas and  in ambient radiation and magnetic fields. We also have inspected   the main features of cross-sections of relevant processes that initiate  and support cascade developments  in these substances.  The cascade curves of electrons and photons in the photon gas and magnetic field have features quite different  from the cascade curves in matter. The energy spectra of cascade particles are also considerably different from the conventional cascade spectra in matter. The spectra for the magnetic field have properties intermediate between those for cascade spectra in matter and in the photon field. Although  for  certain astrophysical scenarios  the development of cascades in ``pure'' environments  can be considered as an appropriate and fair approximation, at some conditions the interference of processes associated with interactions of cascade electrons and \\grs with the  ambient photon gas and magnetic field (or matter) can significantly change the character of cascade development, and consequently the spectra of observed $\\gamma$-rays. The impact is very complex and quite sensitive to the choice of specific parameters. Therefore each practical case should be subject to  independent studies."
    },
    "0208/astro-ph0208218_arXiv.txt": {
        "abstract": "We propose a new empirical method to estimate the total far-infrared flux of galaxies from the spectral energy distribution (SED) at wavelengths $\\lambda \\leq 100\\; \\micron$. It is difficult to derive the total far-infrared luminosity from only the IRAS data, though it is one of the most important properties of galaxies. Observations by Infrared Telescope in Space (IRTS) indicate that the SED of the diffuse emission from the Galactic plane in this wavelength region can be derived from the $60\\; \\micron$ to $100\\; \\micron$ color. This empirical SED relation was improved in order to obtain a better fit to the Galactic plane data for $I_\\nu(60\\;\\micron)/I_\\nu(100\\;\\micron) > 0.6$, and applied to 96 IRAS galaxies for which ISOPHOT and KAO data are available at $\\lambda > 100\\; \\micron$. As a result, the empirical relation describes well the far-infrared (FIR) SED for a majority of the galaxies. Additionally, the total FIR flux for $\\lambda \\geq 40\\; \\micron$ is derived from the flux densities at 60 and $100\\; \\micron$ by using this model. For the 96 IRAS galaxies, the uncertainty in the total far-infrared flux of the present method is 26~\\%. The present method is more accurate than the previous one widely used to derive the total infrared flux from the IRAS 60 and $100\\; \\micron$ data. ",
        "introduction": "\\label{sec:intro}  In the interstellar space of galaxies, there exist solid particles of sub-micron sizes, so called ``interstellar dust grains.'' Earlier work (e.g., \\cite{beichman}; \\cite{soifer1}) has revealed that the far-infrared emission from galaxies is often dominated by thermal emission from such dust grains. Therefore, the total luminosity in the far-infrared from the interstellar dust grains is the most important indicator of star-formation activity or activity of hidden AGNs. Moreover, the far-infrared color is the most direct indicator of the strength of interstellar radiation field (ISRF).  Over the past two decades, many studies have been made to understand the properties of infrared emission from interstellar dust grains. In order to explain the observed spectral energy distribution (SED) of the Galactic plane, \\citet{desert} proposed a three-component model that consists of large grains (LGs), very small grains (VSGs), and polycyclic aromatic hydrocarbons (PAHs). The large grain is the classical interstellar dust grain of sub-micron size, which is thought to be in equilibrium with the interstellar radiation field. Bulk of the emission from the large grains is radiated in the far-infrared and sub-millimeter wavelength regions. The very small grain is the dust grain heated transiently up to $\\mbox{a few}$ hundred~K by the incidence of a single UV photon. A number of dust models (e.g., \\cite{draine2}) have shown that both stochastically heated VSGs with radii of less than 10~nm and very large molecules with up to a few hundred atoms could be the source of the excess emission shorter than 100 $\\micron$. \\citet{dwek} showed that the mean infrared SED of the Galactic plane was fitted with a composite model of silicate and graphite grains. If we take the silicate dust grain to correspond to the large grain, and the graphite to the very small grain, these two models have essentially the same approach to reproduce the observed SED of the Galactic plane.  By using the Infrared Telescope in Space (IRTS) and the Infrared Astronomical Satellite (IRAS) data, \\citet{okumura2} and \\citet{shibai} proposed another approach. They assumed that the SED between 100 and 200 $\\micron$ is dominated by a single-component dust with an emissivity index of 2 at a nearly constant temperature. They showed that the observed SED can successfully be fitted with the single component model at a constant temperature, which means that the strength of the interstellar radiation field (ISRF) is nearly a constant, except for the high-temperature regions around distinct H\\emissiontype{II} regions. However, the diffuse infrared emission at wavelengths $\\lambda$ shorter than $100\\; \\micron$ cannot be explained by the thermal emission from the large grains, and shows the intensity is significantly larger. \\citet{okumura2} found that the infrared SED varies with the strength of interstellar radiation field, and the $12\\mbox{--}60\\; \\micron$ SED is strongly correlated with that at $\\lambda \\geq 100\\; \\micron$. \\citet{shibai} obtained the same result from the Galactic plane data ($|b| < 5$ degree) of COBE/DIRBE. These results suggest that the temperature of the large grain is a key parameter to characterize the Galactic diffuse infrared emission.  Extending the Galactic plane result to external galaxies, one would like to use the largest available database i.e., the IRAS survey (e.g., \\cite{soifer1}). But because of the lack of photometric band at $\\lambda > 100\\; \\micron$, the IRAS observations alone are not adequate to characterize the emission from dust cooler than 30~K. The ISO and the Submillimeter Common-User Bolometer Array (SCUBA) data have shown that the far-infrared emission from late-type galaxies has a substantial contribution from large grains whose temperature is about 15--25 K (\\cite{alton}; \\cite{haas}; \\cite{dunne}). If sub-millimeter fluxes are included, the total far-infrared emission increases by a factor of 2--3 with respect to previous estimates based on IRAS data alone (e.g., \\cite{helou2}). The derived large grain temperatures of galaxies are significantly lower than the previous results estimated from the IRAS $60\\; \\micron$ and $100\\; \\micron$ flux densities (e.g., \\cite{devereux}). Additionally, \\citet{devereux} and \\citet{alton} also suggested that the dust masses of spiral galaxies derived from sub-millimeter/ISO fluxes are ten times larger than those derived only from the IRAS data. Therefore, the SED at $\\lambda > 100\\;\\micron$ is crucial to estimate the thermal emission from large grains correctly, and to understand the physical environments in galaxies.  To derive the total infrared luminosities of IRAS galaxies, \\citet{xu2} have proposed an empirical relation between the ratio of the integrated far-infrared flux (40 -- 120 $\\micron$) to the total luminosities (8--1000~$\\micron$) and the 60-to-100 $\\micron$ flux densities ratio from 13 late type galaxies. Here, we chose another approach based on the dust emission properties to derive the total infrared luminosities of IRAS galaxies. We derived the large grain temperatures and the total radiation energies from external galaxies in the far-infrared region by using the relationship between the COBE/DIRBE 60 $\\micron$, 100 $\\micron$ and 140 $\\micron$ intensities in the Galactic plane derived by \\citet{shibai}. We also adopted a new empirical model. The present method improves not only the previous evaluation of the far-infrared total luminosities of IRAS galaxies, but also provides the key parameters to characterize galaxies, e.g., the temperature and the dust mass of galaxies.  In this paper, we describe our empirical method to derive the total infrared luminosities of galaxies. The dataset and analysis are explained in Section 2. The results and discussion are presented in Section 3 and Section 4, respectively. We summarize the results and present conclusions in Section 5. ",
        "conclusions": "We have successfully improved the empirical SED model proposed by \\citet{okumura2} and \\citet{shibai} in order to obtain a better fit to the diffuse infrared emission from the Galactic plane. This new model is more accurate at higher color temperature. We applied the model to the infrared SEDs of nearby galaxies observed by IRAS and ISO. \\begin{enumerate} \\item We found that the majority of the sample galaxies showed a trend similar to the Galactic plane, but some galaxies approach a single temperature Planck function with the emissivity index of 2 when the large grain temperature, $T_{\\rm LG}$, is higher than 25 K. This fact suggests that for galaxies of $T_{\\rm LG}<25$ K the diffusely distributed dust grains at a single temperature heated by the general interstellar radiation field dominate as in the Galactic plane, while, for some galaxies of $T_{\\rm LG}>25$ K, discrete warm components from H\\emissiontype{II} regions significantly contribute to the averaged SED in the far-infrared region and possibly dominate the 60 $\\micron$ and $100\\; \\micron$ bands over the diffuse component. \\\\ \\item The SED at $\\lambda >100\\; \\micron$ can be derived from the flux densities at 60 and $100\\; \\micron$ by the present model. For 96 IRAS galaxies, the uncertainty in the total far-infrared flux for $\\lambda \\geq 40~\\micron$ derived from the present method is 26~\\%. This method allows us to obtain more accurately the important properties of galaxies, such as dust mass, large grain temperature, and total infrared flux of IRAS galaxies. \\end{enumerate}  \\bigskip We thank to Drs. K.\\ Okumura and T.\\ Hirao for their important suggestions, and to the IRTS and ASTRO-F members for useful discussion. We wish to express our gratitude to Dr. T.\\ N.\\ Rengarajan for his critical reading of this paper. We thank Dr. Simone Bianchi, the referee, for his careful reading and useful comments that improved this paper very much. This research has made use of the NASA/IPAC Extragalactic Database (NED), which is operated by the Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space Administration.  One of the authors (Tsutomu T. Takeuchi) is financially supported by the JSPS Fellowship. This research has been supported in part by a Grant-in-Aid for the Scientific Research Fund (10147102) of the Ministry of Education, Science, Sports and Culture in Japan."
    },
    "0208/astro-ph0208442_arXiv.txt": {
        "abstract": "The spatial distribution of star formation within  galaxies strongly affects the resulting feedback processes. Previous work has considered the case of a single, concentrated nuclear starburst, and also that of distributed single supernovae (SNe). Here, we consider ISM structuring by SNe originating in spatially distributed clusters having a cluster membership spectrum given by the observed \\hii\\ region luminosity function.  We show that in this case, the volume of \\hi\\ cleared per SN is considerably greater than in either of the two cases considered hitherto.  We derive a simple relationship between the ``porosity'' of the ISM and the star formation rate (SFR), and deduce a critical $\\SFR_{\\rm crit}$, at which the ISM porosity is unity.  This critical value describes the case in which the SN mechanical energy output over a timescale $t_e$ is comparable with the ISM ``thermal'' energy contained in random motions; $t_e$ is the duration of SN mechanical input per superbubble. This condition also defines a critical gas consumption timescale $t_{\\rm exh}$, which for a Salpeter IMF  and random velocities of $\\simeq 10\\ {\\rm km\\ s}^{-1}$ is roughly $10^{10}$ years.  We draw a link between porosity and the escape of ionising radiation from galaxies, arguing that high escape fractions are expected if $\\SFR \\simgreat \\SFR_{\\rm crit}$.  The Lyman Break Galaxies, which are presumably subject to infall on a timescale $<t_{\\rm exh}$, meet this criterion, as is consistent with the significant leakage of ionising photons inferred in these systems.  We suggest the utility of this simple parameterisation of escape fraction in terms of SFR for semi-empirical models of galaxy formation and evolution and for modeling mechanical and chemical feedback effects. ",
        "introduction": "Feedback from supernovae (SNe) is a major ingredient of most contemporary models for galaxy formation and evolution.  One of the main motives for its inclusion is the need to reconcile the predictions of CDM cosmology with the observed galaxy luminosity function (e.g. Cole et al 1994, 2000; Efstathiou 2000): the overproduction of dwarf galaxies in CDM models can be alleviated if star formation is inefficient in low mass systems, and the explosive energy input from SNe provides an obvious mechanism for ejecting gas from the shallow potentials of dwarf galaxies. SNe also return metals to the ISM, and if this process is coupled with the strong outflows posited above, can also provide a mechanism for enriching the IGM (e.g., Dekel \\& Silk 1986; Madau, Ferrara and Rees 2001).  Another aspect of SN-driven feedback that has received rather less attention is its  possible relation to the escape of ionising radiation from star forming galaxies. It is currently unclear whether stellar sources or quasars are mainly responsible either for the re-ionisation of the Universe at high redshift or for the present day ultraviolet background (Madau, Haardt and Rees 1999), the answer depending critically on the assumed fraction of Lyman continuum photons that are able to escape star forming galaxies (Giallongo et al 1997; Madau and Shull 1996; Bianchi et al 2001). The simplest parameterisation of this problem involves assuming a constant escape fraction for all galaxies, but there are obvious reasons for supposing that in reality the escape fraction should depend on galactic parameters. In particular, the escape fraction, which depends on the distribution of neutral hydrogen along the line of sight, is likely to be strongly affected by the the re-structuring of the ISM effected by SN explosions. At a qualitative level, it would seem evident that higher escape fractions should be expected in systems with a higher star formation rate, since the disintegration of interacting SN-driven bubbles can in principle open up lines of sight through which Lyman continuum photons can leak.  Some observational support for the notion is provided by the recent detection of Lyman continuum emission in the composite spectra of Lyman Break Galaxies (Steidel et al 2001), which include systems undergoing vigorous star formation.  To date, however, calculations of escape fractions in star forming galaxies do not exhibit a strong  dependence of escape fraction on star formation rate (SFR); in the case of photoionisation calculations in a smoothly stratified ISM (Dove and Shull 1994; Ricotti and Shull 2000; Wood and Loeb 2000) the derived low escape fractions are not strongly dependent on the SFR, provided this exceeds the threshold value at which the resulting \\hii\\ regions cease to be ionisation bounded. Recent calculations by Ciardi et al (2001), demonstrate an even weaker dependence of escape fraction on SFR in the case of an inhomogeneous, fractal ISM. In all these calculations, however, the assumed ISM structure is independent of star formation activity. The effects of mechanical feedback on the escape of ionising radiation have so far only been considered in the case of a coeval burst of star formation, where the source of ionising radiation and SN energy originate in the same region (Tenorio-Tagle et al 1999; Dove, Shull and Ferrara 1999): here the temporary trapping of the ionisation front in the wall of the SN-driven bubble and the steep temporal decline of the ionising radiation from the burst combine to also produce a low escape fraction.  In this paper, we consider the case  where the photons from each  OB association impact on an ISM that has been structured by SN-blown superbubbles, reflecting its star formation history over the last  $10^7-10^8$ years. We thus draw a link between the {\\it{porosity}} of the ISM, i.e., the fraction of the ISM that is devoid of \\hi\\ due to the expansion of SN-driven bubbles, and the resulting escape fraction of ionising radiation. In order to quantify this link in more detail, one requires both  3D  hydrodynamic calculations, which model the break up of bubble walls following the interaction of adjoining bubbles, and radiative transfer calculations, which calculate the escape of ionising radiation from the resulting distribution of HI (see Fujita et al 2002).  Here, we attempt neither of these exercises but instead, having posited that the escape fraction should rise steeply as the porosity of the ISM approaches unity, set up a model of the ISM which allows the porosity to be readily calculated as a function of SFR and galactic parameters.  In this work, we focus on the {\\it volume,} $V_{\\rm sn}$ of the ISM that is cleared of neutral hydrogen per unit SN explosion. This value depends strongly on the spatial distribution and clustering properties of the SN progenitors.  Since superbubbles expand to the point that they are in rough pressure equilibrium with the ISM, it follows that an upper limit to this quantity is $V_{\\rm max}$, the volume of ISM originally containing a thermal energy equal to the SN energy ($\\sim 10^{51}$ ergs).  Note that in this work, we consider ``thermal energy'' to refer to both macroscopic and microscopic random motions in the ISM.  To date, most previous analyses have made either one of two extreme assumptions regarding the distribution of SN progenitors: either they are distributed singly throughout the galaxy  (Larson 1974; McKee and Ostriker 1977; Dekel and Silk 1986; Efstathiou 2000) or else, as in  more recent studies (De Young \\& Heckman 1994; Suchkov et al 1994; Mac Low and Ferrara 1999; Strickland \\& Stevens 2000) they are concentrated in a single coeval, cospatial burst of star formation. Each of these assumptions however predicts a volume cleared per SN that is much less than $V_{\\rm max}$, though for different reasons in the two cases. In the case of single SNe, the limiting factor is the cooling of the shocked ISM (Cox 1972; Chevalier 1974; Goodwin et al 2002), following which the bubble expands as a pressure driven snow plough. Due to these radiative losses, the final bubble encloses a volume of ISM that is a few per cent of $V_{\\rm max}$ (Cioffi et al 1988). In the case of clustered SN progenitors, the effect of multiple SNe may be modeled as a continuous input of mechanical energy, analogous to a stellar wind (McCray \\& Kafatos 1987; Mac Low \\& McCray 1988), and in this case it is found that cooling is of marginal importance.  However, in the case of a luminous starburst in a disc galaxy, the superbubble blows out when its size is between one and two disc scale heights. The bubble interior is consequently de-pressured by the loss of hot gas normal to the disc plane, and again the resulting cavity in the ISM implies a volume cleared per SN that is much less than $V_{\\rm max}$.  In this paper we examine the case in which SNe are distributed according to the observed distribution of OB stars in galaxies. Specifically, we consider spatially distributed OB associations and superclusters whose membership numbers are inferred from the observed luminosity function of \\hii\\ regions and OB associations  (Oey and Clarke 1998; McKee and Williams 1997).  The  OB association membership function is such that the number of associations having numbers of stars between $N_*$ and $N_* + dN_*$ is $ \\propto N_*^{-2}$. This distribution  is similar in functional form to the observed mass distribution of clumps within molecular clouds and the membership number function for all types of stellar clusters (Blitz 1991; Elmegreen and Clemens 1985; Harris and Pudritz 1994; Elmegreen \\& Efremov 1997; Meurer et al. 1995).  In a previous paper (Oey and Clarke 1997), we quantified the distribution of \\hi\\ hole sizes predicted by such a model and found it to be in excellent agreement with the observed distribution of \\hi\\ holes in the Magellanic Clouds (Oey \\& Clarke 1997; Kim et al. 1999). This empirical success encourages us to assume that this scale-free distribution of OB association richness is a universal characteristic of the ISM on all scales and in all environments. With such a prescription we can compute the fraction of the ISM that is cleared of \\hi\\ for a given SFR (Oey et al. 2001), and also follow the evolution of this quantity during an episode of star formation. As we show in Section 2.1, this model implies that the volume of ISM swept up per SN is considerably larger than in either of the two limits that have been considered to date, and may be a significant fraction of $V_{\\rm max}$.  Our analysis expands on the premise introduced by Oey et al. (2001), of a critical $\\SFR_{\\rm crit}$, such that lower rates produce a volume filling factor of holes that is considerably less than unity.  Under these circumstances, we surmise that the escape fraction of ionising radiation will be low, as found in photoionisation calculations based on a smoothly stratified ISM.  If the star formation rate exceeds this value, the ISM becomes filled with hot bubbles and we speculate that the escape fraction from the ISM may rise considerably at this point, as the widespread merging of bubbles leads to the break up of shell walls through a variety of instabilities. Disc systems may be able to continue to produce stars at such a rate, but in spheroidal systems, such as molecular clouds or proto-globular clusters, this seems unlikely, as the energy input into the gas from SNe at this point is comparable with the self-gravitational binding energy of the system.  We investigate, in a crude analysis, how the finite time required to clear the star forming region of \\hi\\ affects the number of ionising photons that can escape from the system, taking into account the rapid temporal decline of the ionising luminosity produced by a stellar population. Our approach is thus especially useful in providing estimates of escape fractions and star formation efficiencies for regions that may be below the resolution limit for numerical simulations of galaxies.  The structure of the paper is as follows. In Section 2 we set out the model for the growth of superbubbles and derive an expression for the porosity of the galaxy as a function of time, SFR, and ISM parameters, including an analysis of how the results are modified in disc systems, and establish the existence for a {\\it critical star formation threshold}. In Section 3 we assess the consequences of the model for the star formation efficiency and escape of ionising photons from galaxies. In Section 4 we apply the results of the foregoing sections to a variety of star forming regions, including Lyman break galaxies, starbursts and giant molecular clouds. Section 5 summarises our conclusions.  \\section {Porosity related to star formation:  an analytical approach}  Oey \\& Clarke (1997; hereafter OC97) estimate galactic porosity as the volume of superbubbles generated by OB associations relative to the simple geometric volume of the host galaxy.  Here, we examine the porosity in more detail and relate it specifically to the ISM thermal energy and mass.  \\subsection{The volume of hot gas generated by star formation: steady state}  We take the mechanical luminosity function (MLF) for the SN energies of the OB associations to be, \\begin{equation}\\label{MLF} \\phi(L) = A L^{-2} \\quad , \\end{equation} where $\\phi(L)$ is the fraction of clusters with mechanical luminosity in the range $L$ to $L+dL$.  The power-law slope of --2 is empirically well-determined from the \\hii\\ region luminosity function and stellar cluster mass function, as described above. Such a distribution implies that equal decades in cluster luminosity contribute equally to the total integrated mechanical luminosity, or, equivalently, SFR, of the galaxy.  Thus the total SFR is dominated neither by very populous nor very sparse clusters.  We furthermore assume that this luminosity function extends over a range of luminosities corresponding to a supernova membership number in the range $N_{\\rm min}$ ($\\geq 1$) to $N_{\\rm max}$.  Following OC97, the lifetime of all superbubbles is assumed to be roughly equal to the main-sequence lifetime of the lowest mass SN progenitor, $t_e \\sim 40$ Myr for Population I stars.  Therefore, if constant star-forming conditions are sustained for periods in excess of $t_e$, the differential superbubble size distribution attains a steady state, as derived by OC97: \\begin{equation}\\label{size_s} N(R) = A\\psi L_e^{-1} R_e^{-1}\\ \\Bigl(\\frac{R}{R_e}\\Bigr)^{-3}\\ \\Bigl[2(t_e + t_s) - \\frac{3}{4}\\ t_e\\ \\Bigl(\\frac{R}{R_e}\\Bigr)\\Bigr] \\ ,\\ R \\leq R_e \\end{equation} and \\begin{equation}\\label{size_g} N(R) = 5A\\psi\\ L_e^{-1} R_e^{-1}\\ \\Bigl(\\frac{R}{R_e}\\Bigr)^{-6}\\ \\Bigl[\\frac{t_e}{4} + t_s\\Bigr] \\ , \\ R > R_e \\end{equation} \\noindent where $\\psi$ is the creation rate of the superbubbles, $L_e$ is the luminosity of a bubble that comes into pressure equilibrium with the ambient medium after time $t_e$ and $R_e$ is the corresponding radius of such a bubble at that time.  It is assumed that the growth of bubbles with $L < L_e$ stall by pressure confinement at the point that they come into pressure balance with the ambient medium. After $t_e$, the SN energy stops, and the object is presumed to survive at constant radius for another increment of time $t_s$.  For this steady-state size distribution and constant MLF, the total volume of superbubbles depends only on $\\psi$, or equivalently SFR, and the interstellar conditions that determine $R_e$. The total volume of the superbubbles in a steady state is \\begin{equation} V_{\\rm tot,ss} = \\int_{\\rmin}^{\\rmax} \\frac{4}{3}\\pi R^3\\ N(R)\\ dR \\quad , \\end{equation}  \\noindent where $R_{\\rm min}$ is the stall radius of a bubble containing $N_{\\rm min}$ supernovae and $R_{\\rm max}$ is the size of a bubble containing $N_{\\rm max}$ supernovae at time $t_e$. Thus integrating equations~\\ref{size_s} and \\ref{size_g} gives a total volume, \\begin{equation} V_{\\rm tot,ss} \\simeq 3\\pi A\\psi L_e^{-1} R_e^3\\ (t_e + 2t_s) \\quad . \\end{equation} \\noindent Equation~\\ref{MLF} is a probability distribution, so its integral is unity, and therefore $A \\simeq L_{\\rm min}$, yielding,  \\begin{equation}\\label{Vtot} V_{\\rm tot,ss} \\simeq 3\\pi N_{\\rm tot} \\Bigl(\\frac{L_{\\rm min}}{L_e}\\Bigr)\\ R_e^3 \\quad , \\end{equation} \\noindent where $N_{\\rm tot}$ is the total number of superbubbles in the steady state.   The total number of supernovae contained in this population of bubbles is (from equation~\\ref{MLF})  \\begin{equation}\\label{Nsn} N_{\\rm sn} \\simeq  N_{\\rm tot} N_{\\rm min} {\\rm ln} \\Bigl(\\frac{N_{\\rm max}}{N_{\\rm min}}\\Bigr)\\ \\quad , \\end{equation}   \\noindent so that the {\\it mean volume of ISM cleared per supernova}, $V_{\\rm sn}$ is  \\begin{equation}\\label{Vsn} V_{\\rm sn} \\simeq  \\frac{3 \\pi}{{\\rm ln} \\Bigl(\\frac{N_{\\rm max}}{N_{\\rm min}}\\Bigr)} \\frac{R_e^3}{N_e} \\quad , \\end{equation}  \\noindent where we have used the fact that $ L \\propto N$, and we define $N_e$ as the number of SNe corresponding to mechanical luminosity $L_e$.  We will use these expressions for SN-cleared volume in deriving the interstellar porosity below.  Equation~\\ref{Vsn} shows that $R_e$ dominates $V_{\\rm sn}$.  Note that since $R_e$ is the radius at which a bubble containing $N_e$ supernovae comes into pressure balance with the ISM, one can roughly equate the thermal energy of the ISM contained within $R_e$ with the total energy input from $N_e$ supernovae:  \\begin{equation}\\label{therm} N_e E_{\\rm sn} \\simeq \\frac{4\\pi}{3} R_e^3 u \\quad , \\end{equation}  \\noindent where $E_{\\rm sn} \\sim 10^{51}$ erg is the SN energy and $u$ is the thermal energy density in the ISM\\footnote{Note that throughout this paper we use the term `thermal' energy to denote energy in random motions in the ISM, whether this is dominated by bulk cloud motions or by motions at a molecular (thermal) level.} . We thus deduce that $V_{\\rm sn}$ is within a factor of order unity of $V_{\\rm max}$:  \\begin{equation}\\label{vmax} V_{\\rm max} \\sim  {{E_{\\rm sn}}\\over{u }} \\quad . \\end{equation}  \\noindent which, as discussed in \\S 1, is the mean  volume per supernova for the adiabatic evolution of individual SNe.  The finding that $V_{\\rm sn} \\simeq V_{\\rm max}$ contrasts with the two scenarios considered by previous authors, namely, either distributed individual SNe, or else all SNe concentrated in a single bubble.  The difference may readily be traced to the fact that when one considers a realistic spectrum of cluster richness (i.e. a MLF) there is an important volumetric contribution from bubbles with $L \\simeq L_e$. Such bubbles, which stall after a time $t_e$, remain in the adiabatic expansion phase over their entire SN-producing lifetime and thus represent optimal coupling between the supernova energy and clearing of the ISM. Note also that $V_{\\rm sn}$ is insensitive to any upper cut-off in the richness of OB associations, provided that the MLF extends well beyond $L_e$, since the volumetric contribution of bubbles with $L\\gg L_e$ is small (equation~\\ref{size_g}).    \\subsection{Non-steady star formation}  The above analysis may readily be modified to model an episode of star formation that proceeds at constant rate over a timescale $t <t_e$.  Such a situation is only relevant to star forming systems that can switch on their star formation on timescales $\\ll t_e$ and thus applies to compact systems with short dynamical timescales, such as molecular clouds and globular clusters, as discussed in \\S 4.1.  At time $t (<t_e)$, the transition from the size distribution given by equation~\\ref{size_s} to that of equation~\\ref{size_g} occurs at radius $R_t$ instead of $R_e$, where $R_t$ is the radius of a bubble that just stalls at time $t$. Consequently, the total volume of superbubbles after time $t$ is given by:  \\begin{equation}\\label{vtt} V_{\\rm tot,ns} (t) = 3 \\pi N_{\\rm tot}(t) R_{\\rm min}^2 R_t \\quad . \\end{equation}  \\noindent $N_{\\rm tot}(t)$,  the total number of superbubbles produced in time $t$, is proportional to $t$, whereas one may readily show (OC97) that $R_t \\propto t$. Consequently, the volume of hot gas created varies {\\it quadratically} with time, so that one may write  \\begin{equation}\\label{vtt2} V_{\\rm tot,ns} (t) = V_{\\rm tot,ss} \\Bigl(\\frac{t}{t_e}\\Bigr)^2 \\quad . \\end{equation}    \\subsection {Effect of cooling and finite scale of ISM}  The above analysis is appropriate to an infinite ISM where the bubble evolution remains adiabatic until it comes into pressure equilibrium with the ISM.  We verify the approximate validity of the adiabatic assumption by comparison of  the bubble stall radius with the cooling radius given by Mac Low and McCray (1988).  The former may be written (e.g. see OC97) in the form: \\begin{equation} R_f =  300\\ {\\rm pc}\\ L_{38}^{1/2} P_{\\rm MW}^{-3/4} n_{\\rm MW}^{1/4} \\quad , \\end{equation} where $L_{38}$ is the luminosity normalised to $10^{38}$ erg s$^{-1}$ (equivalent to an OB association with $\\sim100$ SN progenitors), $P_{\\rm MW}$ and $n_{\\rm MW}$ are respectively the pressure and number density of the ISM normalised to their values in the Milky Way ($3 \\times 10^{-12}$ dyne cm$^{-2}$ and $0.5$ cm$^{-3}$). The corresponding expression for the cooling radius of freely expanding bubbles, using a cooling function appropriate to solar metallicity gas, is (Mac Low and McCray 1988): \\begin{equation} R_c \\simeq 540 \\ {\\rm pc}\\ L_{38}^{4/11} n_{\\rm MW}^{-7/11} \\quad . \\end{equation}  These expressions imply that even in the case of solar metallicity gas,  cooling is of marginal importance before the point is reached where the counter-pressure of the ISM is significant in slowing the expansion; for lower metallicity systems, cooling would be of still less importance. This hence justifies  our treatment of the evolution  prior to stalling as approximately adiabatic.  We now consider how the above analysis is modified when one takes into account the finite extent of the ISM. In general, bubbles evolve as described above, provided their sizes remain less than the density scale length of the ISM. Once they grow to larger radii, their evolution is modified by the ambient density gradient: in  particular,  a decreasing gradient causes the bubble expansion to accelerate due to the decreasing inertia of the newly swept up material. In disc-like density distributions, for example, a  number of authors have performed hydrodynamical simulations of bubbles that demonstrate that bubbles `break out' of the disc when they grow to a height between one and two disc scale heights (e.g., Mac Low and McCray 1988). At this point, the contents of the hot bubble interior are vented normal to the disc plane. Thereafter, the bubble evolution in the disc plane is no longer adiabatically driven, but evolves in a momentum-conserving fashion.  We now calculate how the volume of ISM occupied by bubbles is reduced if one takes this into account.  It is convenient to divide the bubble population into low luminosity objects having $ L< L_H$, which stall at sizes less than the disk scale height $H$, and higher luminosity objects that break out of the disc. >From integration of equation~\\ref{size_s} it can be seen that the total volume of bubbles contained in objects with radius less than $R$ is roughly proportional to $R$, and the total volume contained in bubbles that never break out of the disc is a factor $\\simeq H/R_e $ times the total volume of bubbles that would be created in an infinite medium for a given SFR.  We now estimate the total volume swept out by bubbles with $L > L_H$. Such bubbles evolve adiabatically prior to breakout and hence the kinetic energy of the bubbles walls is proportional to the number of supernovae that have gone off at that point. Since bubbles attain a fixed size scale on a timescale $t_H$ that scales as $L^{-1/3}$\\ (OC97), the  kinetic energy of bubbles at breakout scales as $L \\times t_H \\propto L^{2/3}$.  All bubbles with $L > L_H$ break out when the volume of ISM swept up is $\\sim H^3$, so that the mass of ISM swept up at breakout is independent of $L$. Hence the {\\it momentum} of bubbles at breakout scales simply as the square root of the energy, i.e. as $L^{1/3}$.  Thereafter, the bubbles evolve in an approximately momentum-conserving fashion and then stall when their expansion velocities become of order the thermal speed in the ISM.  Thus, it follows that the final volume of the bubble is proportional to the momentum at breakout, and hence also scales as $L^{1/3}$.  We can obtain the normalisation by noting that objects that just stall at size scale $H$ are by definition not going to undergo further momentum-conserving expansion, because their velocity has already declined to thermal values.  Thus we find that the final volume of a bubble of size $L > L_H$ is given by $H^3 (L/L_H)^{1/3}$ (see also Koo \\& McKee 1992), as compared with the final bubble volume in an infinite medium which can be written as $H^3 (L/L_H)^{3/2}$ (OC97). By integrating each of these expressions over the MLF (equation~\\ref{MLF}), we find that bubbles that have broken out contribute a total volume that is a factor $\\simeq H/R_{\\rm e}$ times the total volume filled in the case of an infinite medium. [Note that this analysis does not account for any continued driving by remaining SNe, which, following breakout from the disk, could contribute power following a momentum-conserving shell evolution (Steigman {\\etal} 1975). It can be shown, using the corresponding relations from OC97, that in this case the bubbles with $L>L_H$ then  contribute a volume fraction that exceeds the above estimate by only a logarithmic factor ($\\ln\\frac{R_{\\rm max}}{H}$). The effect of continued SN driving is thus not expected to be large, and we therefore do not consider it further for the purpose of the rough estimates considered here.]  Thus adding together the total contributions from bubbles that do and do not break out, and taking $R_{\\rm max} \\simeq R_e$, we find that the volume of bubbles produced is reduced by a factor \\begin{equation}\\label{fd} $$ f_d \\sim 2 H/R_e $$ \\end{equation} In forthcoming sections, we shall apply this correction factor where necessary in order to reduce the volume of bubbles produced per unit SFR in disc galaxies.  \\subsection{Calculation of galactic porosity}  In order to compute the porosity of the ISM, it is necessary to divide the volume of hot gas produced by  star formation by the effective volume of the star forming system, $V$. Thus from equations~\\ref{Vtot} and \\ref{therm}, the steady state porosity can  be written  \\begin{equation}\\label{qss} Q_{\\rm ss} \\simeq \\frac {f_d V_{\\rm tot,ss}}{V} \\simeq \\frac{9}{4}\\ \\frac{f_d N_{\\rm tot} N_{\\rm min} E_{\\rm sn}}{uV} \\quad , \\end{equation}  \\noindent where $f_d$ is the factor (equation~\\ref{fd}) that takes rough account of the reduction in galactic porosity in the case of disc systems. If the mean mass of stars produced per bubble is $m_*$, then $N_{tot}$ is related to the star formation rate by:  \\begin{equation}\\label{Ntot} N_{\\rm tot} = \\frac{\\SFR\\ t_e}{m_*} \\quad , \\end{equation}  \\noindent whilst the product $uV$ is, by definition, the total thermal energy contained in the ISM of the system, $E_{\\rm ISM}$:  \\begin{equation}\\label{EISM} uV = E_{\\rm ISM} = \\frac{1}{2} M_{\\rm ISM}\\ \\tilde v^2 \\quad , \\end{equation}   \\noindent where $M_{\\rm ISM}$ is the total mass in the ISM and $\\tilde v$ is the `thermal' velocity dispersion. Thus ~\\ref{qss} becomes  \\begin{equation}\\label{qss2} Q_{\\rm ss} \\simeq \\frac{9}{2}\\ \\frac{f_d N_{\\rm min} {\\rm SFR}\\ t_e E_{\\rm sn}}{m_* M_{\\rm ISM}\\ \\tilde v ^2} \\quad . \\end{equation}  For a Salpeter IMF and $N_{\\rm min} =1 $, the mean mass of stars per bubble can be written $m_* = 150\\ {\\rm ln}(N_{\\rm max})\\ M_\\odot$ and is thus weakly (logarithmically) sensitive to any upper cutoff in the MLF. Here we adopt $N_{\\rm max} \\sim 7000$, which corresponds to the largest OB associations in the Milky Way (McKee and Williams 1997), and which is, incidentally about twice  $N_e$ for  Milky Way ISM parameters (OC97). In this case the mean number of supernovae per bubble is $ \\simeq 9 $ and $m_* \\simeq 1350\\ M_\\odot$. (We note that if $N_{\\rm max}$ was an order of magnitude greater than this, $m_*$ would only increase by $25 \\%$). Taking $E_{\\rm sn} \\simeq 10^{51} $ ergs, we obtain:  \\begin{equation}\\label{qss3} Q_{\\rm ss} \\simeq \\frac { 7 f_d\\  {\\rm SFR_\\odot} }{M_{\\rm ISM,{10}}\\ \\tilde v_{10}^2} \\quad , \\end{equation}  \\noindent where $\\rm SFR_\\odot$ is the star formation rate in solar masses per year, $M_{\\rm ISM,{10}}$ is the mass of the ISM in units of $10^{10} M_\\odot$ and $\\tilde v_{10}$ is the thermal velocity of the ISM normalised to $10$ km s$^{-1}$. In a system where star formation has been ongoing for a time $t < t_e$, the porosity is given by (equation~\\ref{vtt2}):  \\begin{equation}\\label{qt} Q(t) = Q_{\\rm ss} \\Bigl(\\frac{t}{t_e}\\Bigr)^2 \\quad , \\end{equation}  Equation~\\ref{qss3} implies that there is a critical star formation rate, $\\SFR_{\\rm crit}$ such that the porosity of the ISM is unity, i.e.  \\begin{equation}\\label{SFRc} \\SFR_{\\rm crit} = 0.15 \\biggl(\\frac{ M_{\\rm ISM,{10}}\\ \\tilde v_{10}^2}{f_d}\\biggr)\\ M_\\odot {\\rm yr}^{-1} \\quad , \\end{equation}  We stress that {\\it $\\SFR_{\\rm crit}$ is the SFR such that the energy output  from SNe, over a timescale $t_e$, is comparable with the energy of the ISM contained in random motions.} The normalisation of equation~\\ref{SFRc}  thus depends only on the assumed IMF and the stellar astrophysics contained in the value of $t_e$ and the energy delivered per SN. (We note, however, that in reality equation~\\ref{SFRc} should be regarded as a very rough guide, since its derivation suffers from the obvious over-simplification that results from approximating the ISM of a galaxy as a smooth homogeneous entity characterised by a single set of physical parameters. In practice, we will find equation~\\ref{SFRc} useful below in dividing highly porous regimes from the marginal case and from situations where the porosity is very low).  If $\\SFR < \\SFR_{\\rm crit}$, then such a SFR can be sustained indefinitely, provided the gas supply is large; star formation can proceed at such a rate over timescales $\\gg t_e$, with the porosity attaining a steady state value of $Q_e < 1$.  If $\\SFR > \\SFR_{\\rm crit}$, then the system attains unit porosity after a time $t_Q$:  \\begin{equation}\\label{tq} t_Q = t_e \\biggl({\\SFR_{\\rm crit}\\over{\\SFR}}\\biggr)^{1/2} \\quad . \\end{equation} We discuss below the consequences of achieving unit porosity, but first note that the {\\it maximum} rate of star formation achievable in a star forming system is \\begin{equation}\\label{SFRd} \\SFR_{\\rm dyn} \\sim {{M_{\\rm ISM}}\\over{t_{\\rm dyn}}} \\quad , \\end{equation} where $t_{\\rm dyn}$ is the dynamical timescale of the star forming region.  \\section {Consequences for star formation efficiency and escape of ionising radiation}  \\subsection{Star formation efficiency}  We have shown that the porosity of a star forming system becomes of order unity at the point that the input of mechanical energy into the ISM (over time $t_e$, or the duration of the burst, whichever is the shorter) is comparable with the {\\it thermal energy content} of the ISM, where we take `thermal' to denote random ISM motions. The critical SFR that must be sustained over a timescale $t_e$ in order to attain unit porosity is given by $\\SFR_{\\rm crit}$ (equation~\\ref{SFRc}).  For a bound spheroidal system, the thermal energy content of the ISM is always of order its gravitational binding energy, whether the gravitational potential derives from the gas itself or is a background potential of dark matter and/or stars. Consequently, when the porosity attains a value $\\sim 1$, the energy input into the ISM is comparable with its gravitational binding energy. As a result, one would not expect spheroidal systems to be able to sustain star formation rates much in excess of $\\SFR_{\\rm crit}$ over timescales $> t_e$.  In the case of compact systems with dynamical timescale $<t_e$, it is however possible for the SFR to exceed this limit temporarily. If, in this case, we consider the star formation event to be self-terminated after time $t_Q$ (equation~\\ref{tq}), when the volume filling factor of superbubbles reaches unity, we can use equations (19) and (23) to derive the  fraction of gas converted into stars during the event as: \\begin{equation}\\label{eps1} \\epsilon = {{2 m_* \\tilde v^2}\\over{9 E_{\\rm sn}}} \\biggl({{\\SFR}\\over{\\SFR_{\\rm crit}}}\\biggr)^{1/2} \\quad . \\end{equation} so that \\begin{equation}\\label{eps} \\epsilon = 6 \\times 10^{-4} \\tilde v_{10}^2 \\biggl({{\\SFR}\\over{\\SFR_{\\rm crit}}}\\biggr)^{1/2} \\quad . \\end{equation} The maximum fraction of the ISM that can be turned into stars increases with the square root of the SFR, and is thus limited by the upper, dynamical, limit to the SFR implied by equation~\\ref{SFRd}, to the value:  \\begin{equation}\\label{epsmax1} \\epsilon_{\\rm max} =   \\biggl({{2 m_* \\tilde v^2}\\over{9 E_{\\rm sn}}}\\biggr)^{1/2} \\biggl({{t_{\\rm dyn}}\\over{t_e}}\\biggr)^{-1/2} \\quad . \\end{equation}  \\noindent from which  \\begin{equation}\\label{epsmax} \\epsilon_{\\rm max} = 0.02  \\tilde v_{10} \\biggl({{t_{\\rm dyn}}\\over{t_e}}\\biggr)^{-1/2} \\quad . \\end{equation}  In a disc system, by contrast, the thermal energy content of the ISM is  $\\ll$ its gravitational binding energy.  Consideration of hydrostatic equilibrium normal to the disc plane demonstrates that the ratio of energy in random motions to gravitational binding energy is of order  $(H/R)$  if the vertical gravity mainly derives from the disc's local self-gravity, or $(H/R)^2$ if it instead derives from the vertical component of the gravity of a central mass concentration. Thus, star formation rates in excess of $\\SFR_{\\rm crit}$ do not imply the wholesale unbinding of the ISM and may not be ruled out on these grounds.  Whether or not $SFR_{crit}$ represents a maximum to the star formation rate in discs systems, or else a point of transition to star formation in a highly porous state, depends on conditions in the cool gas component once $Q \\sim 1$ (i.e., equivalently, whether feedback operates positively or negatively on the cool gas). Several pieces of evidence suggest that star formation may well continue in this state. We will see below (Section 4.3) that some systems, notably Lyman Break Galaxies, apparently sustain star formation rates well in excess of $SFR_{\\rm crit}$ over prolonged periods ($10^8-10^9$ years; Shapley et al 2001).  A more local example is provided by the LMC.  Although the porosity of this system is around unity (Oey et al 2001), continuous vigorous star formation has been ongoing in the LMC disk for at least $10^9$ year, and possibly up to 15 Gyr (e.g., Smecker-Hane {\\etal}2002; Dolphin 2000).  Inspection of the LMC reveals how this situation is achieved: although the bulk of the system volume is filled with hot gas, chiefly in the halo, the bulk of the {\\it mass} is in the cool component in the disc plane, where a high star-formation rate is maintained.  This system shows little evidence for negative feedback effects on star formation, but is well-known to provide examples where star formation appears to be actively triggered in the cool gas constituting bubble walls (e.g., Yamaguchi {\\etal}2001; Oey \\& Smedley 1996; Oey \\& Massey 1995; Walborn \\& Parker 1992).  The specific example of the supergiant shell LMC-4, which is the largest and most studied case of LMC triggered star formation (e.g., Dopita et al. 1985; Braun et al. 1997; Olsen et al. 2001) clearly shows a massive ring of H I and star formation 1.4 kpc in diameter surrounding a large complex of young blue stars. However, the morphology is clearly ring-like, rather than shell-like, suggesting that mechanical feedback on these large scales does not remove the majority of shell gas from the low galactic latitudes, and does promote continued star formation. In what follows, therefore, we assume that in disc systems star formation can in principle continue in the cool component at rates in excess of $SFR_{\\rm crit}$.     We note that the characteristic gas exhaustion timescale for a system in a marginally porous state is given by: \\begin{equation}\\label{texh} t_{\\rm exh} = \\frac{M_{\\rm ISM}}{\\SFR_{\\rm crit}} = 7 \\times 10^{10} {{ f_d }\\over{\\tilde v_{10}^2}}\\ {\\rm years} \\quad , \\end{equation} where the value of $t_{\\rm exh}$ depends on the same assumptions about stellar astrophysics and IMF as detailed for $\\SFR_{\\rm crit}$ (equation~\\ref{SFRc}).  \\subsection{Implications for the escape of ionising radiation}  We have here presented a model of the ISM structured by SN explosions located in spatially distributed clusters, a model that has been successfully tested in the case of nearby galaxies. Our discussion of ionising photon escape from such a a medium is necessarily more speculative, pending detailed hydrodynamic and photoionisation calculations.  Here we  explore the consequences of such a model by the following crude parameterisation of ionising photon escape: we assume that when the porosity of the ISM is $< 1$, no ionising photons can escape and that when the porosity is high ($>1$) all ionising photons can escape.  It is easy to see why these assumptions are wrong in detail.  For example, pure photoionisation codes of disc galaxies, i.e. calculations that assume a smoothly stratified initially neutral medium, suggest escape fractions of a few per cent even in the absence of mechanical energy input from SNe. Likewise, it is well established that populous clusters can create local chimneys in the ISM, thereby launching galactic superwinds and it is reasonable to expect some photon leakage in this case (see, however, Tenorio-Tagle et al 1999; Dove, Shull and Ferrara 1999) even when the global star formation rate is $\\ll \\SFR_{\\rm crit}$. It is also unlikely that the escape fraction is as high as unity even when the ISM is highly porous. Although most of the volume of the ISM is cleared of neutral material in this case,  most of its {\\it mass} is contained in neutral bubble walls. We here assume that a variety of hydrodynamical instabilities break up the bubble walls once the bubbles start to overlap strongly, thus opening up lines of sight through which ionising photons can escape the disc. In order to escape the galaxy, however, such photons also have to propagate through low density material in the halo without encountering significant opacity from neutral hydrogen. Detailed hydrodynamic/radiative transfer calculations are required, which model the input of ionising photons and mechanical energy into the halo from spatially dispersed, non-coeval star formation events in the disc, in order to assess whether the halo can be maintained in a state of sufficient transparency.  Despite the above caveats, we argue that this simple prescription captures an important dependence of the escape fraction of ionising radiation on SFR. Systems maintaining a steady state star formation rate on timescales greater than $t_e$ can exist in two states: if $\\SFR<\\SFR_{\\rm crit}$ the escape fraction is low and star formation can in principle proceed at such a rate until all the gas is exhausted. On the other hand, disc systems which are re-supplied on a timescale less than $t_{\\rm exh}$ (equation~\\ref{texh}) may  sustain SFRs in excess of  $\\SFR_{\\rm crit}$, in which case the escape fraction would be high.  For star-forming systems in which the dynamical timescale is less than  $t_e$, the SFR can vary on a timescale less than $t_e$, offering the possibility that the SFR may temporarily exceed $\\SFR_{\\rm crit}$. During such a star formation episode, the porosity of the system rises, attaining unity at time $t_Q$. We have argued that in spheroidal systems, star formation is self-limited at this point. Ionising photons from the population created prior to this can then escape the immediate vicinity relatively easily; the above crude model posits escape with unit probability.  Given an IMF and relationships between ionising luminosity, mass and lifetime, one may readily calculate an upper limit to the number of ionising photons escaping the region. Specifically, if the number of ionising photons emitted by the population prior to time $t$ is ${\\cal{ N}}_{\\rm ion}(t)$, then this upper limit is given by: \\begin{equation}\\label{fesc} f_{\\rm esc} = 1 - {{{\\cal {N}}_{\\rm ion}(t_Q)}\\over{{\\cal {N}}_{\\rm ion}(\\infty)}} \\quad . \\end{equation} Note that, whereas escape fractions are conventionally defined in terms of rates, in the case of a finite episode, it makes sense to define the escape fraction in terms of {\\it numbers} of ionising photons.  \\begin{figure} \\includegraphics[width=3.5truein,height=3.5truein]{clarkeoeyf1.ps} \\caption{\\label{fesc} Escape fraction (defined by equation~\\ref{fesc} ) as a function of the time, $t_b$, at which the burst of star formation is terminated. The solid and dashed curves are for  Population III and   Population I stars, with a Salpeter IMF (extending up to $100 M_\\odot$) in both cases. } \\end{figure}  Figure 1 illustrates the dependence of $f_{\\rm esc}$, as defined above, on the duration of a star formation burst ($t_b$) under the simple assumption that ionising photons escape only if emitted at times $> t_b$. The two lines denote stars of Population I and III, with an assumed Salpeter IMF in both cases.  Note that star formation is assumed to continue at constant rate until $t_b$, unlike other recent studies of escape of ionising radiation from ageing populations (Dove, Shull and Ferrara 1999; Tenorio Tagle et al 1999) where all the star formation is concentrated in a burst at time $t=0$. Stellar data for Population I stars is taken from Maeder 1990 and D\\'\\i az-Miller et al 1998, whilst for Population III stars, models have kindly been supplied by Chris Tout in advance of publication. The curve for Population II stars would be almost indistinguishable from that for Population I, since the main dependence of ionising luminosity on metallicity occurs for stars of sufficiently low mass that they make a rather small contribution to the total ionising output of the cluster. The Population III curve is rather different, however, since lower mass stars are considerably hotter in this case and make a significantly larger contribution to the total ionising output than for the Population I case. Consequently,  a higher fraction of the ionising photons can escape at late times in the Population III case. The effect, however, is not enormous: for bursts terminated after $\\sim 20$ Myr the escape fraction (equation \\ref{fesc}) for Population III stars is $\\sim 25 \\%$ compared with a value that is roughly a factor of two lower for Population I stars. ",
        "conclusions": "We have developed a model where the ISM porosity, i.e., the fractional volume devoid of HI, is regulated by SN explosions. In this model, the SN progenitors are located in spatially distributed OB associations, membership numbers being dictated by the observed OB association luminosity function.  This model has previously been shown to provide a good fit to the observed size distribution of \\hi\\ holes in nearby galaxies.  We find that such a realistic distribution of SN progenitors ensures that the clearing of the ISM is more effective, per unit star formation rate, than in either the case of distributed single SNe or the case where all SNe are concentrated in a single region. This is because, given the slope of the OB association LF, the porosity in our model is dominated by bubbles that come into pressure equilibrium with the ISM on a timescale that is similar to $t_e$, where $t_e$ is the maximum lifetime of a SN progenitor and hence the timescale over which associations inject mechanical energy into the ISM. Such superbubbles evolve quasi-adiabatically and thus most of the mechanical energy of their SNe is deposited in the ISM. In consequence, for the population of bubbles as a whole, the average volume cleared per SN is within a factor of order unity of its theoretical maximum (equation~\\ref{vmax}), although this is somewhat reduced in the case of disc galaxies (equation~\\ref{fd}). This contrasts with the situation of single SNe, where cooling limits the volume cleared, and also single burst models, where clearing is limited by the breakout of the bubble from the galactic plane.  This model yields a simple relationship between the star formation rate and interstellar porosity. Following the arguments above, the critical star formation rate ($\\SFR_{\\rm crit}$) required to attain a porosity of order unity is just that at which the energy input from SNe, over a timescale $t_e\\sim 40$ Myr, is comparable with the thermal energy content of the ISM. For a given kinetic temperature of the ISM defined by the level of random motions, this implies a simple relationship between the $\\SFR_{\\rm crit}$ and the mass of the ISM (equation~\\ref{SFRc}), and hence a characteristic timescale for gas exhaustion, $t_{\\rm exh}$ (equation~\\ref{texh}). For a Salpeter IMF and ISM velocity dispersion of around $10$ km s$^{-1}$, this critical star formation timescale is roughly $10^{10}$ years.  If spheroidal galaxies form stars at $\\SFR_{\\rm crit}$, the energy input into the ISM over $t_e$ is comparable with the gravitational binding energy of the ISM and one might expect wholesale expulsion of the ISM to ensue. Disc systems, by contrast, can remain in a highly porous state and still retain their ISM.  Although in this case the volume fraction of \\hi\\ is then small, the mass fraction is still large, so we assume that star formation can proceed in the shredded walls of interacting superbubbles.  We furthermore suggest that the porosity of the ISM has an impact on the escape of ionising photons from galaxies, since the disintegration of overlapping bubbles can create channels in the ISM through which ionising photons can escape. This postulate must be assessed through detailed photoionisation calculations in a medium structured by supernova explosions whose progenitor OB associations are appropriately distributed in luminosity and space. We note that the recent analysis by Elmegreen et al (2001) of the morphology of the neutral ISM in the LMC favours the filament/bubble structure that is characteristic of a supernova-structured ISM.  If we tentatively accept this postulate, we would thus expect high escape fractions in galaxies whose star formation rates exceed $\\SFR_{\\rm crit}$, as would appear to be the case in Lyman Break Galaxies. Sustained star formation at such rates however requires that gas is replenished on a timescale less than $t_{\\rm exh}$ (see above). At recent cosmic epochs, the timescale for gaseous infall into galaxies is long, so that one would not expect that galaxies in general should display the high SFRs required to maintain a highly porous ISM. This conclusion is consistent both with measurements of the HI hole size distributions in nearby galaxies and with the low leakage of ultraviolet photons from the Milky Way based on H$\\alpha$ measurements of the Magellanic Stream (Bland-Hawthorn and Maloney 1999) During the assembly of galaxies at high redshift, however, much shorter infall timescales are expected,  and we suggest that it is the continued infall of material into Lyman Break Galaxies that allows them to sustain vigorous star formation levels with a high associated escape fraction. In systems where the infall timescale falls to values less than $t_e$, however, the situation reverses, since the continual replenishment of neutral material into  the star forming region can prevent the porosity from ever attaining high values. We suggest that this is why local starburst nuclei, being compact regions at the bottom of the galactic potential and thus subject to gaseous inflows on a short timescale, have a low escape fraction despite their high rates of star formation per unit gas mass.  We also consider compact systems, with dynamical timescales $< t_e$, in which the SFR may temporarily exceed $\\SFR_{\\rm crit}$ and we have estimated the maximum number of Lyman continuum photons that might be expected to escape these systems. The key factor here is the efficacy of feedback from stellar winds prior to the explosion of the first SN, which depends critically on metallicity. In Giant Molecular Clouds with near solar metallicity, we find that winds provide a very efficient means of cloud dispersal; we estimate that the maximum fraction of the cloud mass that can be converted into stars prior to their dispersal is a few per cent, close to the observationally inferred value. This suggests that stellar winds may be at least as important as photoionisation as a negative feedback mechanism in Giant Molecular Clouds. In Population III systems, by contrast, the mechanical feedback from stellar winds is negligible and clearing of the ISM is delayed until the explosion of the first SN.  Supernovae are not expected for progenitors more massive than $250 M_\\odot$ however, so that an extremely top heavy IMF might result in inefficient clearing and a low escape fraction of ionising radiation.  Finally, we present these  calculations as a first attempt to parameterise the relationship between escape fraction and SFR in star-forming systems and suggest the utility of such a prescription in semi-empirical models of galaxy formation and evolution. The critical star formation rate may also offer a useful means to parameterize mechanical and chemical feedback."
    },
    "0208/astro-ph0208168_arXiv.txt": {
        "abstract": "\\noindent We use a matched filter algorithm to find and study clusters in both N-body simulations artificially populated with galaxies and the 2MASS survey. In addition to numerous checks of the matched filter algorithm, we present results on the halo multiplicity function and the cluster number function. For a subset of our identified clusters we have information on X-ray temperatures and luminosities which we cross-correlate with optical richness and galaxy velocity dispersions.  With all quantities normalized by the spherical radius corresponding to a mass overdensity of $\\Delta_M=200$ or the equivalent galaxy number overdensity of $\\Delta_N=200\\Omega_M^{-1}\\simeq 666$, we find that the number of $L>L_*$ galaxies in a cluster of mass $M_{200}$ is \\begin{displaymath} \\log N_{*666} = (1.44\\pm0.17)+(1.10\\pm0.09)\\log(M_{200}h/10^{15}M_\\odot) \\end{displaymath} where the uncertainties are dominated by the scatter created by three choices for relating the observed quantities to the cluster mass. The region inside the virial radius has a K-band cluster mass-to-light ratio of $(M/L)_K=(116\\pm46)h$ which is essentially independent of cluster mass.  Integrating over all clusters more massive than $M_{200}=10^{14}\\,h^{-1}M_\\odot$, the virialized regions of clusters contain $\\simeq 7\\%$ of the local stellar luminosity, quite comparable to the mass fraction in such objects in currently popular $\\Lambda$CDM models. ",
        "introduction": "\\label{sec:intro}  Clusters of galaxies have become one of our most important cosmological probes because they are relatively easy to discover yet have physical properties and abundances that are very sensitive to our model for the formation and evolution of structure in the universe. Of particular interest to us, a cluster sample provides the means to study the high-mass end of the halo multiplicity function, the average number of galaxies in a halo of mass $M$, which provides important insight into the process of galaxy formation.  In this paper we have two objectives.  First, we will demonstrate the use of a matched filter approach to finding clusters in a redshift survey using both synthetic catalogs and a large sample of galaxies from the 2MASS survey (Skrutskie et al.~\\cite{Skrutskie97}, Jarrett et al.~\\cite{Jarrett00}). With the synthetic data we can test the algorithm and our ability to extract the input halo multiplicity function from the output cluster catalog. Second, we will determine the halo multiplicity function, $N(M)$, from the 2MASS survey.  This study is the first phase in a bootstrapping process -- based on a synthetic catalog known to have problems matching reality in detail we can calibrate an algorithm which when applied to the real data can supply the parameters for an improved model of the data. Then the improved model can be used to improve the algorithm and so on.  In \\S\\ref{sec:sims} we describe the synthetic and real 2MASS data. In \\S\\ref{sec:findcl} we describe our version of the matched filter algorithm.  In \\S\\ref{sec:tests} we test the algorithm on the synthetic catalog, focusing on our ability to determine the halo multiplicity function. In \\S\\ref{sec:clusters} we apply the algorithm to the 2MASS sample. Finally, in \\S\\ref{sec:conclusions} we discuss the steps which can improve both the synthetic catalog and the algorithm. ",
        "conclusions": "\\label{sec:conclusions}  We have applied the matched filter technique to both simulated galaxy catalogs and the 2MASS galaxy catalogue to search for clusters over approximately 90\\% of the sky to a redshift limit of $z\\simeq 0.05$. We have matched our 2MASS derived catalog to existing catalogs in an automated way using the NASA Extragalactic Database. Our algorithm appears to be both robust and efficient, returning quite complete samples of clusters out to distances where the typical cluster contains as few as 3 galaxies. The algorithm finds almost no `false groups', the main source of contamination being the inclusion of objects below the mass cut in the catalog.  The matched filter algorithm gives us a way of estimating cluster membership and hence cluster properties like the multiplicity function, the number function and the velocity dispersion in a new way. While our estimates for the velocity dispersion agree well with earlier work, we typically have far fewer galaxies per cluster than dedicated surveys and hence larger errors.  This is offset by the fact that we have a large number of clusters, over much of the sky, with which to search for correlations between cluster properties such as velocity dispersion and X-ray temperature. Where there is overlap we find quite good agreement with earlier work, though often with improved statistics.  Although it is necessary to be very careful about the definition of $N$ in order to compare with theory, we find that with sufficient care we have been able to estimate the cluster number function, $dn/dN$, over more than 2 orders of magnitude in $N$.  By using the velocity dispersion, X-ray luminosity or X-ray temperature as a surrogate for mass we are able to estimate the multiplicity function $N(M)$. Although there are serious issues in converting observables into cluster masses, all of our results are fairly consistent and suggest that $N(M)\\sim M$ or perhaps slightly steeper, in reasonable agreement with earlier estimates. With all quantities normalized by the spherical radius corresponding to a mass overdensity of $\\Delta_M=200$ or the equivalent galaxy number overdensity of $\\Delta_N=200\\Omega_M^{-1}=666$, we find that the number of $L>L_*$ galaxies in a cluster of mass $M_{200}$ is \\begin{equation} \\log N_{*666} = (1.44\\pm0.17)+(1.10\\pm0.09)\\log(M_{200}h/10^{15}M_\\odot). \\end{equation} The uncertainties in this relation are largely due to the choice made for relating the observed quantities to the cluster mass scale.  For a fixed mass scale the scatter resulting from the different observed correlations is considerably smaller. Correlations of $N$ with other cluster properties, X-ray luminosity, temperature or galaxy velocity dispersion, are given in \\S\\ref{sec:multiplicity}. The region has a K-band cluster mass-to-light ratio of $(M/L)_K=(116\\pm46)h$ which is essentially independent of cluster mass. The uncertainties are again dominated by the choice of the mass scale. This scaling is consistent with $N(M)\\sim M$, though if we take our best fit seriously and $N(M)$ is steeper than $M$ we expect $M/L$ would fall slowly with increasing mass. Integrating over all clusters more massive than $M_{200}=10^{14}\\,h^{-1}M_\\odot$, the virialized regions of clusters contain 7\\% of the local stellar luminosity, quite comparable to the (somewhat theory dependent) mass fraction in such objects in currently popular $\\Lambda$CDM models.  The cluster likelihoods tend to be larger in the real data than our mock catalogs. This difference could have several sources. It could be due to the concentration of redshift measurements in the real catalog towards groups and clusters rather than having the random distribution of the synthetic catalog.  It may also indicate that the cosmology adopted in the underlying simulation is incorrect (the number of clusters is very sensitive to $\\sigma_8$ for example), the normalization of the input $N(M)$ may be too low, our assumed $N(M)$ may have too many galaxies in low-mass halos compared to high-mass halos, or the luminosity function could vary systematically with the parent halo mass rather than being fixed (as we have assumed). Along these lines, both the expectation that light closely trace mass in $K$-band and that the luminosity function change with mass suggest that clusters contain a larger fraction of galaxies than we have assumed in our current generation of mock catalogs. These avenues will be explored in the future with larger simulations and better data.  We view this work as but the first step in an iterative sequence. Based on simulations which we know describe reality imperfectly, we have calibrated our cluster finding algorithm.  When applied to the real data this algorithm allows us to estimate correlations between different properties of a halo which can be used in the next stage to improve the simulated catalogs. As the 2MASS catalog becomes increasingly complete, and more redshifts become available, we will estimate the global relations between different halo parameters by providing them as priors to the fitting, and varying the parameters to optimize the global likelihood. These relations can then be used in the construction of the galaxy catalogs {}from improved simulations which will allow us to further optimize and understand the cluster finding algorithm."
    },
    "0208/astro-ph0208397_arXiv.txt": {
        "abstract": "This chapter is intended to provide a general presentation of the atomic and nuclear processes responsible for X-ray line and gamma-ray line emission in various astrophysical environments. I consider line production from hot plasmas, from accelerated particle interactions, from the decay of radioactive nuclei synthesized in stars and from positron annihilation. Spectroscopic properties of these emissions are discussed in the light of the detection capabilities of modern space instruments. ",
        "introduction": "X- and gamma-ray emission lines are valuable signatures of various high-energy processes at work in the universe, including heating of astrophysical gas to very high temperatures, particle acceleration and nucleosynthesis. Perhaps the most important realization of pioneering high-energy astronomy was that hot plasmas are found essentially everywhere in the universe. More and more detailed observations of thermal X-ray lines provide detailed and often unique information on a wide variety of astrophysical sites, including stellar environments, supernova explosions, accreting compact objects, interstellar and intergalactic media and active galactic nuclei.  Particle acceleration is believed to occur in most of these sites. Our understanding of acceleration mechanisms should greatly benefit from the detection of lines produced through atomic and nuclear interactions of energetic particles with ambient matter. The sun provides valuable examples of these nonthermal radiation processes. The study of nuclear de-excitation lines produced in solar flares has now become a proper domain of solar physics.  Gamma-ray lines are also emitted in the decay of radioactive nuclei. Measurements of gamma-ray activities from cosmic radionuclei testify to the ongoing synthesis of chemical elements and their isotopes in the Galaxy and beyond. These observations are now cornerstones for models of nucleosynthesis in novae, supernovae and stellar interiors.  Positron-electron annihilation radiation is relevant to almost all of high-energy astrophysics. Positrons can be produced by various pair-creation processes in relativistic plasmas, by accelerated particle interactions and by the $\\beta^+$-decay of radioisotopes. Observations of their annihilations offer a unique vision of high-energy accreting sources, nucleosynthesis sites and the global structure of the Milky Way.  The aim of this text is to present the physical processes responsible for these various line emissions. I also consider emission processes which have not been observed with assurance yet: X-ray line production from accelerated particle interactions and from the decay of cosmic radioactivities, and gamma-ray line emission from thermonuclear plasmas. However, given the advent of new X-ray and gamma-ray satellites with unprecedented sensitivities and spectral resolutions (see Barret and Kn\\\"odlseder, this volume), I am optimistic that at least some of these emissions could be detected in the near future.  The scope of this review is restricted to emission lines. In particular, I do not discuss X-ray absorption lines recently observed from a handful of active galactic nuclei, as well as cyclotron absorption features in intense magnetic fields, detected from several X-ray binaries. Nor do I treat the high-energy gamma-ray emission from pion decay.  The plan of this review is the following: in section \\S~2, I present the basic processes of X-ray line emission in both thermally ionized and photoionized plasmas; in \\S~3, I consider nonthermal X-ray line production in interactions of accelerated electrons and ions with ambient gas; in \\S~4, I deal with gamma-ray line emission from nuclear collisions and discuss thermonuclear reactions as well as accelerated ion interactions; in \\S~5, I consider both the gamma-ray lines and the X-ray lines emitted by the decay of radioactive nuclei synthesized in stars; finally, positron annihilation radiation is discussed in \\S~6. ",
        "conclusions": ""
    },
    "0208/astro-ph0208552_arXiv.txt": {
        "abstract": "We study the radial migration of dust particles in accreting protostellar disks analogous to the primordial solar nebula.  Our main objective is to determine the retention efficiency of dust particles which are the building blocks of the much larger planetesimals. This study takes account of the two dimensional (radial and normal) structure of the disk gas, including the effects of the variation in the gas velocity as a function of distance from the midplane. It is shown that the dust component of disks accretes slower than the gas component. At high altitude from the disk midplane (higher than a few disk scale heights), the gas rotates faster than particles because of the inward pressure gradient force, and its drag force causes particles to move outward in the radial direction.  Viscous torque induces the gas within a scale height from the disk midplane to flow outward, carrying small (size $\\la 100 \\ \\micron$ at $10 \\ {\\rm AU}$) particles with it. Only particles at intermediate altitude or with sufficiently large sizes ($\\ga 1 \\ {\\rm mm}$ at $10 \\ {\\rm AU}$) move inward.  When the particles' radial velocities are averaged over the entire vertical direction, particles have a net inward flux.  The magnitude of their radial motion depends on the particles' distance from the central star.  At large distances, particles migrate inward with a velocity much faster than the gas accretion velocity.  However, their inward velocity is reduced below that of the gas in the inner regions of the disk.  The rate of velocity decrease is a function of the particles' size.  While larger particles retain fast accretion velocity until they approach closer to the star, $10 \\ \\micron$ particles have slower velocity than the gas in the most part of the disk ($r \\la 100$ AU).  This differential migration of particles causes the size fractionation.  Dust disks composed mostly of small particles (size $\\la 10 \\ \\micron$) accrete slower than gas disks, resulting in the increase in the dust-gas ratio during the gas accretion phase.  If the gas disk has a steep radial density gradient or if dust particles sediment effectively to the disk midplane, the net vertically averaged flux of particles can be outward. In this case, the accretion of the dust component is prevented, leading to the formation of residual dust disks after their gas component is severely depleted. ",
        "introduction": "Planets form in circumstellar disks.  In the standard scenario, formation of earth-size planets or planetary cores occurs through coagulation of small dust particles (e.g., Weidenschilling \\& Cuzzi 1993).  Thus, the total amount, radial density distribution, and size distribution of dust particles in the disks are the important initial conditions of planet formation.  In many of the previous studies of planet formation, the dust-gas ratio of circumstellar disks is assumed to be similar to the solar value ($\\sim 10^{-2}$), and dust particles are considered to be well mixed with gas, i.e., the dust-gas ratio is constant throughout the disk (Hayashi, Nakazawa, \\& Nakagawa 1985).  However, the motion of dust particles is different from that of gas. Dust particles have a radial motion which is induced by the gas drag force.  Adachi, Hayashi, \\& Nakazawa (1976) and Weidenschilling (1977) studied the motion of particles in gas disks.  Due to radially outward pressure gradient force, the rotation velocity of the gas is generally slower than particles on nearly Keplerian circular orbits. Consequently, the gas drag force takes angular momentum from particles, resulting in their inward migration (Whipple 1972).  At a few AU from the central star, the orbital decay time is estimated to be $\\sim 10^6 \\ {\\rm yr}$ for $100 \\ \\micron$ particles and $10^4 \\ {\\rm yr}$ for $1 \\ {\\rm cm}$ particles. These times are much shorter than the life time of gas disks ($\\sim 10^7 \\ {\\rm yr}$).  Thus, dust component may evolve much faster than the gas in protostellar disks, in which case, the dust-gas ratio would decline.  It appears that the initial conditions of planet formation need not to be protostellar disks in which dust particles are well mixed with the gas and the dust-gas ratio is constant.  In this paper, we start a series of studies to determine the initial distribution of dust particles in the context of planet formation. Here, we study the radial migration of dust particles in gas disks.  We focus our attention on particles smaller than $\\sim 1 \\ {\\rm cm}$ in order to consider evolution of dust disks before the initiation of planetesimal formation.  The migration of these small particles establishes the initial density distribution of the dust component and sets the stage for planetesimal formation. In this phase of disk evolution, the gas component is considered to be turbulent and is accreting onto the star.  The discussion in the above paragraph is based on the study of particle migration under the assumption that particles have completely sedimented on the midplane of laminar disks. However, in turbulent disks, the sedimentation of particles is prevented, because particles are stirred up to high altitude from the midplane by turbulent gas motion.  In such disks, particles are distributed in the vertical direction, and the radial motion of particles depends on the distance from the midplane.  Thus, there is a need for studying the radial migration of particles which reside above the midplane.  There are two important factors causing the vertical variation in the migration velocity of particles.  The first is the variation of rotation velocity of the gas in the vertical direction.  As mentioned above, the gas rotation differs from the Keplerian rotation because of the gas pressure gradient.  As the gas density decreases with the distance from the midplane, the radial pressure gradient varies and under some circumstances even changes its sign.  While at the midplane the pressure gradient force is outward, it is inward near the disk surface because the disk thickness increases with the radius.  The gas drag force on particles also varies with the height, resulting in a variation of particle migration velocities.  In \\S \\ref{sec:vrd}, we will see that particles at high altitude, where the pressure gradient force is inward, flow outward.  The second factor is the variation in the radial gas flow. Small particles ($\\la 100 \\ \\micron$) are well coupled to the gas, so that they migrate with the gas flow, as discussed in \\S \\ref{sec:vrd} below.  That is if the gas is accreted onto the star, the small particles would also be accreted, while if the gas flows outward, particle flow would also be outward.  In accreting protostellar disks, the gas does not always flow radially inward.  Radial velocity of the gas varies with the distance from the midplane.  Viscous stress can cause gas outflow near the midplane (Urpin 1984; Kley \\& Lin 1992). R\\'o\\.zyczka, Bodenheimer, \\& Bell (1994) showed that the outflow occurs if the radial pressure gradient at the midplane is steep enough.  The density (or pressure) gradient at the midplane is steeper than the average spatial (or surface) density gradient, e.g.,  if the surface density varies as $\\Sigma_g \\propto r^{-1}$ and the disk thickness varies as $h_g \\propto r$, the midplane density has a steeper variation corresponding to $\\rho_g \\propto r^{-2}$. Viscous diffusion at the midplane causes the outflow of the gas to reduce the steep density gradient. On the other hand, at high altitude from the midplane, the density gradient is shallow and the viscous torque causes the usual inward flow. Thus, in accretion disks, the outward flow at the midplane is sandwiched by the inflow at the disk surfaces (see \\S\\ref{sec:vrg} for details). Because small particles are carried by the gas flow, they are expected to flow outward near the midplane, and inward at higher altitude.  Thus, if most of dust particles concentrate in the region where they migrate outward, i.e., near the disk surface or around the midplane, the dust component of the disk would accumulate in total mass and expand in size. In reality, the vertical spatial distribution of particles is determined by the degree of sedimentation, and their size distribution is regulated by the coagulation, condensation, and sublimation processes. In this paper, we calculate the vertical distribution of particles in order to identify the mass flux of the dust particles as functions of their size and their distance from the midplane and their host stars.  The results of the present calculation will be used to study the long term dynamical evolution of the dust component of the disk. For the present task, we do not consider the effects of gas depletion, particles' size evolution or their feedback influence on the flow velocity of the gas. These effects will be considered in a future investigation.  There are several works on the particle migration in turbulent disks. Stepinski \\& Valageas (1997) derived the vertically averaged velocity of migration for two limiting cases. One model is constructed for small particles which are well mixed with the gas and the other is for large particles which are totally sedimented to the midplane. The velocity for intermediate sized particles is calculated by the interpolation of these limiting cases.  Supulver \\& Lin (2000) investigated the evolution of particle orbits in turbulent disks numerically.  This paper gives analytical expression of the particle velocity for any size of particles, which are estimated from interpolation by Stepinski \\& Valageas (1997).  Several studies showed that vortices or turbulent eddies can trap particles within some size range (Barge \\& Sommeria 1995; Cuzzi, Dobrovolskis, \\& Hogan 1996; Tanga et al. 1996; Klahr \\& Henning 1997; Cuzzi et al. 2001, see however Hodgson \\& Brandenburg 1998).  Particle trapping by eddies may be important for local collection of size-sorted particles.  In this paper, turbulent eddies are treated just as a source of the particle diffusion and of the gas viscosity.  The effect of particle trapping on their coagulation is not discussed here.  The plan of the paper is as follows.  In \\S 2, the vertical variation of gas flow is derived.  In \\S 3, we describe how the radial velocity of dust particles varies in the vertical direction, and then calculate the net radial migration velocity.  In \\S 4, we discuss the steady distribution of particles assuming no particle growth. We show that the size fractionation of particles occurs as a result of radial migration, and that the distribution of particles differs from that of the gas. ",
        "conclusions": "\\subsection{Steady Density Distribution of Dust Particles \\label{sec:steady_dist}}  Because the inward velocity of particles depends on their size, particles of different sizes accumulate at different locations.  In this subsection, we discuss the density distribution of dust particles as a consequence of their radial flow and show how their size fractionation may occur.  We adopt the models in which the net velocity of particles is always inward (${\\rm Sc}=1$ and $p_s \\ge -1.0$).  We assume that the distribution of particles approaches a steady state after a brief stage of initial evolution.  When a steady state is achieved, the mass flux of dust particles becomes constant in the radial direction.  We calculate the mass flux of particles as \\begin{equation} \\frac{d \\dot{M_d}}{ds} = 2 \\pi r \\langle v_{r,d} \\rangle \\frac{d \\Sigma_d}{ds} \\ , \\label{eq:mflux} \\end{equation} where $\\dot{M_d}$ and $\\Sigma_d$ are the mass flux and the surface density, respectively, of particles smaller than size $s$. For simplicity, we neglect three physical processes.  First, the evolution of the particle size through the coagulation, collisional destruction, condensation, and sublimation is neglected, i.e., we assume the mass flux $d \\dot{M_d}/ds$ for each size range is constant in the radial direction.  Second, in turbulent disks, the particles' mass flux comes not only from the mean flow with an average velocity $\\langle v_{r,d} \\rangle$ but also from the turbulent diffusion of particles, which appears as $\\mbox{\\boldmath $j$}$ in equation (\\ref{eq:dustcont}) for example. We neglect the mass flux from the turbulent diffusion. Third, we assume the structure of the gas component of the disk is entirely determined by the gas itself.  We neglect the feedback drag induced by the particles on the gas.  In the limit that the spatial density of the dust component becomes comparable to that of the gas near the midplane, this effect would speed up the azimuthal velocity of the gas to the Keplerian value and quench the radial migration of the dust (Cuzzi et al. 1993).  Such a dust concentration requires substantial sedimentation of relatively large particles.  Contributions from all of these factors will be investigated in future grain-evolution calculations.  Here, we adopt the simplest assumptions to focus on showing the size fractionation of particles.  The mass flux of particles in all size range is \\begin{equation} \\dot{M}_{d,{\\rm all}} = \\int_{s_{\\rm min}}^{s_{\\rm max}} \\frac{d \\dot{M_d}}{ds} ds \\ , \\end{equation} where $s_{\\rm min}$ and $s_{\\rm max}$ are the minimum and maximum sizes of particles, respectively.  The surface density in the steady state is calculated from equation (\\ref{eq:mflux}) with given $\\dot{M_d}$.  Figure \\ref{fig:surden}$a$ shows the surface density of the dust component composed of single size particles.  Particles of different sizes have different density profiles.  If the dust component is composed of relatively large particles, it would be concentrate at the inner region of the disk.  Thus, as particles grow in size, their surface density distribution becomes more centrally concentrated.  Figure \\ref{fig:surden}$b$ shows the surface density, assuming the size distribution of particles is a power law with index $-3.5$, $s_{\\rm min}=0.1 \\ \\micron$, and $s_{\\rm max}=10 \\ {\\rm cm}$.  The surface density distribution of the dust particles is different from that of the gas.  The power law index of the dust distribution $\\Sigma_d$ is about $-1.5$ for a gas disk with index $p_s=-1.0$ ($-1.2$ for a gas disk with $p_s=-0.5$).  The implied power law index, $-1.5$, is similar to the value anticipated from the present mass distribution of planets in the solar system (Hayashi et al. 1985). However, note that the density distributions in Figure \\ref{fig:surden} are derived assuming no size evolution of particles, and no turbulent diffusion in the radial direction.  The growth of particles during the radial flow adds a source term in the equation of continuity.  The evolution of the dust density profile should be investigated further by taking particle growth and turbulent diffusion into account.  \\subsection{Evolution of the Dust-gas Ratio}  The accretion velocity of dust particles is different from that of the gas.  This difference causes evolution of the dust-gas ratio.  In the standard model, $10 \\ \\micron$ particles between $10$ and $100$ AU have an inflow velocity which is about half of the gas velocity.  Thus, if the mass of the dust disk is dominated by $10 \\ \\micron$ particles (for example, if the particles have size distribution $n \\propto s^{-3.5}$ with maximum size $10 \\ \\micron$), the dust-gas ratio would increase through the gas accretion.  For example, if the initial mass of the gas disk is $0.11 M_{\\sun}$ and it reduces to $0.01 M_{\\sun}$ after viscous accretion, the dust-gas ratio would increase to be 6 times of its initial value.  The increase in the dust-gas ratio speeds up the formation of planetesimals through mutual collisions.  Their enhanced abundance may also cause the dust component to become unstable and promote the planetesimal formation through the gravitational instability (Goldreich \\& Ward 1973; Sekiya 1998).  If the particle growth proceeds to make $1$ mm to $1$ cm particles during the gas accretion phase, such large particles migrate inward rapidly and accumulate in the inner part of the disk (see Fig \\ref{fig:surden}$a$).  This size fractionation causes an increase in the dust-gas ratio at the inner disk, while this process may decrease the dust-gas ratio at the outer disk, resulting in a dust disk concentrated to the inner part of the gas disk.  \\subsection{Summary}  The radial migration of dust particles in accretion disks is studied. Our results are as follows.  1. Dust particles move radially both inward and outward by the gas drag force.  Particles at high altitude ($|z| \\ga 2 h_g$) move outward because they rotate slower than the gas whose pressure gradient force is inward.  Small particles ($s \\la 100 \\ \\micron$ at $10 \\ {\\rm AU}$) near the midplane ($|z| \\la h_g$) are advected by the gas outflow.  On the other hand, particles at intermediate altitude and large particles ($s \\ga 1 \\ {\\rm mm}$ at $10 \\ {\\rm AU}$) move inward.  2. The net radial velocity, averaged in the vertical direction, is usually inward, provided that the radial gradient of the gas surface density is not too steep ($p_s \\ga -1.3$).  Sedimentation removes outflowing particles from high altitudes.  Small particles, which can be advected by the outflowing gas around the midplane, do not concentrate at the midplane.  In the inner part of the gas disk ($r \\la 100$ AU for $10 \\ \\micron$ particles), the inflow velocity of particles is smaller than the gas accretion velocity, resulting in an increase in the dust-gas ratio.  3. The particle sedimentation would be efficient if dust-gas coupling is relatively weak (${\\rm Sc} > 1.0$). If the sedimentation is so efficient (${\\rm Sc} \\ga 10$), the number of outflowing particles around the midplane would be large, and the direction of the net radial velocity of particles would change to outward at some distances from the star.  Accumulation of particles at such locations serves to increase the local dust-gas ratio.  4. The inflow velocity of particles depends on the particle size. Therefore, the inflow causes the size fractionation of particles. Larger particles accumulate at distances closer to the star."
    },
    "0208/astro-ph0208078_arXiv.txt": {
        "abstract": "{The adiabatic index $\\Gamma_1$ for perturbations of dense matter is studied under various physical conditions which can prevail in neutron star cores. The dependence of $\\Gamma_1$ on the composition of matter (in particular, on the presence of hyperons), on the stellar pulsation amplitude, and on the baryon superfluidity is analyzed. Timescales of damping of stellar pulsations are estimated at different compositions, temperatures, and pulsation amplitudes. Damping of pulsations by bulk viscosity in the neutron-star cores can prevent the stars to pulsate with relative amplitudes $ \\ga (1-15)\\%$ (depending on the composition of matter). ",
        "introduction": "The adiabatic index $\\Gamma_1= (n_{\\rm b}/P)({\\rm d}P/{\\rm d}n_{\\rm b})$ for density perturbations determines the changes of pressure $P$ associated with variations of the local baryon density $n_{\\rm b}$ (e.g., Shapiro \\& Teukolsky 1983). It enters the equations governing small-amplitude neutron-star pulsations (Thorne \\& Campolattaro 1967, Thorne 1968, 1969) as well as the criteria of stability of cold relativistic stars (Meltzer \\& Thorne 1966, Chanmugan \\& Gabriel 1971, Chanmugan 1977, Gourgoulhon et al. 1995). For the time-dependent perturbations, like neutron-star pulsations, $\\Gamma_1$ has to be calculated taking into account the slowness of various equilibration channels in dense matter. The same factors strongly affect viscous damping of neutron-star pulsations.  In the present paper, we study three main factors which regulate $\\Gamma_1$ and the viscous damping of pulsations: composition of dense matter, pulsation amplitude, and superfluidity of baryons. Equilibration processes for various compositions of neutron-star cores are studied in Sect. 2. In Sect. 3, we calculate the adiabatic index under various conditions of  density, composition, and temperature. We separately study two different regimes: first, the regime with the perturbations of chemical potentials of particles much smaller than $T$, and second, the regime with the perturbations much larger than $T$ (we use the units in which the Boltzmann constant $k_{\\rm B}=1$); we also consider the effect of baryon superfluidity. In Sect.\\ 4 we estimate the timescales of the viscous damping of density pulsations under various physical conditions which can be realized in the neutron-star cores. Finally, in Sect.\\ 5 we summarize our results and briefly discuss the problems which remain to be solved. ",
        "conclusions": "We have outlined the most important factors which affect the adiabatic index $\\Gamma_1$ for density perturbations and the viscous damping of pulsations in the neutron-star cores. First, non-leptonic strangeness-changing processes with hyperons may be rapid enough to establish partial hyperonic equilibrium over pulsation periods and decrease $\\Gamma_1$ slightly below the ``frozen'' adiabatic index, $\\Gamma_{\\rm FR}$, in hyperonic matter. In addition, they induce a rapid viscous damping even in the subthermal pulsation regime ($\\delta \\mu_j \\la T$). Second, suprathermal but still linear perturbations ($T \\ll \\delta\\mu_j \\ll \\mu_j$)   may produce a very rapid equilibration in various relaxation channels which reduces $\\Gamma_1$ to the fully equilibrium value $\\Gamma_{\\rm EQ}$, enhances the bulk viscosity and damps the pulsations quickly to lower amplitudes. Third, superfluidity of baryons in the subthermal regime increases relaxation times in various channels, brings $\\Gamma_1$ closer to $\\Gamma_{\\rm FR}$, and reduces the viscous dissipation.  Therefore, a proper calculation of neutron star pulsations and their dynamical evolution represents a complicated problem. Much work is required to study this problem in full detail. Our order-of-magnitude estimates have to be replaced with accurate numerical solutions of the equations of stellar pulsations taking into account proper boundary conditions and joint effect of various factors in all neutron-star layers, from the surface to the center. Our assumption of one typical pulsation amplitude of chemical potentials, $\\delta \\mu$, in Eq.\\ (\\ref{newT}) is an oversimplification. In reality, one has to deal with the number of {\\it density dependent} amplitudes $\\delta \\mu_j$ for different particle species $j$. The relaxation in different equilibration channels and the viscous damping of pulsations can be strongly nonuniform; one cannot exclude the existence of thin layers in the neutron star cores, where the damping is exceptionally strong. For instance, they may be the layers where new hyperons appear, with sufficiently small chemical potentials $\\mu_j$ just beyond their appearance threshold. Neutron star pulsations in these layers may be suprathermal, producing enhanced damping. The problem of suprathermal pulsation regime, without and with superfluidity of baryons, is of special importance. This regime will be accompanied by huge energy release which will heat the star."
    },
    "0208/astro-ph0208287_arXiv.txt": {
        "abstract": "We have discovered an Ofpe/WN9 (WN11 following Smith et al.) star in the Sculptor spiral galaxy NGC~300, the first object of this class found outside the Local Group, during a recent spectroscopic survey of blue supergiant stars obtained at the ESO VLT. The light curve over a five-month period in late 1999 displays a variability at the 0.1 mag level. The intermediate resolution spectra (3800-7200~\\AA) show a very close resemblance to the Galactic LBV AG Car during minimum.  We have performed a detailed non-LTE analysis of the stellar spectrum, and have derived a chemical abundance pattern which includes H, He, C, N, O, Al, Si and Fe, in addition to the stellar and wind parameters. The derived stellar properties and the He and N surface enrichments are consistent with those of other Local Group WN11 stars in the literature, suggesting a similar quiescent or post-LBV evolutionary status. ",
        "introduction": "With the new telescopes of the 8-10 meter class stellar astronomy is branching out beyond the Local Group. Ideal targets for our understanding of young stellar populations in distant galaxies are hot massive stars. These objects have strong stellar winds producing broad and easily detectable spectral features distributed over the whole wavelength range from the UV to the IR, and providing unique information on chemical composition, galactic evolution and extragalactic distances. With these perspectives in mind we have recently begun a systematic spectroscopic study of luminous blue stars in galaxies beyond the Local Group, and presented spectral classification and first quantitative results for A supergiants in NGC~3621 (6.7~Mpc, \\citealt{bresolin01}) and NGC~300 (2.0~Mpc, \\citealt{bresolin02}). Here we report on the discovery and detailed quantitative analysis of the first example of an Ofpe/WN9 star outside of the Local Group. We will present a detailed chemical abundance pattern -- the first in a galaxy beyond the Local Group -- together with stellar parameters and a determination of the stellar wind properties.  The Ofpe/WN9 class was introduced to include objects which show in their spectra high excitation emission lines from He\\II\\/ and N\\III, typical of Of stars, together with low excitation lines from He\\I\\/ and N\\II, seen in late WN stars (\\citealt{walborn82}, \\citealt{bohannan89}). Objects of this class have so far been identified in the Galaxy (possibly several stars in the Galactic center: \\citealt{allen90}, \\citealt{najarro97b}, \\citealt{figer99}), the LMC (ten stars: \\citealt{bohannan89}), M33 (seven stars: \\citealt{massey96}, \\citealt{crowther97b}) and M31 (one star: \\citealt{massey98}).  The importance of Ofpe/WN9 stars as objects in a transitional stage of evolution between O and W-R stars has been recognized in the last decade, and a connection to the LBV class has been suggested, Ofpe/WN9 stars being observed during a quiescent or post-LBV phase (\\citealt{crowther97}, \\citealt{pasquali97}). Indeed, in at least a couple of instances Ofpe/WN9 stars have been observed to turn into LBVs (R127: \\citealt{stahl83}; HDE 269582: \\citealt{bohannan89b}).  The LBV AG Car is also known to show an Ofpe/WN9-like spectrum during its hot phase at visual minimum (\\citealt{stahl86}).  The discovery of ejected circumstellar nebulae, in some cases measured in a state of expansion, associated with some of the LMC Ofpe/WN9 stars by \\citet{nota96} and \\citet{pasquali99} brings forward strong evidence for the occurence of violent episodes of mass loss in these stars, similar to the shell-producing, eruptive outbursts of LBVs.   \\citet{smith94} revised and extended the classification of late WN stars to include lower excitation objects.  In their scheme, stars showing spectra like AG Car at minimum, where no N\\III\\/ is detected, are reclassified as WN11. Given the spectral resemblance of the NGC 300 star we analyze here to AG Car at visual minimum (see \\S~4) we will adopt the WN11 classification for the remainder of this Letter. We describe the observational data in \\S~2, and the photometry in \\S~3. In \\S~4 we present our VLT spectra, together with the stellar and wind properties derived from a quantitative spectral analysis. ",
        "conclusions": ""
    },
    "0208/astro-ph0208414_arXiv.txt": {
        "abstract": "Various instabilities have been proposed as candidates to prompt the condensation of giant, star-forming cloud complexes from the diffuse interstellar medium. Here, we use three-dimensional ideal MHD simulations to investigate nonlinear development of the Parker, magneto-Jeans (MJI),  and swing mechanisms in galactic disk models. The disk models are local, isothermal, and begin from a vertically-stratified magnetohydrostatic equilibrium state with both gaseous and stellar gravity. We allow for a range of surface densities and rotational shear profiles, as well as unmagnetized control models. We first construct axisymmetric equilibria and examine their stability. Finite disk thickness reduces the critical Toomre stability parameter below unity; we find $Q_c\\sim0.75$, 0.72, and 0.57 for zero, sub-equipartition, and equipartition magnetic field cases, respectively. We then pursue fully three-dimensional models. In non-self-gravitating cases, the peak mid-disk density enhancement from the ``pure'' Parker instability is less a factor of two. The dominant growing modes have radial wavelengths $\\lambda_x$ comparable to the disk scale height $H$, much shorter than the azimuthal wavelength ($\\lambda_y\\sim10-20H$). Shearing disks, being more favorable to midplane-symmetric modes, have somewhat different late-time magnetic field profiles from nonshearing disks, but otherwise saturated states are similar. Late-time velocity fluctuations at 10\\% of the sound speed persist, but no characteristic structural signatures of Parker modes remain in the new quasi-static equilibria. In self-gravitating cases, the development of density structure is qualitatively similar to our previous results from thin-disk simulations. The Parker instability, although it may help seed structure or tip the balance under marginal conditions, appears to play a secondary role -- not affecting, for example, the sizes or spacings of the bound structures that form. In shearing disks with $Q$ less than a threshold level $\\approx 1$, swing amplification can produce bound clouds of a few times the local Jeans mass. The most powerful cloud-condensing mechanism, requiring low-shear conditions as occur in spiral arms or galactic centers, appears to be the MJI. In thick disks, the MJI occurs for $\\lambda_y\\simgt 2\\pi H$. Our simulations show that condensations of a local Jeans mass ($\\simlt3\\times 10^7\\,\\Msun$) grow very rapidly, supporting the idea that MJI is at least partly responsible for the formation of bound cloud complexes in spiral galaxies. ",
        "introduction": "Nearly half of the interstellar medium (ISM) in the inner part of the Milky Way is estimated to be in the molecular component (e.g., \\citealt{dam93}), with the largest portion of the molecular ISM in giant molecular clouds or cloud complexes (GMCs) of masses $\\sim10^4-10^6\\,\\Msun$ \\citep{dam87,sol87}, where most of star formation occurs (e.g., \\citealt{wil97}). GMCs are generally turbulent, self-gravitating, and magnetized (e.g., \\citealt{bli93,cru99}). In external galaxies, they often appear in clusters, forming giant molecular associations (GMAs) (e.g., \\citealt{gra87, vog88, ran90, ran93, sak99}), or within the boundaries of \\ion{H}{1} superclouds (e.g., \\citealt{elm95}). These giant clouds in spiral galaxies are mostly associated with spiral arms, but they sometimes appear in the interarm regions (e.g., \\citealt{sol85,ran93,kenney97,hey98}). By regulating the rate and mode of star formation in disk galaxies, GMCs play a fundamental role in controlling galactic evolution and may be a key to understanding the nature of the Hubble sequence. But how GMCs form is still not clearly understood.  Traditionally, proposed mechanisms for giant cloud formation fall into two categories: stochastic coagulation of smaller clouds (following \\citealt{oor54}; e.g., \\citealt{kwa79}) or collective effects involving instabilities (see e.g., \\citealt{elm96}). Existing work suggests that collisional agglomeration may proceed too slowly (e.g., \\citealt{sco79,elm90,elm95}). Collisions in most cases result in disruption rather than merger (e.g., \\citealt{lat85,kim99,kle01}); the available mass in smaller clouds is not enough to form GMCs through collisional build-up \\citep{hey98,bli99}; collisional agglomeration would lead to magnetically supercritical clouds and active star formation before GMC-scale masses are reached \\citep{ost99,ost01}; GMC chemistry may be inconsistent with an extended build-up phase (e.g., \\citealt{vandis93}). All of these and other difficulties point to the need for collective effects \\citep{bli80}. This second means of building GMCs and GMAs involves large-scale dynamical instabilities that coherently amass material over scales large compared to the sizes of diffuse atomic clouds (see e.g., \\citealt{elm95}).  The small-scale structure of the medium plays a secondary role, with the modal gathering of material operating collectively on the whole distribution.  Coherently-growing unstable modes that produce condensations depend on various effects, including self-gravity, the Coriolis force, sheared azimuthal motion, magnetic tension, and magnetic buoyancy, all of which have been shown to interact in unstable fashion.\\footnote{Thermal instabilities involving changes in the microphysical kinetic temperature are important in forming cold clouds on small scales (\\citealt{fie65}; see also e.g., \\citealt{hen99} and \\citealt{bur00} for recent simulations); if dissipation of kinetic energy in collisions of cloudlets acts effectively like microscopic cooling for the turbulent ``cloud fluid'', then analogous mechanisms may also be important on large scales (e.g., \\citealt{elm89, elm91}).}  The two most well-known condensation instabilities are the Parker instability \\citep{par66}, in which buoyancy causes dominantly in-plane magnetic fields to buckle, with matter collecting in valleys; and the swing amplifier, in which epicyclic motion conspires with galactic shear to enable self-gravitating enhancement of overdense regions \\citep{gol65b,too81}.  Another condensation instability that has been shown to be efficient in low-shear, magnetized regions is the magneto-Jeans instability (MJI), in which magnetic tension of in-plane fields counters the stabilizing effect of the Coriolis force, allowing self-gravitating contraction of overdense regions \\citep{elm87,kim01}. Because the growth rates of Parker and MJI modes increase with magnetic field strength and density, the compression of the ISM induced by passage through stellar spiral arms may be important in triggering GMC and GMA formation (e.g., \\citealt{mou_etal74,bli80}). Observed distributions of \\ion{H}{2} regions and OB star complexes in a ``beads on a string'' pattern along spiral arms may be the direct consequence of cooperation among the Parker instability,  Jeans instability, and spiral arm potentials (e.g., \\citealt{elm83}).  In \\citet[hereafter Paper I]{kim01}, we presented studies of detailed nonlinear evolution of self-gravitating modes in shearing, razor-thin galactic disks. We showed that formation of bound condensations via the swing instability is subject to ``threshold'' behavior, with models having sufficiently small values of the Toomre $Q$ parameter (see eq.\\ [\\ref{Toomre_Q}] below) producing bound clouds, and models with larger values of $Q$ remaining nonlinearly stable.  We found that the threshold value of $Q$ is in the range $1.2-1.4$ (for magnetic field strengths from zero up to thermal-equipartition values), consistent with an apparent mean threshold $Q_{\\rm th}=1.4$ for star formation evident from observational studies \\citep{ken89,mar01}. In Paper I, we also elucidated the role of MJI in forming bound condensations by showing that in low-shear regions where the swing amplifier cannot operate, the mechanism that leads to growth of perturbations is quite different from that in high-shear regions, and that the presence of magnetic fields is crucial to the instability. In the absence of magnetic fields, specific vorticity is conserved; contraction of vorticity-conserving masses leads to increased Coriolis forces that exceed the increase in self-gravity such that further contraction is prevented (in the absence of a shearing background). In magnetized systems, however, tension forces can transfer vorticity from one region to another, such that condensation driven by self-gravity can proceed.  Effects of spiral arm gravity on the growth of giant clouds were explicitly addressed by \\citet[hereafter Paper II]{kim02} using two-dimensional models. Paper II showed that for a spiral arm inside corotation, the enhanced surface density and shear gradients in the arm region promote the growth of gaseous spurs. These spurs jut nearly perpendicularly from the outer (downstream) side of the arm, and trail in the same sense as the arm itself; this is consistent with observed spur-like dust lanes in M51 \\citep{sco01}. We argued that the formation of spurs can be understood as a natural consequence of the MJI within spiral arms, where compression leads to a reduced (or negative) shear gradient that allows these self-gravitating modes to grow over an extended time. Paper II showed that bound clouds of typical mass $\\sim 4\\times 10^6 \\,\\Msun$ can form from spur fragmentation; such condensations potentially represent structures that would initially be observed as GMAs, and subsequently evolve to become bright OB associations.  Because of their restriction to razor-thin geometry, our previous two-dimensional models could not capture the potential consequences of the Parker instability on cloud formation. While previous work on the Parker instability has studied the weak-shear limit (see linear-theory analysis of \\citealt{shu74}, \\citealt{fog94,fog95}) and the case of solid-body rotation (see linear theory of \\citealt{zwe75} and simulations by \\citealt{cho00} and \\citealt{kimj01}), and indeed shown that rotation and shear tend to be stabilizing, no previous study focused primarily on Parker modes has incorporated shear at a realistic ``average'' galactic level. Although the Parker instability has been proposed to prompt the formation of molecular clouds (e.g., \\citealt{mou_etal74,bli80}), existing numerical simulations indicate that the Parker instability alone is unable to produce structures of high enough surface density to represent GMCs \\citep{bas97,kimj98_2,kimj00,san00,kimj01}.\\footnote{In addition, a random component of galactic magnetic fields whose strength is almost comparable to the mean uniform component is shown to suppress the Parker instability significantly \\citep{kimj_ryu01}.}  To study the coupling of the Parker instability with self-gravitating modes, \\citet{elm82a,elm82b} performed linear stability analyses for combined Parker-Jeans modes and concluded that clouds of mass $\\sim 10^5-10^6\\,\\Msun$ form in the density regime for which self-gravity significantly modifies the Parker instability; this regime of densities is expected to be found in spiral arm regions only. \\citet{han92} included solid-body rotation in a linear-theory analysis of the Parker-Jeans instability, and concluded that the transfer of angular momentum out of growing condensations may be a limiting factor. This transfer of angular momentum may be accomplished by magnetic stresses much as it occurs in the two-dimensional MJI. A nonlinear simulation of the Parker-Jeans instability in a disk with solid-body rotation was among the models performed by \\citet{cho00}, who found that filamentary structures tend to form perpendicular to the mean magnetic field.  To our knowledge, there have previously been neither linear-theory nor numerical studies that incorporate all of the aforementioned elements that can contribute to three-dimensional instabilities in galactic disks: buoyant (horizontal) magnetic fields, self-gravity, rotation, and {\\it shear}. In this paper, we investigate nonlinear evolution of galactic disks allowing for all these fundamental physical agents. This work extends our previous studies (Paper I) of structure formation via gravitational instability by including the effects of the Parker instability. Our primary objectives are to study how the Parker instability develops in the presence of strong shear, and to compare the eventual evolution of disks under the various instabilities separately or in combination. We can thus assess the potential for gravitationally-bound cloud formation from a variety of dynamical mechanisms. Since vertically-integrated, razor-thin disks are known to overestimate self-gravity at the midplane (e.g., \\citealt{too64}), we will also study analytically and numerically how the stability of three-dimensional disks to in-plane self-gravitating modes changes as their thickness varies.  In our model, disks are local (see \\S2.1), isothermal in both space and time, and vertically stratified. Initial magnetic fields are plane-parallel, pointing in the azimuthal direction. To model realistic conditions in disk galaxies, we include both external gravity arising from stars (as a time-independent potential) and gaseous self-gravity, which together counterbalance thermal and magnetic pressure gradients in an unperturbed state. We do not, however, consider in this paper other features such as stellar spiral arms and effects of cosmic rays that may enhance the magnetic buoyancy force. We begin by solving the governing equations to obtain initial static equilibrium density distributions, and then explore their gravitational stability to axisymmetric perturbations. We next select several parameter sets that represent ``average disk'' or ``spiral arm'' regions. After applying small-amplitude perturbations to the corresponding density profiles, we follow the nonlinear dynamical development with three-dimensional direct numerical simulations.  The paper is organized as follows: In \\S2, we describe the basic MHD equations in the local ``shearing-sheet'' approximation, the computational methods we use, and our model parameters. In \\S3, we present initial magnetohydrostatic equilibria in the presence of both self-gravity and external gravity, and analytically examine their axisymmetric stability by making simplifying assumptions. In \\S4, we present the results of three-dimensional simulations for the Parker instability without self-gravity (\\S4.1), generalized (Parker/swing) gravitational instability in shearing regions (\\S4.2), and MJI in nonshearing regions (\\S4.3). We summarize our results and discuss the implications of present work for galactic cloud formation in \\S5. ",
        "conclusions": "\\subsection{Summary}  A variety of evidence supports the idea that giant, star-forming clouds form as the result of large-scale collective effects -- i.e., dynamical instabilities. Candidate mechanisms that have been proposed on the basis of linear-theory analysis include the swing amplifier \\citep{gol65b,jul66}, magneto-Jeans instabilities (\\citealt{lyn66,elm87}; Paper I), and the Parker instability \\citep{par66}. The first two of these involve primarily in-plane motions and are fundamentally driven by self-gravity, with either shear or magnetic tension acting to neutralize the stabilizing tendency of the Coriolis force. For the third mechanism, vertical motion, driven by magnetic buoyancy, is crucial.  Based on linear-theory growth rates and characteristic spatial scales, all of the above mechanisms could in principle be dynamically important under various galactic conditions. In practice, because giant clouds represent highly overdense regions compared to mean ISM properties, it is important to understand how the candidate instability mechanisms develop in the {\\it nonlinear} regime: i.e., whether and/or how an instability saturates. Addressing these questions requires direct numerical simulations to evolve the fluid dynamics equations into the nonlinear regime.  In previous work, we have investigated the nonlinear development of the swing and magneto-Jeans mechanisms in a thin-disk (two-dimensional) approximation (Papers I, II). Here, we extend those studies by allowing for fully three-dimensional dynamics, which also enables us to investigate the nonlinear development of the Parker instability allowing for rotation, shear, and self-gravity. Our primary goals were to compare dynamical development, and especially ``final-state'' outcomes, of model disks under various conditions. By controlling for different physical effects (i.e., turning gravity, rotation, shear, and magnetic fields ``off'' and ``on''), and exploring parameter space, we were able to assess the potential consequences for bound cloud formation of the various instabilities alone and in combination.  The numerical models we adopt are radially localized, vertically stratified, shearing, isothermal disks with embedded magnetic field lines (see \\S2.1). The initial rate of shear in the background flow is measured by $q\\equiv -d\\ln\\Omega/d\\ln R$, and this overall shear gradient is maintained by enforcing sheared-periodic radial boundary conditions; we study cases with $q=1$ and 0. The initial magnetic field is assumed to be azimuthal and its vertical stratification is determined such that the Alfv\\'en speed $\\vA$ is spatially constant. The density and magnetic field strength in our model disks are characterized by the Toomre parameter $Q$ (see eq.\\ [\\ref{Toomre_Q}]) and the plasma parameter $\\beta$ (see eq.\\ [\\ref{beta}]), respectively. Magnetic fields are either initiated at an equipartition level ($\\vA/\\cs=1$)\\footnote{Thus magnetorotational instability cannot grow; see below.}, or set to zero. We set up initial magnetohydrostatic equilibria with thermal and magnetic pressure gradients balancing gaseous self-gravity and external gravity from a stellar component. We adopt a linear model for the external (stellar-disk) gravity (see eq.\\ [\\ref{extg}]) and parameterize its magnitude using $s_0$ (see eq.\\ [\\ref{s0}]), which corresponds roughly to the ratio of the gas disk's vertical gravity to the external gravity at one scale height. The fiducial sets of parameters our model disks adopt are $Q=1.5$, $s_0=1$ for the ``average disk'', and $Q=0.7$, $s_0=25$ for highly compressed spiral arm regions; we also perform additional simulations with $Q=1$ and $s_0=1$.  Our simulation boxes are 8 scale-heights ($H$) thick, with area either $17H\\times17H$ or $25H\\times25H$ in the $x$--$y$ plane. The relation between dimensionless simulation parameters and physical scales is explained in \\S2.2. For all our simulations, we introduce a spectrum of low-amplitude random perturbations, and numerically integrate the ideal MHD equations up to four or five orbital periods. In \\S3, we first constructed the density profiles for vertical static equilibria and calculated the variation of disk thickness as a function of $s_0$ (\\S3.1), and then studied axisymmetric gravitational instability of these thick disk models (\\S3.2). Finite disk thickness reduces the critical value of the Toomre stability parameter relative to the zero-thickness limit, and also increases the spatial scale required for instability. We found that for $s_0=1$, $Q_c\\sim0.75$, 0.72, and 0.57 for $\\beta=\\infty$, 10, and 1 cases, respectively, suggesting that the average disk is highly stable to axisymmetric perturbations. We also showed that a simple modification of the axisymmetric dispersion relation to account for finite thickness (eq.\\ [\\ref{disp}]) yields excellent agreement with the results of simulations for the value of $Q_c$.  In \\S4, we present nonlinear evolution of fully three-dimensional disk models, first considering the Parker instability in isolation (\\S4.1), then investigating evolution of the swing and Parker-swing mechanisms (\\S4.2), and finally considering how self-gravitating perturbations grow under low shear conditions (\\S4.3).  The main results drawn from these three-dimensional numerical simulations can be summarized as follows:  1. In the absence of self-gravity, the dominant growing Parker mode under uniform rotation is found to be antisymmetric in density with respect to $z=0$ and tailing with radial wavelength $\\lambda_x = 1.9H$ and azimuthal wavelength $\\lambda_y=8.5H$, and has a growth rate $\\sim0.4\\cs/H$. Density fluctuations grow as over-dense regions slide toward the midplane along undulating magnetic field lines and unloaded portions of the field buoyantly rise, but growth slows as nonlinear saturation sets in. The maximum density at saturation is less than twice the initial midplane density. After saturation of the dominant Parker mode, nonlinearly developed small-scale modes interact with each other, dispersing density structures characteristic of the Parker instability. Escape of magnetic flux across the vertical boundaries and likely reconnection of the twisted magnetic fields prompted by small-scale motions raise the mass-to-flux ratio near the midplane. The system progressively moves to a new global vertical equilibrium in which the magnetic field is nearly uniform, hence magnetic pressure support against gravity is negligible. In the saturated state, small scale fluctuating velocities are still present, but their density-weighted dispersions are less than 10\\% of the isothermal speed speed.  2. When a disk is subject to differential rotation, midplane symmetric (in density) Parker modes with $\\lambda_y=17H$ are the first to dominate the evolution. This initial dominance is explained by the large initial amplitude of longer-wavelength perturbations, combined with the ability of shear to increase $\\kx$ to the regime where Parker instabilities are more efficient. Smaller-$\\lambda_y$ antisymmetric modes start with lower amplitudes than the symmetric modes but grow more rapidly, moving gradually downwards from high altitudes to interact with the symmetric mode. The mode interaction reduces magnetic field line inflation and loss compared to the non-shearing case. The final state of the non-self-gravitating shearing disk model is similar to that of the non-shearing disk, consisting of an overall vertical equilibrium with small-scale, low-amplitude fluctuations in density and velocity fields.  3. When we include both self-gravity and uniform shear in our models, the nonlinear evolution of structure is principally dominated by the swing amplification mechanism, although Parker modes play a role in seeding structure. Low-$Q$ ($Q=0.7$) disk models end in gravitational runaway, while high-$Q$ ($Q=1.5$) disk models remain extremely stable. In both high and low $Q$ limits, the ultimate outcome is decided independent of the presence or absence of magnetic fields. When $Q$ is marginal ($Q=1$), on the other hand, we find that only magnetized models form bound condensations, suggesting that Parker instability may play a supplementary role in destabilization. The Parker instability raises the level of density fluctuation and may help channel power in the saturated state into small wavenumber modes. It is the ``rejuvenated'' swing amplification (see Paper I) that ultimately drives the system to form dense condensations, however. For cases in which bound clouds do form, the sizes and spacings are not characteristic of Parker modes, but comparable to the results from thin disk models (with typical masses $M\\sim (2-3)$ times the thin disk Jeans mass).  4. Weakly shearing thick disks, like thin disks, are susceptible to the magneto-Jeans instability (MJI), in which tension forces from azimuthal magnetic fields vitiate the stabilizing Coriolis force. The azimuthal wavelength criterion for MJI in finite-thickness disks is similar to that of the Parker instability ($\\ky H\\simlt1$), and modal growth rates are also similar. Our nonlinear simulations show, however, that MJI under weak-shear (spiral arm) conditions is ultimately much more violent than the Parker instability, and can form dense condensations within one orbital time. This is because the Parker instability is self-limiting, whereas the MJI is a runaway process. The mass of dense condensations amounts typically to the thick-disk Jeans mass ($\\sim$ 4 times the thin-disk Jeans mass).  \\subsection{Discussion}  The numerical simulations in this paper suggest that formation of bound clouds cannot occur as a direct consequence of Parker instability, although Parker instability may play a role in seeding swing amplification in high-shear environments or MJI in regions of weak shear. There are several reasons why the Parker instability is ineffective at producing large, dense clouds. First, although the growth time of the Parker instability from linear-theory analyses is a few times $10^7$ yrs (cf, \\citealt{kimj98_1}), the process does not proceed in a runaway fashion, but is stabilized at a relatively modest density by magnetic tension forces (e.g., \\citealt{mou74}). Second, while the undular Parker instability prefers wavelengths in the azimuthal direction $\\lambda_y\\sim 2\\pi H$ (i.e., $\\ky H\\sim 1$ in rough agreement with observed spacing between molecular complexes), there is no similar preferred scale in the radial direction (e.g., \\citealt{par67,giz93,kimj98_2}). Growth rates increase slowly with $\\kx$, and are nearly uniform for $\\kx H \\simgt 5$. Indeed, even though the initial perturbation amplitudes in our simulations increase toward large scale, we find that small-$\\lambda_x$ modes dominate the evolution. The results is long filamentary structures with a radial spacing about 10 times smaller than the azimuthal spacing in surface density maps during the initial growing phase (see Fig.\\ \\ref{col1}).\\footnote{The corresponding radial wavelengths are one or two times the disk scale height for symmetric and antisymmetric structures, respectively, possibly indicating nonlinear selection of these modes.} Because the perturbed self-gravitational potential $\\delta\\Phi_s\\propto - \\delta\\rho /k^2$ is increasingly small for large $\\kx$, effects of self-gravity on development of the Parker-Jeans instability are weak for the dominant nonlinear modes.  Third, while two-dimensional simulations for the Parker instability inarguably reach \\citet{mou74}-type final static equilibrium configurations (cf, \\citealt{san00}), three-dimensional simulations (\\citealt{kimj98_2,kimj00}, and this work) indicate that such equilibrium configurations are transient at best, evolving rapidly into a state where density is relatively smooth in the horizontal direction. It is unclear whether this transition is initiated by an instability of the static undulating configurations, as claimed by \\citet{lac80}, by active magnetic reconnection near the midplane caused by small-scale chaotic motions, or simply by comparable nonlinear saturation levels of many high-$\\kx$ modes. When shear is present, kinematically driven phase mixing contributes to washing out large-scale radial structures. In any case, the three-dimensional simulations show that it is almost impossible to find any signature of the undular Parker instability in the later stages of evolution.  The result that Parker instability cannot {\\it directly} account for GMC formation has previously been emphasized by other authors. \\citet{kimj01}, for example, find only a maximum factor of 1.5 enhancement in surface density in their three-dimensional non-self-gravitating simulations with solid-body rotation, and conclude that Parker mode {\\it alone} cannot be responsible for forming molecular clouds (see also e.g., \\citealt{bas97,kimj98_2,kimj00,san00}). These authors speculate on the possibility of the Parker instability coupled with other physical processes such as self-gravity, differential rotation, radiative cooling, and spiral arm potentials. We show in the present paper that self-gravity and galactic differential rotation are not much help. Therefore, we argue that {\\it the Parker instability, even combined with gaseous self-gravity and differential rotation, cannot be regarded as the primary formation mechanism for giant molecular clouds at least in galactic disks at large}. While effects on the Parker instability of stronger self-gravity and rapidly varying shear inside spiral arms have yet to be studied in detail, we do not anticipate significant differences in the main conclusions. Although the width of a spiral arm might help in promoting Parker modes of similar $\\lambda_x$ over shorter-wavelength competitors (cf., \\citealt{fra02}), it is not yet clear how effective this can be if the arm transit time is comparable to the growth time.  While larger or smaller shear rates do not seem to alter the basic character of Parker instabilities appreciably, the shear rate is crucial for selecting which type of (in-plane) gravitating amplifier can operate. In a weakly shearing regions, swing is absent, but the MJI is very powerful at initiating self-gravitational condensations. In Paper I, we showed MJI is a potential means to prompt observed active star formation near the central parts of galaxies where rotation curves are nearly rigid-body (see also \\citealt{elm87}). In Paper II, we argued that observed spur structures jutting radially outwards from spiral arms can be understood as a direct consequence of the MJI occurring within the arms. A spiral arm potential has three important effects on the gaseous medium. (1) By compressing gas, the Jeans length -- which determines the preferred MJI scale along the mean magnetic field -- is reduced; the growth time for self-gravitating modes also drops. (2) The spatially-varying density distribution causes a gradient in the shear profile perpendicular to the arm; since transiting gas first experiences reversed shear and then returns to normal shear, the spatially-varying profile tends to keep radial wavenumbers small within the arm, which enhances the efficiency of MJI. (3) It reduces the radial scale length for mass collection by MJI to at most  the arm width, so that the masses of clumps produced are limited.  Compared to our previous results from razor-thin disks, the weaker self-gravity in three dimensions could imply somewhat more massive ($\\sim10^7\\,\\Msun$) clouds would tend to form from MJI under realistic conditions. On the other hand, the weaker self-gravity may allow stronger spiral potential perturbations that give a higher density contrast between arm and interarm regions in stable quasi-axisymmetric spiral shock profiles (cf.\\ Paper II). The Jeans mass at the spiral arm density peak could be correspondingly smaller, producing lower mass clouds in three-dimensional spiral arm regions. Since the marginal wavenumber of MJI parallel to $\\mathbf{B}$ is about the same as that of the Parker instability in spiral arm conditions (see \\S4.3), one might expect some cooperation between MJI and the Parker instability inside spiral arms. The present models under uniform in-plane background conditions do not point to significant modifications of the nonlinear MJI by Parker modes. Effects from Parker-MJI interactions could be more prominent with a realistic spiral arm model (cf. \\citealt{mar98}), however, and it would be particularly interesting to learn whether the final mass scale changes markedly compared to predictions from uniform disk models.  By focusing on cases with relatively strong ($\\beta$=1) equilibrium azimuthal magnetic fields in our MHD models, we have concentrated solely on effects of the Parker instability and MJI in the present work. An additional dynamical process prevalent in magnetized disk systems is the magnetorotational instability (MRI; \\citealt{bal91,bal98}). Because MRIs are most vigorous in their small-scale growth when the thermal pressure is large compared to magnetic pressure ($\\beta\\gg 1$), they have been studied primarily in the context of accretion disks. However, \\citet{sel99} have argued that the MRI may also be important in producing turbulence in galactic disks where the effective value of $\\beta$ is close to unity, citing as evidence the significant nonthermal velocity dispersions ($\\sim 6\\,{\\rm km\\, s^{-1}}$) present in outer \\ion{H}{1} disks with low local star formation rates and apparent gravitational stability. It is of great interest to investigate whether coupling between the MRI and self-gravitating modes under galactic conditions could represent an important new mechanism for forming large-scale ISM condensations."
    },
    "0208/astro-ph0208308_arXiv.txt": {
        "abstract": "Galaxy clusters are potentially powerful probes of the large-scale velocity field in the Universe because their peculiar velocity can be estimated directly via the kinematic Sunyaev-Zeldovich effect (kSZ).  Using high-resolution cosmological simulations of an evolving cluster of galaxies, we evaluate how well the average velocity obtained via a kSZ measurement reflects the actual cluster peculiar velocity. We find that the internal velocities in the intracluster gas are comparable to the overall cluster peculiar velocity, 20 to 30\\% of the sound speed even when a cluster is relatively relaxed. Nevertheless, the velocity averaged over the kSZ map inside a circular aperture matched to the cluster virial region provides an unbiased estimate of a cluster's radial peculiar velocity with a dispersion of 50 to 100 km/s, depending on the line of sight and dynamical state of the cluster.  This dispersion puts a lower limit on the accuracy with which cluster peculiar velocity can be measured. Although the dispersion of the average is relatively small, the velocity distribution is broad; regions of low signal must be treated with care to avoid bias.  We discuss the extent to which systematic errors might be modelled, and the resulting limitations on using galaxy clusters as cosmological velocity tracers. ",
        "introduction": "\\label{sec:intro}  Accurate maps of the cosmic velocity field would provide powerful constraints on the formation of structure in the Universe. A velocity field can be compared directly with measured galaxy density field, testing whether the fundamental picture of gravitational collapse is correct \\citep{branchini01,dacosta98,dore02}, or with velocity fields predicted by cosmological models, providing useful constraints on cosmological parameters and an independent measurement of the bias parameter on a range of scales. Indeed, intensive theoretical effort in the last decade produced very accurate predictions of various properties of the velocity fields from the linear to the highly nonlinear regime \\citep[e.g.,][]{jing98,freudling99,juszkiewicz99,colin00}.  However, despite clear theoretical importance, velocity surveys have been overshadowed by recent redshift surveys because peculiar velocities are far more difficult to measure than redshifts.  Existing velocity surveys are thus more prone to systematic errors than redshift surveys.  The main difficulty in measuring peculiar velocity is, of course, obtaining the distance to an object, which is necessary if the peculiar velocity is obtained by subtracting the Hubble flow velocity from a measured redshift velocity.  Since distance errors tend to increase with distance, the reliability of redshift-based peculiar velocity surveys inevitably degrades with distance, making cosmological tests on large scales difficult.  Several groups have proposed using clusters of galaxies as tracers of the cosmic velocity field \\citep{haehnelt96,lange98,kashlinsky00,aghanim01,peel02}. The peculiar velocity of a cluster can be measured in a single step, independent of the cluster distance, via the kinematic Sunyaev-Zeldovich effect \\citep[SZE :][]{zeldovich69,sunyaev80}; for recent reviews see \\citet{birkinshaw99} and \\citet{carlstrom02}.  The electrons of the hot ionized gas in clusters Compton-scatter passing microwave background photons, resulting in a frequency-dependent redistribution of the photon energies (the thermal SZ effect).  A smaller but measurable blackbody distortion arises from the Doppler shift due to bulk motion of the electrons with respect to the rest frame of the CMB photons (the kinematic SZ effect, hereafter kSZ). In principle, these effects can be distinguished by their spectral signatures. With sufficiently high angular resolution observations the kSZ signal can be distinguished from intrinsic microwave background temperature fluctuations because galaxy clusters subtend angular scales small compared to characteristic primordial fluctuations. First measurements of the kSZ signal in clusters have been reported \\citep{holzapfel97}, and several studies of the systematic errors related to disentangling the kSZ signal from other microwave fluctuations have been performed \\citep{haehnelt96,aghanim01,diego02}.  If clusters were simple spherical objects with negligible internal structure and motions (as has been assumed in several previous papers investigating the effect), then measurements of the kinematic SZ effect would provide a relatively straightforward method for measuring the cluster peculiar velocity. However, we know from high-resolution X-ray observations \\citep[i.e.,][]{markevitch00,vikhlinin01,sun02,mazzotta02} and numerical simulations \\citep[i.e.,][]{norman99,ricker01} that galaxy clusters are actually complex objects with significant internal flows driven by frequent mergers. The observed kinematic SZ signal, which measures the density-weighted peculiar velocity of gas, will thus be some average over the bulk velocity and the internal velocities, which raises the question of how accurately an observed SZ signal will reflect the actual peculiar velocity of the cluster. Potential systematic errors can arise because the gas itself does not necessarily reflect the bulk velocity of the matter.  In this paper we use high-resolution simulations of a galaxy cluster formed in a $\\Lambda$CDM cosmological model to investigate the systematic errors that can arise from the complex internal structure and motion in estimates of the cluster velocity via the kSZ effect. In $\\S$2, we describe the cluster simulation used in our analysis.  In $\\S$3, we illustrate the structure and magnitude of internal flows within the cluster.  $\\S$4 explains how to estimate the peculiar velocity from kSZ maps, discusses definitions of peculiar velocity and analyzes the systematic errors in the kSZ estimate. We summarize our results and discuss related issues in the final section. ",
        "conclusions": "\\label{sec:discuss}  Using a high-resolution N-body+gasdynamic cluster simulation in a $\\Lambda$CDM cosmology, we have shown that  \\newcounter{bean}  \\begin{list} {\\roman{bean}}{ \\usecounter{bean} \\setlength{\\parsep}{+0.03in} \\setlength{\\leftmargin}{+0.15in} \\setlength{\\rightmargin}{+0.15in}}  \\item{} Arcminute-scale kSZ measurements will provide a generally accurate (with typical errors of $\\lesssim 100$ km/s) tracer of cluster peculiar velocities on scales of the cluster virial radius, but show significant discrepancies on scales smaller than half of the virial radius.  \\item{} In an ideal situation with an exact recovery of kSZ signal, the velocity averaged over the kSZ map pixels inside a circular aperture matched to the cluster virial region provides a statistically unbiased estimate of a cluster's radial peculiar velocity with a small dispersion of $\\lesssim 50$\\kms.  When analyzing an actual kSZ map with noise, one may want to use a smaller aperture to include only the region of the map with higher signal-to-noise; in such case, the velocity measurement is still unbiased, but with larger errors.   \\item{} While accurate cluster velocity estimates might be obtained in a statistical sense from kSZ estimates, any individual cluster can give a significant velocity error due to random subclumps with velocities along the line of sight. Comparisons of high-resolution kSZ and X-ray maps can identify such subclumps, although it is unclear how accurately their effect can be modelled on a case-by-case basis.  \\end{list}  Our estimated error amplitude agrees with the conclusions of \\citet{holder02}.  Although our analysis shows that the accurate measurements of radial peculiar velocity of clusters are possible, careful observational and theoretical studies of potential biases are required before we can use the kSZ peculiar velocity measurements to extract cosmological information.  Observationally, one needs thermal SZ effect and cluster gas temperature measurements to determine the beam-smeared column density, $\\langle \\tau^B \\rangle = ( m_e c^2 /\\sigma_T k_B) \\times [\\langle y^B \\rangle / \\langle T_e \\rangle_{n_e}^B ]$.  However, the density-weighted electron temperature $\\langle T_e \\rangle_{n_e}^B$ is not an observable; instead, the emission-weighted temperature of cluster gas $\\langle T_X \\rangle$ from the \\xray spectroscopy is often used.  If the intracluster medium is isothermal throughout, then $\\langle T_e \\rangle_{n_e}^B = \\langle T_X \\rangle$.  However, this is clearly an incorrect assumption: recent high-resolution \\xray cluster observations revealed the non-isothermality of the intracluster medium \\citep{markevitch98,peterson01,tamura01,degrandi02}.  Nevertheless, in the absence of a theoretical model of temperature structure, the isothermality assumption is likely an unavoidable one in the analysis of the real observations, particularly for distant clusters where detail temperature structure is difficult to study.  Clearly, such an incorrect assumption could introduce an additional bias in the kSZ measurements of the cluster peculiar velocity.  Using the high-resolution adiabatic simulation used in this paper, we find that the use of $\\langle T_X \\rangle$ leads to overestimate of the inferred peculiar velocity measurement at the maximum level of $\\sim$20\\% due to monotonically increasing temperature profile toward the cluster center.  \\citet{yoshikawa98}, on the other hand, reported that the use of $\\langle T_X \\rangle$ leads to underestimate of similar magnitude due to temperature drop in the center of the simulated cluster in their rather low-resolution adiabatic cluster simulations.  However, the real clusters \\citep[e.g.,][]{degrandi02} and clusters simulated with starformation and cooling \\citep[e.g.,][]{valdarnini02} seem to have more isothermal core than the simulated cluster analyzed here and this bias will be correspondingly smaller.  Further analyses using simulations with cooling and starformation are needed to evaluate the effect, while reproducing the observed temperature profiles of clusters.  The Sunyaev-Zeldovich effect is independent of redshift and unable to distinguish between objects at different redshifts along the same line of sight. While the chance of two massive galaxy clusters being aligned is negligible \\citep{voit01a}, the probability for random alignment of a galaxy group ($M\\simeq 10^{13}$ to $10^{14}$ $M_\\sun$) with a cluster will be somewhat higher and will appear like additional substructure in the kSZ map. If the group has a large line-of-sight velocity with the same sign as the cluster peculiar velocity, it will be detectable in the same way as a high-velocity merging subclump, and might be modelled out in the same way. Groups with smaller peculiar velocities will bias the velocity determination and be harder to pick out in the kSZ maps. This systematic error will be small, considering that the high-velocity merging subclump in our simulation biases the velocity determination by around 50 km/s and a random galaxy will give a substantially smaller signal. This estimate is consistent with an assessment of this effect by \\citet{aghanim01} who argue that random superposition of clusters with other objects along the line of sight leads to typical {\\it rms} velocity errors of $\\lesssim 100$~km/s.  A more accurate estimation of this bias requires modelling with large-volume simulations.  We mention in passing that significant internal bulk velocities with characteristic amplitudes of 100 to 300 km/sec will greatly complicate measurements of cluster rotations with characteristic kSZ signals at the $\\mu$K level, as proposed by \\citet{cooray02}. A realistic assessment of this effect would need to use high-resolution simulated clusters, such as the one studied here.  Measuring blackbody kSZ distortions at the few $\\mu$K level of course requires overcoming other systematics unrelated to the kSZ signal itself. Most importantly, galaxy clusters will possess a dominant thermal Sunyaev-Zeldovich distortion much larger than the kSZ signal. This thermal signal can, to some extent, be extracted via its departure from a blackbody spectrum using multiple frequency measurements, but relativistic corrections \\citep{rephaeli95,itoh01} or departures from kinetic equilibrium \\citep{blasi00} can complicate this analysis \\citep[see][]{holder02}.  Second, the blackbody kSZ distortion must be separated from the blackbody primary temperature fluctuations of the microwave background. This can be accomplished via spatial filtering, since the primary microwave background fluctuations possess little power on cluster scales. The errors in filtering will give some unavoidable systematic error. Detailed simulations of the filtering process were done by \\citet{haehnelt96} and updated for currently envisioned experimental capabilities by \\citet{holder02}.  \\citet{aghanim01} apply a simple filter to simulated Planck satellite maps, finding velocity errors due to the filtering of 300 to 600 km/s due to Planck's angular resolution.  Upcoming ground-based experiments are likely to have higher sensitivity and resolution than Planck, although with complications from atmospheric emission. The extent to which the kSZ signal can be isolated from the background fluctuations in arcminute resolution maps at high sensitivity needs to be studied in more detail. We anticipate systematic errors comparable to those from internal motions. Galactic dust emission and radio point sources are unlikely to be major problems at 200 GHz frequencies and arcminute angular scales, but lensed images of distant dusty galaxies could contribute a significant confusion noise \\citep{blain98} and may need to be imaged at higher frequencies to extract the kSZ signal accurately.  Our simulated galaxy cluster shows that, perhaps contrary to naive expectation, the internal bulk flows in galaxy clusters do not present an insurmountable source of systematic error for peculiar velocity estimates based on the kinematic Sunyaev-Zeldovich effect. Aperture-averaged velocity estimates are largely unbiased. High-velocity subclumps merging with the cluster along the line of sight will induce systematic errors, but many of these clumps can be identified as particularly bright kSZ signals, and can be removed via comparison with X-ray maps. It is not unreasonable to hope that SZ cluster surveys will eventually produce cluster peculiar velocity catalogs with systematic velocity errors at a level of 50 km/s independent of the cluster distance.  Velocity catalogs of such accuracy would provide a powerful probe of the growth of structure in the mildly nonlinear regime."
    },
    "0208/astro-ph0208372_arXiv.txt": {
        "abstract": "{ We present new theoretical calculations of the total line profiles of  Lyman~$\\alpha$ and Lyman~$\\beta$  which include perturbations by both neutral hydrogen {\\em and} protons and all possible quasi-molecular states of H$_2$ and H$_2^+$. They are used to improve theoretical modeling of synthetic spectra for cool DA white dwarfs. We compare them with \\fuse\\ observation of \\g231.  The appearance of the line wings between Lyman~$\\alpha$ and  Lyman~$\\beta$ is shown to be sensitive to the relative abundance of hydrogen ions and neutral atoms, and thereby to provide a temperature diagnostic for stellar atmospheres and laboratory plasmas. ",
        "introduction": "Structures in the Lyman~$\\alpha$ and Lyman~$\\beta$ line wings have been identified with free-free transitions which take place during binary close collisions of the radiating H atom and a perturbing atom or ion (Allard~et al.~1998a, 1998b, 1999). The characteristics of these features (position, amplitude, and shape), due to the formation of quasi-molecules during collisions between the radiating atom and perturbers, depend directly on the potential energy curves correlated to the atomic levels of the transition (Allard \\& Kielkopf 1982).  Two satellite absorption features at 1058~\\AA\\ and 1076~\\AA\\ due to collisions of atomic hydrogen with protons were first identified in the  spectrum of the DA white dwarf Wolf$\\,$1346, as observed with the Hopkins Ultraviolet Telescope (Koester~et al.~1996). These satellites in the red wing of Lyman~$\\beta$ are in the Far Ultraviolet Spectroscopic Explorer (\\fuse) spectral range (Moos et al.~2000); furthermore, Lyman~$\\beta$ profiles are also the subject of an ongoing study of the far ultraviolet spectrum of dense hydrogen~plasmas.   In Allard~et al.~(1998a) we presented  theoretical profiles of Lyman~$\\beta$ perturbed  solely by protons. The calculations were based on the accurate theoretical H$_2^+$ molecular potentials of Madsen \\& Peek~(1971) to describe the interaction between radiator and perturber, and dipole transition moments of Ramaker \\& Peek~(1972). The line profiles were included as a source of opacity  in model atmospheres for hot white dwarfs, and the predicted spectra compared well with the observed {\\it ORFEUS} and \\fuse~spectra (Koester~et al.~1998; Wolff~et al.~2001).   {\\it Ab initio} calculations of Drira~(1999) of electronic transition moments for excited states of the H$_2$ molecule and molecular potentials of  Detmer~et al.~(1998) allowed us to compute Lyman~$\\beta$ profiles perturbed by neutral atomic hydrogen (Allard~et al.~2000). The appearance of a broad  satellite situated at 1150~\\AA\\ makes necessary to take into account the total contribution of both the  Lyman~$\\alpha$ and Lyman~$\\beta$ wings of H perturbed simultaneously by neutrals and protons.  We show  that the shape of the wings in the region between Lyman~$\\beta$ and Lyman~$\\alpha$ is particularly sensitive to the relative abundance of the neutral and ion perturbers responsible for the broadening of the lines.  These new profiles  have been used to predict synthetic spectra for cool DA white dwarfs which present  structures at 1600~\\AA\\ and 1400~\\AA\\ in the Lyman~$\\alpha$ wing due respectively to quasi-molecular absorption of the H$_2$ and H$_2^+$ molecules. These last two structures have been demonstrated to be a very sensitive temperature indicators in DA white dwarfs. The relative strength of these two satellite features depends very strongly on the degree of ionization in the stellar atmosphere, and thus on the stellar parameters $T_{\\rm eff}$ and $\\log g$ (Koester \\& Allard~1993; Koester~et al.~1994; Bergeron~et al.~1995). ",
        "conclusions": ""
    },
    "0208/astro-ph0208528_arXiv.txt": {
        "abstract": "We discuss the issue of toy model building for the dark energy component of the universe. Specifically, we consider two generic toy models recently proposed as alternatives to quintessence models, respectively known as Cardassian expansion and the Chaplygin gas. We show that the former is entirely equivalent to a class of quintessence models. We determine the observational constraints on the latter, coming from recent supernovae results and from the shape of the matter power spectrum. As expected, these restrict the model to a behaviour that closely matches that of a standard cosmological constant $\\Lambda$. ",
        "introduction": "Introduction}  Currently available observations, especially from high-$z$ type Ia supernovae combined with cosmic microwave background (CMB) results \\cite{Jaffe}, suggest that about one third of the critical energy density of the Universe is in the form of ordinary matter (including classical dark matter), while the remaining two thirds are in an unclustered form which is commonly called dark energy. Among other effects, this unknown component produces a recent accelerated expansion, a behaviour which standard decelerating Friedmann models are unable to match---though see \\cite{Meszaros}. The cosmological constant $\\Lambda$ is arguably the simplest candidate for this dark energy, although it is well known that theoretical predictions for its value are many orders of magnitude off from observationally acceptable values.  Noteworthy among the many proposed alternatives are time varying scalar fields, dubbed quintessence \\cite{Caldwell,Wanga}. Quintessence models typically involve a scalar field function and in some cases more than one. However, quintessence models often suffer from a major problem that also afflicts the cosmological constant: fine-tuning. This is often referred to as the `why now?' problem: why is the cosmological constant (or a quintessence field) so small, and why does it become dominant over the matter content of the Universe right about the present day? There are so-called `tracking' models where one obtains that quintessence energy density is reasonably independent of initial conditions, but on the other hand one does have to tweak some parameters in the scalar field potential in order to obtain the desired behaviour, so this can't really be claimed as an satisfactory solution to the problem.  On the other hand, given that one has yet to see a scalar field in action, it is clear that all such toy models are not much better justified than a classical cosmological constant (despite some claims to the contrary). This is further compounded by the fact that, given some time dependence for the scale factor and energy density, one is always able to \\emph{construct} a potential for a quintessence-type model which reproduces them (see for example \\cite{Padmanabhan}). One is therefore reminded of Occam's razor and can legitimately ask if observational data provide any strong justification for them, as compared to the conceptually simpler cosmological constant.  Here we make a contribution to this ongoing discussion, by studying two particularly illuminating such toy models. Cardassian expansion \\cite{Freese,Freesea} has recently been suggested as a model for an accelerating flat Universe without any use of a cosmological constant or vacuum energy whatsoever but solely depending on a purely matter driven acceleration. This has been accomplished by the use of a modified version of the Friedmann equation where an additional \\emph{empirical} term has been added. Unfortunately, as we show here, the Cardassian model doesn't bring anything particularly new since, for most practical purposes, it reduces to a class of quintessence models. An example of an alternative to quintessence is the so-called Chaplygin gas \\cite{Kamenshchik}, which obeys a physically exotic equation of state (although it can be motivated, at some level, within the context of higher dimensions brane theory). In this case we will show that, as expected, current observations restrict the model parameters to a gravitational behaviour that is very similar to that of a cosmological constant. ",
        "conclusions": ""
    },
    "0208/astro-ph0208002_arXiv.txt": {
        "abstract": "We examine the prospects for measuring the dark energy equation of state parameter $w$ within the context of the still uncertain redshift evolution of galaxy cluster structure.   We show that for a particular X--ray survey (SZE survey) the constraints on $w$ degrade by roughly a factor of 3  (factor of 2) when one accounts for the possibility of non--standard cluster evolution.  With followup measurements of a cosmology independent, mass--like quantity it is possible to measure cluster evolution, improving constraints on cosmological parameters (like $w ~\\& ~\\Omega_M$).  We examine scenarios where 1\\%, 10\\% and 100\\% of detected clusters are followed up, showing that even a modest followup program can enhance the final cosmological constraints.   For the case of followup measurements on 1\\% of the cluster sample with an uncertainty of $30\\%$ on individual cluster mass--like quantities,  constraints on $w$ are improved by a factor of 2 to 3. For the best case scenario of a zero curvature universe,  these particular X--ray and SZE surveys can deliver uncertainties on $w$ of $\\sim$4\\% to 6\\%. ",
        "introduction": "Galaxy clusters have been used extensively to determine the cosmological matter density parameter and the amplitude of density fluctuations.   Cluster surveys in the local universe are particularly useful for constraining a combination of the matter density parameter $\\Omega_M$ and the normalization of the power spectrum of density fluctuations \\citep[we describe the normalization using $\\sigma_8$, the {\\it rms} fluctuations of overdensity within spheres of 8$h^{-1}$~Mpc radius; i.e.][]{henry97,viana99, reiprich02}; surveys that probe the cluster population at higher redshift are sensitive to the growth of density fluctuations, allowing one to break the $\\Omega_M$-$\\sigma_8$ degeneracy that arises from local cluster abundance constraints \\citep{eke96,bahcall98}.      \\citet{wang98} argued that a measurement of the changes of cluster abundance with redshift would provide constraints on the dark energy equation of state parameter $w\\equiv p/\\rho$.  Describing the problem in terms of cluster abundance only makes sense in the local universe , because, of course, one cannot measure the cluster abundance without knowing the survey volume; the survey volume beyond $z\\sim0.1$ is sensitive to cosmological parameters that affect the expansion history of the universe-- namely, the matter density $\\Omega_M$, the dark energy density $\\Omega_E$ and the dark energy equation of state $w$.  A cluster survey of a particular piece of the sky with appropriate followup actually delivers a list of clusters with mass estimates and redshifts-- that is, the redshift distribution of galaxy clusters above some detection limit.  Recently, it has been recognized that with current instrumentation it is possible to use such surveys of galaxy clusters extending to redshifts $z>1$ to precisely study the amount and nature of the dark energy \\citep{haiman01}.  Clusters are promising tools for precision cosmological measurements, because they exhibit striking regularity and they exist throughout the epoch of dark energy domination.  Moreover, their use is complementary to studies of cosmic microwave background (CMB) anisotropy and SNe Ia distance measurements \\citep{haiman01,levine02,hu02}.  Following \\citet{haiman01}, a series of analyses appeared that explore the theoretical and observational obstacles to precise cosmological measurements with cluster surveys : \\citet{holder01b} applied the Fisher matrix formalism to the cluster survey problem and showed that high yield SZE cluster surveys can provide precise constraints on the geometry of the universe through simultaneous measurements of $\\Omega_E$ and $\\Omega_M$.  \\citet{weller01} demonstrated that future SZE surveys might constrain the variation of the dark energy equation of state $w(z)$.  \\citet{hu02} examined the effects of cosmic variance on cluster surveys as well as including the effects of imprecise knowledge of a more complete list of cosmological parameters. \\citet{levine02} examined an X--ray cluster survey, showing that a sufficiently large survey allows one to measure cosmological parameters and constrain the all--important cluster mass--observable relation simultaneously.  An important caveat to these works is that the authors assumed that the evolution of cluster structure with redshift was perfectly known. In this paper, we examine the effects of uncertainties about cluster structural evolution on cosmological constraints from cluster surveys, finding that current survey projections that ignore this evolution uncertainty overstate the cosmological sensitivity of the survey.  Furthermore, we examine the effects of survey followup to measure a cluster mass--like quantity $M_f$, demonstrating that an appropriately designed survey can overcome this evolution uncertainty.  In addition, our calculations underscore the importance of incorporating information from multiple observables into future cluster surveys. Clusters of galaxies are dark matter dominated objects with baryon reservoirs in the form of an intracluster medum (ICM) and a galaxy population.  Clusters can be found through the light the galaxies emit, the gravitational lensing distortions the cluster mass introduces into the morphologies of background galaxies, the X--rays emitted by the energetic ICM, the distortion that the hot ICM introduces into the cosmic microwave background spectrum (SZE), and the effects that the ICM has on jet structures associated with active galaxies in the cluster.  These methods are largely complementary, each having different strengths.  It appears that X--ray and SZE signatures of clusters are higher contrast observables than are weak lensing or galaxy light.  That is, massive galaxy clusters are more prominent relative to the far more abundant lower mass halos and the large scale filaments when viewed with the SZE and X-ray;  projection effects are a far more serious concern when using galaxy light or weak lensing signatures.  Studies of the highest redshift galaxy clusters will likely be done with the SZE, because of the redshift independence of the spectral distortion in the CMB.  Any effort to carry out a precise cosmological study using galaxy clusters will undoubtedly be most effective through some combination of these complementary, cluster observables.  The paper is arranged in the following way.  In $\\S$\\ref{sec:surveys} we describe two representative surveys and survey followup.  Section~\\ref{sec:fisher} contains a description of our estimates of the survey sensitivity when followup is included as well as a description of our fiducial model.  Results are presented in $\\S$\\ref{sec:results} and discussed further in $\\S$\\ref{sec:conclusions}. ",
        "conclusions": "\\label{sec:conclusions} Any attempt to precisely measure the dark energy equation of state $w$ with cluster surveys will require (i) a strong external prior on the curvature (presumably from CMB anisotropy studies) and (ii) an understanding of the evolution of the relation between cluster halo mass and observable properties like the X--ray luminosity, SZE luminosity, galaxy light or weak lensing shear.  We have examined the effects of current uncertainties about cluster structural evolution; for two recently proposed cluster surveys the estimated constraints on $w$ are $\\sim$2--3 times weaker than if one assumes full knowledge of cluster evolution.  Constraints on other interesting cosmological parameters are also weakened (see Table 1).  Followup observations to measure cluster masses directly will enable one to solve for cluster structure evolution and to enhance cosmological constraints.  We have examined the effects of followup mass observations from hydrostatic or dynamical methods, and we find that even modest followup of 1\\% of the cluster sample can improve survey constraints.  Full followup with mass measurements that are 30\\% uncertain, on average, provide cosmological constraints that match or surpass those possible through $dN/dz$ alone with full knowledge of cluster evolution.  Full followup with weak lensing mass measurements is currently being planned for the SPT SZE survey.  The implications are quite interesting.  Essentially, to do precision cosmology with cluster surveys and followup, we need only know that clusters conform to mass--observable scaling relations, and that these relations evolve in some well behaved manner.  Then, together with our well established theoretical framework for structure formation, the observed cluster redshift distribution and followup masses of as few as 1\\% of the sample then provide enough information to deliver precise constraints on cosmological parameters and the character and evolution of the mass--observable relation simultaneously.  In a sense cluster surveys with limited followup are self--calibrating:  one gains detailed knowledge of the structure of the tracers (i.e. galaxy clusters) and  detailed knowledge of the evolution of the universe from the same dataset.  We have focused here on the mean equation of state parameter $w$, and for the two surveys considered, we examine the redshift variation of the sensitivity to $w$.  In the SPT SZE survey, the sensitivity to the dark energy equation of state peaks at $z\\sim1$ with sensitivity at or above half the peak for $0.65\\le z\\le1.5$.  In the DUET X-ray survey, the sensitivity peaks at $z\\sim0.7$ with sensitivity at or above half the peak for $0.45\\le z\\le 1.0$.  In the case of both surveys the $w$ sensitivity of followup mass measurements peaks near redshift $z\\sim0.35$.  In the case of the cluster redshift distribution, the most information about $w$ is provided by the highest redshift clusters, and so deeper, more sensitive surveys will in general be better for studies of the dark energy equation of state.  One interesting feature of our analysis is the orientation of the elliptical constraints on $w$ and $\\Omega_M$ (see Fig.~\\ref{fig:cosmo}).   In general, the rotation of the parameter degeneracy can be understood as the result of competing effects of changes in the volume element and the growth factor as parameters vary.  Variations in $w$ (and $\\Omega_M$) affect the survey yield in different ways at different redshifts, and so the $w$-$\\Omega_M$ degeneracy depends on the redshift distribution of a particular survey.  Rotations of parameter degeneracies occur as the maximum redshift of the survey is varied \\citep{levine02,hu02}.  We have further found that changing the prior on $\\Omega_{tot}$ and changing the degree of mass followup on a survey also result in rotations of the parameter degenaracy.  This behavior has interesting implications for the design of cluster surveys that are optimally complementary to CMB anisotropy and SNe Ia distance measurements, and it deserves further study.  In addition, we emphasize that the final constraint on the determination of cosmological parameters depends sensitively on the survey strategy and also the details of the followup. For example, a different definition of $M_f(\\theta)$ would lead to slightly different uncertainties. Changing $\\theta$ from a  quantity that varies with redshift to some fixed value leads to modest variations of the constraints. In general, best results can be obtained by optimizing $M_f(\\theta)$ so that one probes as mush of the virial mass as possible."
    },
    "0208/astro-ph0208220_arXiv.txt": {
        "abstract": "Intrinsic alignments of galaxies can mimic to an extent the effects of shear caused by weak gravitational lensing.  Previous studies have shown that for shallow surveys with median redshifts $z_m \\sim 0.1$, the intrinsic alignment dominates the lensing signal.  For deep surveys with $z_m \\sim 1$, intrinsic alignments are believed to be a significant contaminant of the lensing signal, preventing high-precision measurements of the matter power spectrum.  In this paper we show how distance information, either spectroscopic or photometric redshifts, can be used to down-weight nearby pairs in an optimised way, to reduce the errors in the shear signal arising from intrinsic alignments. Provided a conservatively large intrinsic alignment is assumed, the optimised weights will essentially remove all traces of contamination. For the Sloan spectroscopic galaxy sample, residual shot noise continues to render it unsuitable for weak lensing studies.  However, a dramatic improvement for the slightly deeper Sloan photometric survey is found, whereby the intrinsic contribution, at angular scales greater than 1 arcminute, is reduced from about 80 times the lensing signal to a 10\\% effect.  For deeper surveys such as the COMBO-17 survey with $z_m \\sim 0.6$, the optimisation reduces the error from a largely {\\em systematic} $\\sim 220$\\% error at small angular scales to a much smaller and largely {\\em statistical} error of only 17\\% of the expected lensing signal. We therefore propose that future weak lensing surveys be accompanied by the acquisition of photometric redshifts, in order to remove fully the unknown intrinsic alignment errors from weak lensing detections. ",
        "introduction": "Weak gravitational lensing by intervening large-scale structure introduces a coherent distortion in faint galaxy images.  Several independent surveys have measured this `cosmic shear' effect and are now able estimate the bias parameter, b, \\cite{HYG01,HvWGMY} and set joint constraints on the values of the matter density parameter $\\Omega_{\\rm m}$ and the amplitude of the matter power spectrum, $\\sigma_8$ \\cite{BMRE,MLB02,HYG02,Maoli,Rhodes,vWb01}.  A potential limitation of this technique is the intrinsic correlation of the ellipticities of nearby galaxies which can result from gravitational interactions during galaxy formation.  This `tidal torquing' leads to a net alignment of nearby galaxies which could produce a spurious signal similar to that induced by weak gravitational lensing.  A number of studies have investigated the amplitude of this intrinsic alignment effect with some broad agreement, (\\pcite{HRH00}, hereafter HRH, \\pcite{BTHD02,CKB01,CNPT01,CM00,HZ02,LP01,Jing,Porciani}), but it is fair to say that an accurate estimate of this effect eludes us.  For low-redshift surveys, for example SuperCOSMOS and the Sloan spectroscopic galaxy sample, where the median redshift is $z_m \\sim 0.1$, the correlation of ellipticities due to intrinsic alignment is far greater than the expected lensing signal.  As the survey depth increases, galaxies at fixed angular separation are distributed over wider ranges of redshifts, leaving only a small proportion of the galaxy pairs close enough for tidal interactions to correlate the ellipticities.  This reduction in the total angular intrinsic alignment signal, combined with an increasing line-of-sight distance boosting the lensing signal, leaves a smaller intrinsic alignment contamination for deeper surveys.  The majority of studies find that for surveys with depths $z_m \\sim 1$ the estimated contamination contributes to up to $\\sim 10\\%$ of the lensing signal, although \\scite{Jing} has argued that the contamination could be much higher.  As cosmic shear measurements become more accurate these low levels of contamination cannot be considered negligible.  If an accurate estimate of the intrinsic ellipticity correlation strength did exist, then it would be possible to subtract it, leaving the correlation induced purely by lensing.  In the absence of a good estimate for the intrinsic  correlation function, it is obvious that one can improve upon a straightforward ellipticity correlation by down-weighting galaxy pairs close in redshift and angular sky separation.  This can be done at the expense of increasing the shot noise contribution from the distribution of individual galaxy ellipticities.  In this work, we deduce the optimal pair weighting for an arbitrary survey, given accurate information about the galaxy distances. We use these results on the Sloan spectroscopic survey design to show that the optimal weighting scheme can largely remove the contamination by intrinsic alignments leaving almost pure shot noise. The shallow depth of this spectroscopic sample still prevents its use for weak lensing studies as the remaining shot noise exceeds the expected weak lensing signal.  For multi-colour surveys with photometric redshift information we propose a semi-optimised procedure of excluding pairs which are likely to be close in three dimensions.  Comparison with the optimal method shows this can be very effective.  We apply this technique to the Sloan photometric survey (SDSS) with $z_m \\sim 0.2$, and three deeper surveys, the Red-Sequence Cluster Survey (RCS) with $z_m \\sim 0.56$, \\cite{HYGBHI}, a sample of the COMBO-17 survey with $z_m \\sim 0.6$, \\cite{C17} and the Oxford Dartmouth Thirty survey (ODT) with $z_m \\sim 1.0$, \\cite{Allen}.  We find that even with fairly inaccurate photometric redshifts it is possible to reduce the contamination from intrinsic alignments significantly.  For the photometric SDSS, the improvement is dramatic, enabling it to be used as a weak lensing survey.  For the deeper surveys we show that using this weighting scheme, intrinsic alignment contamination can be reduced by several orders of magnitude.  The effect of the non-uniform weighting on the weak lensing shear correlation function is calculated.  We find that galaxy pair weighting slightly reduces the lensing signal, the reduction dependent on the photometric redshift accuracy and the survey depth.  This paper is organised as follows. In $\\S$\\ref{sec:theory} we briefly describe related weak lensing theory. In \\S\\ref{sec:method} we derive the optimal pair weighting for a spectroscopic survey.  We discuss the results of pair weighting for the Sloan spectroscopic sample in $\\S$\\ref{sec:res_sloan}, comparing HRH and Jing intrinsic alignment models.  In $\\S$\\ref{sec:photoz} we propose an alternative scheme for multi-colour surveys with photometric redshift information and present the results in $\\S$\\ref{sec:results}.  These results are compared with the expected weighted lensing signal derived in the Appendix.  In $\\S$\\ref{sec:conc} we discuss the results and the implications for the design of future weak lensing surveys. ",
        "conclusions": "\\label{sec:conc} In this paper we have shown how distance information can be used to reduce the contamination of the weak gravitational lensing shear signal by the intrinsic alignments of galaxies.  Our principal finding is that the level of intrinsic ellipticity correlation in the shear correlation function can be reduced by up to several orders of magnitude, by down-weighting appropriately the contribution from pairs of galaxies which are likely to be close in three dimensions.  For spectroscopic surveys we have derived an optimal galaxy pair weighting which reduces the contamination to a negligible component, leaving almost pure shot noise.   Application of the technique to the relatively shallow Sloan spectroscopic galaxy sample reduces the error by up to two orders of magnitude, but still leaves the lensing signal undetectable, dominated now by shot noise.  For multi-colour surveys with photometric redshift estimates  we have proposed a partially-optimised method, which removes pairs with close photometric redshifts from the computation of the shear correlation function. We find that with accurate photometric redshifts this simple method is almost as effective as the fully optimised method.  We show that, even with relatively crude photometric redshift estimates, the contamination by intrinsic alignments can be significantly reduced.  Similar conclusions were independently drawn by \\scite{KingSch02} who simultaneously proposed a weighting scheme based on photometric redshifts. Our technique however requires the assumed knowledge of a model for the intrinsic alignments.  Provided one is conservative, assuming the largest feasible model, this weighting scheme will reduce the true intrinsic alignment contamination to lensing correlation signals to a negligible level.  We have applied the method to four multi-colour survey designs.  The shallow photometric SDSS survey with $z_m \\sim 0.2$,  shows the most dramatic of improvements.  The intrinsic alignment signal is expected to dominate completely the weak lensing shear signal from a survey of this depth, but with judicious removal of pairs within $\\sim 0.14$ in estimated redshift, the error from shot noise and intrinsic alignments is reduced to only $\\sim 10\\%$ of the expected shear signal.  This opens up the possibility of using the SDSS for future cosmic shear studies. With deeper surveys, such as the RCS, COMBO-17 and the ODT survey, intrinsic alignments in the weak lensing shear signal are significantly reduced. With the highly accurate photometric redshifts of COMBO-17, the reduction is close to optimal, reducing the largely systematic $\\sim 220\\%$ error at small angular scales to a much smaller and largely statistical error of $\\sim 17\\%$ of the expected lensing shear correlation. This limiting error is predominantly shot noise which will decrease as the survey size grows. The ODT and RCS have less accurate distance estimates but even with current photometric redshift estimates, which undoubtedly will improve with time, the reduction is quite significant.  For the wide RCS, the 130\\% contamination at angular scales $\\theta \\sim 2'$, is reduced to an error $\\sim 20\\%$. For deep multi-colour surveys such as the ODT, the intrinsic alignment signal is almost completely removed, leaving noise at a level of $\\leq 2$\\% of the lensing signal.  The aim for weak lensing surveys to date has been to go as deep and as wide as possible.  With an increase in depth comes an increase in the expected lensing signal and for purely geometrical reasons we also see a decrease in intrinsic alignment correlations.  To reduce shot noise, the error introduced by the intrinsic ellipticity distribution of galaxies, surveys can go wide and/or deep.  This increases the number of lensed sources, in the limit of an infinite number of galaxies the shot noise goes to zero.  One additional source of error so far unmentioned is cosmic variance which, for small area surveys, can dominate on scales larger than a few arcminutes, \\cite{SchvWKM02}. This can be reduced by sampling different areas of the sky and combining to produce a wide field survey. Note that downweighting close galaxy pairs affects this random error little, whilst effectively removing the systematic errors from intrinsic contamination.  As the size of future lensing surveys increase, control of systematic errors becomes increasingly important. Intrinsic alignment contamination is a potentially large source of error, even if it is at a level much lower than assumed here. This paper has shown that with the addition of accurate photometric redshift information, the presence of intrinsic alignments can be effectively removed.  \\scite{HYGBHI} have shown that it is possible to extract a low weak lensing signal from a survey with a median redshift of 0.56 and argue that using this shallower survey enables more accurate star-galaxy separation and provides a fairly well determined redshift distribution for the sources, another uncertainty in weak lensing measurements to date.  The addition of depth information to weak lensing studies also opens new possibilities for the application of lensing tomography \\cite{Hu02,Hu99}, and reconstruction of the full 3D cosmological mass distribution \\cite{AndyT,HuKeeton}, but here we have concentrated only on its benefits in reducing intrinsic alignment contamination. In view of the dramatic improvements it offers lensing signal detection we therefore propose that for future weak lensing surveys, emphasis should be placed on acquiring multi-colour data in a wide area.  In the case of restricted telescope time, moderate depth surveys, $z_m \\sim 0.6$, may then become a more attractive alternative to ultra-deep observations, $z_m > 1.0$, that can often be limited by seeing."
    },
    "0208/astro-ph0208016_arXiv.txt": {
        "abstract": "{ In this paper we present single stellar population models of intermediate age stellar populations where dust-enshrouded Asymptotic Giant Branch (AGB) stars are introduced. The formation of carbon stars is also accounted for, and is taken to be a function of both initial stellar mass and metallicity. The effects of the dusty envelopes around AGB stars on the optical/near-infrared spectral energy distribution were introduced using semi-empirical models where the mass-loss and the photospheric chemistry determine the spectral properties of a star along the AGB sequence. The spectral dichotomy between oxygen-rich stars and carbon stars is taken into account in the modelling. \\\\ We have investigated the AGB sequence morphology in the near-infrared colour-magnitude diagram as a function of time and metallicity. We show that this diagram is characterized by three morphological features, occupied by optically bright oxygen-rich stars, optically bright carbon stars, and dust-enshrouded oxygen rich and carbon stars respectively. Our models are able to reproduce the distribution of the three AGB subtype stellar populations in colour-colour diagrams. Effects of dusty envelopes on the luminosity function are also investigated.\\\\ We have extended our investigations to the integrated spectro-photometric properties of stellar populations. We find that the contribution of AGB stars to the near-infrared integrated light decreases, making optical/near-infrared colours of intermediate age populations bluer than what is expected from pure photospheric emission models. ",
        "introduction": "Stars in the AGB phase are a dominant source of the near-infrared light of intermediate-age (0.1\\,Gyr$\\,\\le\\,$age$\\,\\le\\,2\\,$Gyr) stellar populations. As an example, AGB stars (defined as stars with M$_{bol}\\,\\le\\,-3.6$) contribute up to 60\\% of K-band light at 1\\,Gyr (Ferraro et al. 1995, Mouhcine \\& Lan\\c{c}on 2002a).  AGB stars develop at the end of their life a strong mass loss. The optical light of AGB stars with the highest mass loss rate is almost entirely absorbed by their dusty circumstellar envelopes and re-emitted at longer wavelengths. These obscured objects become very bright infrared sources with particularly red near-infrared colours. Omont et al. (1999) have shown that at least 25\\% of the RGB stars in the galactic bulge are losing mass, based on their very red ISO colours.  Another characteristic of AGB stars is the formation of carbon stars (i.e., AGB stars with C/O$\\,>\\,1$). It has been well known from counts in the Magellanic Cloud clusters (e.g. Aaronson \\& Mould 1980, Frogel et al. 1990, Rebeirot et al. 1993) that carbon stars represent a significant fraction of the AGB population of intermediate age clusters, at least at sub-solar metallicity. Hence, they are responsible of a significant fraction of the integrated light of intermediate age stellar populations (Costa \\& Frogel 1996). In addition, carbon stars tend to be the intrinsically brightest of the AGB stars present, as they usually correspond to later evolutionary stages than their M-type counterparts. It is well established observationally, at least for simple stellar populations, that carbon stars are redder than M stars (Persson et al. 1983).  Newly large and homogeneous statistical sample of photometric data sets in the near-infrared (2MASS, Skrutskie 1998, and DENIS, Cioni et al 1997) and in the mid-infrared (ISOGAL, Omont et al 1999), and in the optical (Zaritsky et al. 1997) give us a unique opportunity to study the late-type stellar content of our galaxy and the Magellanic Clouds, and to draw conclusions about their formation history.  The observed structure of the Hertzprung-Russell (HR) diagram of resolved galaxies is a convolution of two functions. One function describes the rate of the formation of the stars which are present today in a galaxy. The second function describes the evolution of a star in the observed HR diagram (i.e., lifetime of different stellar phases, the location of those phases in the HR diagram). At identical star formation histories, the apparent morphology is controlled mostly by the second function. Accurate stellar evolutionary models that cover the necessary range of ages, metallicities and evolutionary phases are needed to retrieve information about the star formation history of a galaxy. Unfortunately, the actual state of our understanding of late-type stellar evolution is still far from being reliable enough to perform such a goal accurately. The work presented in this series of papers aims at improving our knowledge of the effect of late type stars on stellar population properties.  Effects of the dusty envelopes surrounding AGB stars, in addition to those related to the formation of carbon stars, on (i) the morphology of the AGB sequence in the observational HR diagram of resolved stellar populations and (ii) integrated properties of unresolved populations have been neglected for a long time. Mouhcine \\& Lan\\c{c}on (2002a) have presented a new spectral library of single stellar populations where the formation of carbon stars was taken into account. They showed that the presence of carbon stars in the stellar populations lead to redder integrated near-infrared colours. Bressan et al (1998) have presented single burst population models taking into account the effect of the formation of dusty envelopes associated with AGB mass loss. They have used a radiative transfer model to correct the stellar spectra predicted by standard evolutionary tracks, assuming that all circumstellar envelopes have the same chemistry (i.e., silicate or carbonaceous grains). Hence when they consider carbon rich dust, the whole dust-enshrouded stellar population have the same properties, even low core mass stars that never form carbon stars. In addition, they have not taken into account the spectral dichotomy between oxygen rich and carbon stars, considering giant star spectra to be representative of the spectral energy distribution of AGB stars.  Based on the models of Mouhcine \\& Lan\\c{c}on (2002a), we constructed a set of theoretical models which account simultaneously for the formation of carbon stars and the dust shells around thermally pulsating-AGB (TP-AGB hereafter) stars, and obtained a new set of isochrones suited for the analysis of the near-infrared data. Our main goal, as noted above, is to study the evolution of the AGB sequence morphology and integrated near-infrared properties.  In Sect.\\,\\ref{syn} we present the grid of stellar evolutionary tracks which we use in order to calculate isochrones in the theoretical plane (i.e. luminosity vs. effective temperature). We recall briefly the modelling of the TP-AGB evolutionary phase, which is included by means of a synthetic model including all relevant physical processes that control the evolutionary properties that are related to their effects on integrated properties.  We describe the dusty envelope model used to calculate the effects of circumstellar shells around late-type stars on the energy distribution, and calculate the isochrones in the observational plane (i.e., magnitude vs. colour).  In Sect.\\,\\ref{pro} we study the effects of various TP-AGB subtypes on the evolution  of near-infrared properties of both resolved and unresolved stellar populations. We discuss how they depend on age and metallicity. Finally, in Sect.\\,\\ref{con} our conclusions are drawn. ",
        "conclusions": "\\label{con} In this paper we have modelled the effects of both carbon stars and of the presence of a steady-state outflow of matter surrounding TP-AGB stars on the optical and near-IR emission of intermediate-age stellar populations. To this aim we coupled stellar evolution models and dusty shell models, in order to derive the evolution of the extinction due to circumstellar shells along the TP-AGB phase. Effects on the luminous and cool region of the HR diagram of resolved stellar populations and on integrated near-infrared properties of unresolved intermediate-age stellar populations were investigated.  The effects of both AGB sub-populations are predicted to be strong on the distribution of AGB stars in the colour-magnitude diagram. Indeed, the AGB sequence is expected to split into three morphological features occupied respectively by oxygen-rich Early-AGB/TP-AGB stars, carbon stars, and dust-enshrouded stars with little overlap, in good agreement with the feature-rich 2MASS LMC colour-magnitude diagram.  As a test of our models, we compared our isochrones with a sample of galactic M stars, Mira stars, and extreme dust-enshrouded objects in the Galaxy and in the LMC. In the near infrared colours the models reproduce the location of both optically bright and dust-enshrouded stars in a (J-H) vs. (H-K) diagram. The models predict that the dust-enshrouded carbon star population populates a sequence with redder (H-K) than dust-enshrouded M-star population at fixed (J-H).  We show that the K-band luminosity function of carbon stars is sensitive to the inclusion of the effect of dust-shells around TP-AGB stars. The peak of the luminosity function is expected to shift to fainter magnitudes and faint tails populated by cool, luminous but obscured stars close to the end of their AGB phase are predicted.  Regarding integrated properties of stellar populations, the inclusion of dust shells around AGB stars reduces the contribution of these stars to near-IR light because of the large obscuration affecting them. This will lead to bluer optical/near-IR colours in comparison to what is predicted by models where the formation of circumstellar shells is neglected. This effect decreases as the stellar population grows older."
    },
    "0208/astro-ph0208366_arXiv.txt": {
        "abstract": "{ A realistic EOS (equation of state) leads to strange stars (ReSS) which are compact in the mass radius plot, close to the Schwarzchild limiting line \\cite{d98}. Many of the observed stars fit in with this kind of compactness, irrespective of whether they are X-ray pulsars, bursters or soft $\\gamma$ repeaters or even radio pulsars. We point out that a change in the radius of a star can be small or large, when its mass is increasing and this depends on the position of a particular star on the mass radius curve. We carry out a stability analysis against radial oscillations and compare with the EOS of other SS models. We find that the ReSS is stable and an M-R region can be identified to that effect.} ",
        "introduction": "Recently there has been some excitement about the possibility that some compact stars are made from unusual forms of matter \\cite{jjd02}. Further, from an analysis of over 1 million seismic data reports sent to the U.S. Geological Survey in the years 1990-93, which were not associated with traditional epicentral sources, Anderson et al. (2002) claim to have found two events that `have the properties predicted for the passage of a strange quark nugget' through the earth \\cite{anderson}. Approximately 8000 separate seismic stations around the world are included in the database. If confirmed this would be a discovery similar to the detection of gamma ray bursts .  The best observational evidence for the existence of quark stars seems to come from some compact objects like the X-Ray burst sources SAX~J1808.4$-$3658 (the SAX in short) and 4U~1728$-$34, the X-ray pulsar Her X-1 and the super burster 4U~1820$-$30. Among these the first is the most stable pulsating X-ray source known to man as of now. This star is claimed to be an ReSS with mass \\cite{Li99a} $\\sim 1.3 ~M_\\odot~$ and a radius of about 7 km. The mass of 4U~1728$-$34 is claimed to be less than 1.1 $M_\\odot$ in Li et al. \\cite{Li99b}, which places it much lower in the M-R plot (Fig.\\ref{MR}). So it could be still gaining mass and shift to another stable point on the M-R graph. Thus in the model proposed in \\cite{d98} there is a possible answer to the question posed by Franco \\cite{fr01}: why are the pulsations of SAX not attenuated, as they are in the 4U~1728$-$34 ?  Fig.\\ref{MR} presents M-R relations for neutron as well as strange stars. The current phenomenology of compact objects could be interpreted to indicate that the mass of a star increases due to accretion so that the radius of a star changes from one point of the stable M-R curve to another. For neutron stars, exemplified by the EOS BBB {\\footnote {this is one of the set calculated by Baldo, Bombaci and Burgeo \\cite{bbb97aa} using a realistic nuclear equation of state.}} - the curve on the right in Fig.\\ref{MR} -  a smaller mass would imply a larger radius for the star. If the mass of the star increases it should contract. Therefore, expansion due to an increase of mass, subsequent to accretion, is an unstable process for a neutron star.  The expected behaviour for SS is directly opposite to that of neutron stars as Fig.\\ref{MR} shows and therefore may support the claim that some compact stars are ReSS.  Coupled to our claim are various other evidences for the existence of ReSS, such as explaining the compact M-R relations of the two candidates given in \\cite{d98} (namely, the Her X-1 and the 4U~1820$-$30), as also the possible explanation of two kHz quasiperiodic oscillations in the 4U 1728~-~34 \\cite{Li99b}.  Recently we found that the matter of an ReSS may have diquarks on the surface which could account for the delayed emission of huge amounts of energy, after a thermonuclear catastrophe. This could be a possible scenario for superbursts \\cite{sinha}.  Some of the stars like the SAX~J1808.8$-$3658  or the PSR~1937$+$21 are fast rotors. ReSS have the possibility of withstanding high rotations which neutron stars or even bag SS cannot sustain. The maximum frequencies for the two EOS of D98 are 2.6 and 2.8 kHz respectively when they are on the mass shed limit (supramassive model) and 1.8 kHz and 2 kHz when they are in the normal evolutionary sequence as shown in Gondek-Rosi\\'nska et al. \\cite{gbzgrdd00aa}.  In the present paper we further show that the ReSS are not only stable under fast rotation but also against radial oscillations.  The strange matter hypothesis has been used to postulate a scenario whereby a neutron star collapses to a quark star, simultaneously with a gamma ray burst - the so called phenomenon of a quark nova \\cite{qnova}.  In the cosmic separation of phase scenario of Witten \\cite{wit} SS are created along with baryons in a hot environment during the expansion of the early universe. The problem was investigated for ReSS \\cite{r2000} and it was found that they are formed at a temperature $T~\\sim~70$ $MeV$. Since it is self-sustaining, the system expands as the Universe cools, the ReSS expands like the Universe itself with cooling \\cite{r2000}. The calculation also suggests that the shift in entropy due the change from normal to strange matter is not very large, indicating that the phase transition is relatively smooth.  We note that in a series of early papers van Paradijs (\\cite{p78n}) had noted that (i) the X-rays for bursters originate from stars with radii around 7 kms, assuming a canonical mass of 1.4 $M_\\odot$ for them and (ii) if one assumes a lower mass  the estimated radii also becomes lower, which fits the M-R relation for the ReSS (Fig. \\ref{MR})  Because of the extremely strong electric field stretching outside the stellar surface within $\\sim10 ^{-10}$ cm \\cite {a91npb} Usov \\cite{u1} suggested that at a finite surface temperature ($5\\times 10^8~K$) of an SS one expects to have the creation of $e^+ e^-$ pairs on its surface. Such an effect could be an additional observational signature of SS with nearly bare quark surfaces. The strange matter equilibrium will not allow the pair production to quench, contrary to the  comment in  \\cite{m98} as pointed out in \\cite{u3}. Usov claimed that the chemical potential of the electrons at the surface is large compared to their mass, being around 20 MeV and the mean velocity of the electrons is very high, so that the electric field will be restored very quickly. In our calculations we specifically find this to be $\\sim$29 MeV, thus strongly supporting the conclusions of Usov.  Usov has further claimed that from the nature of the two step process, viz. $e^+ e^-$ pair production and subsequent $\\g$ emission, the  soft $\\g$ repeaters are indeed very young SS \\cite{u4} and their genesis may be due to the impact of comet like objects with these stars \\cite{u5}.   We must also add that radio pulsars may be SS as was suggested recently by Xu et al \\cite{xqz99apj} for the PSR 0943+10 and all other drifting pulsars. Further, Kapoor and Shukre \\cite{ks} used a remarkably precise observational relation for pulsar core component widths of radio pulsars to get stringent limits on pulsar radii, strongly indicating that some pulsars are strange stars. This is achieved by including general relativistic effects due to the pulsar mass on the size of the emission region needed to explain the observed pulse widths. A recent paper supporting their ideas is Xu, Xu and Wu \\cite{xxw} for PSR 1937$+$21,the fastest known radio pulsar.  The calculations for cold ReSS by Dey et al. \\cite{d98} enables us to draw conclusions about chiral symmetry restoration (CSR) in QCD when the EOS is used to get ReSS fitting definite M-R relations \\cite{d98, Li99a, Li99b}. The empirical M-R relations were derived from astrophysical observations like luminosity variation and some properties of quasiperiodic oscillations from compact stars. The density dependence of the strong coupling constant  can be deduced from the CSR described above \\cite{srdd00mpla}.  Recently Glendenning \\cite{g00prl} has argued that the SAX could be explained as a neutron star rather than a bare SS, not with any of the existing known EOS, but with a hypothetical one, satisfying however, the well-accepted restrictions based on general physical principles and having a core density about 26 $\\rho_0$. Of course, such high density cores imply hybrid strange stars, subject to Glendenning's assumption that such stars can exist with matching EOS for two phases. There is the further constraint that if the most compact hybrid star has a given mass, all lighter stars must be larger. It was found in Li et al. \\cite{Li99b} that the star 4U~1728$-$34 may have a mass less than that of the SAX and yet have a smaller radius. Another serious difference is that the EOS of D98, using the formalism of large $N_{\\rm c}$ approximation, indeed shows a bound state in the sense of having a minimum at about $4.8~n_0$, whereas in Glendenning \\cite{g00prl} one of the assumptions is that strange matter has no bound state. ",
        "conclusions": "In summary, evidence for the existence of strange stars have been accumulating. In the present paper we review  possible candidates  and suggest that the  properties peculiar to  some of the compact stars can be explained if they are SS. In particular we point out that mass accumulation due to accretion does not lead to an increase in the radius for stars like the 4U 1728$-$34, claimed to be low mass SS from accretion data \\cite{Li99b}. If some compact objects are proven to be ReSS then parameterizations of QCD chiral symmetry restoration at high densities for quarks, the smallest particles known, can be achieved with the help of data from some of the heaviest objects in the Universe.   ReSS are stable against radial oscillations close to the maximum attainable mass. For example, the EOS of SS1 sustains gravitationally, $M_{max} \\sim 1.4 M_{\\odot}$, R=7 km with a central number density $n_c ~\\sim 16 n_0$. However, the fundamental frequency of radial oscillations becomes zero at around $n_c~9.5\\sim n_0$,  destabilizing the star after M=1.36 $M_{\\odot}$ with R= 7.24 km  (Table \\ref{eos1}). It is still on the $\\frac{dM}{dR}>0$ region. Thus the maximum mass star which is stable against radial oscillations has a  number density $\\sim 9.5 n_0$ at the centre and $\\sim 4.7 n_0$ at the surface. Macroscopically,  upto this density small vibrations may be sustained.  A corresponding linear fit has the stable values M=1.34 $M_{\\odot}$ and R=7.35 km(Table \\ref{ss1}). Radial oscillation is rather sensitive and the fit is better than 3$\\%$(Tables \\ref{eos1} and \\ref{ss1}). The same pattern is seen for SS2 and the corresponding linear fit (Tables \\ref{eos3} and \\ref{ss2}). Thus we see that almost all of the M-R region is stable.  There has been a recent controversy about the star RXJ1856.$-$3754 $-$  whether it can be inferred to be a strange star \\cite{jjd02, pons}. Analysis of the observational data of this ``no pulsar'', 120 pc away, cool  star is not conclusive. As noted by \\cite {pr}  the stable portion of the $M-R$ region shown in Fig. \\ref{reg} can accommodate this star very easily. This could be a possible candidate of our SS model provided its mass and radius are established beyond controversy.  Acknowledgements:  The sad demise of our collaborator Dr. Ranjan Ray was painful for us. We dedicate this paper to his memory. MD and MS thank the RRI for hospitality and JD, MS and SB thank the IUCAA."
    },
    "0208/astro-ph0208199_arXiv.txt": {
        "abstract": "We describe our NLTE codes which allow the computation of synthetic spectra of hot stars and accretion disks. They can be combined to compute ionizing fluxes from the hot component in symbiotic stars. ",
        "introduction": "Symbiotic stars are interacting binaries with a relatively large separation when compared to cataclysmic variables. Their orbital periods are of the order of months up to many decades. Symbiotic binaries consist of a cool giant star and a hot ionizing radiation source (see e.g.\\ Kenyon 2001). The giant looses mass by a wind and not by Roche-lobe overflow (M\\\"urset \\& Schmid 1999) at a rate of typically $10^{-6}$\\,M$_{\\odot} {\\rm yr}^{-1}$. Only a few percent of this amount is accreted by the companion. The hot component (in quiescent symbiotics) is in most cases a hot white dwarf. Two known systems are hard X-ray sources and contain accreting neutron stars (Chakrabarty \\& Roche 1997, Masetti \\etal 2002). It is not clear if an accretion disk is present in all systems. The possible formation of stable disks is supported by hydrodynamic simulations (Mastrodemos \\& Morris 1998, Dumm \\etal 2000).  The observed high luminosity of the hot component is generated by mass accretion from the giant wind. An accretion rate of \\mdot\\,$\\approx10^{-7}\\,$M$_{\\odot} {\\rm yr}^{-1}$ onto a white dwarf yields a luminosity of 100 L$_{\\odot}$. It is believed that hydrogen shell burning of the accreted material can be sustained above the white dwarf core and in this case \\mdot\\,$\\approx10^{-8}\\,$M$_{\\odot} {\\rm yr}^{-1}$ is sufficient to generate that amount of luminosity.  The white dwarf is dominating the hot component in those symbiotics where steep UV continua are observed. The derived white dwarf parameters are $R \\approx 0.1\\,$R$_{\\odot}$, $M$\\,=\\,0.5--1 M$_{\\odot}$, and the effective temperatures range between \\Teff\\,=\\,30\\,000 up to 200\\,000\\,K (see e.g.\\ M\\\"urset \\etal 1991). The accretion disk is probably dominant in about 10\\% of the systems with flat UV continua (Kenyon 2001). The disk temperature reaches 100\\,000\\,K at the inner boundary layer and decreases outwards.  It is obvious that the analysis of emission line spectra from symbiotic stars requires a realistic modeling of the white dwarf and the disk. Occasionally, the collision of winds from the white dwarf and the disk as well as jets can contribute significantly to the ionizing radiation. In the following we present our approach to calculate synthetic spectra of the hot component. In Sect.\\,2.\\ we describe the construction of white dwarf and neutron star NLTE model atmospheres, assuming planar geometry and hydrostatic equilibrium. Then we present our method to compute NLTE disk models, which assume a radial Keplerian $\\alpha$-disk structure and a detailed vertical structure (Sect.\\,3.). Up to now we are neglecting stellar winds and disk winds. We also disregard jets which are observed in some symbiotics and which could result from outbursts, disk instabilities, or very high accretion rates. ",
        "conclusions": "We have modeling tools at hand which can be used to calculate ionizing spectra of the hot component in symbiotic stars. We can compute synthetic spectra emerging from the white dwarf (or neutron star) and from the accretion disk. Our immediate aim for the near future is the inclusion of irradiation effects of the stellar spectrum onto the disk. We also want to include effects of a disk wind onto the emergent spectrum. Wind models for the compact star are already available (see e.g.\\ Jordan \\etal 1996) and must be utilized at least for analyses of those symbiotic systems where hot stellar winds are observed (Schmid 2000)."
    },
    "0208/astro-ph0208150_arXiv.txt": {
        "abstract": "The distribution of core radii of rich clusters in the Large Magellanic Cloud (LMC) systematically increases in both upper limit and spread with increasing cluster age.  Cluster-to-cluster variations in the stellar initial mass function (IMF) have been suggested as an explanation.  We discuss the implications of the observed degree of mass segregation in our sample clusters for the shape of the initial mass function.  \\\\ Our results are based on {\\sl Hubble Space Telescope}/WFPC2 observations of six rich star clusters in the LMC, selected to include three pairs of clusters of similar age, metallicity, and distance from the LMC centre, and exhibiting a large spread in core radii between the clusters in each pair.  \\\\ All clusters show clear evidence of mass segregation: (i) their luminosity function slopes steepen with increasing cluster radius, and (ii) the brighter stars are characterized by smaller core radii.  {\\it For all sample clusters}, both the slope of the luminosity function in the cluster centres and the degree of mass segregation are similar to each other, within observational errors of a few tenths of power-law slope fits to the data.  This implies that their {\\it initial} mass functions must have been very similar, down to $\\sim 0.8 - 1.0 M_\\odot$. \\\\ We therefore rule out variations in the IMF of the individual sample clusters as the main driver of the increasing spread of cluster core radii with cluster age. ",
        "introduction": "\\label{intro.sec}  The Large Magellanic Cloud (LMC) contains massive star clusters at all stages of their evolution, exhibiting a wide range of intrinsic physical properties.  The focus of this paper is a detailed comparison among the stellar populations in six rich LMC star clusters, which were chosen in three pairs of similar age, mass, metallicity, and distance from the centre of the LMC, but exhibiting a large range in core radii.  We have chosen pairs of clusters with very different core radii at the same age to test directly if variations in the initial mass function (IMF) are the cause of the systematic increase in both the upper limit and spread of the cluster core radii with increasing age seen in the rich clusters in the Magellanic Clouds (e.g., Mackey \\& Gilmore 2002 and references therein).  \\subsection{The distribution of LMC cluster core radii}  In Fig.  \\ref{lmcclusters.fig}, we show the distribution of cluster core radii as a function of age in the LMC, using the most recent determination of these properties by Mackey \\& Gilmore (2002), based on a randomly selected sample of 53 LMC clusters observed with the {\\sl Hubble Space Telescope (HST)}.  These authors confirm the observational trend that the upper limits of the core radii systematically increase with cluster age, as previously discussed by Elson, Freeman \\& Lauer (1989b), Elson (1991, 1992), and van den Bergh (1994), based on smaller cluster samples observed from the ground.  This trend reflects true physical evolution of the LMC cluster population, with some clusters experiencing little or no core expansion, while others undergo large-scale expansion due to some unknown process.  One possible explanation is cluster-to-cluster variations in the IMF (e.g., Elson et al.  1989b), and therefore different expansion rates of the clusters due to varying mass loss rates of the evolving stellar population (Chernoff \\& Weinberg 1990).  However, the IMF slopes required to explain the largest core radii are too flat to allow these clusters to survive stellar mass loss beyond several $10^7$ yr (Elson 1991, Mackey \\& Gilmore 2002), while an increasing body of evidence points towards the universality of the IMF (see Gilmore 2001 for a review).  Alternative explanations for generating the largest core radii include the dynamical effects of the binary stellar population in the cluster, the merger of binary pairs of clusters (e.g., de Oliveira, Bica \\& Dottori 2000), and expansion due to tidal forces.  We will evaluate the observational evidence in terms of these core expansion mechanisms in Section \\ref{tidal.sec}.  \\begin{figure} \\psfig{figure=lmcclusters.ps,width=9cm} \\caption{\\label{lmcclusters.fig}Distribution of core radius versus age for all LMC clusters in the sample of Mackey \\& Gilmore (2002).  The clusters observed as part of our {\\sl HST} programme GO-7307 are indicated; pairs of our sample clusters spanning a large range of core radii at (roughly) similar age are connected by dotted lines.  The solid lines indicate the expected core evolution generated by an IMF with slope $\\alpha$.} \\end{figure}  \\subsection{Effects of mass segregation} \\label{mseffects.sec}  Over the lifetime of a star cluster, encounters between its member stars gradually lead to an increased degree of energy equipartition throughout the cluster.  The most significant consequence of this process is that the higher-mass cluster stars gradually sink towards the cluster centre and in the process transfer their kinetic energy to the more numerous lower-mass stellar component, thus leading to mass segregation.  The time-scale on which a cluster will have lost all traces of its initial conditions is, to first order, well-represented by its characteristic (half-mass) relaxation time, $t_{\\rm r,h}$.  The relaxation time-scale of a specific stellar species is directly related to its mean mass.  Thus, significant mass segregation among the most massive stars in the cluster core, occurs on the local, central relaxation time-scale (comparable to just a few crossing times, depending on the stellar mass, see Bonnell \\& Davies 1998), whereas a time-scale $\\sim t_{\\rm r,h}$ is required to affect a large fraction of the cluster mass.  However, the time-scale for a cluster to lose all traces of its initial conditions also depends, among other factors, on (i) the smoothness of its gravitational potential or, equivalently, the number of stars (Bonnell \\& Davies 1998); (ii) the degree of energy equipartition reached (e.g., Hunter et al.  1995); and (iii) the slope of the mass function (MF; e.g., Lightman \\& Shapiro 1978, Inagaki \\& Saslaw 1985, Pryor, Smith \\& McClure 1986, Sosin 1997).  As the dynamical evolution of a cluster progresses, low-mass stars will, on average, attain larger orbits than the cluster's higher-mass stars, and the low-mass stars will thus spend most of their time in the cluster's outer regions, at the extremes of their orbits.  For this reason alone, we would not expect to achieve global equipartition in a cluster (e.g., Inagaki \\& Saslaw 1985).  In these outer parts, the cluster's gravitational potential is weaker and constantly changing due to the ongoing redistribution of mass (Chernoff \\& Weinberg 1990), and it is more easily affected by the tidal field in which the cluster resides.  In these circumstances, two effects will enhance the mass segregation signatures observed in old, evolved clusters, (i) evaporation and ejection across the cluster's tidal boundary of (preferentially) low-mass stars, because of their higher velocity dispersion and number density (Chernoff \\& Weinberg 1990, Giersz \\& Heggie 1997), and (ii) tidal stripping by the external gravitational field of the low-mass stars sent to the cluster's outer regions by the relaxation process in the inner regions.  We will discuss the effects of the tidal field on a cluster's degree of mass segregation in relation to its size in Section \\ref{tidal.sec}.  In Section \\ref{observations.sec}, we present our sample of six rich LMC clusters, for which we analyse the degree of mass segregation attained over their lifetimes in Section \\ref{masssegr.sec}, based on the clusters' luminosity functions (LFs) derived in Section \\ref{mass.sec}. ",
        "conclusions": ""
    },
    "0208/astro-ph0208050.txt": {
        "abstract": "{In the course of a weak gravitational lensing survey of 39 clusters of galaxies, covering a total sky area of $\\sim 1$ square degree, we have serendipitously discovered mass concentrations in the fields of \\objectname{A1705} and \\objectname{A1722} which are most probably not associated with the main cluster target. By combining weak lensing information with two-color galaxy photometry in fields centered on our sample clusters, we identify a new cluster candidate at $z \\sim 0.5$ in the field of \\objectname{A1705}. This cluster candidate also displays strong lensing in the form of a giant luminous arc. The new mass concentration in the field of \\objectname{A1722} also seems to be associated with an optically luminous cluster of galaxies at $z \\sim 0.5$, but in this case there is some evidence for additional structures along the line of sight that may contribute to the lensing signal. A third cluster, \\objectname{A959}, has a dark sub-clump which shows interesting morphological evidence in the mass map for being associated with the main cluster. This is the first case where there is any significant evidence for a physical association between a dark sub-clump (discovered from weak lensing) and a normal cluster. Analysis of archival X-ray data shows that the three new mass concentrations are not firmly detected in X-rays and that they are X-ray underluminous. } ",
        "introduction": "Weak gravitational lensing provides a powerful way to identify cluster-sized density peaks in the Universe, independent of their baryonic content. Given the currently modest sky coverage of optical imaging surveys with the depth and image quality required to detect new clusters by their weak lensing effect, the number of currently known mass-selected clusters is very small. It is still an open question whether the mass-selection will lead to the identification of a population of clusters which are physically different from optically selected clusters or X-ray detected clusters. If a population of ``baryon-poor'' clusters is found to exist, they may be very useful laboratories for the study of dark matter properties. For instance, some dark matter candidates such as sterile neutrinos may produce an observable signature from their decay (Abazajian, Fuller, \\& Tucker 2001). The best places to detect such a signature would be in baryon-poor clusters -- if such objects exist -- where the spectral line corresponding to dark matter decay would be relatively more conspicuous compared to the emission produced by bremsstrahlung in the hot intra-cluster gas (Hansen et al.\\ 2002). In any case, the existence of baryon-poor clusters or sub-clusters would pose a challenge to current models for structure formation. Furthermore, any previously unrecognized population of clusters with high mass-to-light ratios would have to be taken into account when using the measured average mass-to-light ratios of clusters to estimate the density parameter $\\Omega_m$.  The present sample of weak lensing-detected clusters is small and contains both clusters with ``normal'' mass-to-light ratios and objects which appear to be optically dark. From weak lensing observations in the field of the cluster \\objectname{A1942}, Erben et al.\\ (2000) find a secondary mass peak $\\sim 7\\arcmin$ south of the cluster center which does not correspond to any strong concentration of bright galaxies. From deep $H$-band imaging of this region, Gray et al.\\ (2001) constrain the bolometric mass-to-light ratio to be $M/L_{\\rm Bol} > 1000 h$ in solar units for any reasonable lens redshift. Umetsu \\& Futamase (2000) find a dark mass concentration $1\\farcm 7$ south of the high-redshift cluster \\objectname{CL1604+4304} ($z=0.90$) with an estimated mass of $1.2 \\times 10^{14} h^{-1} M_{\\odot}$ and $M/L_B \\geq 1000 h$ in solar units, if it is located at the redshift of \\objectname{CL1604+4304}. At present, it is not clear whether the dark clumps found by Erben et al.\\ and Umetsu \\& Futamase are physically associated with the nearby clusters, or whether they represent chance alignments on the sky of objects at different redshifts. Most recently, Miralles et al.\\ (2002) have reported evidence for another dark cluster from a conspicuous alignment of faint galaxies in a parallel STIS pointing adjacent to the local Seyfert galaxy~\\objectname{NGC~625}.  Wittman et al.\\ (2001) report the discovery of a more ``normal'' cluster in a ``blank sky'' field through a combination of weak gravitational lensing and photometric data. Their $BVRI$ photometry shows a concentration of elliptical galaxies close to the lensing-derived mass peak corresponding to the cluster, and spectroscopic follow-up of candidate cluster members reveals a cluster with modest galaxy velocity dispersion ($\\sigma_v = 615 \\pm 150~{\\rm km}~{\\rm s}^{-1}$) at $z=0.28$. The mass-to-light ratio of this cluster is $M/L_R = 560\\pm 200 h$, which is somewhat high compared to average values of $M/L_B \\approx 300 h$ obtained from both virial and weak lensing analyses of X-ray selected clusters (e.g., Carlberg et al.\\ 1997; Hoekstra et al.\\ 2002; Dahle et al.\\ 2003, in prep.), but there are some X-ray selected clusters with similar lensing-derived $M/L$, such as \\objectname{MS~1224.7+2007} at $M/L_R = 640 \\pm 150$ (Fahlman et al.\\ 1994; Fischer 1999), and \\objectname{A68} and \\objectname{A697} at $M/L_R = 680 \\pm 230 h$ and $M/L_R = 450 \\pm 115 h$, respectively (Dahle et al.\\ 2003, in prep).  In Paper I in this series (Dahle et al.\\ 2002) we presented weak lensing measurements of a sample of 39 X-ray selected clusters. The results were presented in the form of maps of the reconstructed projected matter density ($\\kappa = \\Sigma / \\Sigma_{\\rm crit}$, where $\\Sigma_{\\rm crit}$ is the critical surface mass density for lensing) and in the form of radial mass profiles around each (lensing-determined) cluster center.  Several mass maps show evidence for sub-peaks in the mass distribution which are not associated with obvious sub-clumping of optically luminous galaxies inside the cluster. In this paper, we investigate the properties of the most significant of these sub-peaks. We attempt to constrain the redshift and mass-to-light ratio of these systems and discuss whether they are likely to be physically associated with their apparent ``host clusters''.  In \\S 2 we describe the selection criteria for cluster candidates, which we describe individually in \\S 3. In \\S 4 we compare our new (sub-)cluster candidates to those found by other groups, and compare the observed abundances and physical properties of such objects to recent theoretical predictions.  The numbers in this paper are given for an Einstein-de Sitter ($\\Omega_m = 1$, $\\Omega_{\\Lambda} = 0$) cosmology. The Hubble parameter is given by $H_0 = 100 h\\, {\\rm km}\\, {\\rm s}^{-1} {\\rm Mpc}^{-1}$, and all celestial coordinates are given in J2000.0. ",
        "conclusions": "Using weak gravitational lensing data, we have in the previous paragraphs identified three prominent (projected) mass concentrations representing new galaxy cluster (or sub-cluster) candidates. We use two-color, $V$- and $I$-band photometry to investigate the nature of these structures and make rough redshift estimates.  The first mass concentration, \\objectname{WL 1017.3+5931}, is the most enigmatic of these objects. The morphology of the mass peak and the associated X-ray emission suggest a possible association with the nearby $z = 0.29$ cluster \\objectname{A959}, but we find no overdensity of early-type cluster galaxies at the location of \\objectname{WL 1017.3+5931}. There is also no strong evidence for clustering of galaxies at any other redshift at this position. This object remains a good candidate for an optically ``dark'' (sub-)cluster. The upper limit on the X-ray luminosity leaves the possibility that it is an X-ray (underluminous) cluster at or beyond the redshift of \\objectname{A959}. Deep X-ray data of this system woulde be particularly interesting in order to accurately measure its hot gas content.  The second mass concentration, \\objectname{WL 1312.5+7252}, appears to be associated with a rich cluster at $z \\sim 0.55$ which also acts as a strong lens. This structure is associated with a prominent overdensity of red galaxies and has a prominent red galaxy sequence. Thus, it is likely to constitute a single strong physical overdensity of dark matter and galaxies, rather than being caused by a line-of-sight projection of lesser structures. Its $M/L_B$ value is similar to the typical values of clusters selected from baryonic tracers. The conservative upper limit on the X-ray luminosity of \\objectname{WL 1312.5+7252} is consistent with a cluster at $z \\sim 0.55$ with the mass determined from weak lensing. However, \\objectname{WL 1312.5+7252} could well be an X-ray dark mass concentration, containing only a small amount of hot gas.  The third mass concentration, \\objectname{WL 1320.4+6959}, is associated with an overdensity of galaxies at an estimated redshift of $z \\sim 0.45$, but these galaxies show a larger spread in $V-I$ color than the galaxies in the \\objectname{WL 1312.5+7252} cluster. However, the upper limit on its X-ray emission does not allow us to rule out that \\objectname{WL 1320.4+6959} is a single cluster at $z\\sim 0.45$ which is X-ray underluminous and possibly X-ray dark. It is thus unclear whether this density peak represents a chance superposition of objects at different redshifts or whether it represents a single cluster at $z \\sim 0.45$ which is still forming and is not yet virialized. Weinberg \\& Kamionkowski (2002) estimate the abundances of non-virialized, X-ray underluminous protoclusters that will be detectable in weak lensing surveys, and they find that such systems are likely to be detected in surveys comparable to ours. The $M/L_B$ value we find is typical for a normal optically luminous cluster.  At this point we may speculate whether it is significant that we find three mass concentrations with $\\kappa > 0.15$ based on 1.0 deg$^2$ of imaging of fields containing massive clusters, while Wilson et al.\\ (2001) find no such objects from similar data (1.5 deg$^2$ of imaging) of blank fields. We also recall that the two optically dark mass concentrations previously discovered by Erben et al.\\ (2000) and Umetsu \\& Futamase (2000) were both found in the fields of massive clusters. In the case of \\objectname{WL 1312.5+7252} we are seeing a mass concentration which is clearly physically unrelated to the nearby Abell cluster. For \\objectname{WL 1320.4+6959} it is hard to draw an equally firm conclusion, since, as noted in \\S~\\ref{sec:a1722field}, there may be a significant contribution to the projected mass density from structures at the redshift of \\objectname{A1722}.   \\ifthenelse{\\equal{\\version}{_apj}} {}{ \\begin{figure*} \\centering\\epsfig{file=f7.eps} \\caption[Area around secondary mass peak in A1722.] {The plots show a $225\\arcsec \\times 150\\arcsec$ region around the \\objectname{WL 1320.4+6959} mass peak in the field of \\objectname{A1722}. Top: ``True color'' image based on 4.5h of integration in the $V$-band and 4h of integration in the $I$-band. Bottom: Solid lines show contours of the projected mass density $\\kappa$ indicating the \\objectname{WL 1320.4+6959} mass peak. Contour levels start at $\\kappa = 0.1$ and are plotted at intervals of $0.02$ in $\\kappa$. The dashed lines are contours of the smoothed galaxy density distribution in the $2.2 < V-I < 2.5$ subpanel in Figure~\\ref{fig:a1722_colnumdens}.} \\label{fig:a1722darkmass} \\end{figure*} }  The photometric data for \\objectname{WL 1017.3+5931} do not provide any strong evidence for an association with nearby \\objectname{A959}, but the morphology of the mass peak does suggest a possible physical link between the two objects. In a recent study, White, van Waerbeke \\& Mackey (2002) show that significant peaks in the projected density distribution, resembling clusters with virial masses of $10^{14} - 3\\times 10^{14} h^{-1} M_{\\odot}$ may be generated by line-of-sight projections of multiple correlated structures with $M < 10^{14} h^{-1} M_{\\odot}$. It is possible that \\objectname{WL 1017.3+5931} is such an object, generated by a superposition of $\\sim 10^{13} h^{-1} M_{\\odot}$ halos within the overdensity associated with \\objectname{A959}. This scenario would naturally explain the low X-ray luminosity, but does not predict an unusually high $M/L_B$ value or the lack of an associated peak in the galaxy density distribution.  Clearer answers to the nature of weak lensing-detected mass concentrations may come soon from systematic cluster searches in the deep wide-field imaging data sets currently used for measurements of ``cosmic shear''. The sky area collectively probed by such surveys to similar depth is now at least an order of magnitude larger than the sky area we study here (see e.g., van Waerbeke et al.\\ 2001).  Our results also demonstrate the usefulness of multi-color photometry and color slicing techniques when interpreting results from weak lensing cluster searches. At redshifts $z \\sim 0.5$ and higher, even rich clusters do not represent strong galaxy density enhancements in single-passband imaging data. In the case of \\objectname{WL 1320.4+6959}, an optical counterpart to the mass concentration could not be identified before data in a second passband had been obtained. It is also clear that X-ray data from either Chandra or XMM-Newton are required in order to detect or tightly constrain the hot gas content of the detected (sub-)clusters.  The methodology for cluster searches may be further refined in the future by developing a more objective and quantitative search algorithm that combines weak gravitational lensing information with e.g, the cluster-red-sequence method of Gladders \\& Yee and/or X-ray data. The X-ray data would be particularly useful for separating line-of-sight superpositions of less massive objects (which would be a significant source of noise and bias for both optical and weak lensing data; see e.g., Hoekstra 2001; White et al. 2002) from genuine deep potential wells. This would greatly improve the power of such surveys to constrain cosmological models."
    },
    "0208/astro-ph0208400_arXiv.txt": {
        "abstract": "{ We develop a model of thin turbulent accretion discs supported by magnetic pressure of turbulent magnetic fields. This applies when the turbulent kinetic and magnetic energy densities are greater than the thermal energy density in the disc. Whether such discs survive in nature or not remains to be determined,  but here we simply demonstrate that self-consistent solutions exist when the $\\alpha$-prescription for the viscous stress, similar to that of the original Shakura--Sunyaev model, is used. We show that $\\alpha \\sim 1$ for the strongly magnetized case and  we calculate the radial structure and emission spectra from the disc in the regime when it is optically thick. Strongly magnetized optically thick discs can apply to the full range of disc radii for objects  $\\la 10^{-2}$ of the Eddington luminosity or for the outer parts of discs in higher luminosity sources. In the limit that the magnetic pressure is equal to the thermal or radiation pressure, our strongly magnetized disc model transforms into the Shakura--Sunyaev model with $\\alpha=1$. Our model produces spectra quite similar to those of standard Shakura--Sunyaev models. In our comparative study, we also discovered a small discrepancy in the spectral calculations of Shakura \\& Sunyaev (1973). ",
        "introduction": "\\label{sec1}  The well known and most widely used model of the accretion disc was proposed and elaborated by \\citet{shakura72} and \\citet{shakura73}. In this model the disc is vertically supported by the thermal pressure \\citep{shakura73}. Turbulent viscosity is invoked in the Shakura--Sunyaev model to explain the angular momentum transfer required by the accretion flow. As originally pointed out in \\citet{lyndenbell69} and \\citet{shakura73} a magnetic field can also contribute to the angular momentum transport. A robust mechanism of the excitation of magnetohydrodynamical (MHD) turbulence was shown to operate in accretion discs due to the magneto-rotational (MRI) instability \\citep{balbus98}. The growth of the MRI leads to the excitation of turbulent magnetic fields and self-sustained MHD turbulence. The contribution of Maxwell stresses to the transport of angular momentum is usually larger than Reynolds stresses. However, the magnetic energy observed in many numerical experiments was smaller than the thermal energy of the gas in the disc \\citep{brandenburg98}. Simulations of the non-linear stage of MRI are typically local simulations in a shearing box of an initially uniform small part of the disc. Attempts to expand the computational domain to include a wider area of radii and azimuthal angle \\citep*{hawley01a, hawley01b, armitage01} are underway. However, even before the recent focus on the MRI \\citet*{shibata90} observed the formation of transient low $\\beta$ state in a shearing box simulations of the non-linear Parker instability in an accretion disc.  Vertical stratification has been considered in the shearing box approximation \\citep{brandenburg95, miller00}. In particular, \\citet{miller00} investigated discs with initial Gaussian density profiles supported by thermal pressure. The initial seed magnetic field grows and starts to contribute to the vertical pressure balance. The computational domain extends over enough vertical scale heights to enable \\citet{miller00} to simulate the development of a magnetically dominated corona above the disc surface. In the case of an initial axial magnetic field, \\citet{miller00} observed that the saturated magnetic pressure dominates thermal pressure not only in the corona but everywhere in the disc. As a consequence, the thickness of the disc increases until it reaches the axial boundaries of the computational box. The formation of low $\\beta$ filaments in magnetized tori was also observed in global MHD simulations by \\citet*{machida00}. Although further global MHD simulations of vertically stratified accretion discs are needed, this numerical evidence suggests that magnetically dominated thin discs may exist.  Previously, analytic models of thin accretion discs with angular momentum transfer due to magnetic stresses were considered by \\citet{eardley75} and \\citet{field93a, field93b}. Both these works included magnetic loops with size of the order of the disc thickness. In \\citet{eardley75}, the magnetic loops were confined to the disc. Loop stretching by differential rotation was balanced by reconnection. The reconnection speed was a fraction of the Alfv\\'en speed. Radial magnetic flux was considered as a free function of the radius. Vertical equilibrium and heat transfer were treated as in \\cite{shakura73}, with the addition of the magnetic pressure in the vertical support. No self-consistent magnetically dominated solutions were found in model of \\citet{eardley75}.  In contrast, dominance of the magnetic pressure over the thermal and radiation pressure was postulated from the beginning by \\citet{field93a,field93b} and verified at the end of their work. These authors assumed that the ordered magnetic field in the disc, amplified by differential rotation, emerges as loops above the surface of the disc due to Parker instability. Because the radial magnetic field in the disc has an intially sectorial structure, the loops above the disc  come to close contact and reconnect. All dissipation of magnetic field occurs in the corona in the model of \\citet{field93a,field93b}. Such a corona was assumed to be consisting of electrons and some fraction of positrons and no outflow from the disc is present. Electrons and positrons are accelerated to relativistic energies at the reconnection sites in the disc corona and subsequently emit synchrotron and inverse Compton photons. Because reconnection was assumed to occur at loop tops, \\citet{field93a,field93b} found that up to 70 per cent of the energy released in reconnection events in the corona will be deposited back to the surface of the disc in the form of relativistic particles and radiation. Only thin surface of optically thick disc is heated and cools by the thermal emission, which is the primary source of soft photons for the inverse Compton scattering by relativistic particles in the corona.  Since the characteristic velocity of rise of the loops of the buoyant magnetic field is of the order of the Alfv\\'en speed, it takes about the time of a Keplerian revolution for the loop of the magnetic field to rise (e.g., \\citealt{beloborodov99}). This is also about the characteristic dissipation time of the magnetic field in shocks inside the disc (see section~\\ref{sec2}). The  model we explore here differs from that of \\citet{field93a,field93b} in that the dissipation of the magnetic energy occurs essentially inside the disc and the heat produced is transported to the disc surface and radiated away. Observations of hard X-ray flux indicate the presence of hot coronae where a significant fraction of the total accretion power is dissipated. For example, the X-ray band carries a significant fraction of the total luminosity of Seyfert nuclei: the flux in the 1--10~keV band is about $1/6$ of the total flux from {\\it IR} to X-rays, and the flux in 1--500~keV band is about 30--40 per cent of the total energy output \\citep*{mushotzky93}. Another example is the low/hard state of galactic black hole sources, where the borad band spectrum is completely dominated by a hard X-ray power law, rolling over at energies of $\\sim 150\\,\\mbox{keV}$ \\citep{nowak95, done02}. Also, in the so called very high state, some of galactic black hole X-ray sources show both thermal and non-thermal (power law) components, with the ratio of non-thermal to total luminosity of 20--40 per cent \\citep{nowak95}. Reconnection events and particle acceleration should also happen in rarefied strongly magnetized corona of the disc in our model and could cause observed X-ray flaring events. However, we do not consider the coronal dynamics here, and instead just focus on the structure and the emission spectrum of the disc itself.  Models of magnetized accretion discs with externally imposed large scale vertical magnetic field and anomalous magnetic field diffusion due to enhanced turbulent diffusion have also been considered \\citep{shalybkov00, campbell00, ogilvie01}. The magnetic field in these models was strong enough to be dynamically important. But those models are limited to the subsonic turbulence in the disc and the viscosity and magnetic diffusivity are due to hydrodynamic turbulence. Angular momentum transport in those models are due to the large scale global magnetic fields. Both small scale and large scale magnetic fields should be present in real accretion discs. Here we consider the possibility that the magnetic field has dominant small scale component, that is magnetic field inside the disc consists mostly of loops with size less than or comparable to the thickness of the disc.  We consider vertically integrated equations describing the radial structure of the magnetically dominated turbulent accretion disc and provide the solutions for the radial dependences of the averaged quantities in section~\\ref{sec2}. In section~\\ref{sec3} we analyse the conditions for a magnetically supported disc to be self-consistent. In section~\\ref{sec4} we calculate thermal emission spectra of magnetically supported disc taking into account scattering by free electrons. ",
        "conclusions": "We have found self-consistent solutions for thin, magnetically supported turbulent accretion discs assuming the tangential stress $\\displaystyle f_{\\phi}=\\alpha(r) \\frac{B^2}{4\\pi}$. When compared to the standard $\\alpha$-disc models \\citep{shakura73} magnetically dominated discs have lower surface and volume densities at the same accretion rate. This is due to the more efficient angular momentum transport by supersonic turbulence and strong magnetic fields than the subsonic thermal turbulence of the standard model. As a result, magnetically dominated discs are lighter and are not subject to self-gravity instability. In the limit of plasma $\\beta=1$, magnetic pressure is comparable to the largest of radiation or thermal pressures and our strongly magnetized disc model transforms into the Shakura--Sunyaev model with $\\alpha=1$.  When we derived the disc structure, we made no explicit distinction between turbulent and magnetic pressure support and angular momentum transfer. As such, our model would be valid in any situation in which the magnetic and turbulent kinetic energies are comparable to, or greater than the thermal energy density. The assumption that the kinetic and magnetic energies are nearly comparable is natural because turbulence should result in the amplification of small scale magnetic fields in highly conducting medium due to dynamo action. Typically, in a sheared system, the magnetic energy can be even slightly larger than turbulent kinetic energy since the magnetic energy gains from the additional shear. We find that the thermal spectrum from the surface of the magnetically dominated disc in the optically thick regime is close to the spectrum of the standard Shakura--Sunyaev disc.  The issue  arises as to how the magnetic field could reach sonic or supersonic energy densities. To obtain sonic turbulence and produce a $\\beta=1$ disc, the MRI might be sufficient. To obtain a $\\beta < 1$ supersonic turbulence may require something else. One possibility in AGN appeals to the high density of stars in the central stellar cluster surrounding AGN accretion  discs. Passages of stars through the disc might be an external source of supersonic turbulence analogous to the supernovae explosions being the source of supersonic turbulence in the Galaxy. Stars pass through the disc with the velocities of order of Keplerian velocity, which is much larger than the sound speed in the disc. We consider the support of turbulence by star-disc collisions in Appendix~B and find that statistically speaking, star-disc collisions are unlikely to provide enough energy to sustain supersonic turbulence in most AGN accretion discs, however the possibility remains that a small number out of a large population could become magnetically dominated.  Indeed whether a disc could ever really attain a magnetically dominated state is important to understand. The  present answer from simulations is not encouraging, but not completely ruled out. Further global MHD simulations of turbulence in vertically stratified accretion discs with realistic physical boundary conditions are needed along with more interpretation and analysis. Magnetic helicity conservation for example, has not been fully analyzed in global accretion disc simulations to date, and yet the large scale magnetic helicity can act as a sink for magnetic energy since magnetic helicity inverse cascades.  As an intermediate step in assessing the viability of low $\\beta$ discs,  it may be interesting to assess whether they are stable. One can take, as an initial condition, the stationary model of the magnetically dominated accretion disc given by expressions~(\\ref{eqn14}), (\\ref{eqn17}--\\ref{eqn19}) with the initial magnetic field satisfying all constraints of our model and falling into the shadowed regions on plots in Figs.~\\ref{fig_m8d54}--\\ref{fig_m1d54}.  One point of note is that magnetically dominated discs may be helpful (though perhaps not essential, if large scale magnetic fields can be produced \\citep{blackman02, blackman03}) in explaining AGN sources in which 40\\% of the bolometric luminosity comes from hot X-ray coronae. If the non-thermal component in galactic black hole sources is attributed to the magnetized corona above the disc (e.g., \\citet{dimatteo99}, also \\citet{beloborodov99} discusses possible alternatives), then magnetically dominated discs can naturally explain large fractions, up to 80~\\% \\citep{dimatteo99}, of the accretion power being transported into coronae by magnetic field buoyancy (although $\\beta\\la 1$ disc solutions are also possible, \\citep{merloni03}). Though coronae can form in systems with high $\\beta$ interiors, the percentage of the dissipation that goes on in the interior vs. the coronae could be $\\beta$ dependent.  The main purpose of our study was simply to explore the consequences of making a magnetically dominated analogy to Shakura and Sunyaev, and filling in the parmeter regime which they did not consider. In the same way that we cannot provide proof that a disc can be magnetically dominated, they did not present proof that a disc must be turbulent, but investigated the  consequences of their assumption. We also realize that the naive $\\alpha$ disc formalism itself can be questioned and its ultimate validity in capturing the real physics is limited. Nevertheless it still has an appeal of simplicity.  Finally, we emphasize that our model does not describe dissipation in the corona and interaction of the corona with the disc. Further work would be necessary to address relativistic particle acceleration and emission, illumination of the disc surface by X-rays produced in the corona and subsequent heating of top layers of the disc, and emergence of magnetized outflows."
    },
    "0208/astro-ph0208293_arXiv.txt": {
        "abstract": "We present the results from three dimensional hydrodynamical simulations of decaying high-speed turbulence in dense molecular clouds. We  compare our results, which include a detailed cooling function, molecular hydrogen chemistry and a limited C and O chemistry, to those previously obtained for decaying isothermal turbulence.  After an initial phase of shock formation, power-law decay regimes are uncovered, as in the isothermal case. We find that the turbulence decays faster than in the isothermal case because the average Mach number remains higher, due to the radiative cooling. The total thermal energy, initially raised by the introduction of turbulence, decays only a little slower than the kinetic energy.  We discover that molecule reformation, as the fast turbulence decays, is several times faster than that predicted for a non-turbulent medium. This is caused by moderate speed shocks which sweep through a large fraction of the volume, compressing the gas and dust. Through reformation, the molecular density and molecular column appear as complex patterns of filaments, clumps and some diffuse structure. In contrast, the molecular fraction has a wider distribution of highly distorted clumps and copious diffuse structure, so that density and molecular density are almost identically distributed during the reformation phase. We conclude that molecules form in swept-up clumps but effectively mix throughout via subsequent expansions and compressions. ",
        "introduction": "\\label{sec:intro} An understanding of turbulence is of fundamental importance to many research areas in astrophysics. In particular, compressible turbulence plays a central role in the process by which stars are formed (reviewed by \\cite{VazSem00, Padoan01}). Stars form out of dense clouds of molecular gas, which have formed out of diffuse clouds of atomic gas. The turbulent energy deduced from observations of the molecular gas is sufficient to delay the gravitational collapse, making molecular turbulence of great interest. The purpose of this study is to provide insight through the first three-dimensional simulations of molecular turbulence.  Isothermal or polytropic equations of state have been the central premise of previous investigations of turbulent models for molecular clouds (e.g. \\cite{VazSem96, M-MML98, Padoan98, Stone98}). This simplification has the advantage that parameter space can be explored with numerical simulations, and the influence of magnetic fields and gravity can be determined in some detail  (e.g. \\cite{Bals01, Passot95, Kless00a, Heitsch01} ). The isothermal approximation is indeed valid in many specific models where molecular cooling is efficient and low gas temperatures are maintained. Provided the temperature is low, one can argue that an isothermal equation of state is as good as any other, with little feedback from the the thermal pressure on the dynamics.  However, for cloud formation and evolution molecular chemistry and cooling is critical \\citep{Lang00, Lim01, Lim99}. Molecular hydrogen forms most efficiently where the gas is warm but the grains are cool (H$_2$ forms mainly when atoms combine after colliding and sticking to dust grains).  Simple molecules like OH, CO and H$_2$O form in the gas phase with H$_2$ as the reactive agent. These molecules are not only important coolants, but associated emission lines provide a means of measuring the cloud properties. Molecules are dissociated as a consequence of fast shocks, UV radiation, X-rays and cosmic rays \\citep{Herbst00}. We thus need to study molecular turbulence to determine the distribution and abundances of molecular species.  Even in cool optically thin regions, as assumed here, molecular pressures may influence the dynamics. Strong shock waves may be driven from hot atomic regions, formed by shock waves, into the proximity of cold molecular gas. Areas devoid of dust, or in which molecule formation is inefficient (e.g. because atoms will evaporate off warm grains rather than recombine) may attain and maintain pressures as high as turbulent pressures.  The rate of decay of supersonic turbulence is important to the theory of molecular clouds. A possible consequence of the rapid decay of kinetic energy is that the turbulent clouds we observe have short lives. The short lives, however, may not provide sufficient time for relaxation into thermodynamic and chemical equilibrium. Hence, cloud structure and content evolve simultaneously  as a cloud evolves dynamically, rather than proceeding through a series of equilibrium states. Therefore, conclusions based on assuming a molecular fraction to be a function of density alone (e.g. \\cite{Balles99a}) might not always be accurate. We note, however, that the conclusion reached by \\cite{Balles99a} of rapid cloud formation is supported by this study.  Decaying supersonic turbulence is the subject of this initial study. The work is a direct extension, in terms of code and initial conditions, of  isothermal simulations presented by \\cite{M-MML98}. Our goals here are as follows: \\begin{itemize}  \\item To directly compare the decay of the kinetic energy $E_{kin}$ with those derived in the papers of \\citep{M-MML98} and \\citep*{Smith00I, Smith00II}. \\item To determine the distribution of the molecular column density as opposed to the total column density \\citep{VazSem01} \\item To study the rate of formation of molecules. \\end{itemize} We omit in this study magnetic field, self-gravity and photodissociating radiation. We begin with a fully molecular cube of dense gas and apply an initial turbulent field of velocity perturbations. The turbulence was chosen to be sufficient to dissociate the gas in one case (root mean square (rms) speed of 60\\,km\\,s$^{-1}$) and to leave the gas molecular in another case (rms speed of 15\\,km\\,s$^{-1}$). ",
        "conclusions": "\\subsection{Summary}  We  have presented the properties of a specific model for molecular turbulence. We carried out three dimensional hydrodynamical simulations of decaying supersonic turbulence in molecular gas. We included a detailed cooling function, molecular hydrogen chemistry and equilibrium C and O chemistry. We studied three cases in which the applied velocity field straddles the value for which wholesale dissociation of molecules occurs. The parameters chosen ensure that for the high-speed turbulence, the molecules are initially destroyed in shocks and gradually reform in a distinct phase.  We find the following. \\begin{itemize} \\item An extended phase of power-law kinetic energy decay, as in the isothermal case, after an initial phase of slow dissipation and shock formation. \\item  The thermal energy, initially raised by the introduction of turbulence, decays only a little slower than the kinetic energy. \\item The reformation of hydrogen molecules, as the fast turbulence decays, is several times faster than expected from the average density. This is expected in a non-uniform medium, as a consequence of the volume reformation rate being proportional to the density squared. \\item The molecules reform into a pattern of  filaments and small clumps, enveloped in diffuse structure. \\item  During reformation, the remaining turbulence redistributes the gas so that the fraction of molecules is distributed relatively evenly. Hence, the density and molecular density are almost identically distributed at any one time after 100 yr. \\item The correlation length in our simulations grows with time as has been  predicted for a gas in general. \\item The probability density functions sampled from regions with size of about one correlation length are consistent with observations. \\end{itemize} \\subsection{Discussion} We mainly wish here to emphasise the insight these simulations provide into how molecular chemistry and supersonic dynamics combine. We have found that isothermal simulations are indeed very useful, not only for the rate of energy decay but also to trace the molecules.  A simple reason for the fast decay is that a sufficient number of strong shocks survive. As shown by \\cite{Smith00I}, the rate of energy decay in {\\em decaying} turbulence is dominated by the vast number of weak shocks. These shocks are less efficient at energy dissipation. A second possible reason is that the curved shock structures create small scale vorticity, which leads to enhanced dissipation of kinetic energy. It is not clear, however, why relatively more vorticity should be created when stronger shocks are present. Simulations of isolated curved shocks would help clarify the dissipation paths.  It is plausible that the high Mach number turbulence could create more small-scale structure, leading to a faster decay rates as argued by \\cite{M-MML99}.  The simulations presented here may directly aid our interpretation of fast molecular shocks in dense regions, such as Herbig-Haro objects. For example, supersonic decay would appear to occur in the wake of bow shocks such as in DR~21 \\citep{Davis96}. These simulations suggest that the turbulence decays rapidly and bow shock wakes will be bright but short.  The fast reformation of molecules indicates that molecules may form on shorter time scales than previously envisaged. Hence molecular clouds may appear out of atomic clouds several times faster than anticipated in non-turbulent scenarios. Recently, some evidence has been found for rapid cloud formation and dissipation \\citep{Balles99a, Hart01}. This implies that much of the interstellar medium may be undergoing both rapid dynamical and chemical changes, driven by sources of supersonic turbulence.  The simulations analysed here provide a basis for much further work. For example, we can now investigate the properties of hydrodynamic clouds in which dust or chemical abundances are non-uniform, clouds in which the turbulence is driven uniformly or non-uniformly, and clouds which are initially atomic."
    },
    "0208/astro-ph0208546_arXiv.txt": {
        "abstract": "An open question in the field of solar and stellar astrophysics is the source of heating that causes stellar coronae to reach temperatures of millions of degrees.  One possibility is that the coronae are heated by a large number of small flares.  On the Sun, flares with energies as low as those of microflares are distributed with energy as a power\\,law of the form $\\frac{dN}{dE} \\propto E^{-\\alpha}$ with $\\alpha\\approx1.8$, and $\\alpha$ appears to increase to values 2.2-2.7 for flares of lower energy.  If the slope exceeds the critical value of 2, then in principle the entire coronal energy input may be ascribed to flares that are increasingly less energetic, but are more numerous.  Previous analyses of flares in light-curves of active stars have shown that this index is generally $>2$, though it may be as low as 1.6 when strong flares alone are considered.  Here we investigate the contribution of very weak flares, covering the milliflare energy range, to the coronal luminosity of low-mass active stars. We analyze {\\sl EUVE}/DS events data from \\fkaqr, \\wolf, and \\adleo\\ and conclude that in all these cases the coronal emission is dominated by flares to such an extent that in some cases the entire emission may be ascribed to flare heating. We have developed a new method to directly model for the first time stochastically produced flare emission, including undetectable flares, and their effects on the observed photon arrival times. We find that $\\alpha_{\\rm\\fkaqr} = 2.60 \\pm 0.34$, $\\alpha_{\\rm\\wolf} = 2.74 \\pm 0.35$, $\\alpha_{\\rm\\adleo} = 2.03 - 2.32$, and that the flare component accounts for a large fraction (generally $>50\\%$) of the total flux. ",
        "introduction": "\\label{s:intro}  The source of heating of solar and stellar coronae still eludes understanding even after decades of study (e.g., Schrijver et al.\\ 1999). Despite significant evidence that magnetic activity is the prime driver for transferring energy into the corona, the mechanism by which this transfer occurs is not established in either the case of the Sun or other stars (see e.g., Rosner, Golub, \\& Vaiana 1985, Narain \\& Ulmschneider 1996). Numerous heating mechanisms, such as acoustic wave dissipation (Stepien \\& Ulmschneider 1989), Alfv\\'en wave dissipation (Cheng et al.\\ 1979, Narain \\& Ulmschneider 1990), magnetic reconnection phenomena (Parker 1988, Lu \\& Hamilton 1991) have been proposed, all of which might play some role in the overall heating.  Recent work in the solar case has lent strong credence to the possibility of coronal heating being dominated by small-scale explosive events suggestive of Parker's {\\sl nanoflare} model, which is based on magnetic reconnections releasing energies $\\sim 10^{24}$ erg event$^{-1}$.  It has been well known that solar microflares and milliflares\\footnote{ Because of the vast range of flare energies encountered, the energy ranges of the different flare types are not well defined. We adopt the convention (see e.g., Aschwanden et al.\\ 2000) that milliflares cover the range $E \\sim 10^{29-32}$ ergs, microflares $E \\sim 10^{26-29}$ ergs, and nanoflares $E \\sim 10^{23-26}$ ergs. We consider all events down to the microflare regime to be `normal' X-ray flare events, with similar origin, parameters, and effects, except for the differences in energy deposition. } are distributed in number as power\\,laws of their energy output (Lin et al.\\ 1984, Hudson 1991), \\begin{equation} \\label{e:powerlaw} \\frac{dN}{dE} = k E^{-\\alpha} \\end{equation} where $E$ is the energy of the flare and $k$ is a constant.  This relation has been verified and extended to lower energies by various authors, but despite the near universal acceptance of the form of the function in Equation~\\ref{e:powerlaw} (e.g., Crosby, Aschwanden, \\& Dennis 1993; see Kopp \\& Poletto 1993, Shimizu \\& Tsuneta 1997 for a different perspective), neither the index $\\alpha$ nor the normalization $k$ are well determined. For instance, $\\alpha = 1.6-1.8$ in the HXR to microflare energy range, and is variously measured to lie in the range 1.8-2.9 at lower energies (Shimizu 1995 [$\\alpha=1.5-1.6$], Porter, Fontenla, \\& Simnett 1995 [$\\alpha=2.3$], Krucker \\& Benz 1998 [$\\alpha=2.3-2.6$], Parnell \\& Jupp 2000 [$\\alpha=2.0-2.6$], Aschwanden et al.\\ 2000 [$\\alpha=1.8$], Winebarger et al.\\ 2001 [$\\alpha=2.9 \\pm 0.1$], Veronig et al.\\ 2002 [$\\alpha=2.03 \\pm 0.09$]). Recently Aschwanden \\& Parnell (2002) have used scaling laws based on energy balance arguments to conclude that $\\alpha$ must be $\\sim 1.6$ on the Sun. The precise value of $\\alpha$ is of considerable interest because if the power\\,law is steep enough ($\\alpha>2$), then in principle a multitude of small impulsive events would be sufficient to account for the energy output of the entire corona.  Here we reconsider in particular an outstanding question in stellar X-ray astronomy, which is the nature of the apparently quiescent emission from active stars: does this emission actually arise from a superposition of a multitude of impulsive events (such as milliflares and microflares), or from truly quiescent plasma?  Previous work based on detecting flares in EUV data (see e.g., Audard et al.\\ 2000) suggests that flare contribution is indeed an important factor.  Further, correlations of quiescent  X-ray flux with time-averaged U-band flare flux (Skumanich 1985, Doyle \\& Butler 1985) and the synchrotron radio luminosity (G\\\"{u}del \\& Benz 1993), together with the similar correlations found in the solar case (Benz \\& G\\\"{u}del 1994) strongly suggest a link between the apparently quiescent emission and flares.  In addition, spectroscopic evidence for high temperature plasma ($T \\gtrsim 10^7$ K) during the quiescent phase (Butler et al.\\ 1986, Kashyap et al.\\ 1994, Drake 1996, G\\\"{u}del et al.\\ 1997, G\\\"{u}del 1997) indicates that this quiescent emission could in fact be very similar to flare emission in origin.  Thus, apparently quiescent coronae of active stars could be composed of a continuum of small unresolved flares, presumably distributed as power\\,laws analogous to the Sun.  This view is also supported by the double-peaked Differential Emission Measures (DEMs) that result when an ensemble of flaring, hydrodynamically evolving loops are modeled on active solar analogs (G\\\"{u}del et al.\\ 1997, G\\\"{u}del 1997).  The possibility of stellar coronal heating due to small flares was considered by Ambruster, Sciortino, \\& Golub (1987) who searched for variability in {\\sl Einstein} data of active stars and discussed the contribution of low-level flaring to heating stellar coronae. They concluded that while flaring must contribute at some level, the evidence does not justify extending the solar power-law distributions to the stellar microflare case.  Later studies of ensembles of strong stellar flares seen with {\\sl EXOSAT} and {\\sl EUVE} have shown these are distributed as power\\,laws with index $\\alpha=1.6-1.8$ (Collura et al.\\ 1988, Pallavicini et al.\\ 1990, Osten \\& Brown 1999), thus ruling out low-intensity flares as a significant contributor to the heating budget. In contrast, using a more sensitive method to detect fainter flares (see Crawford et al.\\ 1970), Robinson et al.\\ (1995, 1999, 2001) find that for stellar chromospheric and transition region events observed with the high-speed photometer and the imaging spectrograph on the {\\sl HST}, $\\alpha \\sim 1.76-2.17$ in the chromosphere of the active dMe star CN\\,Leo; $\\alpha=2.25 \\pm 0.1$ in the chromosphere of the dMe star YZ\\,CMi; and $\\alpha \\sim 2.2-2.8$ in the transition region of the dM0e flare star AU\\,Mic. (Note however that chromospheric and transition-region flare distributions have no known direct correspondence with the coronal case.) Applying a similar method to {\\sl EUVE}/DS data, and also correcting for overlaps in flares, Audard et al.\\ (1999) find that for solar analogs EK\\,Dra and 47\\,Cas, $\\alpha\\approx~2.2\\pm0.2$. This analysis was further extended by Audard et al.\\ (2000) to a larger sample of cool stars, and they find that $\\alpha$ ranging from 1.5 to 2.6, with the majority of the measurements having $\\alpha>2$. Similar results are obtained for AD\\,Leo (G\\\"{u}del et al.\\ 2001,2002).  Note that the above studies are limited to relatively large flares ($E \\gtrsim 10^{31}$ ergs) because of instrument sensitivity, and also because the more numerous weaker flares are harder to detect in the presence of ``contamination'' by other weak flares.  Thus, the low-energy end of the flare distribution is subject to large uncertainties.  We have developed a new method to {\\sl model} the undetectable stellar flares and thus derive estimates of flare indices covering the milliflare regime as well as directly estimating the flare contribution to the observed flux.  We apply this method to active low-mass stars \\fkaqr, \\wolf, and \\adleo. The datasets used are described in \\S\\ref{s:data}. The analysis method is detailed in \\S\\ref{s:analyz} (a glossary of the terms used is given in Appendix~\\ref{s:glossary}). The results are given in \\S\\ref{s:result}, and are summarized in \\S\\ref{s:summary}. ",
        "conclusions": "\\label{s:summary}  We have modeled the event arrival times from active stars \\fkaqr, \\wolf, and \\adleo\\ with particular attention to the component that arises from flare-like events.  On the Sun, flares are known to be distributed as a power\\,law in energy (Hudson 1991; also see Aschwanden et al.\\ 2000 and references therein), and numerous studies have established that strong stellar flares also follow a power\\,law distribution, with indices ranging from 1.5 - 2.5 (see Audard et al.\\ 2000 and references therein).  This is of considerable interest because if the power\\,law index $\\alpha$ is $>2$, then the coronal X-ray losses could in principle be ascribed to weak flare events that are nevertheless numerous enough to dominate the emission.  We consider active dMe stars with known variability in their light curves where numerous flares are seen.  Audard et al.\\ (2000) find that in general dF and dG stars tend to have $\\alpha>2$ while dK and dM stars tend to have lower $\\alpha$.  In particular, they analyze an older dataset of \\adleo\\ (from May 1996) and show that detectable flares have $\\alpha \\in [1.18,2.35]$ or $\\alpha=2.02\\pm0.28$ using different methods.  A more detailed analysis by G\\\"{u}del et al.\\ (2002) based on a larger sample of the dataset used here shows $2.0<\\alpha<2.5$ .  We model the event arrival times using a simple two-component model comprising of a constant rate component and a statistical ensemble of flare components, with the flare energies distributed as a power\\,law. In general, the simplifying assumptions we make (e.g., constancy of decay timescales, ignoring the rise times, assuming a constant counts-to-energy conversion factor, including the background directly in the model, etc.) are conservative, and tend to underestimate the value of $\\alpha$.  We find that all the stars in our sample clearly have $\\alpha>2$: for \\adleo, $\\alpha$ lies in the range 2.06 - 2.32; for \\fkaqr, $\\alpha=2.60\\pm0.34$; and for \\wolf, $\\alpha=2.74\\pm0.35$. We thus conclude that coronal heating on these stars is dominated by impulsive energy release events whose energy output is $\\gtrsim~2-3~\\times~10^{29}$ ergs, reaching to the microflare range. These results are in contrast to the solar case, where over similar flare energy ranges the observed distribution of flares is shallower, with $\\alpha \\approx 1.8$, i.e., below the critical value of 2.  Further, we directly estimate the contribution of the flare emission to the total observed count rate, as one of the parameters defining the model.  Because the energy range over which the model is defined spans over 4 orders of magnitude, and the power\\,law indices indicate steep distributions, we expect that the flare component should contribute significantly to the total emission.  Indeed, we find this to be generally $>50\\%$, and in some cases being $>80\\%$.  Thus, there appears to be no truly ``quiescent'' emission on some of these low-mass active stars, i.e., emission from apparently stable active region loops as on the Sun is not a dominant component of the observed emission.  We have also explored the possible dependence of the various model parameters on flare energies.  The long observation of \\adleo\\ suggests that $\\alpha$ increases when strong flares are not evident in the data suggesting that the flare distribution steepens for flares of smaller energies, though we cannot rule out the possibility of a statistical fluctuation that mimics this trend. We have also searched for, but do not find, evidence of strong dependence of the decay timescale on flare energies.  We find that if the flare distributions extend to the microflare regime, the energy output due to these weak flares is quite sufficient to account for the entire coronal emission in the {\\sl EUVE}/DS passband. However, it must be noted that the error bars on the parameters derived for the fainter stars \\fkaqr\\ and \\wolf\\ are quite large (for instance, values of $\\alpha<2$ cannot be completely ruled out for \\fkaqr, and a firm upper bound on $\\alpha$ cannot be set for \\wolf) and it would be of considerable interest to verify and improve these results (and also to extend them to a larger sample of stellar types) using high-quality data such as those obtainable with {\\sl Chandra} and {\\sl XMM-Newton}. Data from these observatories are characterized by good time resolution and in general larger count rates, and will therefore allow us to explore the arrival time difference distribution functions $g(\\tdel)$ at smaller values of $\\tdel$, thereby extending the range of values of $\\alpha$ that the method is sensitive to."
    },
    "0208/astro-ph0208037_arXiv.txt": {
        "abstract": "s{This is the dawning of the age of precision cosmology, when all the important parameters will be established to one significant figure or better, within the cosmological model. In the age of accurate cosmology the model, which nowadays includes general relativity theory and the CDM model for structure formation, will be checked tightly enough to be established as a convincing approximation to reality. I comment on how we might make the transition. We already have some serious tests of gravity physics on the length and time scales of cosmology. The evidence for consistency with general relativity theory is still rough, but impressive, considering the enormous extrapolation from the empirical basis, and these probes of gravity physics will be considerably improved by work in progress on the cosmological tests. The CDM model has some impressive observational successes too, and some challenges, not least of which is that the model is based on a wonderfully optimistic view of the simplicity of physics in the dark sector. I present as a cautionary example a model for dark matter and dark energy that biases interpretations of cosmological observations that assume the CDM model. In short, cosmology has become an empirically rich subject with a well-motivated standard model, but it needs work to be established as generally accurate.} ",
        "introduction": "Our colleagues in the more exact sciences distinguish the precision of a measurement, which is indicated by the number of significant figures, from the accuracy, which is what remains after due account of the interference by systematic errors. In cosmology we have to worry about systematic errors in the astronomy and, it is less commonly emphasized, in the physics. In the standard cosmology the latter includes general relativity and the rest of textbook physics, along with the cold dark matter model for structure formation. All this physics is a considerable extrapolation from the empirical basis. This means that, unlike the standard model for particle physics, in cosmology it is not a matter of measuring parameters in a reliably established theory: we have to check the physics too.  We have checks of the physics and the astronomy, from the growing network of cosmological tests. For example, the SNeIa redshift-magnitude measurements, combined with the CDM model interpretation  of the anisotropy of the 3~K cosmic background radiation (the CBR), indicate the mean mass density, $\\rho _m$, in low pressure matter is about one quarter of the critical Einstein-de Sitter value. A similar number follows from most dynamical analyses of galaxy peculiar velocities. These two approaches depend on very different astronomy, and they apply quite different aspects of the physics of the relativistic Friedmann-Lema\\^\\i tre cosmology. If the physics or the astronomy failed, the consistency of these estimates of $\\rho _m$ would seem  unlikely. Accidental coincidences do happen, of course, and we have to remember the natural human tendency to stop working so hard on an analysis when it approaches the wanted answer. Thus it is important that similar estimates of $\\rho _m$ follow  from still other lines of evidence: weak gravitational lensing, the baryon mass fraction in clusters of galaxies, the abundance of clusters as a function of mass and redshift, and the power spectrum of the galaxy distribution. If this concordance survives further scrutiny that explains the remaining discrepant indications, for significantly larger and smaller values of $\\rho _m$, it will eliminate the hypothesis of canceling errors.  What do we learn from this evidence for concordance? It certainly encourages the view that the mass density parameter, $\\Omega _m=8\\pi G\\rho _m/3H_o^2$, is a physically meaningful number, and that we know its value to a factor of two or so. But it is useful to be more specific, by considering what aspects of gravity physics are probed by this concordance, and by all the other cosmological tests. I review four tests of gravity physics on the scales of cosmology in Sec.~2. All agree with GR so far. This is not surprising: we have no substantial reason within fundamental theory (apart maybe from brane worlds) to suspect GR fails on cosmological length scales. But positive empirical evidence is the thing.  Many of the cosmological tests assume the CDM model for structure formation. Is this model adequate for precision cosmology at the ten percent level? I discuss aspects of this issue in Sec.~3. The estimates of $\\Omega _m$ based on CDM generally agree with independent indications from dynamics, and the successful fit to the 3~K CBR temperature anisotropy is also impressive. This is serious evidence that the CDM model is a useful approximation to reality. But the present precision of the evidence allows considerably more complicated physics in the dark sector, and more complicated physics may be indicated by the observational challenges from galaxy structure and formation. In short, significant adjustments to the CDM model would not be surprising, and a major shift not inconceivable. Until this is sorted out structure formation is a hazardous basis for cosmological tests.  These topics are discussed at length, with many references, in a paper with Bharat Ratra (in astro-ph/0207347). I refer the reader to this paper for details and references. Here I indulge in the luxury of a reference-free overview. ",
        "conclusions": "It is standard and efficient practice to stick with the theory that has brought us this far until it fails. Experience reenforces the strategy; GR is a good example. Einstein's modest empirical basis came from laboratory physics: Maxwell's equations, that contain special relativity, and the evidence for the equivalence principle. Beginning with Einstein's calculation of the precession of the perihelion of Mercury, GR has been shown to pass searching tests out to the much larger scales of the Solar System. We are now seeing that the theory passes nontrivial tests on the enormous scales of cosmology. One might argue that this is to be expected, from the compelling physical logic of GR. I respect the logic, but am much more impressed by the prospect of actually weighing the physics of GR on the observational scales of cosmology.  The physics of the CDM model for structure formation is not as logically compelling as GR, as witness the broad interest in the warm and self-interacting variants. Alternatives that upset the cosmological tests are less widely discussed, but certainly will be useful, maybe as foils to help establish the CDM model, maybe as leads to better physics in the dark sector.  We may be lucky enough to get a laboratory detection and exploration of the properties of dark matter, but most of the physics in the dark sector and the rest of cosmology will have to be established in quite indirect ways, like much of physical science these days. One way to organize this follows the PPN approach to tests of GR: assign parameters to the aspects of gravity physics that are of interest to cosmology, as discussed in Sec.~2, other parameters for such physics in the visible sector as rolling coupling constants, more parameters for physics in the dark sector, and still more for initial conditions. Overconstraining them all will be quite a challenge, but Nature has provided opportunities for lots of observations, the pursuit of which we may hope will show us when we have arrived at the dawning of the age of accurate cosmology."
    },
    "0208/astro-ph0208201_arXiv.txt": {
        "abstract": "Using  the   HST  WFPC2  we   perform  deep  $I$-band  imaging   of  9 radio-selected ($\\rm F(8.5~GHz)\\geq 14\\mu \\rm Jy$) faint galaxies from the  Roche, Lowenthal and  Koo (2002)  sample. Two are also observed in $V$ using HST STIS.  Six of the galaxies have  known redshifts, in the range $0.4<z<1.0$. Radial intensity profiles indicate that 7 are disk galaxies and 2 are bulge-dominated. Four of the six with redshifts have a high optical  surface brightness in comparison with previous studies of disk galaxies at similar redshifts (e.g. Lilly et al. 1998). The HST imaging reveals that 2 of the 9 galaxies are in close interacting pairs and another 5 show morphological evidence of recent interactions -- two are very asymmetric ($A_{asym}\\sim 0.4$) and, three have large, luminous rings resembling the collisional starburst rings in the Cartwheel galaxy. For the two ring galaxies with redshifts, we measure ring radii of 7.05 and 10.0 $h^{-1}_{50}$ kpc, which suggest post-collision ages 0.1--0.2 Gyr. One has a fainter inner ring, like the original Cartwheel. The remaining two appear to be late-type  barred galaxies  and relatively undisturbed.     Our HST imaging confirms the high incidence of interactions and dynamical disturbance in faint  radio-selected galaxies, as  reported by e.g.   Windhorst et al.  (1995) and  Serjeant. et  al.  (2000).  In the great majority of  these galaxies the high radio luminosities are probably  the result of interaction-triggered starbursts. However, one interacting galaxy is a very radioluminous giant elliptical, with red $V-I$ colours, a normal surface  brightness  and no  evidence of  star-forming regions, so its radio source is probably an obscured AGN. The mixture of observed morphologies suggests that enhanced radio luminosities often persist to a late stage of interaction, i.e. at least   $\\sim 0.2$  Gyr  after the  perigalactic encounter. ",
        "introduction": "The radio emission from star-forming galaxies is of great importance as an  apparently unbiased tracer of the star-formation rate (SFR), being neither obscured nor enhanced by the presence of dust (e.g. Condon 1992). At sub-milliJansky fluxes, deep radio surveys detect large number of star-forming galaxies, which overtake radio galaxies and other active galactic nuclei (AGN) as the most numerous type of source. Hence a  deep radio-selected source sample will efficiently select the most intensely star-forming galaxies in the survey volume.  Hammer et al. (1995) investigated  the optical IDs of a sample of radio sources detected in  a deep  Very Large Array  (VLA) survey of one of the  Canada-France Redshift  Survey fields, to a limit in integrated flux $F(5.0~\\rm GHz)\\geq  16\\mu Jy$. Spectroscopy revealed these galaxies to be  mixture of  (i) giant ellipticals,  presumably  with AGN,  (ii)  star-forming disks at redshifts out to $z=1.16$, and (iii) lower redshift, high-excitation emission-line galaxies. Windhorst et al. (1985) and Richards et al. (1998) obtained similar findings -- although with fewer AGN and more disk galaxies -- at even fainter flux limits. Using HST imaging, they estimated that at least  $\\sim 60$ per cent  of the radio-selected galaxies were interacting.  In a previous paper (Roche, Lowenthal and Koo 2002; hereafter RLK02) we identified the most probable optical IDs for 50 out of 51 $F(8.5~\\rm GHz)=7$--900 $\\rm \\mu Jy$ sources, detected in three deep VLA surveys. These galaxies were imaged  in $BRI$ using the Keck Low Resolution Imaging Spectrograph (LRIS), and in $K$ with the Near Infra-red Camera (NIRC). Spectra and redshifts were obtained for 17, using LRIS in spectroscopic mode. Of these, 14 were found to be emission-line galaxies at redshifts up to $z\\sim 1$, and the other 3 were QSOs at $z>1$. A further 9 of the radio ID galaxies had known redshifts from previous observations, giving a total of 26 with known redshifts. RLK02 concluded that more than half of these 26  were disk galaxies with enhanced radio luminosities, resulting from major starbursts ($\\rm SFR\\sim 100 M_{\\odot}yr^{-1}$). The remainder were apparently normal, non-starburst spirals and ellipticals at lower redshifts ($z<0.4$) ($\\sim 20$ per cent), QSOs ($\\sim 15$ per cent), or giant radioluminous ellipticals suspected to contain obscured AGN ($\\sim 8$ per cent).   The LRIS spectra of the 14 non-QSO  galaxies showed  the expected emission lines, $\\rm [OII]3727\\AA$, $H\\beta$ and $\\rm [OIII]5007\\AA$, but the line luminosities typically correspond to SFRs an order of magnitude lower than the radio luminosities imply. Other surveys of radio-selected galaxies have found a similar discrepancy between emission-line and radio fluxes  (e.g. Smith et al. 1996;  Beck, Turner and Kovo  2000; Serjeant et al. 2000). The simplest explanation is a high dust extinction ($A_V\\simeq 2$--3 mag) of the starburst regions. We also found that  11 of the 26 galaxies  were in  close  pairs and several others looked disturbed. However, it was clear that to adequately investigate the interaction status of these galaxies,  much higher resolution imaging, e.g WFPC2, would be required.  Serjeant et al. (2000) performed an WFPC2 imaging survey of radio-selected star-forming galaxies, and describe the first four. These galaxies are at lower redshifts, $z\\sim 0.2$, than the RLK02 sample but have comparable radio luminosites. They found at least one galaxy to be  interacting and all 4 to be disturbed to some degree, and quantified morphological disturbance in terms of a rotational asymmetry parameter $A_{asym}$ (Conselice, Bershady and Jangren 2000). They concluded that radio-selected galaxies are generally more asymmetric than optically-selected galaxies at similar magnitudes. Similarly, and at higher redshifts, Fomalont et al. (2002) identified 37/63 faint $\\rm F(8.5~GHz)\\geq7.5\\mu \\rm Jy$ radio sources with $I\\leq 23.3$ galaxies, and using WFPC2 imaging found a high proportion, 46 per cent, to be multiple or interacting.  In this paper we perform a similar WFPC2 study of a subsample of the RLK02 radio IDs. Our HST program was allocated a total of 24 orbits, which was devoted to WFPC2 $I$-band imaging of four fields, containing a total of 9 of the RLK02 galaxies. For the first two fields we imaged in  parallel with STIS, and obtained images in the broad STIS $V$-band of two of the galaxies. Section 2 of this paper describes the observations and data reduction. Section 3 catalogs the observed galaxies and presents HST images. In Section 4, radial intensity profiles are presented, model profiles fitted, and   surface brightness compared with local galaxies. In Section 5 the evidence for interactions is investigated. Asymmetry parameters are evaluated and model profiles subtracted from the galaxies to highlight `residual' features. In Section 6, galaxy colours are interpreted.  Section 7 is a discussion of the nature of these galaxies and the source of their radio luminosity. ",
        "conclusions": "\\subsection{Origin of the Radio Emission}  S4, the most radioluminous, is a giant elliptical, with a much fainter blue disk companion. The colours and SB are consistent with a passively evolving elliptical and there are no obvious star-forming regions. We find no indication that it is undergoing the extreme  star formation -- at least $\\sim 1000 M_{\\odot}$ $\\rm yr^{-1}$ -- needed for this to account for its radio luminosity. The radio emission is probably from an obscured AGN, and it seems plausible that the interaction with the companion triggered the current  nuclear activity.  S10, the second most radioluminous, is a very different object, blue in colour with a very asymmetric disk profile. The bright knots, high  optical SB and  strong emission lines are evidence for an extensive starburst. It has a  much redder nucleus, suggesting it was initially a spiral. Interaction-triggered star-formation can probably account for the high radio luminosity.  16V21 is also a highly disturbed disk galaxy with bright knots, and may be interacting or merging.   16V22 is a sub-$L^*$ galaxy of estimated type $\\sim \\rm SBm$. The SB is near average for a disk of its redshift, and is not visibly interacting or disturbed. However, very blue colours,  strong $\\rm [OIII]5007 \\AA$ emission, and a high excitation ratio,  $F({\\rm [OIII]})/F({\\rm H\\beta})=2.93\\pm 0.14$ (RLK02) imply starbursting, or possibly an AGN. The deep radio sample of Hammer et al. (1995) contained some similar high-excitation, sub-$L^*$  galaxies, which were of uncertain nature having line diagnostics near the HII/Seyfert/LINER divide. In the case of 16V22, we find no sign  of a central point source on our image, so it seems most likely that star-formation accounts for the radio emission.  16V25 appears to be a barred $\\sim\\rm SBbc$ spiral , with a high disk and central SB. The radio emission may be due to strong star-formation, possibly enhanced  by a minor interaction as the central bar appears asymmetric. It is also possible that some of the emission is from an AGN, but any active nucleus must be heavily obscured as there is no visible point-source (the redshift was too high for RLK02 to obtain an excitation ratio).  16V26 appears to be a disturbed disk galaxy, like S10, and on the basis of its faintness and colours is likely to be at $z>1$.  16V30 could be (i) a very early-type spiral with radio emission from a collisional starburst ring, or (ii)  a giant elliptical that has accreted a smaller galaxy, forming a ring. Of these 9 galaxies, it is the second brightest in $I$ but has the lowest radio flux. The redshift is unknown but if the optical luminosity is high ($M_B\\sim-22.5$) the radio emission would be consistent with a normal giant elliptical, following the Sadler, Jenkins and Kotanyi (1989) relation.  16V31 and 16V34 are disks with rings resembling those in the Cartwheel galaxy, and on the basis of the high SFR and radio flux of the original Cartwheel, it is highly probable that  collisional starbursting in these rings accounts for their radio luminosities.  \\subsection{The Role of Interactions} Our  HST imaging survey confirms the high incidence of interactions amongst faint radio-selected galaxies, as previously reported by e.g.  Windhorst et al.  (1995) and Serjeant. et al.  (2000), with the strong radio emission resulting from starbursting in the majority of galaxies (at least 6 of this sample of 9) and obscured AGN in a significant minority (at least 1/9 here). Probably the  most interesting new findings are the Cartwheel-type rings in at least 2 and possibly 3 of these galaxies.  These rings, together with  the asymmetry of  S10  and 16V26, imply that a rather large fraction, 5/9, of the sample are post-encounter interacting galaxies, whereas  only two appear to be in pre-encounter pairs. This can be compared with the ULIRGs ($L_{60\\rm \\mu m}>10^{12}h^{-2}_{50}L_{\\odot}$ galaxies) of  Clements et al. (1996),  of which 55/60 are visibly disturbed or merging, and double nuclei could be resolved in 28, i.e. about equal numbers are observed    before    and   after    nuclear coalescence/collision. A high post-encounter fraction amongst radio-selected galaxies might be expected on the basis of a model of  Lisenfeld et al. (1996), in which the synchrotron radio emission from a starburst remains strong for at least $\\sim 80$ Myr after star formation ceases. This also means that  galaxies that  have already been starbursting for $>80$ Myr would have higher radio luminosities per unit SFR, and higher radio/FIR ratios, than near the onset of star formation.   16V31 has strong $\\rm H\\delta$ absorption which could be fit by a Delgado et al. (1999) post-starburst model at age $\\sim 27$ Myr, but could also result from `age-dependent extinction', with starbursting ongoing for $>10^8$ yr with only the most recent star-formation ($<10^7$ yr)   obscured by dust (Poggianti, Bressan and Franceschini 2001). The LRIS spectroscopy of RLK02 did not provide the line ratios, e.g. the Balmer decrement, needed to distinguish these possibilities. However, the  close resemblance to the Cartwheel is an important clue, as the Cartwheel is known  to have a very high current SFR combined with $A_V\\simeq 2$ mag extinction of the starburst emission lines. A similar picture is seen for the other 8 collisional ring galaxies studied by Bransford et al. (1998), including two with strong Balmer absorption lines.  This is significant, as prolonged starbursting with age-dependent extinction can (in addition to producing strong Balmer lines), explain the very low emission line to radio flux ratios of $\\rm \\mu Jy$ radio-selected galaxies in general (see RLK02).  Collisional ring galaxies may prove to be quite prominent in deep  $\\rm \\mu Jy$ radio surveys,  as they are more numerous at higher redshifts (Lavery et al. 1996) and the long synchrotron lifetime  would favour the radio-detection of post-encounter, $\\geq 10^8$ yr age,  prolonged  starbursts, of which Cartwheel-type galaxies are  prime examples.  \\subsection*{Acknowledgements}Based on observations with the NASA/ESA Hubble Space Telescope obtained at the Space Telescope Science Institute, which is operated by the Association of Universities for Research in Astronomy, Incorporated, under NASA contract NAS5-26555. . DCK was supported by an NSF PYI grant AST-8858203 and a research grant from UC Santa Cruz. NR is currently supported by a PPARC research associateship. We thank Matthew Bershady and Christopher Conselice for providing their software for calculating galaxy asymmetries."
    },
    "0208/astro-ph0208171_arXiv.txt": {
        "abstract": "Spectral fitting of the radio through hard X-ray emission of BL~Lac objects has previously been used to predict their level of high-energy (GeV -- TeV) emission. In this paper, we point out that such spectral fitting can have very large uncertainties with respect to predictions of the VHE emission, in particular if no reliable, contemporaneous measurement of the GeV flux is available and the $\\nu F_{\\nu}$ peak (flux and frequency) of the synchrotron component is not very precisely known. This is demonstrated with the example of the radio-selected BL~Lac object W~Comae, which is currently on the source list of the STACEE and CELESTE experiments, based on extrapolations of the EGRET flux measured from this source, and on model predictions from hadronic blazar jet models. We show that the best currently available contemporaneous optical -- X-ray spectrum of W Comae, which shows clear evidence for the onset of the high-energy emission component beyond $\\sim 4$~keV and thus provides a very accurate guideline for the level of hard X-ray SSC emission in the framework of leptonic jet models, still allows for a large range of possible parameters, resulting in drastically different $> 40$~GeV fluxes. We find that all acceptable leptonic-model fits to the optical -- X-ray emission of W Comae predict a cut-off of the high-energy emission around $\\sim 100$~GeV. We suggest that detailed measurements and analysis of the soft X-ray variability of W Comae may be used to break the degeneracy in the choice of possible fit parameters, and thus allow a more reliable prediction of the VHE emission from this object. Using the available soft X-ray variability measured by {\\it BeppoSAX}, we predict a $> 40$~GeV flux from W~Comae of $\\sim$~(0.4 -- 1)~$\\times 10^{-10}$~photons~cm$^{-2}$~s$^{-1}$ with no significant emission at $E \\gtrsim 100$~GeV for a leptonic jet model. We compare our results concerning leptonic jet models with detailed predictions of the hadronic Synchrotron-Proton Blazar model. This hadronic model predicts $> 40$~GeV fluxes very similar to those found for the leptonic models, but results in $> 100$~GeV emission which should be clearly detectable with future high-sensitivity instruments like VERITAS. Thus, we suggest this object as a promising target for VHE $\\gamma$-ray and co-ordinated broadband observations to distinguish between leptonic and hadronic jet models for blazars. ",
        "introduction": "Introduction}  After the detection of 6 high-frequency peaked BL~Lac objects (HBLs) with ground-based air \\v Cerenkov telescope facilities, the field of extragalactic GeV -- TeV astronomy is currently one of the most rapidly expanding research areas in astrophysics. The steadily improving flux sensitivities of the new generation of air \\v Cerenkov telescope arrays \\citep{konopelko99,weekes02} and their decreasing energy thresholds, provides a growing potential to extend their extragalactic-source list towards intermediate and even low-frequency peaked BL~Lac objects (LBLs) with lower $\\nu F_{\\nu}$ peak frequencies in their broadband spectral energy distributions (SEDs). Detection of such objects at energies $\\sim 40$ -- 100~GeV might provide an opportunity to probe the intrinsic high-energy cutoff of their SEDs since at those energies, $\\gamma\\gamma$ absorption due to the intergalactic infrared background is still expected to be negligible at redshifts of $z \\lesssim 0.2$ \\citep{djs02}.  Theoretical predictions of the high-energy emission of BL~Lac objects on the basis of their emission at lower frequencies \\citep{stecker96,cg02} are essential for careful planning of future observations by ground-based VHE $\\gamma$-ray observatories. Such studies have generally been restricted to considerations of the observed broadband spectral properties of potential candidate sources and have mostly been based on non-simultaneous spectral measurements alone. In this paper, we point out that such considerations can have very large uncertainties and ambiguities with respect to the predicted VHE emission. The importance and potential scientific return of including detailed variability information in the modeling of HBLs has been pointed out by \\cite{ca99} who have shown the wide variety of correlated X-ray and VHE $\\gamma$-ray variability patterns which can result in time-dependent synchrotron-self-Compton models for blazars. In particular, they have pointed out that combined X-ray spectral and variability information may be sufficient to predict the level of intrinsic TeV emission in HBLs, even if no direct measurements at GeV -- TeV energies are available. Here, we will demonstrate that similar conclusions hold for LBLs, and investigate the example of the radio-selected BL Lac object W~Comae (= ON~231 = 1219+285; z = 0.102), which is currently on the source list of the STACEE and CELESTE experiments. Its GeV -- TeV source candidacy is based on the fact that W~Comae has been detected by the EGRET instrument on board the {\\it Compton Gamma-Ray Observatory} at energies above 100~MeV, exhibiting a very hard spectrum \\citep{vm95,sreekumar96}. A power-law extrapolation of the average EGRET 0.1 -- 10~GeV flux into the multi-GeV -- TeV range yields a VHE flux well above the current detection threshold of both STACEE and CELESTE (see Fig. \\ref{sscgraph}). \\cite{db01} also report on the detection of a 27.3~GeV photon from this source by EGRET in April 1993. Furthermore, \\cite{mannheim96} has predicted a TeV flux near the detection limit of the Whipple air \\v Cerenkov telescope at the time, based on a proton-blazar model fit to a non-simultaneous broadband spectrum of W~Comae.  The source was observed in a multiwavelength campaign in February 1996, covering the electromagnetic spectrum from GHz radio frequencies to TeV energies \\citep{maisack97}. No TeV emission was detected by either Whipple or HEGRA.  While W~Comae is generally observed to exhibit a typical one-sided jet morphology in VLBI images, \\cite{massaro01} report the detection of a weak apparent counter-jet component in 1999.13, if the brightest jet component with the flattest radio spectrum is identified with the core. Such a feature has not been found in any previous or later radio maps of the source. \\cite{massaro01} demonstrate that it is implausible that this component is actually the emission from the counter-jet. Alternatively, they suggest, e.g., that it could be due to a small-angle displacement of the jet direction in a general configuration in which the jet is directed at a very small average angle with respect to the line of sight.  The most detailed currently available simultaneous broadband spectrum of W~Comae has been measured in May 1998 \\citep{tagliaferri00} and is shown in Figs. \\ref{sscgraph} and \\ref{blrgraph}. The X-ray spectrum has been measured by {\\it BeppoSAX} and shows clear evidence for the intersection of the low-frequency (synchrotron) component and the high-frequency (Compton) component of the SED of W~Comae at $\\sim 4$~keV. There was clear evidence for variability on a $\\sim 10$~hr time scale in the LECS count rate at photon energies of 0.1 -- 4~keV, while no evidence for variability was found in the MECS count rate at 4 -- 10~keV and the PDS count rate at 12 -- 100~keV. A 3~$\\sigma$ upper limit of 40~\\% on the short-term variability amplitude in the MECS count rate could be derived in the May 1998 observations \\citep{tagliaferri00}. The 4 individual spectral points in the EGRET energy range have been measured in March 1998, and are not strictly simultaneous to the {\\it BeppoSAX} observations. The EGRET detection was at low significance (2.7~$\\sigma$), and allowed only a rather crude source localization to within 1.5$^o$. We have calculated the spectrum of the source using 4 broad energy bins, as described more fully in \\S \\ref{egret}.  The remainder of this paper is organized as follows: The re-analysis of the available EGRET data is presented in \\S \\ref{egret}. In \\S \\ref{models} we describe the leptonic jet model which we use to reproduce the broadband spectrum of W~Comae. The modeling results and the model-dependent predictions for VHE emission from W~Comae will be presented in \\S \\ref{vhe}. In \\S \\ref{variability} we discuss, how a detailed measurement and analysis of the photon-energy dependent fast X-ray variability of W~Comae and other BL~Lac objects might be used to break the degeneracy of model parameters still present in the pure spectral modeling using a leptonic jet model. A comparison to the modeling results and predictions of a hadronic jet model are presented in \\S \\ref{hadrons}. We summarize in \\S \\ref{summary}. ",
        "conclusions": ""
    },
    "0208/astro-ph0208347_arXiv.txt": {
        "abstract": "{The cumulative light curves of a large sample of gamma ray bursts (GRBs) were obtained by summing the BATSE counts. The smoothed profiles are much simpler than the complex and erratic running light curves that are normally used.  For most GRBs the slope of the cumulative light curve (S) is approximately constant over a large fraction of the burst.  The bursts are modelled as relaxation systems that continuously accumulate energy in the reservoir and discontinuously release it.  The slope is a measure of the cumulative power output of the central engine. A plot of S versus peak flux in 64 ms (P$_{64 {\\rm ms}}$) shows a very good correlation over a wide range for both long and short GRBs. No relationship was found between S and the GRBs with known redshift.  The standard slope (S$^{\\prime}$), which is representative of the power output per unit time, is correlated separately with P$_{64 {\\rm ms}}$ for both sub-classes indicating more powerful outbursts for the short GRBs.  S$^{\\prime}$ is also anticorrelated with GRB duration. These results imply that GRBs are powered by accretion into a black hole. ",
        "introduction": "Cosmological gamma ray bursts (GRBs) emit an extraordinary amount of energy in gamma rays \\citep{cfpc:1997,vanpara:1997,piran:1999}. The source of this energy may be a cataclysmic event involving mergers of compact objects such as neutron star binaries or neutron stars and black hole binaries \\citep{ruffjan:1999} or the formation of a black hole during or after the collapse of massive stars \\citep{reemes:1994,pacy:1998,vietri:1998,macfad:1999,reeves:2002}. GRB light curves are complex and erratic \\citep{fishman:1995}.  There are a number of recent results on the properties of the pulses in GRBs and their relationship to the duration T$_{90}$ (Ramirez-Ruiz \\& Fenimore, 2000; Salmonson, 2000; Norris, 2002; McBreen et al., 2002a; Gupta et al., 2002; Quilligan et al., 2002 and references therein). The correlated pulses in GRBs have a unique set of properties. In this paper we show in sections 3 and 4 that the slope of the cumulative light curve of most short and long GRBs is approximately linear and the bursts can be modelled as relaxation systems.  In addition the slopes are highly correlated with the peak flux and anticorrelated with GRB duration. The large and uniform BATSE sample of GRBs were used in this analysis \\citep{fishman:1995,kmf:1993}. ",
        "conclusions": "\\subsection{GRBs as relaxation systems}  It can be assumed that the sum of the counts in the bursts (Fig. 1.) is a good measure of the integrated energy emitted by the source because the peak energy lies well within the BATSE band \\citep{fishman:1995}.  Many models of GRBs consist of a newly formed black hole that accretes from a remnant torus that is cooled by neutrino emission \\citep{popham:1999,narayan:2001,leeram:2002}. The energy to drive the relativistic jets and bursts may be extracted from the disk and spinning black hole by MHD processes and neutrino annihilation. The efficiency of these processes in creating relativistic jets is on the order of one percent \\citep{macfad:1999}  Models of this type can be usefully compared with a relaxation system \\citep{palmer:1999} which is taken to be one that continuously accumulates energy from the accretion process and discontinuously releases it. The energy in the reservoir at any time t is \\begin{equation} E(t) = E_{o} + \\int_{o}^{t} R(t)dt - \\Sigma S_{i} \\end{equation} where E$_{o}$ is the energy stored in the reservoir that accumulates energy at a rate R(t) and discontinuously releases events of size S$_{i}$.  The simplest system is referred to as a relaxation oscillator where there is a fixed level or trip-point that triggers a release of the energy when E = E$_{\\rm max}$.  The soft gamma ray repeater (SGR) \\citep{palmer:1999,gwk:2000,hmrs:1994} and GRB pulses are not consistent with this oscillator.  More complicated behaviour occurs when the accumulation rate, trigger rate or release strength are not constant.  If the system starts from a minimum level E = E$_{\\rm min}$, accumulates energy at a constant rate R = r, the sum of the releases is approximately a linear function of time i.e. \\(\\Sigma S_{i} \\propto rt\\).  This model can account for the approximately linear increase in cumulative counts from GRBs (Fig 1).  The pulses in GRBs have a tendency to keep the cumulative count close to a linear function and maintain a steady state situation.  \\citet{rm:2001} found a correlation between the duration of an emission episode in a multi-peaked burst and the duration of the preceding quiescent time which is similar to the above scenario. The system could build up its energy probably via an MHD instability driven dynamo and reach a near critical or metastable level.  A local instability could cause a rapid dissipation of all the stored energy.  The system will tend to return to a more stable configuration characterised by a certain threshold energy E$_{\\rm o}$, or a sub-critical magnetic field configuration.  The source then becomes quiescent.  Interestingly both models can be unified by noting that each time the system is completely drained by a total release of accumulated energy, the central engine goes quiescent but otherwise the energy extraction is usually in episodes that are incomplete releases of energy. In this case the longer the quiescent time, the higher the stored energy from the next episode.  Such a situation may give rise to the observed correlations between long quiescent times \\citep{rm:2001}, correlated pulse properties and intervals between pulses \\citep{quilligan:2002,smcb:2002,nakar:2002}.  This is a different mechanism from any relaxation oscillator which forces a release of energy when the system reaches an upper level.   \\begin{figure}[ht] \\leavevmode  \\begin{center} \\psfrag{Slope}[t]{\\large Slope (Cumulative Counts s$^{-1}$) } \\psfrag{P64ms}[t]{\\large P$_{\\rm 64 ms} ($ photons cm$^{-2}$ s$^{-1}$)} \\ \\resizebox{0.9\\columnwidth}{0.22\\textheight}{\\includegraphics{./ee241_f2a.eps}}\\\\[7.5pt]  \\psfrag{Sprime}[t]{\\large Standard Slope (Cumulative Counts s$^{-2}$) } \\resizebox{0.9\\columnwidth}{0.22\\textheight}{\\includegraphics{./ee241_f2b.eps}}\\\\[7.5pt]  \\psfrag{T50}[t]{\\large T$_{50}$ (sec) } \\psfrag{Sprime}[t]{\\large Standard Slope (Cumulative Counts s$^{-2}$) } \\resizebox{0.9\\columnwidth}{0.22\\textheight}{\\includegraphics{./ee241_f2c.eps}}    \\caption{ The values of P$_{64 \\rm ms}$ are plotted versus a) the slope S and b) the standard slope S$^{\\prime}$ of the GRB cumulative light curve for three categories of GRBs i.e. T$_{90} <$ 2 s (red), T$_{90} >$ 2 s (green) and the additional sample with T$_{90} >$ 100 s (blue).  T$_{50}$ is plotted versus S$^{\\prime}$ in c) for the same three catagories. The seven GRBs with known redshift and detected by BATSE are labelled (van Paradijs et al., 2000; Castro-Tirado, 2001 and references therein). The BATSE trigger numbers and redshifts are given in the top figure.  An extension of the peak flux limited sample with T$_{90} >$ 2 s to lower values should populate the region containing GRBs 1 and 2 with known z in b).} \\end{center} \\end{figure}  \\subsection{Relationships between the slopes and peak flux}  There is a significant correlation, that extends over a range of $\\sim10^{3}$, between P$_{64 \\rm ms}$ and S (Fig. 2a and Table 1).  The seven GRBs, detected by BATSE, with measured redshifts are also plotted in Fig. 2.  These sources cover a wide range with no obvious relationship between S, P$_{\\rm 64 ms}$ and z.  The separation of GRBs into two sub-classes was characterised by durations $>$ 2 s and $<$ 2 s \\citep{kmf:1993}. The plot of the standard slope versus P$_{\\rm 64 ms}$ (Fig. 2b) also separates the GRBs into two classes with the short GRBs having a more powerful output power per unit time.  The cosmological GRBs with known redshift have values of S and S$^{\\prime}$ that range from $7 \\times 10^{44}$ to $5 \\times 10^{47}$ Watts and $1.5 \\times 10^{44}$ to $2 \\times 10^{46}$ Watts per second respectively. There is no evidence for a new class of GRBs with a different relationship between the standard slope and P$_{\\rm 64 ms}$.  \\subsection{Relationships between the slopes and durations}  The data presented in Fig. 2c shows that as the standard slope increases the values of T$_{50}$ decrease and this effect is present in both sub-classes (Table 1).  The trend is quite revealing and shows that the smaller the value of T$_{50}$ the greater the standard slope or the cumulative power output per second from the source.  The standard slope plays an important role in determining T$_{50}$. The median values of the pulse properties and time intervals between pulses were found to increase with T$_{90}$ \\citep{smcb:2002}. The opposite relationship exists here between S$^{\\prime}$ and burst duration (Table 1) implying that as the standard slope increases there is a corresponding decrease in the pulse properties. GRBs with high accretion rates have large values of the standard slope, fast pulses and short durations whereas lower accretion gives lower values of S$^{\\prime}$, slower pulses that are further apart and larger values of T$_{90}$/T$_{50}$. These results provide strong evidence that GRBs are powered by hyperaccretion into a black hole from a standard type engine \\citep{sal:2000,frail:2001,panait:2001,piran:2001}.  \\small"
    },
    "0208/astro-ph0208567_arXiv.txt": {
        "abstract": "We review the current status of neutrino cosmology, focusing mainly on the question of the absolute values of neutrino masses and the possibility of a cosmological neutrino lepton asymmetry. \\vspace{1pc} ",
        "introduction": "The absolute value of neutrino masses are very difficult to measure experimentally. On the other hand, mass differences between neutrino mass eigenstates, $(m_1,m_2,m_3)$, can be measured in neutrino oscillation experiments. Observations of atmospheric neutrinos suggest a squared mass difference of $\\delta m^2 \\simeq 3 \\times 10^{-3}$ eV$^2$ \\cite{Fukuda:2000np,Fornengo:2000sr}. While there are still several viable solutions to the solar neutrino problem the so-called large mixing angle solution gives by far the best fit with $\\delta m^2 \\simeq 5 \\times 10^{-5}$ eV$^2$ \\cite{sno,Bahcall:2002hv} (see also contributions by A. Hallin and A. Smirnov in the present volume).  In the simplest case where neutrino masses are hierarchical these results suggest that $m_1 \\sim 0$, $m_2 \\sim \\delta m_{\\rm solar}$, and $m_3 \\sim \\delta m_{\\rm atmospheric}$. If the hierarchy is inverted \\cite{Kostelecky:1993dm,Fuller:1995tz,Caldwell:1995vi,Bilenky:1996cb,King:2000ce,He:2002rv} one instead finds $m_3 \\sim 0$, $m_2 \\sim \\delta m_{\\rm atmospheric}$, and $m_1 \\sim \\delta m_{\\rm atmospheric}$. However, it is also possible that neutrino masses are degenerate \\cite{Ioannisian:1994nx,Bamert:vc,Mohapatra:1994bg,Minakata:1996vs,Vissani:1997pa,Minakata:1997ja,Ellis:1999my,Casas:1999tp,Casas:1999ac,Ma:1999xq,Adhikari:2000as}, $m_1 \\sim m_2 \\sim m_3 \\gg \\delta m_{\\rm atmospheric}$, in which case oscillation experiments are not useful for determining the absolute mass scale.  Experiments which rely on kinematical effects of the neutrino mass offer the strongest probe of this overall mass scale. Tritium decay measurements have been able to put an upper limit on the electron neutrino mass of 2.2 eV (95\\% conf.) \\cite{Bonn:tw} (see also the contribution by Ch. Weinheimer in the present volume). However, cosmology at present yields an even stronger limit which is also based on the kinematics of neutrino mass.  Neutrinos decouple at a temperature of 1-2 MeV in the early universe, shortly before electron-positron annihilation. Therefore their temperature is lower than the photon temperature by a factor $(4/11)^{1/3}$. This again means that the total neutrino number density is related to the photon number density by \\begin{equation} n_{\\nu} = \\frac{9}{11} n_\\gamma \\end{equation}  Massive neutrinos with masses $m \\gg T_0 \\sim 2.4 \\times 10^{-4}$ eV are non-relativistic at present and therefore contribute to the cosmological matter density \\cite{Hannestad:1995rs,Dolgov:1997mb,Mangano:2001iu} \\begin{equation} \\Omega_\\nu h^2 = \\frac{\\sum m_\\nu}{92.5 \\,\\, {\\rm eV}}, \\end{equation} calculated for a present day photon temperature $T_0 = 2.728$K. Here, $\\sum m_\\nu = m_1+m_2+m_3$. However, because they are so light these neutrinos free stream on a scale of roughly $k \\simeq 0.03 m_{\\rm eV} \\Omega_m^{1/2} \\, h \\,\\, {\\rm Mpc}^{-1}$ \\cite{dzs,Doroshkevich:tq,Hu:1997mj}. Below this scale neutrino perturbations are completely erased and therefore the matter power spectrum is suppressed, roughly by $\\Delta P/P \\sim -8 \\Omega_\\nu/\\Omega_m$ \\cite{Hu:1997mj}.  This power spectrum suppression allows for a determination of the neutrino mass from measurements of the matter power spectrum on large scales. This matter spectrum is related to the galaxy correlation spectrum measured in large scale structure (LSS) surveys via the bias parameter, $b^2 \\equiv P_g(k)/P_m(k)$. Such analyses have been performed several times before \\cite{Croft:1999mm,Fukugita:1999as}, most recently using data from the 2dF galaxy survey \\cite{Elgaroy:2002bi}. This investigation finds an upper limit of 1.8-2.2 eV for the sum of neutrino masses. However, this result is based on a relatively limited cosmological parameter space. For the same data and an even more restricted parameter space an upper limit of 1.5 eV was found \\cite{Lewis:2002ah}.    It should also be noted that, although massive neutrinos have little impact on the cosmic microwave background (CMB) power spectrum, it is still necessary to include CMB data in any analysis in order to determine other cosmological parameters.  When calculating bounds on the neutrino mass from cosmological observations great care must be taken, because if the analysis is based on a too restricted parameter space, possible parameter degeneracies cannot be studied and the bound on $m_\\nu$ can become artificially strong.  \\begin{figure}[h] \\begin{center} \\vspace*{-1cm} \\hspace*{-1cm}\\includegraphics[width=20pc]{fig1.ps} \\vspace*{-1cm} \\end{center} \\vspace*{-1.5cm} \\caption{Values of the parameter $r_{ij}$, defined in Eq.~(\\ref{eq:rij}).} \\label{fig:1} \\end{figure}  In Fig. 1 the degeneracy between different cosmological parameters is shown in form of the quantity \\begin{equation} r_{ij} = \\frac{\\sigma_{j \\,\\, {\\rm fixed}}(\\theta_i)}{\\sigma (\\theta_i)} \\leq 1, \\label{eq:rij} \\end{equation} i.e.\\ the decrease in uncertainty of a measurement of parameter $i$ when parameter $j$ is fixed. From this figure it can be seen that there are significant degeneracies between $\\Sigma m_\\nu$ and other parameters, most notably $b$, the bias parameter, and $\\Omega_m$, the matter density.   In Ref.~\\cite{han02} a full numerical likelihood analysis was performed using a slightly restricted parameter space with the following free parameters: $\\Omega_m$, $\\Omega_b$, $H_0$, $n_s$, $Q$, $b$, and $\\tau$. The analysis was further restricted to flat models, $\\Omega_k=0$. This has very little effect on the analysis because there is little degeneracy between $m_\\nu$ and $\\Omega_k$. In order to study the effect of the different priors three different cases were calculated, the priors for which can be seen in Table I. The BBN prior on $\\Omega_b h^2$ comes from Ref.~\\cite{Burles:2000zk}. The actual marginalization over parameters other than $\\Omega_\\nu h^2$ was performed using a simulated annealing procedure \\cite{Hannestad:wx}.  Fig.~2 shows $\\chi^2$ for the three different cases as a function of the $m_\\nu$. The best fit $\\chi^2$ values are 24.81, 25.66, and 25.71 for the three different priors respectively. In comparison the number of degrees of freedom are 34, 35, and 36, meaning that the fits are compatible with expectations, roughly within the 68\\% confidence interval.  \\begin{figure}[h] \\begin{center} \\hspace*{-1cm}\\includegraphics[width=20pc]{fig2.ps} \\end{center} \\vspace*{-1cm} \\caption{$\\chi^2$ as a function of $\\Omega_\\nu h^2$, plotted for the three different priors. The dotted curve is for CMB+LSS, the dashed for CMB+LSS+BBN+$H_0$, and the full curve for CMB+LSS+BBN+$H_0$+SNIa.} \\label{fig:2} \\end{figure}   The 95\\% confidence limit on $m_\\nu$ was identified with the point where $\\Delta \\chi^2 = 4$. These limits are shown in Table II. For the most restrictive prior we find a 95\\% confidence upper limit of $\\sum m_\\nu \\leq 2.47$ eV. This is compatible with the findings of Ref.~\\cite{Elgaroy:2002bi} who derived that $\\sum m_\\nu \\lesssim 1.8-2.2$ eV for a slightly more restrictive parameter space.  Based on the present analysis we consider $\\sum m_\\nu \\leq 3$ eV (95\\% conf.) a robust upper limit on the sum of the neutrino masses. This corresponds roughly to the value found for the CMB+LSS data alone without any additional priors. Even though this value is significantly higher than what is quoted in Ref.~\\cite{Elgaroy:2002bi}, it is still much more restrictive than the value $\\sum m_\\nu \\leq 4.4$ eV \\cite{WTZ} found from CMB and PSCz \\cite{pscz} data. As is also discussed in Ref.~\\cite{Elgaroy:2002bi} the main reason for the improvement is the much greater precision of the 2dF survey, compared to the PSCz data \\cite{pscz}.    \\begin{table*}[htb] \\caption{The different priors on parameters other than $\\Omega_\\nu h^2$ used in the analysis of Ref.~\\protect\\cite{han02}.} \\label{table:1} \\newcommand{\\m}{\\hphantom{$-$}} \\newcommand{\\cc}[1]{\\multicolumn{1}{c}{#1}} \\renewcommand{\\tabcolsep}{2pc} % \\renewcommand{\\arraystretch}{1.2} %  \\begin{tabular}{lccc} \\hline Parameter & CMB + LSS & CMB + LSS & CMB + LSS + BBN  \\\\ && + BBN + $H_0$ & + BBN + $H_0$ + SNIa \\\\ \\hline $\\Omega_m$ & 0.1-1 & 0.1-1 & $0.28 \\pm 0.14$ \\\\ $\\Omega_b h^2$ & 0.008 - 0.040 & $0.020 \\pm 0.002$ & $0.020 \\pm 0.002$ \\\\ $h$ & 0.4-1.0 & $0.70 \\pm 0.07$ & $0.70 \\pm 0.07$ \\\\ $n$ & 0.66-1.34 & 0.66-1.34 & 0.66-1.34 \\\\ $\\tau$ & 0-1 & 0-1 & 0-1 \\\\ $Q$ & free & free & free \\\\ $b$ & free & free & free \\\\ \\hline \\end{tabular}\\\\[2pt] \\end{table*}    \\begin{table*}[htb] \\caption{Best fit $\\chi^2$ and upper limits on $\\sum m_{\\nu,{\\rm max}}$ for the three different priors.} \\label{table:2} \\newcommand{\\m}{\\hphantom{$-$}} \\newcommand{\\cc}[1]{\\multicolumn{1}{c}{#1}} \\renewcommand{\\tabcolsep}{2pc} % \\renewcommand{\\arraystretch}{1.2} % \\begin{tabular}{lcc} \\hline prior type & best fit $\\chi^2$ & $\\sum m_{\\nu,{\\rm max}}$ (eV) (95\\%) \\\\ \\hline CMB + LSS & 24.81 & 2.96 \\\\ CMB + LSS + BBN + $H_0$ &  25.66 & 2.65 \\\\ CMB + LSS + BBN + $H_0$ + SNIa & 25.71 & 2.47 \\\\ \\hline \\end{tabular}\\\\[2pt] \\end{table*} ",
        "conclusions": "Cosmology offers an interesting probe of neutrino physics which is complementary to terrestrial experiments.  Observations of the large scale structure power spectrum has already given bounds on the absolute value of neutrino masses which are comparable to, or stronger than the present bound from tritium endpoint measurements. For the present data an upper limit to the neutrino mass of $\\sum m_\\nu \\lesssim 2.5-3$ eV can be derived. This can be compared to the present bound from the Mainz experiment of $m_{\\nu_e} = \\sum_i  \\left(|U_{ei}|^2 m_i^2\\right)^{1/2} = 2.2$ eV.  However, it should be noted that an even stronger upper bound can be put on neutrino masses if they are Majorana particles. In that case neutrinoless double beta decay is possible because lepton number is not a conserved quantity. The non-observation of such events has led to the bound \\begin{equation} m_{ee} = \\sum_j U^2_{ej} m_{\\nu_j} < 0.27 \\,\\, {\\rm eV}, \\end{equation} where $U$ is the neutrino mixing matrix \\cite{klapdor}. From observations of neutrinoless double beta decay it has indeed been claimed that positive evidence for non-zero neutrino masses has been obtained, with a favoured value in the range 0.11-0.56 eV \\cite{Klapdor-Kleingrothaus:2001ke}. However, this claim is highly controversial and has been refuted by a number of other authors \\cite{Aalseth:2002dt}. At present it therefore seems safest to regard neutrinoless double beta decay experiments as yielding only an upper limit on $N_\\nu$.  In the coming years the large scale structure power spectrum will be measured even more accurately by the Sloan Digital Sky Survey, and at the same time the CMB anisotropy will be probed to great precision by the MAP and Planck satellites. By combining these measurements it was estimated by Hu, Eisenstein and Tegmark that a sensitivity of about 0.3 eV could be reached \\cite{Hu:1997mj}. Currently an upgrade of the Mainz experiment by an order of magnitude, the KATRIN experiment, is planned. Such an experiment should take the limit on the electron neutrino mass down to about 0.2 eV. The prospects for measuring a neutrino mass of the order 0.1 eV, as suggested by oscillation experiments is therefore almost within reach.  Another cosmological probe of the neutrino mass is the so-called Z-burst scenario for ultrahigh energy cosmic rays \\cite{Weiler:1997sh,Fargion:1997ft}. Neutrinos are not subject to the GZK cut-off which applies to protons \\cite{Greisen:1966jv,Zatsepin:1966jv}. Therefore it is in principle possible that the primary particles for super GZK cosmic rays are neutrinos. One possibility which has been explored is that the neutrino-nucleon cross-section increases drastically at high CM energies, for instance due to the presence of large extra dimensions. The other possibility is that neutrinos have rest mass in the eV range. In that case high energy neutrinos can annihilate on cosmic background neutrinos with a large cross section if the CM energy is close to the Z-resonance, corresponding to a primary neutrino energy of $E_\\nu \\simeq 4 \\times 10^{21} m_{\\rm eV}^{-1} \\, {\\rm eV}$. This annihilation would produce high energy protons which could then act as primaries for the observed high energy cosmic rays. The observed ultrahigh energy cosmic ray flux can be explained if the heaviest neutrino has a mass larger than $\\sim 0.1$ eV. Therefore, if the Z-burst scenario turns out to be correct, it is in principle possible to measure a neutrino mass in this range \\cite{Pas:2001nd,Fodor:2001qy,Ringwald:2001mx}.  With regards to the neutrino relativistic energy density, the present bound from BBN is roughly $N_\\nu \\lesssim 3.5-4$. The prospects for improving this bound in the future do not seem too bright. The present bound is most likely dominated by systematic effects in the measurement of helium abundances. BBN is also a very powerful probe of a possible neutrino lepton asymmetry, particularly if, as is indicated by data from SNO and Super-Kamiokande \\cite{sno,Bahcall:2002hv}, the LMA mixing solution is correct. In this case, an upper bound on the lepton asymmetry for any flavour is roughly $\\mu_{\\nu_i}/T \\lesssim 0.07$ \\cite{Dolgov:2002ab}.  The combination of CMB and large scale structure data can also be used for constraining the neutrino relativistic energy density. At present the CMB data are not of sufficient accuracy to yield a bound which is competitive with that from BBN, the present bound being $N_\\nu = 6^8_{4.5}$ (95\\%) (see also Table 3). However, this bound applies to a very different epoch and therefore puts significant constraints on possible entropy production after BBN but before CMB formation.  \\begin{table*}[htb] \\caption{Best fit values and $2\\sigma$ (95\\%) limits on $N_\\nu$ for different priors and two different data sets. Both priors and data sets are discussed in Ref.~\\protect\\cite{Hannestad:2001hn}} \\label{table:3} \\newcommand{\\m}{\\hphantom{$-$}} \\newcommand{\\cc}[1]{\\multicolumn{1}{c}{#1}} \\renewcommand{\\tabcolsep}{2pc} % \\renewcommand{\\arraystretch}{1.2} % \\begin{tabular}{lcc} \\hline prior type & WTZ & COBE+Boomerang \\\\ \\hline CMB only & $8^{+11}_{-8}$ & $7^{+17}_{-7}$ \\\\ CMB + BBN + $H_0$ & $8^{+9.5}_{-7}$ & $4^{+13}_{-4}$ \\\\ CMB + BBN + $H_0$ + LSS & $6^{+8}_{-4.5}$ & $9^{+8}_{-6.5}$  \\\\ \\hline \\end{tabular} \\end{table*} In the near future a much more accurate determination of $N_\\nu$ from CMB measurements will become possible thanks to the satellites MAP and Planck. It was estimated by Lopez et al. \\cite{Lopez:1999aq} that it would be possible to measure $\\Delta N_\\nu \\sim 0.04$ using Planck data. However, this is probably overly optimistic and a more reasonable estimate seems to be $\\Delta N_\\nu \\sim 0.1-0.2$ \\cite{Bowen:2001in}.  This will also allow for a possible detection of sterile neutrinos mixing with ordinary neutrinos in the early universe over a wide range of parameter space \\cite{Hannestad:1998zg}."
    },
    "0208/astro-ph0208084_arXiv.txt": {
        "abstract": "{\\small Astrophysical jets exist from the stellar scale up to AGN, and seem to share common features particularly in the radio.  But while AGN jets are known to emit X-rays, the situation for XRB jets is not so clear. Radio jets have been resolved in several XRBs in the low/hard state, and it seems likely that some form of outflow is present whenever this state is achieved.  Interestingly, the flat-to-inverted radio synchrotron emission associated with these outflows strongly correlates with the X-ray emission in several sources, suggesting that the jet plasma plays a role at higher frequencies.  In this same state, there is also increasing evidence for a turnover in the IR/optical where the flat-to-inverted spectrum seems to connect to an optically thin component extending into the X-rays.  We discuss how jet synchrotron emission is likely to contribute to the X-rays, in addition to inverse Compton up-scattering, providing a natural explanation for these correlations and the turnover in the IR/optical band.} ",
        "introduction": "Active galactic nuclei (AGN) jets have been extensively imaged in the radio, and also emit significantly at higher frequencies including the X-rays via synchrotron and inverse Compton (IC).  It turns out that black hole candidate (BHC) X-ray binaries (XRBs) also produce collimated outflows, at least when in the low/hard state (LHS) \\cite{Fender2001a}.  The jets in several persistent Galactic sources have already been resolved in the radio (e.g., 1E1740.7-2942, \\cite{Mirabeletal1992}; Cyg X-1, \\cite{Stirlingetal2001}).  Analogous to the ``signature'' emission of compact radio cores in AGN (e.g., \\cite{BlandfordKoenigl1979}), XRB jets contribute a flat-to-inverted radio synchrotron component to the LHS spectra.  Beyond this radio signature, a typical LHS spectrum shows a weak thermal contribution and a hard power-law at higher frequencies.  This has generally been interpreted in terms of a Standard Thin Disk (SD; \\cite{ShakuraSunyaev}) which either transitions at some radius to an optically thin, non-radiative flow, or is underlying a corona (for reviews see \\cite{Poutanen,NowakWilmsDove}).  The hotter plasma is believed to account for the hard power-law via IC upscattering of the thermal SD photons.  While variations of this picture can successfully explain the X-ray features, there is increasing evidence that---at least in the LHS and likely also the quiescent state---the jet is also playing a role outside the radio band.  The extent of this role is not yet clear, but we have found that emission from the jets alone can actually account for the majority of the broadband LHS spectrum (excluding the thermal SD contribution) in sources where simultaneous radio, X-ray and sometimes infrared (IR)/optical data are available.  The models for these various sources also show a surprising degree of similarity in their input parameters, and is the only model yet which can explain the multi-wavelength correlations which are now being seen in several sources.  It is therefore important to begin exploring ways in which the jets can be incorporated into the previously disk/corona-only X-ray picture, and to know the extent to which they can reasonably contribute. ",
        "conclusions": ""
    },
    "0208/hep-ph0208046_arXiv.txt": {
        "abstract": "We study the scattering of fermions off a finite width kink wall during the electroweak phase transition in the presence of a background hypermagnetic field. We derive and solve the Dirac equation for such fermions and compute the reflection and transmission coefficients for the case when the fermions move from the symmetric to the broken symmetry phase. We show that the chiral nature of the fermion coupling with the background field in the symmetric phase generates an axial asymmetry in the scattering processes. We discuss possible implications of such axial charge segregation for baryon number generation. ",
        "introduction": "\\label{I}  The possible existence of magnetic fields in the early universe has recently become the subject of intense research due to the many interesting cosmological implications that these entail~\\cite{reviews}. For instance, magnetic fields can influence big bang nucleosynthesis (BBN), affecting the primordial abundance of light elements and the rate of expansion of the universe. The success of the standard BBN scenario can be used to set limits on the strength of the magnetic fields at this epoch. Moreover, at decoupling, long range magnetic fields can induce anisotropies in the cosmic microwave background radiation. Temperature anisotropies from COBE results place an upper bound $B_0\\sim 10^{-9}\\ $G for homogeneous fields ($B_0$ refers to the intensity that the field would have today under the assumption of adiabatic decay due to the Hubble expansion)~\\cite{Barrow}. In the case of inhomogeneous fields their effect must be searched for in the Doppler peaks~\\cite{Adams} and in the polarization of the CMBR~\\cite{Kosov}. The future CMBR satellite missions MAP and PLANCK may reach the required sensitivity for the detection of these last signals.  Another interesting cosmological consequence is the effect that primordial magnetic fields could have had on the dynamics of the electroweak phase transition (EWPT) at temperatures of the order of $T\\sim 100$ GeV. In fact, it has been recently pointed out that, provided enough {\\it CP} violation exists, large scale primordial magnetic fields can be responsible for a stronger first order EWPT~\\cite{{Giovannini},{Elmfors},{Giovannini2}} (see however Ref.~\\cite{Skalozub}). The situation is similar to a type I superconductor where the presence of an external magnetic field modifies the nature of the superconducting phase transition due to the Meissner effect.  Recall that for temperatures above the EWPT, the SU(2)$\\times$U(1)$_Y$ symmetry is restored and the propagating, non-screened vector modes that represent a magnetic field correspond to the U(1)$_Y$ group instead of to the U(1)$_{em}$ group, and are therefore properly called {\\it hypermagnetic} fields.  In a previous work~\\cite{Ayala2}, we have shown, by using a simplified picture of a first order EWPT, that the presence of such fields also provides a mechanism, working in the same manner as the existence of additional {\\it CP} violation within the SM, to produce an axial charge segregation in the scattering of fermions off the true vacuum bubbles The asymmetry in the scattering of fermion axial modes is a consequence of the chiral nature of the fermion coupling to hypermagnetic fields in the symmetric phase. The simplification consisted in considering the limit of an infinitely thin bubble wall. This assumption allowed us to formulate the problem in terms of solving the Dirac equation with a position dependent fermion mass, proportional to a step function, this last being zero in the false phase and non-vanishing in the broken symmetry phase. This treatment rendered analytic solutions from where reflection and transmission coefficients for axial modes were straightforward computed.  In spite of the relative ease for the computation in such a scheme, there are two limitations related to the sudden change in the Higgs field profile that needed to be addressed. First, it is well known that the negative energy solutions of the Dirac equation become important in situations where the potential energy term changes over distances smaller than the particle's Compton wave length. Second, the height and width of the wall are typically related to each other in such a way that it is not entirely realistic to vary one without affecting the other.  In this paper we overcome the above limitations by allowing a finite width of the Higgs field profile. Working in the thin wall regime, we use the kink solution of the Higgs field to formulate and solve the Dirac equation in the presence of hypermagnetic fields. We compute explicitly transmission and reflection coefficients for the axial modes incident on the wall from the symmetric phase. Since these are related to the corresponding coefficients for fermions incident from the broken symmetry phase by CPT and Unitarity, we find that the axial charge segregation still happens during fermion scattering of this wall. The existence of such asymmetric reflection for the axial modes provides a bias for baryon over antibaryon production. In the absence of hypermagnetic fields, this mechanism has been proposed and studied in Refs.~\\cite{{Dine},{Cohen},{Nelson}} in extensions of the SM.  The outline of this work is as follows: In Sect.~\\ref{II}, we briefly review how the kink solution for the spatial profile of the Higgs field is obtained from a finite temperature effective potential. In Sect.~\\ref{III}, we set up the Dirac equation for fermions moving in this background Higgs field in the presence of an external hypermagnetic field. Section.~\\ref{IV} is devoted to a rather technical discussion about the solutions of this equation and their properties. In Sect.~\\ref{V}, we use the above solutions to compute reflection and transmission coefficients for axial fermion modes moving from the symmetric phase toward the broken symmetry phase. We show that these coefficients differ for the two distinct helicity modes. Finally in Sect.~\\ref{VI}, we conclude by looking out at the possible implications of such axially asymmetric fermion reflection and transmission. ",
        "conclusions": "\\label{VI}  In this paper we have derived and solved the Dirac equation for fermions scattering off a first order EWPT bubble wall with a finite width in the presence of a magnetic field directed along the fermion direction of motion. In the symmetric phase, the fermions couple chirally to the magnetic field, which receives the name of {\\it hypermagnetic}, given that it belongs to the $U(1)_Y$ group. We have shown that the chiral nature of this coupling implies that it is possible to build an axial asymmetry during the scattering of fermions off the wall. We have computed reflection and transmission coefficients showing explicitly that they differ for left and right-handed incident particles from the symmetric phase. The results of this more realistic, albeit numerical calculation where we allow for a finite wall width are in qualitative and quantitative agreement with those previously found in Ref.~\\cite{Ayala2}, where the wall was modeled as a step function.  \\begin{figure}[t] % \\vspace{-0.8cm} {\\centering\\rotatebox{-90}{ \\resizebox*{0.35\\textwidth}{!} {\\includegraphics{coefb0.5xi3.5-left.ps}}}\\par} {\\centering\\rotatebox{-90}{ \\resizebox*{0.35\\textwidth}{!} {\\includegraphics{coefb0.5xi3.5-right.ps}}}\\par} \\caption{Reflection and transmission coefficients as a function of the energy parameter $\\epsilon$ scaled by twice the height of the barrier $2\\xi$ for $b=0.5$ and $\\xi=3.5$, $y_R=4/3$, $y_L=1/3$, $g'=0.344$. Figure~2a (upper panel) shows the coefficients for incident, left-handed helicity modes and Fig.~2b (lower panel) for incident, right-handed helicity modes. In both figures, the dots represent the computed values.} \\end{figure}  It could be thought that the asymmetric reflection found in this work could be washed out when considering the averaging over the different angles of incidence of the fermion flux. This is not the case as we proceed to show. Let us first look a the situation in which the direction of the magnetic field is reversed with respect to the case studied here. This a physically relevant scenario since during the phase transition, fermions are scattered on opposite sides of the bubbles and if the sign of the asymmetry would depend on the direction of the magnetic field with respect to the direction of fermion incidence, then the building of an axial charge density in one side of the bubble would compensate the building of this charge on the other side, thereby canceling the effect. However, it is easy to convince oneself that this is not the case. By looking at Eqs.~(\\ref{set1}) and~(\\ref{set2}), we see that changing $B$ to $-B$ interchanges one set of equations with the other, leaving intact the coupling. Physically this is also easy to understand since the fermion coupling with the external field is through its spin. Changing the direction of the field exchanges the role of each spin component but since each chirality mode contains both spin orientations, it does not affect the final probabilities.  Now suppose that the original direction of motion of the fermion is not parallel to the direction of the magnetic field and therefore its velocity vector contains a component perpendicular to the direction of the field. In this case, due to the Lorentz force, the particle circles around the field lines maintaining its velocity along the direction of the field. The motion of the particle is thus described as an overall displacement along the field lines superimposed to a circular motion around these lines. In the three dimensional quantum mechanical treatment of the problem, these circles correspond to the different Landau levels. We see that the originally different angles of incidence all result in the same overall direction of incidence. Nonetheless, it is certainly true that these circular trajectories could be regarded as the paths where the wave function of the particle picks up a phase in the same manner as in the Aharanov-Bohm effect. However, since there is no definite phase relation of the incident fermions, these phases have to be regarded as randomly distributed. Thus, the addition of the wave functions at the interference point (minus infinity for the reflected waves and plus infinity for the transmitted waves) has to be done incoherently which precludes any possible destructive effect of these phases on the overall particle fluxes.  We also emphasize that, under the very general assumptions of CPT invariance and unitarity, the total axial asymmetry (which includes contributions both from particles and antiparticles) is quantified in terms of the particle (axial) asymmetry. Let $\\rho_i$ represent the number density for species $i$. The net densities in left-handed and right-handed axial charges are obtained by taking the differences $\\rho_L-\\rho_{\\bar{L}}$ and $\\rho_R-\\rho_{\\bar{R}}$, respectively. It is straightforward to show~\\cite{Nelson} that CPT invariance and unitarity imply that the above net densities are given by \\be \\rho_L-\\rho_{\\bar{L}}&=&(f^s-f^b) (R_{r\\rightarrow l} - R_{l\\rightarrow r})\\nonumber\\\\ \\rho_R-\\rho_{\\bar{R}}&=&(f^s-f^b) (R_{l\\rightarrow r} - R_{r\\rightarrow l})\\, , \\label{net} \\ee where $f^s$ and $f^b$ are the statistical distributions for particles or antiparticles (since the chemical potentials are assumed to be zero or small compared to the temperature, these distributions are the same for particles or antiparticles) in the symmetric and the broken symmetry phases, respectively. From Eq.~(\\ref{net}), the asymmetry in the axial charge density is finally given by \\be (\\rho_L-\\rho_{\\bar{L}}) - (\\rho_R-\\rho_{\\bar{R}})= 2(f^s-f^b)(R_{r\\rightarrow l} - R_{l\\rightarrow r}). \\label{final} \\ee This asymmetry, built on either side of the wall, is dissociated from non-conserving baryon number processes and can subsequently be converted to baryon number in the broken symmetry phase where sphaleron induced transitions are taking place with a large rate. This mechanism receives the name of {\\it non-local baryogenesis}~\\cite{{Dine},{Nelson},{Cohen},{Joyce}} and, in the absence of the external field, it can only be realized in extensions of the SM where a source of {\\it CP} violation is introduced {\\it ad hoc} into a complex, space-dependent phase of the Higgs field during the development of the EWPT~\\cite{Torrente}.  Since  another consequence of the existence of an external magnetic field is the lowering of the barrier between topologically inequivalent vacua~\\cite{Comelli}, due to the sphaleron dipole moment, the use of the mechanism discussed in this work to possibly generate a baryon asymmetry is not as straightforward. Nonetheless, if such primordial fields indeed existed during the EWPT epoch and the phase transition was first order, as is the case, for instance, in minimal extensions of the SM, the mechanism advocated in this work has to be considered as acting in the same manner as a source of {\\it CP} violation that can have important consequences for the generation of a baryon number. These matters will be the subject of an upcoming work~\\cite{Ayala3}."
    },
    "0208/astro-ph0208059_arXiv.txt": {
        "abstract": "We examine the host galaxies of high redshift type~Ia supernovae (SNe~Ia) using archival I and R band data from the Hubble Space Telescope.  The SNe~Ia host galaxies show a wide variety of morphologies, including undisturbed ellipticals, spirals and disturbed systems. SNe~Ia are also found over a wide range of projected distances from the host galaxy centres, ranging from $3$kpc to $\\sim30$kpc.  For a sample of 22 SNe~Ia at $\\langle z\\rangle = 0.6$, $\\sim70\\%$ are found in spiral galaxies and $\\sim30\\%$ are found in elliptical systems, similar to the proportions observed locally.  Including data from \\citet{ell}, we find no significant difference in the average light-curve-shape-corrected $M_{B}^{peak}$ for high-z SNe~Ia between spirals and ellipticals.  These results are consistent with predictions based on the locally-derived understanding of SNe~Ia physics and the influence of progenitor mass and metallicity.  We also construct colour maps for two host galaxies and find that both show a non-uniform colour structure with typical variations of rest-frame $B-V\\sim0.5$.  This is most plausibly attributed to the presence of, and variation in dust extinction in these galaxies.  Moreover, we find no evidence that the SNe~Ia are preferentially found in outer regions ($>10$~kpc) of the host galaxies where extinction would be low.  This suggests that the range of host galaxy extinctions of SNe~Ia at $z\\sim0.6$ should be comparable to those of local SNe~Ia.  Although observational bias cannot be completely ruled out, this appears to be in conflict with the finding of low extinction for SNe~Ia found in the high-$z$ supernova search studies. ",
        "introduction": "Observations of type~Ia supernovae (SNe~Ia) at high redshift has led to the astonishing deduction that the universal expansion is accelerating \\citep{rie1,per1}. A key factor in this discovery has been the painstaking calibration of the relation between the SNe~Ia luminosity and the behaviour of the light curves \\citep{phi0,ham,rpk,per0,tri,phi,sah}.  However, the size of the effect (about 0.25~mags) which indicates acceleration is several times smaller than the intrinsic range of luminosity in local SNe~Ia {\\it i.e.} the correction that has to be made using the calibration relation is large.  Given that the calibration relation is derived empirically, and that the physics of SNe~Ia is only partially understood, it is therefore important to exploit every opportunity to test for intrinsic differences in the SNe or in their environments between the local and high-z universe.  In this letter we consider the rates, luminosities and extinction of SNe~Ia at $z\\sim0.6$.  The current consensus on the diversity of SNe~Ia is that it arises primarily from the mass of $^{56}$Ni produced in the explosion, with a higher relative $^{56}$Ni producing a more luminous and more slowly declining lightcurve.  Close-binary evolutionary calculations \\citep{ume1} of possible progenitors indicate that the C/O ratio is lower for higher mass CO cores.  It has been suggested, although not demonstrated, that a smaller C/O ratio in the progenitor white dwarf yields less $^{56}$Ni and so a lower luminosity \\citep{hwt,ume1}. Thus, CO white dwarf progenitors arising from higher mass main-sequence stars may yield less luminous SNe~Ia.  Metallicity can have a significant effect via the white dwarf wind which affects the minimum CO white dwarf mass that can accrete as far as the Chandrasekhar Limit \\citep{kob}. A high metallicity implies a higher white dwarf wind.  This allows white dwarfs with a lower CO mass to reach the Chandrasekhar Limit and hence produce brighter SNe~Ia. Thus, there will be a tendency for a larger number of, and more luminous SNe~Ia to occur in high metallicity populations and vice versa.  It is predicted that metallicity in spirals will decrease with increasing redshift \\citep{cal}.  Such a decrease is not expected to occur in ellipticals since they have had essentially no star formation over this look-back time.  Consequently we might expect that, relative to ellipticals, SNe~Ia in high-$z$ spirals should exhibit reduced luminosities and lower rates.  Conversely, modelling of the total SNe~Ia rate (per $10^{10}$ $L_{B\\odot}$) with redshift predicts little change in the total SNe~Ia rate with increasing redshift up to $z\\sim1.6$ \\citep{ktn}. This prediction has been shown to be true up to at least $z\\sim0.55$ by independent observations \\citep{pai}. Moreover, \\citet{ktn} also model the variation in SNe~Ia rate with redshift for host galaxies of different morphological types, and find that significant changes are anticipated only beyond z$\\sim$1.6.  At lower redshifts, the absolute and relative rates are predicted to remain roughly the same as the local values for spirals and ellipticals.  Another important issue in current studies of high-$z$ SNe~Ia is that of extinction.  Both supernova search teams correct for galactic extinction but find that very few of their supernovae show a residual colour excess, implying negligible extinction in the host galaxies. No significant extinction is found in 14 out of 16 SNe~Ia discovered by \\citet{rie1} and in 40 out of 42 SNe~Ia discovered by \\citet{per1}. \\citet{rie1} also argue against the presence of neutral extinction. They point out that the observed dispersion in MLCS-derived distances is substantially smaller than would be expected if grey dust extinction were the main cause of the lower brightness of high-$z$ SNe~Ia.  However, \\citet{lei} has highlighted inconsistencies in the treatment of host galaxy extinctions between the two SNe~Ia search teams.  Moreover, \\citet{mrr} has shown that by employing a consistent treatment of extinction and excluding those SNe~Ia observed only after maximum light, a positive $\\Lambda$ is only required at the $\\sim3\\sigma$ level.  The apparent lack of host galaxy extinction is, perhaps, surprising for two reasons. Firstly, local SNe~Ia are observed with a range of host galaxy extinctions spanning $0 < E(B-V) < 2.0$, with most lying in the range $0 < E(B-V) < 0.5$ \\citep{phi,mkm}.  Secondly, the increasing gas content in galaxies with increasing redshift means that, contrary to the findings of the SN search teams, we might expect mean extinction levels to {\\it increase} with redshift due to increased dust opacities caused by the higher gas column densities \\citep{cal}.  However, it is possible that the apparent lack of extinction at high-z is due to observational bias.  SNe~Ia behind a dust screen will be fainter, and thus would be selected against in searches near the magnitude limit.  In addition, follow-up spectroscopy favours those SNe~Ia that are a large distance from the host galaxy centres in order to minimise contamination from host galaxy light. These SNe~Ia may be expected to have less extinction than those found near the host galaxy centres.  \\begin{table} \\caption{High redshift type Ia supernovae host galaxies \\label{sne1ahosts}} \\begin{tabular}{@{}lccccc} \\hline \\hline Name            & RA         & Dec                      & $z$  & Host   & $r$$^{a}$  \\\\ & hh mm ss   & \\degr\\ \\arcmin\\ \\arcsec\\ &      &        & Kpc  \\\\ \\hline \\hline 1997ce$^{b}$ & 17 07 48.3 & +44 01 26.2              & 0.44 & Spiral    & 3.1  \\\\ 1997cj$^{b}$ & 12 37 04.3 & +62 26 24.9              & 0.50 & E/S0      & 4.8  \\\\ 1997ek$^{c}$ & 04 56 11.6 & -03 41 26.0              & 0.86 & Spiral    & 6.3  \\\\ 1997eq$^{c}$ & 04 58 56.3 & -03 59 29.4              & 0.54 & Sab       & 13.8 \\\\ 1997es$^{c}$ & 08 18 40.7 & +03 13 36.5              & 0.65 & Disturbed & 3.0  \\\\ 1998ba$^{d}$ & 13 43 36.9 & +02 19 30.6              & 0.43 & Spiral    & 3.8  \\\\ 1998bi$^{d}$ & 13 47 44.7 & +02 20 57.2              & 0.75 & Spiral    & 6.2  \\\\ 1998J$^{e}$  & 09 31 10.5 & -04 45 36.5              & 0.83 & Spiral?   & 3.3  \\\\ 1998M$^{e}$  & 11 33 44.4 & +04 05 13.4              & 0.63 & E/S0      & 28.9 \\\\ 1999U$^{f}$  & 09 26 43.0 & -05 37 57.8              & 0.50 & E/S0      & 8.6  \\\\ 2000dz$^{g}$ & 23 30 41.4 & +00 18 42.7              & 0.50 & Spiral    & 3.5  \\\\ 2000ea$^{g}$ & 02 09 54.0 & -05 28 17.8              & 0.42 & Spiral    & 9.0  \\\\ 2000ec$^{g}$ & 02 11 32.0 & -04 13 56.1              & 0.47 & E/S0      & 20.7 \\\\ 2000ee$^{g}$ & 02 27 34.5 & +01 11 49.4              & 0.47 & Spiral    & 4.8  \\\\ 2000eg$^{g}$ & 02 30 21.1 & +01 03 48.5              & 0.54 & Spiral    & 6.4  \\\\ \\hline \\hline \\end{tabular}  \\medskip  J2000 coordinates and redshifts are taken from the relevant IAU circulars. $^{a}$Projected distance in kpc of the SNe~Ia from the host galaxy centres. $^{b}$\\citet{rie1}, $^{c}$\\citet{nug}, $^{d}$\\citet{gre}, $^{e}$\\citet{gar} $^{f}$\\citet{gar2} $^{g}$\\citet{sch}  \\end{table}  In this letter we use HST archive data, and previously published HST observations, to test whether or not SNe~Ia rates and luminosities at $z\\sim0.6$ show a different distribution with host morphology.  We also examine the issue of extinction via the radial distribution of the SNe, and by making use of colour maps of two of the high-z host galaxies.  Sample selection, observations and data analysis are described in \\S2. Results are presented in \\S3, and discussion and conclusions are given in \\S4. We adopt $H_{0}=65$ km s$^{-1}$ Mpc$^{-1}$, $\\Omega_{0}=0.3$ and $\\Lambda=0.7$. ",
        "conclusions": "\\subsection{Supernova rates} To improve the statistics we added seven more SNe~Ia host galaxies at a similar redshift, imaged by \\citet{ell} using STIS on HST. These galaxies are presented in Table \\ref{ellishosts}. Two of these are classified as E/S0 and five as spirals. An eighth galaxy from \\citet{ell}, the host of SN~1997eq, is common to both samples. We agree with \\citet{ell} that it is a spiral type.  Thus, for a combined sample of 22 SNe~Ia host galaxies, $30\\%$ are E/S0 systems and $70\\%$ are spirals or disturbed systems. These proportions are similar to that observed locally by \\citet{cet}. These authors also find a ratio of about $30\\%$ to $70\\%$ when their results are expressed as SN rate per unit bolometric luminosity of the host galaxy.  Adopting the same conversion between SNe~Ia counts per galaxy type and SNe~Ia rates per unit galaxy luminosity as \\citet{cet}, we infer that the relative SNe~Ia rate in spirals and ellipticals has not changed between $z\\sim0.6$ and the local universe.  This is consistent with the prediction of \\citet{ktn} that the relative rates should be unchanged in going from z=0.6 to the local universe.  \\subsection{Supernova peak magnitudes} Peak rest-frame B band magnitudes for the SNe in our sample have so far been published for only two objects \\citep{rie1}; SN 1997ce ($M_{B}^{peak}$=--19.26) and SN 1997cj (--19.25). These are the magnitudes obtained following application of MLCS correction. However, stretch-factor-corrected B band peak magnitudes for 7 out of the 8 host galaxies presented in \\citet{ell} have been published in \\citet{per1}.  These magnitudes are listed in Table \\ref{ellishosts}. The total of 9 SNe~Ia comprises 3 in ellipticals and 6 in spirals. Although the two teams observe their supernovae using different filter sets, the light curve fitting methods employed by both teams give stretch factor corrected magnitudes in the same filter, namely B band in the rest-frame of the supernovae. We thus intercompare published magnitudes from both teams. The mean peak B~band magnitude for the SNe~Ia in the 6 spiral systems is $-19.22^{-0.13}_{+0.14}$. For the 3 SNe~Ia in elliptical systems the mean peak B~band magnitude is $-19.53^{-0.15}_{+0.19}$.  Thus, we find no significant difference in the average light-curve-shape-corrected $M_{B}^{peak}$ for high-z SNe~Ia between spirals and ellipticals. We thus find that light-curve shape correction methods \\citep{sch0} appear to be valid for SNe~Ia at $z\\sim0.6$ in all host galaxy types.  \\subsection{Extinction} The irregular colour structure in two of the host galaxies (Figure \\ref{sn_colmaps}) is most plausibly interpreted as being due to variation in a significant dust extinction (this also supports their classification as spirals). We also note that simulations of host galaxy extinctions and radial distributions of supernovae \\citep{hbd} have shown that SNe~Ia within $10$kpc of the centres of spiral galaxies will be observationally dimmer and have a larger magnitude dispersion than SNe~Ia in the outer regions, although part of the observed dispersion will probably be due to projection effects. Inspection of Figure \\ref{sn_images} shows that the SNe~Ia are found over a range of projected distances from the host galaxy centre. These distances are listed in Table \\ref{sne1ahosts}. It is possible that the real distances from the host galaxy centres are in fact all $>10$kpc and that the observed distances are simply due to projection effects, however we argue against this for two reasons. Firstly, 10 out of the 15 SNe in our sample have projected distances from the host galaxy nuclei of $<7$kpc, which argues against a distribution in a 'shell' of $>10$kpc in radius. Secondly, the distances from galaxy centres for local SNe~Ia range from $\\sim1$kpc to $\\sim30$kpc, and there is no compelling physical reason why the distribution of galactocentric distances of SNe~Ia at z=0.6 should differ from the distribution observed locally.  Overall, our results are consistent with a similar range of real separations of SNe~Ia from the centres of their host galaxies as is observed locally.  We find no evidence to suggest that SNe~Ia at $z\\sim0.6$ are found preferentially far out in the host galaxies where extinction levels might be expected to be small. Moreover, current observational and theoretical evidence does not favour any significant changes in typical galaxy size at $z\\sim0.6$ relative to that observed locally \\citep{lil,boi}. Consequently we can also state that there is no difference in the distribution with galaxy scalelength between local SNe~Ia and those at $z\\sim0.6$. We are led to suspect that a significant fraction of the 22 events may be subject to some extinction.  Given that dustier environments at higher redshifts are also to be expected \\citep{eal,fox}, the general lack of extinction found in high-$z$ SNe~Ia discovered by \\citet{rie1} and \\citet{per1} remains puzzling. We do however note the following two caveats. Firstly, the possibility of some observational bias ({\\it cf.} Section 1) operating within a region closer than $\\sim$10~kpc is not ruled out. Secondly, details about the SNe~Ia in our two ``dusty'' spirals (Figure \\ref{sn_colmaps}) have yet to be published, and so we are unable to comment directly on the degree of extinction that these events might have encountered.  In summary, we find that, at $z\\sim0.6$, the observed relative rates and relative luminosities of type Ia supernovae in elliptical and spiral galaxies are consistent with predictions based on the locally-derived understanding of SNe~Ia physics and the influence of progenitor mass and metallicity \\citep{ume1,ktn}.  We find no reason to question this understanding.  Unless some level of observational bias is present, there is however still some difficulty with the apparently low observed extinction towards SNe~Ia at high-z."
    },
    "0208/astro-ph0208090_arXiv.txt": {
        "abstract": "We determine the total enclosed mass profile from 0.7 to 35 kpc in the elliptical galaxy NGC 4636 based on the hot interstellar medium temperature profile measured using the {\\it Chandra} X-ray Observatory, and other X-ray and optical data. The total mass increases as $r^{1.2}$ to a good approximation over this range in radii, attaining a total of $\\sim 1.5\\times 10^{12}$ M$_{\\odot}$ (corresponding to $M_{\\rm tot}/L_V=40$) at 35 kpc. We find that at least half, and as much as 80\\%, of the mass within the optical half-light radius is non-luminous, implying that NGC 4636 has an exceptionally low baryon fraction. The large inferred dark matter concentration and central dark matter density, consistent with the upper end of the range expected for standard cold dark matter halos, imply that mechanisms proposed to explain low dark matter densities in less massive galaxies (e.g., self-interacting dark matter, warm dark matter, explosive feedback) are not effective in elliptical galaxies (and presumably, by extension, in galaxy clusters). The composite (black hole, stars, and dark matter) mass distribution has a generally steep slope with no core, consistent with gravitational lensing studies. ",
        "introduction": "\\subsection{Context}  The presence of extended dark matter halos in late-type galaxies inferred from analysis of gas and stellar dynamics is well-established; and, attention is now focussed on comparing the detailed mass distribution with theoretical predictions of galactic dark halo structure. In a universe predominantly composed of non-baryonic matter with specified characteristics, the mass distribution is determined by the cosmological world model, the initial density perturbation spectrum, and -- at least on galactic scales -- the dynamical response to the reconfiguration of the secondary, baryonic component that occurs during galaxy formation. Models where non-interacting (except for gravity) cold dark matter (CDM) constitutes the primary mass component are highly successful in providing an explanatory framework for a diverse range of phenomena in the cosmological setting, and are now standard. However a critical re-examination of the CDM paradigm is underway, due in large part to its confrontation with measurements of late-type galaxy mass distributions indicating that dark matter is less concentrated than expected.  Dark matter density distributions in the centers of collapsed structures in CDM-dominated model universes increase at least as steeply as $r^{-1}$ (e.g., Navarro, Frenk, \\& White 1997; Ghigna et al. 2000). However, over a wide range of galaxy luminosity and morphological type -- including luminous spiral, dwarf, and low surface brightness galaxies -- observational data favor mass models with dark matter cores. Some attempt to explain this discrepancy by proposing an alteration in the nature or perturbation spectrum of dark matter in such a way that the density distribution evolves to develop a core, or is initially more diffuse on galaxy scales. Whether any of these variants offer success over the full range of mass scales from dwarf galaxies to galaxy clusters, without conflicting with other astrophysical and particle physics considerations, is an open question \\citep{sil02}. In a separate class of alternative scenarios, the assumption of the cold, non-interacting nature of dark matter is retained; but, dynamical effects from other mass constituents during the galaxy formation epoch reduce the central dark matter concentration. This feedback may take the form of core heating via mergers of two systems with supermassive black holes, or gravitational coupling to powerful outflows in a baryon dominated central proto-galactic region. \\S 5 of the present paper includes details on the observational situation and these proposed explanations.  The form of the dark matter distribution in elliptical galaxies could prove decisive in this debate, since these are the most massive galaxies with the largest supermassive black holes, display strong evidence for undergoing early outflows and, have high central stellar densities. Various alternative explanations for the absence of dark matter cusps in late-type galaxies may very well diverge in their expectations of whether low central dark matter concentrations occur in ellipticals as well.  \\subsection{Dark Matter in Elliptical Galaxies}  A consensus affirming the presence of dark matter in elliptical galaxies is finally emerging from a diversity of observational programs, including optical studies of ionized gas disks \\citep{b93}, dynamical modeling of stellar kinematical data (e.g., Gerhard et al. 2001, Magorrian \\& Ballantyne 2001, \\S 5.3; but, see Baes \\& Dejonghe 2002), gravitational lensing statistical studies, (e.g., Keeton 2001, \\S 5.1.2), and analysis of extended X-ray emitting gas distributions (e.g., Loewenstein \\& White 1999). As is the case for spiral galaxies, dark matter in elliptical galaxies comprises an increasing fraction of the total mass with distance from the galactic nucleus, and does not obviously become the dominant constituent until beyond the optical half-light radius $r_e$.  Progress on constraining the dark matter distribution in elliptical galaxies was impeded by the absence of fine angular resolution measurements, over a sufficiently broad dynamical range in radii, of mass tracers in individual systems.  Data with these qualities are now obtainable, following the launch of the {\\it Chandra} X-ray Observatory with its capability for imaging and imaging spectroscopy on sub-arcsecond scales over a $\\sim 10'$ field-of-view.  In this paper we utilize analysis of {\\it Chandra} observations of NGC 4636 to investigate the dark matter distribution in this elliptical galaxy. Observations and data analysis are described in \\S 2, the mass modeling procedure explained in \\S 3, and the resulting dark matter constraints presented in \\S 4. In \\S5 we compare these constraints with theoretical expectations in the context of standard CDM, with results of gravitational lensing studies, and with observations of galaxies of other Hubble types -- as well as with expectations for elliptical galaxies based on various explanations for the low central dark matter densities in these galaxies. Our conclusions follow in \\S 6. A distance of 15 Mpc to NGC 4636 is adopted \\citep{t01}. ",
        "conclusions": "We have constructed mass models of the elliptical galaxy NGC 4636 based primarily on the density and temperature distributions of the hot gas measured with the {\\it Chandra} X-ray Observatory and the stellar light density distribution measured in the optical. Secondary, observational inputs are the {\\it XMM-Newton}/{\\it ASCA} X-ray temperature profile and the stellar velocity dispersion profile. We derive accurate constraints on the total mass distribution from 0.7--35 kpc.  The total mass increases as $r^{1.2}$ to a good approximation over this range in radii, attaining a total of $\\sim 1.5\\times 10^{12}$ M$_{\\odot}$ (corresponding to $M_{\\rm tot}/L_V=40$, Figure 3) at the outermost point we consider.  Of the models we investigate, the temperature profile is most accurately fit using a dark matter distribution that consists of two-components -- one with a flat core and one of the generalized NFW form with a cusp (equation 4).  We consider this an indication that the dark matter distribution may be more complex than a single component following equation (4). There is no unique mass decomposition into luminous and dark matter components (see Figures 5 and 7); steeper assumed values of the central dark matter density slope ($\\zeta$) imply lower stellar mass-to-light ratios (such a degeneracy exists even for dark matter dominated dwarf galaxies; van den Bosch \\& Swaters 2001). However, constant mass-to-light models and models with $\\zeta\\ge 2$ are too steep, and are ruled out by the {\\it Chandra} data. $M_{\\rm stars}/L_V\\le 6.6$ is indicated from fitting models with dark matter cores, and models with $M_{\\rm stars}/L_V\\le 5.5$ -- consistent with the stellar population in NGC 4636 -- are favored.  While a wide range of central dark matter densities and dark-to-luminous mass ratios are allowed (Figure 4 and 6), all of these are sufficiently restrictive so as to have important implications. The mass profile significantly departs from that of the light inside a few kpc ({\\it i.e.} $\\sim 0.5\\times$ the half-light radius $r_e$). The dark matter fraction is $\\sim 0.5-0.8$ within $r_e$, and therefore on the high concentration end of expectations for CDM halo formation models. Moreover, the central dark matter density is at least an order of magnitude -- and possibly more than two orders of magnitude (at $\\sim 2$ M$_{\\odot}$ pc$^{-3}$, based on the best-fit two-component models) -- greater than reported as typical for a variety of spiral and less luminous, dark matter dominated galaxies.  Non-parametric measures of the dark matter central density also indicate that dark matter in NGC 4636 is more concentrated than average for a standard CDM halo, though within the scatter. Thus, the effects of adiabatic contraction and expansion via explosive feedback are negligible, effectively cancel each other out, or operate on a halo that initially differs from those computed for CDM halos.  The dark matter distribution in NGC 4636 is at odds with many alternative models designed to explain low central densities in galaxies of other Hubble type. It is instructive to consider two classes of models from this perspective -- those with flat dark matter cores, and those with cuspy dark matter cores (that provide better fits to the X-ray data) interpreted as contracting from initially flat cores due to baryonic infall. For the flat-core models, the high central dark matter mass density and large cores are contrary to the expected scaling relations for self-interacting dark matter and other models where dark matter structure is driven by dark matter particle interaction. For the cuspy-core models, the central phase-space density, conserved during adiabatic contraction, is too high. Models where the flattening of dark matter cores occurs only at relatively low mass scales (perhaps, at the low surface brightness galaxy scale; van den Bosch et al. 2000) -- including some involving explosive feedback and warm dark matter -- remain most viable. If this transitional mass is indeed on the giant elliptical galaxy scale (or below), cuspy dark matter distributions in galaxy clusters are implied.  If the results for dark matter dominated, low mass galaxies are correct, there is a reversal from the trend naively expected from a standard bottom-up hierarchy where less massive objects collapse earlier in a more dense universe and retain that higher density.  Perhaps some modification to the standard CDM scenario is required. It should be noted, however, that considerable scatter in dark matter concentration is predicted for a given mass range -- particularly on galaxy scales. Moreover, this scatter is not purely random, but represents the effects of variations in age and assembly history \\citep{w02}. This introduces a bias such that galaxies of the types where low central dark matter densities are derived may represent relatively recently formed systems, and/or particularly fragile galaxies that have experienced relatively tranquil assembly histories. Conversely, giant elliptical galaxies such as, or perhaps particularly, NGC 4636 may have the highest formation redshifts and the most prominent merger histories.  Gravitational lensing probabilities are sensitive to the inner total mass density slope \\citep{wb00,bks,tc02,wts,ots,lo02}, and explainable if the total mass in clusters follows an NFW profile and that in elliptical galaxies a singular isothermal sphere profile \\citep{lo02}. This is consistent with our results, once one takes into account the stellar contribution. If one is to use lensing statistics to constrain cosmological parameters, the correct composite mass model -- derived empirically as we do here -- should be used. Moreover, these studies assume a constant dark matter distribution for galaxies of a given optical luminosity, while our results indicate the likelihood of a significant cosmic variance.  In fact, we suggest that NGC 4636 is an exceptionally dark matter dominated elliptical galaxy, a consequence of a low global baryon fraction or a high concentration of dark, relative to luminous, matter. Although one must therefore exercise caution in generalizing from our results, the most stringent constraints on alternative models for dark halo structure, that presumably universally apply, emerge. Results, in progress, of similar investigations for other galaxies (e.g., NGC 1399, NGC 4472) are anticipated with great interest."
    },
    "0208/astro-ph0208573_arXiv.txt": {
        "abstract": "{ Model computations of $\\delta$~Scuti stars, located in the vicinity of the red edge of the classical instability strip, suggest amplitudes of solar-like oscillations larger than in cooler models located outside the instability strip. Solar-like amplitudes in our $\\delta$~Scuti models are found to be large enough to be detectable with ground-based instruments provided they can be distinguished from the opacity-driven large-amplitude pulsations. There would be advantages in observing simultaneously opacity-driven and stochastically excited modes in the same star. We anticipate their possible detection in the context of the planned asteroseismic space missions, such as the French mission COROT (COnvection ROtation and planetary Transits). We propose known $\\delta$~Scuti stars as potential candidates for the target selection of these upcoming space missions. } ",
        "introduction": "The $\\delta$~Scuti stars are in general main sequence stars with masses between 1.5\\,M$_\\odot$ and 2.5\\,M$_\\odot$. They are located  inside the classical instability strip (IS hereafter) where the $\\kappa$-mechanism drives low-order radial and nonradial modes of low degree to measurable amplitudes (opacity-driven unstable modes).  Only a small number of opacity-driven modes are observed in $\\delta$~Scuti stars \\citep[for a review see e.g.][]{Gautschy96}, but their amplitudes, which are limited by nonlinear processes, are much larger than stochastically driven intrinsically stable solar-like p modes.  For main-sequence stars with surface convection zones, located outside the IS, model computations suggest all modes to be intrinsically stable but excited stochastically by turbulent convection; for models located near the red edge of the IS the predicted velocity amplitudes become as large as 15 times the solar value \\citep{Houdek99}. Moreover, these computations suggest that models located inside the IS can pulsate simultaneously with modes excited both by the $\\kappa$-mechanism and by the turbulent velocity field.  { Provided that many modes can be detected, high-frequency p modes are more easily identified than low frequency p modes. Hence there are advantages of observing simultaneously both types of modes in the same star. As a first step, high-frequency p modes can help to determine the fundamental stellar parameters (e.g., luminosity, effective temperature) more accurately, whereas low-frequency modes, which are strongly sensitive to the properties of the deep layers of the star, can then be used as a diagnostic for the inner properties of the model. Such developments are outside the scope of the present paper and we only outline briefly the underlying idea.  The nearly regular frequency spacing of solar-like modes of high order (i.e., the large frequency separation) depends predominantly on the structure of the surface layers and consequently provides further constrains on the equilibrium models. Their  degree $l$ and azimuthal order $m$ can be identified with the help of the classical echelle diagram method; this method was successfully tested by the simulation results of the COROT Seismic Working Group (Appourchaux, 2002, personal communication); this severely constraints the fundamental stellar parameters (mass, age, chemical composition) of models for which the frequencies of computed oscillation modes are similar to the observed high-order modes \\citep{Berthomieu02}. % Moreover, solar-type modes also provide information on the star's mean rotation rate.  A nearly regular spacing in frequency is also observed for opacity-driven low-frequency modes \\citep{Breger99}; the large separation of these low-frequency modes has to be similar between observations and theoretical models which satisfy also the properties of the observed high-frequency solar-type p modes. However, some of the opacity-driven modes deviate from the mean value of the large frequency separation; these modes are so-called mixed modes which provide details of the stellar core and of the precise evolutionary stage of the observed star \\citep[][ and references therein]{Unno89}. This deviation from the mean value of the large frequency separation could suggest the presence of mixed modes. The problem is further complicated by the fact that the rotational splitting frequency components are no longer equidistant for these fast rotators, i.e. these frequencies could erroneously be identified as frequencies of mixed modes. However, knowing the mean rotation rate from the high-frequency splittings of solar-type p modes, the frequency splittings of the low-frequency opacity-driven modes can be computed in the manner of \\citet{Dziembowski92} \\citep[see also ][]{Soufi98}. }  The understanding of the physics responsible for the return to stability of opacity-driven modes at the red edge of the IS is still in its infancy. As the star becomes cooler the extent of the surface convection zone increases, thereby making the effect of convection-pulsation coupling on mode stability progressively more important. Several authors have tried to model the location of the red edge, e.g., \\citet{Baker79}, \\citet{Bono95} for RR Lyrae stars and e.g., \\citet{Houdek96} and Xiong \\& Deng (2001) for $\\delta$~Scuti stars. Although the authors assumed various models for the time-dependent treatment of convection, they all concluded that convection dynamics crucially effect the location of the red edge; however, different results were reported as to whether the convective heat flux (e.g., \\citealp{Bono95}), the momentum flux (e.g., Houdek,\\,1996) or turbulent viscosity (Xiong \\& Deng,\\,2001) is the crucial agent for stabilizing the modes at the red edge. In all these investigations, the predicted position of the red edge depends crucially on the assumed convection parameters, such as the mixing-length parameter or whether acoustic emission is included or neglected in the equilibrium model \\citep{Houdek00}.  Although it is possible from Fig.~(13) of \\citet{Houdek99} to conclude that both types of modes can be excited simultaneously in the same star, amplitudes of stochastically excited modes for stars located inside the instability strip were not explicitly carried out by \\citet{Houdek99} and their possible detection were not addressed.  The aim of this paper is to demonstrate that models of stars, located inside the IS and near the red edge, can exhibit both opacity driven modes and solar-like oscillations with sufficiently large amplitudes to be detectable with today's ground-based instruments. Consequently the planned asteroseismology space missions, such as COROT \\citep[COnvection ROtation and planetary Transits, ][] {Baglin98} or Eddington \\citep{Favata00}, will detect these oscillations even more easily.  Sect.~2 describes the equilibrium models, and the linear analysis results are discussed in Sect.~3., which are obtained from solving the equations of linear nonadiabatic oscillations in which convection is treated with the time-dependent, nonlocal formalism by \\citet[][hereafter G'MLT]{Gough76,Gough77}. Furthermore, the effect of acoustic radiation in the equilibrium model on the stability properties is taken into account in the manner of \\citet[ and references therein]{Houdek00}. In this paper we consider only radial p modes.  Amplitudes of solar-like oscillations result from the balance between damping and stochastic driving by turbulence. The rate at which the turbulence injects energy into the p modes is estimated in the manner of \\citet[][ Paper~I hereafter]{Samadi00I} and is discussed in Sect.~4.  In Sect.~5 we address the possibilities and conditions for detecting solar-type oscillations in $\\delta$~Scuti stars with ground-based instruments and propose possible candidates, some of which are listed in the catalogue by \\citet{Rodriguez00}. Conclusions are given in Sect.~6. ",
        "conclusions": "We studied oscillation properties in $\\delta$~Scuti stars located near the observed red edge of the classical instability strip. Such stars can pulsate with both opacity-driven modes and intrinsically stable stochastically driven (solar-like) modes. The estimated velocity amplitudes of the stochastically driven modes in our $\\delta$~Scuti models are found to be larger than in cooler and pulsationally stable models lying outside the IS. This result supports the idea that solar-like oscillations in $\\delta$~Scuti stars may be detected.  Including a model for the acoustic radiation in the equilibrium model results in a cooler red edge and does effect the properties of the excitation rate of p modes (see also \\citealp{Houdek98}, \\citealp{Houdek00}); in particular the pulsation amplitudes do become larger and are predicted to be largest for a model with the largest acoustic flux $F_{\\rm ac}$ (i.e., model A2). {Moreover, for the $\\delta$~Scuti models considered in this paper, overstable modes were predicted only if either acoustic emission in the mean stratification was included or if the mixing-length parameter was reduced to a value smaller than suggested by a calibrated solar model.}  A potential target star should neither be too cool (i.e., no opacity-driven modes) nor too hot (i.e., stochastically excited modes with amplitudes too small to be detectable). We quantify this with the illustrative case of our $\\delta$~Scuti models with a mass $M =1.68\\,$M$_\\odot$ and we identify the following $\\delta$~Scuti stars from the \\citet{Rodriguez00} catalogue, located near the red edge, as potential candidates for the target selection of upcoming observing campaigns: HD57167, HD14147, HD208999 and HD105513.  Although the amplitudes of the solar-type oscillations, predicted in our $\\delta$~Scuti models, are large enough to be detected from ground, today's ground-based instruments will detect such oscillations only in brighter $\\delta$~Scuti stars with an apparent magnitude of up to $V\\sim 3-4$ (Bouchy, 2001, personal communication). However, new ground-based observing campaigns, such as the HARPS project \\citep{Bouchy01} will be able to detect stochastically excited oscillations in  $\\delta$~Scuti  stars with an apparent magnitude of up to $V \\sim 4-5$. Unfortunately, there are no such bright stars in the \\citet{Rodriguez00} catalogue which are located near the red edge, although some bright stars near the red edge may have opacity-driven modes with amplitudes too small to be detectable with today's ground-based instruments and are therefore not classified as $\\delta$~Scuti stars.  The forthcoming space missions for asteroseismology, such as COROT \\citep{Baglin98} and Eddington \\citep{Favata00} will be able to detect solar-like oscillations in faint $\\delta$~Scuti stars. The large instrument on the Eddington spacecraft will measure stellar oscillations with amplitudes as small as $1.5$~ppm in stars with an apparent magnitude of $V \\simeq 11$ assuming an observing period of 30 days. Moreover, Eddington's large field of view % will allow it to monitor a large number of stars simultaneously. This will be helpful for detecting and classifying new $\\delta$~Scuti stars and for measuring the location of the red edge of the IS with greater precision than it was possible before."
    },
    "0208/astro-ph0208353_arXiv.txt": {
        "abstract": "We develop the formalism to include substructure in the halo model of clustering. Real halos are not likely to be perfectly smooth, but have substructure which has so far been neglected in the halo model --- our formalism allows one to estimate the effects of this substructure on measures of clustering. We derive expressions for the two-point correlation function, the power-spectrum, the cross-correlation between galaxies and mass, as well as higher order clustering measures.  Simple forms of the formulae are obtained for the limit in which the size of the substructure and mass fraction in it is small. Inclusion of substructure allows for a more accurate analysis of the statistical effects of gravitational lensing.  It can also bring the halo model predictions into better agreement with the small-scale structure seen in recent high resolution simulations of hierarchical clustering. ",
        "introduction": "Sheth \\& Jain (1997) described how the halo model for clustering can allow one to model the distribution of matter in the highly nonlinear regime.  The model falls within the broader framework described by Neyman \\& Scott (1954) and Scherrer \\& Bertschinger (1991).  It combines results from Peebles (1974) and McClelland \\& Silk (1977) with the work of Press \\& Schechter (1974), and is able to provide a good description of nonlinear clustering seen in numerical simulations of hierarchical gravitational clustering. The halo model assumes that most of the mass in the Universe is bound up in virialized dark matter halos, and that statistical measures of clustering on small scales are dominated by the internal structure of the halos. The agreement with simulations shows that is possible to provide an accurate description of clustering in the small-scale nonlinear regime even if one has no knowledge of if and how the halos themselves are clustered.  This halo--model of clustering has been the subject of much recent interest (e.g. Seljak 2000; Peacock \\& Smith 2000; Ma \\& Fry 2000; Scoccimarro et al. 2001; Cooray \\& Sheth 2002).  To date, almost all analytic work based on the halo--model approach assumes that halos are spherically symmetric, and that the density run around each halo center is smooth.  Halos which form in numerical simulations of hierarchical clustering are neither spherically symmetric nor smooth (e.g., Navarro, Frenk \\& White 1997; Moore et al. 1999; Jing \\& Suto 2002). About ten percent of the mass of a halo is associated with subclumps (Tormen, Diaferio \\& Syer 1998; Ghigna et al. 1999). The main purpose of the present work is to derive a model which accounts for this substructure.  Section~\\ref{themodel} shows how to compute two-point statistics, the correlation function and its Fourier transform, the power spectrum, when substructure is important.  Section~\\ref{higher} shows that the model can be easily extended to estimate higher-order statistics. Section~\\ref{averages} shows how our formalism can incorporate a range of parent halo and subclump masses, and Section~\\ref{details} provides a few explicit examples.  This section includes a discussion of how to model the shape of the subclump mass function. Section~\\ref{discuss} summarizes our results, and suggests various other applications of our formalism. ",
        "conclusions": "\\label{discuss}  We have shown how to incorporate the effects of substructure into the halo model description of the nonlinear density field.  Accounting for this substructure is important on scales smaller than the virial radii of typical halos.  The effects are more pronounced for statistics which treat the subclumps preferentially, such as the power spectrum measured in studies of weak galaxy--galaxy gravitational lensing. Substructure will also change the dynamics within halos. Although we have not done so here, it is straightforward to insert our model for substructure into the halo model of the cosmic virial theorem, and the mean pairwise velocity and velocity dispersion developed in Sheth et al. (2001).  The stable clustering limit is a physically appealing description of clustering on small scales (Peebles 1980).  It has been argued that a model with smooth halos is inconsistent with this limit (Ma \\& Fry 2000; Scoccimarro et al. 2001). Substructure changes the shape of the small scale power spectrum (c.f., Fig.~\\ref{pksubclumps}); at least in principle, it can bring the halo model predictions into agreement with the stable clustering solution.  However, it is not obvious that stable clustering is, indeed, the correct physical limit.  Smith et al. (2002) argue that the stable clustering assumption is inconsistent with the results of high resolution numerical simulations.  They also find that the simulations do not follow the small scale scaling predicted by models in which halos are smooth.  Once an accurate model of the subclump mass function is available, it will be interesting to compare the predictions of our description of substructure with their results.  Although we have focussed primarily on the implications of substructure for the halo model of nonlinear clustering, our results have a wide range of other applications.  For example, excess power in the Fourier transforms of images of galaxies or distant clusters can be used to infer the existence of spiral arms or substructure.  This is the subject of work in progress.  Closely related is the question of what images of high redshift galaxies may look like.  Observations through a filter which has a fixed wavelength range probe the emission from high redshift galaxies at shorter restframe wavelengths than for galaxies at low redshift. If obscuration by dust is not a problem, and the UV luminosity is dominated by patchy star forming regions, then the images of high redshift galaxies should show considerable substructure.  Our results suggest that, in this case, the power spectrum obtained by Fourier transforming the image of a high redshift patch of sky should show an increase in small scale power.  In addition, although we have phrased the entire discussion of substrucutre in terms of spatial statistics, this is not really necessary.  Large databases describing various observed characteristics of galaxies are now becoming available (e.g., the 2dFGRS and SDSS surveys).  If some of $n$ observables are correlated with others, the data will not fill the full $n-$dimensional space available:  the data set itself can be thought of as being clumpy, and the various clumps in dataspace may themselves have substructure.  The formalism developed here provides a way of discovering, quantifying and modelling such substructure.  \\bigskip  \\noindent We thank Masahiro Takada for helpful discussions. B.J. acknowledges financial support from a NASA-LTSA grant and a Keck foundation grant."
    },
    "0208/astro-ph0208486_arXiv.txt": {
        "abstract": "We use deep near-infrared and submillimeter observations of three massive lensing cluster fields, A370, A851, and A2390, to determine the average submillimeter properties of a $K'$-selected sample. The 38 Extremely Red Objects (EROs; $I-K'>4$) with $K'<21.25$ have a very significant error-weighted mean 850~$\\mu$m flux of $1.58\\pm0.13$~mJy. The ERO contribution to the 850~$\\mu$m background is $1.88\\pm 0.16\\times 10^4$~mJy~deg$^{-2}$, or about half the background light. The 17 Very Red Objects (VROs; $3.5<I-K'<4$) are also significantly detected ($1.32\\pm 0.19$ mJy), bringing the combined VRO and ERO contribution to $2.59\\pm 0.19 \\times 10^4$~mJy~deg$^{-2}$. There is a substantial systematic uncertainty (about a factor of two) in this value due to field-to-field variation, but even with this uncertainty it is clear that a large fraction of the 850~$\\mu$m background arises from red objects. An analysis of the VRO and ERO number counts shows that half of the population's 850~$\\mu$m light arises in objects with demagnified magnitudes $K'<20$ and half in fainter objects. On the basis of the $I-J$ versus $J-K'$ colors of the galaxies, the bulk of the submillimeter signal appears to arise from the dusty starburst galaxies in the red object population rather than from the high-redshift elliptical galaxies. ",
        "introduction": "\\label{secintro}  Extremely Red Objects (EROs) have been the subject of many studies since their discovery (e.g., \\markcite{cowie90}Cowie et al.\\ 1990; \\markcite{hu94}Hu \\& Ridgway 1994). Two scenarios can explain the unusally red colors of the EROs: old stellar populations at $z>1$ (e.g., \\markcite{dunlop96}Dunlop et al.\\ 1996; \\markcite{spinrad97}Spinrad et al.\\ 1997; \\markcite{soifer99}Soifer et al.\\ 1999; \\markcite{cimatti02}Cimatti et al.\\ 2002) or dust-enshrouded galaxies (e.g., \\markcite{graham96}Graham \\& Dey 1996; \\markcite{cimatti98}Cimatti et al.\\ 1998; \\markcite{smith01}Smith et al.\\ 2001). The starlight from dusty galaxies is reprocessed and reemitted in the rest wavelength far-infrared band. For high redshift sources, this light is redshifted into the submillimeter. Recently, a number of possible ERO counterparts to submillimeter detected sources have been discovered (e.g., \\markcite{smail99}Smail et al.\\ 1999, 2002; \\markcite{barger00}Barger, Cowie, \\& Richards 2000; \\markcite{ivison02}Ivison et al.\\ 2002).  The {\\it COBE} satellite found that the submillimeter extragalactic background light (EBL) has approximately the same integrated energy density as the optical EBL (\\markcite{puget96}Puget et al.\\ 1996; \\markcite{fixsen98}Fixsen et al.\\ 1998); thus, a large fraction of starlight is reradiated by dust. To determine the high-redshift star formation, the sources that comprise the submillimeter EBL need to be determined. Here we use deep submillimeter and near-infrared (NIR) imaging of three massive lensing cluster fields to determine the submillimeter properties of the ERO population. Two advantages of observing cluster fields are that background sources are magnified and mean source separations on the sky are increased.  In \\S~\\ref{secdata} we briefly describe our $I$, $J$, $K'$, and 850~$\\mu$m data. The $15''$ beam size of SCUBA (\\markcite{holland99}Holland et al.\\ 1999) on the 15~m James Clerk Maxwell Telescope\\footnote{The James Clerk Maxwell Telescope is operated by the Joint Astronomy Centre on behalf of the UK Particle Physics and Astronomy Research Council, the Netherlands Organization for Scientific Research, and the Canadian National Research Council} prohibits us from making individual matches between submillimeter and optical/NIR sources. However, in \\S~\\ref{secebl} we analyze our data statistically and find that EROs, as a class, have a significant submillimeter flux and produce much of the 850~$\\mu$m EBL.  In \\S~\\ref{secfaint} we use our NIR photometry and lensing corrections to calculate the ERO surface densities to $K'=24$. We analyze where in $K'$ magnitude the 850~$\\mu$m flux arises. Using the $I$, $J$, $K'$ colors, we differentiate between evolved galaxies and dust-reddened starbursts to show that the 850~$\\mu$m light primarily arises in the latter class of object. ",
        "conclusions": "From an optical, NIR, and submillimeter survey of three massive lensing cluster fields we find that VROs and EROs mark, on average, strong submillimeter emitters. The $I-K'>3.5$ population contributes $1.94\\pm 0.15\\times 10^4$~mJy~deg$^{-2}$ to the 850~$\\mu$m EBL if we exclude the A2390 field because of the anomalously high number of red objects, and $2.59\\pm 0.19\\times 10^4$~mJy~deg$^{-2}$ if we include the A2390 field. These contributions are, respectively, $44-62$ percent and $59-83$ percent of the 850~$\\mu$m EBL (the ranges depend on the measurement of the EBL assumed) and are much larger than the $12-16$ percent from galaxies containing bright AGN (\\markcite{barger01}Barger et al.\\ 2001). ERO clustering is an important effect that must be accounted for if we are to obtain a better estimate. The median magnitude of the sources giving rise to the EBL is $K'=20$ to $K'=20.3$, depending on the cluster sample used. The 850~$\\mu$m EBL contribution primarily arises in the subset of the population corresponding to dusty starbursts.  \\newpage"
    },
    "0208/astro-ph0208165_arXiv.txt": {
        "abstract": "We report the first submillimeter interferometric observations of an ultraluminous infrared galaxy. We observed Arp~220 in the CO J=3-2 line and 342~GHz continuum with the single baseline CSO-JCMT interferometer consisting of the Caltech Submillimeter Observatory (CSO) and the James Clerk Maxwell Telescope (JCMT). Models were fit to the measured visibilities to constrain the structure of the source. The morphologies of the CO J=3-2 line and 342~GHz continuum emission are similar to those seen in published maps at 230 and 110~GHz. We clearly detect a binary source separated by $\\sim1\\arcsec$ in the east-west direction in the 342~GHz continuum. The CO J=3-2 visibility amplitudes, however, indicate a more complicated structure, with evidence for a compact binary  at some velocities and rather more extended structure at others. Less than 30\\% of the total CO J=3-2 emission is detected by the interferometer, which implies the presence of significant quantities of extended gas. We also obtained single-dish CO J=2-1, CO J=3-2 and HCN J=4-3 spectra. The HCN J=4-3 spectrum, unlike the CO spectra,  is dominated by a single redshifted peak. The HCN J=4-3/CO J=3-2, HCN J=4-3/HCN J=1-0 and CO J=3-2/2-1 line ratios are larger in the redshifted (eastern) source, which suggests  that the two sources may have different physical conditions. This result might be explained by the presence of an intense starburst that has begun to deplete or disperse the densest gas in the western source, while the eastern source harbors undispersed high density gas. ",
        "introduction": "\\noindent  Ultraluminous infrared galaxies contain extraordinary nuclear starbursts and, at least in some cases, an active galactic nucleus, all hidden within a dense shroud of gas and dust. Understanding the mechanisms which produce and power these luminous galaxies has taken on added urgency with the discovery of their young counterparts at cosmological distances \\citep{ivison}. The nearest and prototype ultraluminous infrared galaxy, Arp~220, is located at a distance of 73 Mpc ($H_o = 75$ km~s$^{-1}$) and is one of the best studied of this class of galaxies. The presence of tidal tails observed in the optical \\citep{arp, joseph} as well as two compact emission peaks at near-infrared \\citep{scoville98}, millimeter (Scoville, Yun, \\& Bryant 1997, hereafter \\citet{scoville97}; Downes \\& Solomon 1998, hereafter \\citet{downes}; Sakamoto et al. 1999, hereafter \\citet{sakamoto}), and radio wavelengths \\citep{becklin, norris, sopp} suggests that Arp~220 is a recent merger. Given the observed correlation between high infrared luminosity and disturbed optical morphologies indicative of galaxy interactions \\citep{mirabel}, this merger is likely responsible for the extremely high infrared luminosity of Arp~220 [$1.4\\times10^{12}$~L$_\\sun$, \\cite{soifer}]. There has been much debate as to whether the high infrared luminosity is due to a starburst or an active nucleus or a combination of the two, both for Arp~220 in particular and for ultraluminous infrared galaxies in general \\citep{genzel, scoville97,lutz}. Recent radio studies of Arp~220 provide support for the starburst hypothesis: the 18 cm flux seen in VLBI observations is thought to be emitted by luminous radio supernovae \\citep{smith} and the effects of these supernovae winds are seen in X-rays \\citep{heckman}. Near-infrared \\citep{emerson, rieke,  sturm, lutz} and earlier radio observations \\citep{condon, sopp, baan} further support the starburst scenario.  Besides its extremely high infrared luminosity, Arp~220 also contains large amounts of dust and molecular gas ($\\sim 9\\times 10^{9}M_\\sun$ \\citet{scoville97}); large gas masses of $4-40\\times 10^9 M_\\sun$ are typical for ultraluminous infrared galaxies \\citep{sanders}. The extinction at optical wavelengths is estimated to be at least $A_V \\sim 50$ and possibly as high as $A_V \\sim 1000$ \\citep{sturm, downes}; even at near-infrared wavelengths (2.2$\\mu$m), dust lanes obscure the possible nuclei \\citep{scoville98} These high extinctions mean that radio observations are needed to probe the deep interior regions. \\citet{downes} have mapped Arp~220 in CO J=2-1 and 1.3 mm continuum; besides detecting two emission peaks, they also see an extended disk (with full-width half-maximum extent of $2\\arcsec \\times 1.6\\arcsec$). They interpret the emission peaks as the nuclei of the merging galaxies embedded in a more extended disk of molecular gas. \\citet{sakamoto} refine the model further using their CO J=2-1 and continuum observations; in this model, the nuclei are each embedded in their own gas disk which is counter-rotating in the larger common disk. An alternative interpretation of the emission peaks as being due to crowding in the orbits of molecular gas and stars is given by \\cite{eckart}.   This paper presents the very first interferometric submillimeter observations of Arp~220 in the CO J=3-2 line and 0.88~mm continuum. In fact, it is the first paper to present data from submillimeter interferometry of any extragalactic source. We also present single dish observations of Arp~220 of HCN J=4-3, a high density tracer. Furthermore, to determine the total CO flux of Arp~220, which could be partly resolved out by the interferometer, complementary single dish data were taken in CO J=3-2 as well as in the CO J=2-1 line. These submillimeter observations allow us to penetrate deep into the interior of the nuclei, while at the same time tracing hotter gas, which may be more closely associated with the source of the ultraluminous infrared luminosity. We describe the observations in \\S~\\ref{sec: observations} and the data analysis in  \\S~\\ref{sec: modeling}. The interferometric data were obtained with the single baseline CSO-JCMT interferometer, the only currently available submillimeter interferometer with sufficient bandwidth to observe the broad emission lines of Arp~220. With data from a single baseline, mapping is not possible and so we analyze the data by making fits in the visibility plane. The data are compared to published data at lower frequencies and to single dish data in \\S~\\ref{sec: discussion} and the conclusions are presented in \\S~\\ref{sec: conclusion}. ",
        "conclusions": "\\label{sec: conclusion}  We have presented the first interferometric observations of Arp~220 at submillimeter wavelengths. The interferometric visibilities of the CO J=3-2 line and 342~GHz continuum are largely consistent with the emission morphology seen previously at lower frequencies. We clearly detect continuum and CO J=3-2 emission from at least two sources separated by $\\sim 1\\arcsec$ at P.A. $\\sim80^o$. The CO J=3-2 visibility amplitudes show additional extended structure with a complex morphology. Masses, column densities, volume densities and optical extinction calculated for both emission sources agree with previous estimates within the errors and underline that the center of Arp~220 contains large amounts of molecular gas ($\\sim8~10^9$ M$_\\odot$).   Single-dish data indicate that the CO J=3-2 emission is moderately extended compared to the 15$\\arcsec$ beam of the JCMT. Though the continuum visibilities show no sign of an extended source, the single dish continuum flux is about twice that detected with the interferometer. In single dish data, the HCN J=4-3/J=1-0, HCN J=4-3/CO J=3-2 and CO J=3-2/J=2-1 ratios are all larger for the redshifted portion of the line than for the blueshifted portion. These observations suggest that the redshifted eastern source is denser and/or hotter than the blueshifted western source. This results could provide support for \\citet{downes} hypothesis that the western source is currently undergoing an intense starburst that has dispersed the dense gas, whereas the eastern source still harbors dense molecular material."
    },
    "0208/astro-ph0208215_arXiv.txt": {
        "abstract": "We study the number and interaction rates of supermassive black holes in galactic bulges as predicted by hierarchical models of galaxy formation in which the spheroidal components of galaxies are formed by mergers. In bright ellipticals, the number of events that can eject a central supermassive binary black hole is large.  Central binaries must  therefore merge in less than a Hubble time -- otherwise there will be too much  scatter in the \\msigma relation and too many off-center galactic nuclei. We propose that binary black holes are able to merge during the major gas  accretion events that trigger QSO activity in galaxies. If this is the case, less than  10 percent of faint ellipticals and 40 percent of bright ellipticals are predicted to harbour binary black holes with near equal masses at their centres. This binary may be ejected away from the centre of the galaxy or even into intergalactic space in up to 20\\% of the  most luminous ellipticals. The number of low mass black holes that can interact with the central object is predicted to be a strong function of galaxy luminosity. In most faint ellipticals, no black holes fall into the centre of the galaxy after the last major gas accretion event, but in the most luminous ellipticals, an average of 10 low mass black holes interact with the central supermassive object after this time. It is expected that stars will be ejected from galaxy cores as these low mass-ratio binaries harden. Multiple black holes in galactic bulges thus  provide a natural explanation for the strong systematic trends in the observed central density profiles of ellipticals as a function of luminosity. ",
        "introduction": "According to the standard paradigm of structure formation in the Universe, galaxies merge frequently as their dark matter halos assemble. Frequent galaxy mergers will inevitably lead to the  the formation of supermassive binary black holes (e.g. Miloslajevic \\& Merritt 2001). It is not clear if these supermassive binaries will be able to  merge on a  Hubble time (Begelman, Blandford \\& Rees 1980; Milosavljevic \\& Merritt 2001; Yu 2002). If  merging timescales are long,  the binary is likely to interact with new infalling black holes through  gravitational slingshot interactions (Saslaw, Valtonen \\& Aarseth  1974). There is some circumstantial observational evidence that binary black holes do merge (e.g.  Meritt \\& Ekers 2002) , but very little is known about the rate at which these mergers occur.  A number of authors have presented models for the growth  of supermassive black holes in  merging galaxies in a cold dark matter (CDM) Universe (Cattaneo, Haehnelt\\& Rees 1999;  Kauffmann \\& Haehnelt 2000 (KH2000); Monaco, Salucci \\& Danese 2000; Menou,  Haiman \\& Narayanan 2001; Hein\\\"am\\\"aki 2001; Volonteri, Haardt \\& Madau 2002, Cattaneo 2002). Here, we extend the models of KH2000, which  explicitly followed the star formation  histories and dynamical evolution of bulges. In a subsequent paper, Haehnelt \\& Kauffmann (2000) demonstrated that the same model could also reproduce the observed \\msigma relation (Gebhardt et al. 2000, Ferrarese \\& Merrit 2000). KH2000 made the extreme assumption that  black holes merge instantaneously  when a bulge forms through a merger of two galaxies of roughly equal mass. In this Letter we relax this assumption and we discuss some of the issues which will influence our predictions of the multiplicity of black holes  in galactic bulges. We also explore the possibility that the merging of supermassive binary black holes in bright ellipticals is responsible for the observed cusp/core  dichotomy of elliptical galaxies (Gebhardt et al. 1996; Merritt 2001). ",
        "conclusions": "According to hierarchical merger models, the formation of supermassive binary black holes will be common. Unless the majority of these binaries merge faster than a Hubble time, further infall of new black holes will lead to an ejection rate of these  binary systems that is too large  to be consistent with the small scatter in the \\msigma relation. There is also no observational evidence for the existence of a significant fraction of galaxies with off-centre galactic nuclei. The accretion of gas and binary hardening by stars are the two main candidates for driving the merger of these binaries.  We have assumed  that the former is efficient if the accreted gas mass exceeds the total mass of the supermassive binary. This reduces the  fraction of faint ellipticals with binary black holes to less than 10\\% and the fraction of bright galaxies that harbour binaries to less than 40\\%. Up to  20\\% of the brightest ellipticals have  recently ejected a binary from their centres. We  thus predict that a small fraction  of  bright ellipticals do not contain a central black hole with a mass that fits onto the \\msigma relation. We note, however, that the fraction of binaries that are truly ejected from the galaxy depends sensitively on the   time evolution of the hardening and on the detailed orbital structure at the centre of galactic bulges. Gas accretion in cooling flows may also prevent efficient binary ejection.  Finally, we have demonstrated that the total number of expected low mass ratio black hole mergers is a strong function of bulge luminosity. A strong transition from no mergers to many mergers occurs at the  same characteristic luminosity ($M_V =-20$) where ellipticals transition from cuspy to core-dominated central profiles. This supports the hypothesis that the ejection of stars by central supermassive binary black holes is indeed responsible for the shallow central density profiles observed in luminous elliptical galaxies.      \\label{lastpage}"
    },
    "0208/astro-ph0208023_arXiv.txt": {
        "abstract": "We present predictions for two statistical measures of the hydrogen reionization process at high redshift. The first statistic is the number of neutral segments identified in spectra of high redshift QSOs as a function of their length. The second is the cross-correlation of neutral regions with possible sources of ionizing radiation.  These independent probes are sensitive to the topology of the ionized regions. If reionization proceeded from high to low density regions then the cross-correlation will be negative, while if voids were ionized first then we expect a positive correlation and a relatively small number of long neutral segments. We test the sensitivity of these statistics for reionization by stars in high redshift galaxies.  The flux of ionizing radiation emitted from stars is estimated by identifying galaxies in an N-body simulation using a semi-analytic galaxy formation model.  The spatial distribution of ionized gas is traced in various models for the propagation of the ionization fronts. A model with ionization proceeding from high to low density regions is consistent with the observations of Becker et al. (2001), while models in which ionization begins in the lowest density regions appear to be inconsistent with the present data. ",
        "introduction": "Early hydrogen reionization is a particularly interesting process in the high redshift universe and is inevitably linked to the appearance of the first star forming objects, at least those that served as sources of the ionizing radiation.  If it occurred early enough, reionization imprints distinct features in maps of the cosmic microwave background (CMB) on arcminute angular scales \\citep{vishniac87,bruscoli, BNSL}. Several aspects of hydrogen reionization remain uncertain despite the rapidly accumulating data on the high redshift universe. For example, it is unclear what objects produce most of the ionizing radiation, although high redshift galaxies are very strong candidates \\citep{CR86,haiman96,ciardi99}. It is also unclear how the ionized regions develop in space \\citep{miralda}.  The ionizing sources are likely to lie in high density regions, but those regions do not necessarily ionize first; the ionizing photons may tunnel into less dense regions and ionize those first.  Further, the duration of reionization is unknown and only a lower limit on the redshift marking the end of that epoch exists \\citep{becker,gunn65}.  In previous papers (Benson et al. 2001, Liu et al. 2001) we examined how the CMB is affected by the reionization process and how future CMB maps can be used to extract information on that process.  However, hydrogen reionization is currently best probed by spectra of high redshift QSOs. Unlike maps of the CMB which are sensitive to line of sight integrals over the density and velocity of ionized gas, QSO spectra contain direct information on the local distribution of neutral hydrogen.  Recently, Becker et al. (2001) analyzed spectra of a sample of QSOs with redshifts between $z=5.82$ and $z=6.28$. In the spectrum of their highest redshift QSO ($z=6.28$), the transmitted flux in the \\op\\ and \\ob\\ forest in the redshift stretch $5.95<z<6.16$ is consistent with zero, with a lower limit of 20 on the \\op optical depth. This long stretch in redshift corresponds to a comoving distance of $60\\mpc$ in a universe with a cosmological constant of $\\Lambda_0=0.7$ and matter density of $\\Omega_0=0.3$. Becker et al. suggest that this long neutral region is a detection of the end of the hydrogen reionization era.  To increase the signal-to-noise ratio, Becker et al. binned their spectra in $4$\\AA\\ pixels. This prevented them from detecting small scale dark windows that are likely to appear in the spectra as left overs from the reionization epoch.  A high resolution spectrum for one of the quasars observed by Becker et al. was obtained by Djorgovski et al. (2001). The spectrum of this quasar ($z=5.73$) was thorough analyzed by Djorgovski et al. (2001) and was found to contain several small dark windows signifying the detection of the trailing edge of the reionization epoch.  Motivated by the results of Becker et al. (2001) and Djorgovski et al. (2001), we examine here how the key ingredients in the reionization process can be probed by QSO spectra and future high redshift galaxy surveys.  The methodology of the present paper has been previously developed in Benson et al. (2001, hereafter BNSL). As in BNSL, we obtain the distribution of ionized gas in an N-body simulation.  Using a semi-analytic model for galaxy formation \\citep{kauff93,somerville99,coleetal99} we identify mock galaxies in the simulation and estimate the ionizing flux emitted by stars at high redshift.  We then use several schemes to follow the development of ionized regions in the simulation.  Using the output of this procedure we obtain predictions for two statistics. The first is the expected number of neutral segments longer than a given length, and the second is the cross-correlation function between the galaxies and neutral regions. Both of these measures rely on observations of QSO spectra, and the latter also on a sample of high redshift candidates for the sources of ionizing radiation. ",
        "conclusions": "The statistic $N(>L)$ giving the number of neutral segments of length greater than $L$ for a given total length of a QSO spectrum is sensitive to the filling factor and the way in which ionization proceeds.  The cross correlation between candidate ionizing sources and neutral regions is less sensitive to the filling factor but is a more direct and robust probe of the propagation of the ionization fronts. In addition to QSO spectra, the cross-correlation requires a sample of candidates (galaxies and QSO) for the ionizing radiation. Catalogs of galaxies and QSO at high redshift are rapidly accumulating, making it possible to compute the QSO-flux and galaxy-flux correlations. Comparison between galaxy-flux and QSO-flux correlation functions  will tell us whether galaxies or QSO  contributed most of the ionizing radiation.  Current observations do not allow a robust determination of $N(>L)$, The number of spectra needed to determine $N(>L)$ to within a given accuracy can be estimated by noting that the relative error on this function is $1/\\sqrt{M N}$, where $M$ is the number of observed QSO spectra covering the same redshift range.  Nevertheless we still can make general conclusions based on the Becker et. al.  (2001) result, assuming that the long Gunn-Peterson trough they observe is indeed a signature of reionization.  Let us take the length of a spectrum in the \\op forest at $z\\approx 6$ to be $\\sim 250\\mpc$ corresponding  to the comoving distance between \\op and \\ob emission lines.  An inspection of Figure \\ref{figure1} shows that: $1)$ A completely neutral stretch of a comoving length of $60\\mpc$ at $z\\approx 6.2$ is inconsistent with a large filling factor. 2) The observations indicate that the chances of finding long neutral regions at $z<5.94$ are tiny while they are significant at higher redshift.  This behavior seems inconsistent with our theoretical $N(>L)$ computed with constant $\\fesc$. If $\\fesc=0.01$ then there are similar probabilities for finding long segments at $z=6.22$ and $z=5.80$.  If $\\fesc=0.05$ then the box is fully ionized at $5.80$, while at $z=6.22$ long segments are very rare.  Therefore the data favor models in which there is a significant increase in the amount of ionizing radiation in the IGM, either due to an increasing escape fraction over this redshift range, or a much stronger evolution in the galaxy/QSO population than is predicted by our model. And, $3)$ a model in which ionization proceeds from low to high density regions seems to be inconsistent with a $\\sim 60 \\mpc$ neutral region even for escape fractions as low as $\\fesc=0.01$."
    },
    "0208/astro-ph0208509_arXiv.txt": {
        "abstract": "We study the origin of the non-thermal emission from the intracluster medium, including the excess hard X-ray emission and cluster-wide radio haloes, through fitting two representative models to the Coma cluster. If the synchrotron emitting relativistic electrons are accelerated {\\it in situ} from the vast pool of thermal electrons, then a quasi-stationary solution of the kinetic equation with particle acceleration through turbulence at high energies ($>200$\\,keV) naturally produces a population of supra-thermal electrons responsible for the excess hard X-ray emission through bremsstrahlung. Inverse Compton scattering is negligible at hard X-ray energies in this case. The radio halo flux density constrains the magnetic field strength to a value close to that of equipartition $\\sim 1 \\mu$G. Alternatively, if the relativistic electrons are injected from numerous localised `external' sources then the hard X-rays are best explained by inverse Compton scattering from GeV electrons, and little of the hard X-radiation has a bremsstrahlung origin. In this case, the magnetic field strength is constrained to $\\sim 0.1-0.2\\,\\mu$G.  Both models assume that the non-thermal emissions are generated by a single electron spectrum, so that only two free parameters, well constrained by the observed hard X-ray and radio halo spectra, are needed in either case. Measurements of the cluster magnetic field will distinguish between the models. ",
        "introduction": "Since the late 1960s, clusters of galaxies have been known to be strong X-ray sources where the thermal  emission from the intracluster medium (ICM) produces a diffuse X-ray halo. The ICM is diffuse (central electron density of $\\sim 10^{-3}$\\,cm$^{-3}$) with typical temperature $2\\times 10^{7}$ to $2 \\times 10^{8}$\\,K. In the 1970s, similarly diffuse radio emission was found in the Coma cluster. This radio emission was found to have a steep power law index ($\\alpha \\sim 1$ for $S_{\\nu}\\propto \\nu^{-\\alpha}$) which indicates a non-thermal origin. Despite efforts to detect similar radio haloes in other clusters, until recently few were found with diffuse radio emission (see review by Feretti \\& Giovannini 1996). This led to the belief that radio haloes are rare. However, the number of clusters found to have diffuse radio emission has more than doubled over the last couple of years, mainly because of the availability of moderate brightness sensitivity surveys at 1.4\\,GHz, e.g. the NVSS \\cite{giov99}, as well as high brightness sensitivity observations of hot clusters to detect the Sunyaev-Zel'dovich effect (e.g. Herbig \\& Birkinshaw 1994; Liang et al. 2000).   Diffuse cluster radio emission is now divided into two distinct classes: haloes and relics. Haloes permeate the entire cluster and resemble the thermal X-ray emission, but relics are on the cluster periphery. Both haloes and relics exhibit steep spectral indices ($\\alpha \\sim 1$) and are extended over $\\sim 1$\\,Mpc with the haloes having lower surface brightness. Emission from radio relics is strongly polarised (up to $\\sim 20$\\%) but no polarised emission has been detected in radio haloes.  Radio haloes and relics are believed to emit by the synchrotron process.  The non-detection of polarised emission from radio haloes is attributed to  low surface brightness and to Faraday depolarisation effects due to the mixing of thermal and relativistic plasma, especially towards the cluster centres.  However, synchrotron emission from haloes requires a population of relativistic (GeV energy range) electrons and a cluster-wide magnetic field. The origins of both particles and field have been a subject of much debate since the late 1970s (e.g.  Jaffe 1977; Dennison 1980; Roland 1981). Recent evidence on this question has come through the observation that radio haloes are preferentially found in high X-ray luminosity clusters (Giovannini et al. 1999), and that there may be a steep correlation between the radio power of haloes and the ICM temperature (e.g. Liang et al. 2000). No upper limits so far contradict this correlation in the temperature range 5--15\\,keV (Liang 2000). The relativistic particles cannot simply leak out of radio sources in the cluster, because the diffusion path length ($\\sim 10$\\,kpc assuming Alfv\\'en speed propagation) is short compared to the size of clusters ($> 1$\\,Mpc).  Giovannini \\& Feretti (2000) found that while diffuse radio sources were found in bright X-ray clusters, none of the 11 clusters selected for the presence of a tailed radio source within the central 300\\,kpc showed any diffuse radio emission. The presence of diffuse radio emission is therefore more closely connected with the properties of the thermal plasma in the ICM than the presence of tailed radio sources which were thought to provide relativistic particles. This gives observational evidence that particles escaping from radio sources in the cluster are unlikely to be accelerated to the relativistic energies required for the radio halo emission. Instead, we interpret the correlation between halo radio power and ICM temperature, and the similarity in appearance of the X-ray and radio haloes, as showing that the relativistic electrons may be accelerated from the thermal pool. The direct cause of this acceleration may be the merging of subclusters and turbulence in the ICM (e.g. Tribble 1993, Brunetti et al. 2001).  The idea that the same relativistic electrons generate radio emission through their synchrotron losses and X-ray or gamma-ray emission by inverse Compton (IC) scattering was first used by Felten and Morrison (1966) for the interpretation of the galactic diffuse emission. Later, Cooke et al. (1978) used it for the interpretation of the X-ray emission from the lobes of the radio galaxy Centaurus A, assuming that the electrons are scattered by the cosmic microwave background (CMB) photons.  A similar model was used by Rephaeli (1979) for the interpretation of the hard X-ray emission from the Coma cluster: the relativistic electrons responsible for the radio halo emission can scatter the CMB photons up to X-ray energies through IC scattering. Efforts have been made to detect this excess of non-thermal emission above the thermal emission from the hot ICM. It is easiest to detect at hard X-ray energies (above $20$\\,keV) where the thermal spectrum is expected to cut off exponentially. Recent observations from {\\it Beppo-SAX} have shown that such a hard X-ray excess does exist in some clusters, e.g. Coma and A2256 (Fusco-Femiano et al. 1999, 2000).  However, alternative theories of the origin of a hard excess have been proposed by various authors.  En{\\ss}lin et al.  (1999) first suggested that the hard X-ray excess observed in the Coma cluster may be the result of bremsstrahlung of supra-thermal electrons in the ICM accelerated by turbulence. Both En{\\ss}lin et al. (1999) and Sarazin \\& Kempner (2000) have assumed the acceleration process works on particles with energies of a few times $kT_{x}$ (i.e. a few tens of keV), and that the electron energy distribution has a sharp transition from a Maxwellian to a power law at a few times $kT_{x}$. Dogiel (2000) pointed out that the acceleration of particles is only efficient above $\\sim 900$\\,keV (i.e. the power law distribution starts at $\\sim 900$\\,keV), but the distortion of the Maxwellian is already significant at much lower energies (as low as $\\sim 30$\\,keV) due to the flux of thermal particles running away into the region of acceleration (for details see Gurevich 1960). It is the `quasi-thermal' particles just above $\\sim 30$\\,keV that produce the observed hard X-ray excess through bremsstrahlung emission.  However, our analysis of the {\\it in situ} acceleration model given below (Sec. 6) shows that the spectrum of MHD turbulence needed to produce the radio emitting electrons is very steep which may make this model problematic.  On the other hand, a model where the X-ray and radio emitting particles are generated by electrons from an `external' injected population cannot be ruled out. A possible model could be that particles in the intracluster medium are accelerated only in regions with high level of turbulence, for example near collisionless bow shocks generated by motions of hypothetical dark matter haloes (e.g., Bykov et al. 2000b).  Radio observations have not shown significant spatial fluctuations of intensity in radio haloes (e.g. Giovannini et al. 1993, Liang et al. 2000) which means that we must have either {\\it in situ} acceleration of these electrons everywhere in the intracluster medium or injection by numerous sources whose separation is smaller than the path length of the electrons ($\\sim 10$\\,kpc) or the length corresponding to the resolution of the radio maps. In the case of the Coma cluster which is relatively bright and close by, the highest resolution VLA map of the radio halo corresponds to a length scale similar to the diffusion path length of the elecrons. Therefore, the injection model requires a high density of localised `external' sources.  Below we investigate both the local and global acceleration models and determine formally the model parameters required to fit the whole observed spectrum from radio to X-rays   for the Coma cluster, since this is the best-studied cluster in all wave-bands.  Our analysis is performed in the framework of a model where spatial variations of the model parameters are neglected. There are several reasons for this: a) observations show more or less uniform distribution of non-thermal emission in the halo; and b) we do not have enough information about spatial variations of the hard X-ray flux from the Coma halo, therefore we derive averaged parameters of particle acceleration based on the total flux of hard X-rays.   The cosmological parameters adopted are $H_{0}=50$\\,km\\,s$^{-1}$\\,Mpc$^{-1}$, $q_{0}=0.5$, so that for the Coma cluster 1 arcmin is 40\\,kpc. ",
        "conclusions": "We have shown that it is possible to explain both the excess hard X-ray data and the radio halo spectrum with a model where the relativistic electrons are weakly accelerated from the thermal pool by turbulence in the ICM. By using a quasi-stationary solution of the kinetic equation, we arrive at an electron energy distribution that deviates from a Maxwellian at $\\sim 30$\\,keV and reaches a power law at $\\sim 1-10$\\,MeV. In the process of reaching an equilibrium, the collisions between thermal particles produce the supra-thermal electrons at $\\sim 30-100$\\,keV required to give the excess hard X-ray emission through bremsstrahlung emission. The accelerated GeV electrons are then sufficient to produce the observed radio halo spectrum with a magnetic field close to the equipartition value. The model produces a self-consistent electron energy distribution and has only two free parameters, the acceleration parameter $a_{0}$ and the momentum spectral index $q$, which are well constrained by the observed excess hard X-ray spectrum and the radio halo spectral index (Fig.~\\ref{aq}). In the hard X-ray energy range, the contribution from the inverse Compton scattering of the GeV electrons is negligible compared with the non-thermal bremsstrahlung emission for steep radio spectral indices $\\alpha > 1$.  It is also possible to explain the hard X-ray and radio spectrum by a model where accelerated particles are externally injected in numerous localised regions distributed throughout the cluster. In this case, the contribution from IC scattering dominates that of non-thermal bremsstrahlung in the hard X-ray energy range for steep radio spectral indices.  For both the above models a strong assumption was made, that the non-thermal X-ray and radio emission from Coma is generated from a single electron acceleration mechanism. An advantage of this assumption is that it limits significantly the number of free parameters in the model. Of course it is possible to have both {\\it in situ} acceleration and external injection of accelerated particles (or global and local acceleration) at work in the Coma cluster. We have, however, examined the two extremes rather than such a hybrid model. This study has shown that if it is purely {\\it in situ} acceleration that produces the particles necessary for the non-thermal emission in the hard X-ray and radio spectral bands, then the only explanation for the origin of the hard X-ray is bremsstrahlung of supra-thermal electrons; and consequently, to fit both X-ray and radio emission a cluster-wide magnetic field of order $\\mu$G (i.e. close to the equipartition value) is implied. On the other hand, if the accelerated particles are purely provided by numerous, localised external sources of injection, then the only possible origin for the hard X-rays is inverse Compton scattering of CMB photons by relativistic electrons, which implies a cluster-wide magnetic field of 0.1--0.2\\,$\\mu$G. Any independent means of measuring the global cluster magnetic fields would distinguish between the two models.  At present it is not easy to choose between the models of particle acceleration for Coma. Both models: {\\it in situ} acceleration (global) and shock wave acceleration (local) have their advantages and problems. In favour of the {\\it in situ} model is the similar morphology between the thermal X-ray and the radio emission. The radio data have not shown  significant structure in the radio intensity in the Coma halo which would indicate regions of electron acceleration. Analysis of the thermal X-ray emission from the Coma cluster also did not show the strong temperature variations expected from shocks (Arnaud et al. 2001). On the other hand, several features indicating recent and ongoing mergers are found in the Coma cluster (e.g. Donnelly et al. 1999, Arnaud et al. 2001). These mergers often produce multiple large scale shocks and turbulence capable of both magnetic field amplification and {\\it in situ} electron acceleration (see a recent paper by Ohno et al. 2002 on the subject). Whether or not these regions extend far enough to allow electrons to fill the whole volume of the halo, remains to be discovered.  If the origins of the radio and X-ray emitting electrons are different, then a much wider range of possible models can be accommodated. For example, there has been revived interest in the class of secondary electron models where relativistic and thermal protons in the ICM collide to produce relativistic (secondary) electrons (Dennison 1980; Blasi \\& Colafrancesco 1999; Dolag \\& En{\\ss}lin 2000; and Miniati et al. 2001).     We would like to thank R. Fusco-Femiano for providing us with the data points for the PDS spectrum for Coma. We are grateful to the referee T. En{\\ss}lin for helpful comments and suggestions.  VAD thanks the University of Bristol and the Institute of Space and Astronautical Science (Sagamihara, Japan) in particular Prof. H. Inoue for hospitality during his visits.  \\newpage"
    },
    "0208/astro-ph0208100_arXiv.txt": {
        "abstract": "\\vspace{0.2cm} A primary goal for cosmology and particle physics over the coming decade will be to unravel the nature of the dark energy that drives the accelerated expansion of the Universe. In particular, determination of the equation-of-state of dark energy, $w\\equiv p/\\rho$, and its time variation, $dw/dz$, will be critical for developing theoretical understanding of the new physics behind this phenomenon. Type Ia supernovae (SNe) and cosmic microwave background (CMB) anisotropy are each sensitive to the dark energy equation-of-state.  SNe alone can determine $w(z)$ with some precision, while CMB anisotropy alone cannot because of a strong degeneracy between the matter density $\\Omega_M$ and $w$. However, we show that the Planck CMB mission can significantly improve the power of a deep SNe survey to probe $w$ and especially $dw/dz$. Because CMB constraints are nearly orthogonal to SNe constraints in the $\\Omega_M$--$w$ plane, for constraining $w(z)$ Planck is more useful than precise determination of $\\Omega_M$. We discuss how the CMB/SNe complementarity impacts strategies for the redshift distribution of a supernova survey to determine $w(z)$ and conclude that a well-designed sample should include a substantial number of supernovae out to redshifts $z \\sim 2$. ",
        "introduction": "\\label{sec:intro}  Recent observations of Type Ia supernovae (SNe) have provided direct evidence that the Universe is accelerating \\cite{perl99,riess98}, indicating the existence of a nearly uniform dark-energy component with negative effective pressure, $w\\equiv p/\\rho < -1/3$.  Further evidence for dark energy comes from recent cosmic microwave background (CMB) anisotropy measurements pointing to a spatially flat, critical density Universe, with $\\Omega_{0} = 1$ \\cite{CMB}, combined with a number of indications that the matter density $\\Omega_M \\simeq 0.3$ \\cite{omega_m}; the `missing energy' must also have sufficiently negative pressure in order to allow time for large-scale structure to form \\cite{turner_white}. Together, these two lines of evidence indicate that dark energy composes 70\\% of the energy density of the Universe and has equation-of-state parameter $w < - (0.5 - 0.6)$ \\cite{w_constraints}. Determining the nature of dark energy, in particular its equation-of-state, is a critical challenge for physics and cosmology.  At present, particle physics theory provides little to no guidance about the nature of dark energy. A cosmological constant---the energy associated with the vacuum---is the simplest but not the only possibility; in this case, $w=-1$ and is time independent, and the dark energy density is spatially constant. Unfortunately, theory has yet to provide a consistent description of the vacuum: the energy density of the vacuum, at most $10^{-10}\\,{\\rm eV}^4$, is at least 57 orders of magnitude smaller than what one expects from particle physics---the cosmological constant problem \\cite{weinberg}. In recent years, a number of other dark energy models have been explored, from slowly rolling, ultra-light scalar fields to frustrated topological defects \\cite{demodels}. These models predict that $w\\not= -1$, that $w$ may evolve in time, and that there may be small spatial variations in the dark energy density (of less than a part in $10^5$ on scales $\\sim H_0^{-1}$~\\cite{perturbations}). In all models proposed thus far dark energy can be characterized by its equation-of-state $w$. Measuring the present value of $w$ and its time variation will provide crucial clues to the underlying physics of dark energy.  As far as we know, dark energy can only be probed directly by cosmological measurements, although it is possible that laboratory experiments could detect other physical effects associated with dark energy, e.g., a new long-range force arising from an ultra-light scalar field \\cite{longrange}. Dark energy affects the expansion rate of the Universe and thereby influences cosmological observables such as the distance vs.\\ redshift, the linear growth of density perturbations, and the cosmological volume element (see, e.g., \\cite{HT}). Standard candles such as Type Ia supernovae offer a direct means of mapping out distance vs.\\ redshift (see, e.g., \\cite{WA2002}), while the CMB anisotropy can be used to accurately determine the distance to one redshift, the last scattering epoch ($z_{LS}\\simeq 1100$). Because they measure distances at such different redshifts, the SNe and CMB measurements have complementary degeneracies in the $\\Omega_M$--$\\Omega_{\\Lambda}$ and $\\Omega_M$--$w$ planes \\cite{HT,WA2002,CMB+SN}. More recently, Spergel \\& Starkman \\cite{SS} have suggested that this complementarity argues for using supernovae at relatively low redshift, $z\\sim 0.4$, to most efficiently probe dark energy.  In so doing, they used a highly simplified model which did not consider a spread of SNe in redshift, systematic error, possible evolution of $w$, or the finite precision with which planned CMB missions can actually constrain $\\Omega_M$ and $w$.  By including these ``real-world'' effects, this paper clarifies the complementarity of the CMB and SNe and explores strategies for best utilizing it in SNe surveys to probe the properties of dark energy.  We show that dark energy-motivated supernova surveys should target SNe over a broad range of redshifts out to $z \\sim 2$, and that CMB/SNe complementarity in fact strengthens the case for deep SNe surveys. ",
        "conclusions": "\\label{sec:concl}  Unraveling the nature of dark energy is one of the outstanding challenges in physics and astronomy.  Determining its properties is critical to understanding the Universe and its destiny and may shed light on the fundamental nature of the quantum vacuum and perhaps even of space-time. Type Ia supernovae and CMB anisotropy can both probe the dark energy equation-of-state $w$, and we have explored in detail the synergy between the two.  With the MAP mission in progress, the Planck mission slated for launch in 2007, and the design of dedicated SN surveys now underway, such a study is very timely.  CMB anisotropy alone cannot tightly constrain the properties of dark energy because of a strong degeneracy between the average equation-of-state and the matter density.  SNe can probe $w$ with a precision that improves significantly with knowledge of the matter density, because $H_0r(z)$ depends only upon $w$ and $\\Omega_M$.  A key result of this paper is that CMB anisotropy measurements by the upcoming Planck mission have even more potential for improving the ability of SNe to probe dark energy. The reason is simple: in the $\\Omega_M$--$w$ plane (Fig.~\\ref{fig:Fig1}), the CMB constraint is more complementary to the SNe constraint than is determination of $\\Omega_M$.  Compared to the matter density prior $\\sigma_{\\Omega_M} = 0.03$, Planck CMB data reduce the predicted error $\\sigma_w$ (under the assumption of constant $w$) by about a factor of two (Fig.~\\ref{fig:Fig7}). In probing possible variation of $w$ with redshift, the Planck prior is also significantly better than the same matter density prior (Fig.~\\ref{fig:Fig9}).  Given the concern expressed by some (e.g., \\cite{Maoretal}) that a precise measurement of the matter density independent of dark energy properties may be difficult, this is good news.  On the other hand, we find that even if MAP can successfully measure polarization on large scales, its potential for complementarity with SNe falls short of that for Planck and is not as good as the $\\sigma_{\\Omega_M}=0.03$ matter density prior.  We have also explored how the SNe determination of the dark energy equation-of-state, with or without prior information from the CMB or the matter density, depends upon the redshift distribution of the survey, including the effects of systematic error and a realistic spread of SNe redshifts. For either constant or evolving $w$, the optimal strategy calls for significant numbers of SNe above redshift $z\\sim 1$.  For the constant $w$ case with no Planck prior, or for evolving $w$ regardless of prior, these high-redshift SNe are necessary for achieving $\\sigma_w < 0.1$. Observing substantial numbers of SNe at these high redshifts also provides the only hope of probing time evolution of the equation-of-state with reasonable precision.  Moreover, the improvement in $\\sigma_{dw/dz}$ continues to high redshift: $\\sigma_{dw/dz}$ falls by more than a factor of two when $z_{\\rm max}$ increases from $1$ to $2$ (Fig.~\\ref{fig:Fig9}). Since we currently have no prior information about (or consensus physical models which significantly constrain) the time variation of $w$, the design of a SNe survey aimed at probing dark energy should take into account the possibility that $w$ evolves. These conclusions about the need for high-redshift supernovae do not change significantly if we consider a hypothetical survey for which resources are constrained and a redshift-dependent cost is assigned to each supernova.  Ref.~\\cite{SS} raised the question whether a shallow SNe survey is better than a deep one in determining the dark energy equation-of-state, given prior knowledge from the CMB. Our results indicate that it is not, once the SNe and CMB experiments are realistically modelled. On the contrary, CMB/SNe complementarity strengthens the case for a deep SNe survey that extends to redshift $z \\sim 2$."
    },
    "0208/astro-ph0208270_arXiv.txt": {
        "abstract": "{Results of a photometric study of the SW Sex novalike \\object{PX And} are presented. The periodogram analysis of the observations obtained in October 2000 reveals the presence of three signals with periods of 0\\fd142, 4\\fd8 and 0\\fd207. The first two periods are recognized as \"negative superhumps\" and the corresponding retrograde precession period of the accretion disk. The origin of the third periodic signal remains unknown. The observations in September-October 2001 point only to the presence of ``negative superhumps\" and possibly to the precession period. The origin of the ``negative superhumps\" is discussed and two possible mechanisms are suggested. All light curves show strong flickering activity and power spectra with a typical red noise shape. PX And shows eclipses with highly variable shape and depth. The analysis suggests that the eclipse depth is modulated with the precession period and two possible explanations of this phenomenon are discussed. An improved orbital ephemeris is also determined: $T_{\\rm min}[HJD]=49238.83662(14)+0\\fd146352739(11)E$. ",
        "introduction": "\\object{PX And} is probably one of the most complicated SW Sex stars (Thorstensen et al. \\cite{th91}; Hellier \\& Robinson \\cite{hel94}; Still et al. \\cite{still}). Thorstensen et al. (\\cite{th91}) reported shallow eclipses with highly variable eclipse depth and repeating with a period of $\\sim$0\\fd1463533. The authors  assumed a steady-state accretion disk effective temperature distribution $T_{\\rm eff}\\sim r^{-3/4}$ and $q\\simeq0.46$, and adjusted the inclination and disk radius to match the width and depth of the mean eclipse. They obtained $i\\simeq73.8^\\circ$ and $r_{\\rm d}\\simeq0.6R_{L_1}$. No other system parameters estimations have been published and all authors used the above values. Apart from the distinctive characteristics of SW Sex stars (reviewed recently by Hellier \\cite{helrev}), \\object{PX And} shows some other interesting peculiarities. Patterson (\\cite{patt99}) reported \\object{PX And} to show simultaneously ``negative\" and ``positive\" superhumps, and signals with typical periods of 4--5 days. This implies that \\object{PX And} most probably possesses an eccentric and tilted accretion disk. In the view of the expected stream overflow (Hellier \\& Robinson \\cite{hel94}) all this would result in very complex accretion structures.  In this paper we report the results of a photometric study of \\object{PX And}. ",
        "conclusions": "The origin of the ``negative superhumps\" is still an open question. The most plausible model is based on the assumption of a retrograde precession of a tilted accretion disk. While the \"positive superhumps\" have been simulated numerically since the work of Whitehurst (\\cite{wh}), there are some difficulties to simulate the \"negative superhumps\". In fact, all attempts to simulate ``negative superhumps\" failed in the sense that they were not able to produce a significant tilt starting from a disk lying in the orbital plane (Murray \\& Armitage \\cite{mur}; Wood et al. \\cite{wood}). Once tilted, however, the accretion disk starts precessing in retrograde direction (Larwood et al. \\cite{lar}; Wood et al. \\cite{wood}) thus giving two additional clocks: $P_{\\rm prec}$ in inertial co-ordinate system and $P^-_{\\rm SH}$ in a system co-rotating with the binary. $P_{\\rm prec}$ is typically a few days and thus $P^-_{\\rm SH}$ is slightly shorter than the orbital period. The angle at which the accretion disk is seen from the Earth is modulated with the precession period, giving brightness modulations with $P_{\\rm prec}$.  Taking into account foreshortening and limb-darkening, one can estimate the disk tilt needed to produce the observed amplitude of the 4\\fd8 modulation in \\object{PX And}. With a limb-darkening coefficient $u=0.6$ the tilt angle is between 2.5$^\\circ$ and 3$^\\circ$, depending on the assumed system inclination. For comparison the simulations of Wood et al. (\\cite{wood}) were performed with a tilt angle of 5$^\\circ$.  \\begin{figure}[t] \\centering \\includegraphics*[width=8.8cm]{ms2639f7.eps} \\caption{A sketch of the system configuration at $\\phi_{\\rm prec}$=0.13 {\\it and} $\\phi_{\\rm orb}$=0. The disk edge marked with the dashed line lays below the orbital plane. Also shown are the expected position of the SLS, shadow of the secondary, accretion stream and re-impact zone.} \\label{pxmod} \\end{figure}  The mechanism generating the ``negative superhump\" light itself is however rather uncertain. According to the simulations of Wood et al. (\\cite{wood}) the two opposite parts of the disk which are most displaced from the orbital plane are tidally heated by the secondary. The heating is maximal when these parts point to the secondary and this happens twice per superhump cycle. Wood et al. (\\cite{wood}) find that the tidal stress is asymmetric with respect to the disk mid-plane and the disk surface facing the orbital plane is more heated. Thus, if the disk is optically thick the observer sees only one side and a modulation with the superhump period is observed. Let us define precession phase $\\phi_{\\rm prec}$ to be zero at the maximum of the 4\\fd8 cycle, i.e. when the disk is most nearly face-on. The relative phasing between the periodic signals determined from the fit shows that the maxima of the superhumps coincide with the eclipses at $\\phi_{\\rm prec}\\simeq$0.13. Figure\\,\\ref{pxmod} shows a sketch of the system configuration in eclipse at $\\phi_{\\rm prec}=0.13$. The expected position of the superhump light source (SLS) according to Wood et al. (\\cite{wood}) and the line of nodes are also shown. The dashed line marks the part of the disk which lays below the orbital plane. According to the simulations of Wood et al. (\\cite{wood}) the \"negative superhumps\" maxima  occur when the region labelled as \"SLS\" in Fig.\\,\\ref{pxmod} lies on the line connecting the two system components, which means that the superhumps maxima will coincide with the eclipses at $\\phi_{\\rm prec}=0$. Our results show that in \\object{PX And} this is not exactly the case and the superhumps maxima are observed $\\sim0.13P_{\\rm SH}^-$ later.  There is another mechanism which could generate ``negative superhumps\" and at the same time account for the delay. Patterson (\\cite{patt99}) noticed that most of the SW Sex novalikes show ``negative superhumps\" and this could naturally explain the accretion stream overflow thought to be responsible for the SW Sex phenomenon (Hellier \\& Robinson \\cite{hel94}). In a system with a precessing tilted accretion disk, the amount of gas in the overflowing stream will vary with the ``negative superhumps\" period. Correspondingly, the intensity of the spot (shown in Fig.\\,\\ref{pxmod}) formed where the overflowing stream re-impacts the disk should be modulated and might be the ``negative\" SLS. A similar model was proposed by Hessman et al. (\\cite{hess}) for the origin of \"positive superhumps\". The maximal overflowing is expected to take place when the most displaced part of the disk points to the accretion stream. The intensity of the re-impact spot will however reach its maximum later when the gas in the stream reaches the re-impact zone (the moment shown in Fig.\\,\\ref{pxmod}).  This means that the gas in the accretion stream which at the moment shown in Fig.\\,\\ref{pxmod} is near the re-impact zone, has passed over the disk edge earlier, when the system configuration was different and the most displaced part of the disk pointed to the accretion stream.  Let the gas in the overflowing stream travels from the outer disk edge to the re-impact zone in the time $\\Delta t$. Then, because in co-ordinate system rotating with the binary the line of nodes precesses with period $P^-_{\\rm SH}$, at the superhumps maxima the displaced part of the disk will not point to the accretion stream but will be rotated with respect to it by an angle $\\theta\\simeq360\\Delta t/P^-_{\\rm SH}$ deg. Figure\\,\\ref{pxmod} shows that in the case of \\object{PX And} $\\theta\\simeq0.15\\times360=54^\\circ$. Therefore, in order to observe the maxima of the ``negative superhumps\" at $\\phi_{\\rm prec}\\simeq$0.13 {\\it and} $\\phi_{\\rm orb}\\simeq$0 the travel time from the outer disk edge to the re-impact zone has to be $\\sim30$ min. This is slightly longer than expected (Warner \\& Peters \\cite{war}), but we have to take into account that the velocity of the incoming gas is most probably reduced by the first impact with the accretion disk. This model for the origin of the ``negative\" SLS can be independently checked by tracking the intensity of the high-velocity $s$-wave observed in the spectra of SW Sex novalikes.   The origin of the 0\\fd207 signal is quite uncertain. The corresponding peak in the power spectrum could not be a result of an amplitude modulation of the ``negative superhumps\". If this is the case, the power spectrum should show two additional peaks. The secondary crosses the line of nodes twice per superhump cycle. The time between three consecutive passages coincides within 1.5\\% with the observed period. It is however not clear how this would produce periodic brightness modulations, moreover with the observed amplitude of $\\sim$0.15 mag.   The most puzzling observation of \\object{PX And} is its highly variable eclipse depth. Our observations suggest that the eclipse depth is modulated with the precession period (Figs.\\,\\ref{eclch} and \\ref{pxdepth1}), indicating that this phenomenon could be related to the disk precession. In the precessing tilted accretion disk model, in some particular combinations of $q$ and $i$ the eclipsed area of the disk is expected to vary with the precession period and thus to modulate the eclipse depth. We have simulated eclipses of a tilted precessing disk which showed that the eclipse depth indeed varied but with a very low amplitude of 2-3\\%, which is insufficient to explain the observations.  \\begin{figure}[t] \\includegraphics*[width=8.8cm]{ms2639f8.eps} \\caption{Normalized $V$ band eclipses of PX And. ``*\" mark the eclipses obtained in 2001.} \\label{ecl} \\end{figure}  The relatively shallow eclipses  show that the accretion disk in \\object{PX And} is not totally eclipsed. In this case, if there is a significant constant light source which is totally eclipsed, the eclipse depth will be modulated since different fraction of the total light emitted by the system will be eclipsed at different precession phases. The deepest eclipses are expected at the precession cycle minima, as observed. As the constant light source must be eclipsed, its most likely location is close to the white dwarf. The white dwarf itself could not be the source of the constant light as it does not contribute significantly to the total system light. The most obvious candidate is the inner hot part of the disk. Since the brightness modulation with the precession cycle comes from more or less pure geometrical considerations it is not clear why the emission of the inner parts of the disc could be constant. One possible explanation might be that the inner part of the disk is not tilted or is less tilted than the outer part. In this  case the emission from the inner disk will be constant. Another source of the constant light could be the emission from the boundary layer between the disk and the white dwarf surface. This however is not very likely since the temperature of the boundary layer is thought to be rather high to emit significantly in the optical wavelengths. Of course, there exists the possibility that the eclipse depth modulation with the precession cycle is an artifact. If some flickering peaks occur during the eclipse (which is likely because the accretion disk in \\object{PX And} is not totally eclipsed), this will decrease the observed eclipse depth. Figure\\,\\ref{ecl} shows that some eclipses are indeed badly affected by the flickering. Because of the small number of eclipses covered this might introduce a spurious modulation. But since the correlation between eclipse depth and mean magnitude is good, we suggest that the eclipse depth modulation with the precession cycle is real.  \\begin{figure}[t] \\includegraphics*[width=8cm]{ms2639f9.eps} \\caption{Eclipse depth as a function of $F_{\\rm mod}/F_{\\rm const}$ (see text for details). Dots and open circles show the 2000 and 2001 data, respectively.} \\label{pxdepth} \\end{figure}  There is another way to modulate the eclipse depth. As the SLS is non-uniformly distributed over the disk surface and the accretion disk is not totally eclipsed, there are two extreme possibilities when the SLS is (i) totally and (ii) never eclipsed. In both cases the eclipse depth will depend on the actual superhump light at the moment of eclipse, but will have opposite behavior. If the SLS is totally eclipsed, then the deepest eclipses will be observed when superhump maxima coincide with the eclipses. If the SLS is never eclipsed, the eclipse depth will be affected in a way analogous to the {\\it veiling} of the absorption lines in \\object{T Tau} stars. In these stars the additional emission in the continuum reduces the observed depth of the absorption lines. Then the eclipses will be deepest when superhump minima coincide with the eclipses. Using the parameters of the periodic signals obtained from the multi-sinusoidal fit, we calculated the total contribution of the modulated part of the flux to the unmodulated one, $F_{\\rm mod}/F_{\\rm const}$, at the moments of the eclipses. $F_{\\rm const}$ is modulated with the precession cycle and hence may be thought to be constant on the superhump time scale. The $\\sim$0\\fd207 signal was included in the calculations for 2000. The reason is that the eclipse depth variation observed in the 2000 data is rather high to be explained if only the ``negative\" SLS is involved. Figure\\,\\ref{pxdepth} shows the eclipse depth $d_{\\rm e}$ as a function of $F_{\\rm mod}/F_{\\rm const}$. Although the scatter is large, one should keep in mind that any flickering near the mid-eclipses will decrease the eclipse depth and will move the points in Fig.\\,\\ref{pxdepth} downward. If a small number of eclipses are observed then this could easily obscure any correlation. Despite the large scatter, there are some hints of an inverse correlation in Fig.\\,\\ref{pxdepth}. This suggests that the SLS is not eclipsed. In Fig.\\,\\ref{pxdepth} is also shown an arbitrary curve defined by the equation: \\begin{equation} d_{\\rm e}=\\frac{d_{\\rm e,0}}{1+F_{\\rm mod}/F_{\\rm const}} \\label{depth} \\end{equation} which gives the eclipse depth as a function of the additional modulated light in the case  when this light is not eclipsed. Here, $d_{\\rm e,0}$ is the eclipse depth if $F_{\\rm mod}=0$.   Although mathematically the veiling could explain the eclipse depth changes, there are many uncertainties with this interpretation. Most of them are related to the unknown place of origin of the periodic signals. If the conclusion of Wood et al. (\\cite{wood}) about where the ``negative superhumps\" are generated is correct, then the ``negative\" SLS should be at least partially eclipsed. This suggests that the stream overflow could be the mechanism generating the ``negative superhumps\" in \\object{PX And}. Moreover, in a system with grazing eclipses the re-impact zone is not eclipsed. Apart of this, the origin of the $\\sim$0\\fd207 signal is unknown and it is not clear if its source is eclipsed or not. Thus, the question why the eclipse depth in \\object{PX And} varies remains open and we have only pointed to some of the possible explanations.  Thorstensen et al. (\\cite{th91}) have estimated $q$, $i$ and the disk radius of \\object{PX And} by fitting a mean eclipse. It should be noted however that the observations of Thorstensen et al. (\\cite{th91}) point to an average eclipse depth of $\\sim0.5$ mag while our observations show much deeper eclipses. It is clear that the highly variable shape and depth of eclipses could affect significantly any attempt to estimate the system parameters of \\object{PX And}. Because of this  we have not attempted such an estimation from the eclipse profiles. It would also not be correct to define so called \"mean\" eclipse and to analyze it with the eclipse mapping technique. Moreover, a large part of the accretion disk in \\object{PX And} is not eclipsed. The eclipse mapping algorithm could be easily modified to allow analysis of tilted accretion disks. To use this technique however one will need to average at least several eclipses at given precession phase in order to reduce the influence of flickering and other noise.  Generally, if a given cataclysmic variable shows ``positive superhumps\" one could estimate $q$ using the existing $\\epsilon^+(q)$ relations ($\\epsilon^+=(P_{\\rm sh}^+-P_{\\rm orb})/P_{\\rm orb}$). Patterson (\\cite{patt99}) reported that \\object{PX And} shows ``positive superhumps\" with $\\epsilon^+\\simeq$0.09 and this could be used to estimate $q$. Recently, Montgomery (\\cite{mon}) published an analytic $\\epsilon^+(q)$ expression which takes into account the pressure effect. Applying this relation  (Eq.\\,(8) in Montgomery \\cite{mon}) to \\object{PX And} we obtain $q\\simeq$0.27. Although this value of $q$ is more plausible for a system showing \"positive superhumps\", one should keep in mind that the peak in the power spectrum of the CBA data is very weak. The \"negative superhumps\" are much more confidently detected, but unfortunately an $\\epsilon^-(q)$ relation for the \"negative superhumps\" has not been established yet. Patterson (\\cite{patt99}) noticed that $\\epsilon^-\\simeq-0.5\\,\\epsilon^+$. Our periodogram analysis yields $\\epsilon^-\\simeq-0.029$, hence $\\epsilon^+\\simeq0.058$. From Montgomery (\\cite{mon}) relation, one obtains $q\\simeq0.18$. This is not an unrealistic value but we cannot be sure that the relation $\\epsilon^-\\simeq-0.5\\,\\epsilon^+$ holds true for \\object{PX And}.   Based on the analysis of our new photometric observations and on the results of Patterson (\\cite{patt99}) we can with no doubt place the novalike \\object{PX And} among the cataclysmic variables showing permanent \"negative superhumps\". The observations of the star  in 2000 show  two unusual features: the presence of another superhump with a period of $\\sim$0\\fd207 and a modulation of the eclipse depth with the precession period of the accretion disk. Our 2001 observations are however limited in time coverage and it is difficult to say if these two phenomena are typical of the star or are a peculiarity of this particular data set only. As a suggestion  for future work we recommend performing a multi-site photometric observations covering several precession cycles. If the modulation of the eclipse depth with the precession period is confirmed, this phenomenon could be further studied by analysis of the eclipse profiles at different precession phases."
    },
    "0208/astro-ph0208046_arXiv.txt": {
        "abstract": "MHD turbulence consists of waves that propagate along magnetic fieldlines, in both directions.  When two oppositely directed waves collide, they distort each other, without changing their respective energies. In weak MHD turbulence, a given wave suffers many collisions before cascading. ``Imbalance'' means that more energy is going in one direction than the other. In general, MHD turbulence is imbalanced. A number of complications arise for the imbalanced cascade that are unimportant for the balanced one.  We solve weak MHD turbulence that is imbalanced. Of crucial importance is that the energies going in both directions are forced to equalize at the dissipation scale.  We call this the ``pinning'' of the energy spectra.  It affects the entire inertial range.  Weak MHD turbulence is particularly interesting because perturbation theory is applicable. Hence it can be described with a simple kinetic equation. Galtier et al. (2000) derived this kinetic equation. We present a simpler, more physical derivation, based on the picture of colliding wavepackets. In the process, we clarify the role of the zero-frequency mode. We also explain why Goldreich \\& Sridhar claimed that perturbation theory is inapplicable, and why this claim is wrong. (Our ``weak'' is equivalent to Goldreich \\& Sridhar's ``intermediate.'')  We perform numerical simulations of the kinetic equation to verify our claims. We construct simplified model equations that illustrate the main effects. Finally, we show that a large magnetic Prandtl number does not have a significant effect, and that hyperviscosity leads to a pronounced bottleneck effect. ",
        "introduction": "\\label{sec:heuristic}  One of the virtues of weak MHD turbulence is that it can be analyzed in a mathematically rigorous way with perturbation theory; this yields a kinetic equation. Nevertheless, we begin with a qualitative description, which captures most of the features of the turbulent cascade.  \\subsection{Scaling Relation} \\label{sec:scaling}  MHD turbulence can be understood from the dynamics of $\\bwupp$ and $\\bwdownp$ (eqs. [\\ref{eq:eomrmhd1}-\\ref{eq:eomrmhd4}] for reduced MHD, dropping $\\perp$ subscripts). To linear order,  $\\bwupp$ is a wave that propagates up the mean magnetic fieldlines at the Alfv\\'en speed, $v_A$; $\\bwdownp$ propagates down at $v_A$. Each wave perturbs the mean magnetic fieldlines. Nonlinear terms describe the interaction between oppositely directed waves: each wave nearly follows the fieldlines perturbed by its collision partner. \\footnote{The equation for a scalar quantity $f$ that travels upwards at speed $v_A$, while following the magnetic fieldlines of the down-going $\\bwdownp$ is $(\\pt+v_A\\pz+\\bwdownp\\bcdot\\bnablap)f=0$. Equation (\\ref{eq:eomrmhd1}) for the vector $\\bwupp$ differs from this because of the pressure term, which is required to keep $\\bwupp$ incompressible, while conserving the energy $(\\wupp\\ )^2$.  Thus $\\bwupp$ does not exactly follow the fieldlines of $\\bwdownp$.  Nonetheless, this deviation does not greatly affect the behaviour of the turbulence.  Dissipation is a second effect that prevents the following of fieldlines. In the present discussion, we consider lengthscales that are sufficiently large that dissipation can be neglected.}   Consider an upgoing wavepacket that encounters a train of downgoing wavepackets. As the upgoing wave travels up the length of the downgoing train, it is gradually distorted.  It tries to follow the perturbed fieldlines in the downgoing train, but these fieldlines ``wander,'' i.e., the transverse separation between any two fieldlines changes. When the up-wave has travelled through a sufficiently large number of downgoing wavepackets that the amount of fieldline wander is comparable to the up-wave's transverse size, then the up-wave cascades.  To be quantitative, let each downgoing wave in the train have a typical amplitude $\\wdownpl$, a transverse size $\\lp$, and a parallel size $\\lpar$, where ``transverse'' and ``parallel'' refer to the orientation relative to the mean magnetic field. The most important collisions are between wavepackets of comparable transverse size (see \\S \\ref{sec:locality}). So let the upgoing wave have transverse size $\\lp$ as well.  Since each downgoing wavepacket has a typical perturbed magnetic field of magnitude $\\sim\\wdownpl$ (neglecting the factor of $1/2$), it bends the fieldlines by the angle $\\wdownpl/v_A$; the transverse displacement of a fieldline through this wavepacket is $(\\wdownpl/v_A)\\lpar$; and the wander of two typical fieldlines through the wavepacket, if they are initially separated by $\\lp$, is also $(\\wdownpl/v_A)\\lpar$.  In weak turbulence, the wander through a single wavepacket is smaller than the wavepacket's transverse size, \\begin{equation} {\\wdownpl\\over v_A}\\lpar\\ll \\lp \\ \\ {\\rm and} \\ \\ \\ {\\wuppl\\over v_A}\\lpar\\ll \\lp \\ . \\label{eq:inter} \\end{equation} When these inequalities are not satisfied, strong turbulence is applicable;  see \\S \\ref{sec:strong}. Thus in weak turbulence an upgoing wavepacket must travel through many downgoing ones before cascading. After $N$ downgoing wavepackets, fieldlines have wandered a distance $\\sim N^{1/2}(\\wdownpl/v_A)\\lpar$, assuming that wavepackets are statistically independent. The upgoing wavepacket is fully distorted---and hence cascaded---when the fieldlines it is following wander a distance $\\lp$, i.e., when $N\\sim(\\lp v_A/\\lpar\\wdownpl)^2$. Since each downgoing wavepacket is crossed in the time $\\lpar/v_A$, the cascade time of the upgoing wavepacket is \\begin{equation} \\tcasup\\sim\\Big({\\lp v_A\\over \\lpar\\wdownpl}\\Big)^2{\\lpar\\over v_A}\\sim \\Big({\\lp \\over\\wdownpl}\\Big)^2 {v_A\\over\\lpar} \\ . \\label{eq:cascadetime} \\end{equation} In this time, the upgoing wavepacket travels a distance $v_A\\tcasup$, which is much larger than its own length, $\\lpar$.\\footnote {We assume throughout this paper that the upgoing waves' parallel lengthscale is the same as that of the downgoing waves, $\\lpar$; the extension to the case when they differ is trivial, as long as the inequalities (\\ref{eq:inter}) are both satisfied, with the appropriate $\\lpar$'s.} The head and the tail of the upgoing wavepacket are both distorted by the same downgoing wavepackets; so both head and tail undergo nearly the same distortion as they cascade.\\footnote{The head of the upgoing wavepacket slightly distorts each downgoing wavepacket; so the downgoing wavepacket seen by the tail is slightly distorted relative to that seen by the head. Nonetheless, this backreaction is a higher-order correction that can be ignored in weak turbulence; see \\S \\ref{sec:prelim}. } Consequently, as the upgoing wavepacket cascades to smaller transverse lengthscales, it does not cascade to smaller parallel ones: \\begin{equation} \\lpar={\\rm scale \\ independent} \\ . \\label{eq:intscal1} \\end{equation} A proof of this follows from the 3-wave resonance relations (Shebalin, Matthaeus, \\& Montgomery 1983; see also \\S \\ref{sec:compare} of the present paper).  We calculate the steady state energy spectra by using Kolmogorov's picture of energy flowing from large to small lengthscales. The energy in up-waves flows from lengthscales larger than $\\lp$ to those smaller than $\\lp$ at the rate \\begin{equation} \\epsilon^\\uparrow\\sim\\frac{(\\wuppl)^2}{ \\tcasup} \\sim\\Big[\\frac{\\wuppl\\wdownpl}{\\lp}\\Big]^2\\frac{\\lpar} {v_A} \\label{eq:epsilonheuristic} \\ . \\end{equation} We call this simply the ``flux.'' We define $\\epsilon^\\uparrow$ more precisely below (eq. [\\ref{eq:flux}]). In steady state, the flux must be independent of $\\lp$, so \\begin{equation} \\wuppl\\wdownpl\\propto \\lp \\ . \\label{eq:intscal2} \\end{equation}  \\subsection{Insufficiency of Scaling Arguments for the Imbalanced Cascade} \\label{sec:insufficiency}  A balanced cascade has $\\wuppl=\\wdownpl\\equiv w_\\lambda$. Its solution in steady state is simple: $w_\\lambda\\propto \\lp^{1/2}$ and $\\epsilon^\\uparrow=\\epsilon^\\downarrow\\sim w_\\lambda^4\\lpar/\\lp^2 v_A$ (Goldreich \\& Sridhar 1997, Ng \\& Bhattacharjee 1997). However, if the cascade is imbalanced, a number of complications arise.  By the symmetry between up- and down-going waves, the down-going flux is given by the analogue of equation (\\ref{eq:epsilonheuristic}): \\begin{equation} \\epsilon^\\downarrow\\sim \\Big[\\frac{\\wuppl\\wdownpl}{\\lp}\\Big]^2\\frac{\\lpar} {v_A} \\ . \\label{eq:epsilondownheuristic} \\end{equation} Because $\\epsilon^\\uparrow$ and $\\epsilon^\\downarrow$ both depend on the same combination of $\\wuppl$ and $\\wdownpl$---namely their product---the steady state solution is non-trivial. Had this degeneracy not occurred, e.g., had we found \\begin{equation} \\epsilon^\\uparrow\\sim {(\\wuppl)^{2+\\gamma}(\\wdownpl)^{2-\\gamma}\\over\\lp^2} {\\lpar\\over v_A} \\ \\ {\\rm and} \\ \\ \\epsilon^\\downarrow\\sim {(\\wuppl)^{2-\\gamma}(\\wdownpl)^{2+\\gamma}\\over\\lp^2} {\\lpar\\over v_A} \\ , \\end{equation} where $\\gamma\\ne 0$, then the solution would have been simple: $\\wuppl\\propto\\wdownpl\\propto\\lp^{1/2}$ and $(\\epsilon^\\uparrow/ \\epsilon^\\downarrow)\\sim(\\wuppl/\\wdownpl)^{2\\gamma}$, which follow from the constancy of $\\epsilon^\\uparrow$ and $\\epsilon^\\downarrow$ with $\\lp$.  But in weak MHD turbulence $\\gamma=0$ (eqs. [\\ref{eq:epsilonheuristic}] and [\\ref{eq:epsilondownheuristic}]); constancy of $\\epsilon^\\downarrow$ with $\\lp$ is forced by the constancy of $\\epsilon^\\uparrow$, and does not yield new information. One implication is that scaling arguments are insufficient to determine the flux ratio $\\epsilon^\\uparrow/\\epsilon^\\downarrow$. Physically, any flux ratio should be possible. But without the dimensionless coefficients of equations (\\ref{eq:epsilonheuristic}) and (\\ref{eq:epsilondownheuristic}), $\\epsilon^\\uparrow/\\epsilon^\\downarrow$ cannot be determined.  The coefficients cannot be obtained from scaling arguments; they depend on the spectral slopes of $\\wuppl$ and $\\wdownpl$ (which are related through $\\wuppl\\wdownpl\\propto\\lp$, eq. [\\ref{eq:intscal2}]). Galtier et al. (2000) calculated the coefficients using kinetic equations. We explain how in \\S \\ref{sec:steadystatefluxes}. Therefore these authors were able to relate the flux ratio to the spectral slopes.  The arguments presented thus far are still insufficiently constraining. Equations (\\ref{eq:epsilonheuristic}) and (\\ref{eq:epsilondownheuristic}) constrain only the product $\\wuppl\\wdownpl$. There are seemingly an infinite number of solutions with given values of $\\epsilon^\\uparrow$ and $\\epsilon^\\downarrow$, since $\\wuppl$ can be multiplied by any constant as long as $\\wdownpl$ is divided by this same constant. Furthermore, we expect on physical grounds that if the values of $\\wuppl$ and $\\wdownpl$ at a given lengthscale are fixed (instead of the values of $\\epsilon^\\uparrow$ and $\\epsilon^\\downarrow$), the cascade should be completely constrained; however, in this case the constancy of equations  (\\ref{eq:epsilonheuristic}) and (\\ref{eq:epsilondownheuristic}) leaves the $\\lambda$-dependence of $\\wuppl/\\wdownpl$ completely undetermined---even given the coefficients derived by Galtier et al. (2000). Do $\\wuppl$ and $\\wdownpl$ cross? Are they cut off by dissipation at the same scale? All of these problems for the imbalanced cascade can be resolved once the dynamics at the dissipation scale is understood.   \\subsection{Dynamics at the Dissipation Scale: Pinned Spectra} \\label{sec:pinned}  The main result of the present paper is that the energies of the up- and down-going waves are forced to equalize---they are ``pinned''---at the dissipation scale. This completely constrains the cascade. It is unusual that the dynamics at the dissipation scale has such an important influence. In this subsection we explain why pinning occurs. In \\S \\ref{sec:sses}, we give the resulting solution of the steady state cascade.  From equation (\\ref{eq:cascadetime}), the cascade time of the up-going waves is inversely proportional to the energy of the down-going ones: $\\tcasup= (\\lp/\\wdownpl)^2 (v_A/\\lpar)$, and similarly for the downgoing waves. We consider how the spectra evolve if initially, on lengthscales comparable to the dissipation scale,  waves going in one direction are more energetic than the oppositely directed ones.  To facilitate the discussion, we refer to Figure \\ref{fig:ke1} on page \\pageref{fig:ke1}, which presents the results from a numerical simulation that we discuss in detail in \\S \\ref{sec:fixedenergy}. In the middle panels of Figures \\ref{fig:ke1}a-d, we plot $\\eupp(k)\\sim(\\lp\\wuppl)^2$ and $\\edownp(k)\\sim(\\lp\\wdownpl)^2$ as functions of wavenumber $k=1/\\lp$. The initial condition is shown in Figure \\ref{fig:ke1}a. Initially, $\\wuppl>\\wdownpl$, so $\\tcasup>\\tcasdown$.  We consider a lengthscale-dependent dissipation time, $\\tdiss$, that is the same for both up- and down-going waves.\\footnote{For example, in the simulation presented in Figure \\ref{fig:ke1}, $\\tdiss\\simeq\\lp^2/\\nu$, where $\\nu$ is both the viscosity and the resistivity.} On large lengthscales, the dissipation timescale is much longer than the cascade times; $\\tdiss$ decreases faster with increasing $k$ than both $\\tcasup$ and $\\tcasdown$. The effects of dissipation are felt on lengthscales where $\\tdiss$ is comparable to---or smaller than---either $\\tcasup$ or $\\tcasdown$. Since $\\tcasup>\\tcasdown$ in the vicinity of the dissipation scale, the largest lengthscale at which dissipation effects are felt is where $\\tcasup\\sim\\tdiss$.  In Figure \\ref{fig:ke1}a, this is at $k\\sim 4,000$. We now let the spectra evolve; see Figs. \\ref{fig:ke1}b-c. We hold the energies at $k\\simeq 1$ fixed; this does not affect the short-time behaviour shown in Figs. \\ref{fig:ke1}b-c. Since $\\wuppl$ feels the dissipative effects at $k\\sim 4,000$, its spectrum is exponentially cut off at smaller scales.  This implies that the cascade time of the down waves, $\\tcasdown$, increases exponentially towards smaller scales. As a result, down-wave energy that is being cascaded from large to small scales cannot be cascaded fast enough at $k\\gtrsim 4,000$. Therefore the down-waves' energy flux is backed up, and the $\\wdownpl$ spectrum rises.  Furthermore, as $\\wdownpl$ rises, $\\tcasup$ falls, so the cascade time of the up-waves on small scales decreases, and the up-wave spectrum falls.  The final result is that the two spectra are pinned at the dissipation scale.  This pinning occurs very quickly: on the dissipation timescale. ",
        "conclusions": "\\label{sec:heuristic}  One of the virtues of weak MHD turbulence is that it can be analyzed in a mathematically rigorous way with perturbation theory; this yields a kinetic equation. Nevertheless, we begin with a qualitative description, which captures most of the features of the turbulent cascade.  \\subsection{Scaling Relation} \\label{sec:scaling}  MHD turbulence can be understood from the dynamics of $\\bwupp$ and $\\bwdownp$ (eqs. [\\ref{eq:eomrmhd1}-\\ref{eq:eomrmhd4}] for reduced MHD, dropping $\\perp$ subscripts). To linear order,  $\\bwupp$ is a wave that propagates up the mean magnetic fieldlines at the Alfv\\'en speed, $v_A$; $\\bwdownp$ propagates down at $v_A$. Each wave perturbs the mean magnetic fieldlines. Nonlinear terms describe the interaction between oppositely directed waves: each wave nearly follows the fieldlines perturbed by its collision partner. \\footnote{The equation for a scalar quantity $f$ that travels upwards at speed $v_A$, while following the magnetic fieldlines of the down-going $\\bwdownp$ is $(\\pt+v_A\\pz+\\bwdownp\\bcdot\\bnablap)f=0$. Equation (\\ref{eq:eomrmhd1}) for the vector $\\bwupp$ differs from this because of the pressure term, which is required to keep $\\bwupp$ incompressible, while conserving the energy $(\\wupp\\ )^2$.  Thus $\\bwupp$ does not exactly follow the fieldlines of $\\bwdownp$.  Nonetheless, this deviation does not greatly affect the behaviour of the turbulence.  Dissipation is a second effect that prevents the following of fieldlines. In the present discussion, we consider lengthscales that are sufficiently large that dissipation can be neglected.}   Consider an upgoing wavepacket that encounters a train of downgoing wavepackets. As the upgoing wave travels up the length of the downgoing train, it is gradually distorted.  It tries to follow the perturbed fieldlines in the downgoing train, but these fieldlines ``wander,'' i.e., the transverse separation between any two fieldlines changes. When the up-wave has travelled through a sufficiently large number of downgoing wavepackets that the amount of fieldline wander is comparable to the up-wave's transverse size, then the up-wave cascades.  To be quantitative, let each downgoing wave in the train have a typical amplitude $\\wdownpl$, a transverse size $\\lp$, and a parallel size $\\lpar$, where ``transverse'' and ``parallel'' refer to the orientation relative to the mean magnetic field. The most important collisions are between wavepackets of comparable transverse size (see \\S \\ref{sec:locality}). So let the upgoing wave have transverse size $\\lp$ as well.  Since each downgoing wavepacket has a typical perturbed magnetic field of magnitude $\\sim\\wdownpl$ (neglecting the factor of $1/2$), it bends the fieldlines by the angle $\\wdownpl/v_A$; the transverse displacement of a fieldline through this wavepacket is $(\\wdownpl/v_A)\\lpar$; and the wander of two typical fieldlines through the wavepacket, if they are initially separated by $\\lp$, is also $(\\wdownpl/v_A)\\lpar$.  In weak turbulence, the wander through a single wavepacket is smaller than the wavepacket's transverse size, \\begin{equation} {\\wdownpl\\over v_A}\\lpar\\ll \\lp \\ \\ {\\rm and} \\ \\ \\ {\\wuppl\\over v_A}\\lpar\\ll \\lp \\ . \\label{eq:inter} \\end{equation} When these inequalities are not satisfied, strong turbulence is applicable;  see \\S \\ref{sec:strong}. Thus in weak turbulence an upgoing wavepacket must travel through many downgoing ones before cascading. After $N$ downgoing wavepackets, fieldlines have wandered a distance $\\sim N^{1/2}(\\wdownpl/v_A)\\lpar$, assuming that wavepackets are statistically independent. The upgoing wavepacket is fully distorted---and hence cascaded---when the fieldlines it is following wander a distance $\\lp$, i.e., when $N\\sim(\\lp v_A/\\lpar\\wdownpl)^2$. Since each downgoing wavepacket is crossed in the time $\\lpar/v_A$, the cascade time of the upgoing wavepacket is \\begin{equation} \\tcasup\\sim\\Big({\\lp v_A\\over \\lpar\\wdownpl}\\Big)^2{\\lpar\\over v_A}\\sim \\Big({\\lp \\over\\wdownpl}\\Big)^2 {v_A\\over\\lpar} \\ . \\label{eq:cascadetime} \\end{equation} In this time, the upgoing wavepacket travels a distance $v_A\\tcasup$, which is much larger than its own length, $\\lpar$.\\footnote {We assume throughout this paper that the upgoing waves' parallel lengthscale is the same as that of the downgoing waves, $\\lpar$; the extension to the case when they differ is trivial, as long as the inequalities (\\ref{eq:inter}) are both satisfied, with the appropriate $\\lpar$'s.} The head and the tail of the upgoing wavepacket are both distorted by the same downgoing wavepackets; so both head and tail undergo nearly the same distortion as they cascade.\\footnote{The head of the upgoing wavepacket slightly distorts each downgoing wavepacket; so the downgoing wavepacket seen by the tail is slightly distorted relative to that seen by the head. Nonetheless, this backreaction is a higher-order correction that can be ignored in weak turbulence; see \\S \\ref{sec:prelim}. } Consequently, as the upgoing wavepacket cascades to smaller transverse lengthscales, it does not cascade to smaller parallel ones: \\begin{equation} \\lpar={\\rm scale \\ independent} \\ . \\label{eq:intscal1} \\end{equation} A proof of this follows from the 3-wave resonance relations (Shebalin, Matthaeus, \\& Montgomery 1983; see also \\S \\ref{sec:compare} of the present paper).  We calculate the steady state energy spectra by using Kolmogorov's picture of energy flowing from large to small lengthscales. The energy in up-waves flows from lengthscales larger than $\\lp$ to those smaller than $\\lp$ at the rate \\begin{equation} \\epsilon^\\uparrow\\sim\\frac{(\\wuppl)^2}{ \\tcasup} \\sim\\Big[\\frac{\\wuppl\\wdownpl}{\\lp}\\Big]^2\\frac{\\lpar} {v_A} \\label{eq:epsilonheuristic} \\ . \\end{equation} We call this simply the ``flux.'' We define $\\epsilon^\\uparrow$ more precisely below (eq. [\\ref{eq:flux}]). In steady state, the flux must be independent of $\\lp$, so \\begin{equation} \\wuppl\\wdownpl\\propto \\lp \\ . \\label{eq:intscal2} \\end{equation}  \\subsection{Insufficiency of Scaling Arguments for the Imbalanced Cascade} \\label{sec:insufficiency}  A balanced cascade has $\\wuppl=\\wdownpl\\equiv w_\\lambda$. Its solution in steady state is simple: $w_\\lambda\\propto \\lp^{1/2}$ and $\\epsilon^\\uparrow=\\epsilon^\\downarrow\\sim w_\\lambda^4\\lpar/\\lp^2 v_A$ (Goldreich \\& Sridhar 1997, Ng \\& Bhattacharjee 1997). However, if the cascade is imbalanced, a number of complications arise.  By the symmetry between up- and down-going waves, the down-going flux is given by the analogue of equation (\\ref{eq:epsilonheuristic}): \\begin{equation} \\epsilon^\\downarrow\\sim \\Big[\\frac{\\wuppl\\wdownpl}{\\lp}\\Big]^2\\frac{\\lpar} {v_A} \\ . \\label{eq:epsilondownheuristic} \\end{equation} Because $\\epsilon^\\uparrow$ and $\\epsilon^\\downarrow$ both depend on the same combination of $\\wuppl$ and $\\wdownpl$---namely their product---the steady state solution is non-trivial. Had this degeneracy not occurred, e.g., had we found \\begin{equation} \\epsilon^\\uparrow\\sim {(\\wuppl)^{2+\\gamma}(\\wdownpl)^{2-\\gamma}\\over\\lp^2} {\\lpar\\over v_A} \\ \\ {\\rm and} \\ \\ \\epsilon^\\downarrow\\sim {(\\wuppl)^{2-\\gamma}(\\wdownpl)^{2+\\gamma}\\over\\lp^2} {\\lpar\\over v_A} \\ , \\end{equation} where $\\gamma\\ne 0$, then the solution would have been simple: $\\wuppl\\propto\\wdownpl\\propto\\lp^{1/2}$ and $(\\epsilon^\\uparrow/ \\epsilon^\\downarrow)\\sim(\\wuppl/\\wdownpl)^{2\\gamma}$, which follow from the constancy of $\\epsilon^\\uparrow$ and $\\epsilon^\\downarrow$ with $\\lp$.  But in weak MHD turbulence $\\gamma=0$ (eqs. [\\ref{eq:epsilonheuristic}] and [\\ref{eq:epsilondownheuristic}]); constancy of $\\epsilon^\\downarrow$ with $\\lp$ is forced by the constancy of $\\epsilon^\\uparrow$, and does not yield new information. One implication is that scaling arguments are insufficient to determine the flux ratio $\\epsilon^\\uparrow/\\epsilon^\\downarrow$. Physically, any flux ratio should be possible. But without the dimensionless coefficients of equations (\\ref{eq:epsilonheuristic}) and (\\ref{eq:epsilondownheuristic}), $\\epsilon^\\uparrow/\\epsilon^\\downarrow$ cannot be determined.  The coefficients cannot be obtained from scaling arguments; they depend on the spectral slopes of $\\wuppl$ and $\\wdownpl$ (which are related through $\\wuppl\\wdownpl\\propto\\lp$, eq. [\\ref{eq:intscal2}]). Galtier et al. (2000) calculated the coefficients using kinetic equations. We explain how in \\S \\ref{sec:steadystatefluxes}. Therefore these authors were able to relate the flux ratio to the spectral slopes.  The arguments presented thus far are still insufficiently constraining. Equations (\\ref{eq:epsilonheuristic}) and (\\ref{eq:epsilondownheuristic}) constrain only the product $\\wuppl\\wdownpl$. There are seemingly an infinite number of solutions with given values of $\\epsilon^\\uparrow$ and $\\epsilon^\\downarrow$, since $\\wuppl$ can be multiplied by any constant as long as $\\wdownpl$ is divided by this same constant. Furthermore, we expect on physical grounds that if the values of $\\wuppl$ and $\\wdownpl$ at a given lengthscale are fixed (instead of the values of $\\epsilon^\\uparrow$ and $\\epsilon^\\downarrow$), the cascade should be completely constrained; however, in this case the constancy of equations  (\\ref{eq:epsilonheuristic}) and (\\ref{eq:epsilondownheuristic}) leaves the $\\lambda$-dependence of $\\wuppl/\\wdownpl$ completely undetermined---even given the coefficients derived by Galtier et al. (2000). Do $\\wuppl$ and $\\wdownpl$ cross? Are they cut off by dissipation at the same scale? All of these problems for the imbalanced cascade can be resolved once the dynamics at the dissipation scale is understood.   \\subsection{Dynamics at the Dissipation Scale: Pinned Spectra} \\label{sec:pinned}  The main result of the present paper is that the energies of the up- and down-going waves are forced to equalize---they are ``pinned''---at the dissipation scale. This completely constrains the cascade. It is unusual that the dynamics at the dissipation scale has such an important influence. In this subsection we explain why pinning occurs. In \\S \\ref{sec:sses}, we give the resulting solution of the steady state cascade.  From equation (\\ref{eq:cascadetime}), the cascade time of the up-going waves is inversely proportional to the energy of the down-going ones: $\\tcasup= (\\lp/\\wdownpl)^2 (v_A/\\lpar)$, and similarly for the downgoing waves. We consider how the spectra evolve if initially, on lengthscales comparable to the dissipation scale,  waves going in one direction are more energetic than the oppositely directed ones.  To facilitate the discussion, we refer to Figure \\ref{fig:ke1} on page \\pageref{fig:ke1}, which presents the results from a numerical simulation that we discuss in detail in \\S \\ref{sec:fixedenergy}. In the middle panels of Figures \\ref{fig:ke1}a-d, we plot $\\eupp(k)\\sim(\\lp\\wuppl)^2$ and $\\edownp(k)\\sim(\\lp\\wdownpl)^2$ as functions of wavenumber $k=1/\\lp$. The initial condition is shown in Figure \\ref{fig:ke1}a. Initially, $\\wuppl>\\wdownpl$, so $\\tcasup>\\tcasdown$.  We consider a lengthscale-dependent dissipation time, $\\tdiss$, that is the same for both up- and down-going waves.\\footnote{For example, in the simulation presented in Figure \\ref{fig:ke1}, $\\tdiss\\simeq\\lp^2/\\nu$, where $\\nu$ is both the viscosity and the resistivity.} On large lengthscales, the dissipation timescale is much longer than the cascade times; $\\tdiss$ decreases faster with increasing $k$ than both $\\tcasup$ and $\\tcasdown$. The effects of dissipation are felt on lengthscales where $\\tdiss$ is comparable to---or smaller than---either $\\tcasup$ or $\\tcasdown$. Since $\\tcasup>\\tcasdown$ in the vicinity of the dissipation scale, the largest lengthscale at which dissipation effects are felt is where $\\tcasup\\sim\\tdiss$.  In Figure \\ref{fig:ke1}a, this is at $k\\sim 4,000$. We now let the spectra evolve; see Figs. \\ref{fig:ke1}b-c. We hold the energies at $k\\simeq 1$ fixed; this does not affect the short-time behaviour shown in Figs. \\ref{fig:ke1}b-c. Since $\\wuppl$ feels the dissipative effects at $k\\sim 4,000$, its spectrum is exponentially cut off at smaller scales.  This implies that the cascade time of the down waves, $\\tcasdown$, increases exponentially towards smaller scales. As a result, down-wave energy that is being cascaded from large to small scales cannot be cascaded fast enough at $k\\gtrsim 4,000$. Therefore the down-waves' energy flux is backed up, and the $\\wdownpl$ spectrum rises.  Furthermore, as $\\wdownpl$ rises, $\\tcasup$ falls, so the cascade time of the up-waves on small scales decreases, and the up-wave spectrum falls.  The final result is that the two spectra are pinned at the dissipation scale.  This pinning occurs very quickly: on the dissipation timescale.      In this paper we describe imbalanced weak turbulence and solve the steady state cascade. In a future paper, we will extend the result to the strong cascade.  One of our ultimate goals is to develop a theory of imbalanced strong turbulence to apply to the solar wind, where imbalance is observed.  Although strong turbulence is more generally applicable than weak, the latter is a simple and illuminating model. There are a number of issues in strong turbulence that are not understood.   Weak turbulence can be used as the first step in explaining them.  For example, in this paper we examined the effect of a large magnetic Prandtl number on weak MHD turbulence.  We also found that the bottleneck effect appears in weak turbulence, where its interpretation is straightforward. There are a number of other issues that we intend to investigate in weak turbulence as a prelude to understanding them in strong turbulence; for example, reconnection and the turbulent dynamo."
    },
    "0208/astro-ph0208516_arXiv.txt": {
        "abstract": "We report the discovery of \\ion{Ge}{3}~$\\lambda$1088.46 in the planetary nebulae (PNe) SwSt~1, BD+30$^{\\rm o}$3639, NGC~3132, and IC~4593, observed with the \\emph{Far Ultraviolet Spectroscopic Explorer}\\footnote{Based on observations made with the NASA-CNES-CSA \\emph{Far Ultraviolet Spectroscopic Explorer}.  \\emph{FUSE} is operated for NASA by Johns Hopkins University under NASA contract NAS5-3298.}.  This is the first astronomical detection of this line and the first measurement of Ge ({\\it Z} = 32) in PNe.  We estimate Ge abundances using S and Fe as reference elements, for a range of assumptions about gas-phase depletions.  The results indicate that Ge, which is synthesized in the initial steps of the {\\it s}-process and therefore can be self-enriched in PNe, is enhanced by factors of $\\ge$ 3 -- 10. The strongest evidence for enrichment is seen for PNe with Wolf-Rayet central stars, which are likely to contain heavily processed material. ",
        "introduction": "Asymptotic giant branch (AGB) stars, the progenitors of planetary nebulae (PNe), are believed to be the primary synthesis site for heavy element isotopes produced by the slow neutron-capture or {\\it s}-process (K\\\"{a}ppeler et al. 1989; K\\\"{a}ppeler 1999).  Since enhanced surface abundances of \\emph{s}-process products are observed in evolved red giants (e.g. Smith \\& Lambert 1990), it is reasonable to expect such enrichments in PNe as well.  P\\'{e}quignot \\& Baluteau (1994) reported optical emission lines indicating overabundances for a large number of {\\it n}-capture elements in the PN NGC~7027, and Dinerstein (2001) identified two near-infrared lines commonly seen in PNe as fine-structure transitions of the light \\emph{n}-capture elements Kr and Se (\\emph{Z}=34, 36).  Germanium ($Z$=32) is more abundant than Kr and Se in the solar system, and, like these elements, can be synthesized in AGB stars of roughly solar metallicity (Busso et al. 1999; 2001).  In this Letter, we report the discovery of the strong resonance line \\ion{Ge}{3}~$\\lambda$1088.46 in four PNe observed with \\emph{FUSE}, and present evidence that Ge has been self-enriched in their precursor stars. ",
        "conclusions": "We report the discovery of \\ion{Ge}{3}~$\\lambda$1088.46 in four PNe observed with the \\emph{Far Ultraviolet Spectroscopic Explorer}.  This line, which arises in the ionized zone, is used to estimate the abundance of Ge in the nebular gas relative to the elements S and Fe, whose abundances are not altered by the star's evolution. We find convincing evidence for elevated abundances $\\ge$3--10 times solar (depending on assumed depletion factors) and therefore self-enrichment of Ge by {\\it s}-process nucleosynthesis in the progenitor stars.  This result demonstrates the potential of UV absorption-line spectroscopy to shed light on the operation of the {\\it s}-process in PN progenitor stars.  By determining the abundances of various heavy elements in PNe, one can set constraints on models of stellar nucleosynthesis and mixing in late stellar evolutionary stages, and directly investigate the process of chemical enrichment of the ISM by stars that evolve through the AGB and PN phases.   We are grateful to the \\emph{FUSE} operations and science teams, whose exceptional efforts have made this facility available to the astronomical community.  We thank S. McCandliss for providing data on H$_2$, O. De Marco for sharing her STIS spectrum of SwSt~1, and D. Lambert and C. Sneden for helpful comments.  Financial support was provided by NASA contracts NAG5-9239, NAG5-11597, and NSF grant AST 97-31156."
    },
    "0208/astro-ph0208385_arXiv.txt": {
        "abstract": "We report here on a sample of resolved, infrared images of galaxies at z$\\sim$0.5 taken with the 10-m Keck Telescope's Adaptive Optics (AO) system. We regularly achieve a spatial resolution of 0.05$\\arcsec$ and are thus able to resolve both the disk and bulge components. We have extracted morphological information for ten galaxies and compared their properties to those of a local sample. The selection effects of both samples were explicitly taken into account in order to derive the unbiased result that disks at z$\\sim$0.5 are $\\sim$0.6 mag arcsec$^{-2}$ brighter than, and about the same size as, local disks. The no-luminosity-evolution case is ruled out at 90\\% confidence. We also find, in a more qualitative analysis, that the bulges of these galaxies have undergone a smaller amount of surface brightness evolution and have also not changed significantly in size from z$\\sim$0.5 to today. This is the first time this type of morphological evolution has been measured in the infrared and it points to the unique power of AO in exploring galaxy evolution. ",
        "introduction": "Observational constraints on the growth of galaxies and the role of star formation within them over time are finally within the reach of current data. Powerful new techniques, thanks to modern space-based and ground-based telescope facilities, have allowed researchers to make more precise measurements of galaxy properties. In this paper we focus on one aspect of this progress that is especially promising -- resolved images of distant galaxies.  There has been a series of recent studies taking advantage of resolved data to measure the evolution of disk properties, especially B-band disk surface brightness to z$\\sim$1 from Hubble Space Telescope (HST) imaging.  These studies have provided some constraints on the amount of evolution taking place, but there have been contradictory results, mostly due to disagreements about the impact of selection effects. \\citet{schade} and \\citet{roche} found B-band disk surface brightness evolution of $\\sim$1.6 mag to z=0.73 and $\\sim$0.95 mag to z=0.9, respectively, but neither focused on the impact of selection effects on their results. \\citet{lil} also found a high level of surface brightness evolution ($\\sim$0.8 mag to z=0.67), but calculated that as much as 0.3 mag of this could be due to selection effects.  \\citet{vogt} obtained rotation curves of galaxies to z$\\sim$1 and found relatively mild evolution in the Tully-Fischer relation ($\\Delta M_B \\lesssim$0.2 mag), although their selection effects were very complicated.  \\citet{simard} concluded that almost all of the disk surface brightness evolution they detected ($\\sim$1.3 mag to z=1) is due to the selection effects. They also reanalyzed the \\citet{schade} and \\citet{roche} studies and found that selection effects could account for most of the evolution seen there as well. Both \\citet{vogt} and \\citet{simard}, however, do detect (even after accounting for selection effects) a population of galaxies at z$>$0.5 that have higher surface brightness than the locus of nearby galaxies.  The progress derived from this resolved data has, moreover, been limited to observations of optical disk properties.  To date the only resolved images of distant galaxies in the infrared (IR) have been those taken with NICMOS on HST. However these data mostly provided global colors and luminosities \\citep[e.g.][]{tep}.  The resolution of the camera ($\\theta=0.16\\arcsec$) was insufficient to get reliable morphological information and profile fitting routines failed more than 50\\% of the time \\citep{corb}.  Observing in the IR reduces problems due to morphological K corrections that can arise with optical images which sample rest-frame ultraviolet (UV) light at high redshifts \\citep[e.g.][]{bunk}. IR data more directly sample the mass distributions of distant galaxies since such images are less biased towards regions of high star formation.  Although a picture of the structure and kinematics of nearby bulges is beginning to emerge \\citep{bur}, resolved bulge observations have also been rare at high redshift. HST optical data does offer resolved images of bulges to z$\\sim$1, but most analysis of these data has been confined to such things as bulge to disk ratios and bulge colors \\citep[e.g.][]{lil, ell}.   In order to study resolved structures, including bulges, we have started a campaign to exploit high-resolution, IR images of distant galaxies from the 10-m Keck Telescope's Adaptive Optics (AO) system.  These data regularly have a resolution of $\\sim$0.05$\\arcsec$ in the H band, equal to that of the HST optical data and three times higher than that of NICMOS.  This resolution is sufficient to resolve bulges to arbitrary redshifts.  The AO observations of the galaxies in our sample are described in \\S2 along with our methods of obtaining redshifts.  In \\S3 we discuss the procedure used to extract galaxy morphological information from the AO images and in \\S4 we present the comparison of these data to a local sample, taking into account the selection effects for both samples. In \\S5 we discuss the implications of our result for galaxy evolution. Throughout this paper, we use H$_{0}$=65 km s$^{-1}$ Mpc$^{-1}$, $\\Omega_M$=0.25, and $\\Omega_{\\Lambda}$=0.75. ",
        "conclusions": ""
    },
    "0208/astro-ph0208450_arXiv.txt": {
        "abstract": "We present chemical abundance measurements in the $z_{abs}=3.35045$ Damped Lyman-$\\alpha$ (DLA) system observed in the UVES spectrum of the BAL quasar \\astrobj{BR 1117$-$1329}. We measure a neutral hydrogen column density $N (H I) =6.9 \\pm 1.7 \\times 10^{20}$ atoms cm$^{-2}$ and derive mean abundances relative to solar: $[Si/H] = -1.26\\pm0.13$, $[Fe/H]=-1.51\\pm0.13$, $[Ni/H]=-1.57\\pm0.13$, $[Cr/H]=-1.36\\pm0.13$, $[Zn/H]=-1.18\\pm0.13$, $[Al/H]>-1.25$, $[O/H]>-1.25$ and $[N/H]<-2.24$. This is the third measurement of Zn, an element mildly depleted onto dust grain, at $z_{abs}>3$. The iron to zinc and chromium to zinc ratios, $[Fe/Zn]=-0.33\\pm0.05$ and $[Cr/Zn]=-0.18\\pm0.05$ demonstrate that the absorber has a low dust content. The nitrogen ratio $[N/Si]<-0.98$ suggests that the ``secondary'' N production process is taking place in this DLA. Finally, this absorber does not seem to present a convincing $\\alpha$-enhancement as shown by the $\\alpha$ over Fe-peak element ratios: $[Si/Fe]=0.25\\pm0.06$, $[Si/Cr]=0.10\\pm0.06$ and $[Si/Zn]=-0.08\\pm0.06$. ",
        "introduction": "The chemical abundances of the highest H I column density quasar absorbers ($N(HI)> 2 \\times 10^{20}$ atoms cm$^{-2}$), the Damped Lyman-$\\alpha$ systems (DLAs), are commonly expressed using column density weighted measurements. They are used as observational tracers of the cosmological evolution of metallicities, up to very high redshifts (see for example Pettini et al. 1994, 1997; Lu et al.  1996; Prochaska \\& Wolfe 1999, 2000, 2001; Dessauges-Zavadsky \\etal\\ 2001; Ledoux, Bergeron \\& Petitjean 2002 and references therein).  However, recent studies give somehow surprising results: Prochaska \\& Wolfe (2002) claim no evolution of the mean weighted iron metallicity over the redshift range $1.7-3.5$, in contradiction with predictions from essentially all chemical evolution models. A possible explanation for this result is that iron is not the optimal element for tracing metals on cosmological scales, because it is easily depleted onto dust grains.  Dust depletion greatly complicates the interpretation of the metal content of DLAs. Nevertheless at very high-redshift ($z>3.8$), there are evidences that [Fe/H] is beginning to fall, in such a way that at $z=5$ the metallicity is substantially lower than at $z<4$ (Prochaska, Gawiser \\& Wolfe 2001, Songaila \\& Cowie 2002).  \\begin{figure} \\psfig{figure=peroux_fig1.ps,angle=0,width=1.0\\textwidth,height=0.6\\textheight} \\caption{ Voigt profile fits to the Lyman series lines (Ly-$\\alpha$, Ly-$\\beta$ and Ly-$\\gamma$) of the $z=3.35045$ DLA. The vertical bar in each panel indicates the wavelength centroid of the component used for the best fit, which is shown as a solid line together with 1 $\\sigma$ errors (dashed lines). The resulting total HI column density is $N$(HI)$=6.92 \\pm 1.7 \\times 10^{20}$ atoms cm$^{-2}$.} \\label{fig:DLA} \\end{figure}   Pettini \\etal\\ (1994) estimated the dust content of DLAs with the help of a comprehensive survey of Zn measurements, an element which is known to be only slightly depleted onto dust grains, and thus provide an unbiased tracer of metallicities. However such measurements are challenging because of both the paucity and weakness of Zn features in quasar absorbers. At the moment, only two measurements of [Zn/H] have been made in DLAs at z$>$3 (Molaro \\etal\\ 2000, Levshakov \\etal\\ 2001). Here, we present a third detection of Zn in a DLA at z$\\sim$3.35, thus providing clues on the dust-free abundance determination in high-redshift Damped Lyman-$\\alpha$ systems.  We first detail the observational set-ups and data reduction processes of the quasar spectrum. We then present the analysis and chemical abundance determination for a number of elements in the DLA studied, providing a short discussion on the consequences of these additional measurements at high-redshift. ",
        "conclusions": "In the DLA studied here, we observe a velocity shift between the low and the high ionization ionic features. The fact that the high-ionization profiles show much more disturbed velocity structure than the low-ionization profiles has already been noticed in the past (Lu \\etal\\ 1996; Ledoux \\etal\\ 1998). Haehnelt, Steinmetz \\& Rauch (1998) have suggested that such feature could be the signature of merging protogalactic clumps. Indeed they use numerical simulations to model the Si~{\\sc ii}$\\lambda$1808 and C~{\\sc iv}$\\lambda$1548 line profiles and note that the absorption features vary independently in the high ionization and low ionization species since the C~{\\sc iv} absorption arises mainly from the warmer gas outside the self-shielding region of DLAs. Nevertheless, other numerical simulations fail to reproduce the DLA kinematics in a self-consistent manner (e.g. Prochaska \\& Wolfe 2001) and actually require that a significant fraction of DLAs have $v_{circ} \\sim$ 150 km s$^{-1}$ (Maller \\etal\\ 2001) even within the Cold Dark Matter cosmology. In other words, if one wants to interpret the global kinematics of the whole DLA population with a simple model, the kinematics of the strongest systems might rule out a global dwarf galaxy scenario if one adopts $v_{circ} <$ 100 km s$^{-1}$ as the dwarf criterion. Nevertheless, note that in the case of BR~1117$-$1329, the neutral component has a velocity width of about 50~km~s$^{-1}$ and that the strongest C~{\\sc iv} components are seen on both side (at $-$50 and +130~km~s$^{-1}$) of the low-ionization profile. This is suggestive of a galactic wind from a dwarf galaxy.    \\begin{figure} \\psfig{figure=peroux_fig5.ps,angle=-90,width=0.6\\textheight,height=0.45\\textwidth} \\caption{Fe II $\\lambda$ 1608 and the C IV doublet ($\\lambda$ 1548 and $\\lambda$ 1550\\AA) overplotted on the same velocity scale. This figure illustrates the shift between the center of the low and the high ionization ionic features in the DLA. The zero velocity is fixed at the redshift of the absorption feature: $z_{abs}=3.35045$.} \\label{fig:overplot_ions} \\end{figure}  The detection of N~{\\sc i} in this DLA system allows us to probe the chemical history of the system. Indeed, the [N/O] ratio is useful to disentangle the contribution from the ``primary'' and ``secondary'' production processes for N enrichment.  The limits obtained here suggests that [N/O]$\\le-0.9$ which is slightly lower than what one expects (i.e., $\\simeq-0.6$) if the primary process dominates (Vila-Costas \\& Edmunds, 1993). Like in most of the DLAs the [O/N] limits for the measured $\\alpha-$element metallicity is consistent with the delayed production of N compared to O and inconsistent with the primary production of N from the massive stars in the low metallicity gas (Lipman, Pettini \\& Hunstead, 1995; Lu \\etal\\, 1998; Centuri\\'on \\etal\\ 1998; Pettini \\etal\\ 2002 {\\it in press}; Prochaska \\etal\\, {\\it in press}).   The abundances of the DLA have been studied in detail. In this DLA, we find a zinc abundance slightly higher than the iron one: $[Fe/Zn]=-0.33\\pm0.05$ which suggests that the amount of dust present in this absorber is rather low. This is further supported by the chromium over zinc ratio $[Cr/Zn]=-0.18\\pm0.05$, another tracer of dust. Thus in this DLA the abundances of refractory elements (such as Fe and Cr) are in line with the one from the non-refractory element Zn. In addition, Al~{\\sc iii}/Al~{\\sc ii} is less than 5\\%. This means that the effect of ionization correction in the measurement of metallicity in this system is almost negligible (Vladilo et al. 2001) and the absolute abundances derived here do not suffer severe biases due to dust depletion.  We derive various $\\alpha$-chain element to Fe-peak element ratios: $[Si/Fe]=0.25\\pm0.06$, $[Si/Cr]=0.10\\pm0.06$ and $[Si/Zn]=-0.08\\pm0.06$.  Thus the abundance pattern in this system is consistent with no $\\alpha-$enhancement with very little depletion in Fe and Cr with respect to Zn. The results indicate that the Type~II~SNe enrichment process does not dominate in this system, possibly because the Type~I~SNe enrichment process is already important. The fact that we observe the DLA when Type~I~SNe processes start to dominate over Type~II~SNe can not easily be interpreted since the turn-over depends on both the star formation rate and the initial mass function (Matteucci \\& Recchi 2001).   \\begin{table} \\begin{center} \\caption{Summary table of the column densities of the low and high ionization species in the $z_{abs}=3.35045$ DLA.} \\label{tab:summary} \\begin{tabular}{lclc} \\hline \\hline Ion\t&$\\log$ [N(X)/N(H)]$_\\odot$ (Ref)&$\\log$ N(X)&[X/H]\\\\ \\hline H I \t &...\t\t&20.84$\\pm$0.12 &...\\\\ Fe II    &$-$4.50 (1)\t&14.83$\\pm$0.03\t&$-$1.51$\\pm$0.13\\\\ Si II    &$-$4.45 (1)\t&15.13$\\pm$0.05\t&$-$1.26$\\pm$0.13\\\\ Ni II    &$-$5.75 (1)\t&13.52$\\pm$0.03\t&$-$1.57$\\pm$0.13\\\\ Cr II    &$-$6.33 (1)\t&13.15$\\pm$0.03\t&$-$1.36$\\pm$0.13\\\\ Zn II    &$-$7.40 (1)\t&12.26$\\pm$0.03\t&$-$1.18$\\pm$0.13\\\\ Al II    &$-$5.53 (1)\t&$>$14.06      \t&$>-$1.25\\\\ OI\t &$-$3.26 (2)   &$>$16.33\t&$>-$1.25\\\\ NI\t &$-$4.07 (2)\t&$<$14.53\t&$<-$2.24\\\\ C IV     &...    \t&14.13$\\pm$0.01\t&...\\\\ Al III   &...\t\t&12.36$\\pm$0.03\t&...\\\\ \\hline \\hline \\end{tabular} \\vspace{.5cm}  References:\\\\  [1] Grevesse \\& Savage 1998.\\\\  [2] Holweger 2001.\\\\  \\end{center} \\end{table}   We derive a Zn abundance, $[Zn/H]=-1.18\\pm0.13$ at $z_{abs}=3.35$, which is higher than the very few measurements available at these redshifts (corrected to the same solar value as use in the present study): Molaro \\etal\\ (2000) derives $[Zn/H]=-2.00\\pm0.1$ in a DLA at $z_{abs}=3.39$ towards QSO 0000$-$2620 and Levshakov \\etal\\ (2002) find $[Zn/H]=-1.43\\pm0.08$ in an absorber at $z_{abs}=3.02$ towards QSO 0347$-$3819 although the metallicity in this absorber is uncertain and could be larger (see Ledoux, Srianand \\& Petitjean 2002). The resulting column density weighted mean at $z>3$ is $[<Zn/H_{DLA}>]=-1.63$, i.e. dominated by the highest HI column density, which here corresponds to the lowest Zn measurement from Molaro \\etal\\ (2000). For comparison, Pettini \\etal\\ (1997) find $[<Zn/H_{DLA}>]=-1.18$ from $z=0.69$ to $z=3.39$ (where the measures at $z>3$ are upper limits). Although statistics are small, our result suggests an increase of metallicities from $z>3$ to present times. Previous compilations of Zn measurements at all redshifts have shown that the column density weighted metallicity does not evolve with time (Pettini \\etal\\ 1997, Vladilo \\etal\\ 2000, Prochaska \\& Wolfe 2002). These authors argue however that this cannot be directly interpreted as a lack of evolution of the cosmic metallicity of DLAs since measurements are still affected by low number statistics and possible bias against high column density, high metallicity absorbers (as first pointed out by Fall \\& Pei 1993 and Boiss\\'e \\etal\\ 1998). The presence of this bias is controversial at the moment (Ellison \\etal\\ 2001, Prochaska \\& Wolfe 2002, Petitjean \\etal\\ 2002) and many more measurements of undepleted metals in quasar absorbers are needed at $z>3$ before any conclusion can be drawn."
    },
    "0208/astro-ph0208499_arXiv.txt": {
        "abstract": "We have started a survey of galaxies at intermediate redshifts using the HST-STIS parallel fields. Our main goal is to analyse the morphology of faint galaxies in order to estimate the epoch of formation of the Hubble classification sequence. The high resolution of STIS images (0.05$''$) is ideal for this work and enable us to perform a morphological classification and to analyse the internal structures of galaxies. We find that 40\\% of the 290 galaxies are early types and that there are more irregulars and ellipticals at the fainter magnitudes. ",
        "introduction": "One of the key questions in galaxy evolution is the epoch of the assembling of the Hubble types. How and when do galaxies acquire their shape? Are ellipticals assembled in a monolithic collapse in the early universe or do they form from subunits as predicted in the hierarchical clustering scenario? In the latter scenario stars are formed continuously over a wide range of redshift and galaxies are assembled via many generations of mergers of smaller subunits (e.g. White \\& Frenk 1991, Kauffmann et al. 1993). This prediction is strongly supported by the compilation of the cosmic star formation history (a.k.a. Madau plot) where no particular epoch of star formation is seen (Madau et al. 1996, Blain 2001). This has frequently been used against the monolithic collapse theory which predicts rapid star formation at very high redshift (z$>$2) followed by a steep decline in the star formation rates (Eggen et al. 1962; Jimenez et al. 1998). However, in the Madau plot, the $\\rho$$_{SFR}$ at a given redshift could be either due to massive galaxies experiencing modest bursts of star formation or newly formed dwarf galaxies. The distinction between these two types of galaxies is the main difference between the two models. One way to decide between the models is to search for their predictions. It is our goal to construct a database of galaxies at intermediate-{\\it z} which will illustrate the galaxy population at that particular epoch. Therefore our database will be useful when searching for the predicted types of galaxies at intermediate-{\\it z}. ",
        "conclusions": ""
    },
    "0208/astro-ph0208391_arXiv.txt": {
        "abstract": "{ We present the results of mid-resolution spectroscopy in the \\ion{Li}{I} 6708 \\AA~ spectral region of Asymptotic Giant Branch (AGB) stars belonging to young open clusters of the Large Magellanic Cloud. Most stars belong to the clusters \\object{NGC 1866} and \\object{NGC 2031}, which have an age of $\\simeq 150$~Myr. Lithium lines of different strength are detected in the spectra of stars evolving along the AGB, in agreement with theoretical predictions. According to stellar evolutionary models, at the start of the AGB the stars should all show a low residual lithium abundance as a consequence of  dilution during the previous evolutionary phases. The most luminous and cooler thermally pulsating  AGB stars, if they are massive enough, once in the AGB go first through a phase of Li destruction, which is followed by a phase of strong lithium production and further destruction. The production of lithium on the AGB is in particular  explained by the onset of the {}``Hot Bottom Burning\\char`\\\"{} (HBB) process. Our most conclusive results are obtained for the populous cluster \\object{NGC 1866} where: the `early--AGB' stars show a weak Li line, which can be attributed to the dilution of the  initial abundance; one of the two luminous stars seem to have completely depleted lithium, as no line is detected;  the second one shows a deep lithium line, whose strength can be explained by lithium production. The bolometric magnitude of these stars are consistent with the values predicted by the theory, for the mass evolving on the AGB of \\object{NGC 1866}, at which lithium is first destroyed and then produced (\\Mbol $\\simeq-6$). We  also analyze the infrared luminosities (ISOCAM data) of these stars, to discuss if their evolutionary phase precedes or follows the lithium production stage. More intriguing and less clear results are obtained for the most luminous stars in \\object{NGC 2031}: the brightest star seems as well to have destroyed lithium, while the second one shows a strong lithium line. However its bolometric luminosity --derived from the near--IR photometry, is much lower (\\Mbol $\\simeq -5.2 \\pm 0.15$)  than that expected from HBB models. Although low luminosity lithium rich AGB stars are also known, whose appearance is attributed to non--canonical mixing processes, it is not clear why two almost coeval clusters show such a different behaviour. It is also possible that this star does not belong to \\object{NGC 2031}. Finally we suggest the observational tests that could shed further light on this matter.} ",
        "introduction": "The evolution of lithium in intermediate mass stars is a powerful probe of the convective envelope conditions. In particular, during the Asymptotic Giant Branch (AGB) phase in stars of mass large enough for the Hot Bottom Burning (HBB) process to occur, the lithium abundance has a complex behaviour, whose knowledge can shed light on the efficiency of convection and on the occurrence of convective overshooting in the preceding phases. The most suitable targets for such a study are the relatively massive and thermally pulsating (TP) AGB stars in (young) clusters, whose mass can, in principle, be derived by fitting the cluster color-magnitude diagram. The need for a relatively well populated AGB naturally favors the choice of young clusters in the Large Magellanic Cloud.  We carried out, therefore, an observing program to explore the lithium abundance in the TP--AGB phase of stars  belonging to four Large Magellanic Cloud open clusters and whose initial mass was determined by fitting the morphology of the color-magnitude diagram (CMD), turnoff included. We obtained mid-resolution spectra of several stars in \\object{NGC 1866} and \\object{NGC 2031} (which have ages \\( \\simeq 150 \\)~Myr), of one star in the younger (\\( \\simeq 100 \\)~Myr old) cluster \\object{NGC 2214} and of one star in the older \\object{NGC 2107} (\\( \\simeq 250 \\)~Myr) \\citep[e.g.][]{corsi94,girardi95}. The most luminous (three) stars in \\object{NGC 1866} and (two) in \\object{NGC 2031}, and the stars in the other clusters, were selected from the list of \\citet[][ hereafter FMB90]{frogel90}. Being these also the latest spectral type clusters' stars, they are good TP--AGBs candidates. Additional stars were selected in \\object{NGC 1866} and \\object{NGC 2031}, as good candidates for the `early--AGB' phase of evolution on the base of our own near IR photometry.  We looked for and derived the strength of the lithium line at \\( \\lambda = 6707.8 \\) \\AA, and explored its dependence on the AGB luminosity and on the cluster age. The observed spectra were compared with synthetic ones, for evaluation of the lithium abundance. Though a higher dispersion is required for a precise abundance determination, our mid-resolution spectra already provide interesting results.  In the next Sections we present the theoretical background of our project (Sect. 2), the criteria for cluster and target selection (Sect. 3), the observations and data reduction (Sect. 4), the analysis of the lithium abundance (Sect. 5)  and its theoretical implications (Sect. 6), the analysis of the evolutionary stage represented by the selected stars (Sect. 7), and the final discussion and conclusions (Sect. 8). ",
        "conclusions": "\\begin{enumerate} \\item The early AGB stars in \\object{NGC 1866} show lithium abundances which are roughly constant with increasing luminosity, consistent with the lithium dilution expected to have taken place during the previous evolutionary stages. The average abundance found is \\( \\log N(\\mathrm{Li})\\simeq 0.0\\pm 0.5 \\), implying, however, stronger than standard dilution. This result is confirmed by at least  another early AGB star in \\object{NGC 2031}, although the data derived from other stars analyzed in this cluster are inconclusive; \\item We have detected three cool luminous AGB stars in \\object{NGC 1866}. The faintest one (\\#3) shows a lithium abundance consistent with the remnant abundance expected from an AGB star at the beginning of this phase as a consequence of previous lithium dilution. The brightest one (\\#1) does not show any lithium and the upper limit derived  (\\( \\log N(\\mathrm{Li})\\leq -0.5 \\)) suggests that this star plausibly is  in the phase of HBB preceding lithium production (confirmed by the smaller mid-IR excess detected by ISOCAM). The other most luminous AGB star (\\#2) has a larger lithium abundance (\\( \\log N(\\mathrm{Li})\\simeq 1.5\\pm 0.5 \\)), which we can attribute to production by HBB.  On the basis of our models we would expect a few ($\\sim 5$) other AGBs in the field of \\object{NGC 1866}, while none were found, neither luminous in the optical, nor in the mid-IR. This result put interesting constraints on the duration of the AGB phase and the severity of the mass loss during this phase. \\item The most luminous star in \\object{NGC 2031} (\\#1) is found to be similar to star \\#1 in \\object{NGC 1866}. The second brightest star in the K-band in this cluster (\\#2) has the largest lithium abundance in the sample, but Li production is not predicted at its derived luminosity (HBB not active). Further observations are needed to confirm the abundance analysis and the membership of this star. \\end{enumerate} We conclude that, though many points still remain unclear in the interpretation of the observations,  we are in presence of an interesting sample of stars, whose further careful analysis can  shed light on the expected evolutionary paths. Observations at a higher dispersion are needed for these stars to clarify their evolutionary status: for example, the presence of {\\it s}-process elements in the spectra would be an important indicator of how many thermal pulses the stars have gone through. Our best guess is that they still are at the beginning of the TP phase, so that we should \\textit{not} expect a sensible {\\it s}-process abundance enhancement."
    },
    "0208/astro-ph0208358_arXiv.txt": {
        "abstract": "A comprehensive study of relativistic and resonance effects in \\eie of (e+Fe~XVII) is carried out using the \\bprm (BPRM) method in the relativistic \\cc (CC) approximation. Two sets of eigenfunction expansions are employed; first, up to the $n$ = 3 complex corresponding 37 fine-structure levels (37CC) from 21 LS terms; second, up to the \\n = 4 corresponding to 89 fine-structure levels (89CC) from 49 LS terms. In contrast to previous works, the 37CC and the 89CC collision strengths exhibit considerable differences. Denser and broader resonances due to \\n = 4 are present in the 89CC results both above and {\\it below} the 37 thresholds, thus significantly affecting the collision strengths for the primary X-ray and EUV transitions within the first 37 \\n = 3 levels. Extensive study of other effects on the collision strengths is also reported: (i) electric and magnetic multipole transitions E1,E2,E3 and M1,M2, (ii) J-partial wave convergence of dipole and non-dipole transitions, (iii) high energy behaviour compared to other approximations. Theortical results are benchmarked against experiments to resolve longstanding discrepancies --- collision strengths for the three prominent {\\small X}-ray lines 3C, 3D and 3E at 15.014, 15.265, and 15.456~\\AA\\ are in good agreement with two independent measurements on Electron-Beam-Ion-Traps (EBIT). Finally, line ratios from a collisional-radiative model using the new collisional rates are compared with observations from stellar coronae and EBITs to illustrate potential applications in laboratory and astrophysical plasmas. ",
        "introduction": "The ground configuration of \\Fe17 has a stable closed L-shell structure (neon core) rendering \\Fe17 the dominant Fe ion species in many laboratory and cosmic plasmas. Owing to its importance \\Fe17 has also become a benchmark ion, and one of the most extensively studied for its spectral diagnostic potential in plasma- and astro-physics. Both the atomic structure of \\Fe17 and \\eie (EIE) of \\eFe17 are extremely complex. Although EIE of \\Fe17 has been investigated experimentally and theoretically for a long time \\cite{bh92}, there are considerable uncertainties in line intensity comparisons between measurements and theories, even with the most elaborate methods.  Unlike lower ionization stages of Fe, Fe~XVII is a medium-to-highly charged many-electron ion with strong correlation and relativistic effects in both the target dynamics and the electron collision processes \\cite{ch99}, particularly for non-dipole forbidden and intercombination transitions and near-threshold cross sections. Both intermdediate coupling LSJ, and jj-coupling, are needed in atomic structure calculations for proper target level designations. Precise atomic structure calculation is a crucial part of EIE calculations. However, despite a number of experimental and theoretical works, accurate atomic structure information on \\Fe17 is not available in literature. The first 27 levels of \\Fe17 in the configurations 2p$^5$3l (l=s,p,d) have been determined experimentally. But consistent energies for inner-shell excitation configurations 2s2p$^6$3l (l=s,p,d) (from level 28 to 37) are yet to be obtained. The same is true for the \\n = 4 configurations (2p$^5$4l (l=s,p,d,f); 2s2p$^6$4l (l=s,p,d,f)). Systematic theoretical structure calculations have been used to determine energies not available from experimental measurements, with an estimated accuracy of less than a few to ten per cent, and an uncertainty of $\\sim$0.1$\\AA$ in wavelengths to the ground state.  Transition probabilities and \\eie collision strengths of \\Fe17 are more inaccurate. We calculate transition probabilities using the code \\sss, which provides the structure input for the EIE calculations of \\Fe17 \\cite{ei74}, and the  multi-configuration Dirac-Fock (MCDF) method using the {\\small GRASP} code with extensive configuration-expansion \\cite{dy89}. All E1, M1, E2 transitions among the 89 levels are calculated using \\sss and {\\small GRASP} codes. We find some large M2 and E3 A-values from the comprehensive {\\small GRASP} data, and for the first time, point out their importance in spectral formation of \\Fe17.  Similar to the study of the \\Fe17 structure, there is a long history of EIE calculations for \\Fe17. Basically, there are three types of EIE calculations in literature. (i) Distorted-wave (DW) and relativistic distorted-wave (RDW), which couple initial and final channels for a given transition and thus only background cross sections are calculated. (ii) Isolated resonance approximation plus (R)DW: including a limited number of resonances in the isolated resonance approximation (IRA). This approach improves (i) somewhat;  however we show that the final results are not very accurate due to the inevitably limited number of channels in the IRA, and breakdown of IRA due to overlapping of dense and broad resonances in \\Fe17 \\cite{ch02}. In some cases, the results from this approach may be misleading or confusing owing to unknown missing channels. (iii) Close coupling (CC) approach: non-relativistic R-matrix calculations in LS coupling, and relativistic BPRM calculations in intermediate coupling. While this is the most advanced method, and in principle capable of including all important atomic effects, several issues require extreme care, as demonstrated in this work. Interestingly, there is no complete R-matrix calculations in literature for \\Fe17 either. One previous coupled-channel calculation by Mohan \\etal \\cite{mo97} employed 15 LS terms, followed by a pair coupling transformation to obtain fine-structure collision strengths between levels in the first 27 levels. However, Mohan \\etal obtained only the background collision strengths, similar to the various DW calculations, because their calculations were above the highest threshold and therefore no resonances were included. Nontheless their data have been used to compare with experiments and observations, and the conclusion that resonance enhancement is not important has been drawn. Recently, limited BPRM calculations for \\Fe17 were reported \\cite{gu00}. Similar to the work in \\cite{mo97}, only 27CC was employed in the calculations. Also, no detailed resonance structures in collision strengths were presented in this work either and thus it prevents the in-depth investigations. In order to address these and other issues thoroughly, it is therefore necessary to carry out a full set of BPRM calculations for \\eFe17. There are some R-matrix calculations for the neon isoelectronic sequence for other ions. Ne-like selenium has been calculated by fully relativistic Dirac R-matrix approach \\cite{wi91} and BPRM approach \\cite{gu89} using a 27CC expansion in jj-coupling and in LSJ coupling, respectively.  In the present work, relativistic and resonance effects are considered with two sets of calculations: 21 terms or 37 fine-structure levels (37CC) up to the \\n = 3 complex, and 49 terms or 89 fine-structure levels (89CC) including up to the \\n = 4 levels. The 37CC BPRM calculation includes more than 7300 channels, with the largest Hamiltonian matrix of dimension 3659; the 89CC BPRM calculation involves more than 20,000 channels overall, with and the largest the Hamiltonian matrix of dimension 10086 (this is possibly the largest BPRM calculation yet carried out). There are pronounced differences between the two sets of calculations. The results show that the 89CC calculation with levels up to \\n = 4 is necessary to obtain accurate collision strengths even for the transitions up to the \\n = 3, since the \\n = 4 resonances appear both above and, surprisingly,  {\\it below} the \\n = 3 levels and affect the effective collision strengths significantly. The two sets of calculations also show that the backgrounds for some important transitions are also affected. The implications of this study on \\Fe17 for Ne-like ions and other highly charged ions are pointed out.  As mentioned, in addition to theoretical studies there have been several experimental measurements for \\Fe17. For example, experimental line intensity ratios have recently been measured on Electron-Beam-Ion-Traps (EBIT) at Lawrence Livermore National Laboratory (LLNL, Brown \\etal 1998), and at the National Institute for Standards and Technology (NIST, Laming \\etal 2000), for some prominent and strong lines of \\Fe17 that have been observed in the {\\small X}-ray spectra of solar corona and other stellar coronae, active galactic nuclei, {\\small X}-ray binaries, supernovae, and recently in solar-type star Capella from {\\small X}-ray satellites Chandra and XMM-Newton \\cite{sa99}. These 3 lines correspond to (see Fig.~1) {\\small X}-ray lines from M-shell 3d-2p with wavelength at $\\sim$15$\\AA$: (1) 3C $\\lambda$~15.013$\\AA$: 1s$^2$2s$^2$2p$^5$[1/2]3d$_{3/2}$~$^1$P$^\\ro_1$ (level 27)\\lr1s$^2$2s$^2$2p$^6$~$^1$S$_0$ (ground state); (2) 3D $\\lambda$~15.265$\\AA$: 1s$^2$2s$^2$2p$^5$[3/2]3d$_{5/2}$~$^3$D$^\\ro_1$ (level 23)\\lr1s$^2$2s$^2$2p$^6$~$^1$S$_0$; (3) 3E $\\lambda$~15.456$\\AA$: 1s$^2$2s$^2$2p$^5$[3/2]3d$_{5/2}$~$^3$P$^\\ro_1$ (level 17)\\lr1s$^2$2s$^2$2p$^6$~$^1$S$_0$. The line intensities display subtle effects since the 3C is a dipole-allowed transition, but the 3D and 3E lines are intercombination transitions that behave as forbidden transitions (spin forbidden) at low-Z neon-like ions, and as allowed (in jj coupling) E1 transitions for high-Z high ions. Comparison of experimental and theoretical data shows disagreement of up to 50\\% for R1=3C/3D, and a factor of two for R2=3E/3C \\cite{br98,la00}. This means that some atomic mechanisms, and their effect on line formation, have not been considered in previous theoretical works.  The outline of the paper is as follows. In Secs.~2 and 3 the basic theoretical  and computational methods and techniques are briefly described. A schematic illustration of many important \\Fe17 lines are shown in Fig.~1, and discussed in Sec.~4. In Secs.~5 and 6 we present the results for cross sections and line intensities, and demonstrate that: (i) dense resonance structure appeared in all transitions over the entire energy range below the highest threshold in the 89CC BPRM collision strengths; in particular, resonance enhancement generally dominates forbidden and intercombination transitions (but has not heretofore been studied), (ii) the theoretical line intensity ratios for intercombination transitions, over the strongest dipole-allowed transition, agree with two sets of recent EBIT measurements \\cite{br98,la00} to 10\\% or within experimental uncertainties. The conclusions are summarised in Sec.~7. ",
        "conclusions": "The principal features of the present work are as follows.  (I) The hitherto most detailed sets of 37CC and 89CC BPRM calculations show that the \\n = 4 complex explicitly included in the latter calculation has a considerable effect on the collision strengths for \\Fe 17 transitions. In particular, prominent resonances appear in the 89CC calculations at energies above and below the \\n = 3 thresholds, and the effective collision strengths are considerably enhanced relative to the smaller 37CC calculation. The background collision strengths for some transitions are also affected due to inter-channel coupling and consequent re-distribution of flux among the larger number of channels in the 89CC case.  (II) New calculations of atomic structure and transition probabilities for all 89 levels have also been carried out, including E1,E2,E3 and M1 and M2 multipole transitions. Some results are presented, although the primary focus is on electron excitation. Owing to the complexity of Fe~{\\small XVII} relativistic and correlation effects are equally important. We note that neither pure LS-coupling nor pure jj-coupling is appropriate for some transitions. It is found that the M2 and E3 transition probabilities are sufficiently large and may have non-negligible effect on the intensities of some important Fe~{\\small XVII} lines.  (III) The calculated effective collision strengths are benchmarked against EBIT experimental data, and show very good agreement to $\\sim$10\\%, or within experimental uncertainties. Resonance enhancements in the intercombination lines 3D and 3E is much larger than for the dipole-allowed line 3C, and is crucial to spectral formation of these important lines. The strong dependence of line ratios 3C/3D and 3E/3C on electron beam energy is explained by the present results.  \\ack  This work was partially supported by the National Science Foundation and the NASA Astrophysical Theory Program. The computational work was carried out at the Ohio Supercomputer Center, Ohio."
    },
    "0208/astro-ph0208502_arXiv.txt": {
        "abstract": "Echelle spectrophotometry of the 30~Doradus nebula in the LMC is presented. The data consists of VLT UVES observations in the 3100 to 10350 \\AA\\ range. The intensities of 366 emission lines have been measured, including 269 identified permitted lines of H$^{0}$, He$^{0}$, C$^{0}$, C$^{+}$, N$^{+}$, N$^{++}$, O$^{0}$, O$^{+}$, Ne$^{0}$, Ne$^{+}$, S$^{+}$, S$^{++}$, Si$^{0}$, Si$^{+}$, Si$^{++}$, Ar$^{+}$, and Mg$^{+}$; many of them are produced by recombination only while others mainly by fluorescence.  Electron temperatures and densities have been determined using different line intensity ratios.  The He$^+$, C$^{++}$, O$^+$, and O$^{++}$ ionic abundances have been derived from recombination lines, these abundances are almost independent of the temperature structure of the nebula. Alternatively abundances from collisionally excited lines have been derived for a large number of ions of different elements, these abundances depend strongly on the temperature structure. Accurate $t^2$ values have been derived from the Balmer continuum, and by comparing the C$^{++}$, O$^+$, and O$^{++}$ ionic abundances obtained from collisionally excited and recombination lines. The chemical composition of 30~Doradus is compared with those of Galactic and extragalactic \\ion{H}{2} regions. The values of $\\Delta Y$/$\\Delta O$, $\\Delta Y$/$\\Delta Z$, and $Y_p$ are also discussed. \\end {abstract} ",
        "introduction": "\\footnotetext{Based on observations collected at the European Southern Observatory, Chile, proposal number ESO 68.C-0149(A).}  The determination of the chemical composition of \\ion{H}{2} regions has been paramount for the study of the chemical evolution of galaxies and for the determination of the primordial helium abundance, $Y_p$. In recent times the determination of atomic data of higher accuracy and the detection of fainter emission lines with the use of echelle spectrophotometry have permitted to derive more accurate physical conditions for Galactic \\ion{H}{2} regions \\citep[e.g.,][] {est98,est99a,est99b}. For these reasons it was decided to carry out echelle spectrophotometry of 30~Doradus.  Due to its proximity, its high angular dimensions, and its high surface brightness 30~Doradus, NGC~2070, is the most spectacular extragalactic \\ion{H}{2} region and thus it has been the subject of many spectrophotometric studies \\citep*[e.g.,][]{pei74,all74,duf75,pag78,boe80,duf82,sha83,mat85,ros87,ver02,tsa02}.  The main aim of this paper is to make a new determination of the chemical abundances of 30~Doradus including the following improvements over previous determinations: the consideration of the temperature structure that affects the helium and heavy elements abundance determinations, the derivation of the O and C abundances from recombination line intensities of very high accuracy, the consideration of the collisional excitation of the triplet \\ion{He}{1} lines from the $2^3$S level by determining the electron density from many line intensity ratios, and the study the $2^3$S level optical depth effects on the intensity of the triplet lines by observing a large number of singlet and triplet lines of \\ion{He}{1}.  In sections 2 and 3 the observations and the reduction procedure are described. In section 4 temperatures and densities are derived from eight and six different methods respectively; also in this section, four independent values of the mean square temperature fluctuation, $t^2$, are determined by combining the electron temperatures.  In section 5 ionic abundances are determined based on recombination lines that are almost independent of the temperature structure, and ionic abundances based on ratios of collisionally excited lines to recombination lines that do depend on the temperature structure of the nebula.  In section 6 the total abundances are determined and compared with those of NGC~346 (the most luminous \\ion{H}{2} region in the SMC), the Orion nebula, M17, and the Sun.  In section 7 $\\Delta Y/\\Delta O$ and $\\Delta Y/\\Delta Z$ are determined; these ratios are important restrictions for the study of the chemical evolution of galaxies and for the determination of the primordial helium abundance, $Y_p$.  Also in section 7 $Y_p$ is determined based on the abundances of 30 Dor and a value of $\\Delta Y/\\Delta O$ derived from the observations of other objects and from chemical evolution models of irregular galaxies. ",
        "conclusions": "\\subsection{ The \\ion{H}{2} regions and the solar abundances}  Table~\\ref{tta} presents the abundances of four very well observed \\ion{H}{2} regions: Orion and M17 in the Galaxy, NGC~346 in the SMC, and 30~Doradus in the LMC. The Orion, M17, and 30~Doradus values include additions to the C and O gaseous values of 0.1 dex and 0.08 dex respectively to take into account the fraction of these atoms embedded in dust. For the same reason the NGC~346 C and O values include additions of 0.05 dex and 0.04 dex respectively \\citep{rel02}. The \\ion{H}{2} region abundances have been obtained adopting values of $t^2$ larger than 0.00. Further arguments in favor of $t^2 > 0.00$ have been presented elsewhere \\citep{pei02y,pei02a,pei02b}.  In addition Table~\\ref{tta} presents also the solar photospheric values for C, N, O, Ne, and Ar, and the solar abundances derived from meteoritic data for S, Cl, and Fe. For the solar initial helium abundance the $Y_0$ by \\citet{chr98} was adopted, and not the photospheric one because, apparently, it has been affected by settling.  The \\ion{H}{2} region values for Ne/O, S/O, and Ar/O are in excellent agreement with the solar values which implies that the production of these elements is primary and due to massive stars, and that the assumptions involved in the two types of abundance determinations are sound.  The \\ion{H}{2} regions S/O and Cl/O abundances are in better agreement with the solar meteoritic abundances than with the photospheric ones, the solar photospheric values are S/O = -1.38 $\\pm 0.11$ dex, and Cl/H = -3.21 $ \\pm 0.3$ dex.  The 30~Doradus C/O value is intermediate between that of NGC~346 and those of Orion, M17 and the Sun. The differences are significant and imply that even if C is of primary origin part of it is due to intermediate mass stars and part is due to massive stars, and that the C yield increases with the O/H ratio \\citep{gar95,gar99,car02}.  The 30~Doradus N/O value is intermediate between that of NGC~346 and those of Orion, M17 and the Sun. The differences are significant and imply that part of the N is of primary origin and part of secondary origin \\citep[see][and references therein]{hen00}.  The accuracies of the He/H abundances of 30~Doradus, NGC~346, and M17 are higher than that of the Orion nebula because the first three objects have $ICFs$(He) = 1.00, while the $ICF$(He) for the Orion nebula is larger than 1.00 and it is not well determined. Similarly the values for 30~Doradus, NGC~346, and M17 are more accurate than the solar one because they are based on direct determinations, while the solar value is obtained from models that depend on the helium abundance in a more complex way.  \\subsection{The $\\Delta Y$/$\\Delta O$ and $\\Delta Y$/$\\Delta Z$ ratios and chemical evolution}  The determination of the $\\Delta Y$/$\\Delta O$ and $\\Delta Y$/$\\Delta Z$ ratios is crucial for the determination of $Y_p$, and for constraining the models of galactic chemical evolution. The abundance determinations of 30~Doradus are based on emission line intensities of high quality, take into account the temperature structure of the nebula, and include a very accurate He/H value because the degree of ionization of 30~Doradus is relatively high implying an $ICF$(He) very close to unity.  To determine the hydrogen, helium, heavy elements, and oxygen abundance by mass, $X$, $Y$, $Z$, and $O$, presented in Table~\\ref{tta} I proceeded as follows: for the Sun I adopted the initial helium abundance by mass ($Y_0$) of \\citet{chr98}, the $Z/X$ value derived from the C/H, N/H, O/H, Ne/H, and Ar/H values presented in Table~\\ref{tta}, and $0.56 \\times O/X$ for the ratio of the rest of the heavy elements to hydrogen, value obtained from the meteoritic abundances by \\citet{gre98}; for the \\ion{H}{2} regions I adopted the He/H value determined by the different observers, the C/H, N/H, O/H, Ne/H, and Ar/H values presented in Table~\\ref{tta}, and $0.56 \\times O/X$ for the rest of the heavy elements.  {From} Table~\\ref{tdy/dz} it can be seen that the spread among the $\\Delta Y$/$\\Delta O$ and $\\Delta Y$/$\\Delta Z$ values derived from the \\ion{H}{2} regions is smaller for the $t^2 > 0.00$ results than for the $t^2 = 0.00$ results, which is consistent with the idea that the $t^2 > 0.00$ values are better. Moreover the theoretical computations for the chemical evolution of irregular galaxies and for the solar vicinity predict values of $\\Delta Y$/$\\Delta O$ in the 2.9 to 4.6 range with a representative value of $\\Delta Y$/$\\Delta O$ = 3.5 $\\pm 0.9$ \\citep*[see][]{car95,car99,car00,chi97,pei00}, again in better agreement with the results from \\ion{H}{2} regions under the assumption that $t^2 > 0.00$.  \\subsection{The primordial helium abundance}  To determine the $Y_p$ value from 30 Doradus it is necessary to estimate the fraction of helium present in the interstellar medium produced by galactic chemical evolution. For this purpose it was assumed that: \\begin{equation} \\label{DeltaO} Y_p  =  Y({\\rm 30~Dor}) - O({\\rm 30~Dor}) \\frac{\\Delta Y}{\\Delta O}. \\end{equation} As in section 7.2 a $\\Delta Y$/$\\Delta O$ = 3.5 $\\pm 0.9$ will be adopted, which together with the $Y$(30~Dor) and $Z$(30~Dor) values of Table~\\ref{tta} yield $Y_p$ = 0.2345 $\\pm$ 0.0047, where most of the error comes from the adopted $\\Delta Y$/$\\Delta O$ value. This $Y_p$ value is in excellent agreement with the value derived from NGC 346 \\citep*{pei00}.  This agreement is due to the similarity between the adopted $\\Delta Y$/$\\Delta O$ value and that derived from 30 Doradus and NGC 346."
    },
    "0208/astro-ph0208028_arXiv.txt": {
        "abstract": "We propose a new approach for studying the neutron star/supernova remnant associations, based on the idea that the supernova remnants (SNRs) can be products of an off-centered supernova (SN) explosion in a preexisting bubble created by the wind of a {\\it moving} massive star. A cavity SN explosion of a moving star results in a considerable offset of the neutron star (NS) birth-place from the geometrical center of the SNR. Therefore: a) the high transverse velocities inferred for a number of NSs through their association with SNRs can be reduced; b) the proper motion vector of a NS should not necessarily point away from the geometrical center of the associated SNR. Taking into account these two facts allow us to enlarge the circle of possible NS/SNR associations, and could significantly affect the results of previous studies of associations. The possibilities of our approach are illustrated with some examples. We also show that the concept of an off-centered cavity SN explosion could be used to explain the peculiar structures of a number of SNRs and for searches for stellar remnants possibly associated with them. ",
        "introduction": "Usually the evaluation of reliability of claimed NS/SNR associations is based on the use of five criteria formulated by \\cite*{vgvaramadze-B2:kas96}, which come to the following questions:\\\\ \\hskip 5mm -- do independent distance estimates agree?\\\\ \\hskip 5mm -- do independent age estimates agree?\\\\ \\hskip 5mm -- is the implied transverse velocity reasonable?\\\\ \\hskip 5mm -- is there evidense for any interaction between the NS and SNR?\\\\ \\hskip 5mm -- does the proper motion vector of the NS point away from the SNR center?\\\\ The last question is considered the most important one since ``a proper motion measurement has the potential to disprove an association regardless of the answers to the other questions\" (\\cite{vgvaramadze-B2:kas96}).  Sometimes a claimed NS/SNR association is considered as false on the basis of statistical studies of associations (e.g. \\cite{vgvaramadze-B2:gae95}, \\cite{vgvaramadze-B2:lor98}). For example, one of the arguments against the association of PSR \\object{B\\,1706-44} with the SNR \\object{G\\,343.1-2.3} (\\cite{vgvaramadze-B2:nic96}) is based on the suggestion by \\cite*{vgvaramadze-B2:gae95} that young ($< 25\\,000$ yr) NSs cannot overrun their parent SNR shells.  However, these approaches neglect two very important effects: the modification of the ambient medium by the ionizing emission and stellar wind of massive stars (the progenitors of most of SNe), and the proper motion of SN progenitor stars. The first effect is important since it is the subsequent interaction of SN blast waves with their processed ambient medium (a system of cavities and shells) that results in the observed SNRs: their structure and evolution are already known to deviate significantly from those derived from standard models of SNRs based on the Sedov-Taylor solution (e.g. \\cite{vgvaramadze-B2:shu85}; \\cite{vgvaramadze-B2:cio89}; \\cite{vgvaramadze-B2:che89}; \\cite{vgvaramadze-B2:fra91}). The stellar proper motion should be considered since it could result (\\cite{vgvaramadze-B2:gva02} and references therein) in a considerable offset of the SN explosion site from the center of the wind-driven bubble (i.e. from the geometrical center of the future SNR).  Taking into account these two effects could significantly affect the results of previous studies of NS/SNR associations, and allow us to enlarge the circle of possible NS/SNR associations and to search for new associations. ",
        "conclusions": ""
    },
    "0208/astro-ph0208578_arXiv.txt": {
        "abstract": "{ We have assembled a UV-flux selected sample of 82 early-type galaxies and collected additional information at other wavelengths. These data confirm a large spread of the $UV-V$ color in the range 2 to 5. The spread in $UV-V$ is accompanied by a spread in $B-V$ that is mainly attributed to the range of morphological types and luminosities. A large fraction of the objects have red colors, $UV-V = 4 \\pm 0.4$, corresponding to a weak UV-upturn as observed with IUE. If the current interpretation for the UV emission from early-type galaxies is applicable to our sample, the PAGB (Post-Asymptotic Giant Branch) tracks are the most common evolution path for the low-mass stars responsible for the UV emission. A small number of very blue ($UV-V < 1.4$) objects have been found that can be reasonably interpreted as harbouring some low level of star formation. In contrast with a previous sample based on IUE observations, no correlation is found between the $UV-V$ color and the Mg$_2$ spectral line index; possible explanations are reviewed. The potential of a more extended UV survey like GALEX is briefly presented. ",
        "introduction": "The UV emission discovered in early-type galaxies as early as 1969 by the {\\it Orbiting Astronomical Observatory-2} (Code et al. \\cite{cod0}) is now currently interpreted in terms of low-mass, helium-burning  stars in extreme horizontal branch and subsequent stages of evolution.  O'Connell (\\cite{oco}) has extensively reviewed the built-up of that interpretation thanks to the combination of high quality UV data and new generations of theoretical models for advanced stellar evolution (e.g. Greggio \\& Renzini \\cite{gre1}, \\cite{gre2} and references therein). The former include spectra with the Hopkins Ultraviolet Telescope (HUT) (e.g. Ferguson \\& Davidsen \\cite{fer}, Brown et al. \\cite{bro0}) and high angular resolution images with HST (e.g. Brown et al. \\cite{bro2}, \\cite{bro3}). Notorious difficulties for converging on a well accepted interpretation were the large variety of advanced stages of stellar evolution and the sensitivity of UV production to small changes of physical properties.  Because the most detailed observations are both time consuming and intrinsically difficult, the current interpretation of the far-UV radiation from early-type galaxies relies on a small number of objects. In addition to the bulge of M31, only 7 elliptical galaxies were spectroscopically observed with HUT; it is not yet possible to resolve UV-bright stars down to the horizontal branch beyond M31 (bulge) and its companions (Brown et al. \\cite{bro3}). Studying a large sample of early-type galaxies would therefore require the use of cruder approaches, such as broad-band and integrated UV fluxes, but would still be of interest. It would help to understand the generality of the conclusions reached and to distinguish the possibility and frequency of low level of star formation in the population of early-type galaxies. Although this latter phenomenon is now excluded as a general explanation for the UV emission, it may well be present in a number of objects and have implications on galaxy evolution.  A sample of 32 early-type galaxies was already studied by Burstein et al. \\cite{bur} (hereafter BBBFL) and, albeit observed spectroscopically with IUE, was mostly discussed in terms of their ($1550-V$) color (see also Dorman et al. \\cite{dor}). With the availability of several UV imaging surveys performed in the IUE-era (Brosch \\cite{bros2}, O'Connell \\cite{oco}), it is now possible to study a larger sample of early-type galaxies in the far-UV.  Such a sample would have the advantage to be essentially UV-flux selected and to potentially reveal UV emission from unexpected early-type objects. This is a significant difference with the BBBFL sample made of objects with substantial record in the refereed literature and selected for one-by-one spectroscopic investigation. An additional motivation of our approach is to prepare ourselves to the extended UV survey of GALEX (Martin et al. \\cite{mar}) and what should be learnt of the early-type galaxies.  The paper is organized as follows. Section 2 describes how our sample of UV selected early-type galaxies has been built and is complemented by a wealth of data at other wavelengths. The $UV-V$ color distribution and color-color diagram are presented in sections 3 and 4. The analysis follows in section 5. We first take advantage of the fact that the BBBFL sample contains most of the objects that have been studied in details to emphasize a possible relationship between the UV color and the categories of stars responsible for the UV radiation. We then discuss the relative frequencies of these categories of stars in the population of early-type galaxies, the possible cases of recent star formation, the role of global properties such as the luminosity, the relation with the Mg$_2$ spectral line index and the UV light profiles in a few objects.        \\section {The sample}  \\subsection{Data origin}  The sample analysed in this work is composed of all the optically selected early-type galaxies (type $\\leq$ S0a) belonging to the Zwicky catalogue (CGCG, Zwicky et al. \\cite{zwi}) ($m_{pg}$ $\\leq$ 15.7) detected in the UV by the FOCA experiment during the observations of the Coma, A1367 and Cancer clusters (Donas et al. \\cite{don2}, \\cite{don3}, private communication). To these, we add all the early-type galaxies belonging to the Virgo Cluster Catalogue (VCC; Binggeli et al. \\cite{bin}) ($m_{pg}$ $\\leq$ 18) detected by SCAP, FOCA and FAUST in the direction of Virgo (Donas et al. \\cite{don1}, private communication; Deharveng et al. \\cite{deh}). The sample is thus composed primarily of cluster galaxies, even though some background or foreground objects are also included. Galaxies whose UV detection is doubtful because of confusion with nearby objects, (such as VCC 311), unless specified, have been systematically excluded. The sample, largely dominated by objects observed with FOCA (85\\%), is complete to a UV magnitude of about 18 whereas only 7 objects come from the less deep images of SCAP and FAUST. Two additional early-type galaxies identified by Brosch et al. (\\cite{bros1}) in their detailed study of FAUST images in the direction of the Virgo cluster have not been included. In order to preserve homogeneity and UV-flux selection, the sample was not extended with other sources of UV data (UIT archives, O'Connell et al. \\cite{oco0}; Maoz et al. \\cite{mao}; Rifatto et al. \\cite{rif} and references therein).   The final combined sample comprises 82 early-type galaxies, including a few dwarf ellipticals and spheroidals. The accuracy of the morphological classification is excellent for the Virgo galaxies (Binggeli et al. \\cite{bin}). Because of the higher distance, the morphology of galaxies belonging to the other surveyed regions suffers from an uncertainty of about 1.5 Hubble type bins.\\\\  \\subsection {UV data and precision}  Most of the UV data are total integrated magnitudes obtained at 2000 \\AA~ with the FOCA experiment described by Milliard et al. (\\cite{mil}). The FOCA UV magnitudes from Donas et al. (\\cite{don2}, \\cite{don3}) have been reprocessed adopting a new zero-point calibration and a revised version of the data reduction pipeline (Donas et al., private communication). A comparison of the FOCA magnitudes with IUE data (stars and galaxies) has revealed large fluctuations from object to object, with the FOCA fluxes being on average 0.3 mag brighter. Because of this dispersion and various possible explanations on a case by case basis, we decided to stay on the FOCA calibration in order to be consistent with previous works. The comparison with IUE will again be addressed in the specific context of the colors of the galaxies in section 3. The  UV magnitudes at 1650 \\AA~ of the additional galaxies (6) from FAUST have been transformed to 2000 \\AA ~using the relation $UV$(2000 \\AA) = $UV$(1650 \\AA) $+$ 0.2. This relation is intended to account for the average spectral trend of ellipticals between 1650 \\AA~ and 2000 \\AA~ as well as the comparison of FAUST magnitudes with other UV measurements (Deharveng et al. \\cite{deh}). The estimated error on the (FOCA) UV magnitude due to the flux extraction procedure and to the linearisation of the photographic plates is 0.3 mag in general, but it ranges from 0.2 mag for bright galaxies to 0.5 mag for weak sources. This, combined with the previously discussed uncertainty on the zero point, gives errors on the UV magnitudes of $\\sim$ 0.5 mag. This uncertainty should be reminded when discussing color trends in our sample; it is extremely large in comparison with the current range of optical colors (as $B-V$) but should be seen in the context of the much larger range of variation of the UV color. A comparison of 4 galaxies measured with both FOCA and FAUST (for homogeneity only the FOCA data have been retained in the sample) shows the FOCA fluxes 0.55 mag fainter than FAUST fluxes on average. This number suggests a possible systematic effect but remains consistent with our evaluation of the uncertainty of UV magnitudes.  \\subsection{Complementary data}  Optical data, available for 63 objects in the V, 72 in the B and 51 in the U band are from Gavazzi \\& Boselli (\\cite{gav1}) and Boselli et al. (private communication). NIR data, from Nicmos3 observations, are taken from Boselli et al. (\\cite{bos1}) and Gavazzi et al. (\\cite{gav5}, \\cite{gav7}) (74 galaxies). From these data we derive total (extrapolated to infinity) magnitudes $H_T$ determined as described in Gavazzi et al. (\\cite{gav6}) with typical uncertainties of $\\sim$ 10 \\%.  For a few objects we derive the H luminosity from K band measurements assuming an average $H-K$ color of 0.25 mag (independent of type; see Gavazzi et al. \\cite{gav6}) when the true $H-K$ color is not available. The estimated error on the optical and near-IR magnitudes is 0.1 mag.  The multifrequency data used in this work are listed in Table 1, arranged as follow:  \\begin{itemize}  \\item{Column 1: VCC designation, from Binggeli et al. (\\cite{bin}) for Virgo galaxies, or CGCG (Zwicky et al. \\cite{zwi}) for A1367, Coma and Cancer cluster galaxies.} \\item{Column 2: UGC name.} \\item{Column 3: NGC/IC name.} \\item{Column 4: morphological type as given in the VCC for Virgo galaxies or in Gavazzi \\& Boselli (\\cite{gav1}) for the other objects.} \\item{Column 5: photographic magnitude from the VCC for Virgo galaxies, from the CGCG for the other objects.} \\item{Columns 6 and 7: major and minor optical diameters. For VCC galaxies the diameters are measured on the du Pont plates at the faintest detectable isophote. For CGCG galaxies these are the major and minor optical diameters ($a_{25}$, $b_{25}$) (in arcmin) derived as explained in Gavazzi \\& Boselli (\\cite{gav1})} \\item{Column 8: distance, in Mpc. Distances to the various substructures of Virgo are as given in Gavazzi et al. (\\cite{gav3}). A distance of 91.3 and 96 Mpc is assumed for galaxies in the A1367 and Coma clusters respectively. For Cancer cluster galaxies and for background and foreground objects the distance is determined from the redshift assuming $H_0$= 75 $\\rm km~s^{-1}Mpc^{-1}$.} \\item{Column 9: cluster membership as defined in Gavazzi et al. (\\cite{gav3}) for Virgo and in Gavazzi et al. (\\cite{gav4}) for A1367 and Coma. P is for pairs, G for groups, BkgV for galaxies in the background of Virgo, ForC for objects in the foreground of Coma.} \\item{Columns 10 to 15: K, H, J, V, B and U magnitudes determined as in Gavazzi \\& Boselli (\\cite{gav1}), corrected for galactic extinction according to Burstein \\& Heiles (\\cite{burstein}). S0a galaxies are corrected for internal extinction as in Gavazzi \\& Boselli (\\cite{gav1}).} \\item{Column 16 and 17: $UV$ (2000 \\AA) magnitude corrected for galactic extinction according to Burstein \\& Heiles (\\cite{burstein}) assuming $A(UV)= 2.1 \\times A(B)$ (all the targets being high galactic latitude objects, $A(UV) \\leq$ 0.3 mag), and reference.} \\item{Column 18: Mg$_2$ data for the nuclear regions from Golev \\& Prugniel (\\cite{gol}) and Jorgensen (\\cite{jor}). The index is defined as in Worthey (\\cite{wor}).} \\item{Column 19: logarithm of the H band luminosity, in solar units, determined from the relation log$L_H$ = 11.36 + 2log$D$ $-$ 0.4$H_T$, where $H_T$ is the total extrapolated $H$ magnitude and $D$ the distance (in Mpc).} \\item{Column 20: the $C_{31}$ index, defined as the ratio of the radii containing 75\\% to 25\\% of the total H-band light of the galaxy.} \\item{Column 21: comments to individual objects} \\end{itemize}              \\addtocounter{table}{0} \\onecolumn \\clearpage \\scriptsize \\begin{landscape} \\begin{longtable}{rcccccccccccccccccccc} \\caption{The sample galaxies: Virgo}\\\\ \\hline \\hline \\noalign{\\bigskip} VCC  &  UGC &  NGC/IC &  type &  m$_{pg}$ &  a &  b &  Dist &  Cluster & Kmag &  Hmag &  Jmag &Vmag &   Bmag & Umag& UVmag & Ref &  Mg$_2$ &  log L$_H$ &  C$_{31}$&Note\\\\ \\noalign{\\smallskip} (1) & (2) & (3) & (4) & (5) & (6) & (7) & (8) & (9) & (10) & (11) & (12) & (13) & (14) & (15) & (16) & (17) & (18) &(19)&(20)&(21)\\\\ \\hline \\noalign{\\smallskip} 49 & 7203&4168&  E&12.21& 1.76& 1.40&   32&    M & 8.32& 8.84& 9.62&11.72&12.63&13.11&12.50& 2& 0.246&   10.89&    3.71&*\\\\ 288 &    -&   -& dE&17.70& 0.52& 0.31&   17&    N &  -  &  -  &   - &  -  &  -  &  -  &16.82& 8&   -  &     -  &     -  &\\\\ 355 & 7365&4262& S0&12.41& 1.87& 1.63&   17&    A & 8.37& 8.59& 9.33&11.62&12.56&13.10&14.93& 8& 0.294&   10.38&    5.72&\\\\ 389 &    -& 781&dS0&14.21& 1.41& 0.91&   17&    A &  -  &  -  &  -  &  -  &  -  &  -  &16.85& 8&   -  &     -  &     -  &\\\\ 608 &    -&4322& dE&14.94& 1.25& 0.62&   17&    A &11.80&11.70&12.62&14.52&15.24&15.46&16.89& 8&   -  &    9.07&    2.66&\\\\ 616 &    -&4325&  E&14.40& 1.55& 0.97&102.8& BkgV &  -  &  -  &  -  &13.47&14.40&14.85&14.06& 1&   -  &     -  &     -  &\\\\ 715 &    -&3274& S0&14.80& 0.93& 0.46& 90.9& BkgV &  -  &  -  &  -  &  -  &  -  &  -  &17.27& 9&   -  &     -  &     -  &\\\\ 731 & 7488&4365&  E&10.51& 8.73& 6.18&   23&    B & 6.50& 6.78& 7.48& 9.66&10.64&11.25&13.75& 1& 0.312&   11.42&    6.00&*\\\\ 759 & 7493&4371& S0&11.80& 5.10& 2.48&   17&    A & 7.77& 8.05& 8.76&10.87&11.85&12.41&15.61& 6&   -  &   10.62&    3.93&\\\\ 763 & 7494&4374&  E&10.26&10.07&10.07&   17&    A & 6.43& 6.69& 7.43& 9.16&10.16&10.76&13.71& 6& 0.287&   11.16&    4.70&*\\\\ 781 & 7500&3303&dS0&14.72& 1.08& 0.50&   17&    A &12.11&  -  &  -  &14.34&15.03&  -  &16.92& 6&   -  &    8.94&    3.04&*\\\\ 828 & 7517&4387&  E&12.84& 1.84& 0.83&   17&    A & 9.04& 9.32&10.03&12.29&13.19&13.74&16.67& 6& 0.228&   10.08&    4.25&\\\\ 870 &    -&3331&dS0&15.52& 1.16& 0.43&185.7& BkgV &  -  &  -  &  -  &  -  &  -  &  -  &16.33& 6&   -  &     -  &     -  &\\\\ 881 & 7532&4406&  E&10.06&11.37& 7.51&   17&    A & 6.04& 6.27& 6.98& 8.95& 9.94&10.51&14.11& 6& 0.290&   11.38&    7.06&*\\\\ 914 &    -&   -& dE&19.00& 0.25& 0.25&   23&    B &  -  &  -  &  -  &  -  &  -  &20.02&17.29& 9&   -  &     -  &     -  &\\\\ 944 & 7542&4417& S0&12.08& 3.60& 1.00&   23&    B & 8.21& 8.42& 9.20&11.13&12.03&12.52&15.58& 9&   -  &   10.89&   12.52&*\\\\ 951 & 7550&3358&dE/dS0&14.35& 1.43& 0.94&17&    B &  -  &11.54&  -  &13.94&14.58&  -  &16.32& 6&   -  &    9.18&    3.25&\\\\ 1003 & 7568&4429&S0a&11.15& 8.12& 3.52&   17&    A & 6.54& 6.74& 7.43& 9.59&10.58&11.22&14.98& 6& 0.232&   11.01&    5.48&*\\\\ 1010 & 7569&4431&dS0&13.68& 1.58& 0.79&   17&    A &10.53&10.74&11.35&13.23&14.05&14.59&16.91& 6& 0.153&    9.57&    2.86&\\\\ 1030 & 7575&4435& S0&11.84& 2.92& 2.48&   17&    A & 7.68& 7.92& 8.66&10.94&11.82&12.33&15.31& 6& 0.189&   10.74&   10.06&*\\\\ 1111 &    -&   -& dE&17.70& 0.33& 0.20&   17&    A &  -  &  -  &  -  &  -  &  -  &  -  &15.73& 6&   -  &     -  &     -  &\\\\ 1125 & 7601&4452& S0&13.30& 2.92& 0.57&   17&    A & 9.04& 9.31&10.10&11.91&12.87&13.41&16.23& 6&   -  &   10.06&    2.67&\\\\ 1146 & 7610&4458&  E&12.93& 1.80& 1.52&   17&    A & 9.38& 9.69&10.32&12.32&13.18&13.64&16.74& 6& 0.208&   10.06&    7.99&\\\\ 1226 & 7629&4472&  E& 9.31&10.25& 8.11&   17&    S & 5.30& 5.59& 6.31& 8.54& 9.52&10.16&13.35& 9& 0.313&   11.66&    7.34&*\\\\ 1250 & 7637&4476& S0&12.91& 1.89& 0.94&   17&    A & 9.50& 9.82&10.47&12.41&13.23&13.55&15.33& 6& 0.141&    9.88&    3.59&\\\\ 1279 & 7645&4478&  E&12.15& 1.89& 1.43&   17&    A & 8.24& 8.52& 9.17&11.45&12.36&12.77&15.47& 6& 0.233&   10.35&    3.37&\\\\ 1297 &    -&   -&  E&14.33& 0.51& 0.45&   17&    A &10.07&10.34&11.13&13.44&14.42&15.03&17.41& 6& 0.290&    9.72&    3.52&\\\\ 1316 & 7654&4486&  E& 9.58&11.00&11.00&   17&    A & 5.92& 6.19& 7.01& 8.82& 9.82&10.37&12.70& 6& 0.270&   11.34&    4.20&*\\\\ 1327 & 7658&   -&  E&13.26& 1.10& 0.88&   17&    A & 9.07& 9.23& 9.75&11.49&12.13&12.27&14.49& 6&   -  &   10.14&    4.43&*\\\\ 1368 & 7665&4497&S0a&13.12& 2.01& 0.85&   17&    A & 9.60& 9.72&  -  &12.18&13.05&13.46&17.22& 6&   -  &    9.85&    2.81&\\\\ 1499 &    -&3492&  E&14.94& 0.64& 0.46&   17&    A &  -  &12.59&  -  &14.77&15.26&  -  &13.79& 1&   -  &    8.79&    2.74&*\\\\ 1535 & 7718&4526& S0&10.61& 7.00& 2.01&   17&    S & 6.37& 6.65& 7.47& 9.83&10.80&11.36&14.04& 1& 0.272&   11.19&   10.59&\\\\ 1809 & 7825&3631&S0a&14.17& 1.10& 0.67& 37.3& BkgV &  -  &  -  &  -  &  -  &  -  &  -  &10.70& 1&   -  &     -  &     -  &\\\\ \\hline \\newpage \\caption{The sample galaxies: A1367, Cancer, Coma}\\\\ \\hline \\noalign{\\bigskip} CGCG  &  UGC &  NGC/IC &  type &  m$_{pg}$ &  a &  b &  Dist &  Cluster & Kmag &  Hmag &  Jmag& Vmag &   Bmag & Umag& UVmag & Ref &  Mg$_2$ &  log L$_H$ &  C$_{31}$&Note\\\\ \\noalign{\\smallskip} (1) & (2) & (3) & (4) & (5) & (6) & (7) & (8) & (9) & (10) & (11) & (12) & (13) & (14) & (15) & (16) & (17) & (18) &(19)&(20)&(21)\\\\ \\hline \\noalign{\\smallskip} 97125 &    -&   -&S0a&15.60& 0.84& 0.59& 91.3&A1367 &  -  &11.82&  -  &14.67&15.55&15.82&16.05& 3&   -  &   10.67&    5.76&*\\\\ 97127 & 6723&3862&  E&14.00& 1.62& 1.58& 91.3&A1367 & 9.75&10.00&10.72&12.90&13.86&14.43&16.32& 5& 0.282&   11.32&    5.88&*\\\\ 97134 & 6731&3867& S0&14.60& 1.31& 0.44& 91.3&A1367 &  -  &10.36&  -  &13.32&14.28&14.86&18.04& 3&   -  &   11.45&    4.02&\\\\ 98078 &    -&   -&  E&15.20& 0.40& 0.30& 91.2& P    &  -  &12.35&  -  &  -  &  -  &  -  &13.25& 1&   -  &   10.38&     -  &*\\\\ 119030 &    -&   -&  E&15.70& 0.66& 0.44& 31.2&Cancer&  -  &13.07&  -  &15.45&16.03&15.98&16.72& 7&   -  &    9.03&    3.65&*\\\\ 119053 &    -&   -&S0a&15.50& 0.63& 0.50& 66.4&Cancer&12.17&12.47&12.77&14.80&15.30&15.28&14.87& 7&   -  &   10.12&    8.67&*\\\\ 119065 & 4347&2563&  E&13.70& 2.60& 2.21& 66.4&Cancer&  -  & 9.46&  -  &12.13&13.04&13.63&16.01& 7& 0.313&   10.98&    9.44&\\\\ 119086 &    -&   -&  E&15.70& 0.53& 0.40& 89.2&Cancer&  -  &13.47&  -  &15.39&15.77&  -  &15.07& 7&   -  &    9.84&    2.86&\\\\ 127032 & 6663&3821& S0&13.80& 1.77& 1.51& 91.3&A1367 &  -  &10.56&  -  &12.66&13.32&13.59&15.03& 3&   -  &   11.11&    5.45&*\\\\ 127045 & 6725&   -&S0a&14.50& 1.50& 1.20& 91.3&A1367 &  -  &11.15&  -  &  -  &  -  &  -  &16.18& 5&   -  &   10.94&    4.35&*\\\\ 127048 &    -&   -&  E&15.00& 0.50& 0.50& 93.4& G    &  -  &11.20&  -  &  -  &  -  &  -  &16.49& 5&   -  &   10.91&    7.05&\\\\ 160014 &    -&   -&  E&15.70& 0.80& 0.54&   96& Coma &  -  &11.79&  -  &  -  &15.38&  -  &16.56& 5&   -  &   10.58&    3.29&\\\\ 160021 & 8057&4816& S0&14.80& 2.04& 1.41&   96& Coma & 9.89&10.06&10.87&12.77&13.74&14.18&16.89& 4& 0.304&   11.32&    8.21&\\\\ 160028 & 8065&4827& S0&14.10& 1.50& 1.16&   96& Coma &10.08&10.37&  -  &13.16&14.10&  -  &16.92& 5& 0.327&   11.27&    8.71&\\\\ 160038 & 8069&   -&S0a&14.80& 1.18& 0.53&   96& Coma &  -  &10.92&  -  &13.57&14.35&15.03&17.56& 5&   -  &   10.93&    5.84&\\\\ 160039 & 8070&4839&  E&13.60& 3.56& 1.55&   96& Coma & 9.60& 9.85&10.58&12.15&13.15&13.71&16.27& 5& 0.312&   11.53&   10.20&*\\\\ 160042 &    -&4840&  E&14.80& 0.95& 0.85&   96& Coma &10.56&10.84&11.58&13.78&14.73&15.32&17.45& 4& 0.323&   11.08&    8.73&\\\\ 160044 & 8072&4841&  E&13.50& 1.59& 1.55&   96& Coma & 9.74& 9.90&10.70&12.75&13.72&  -  &16.82& 4& 0.317&   11.45&    8.72&*\\\\ 160059 &    -&   -&  E&15.20& 1.31& 0.25&   96& Coma &  -  &11.59&  -  &  -  &15.16&  -  &17.85& 5&   -  &   10.74&    4.51&\\\\ 160063 &    -&4850& S0&15.30& 0.79& 0.61&   96& Coma &11.24&11.54&12.34&14.29&15.22&15.81&16.90& 4& 0.287&   10.75&    3.82&\\\\ 160068 & 8092&4853& S0&14.20& 1.00& 0.78&   96& Coma &10.45&10.91&11.61&13.61&14.30&14.55&15.42& 4& 0.164&   10.98&    3.77&*\\\\ 160077 &    -&3990&S0a&15.00& 1.22& 0.49&   96& Coma &  -  &10.61&  -  &  -  &14.37&  -  &17.48& 4&   -  &   11.05&    7.55&\\\\ 160079 &    -&   -&S0a&15.10& 0.98& 0.38&   96& Coma &10.87&11.18&11.76&  -  &14.56&  -  &17.97& 5& 0.251&   10.83&    5.48&\\\\ 160100 &    -&   -&  E&15.50& 0.72& 0.67&   96& Coma &11.51&11.84&12.55&14.73&15.65&  -  &17.87& 4& 0.279&   10.62&    4.89&\\\\ 160101 &    -&   -&S0a&15.20& 1.03& 0.35&   96& Coma &10.95&11.24&11.88&  -  &14.84&  -  &17.90& 4& 0.286&   10.79&    6.15&\\\\ 160103 & 8142&4926&  E&14.10& 1.31& 0.99&   96& Coma & 9.91&10.25&10.93&13.10&14.08&  -  &17.29& 4& 0.315&   11.28&    8.11&\\\\ 160104 &    -&   -&S0a&15.40& 0.67& 0.28&   96& Coma &12.26&12.55&13.12&  -  &15.26&  -  &16.49& 4&   -  &   10.28&    3.95&*\\\\ 160105 &    -&4927& S0&14.80& 1.07& 0.74&   96& Coma &10.35&10.69&11.50&13.72&14.73&  -  &17.34& 4& 0.348&   11.14&    8.42&\\\\ 160107 &    -&   -&S0a&14.90& 1.06& 0.32&   96& Coma &  -  &10.66&  -  &13.50&14.51&15.78&17.39& 5&   -  &   10.99&    3.66&\\\\ 160109 &    -&   -& S0&15.50& 0.67& 0.54&   96& Coma &11.29&11.48&12.36&  -  &15.45&  -  &17.42& 4&   -  &   10.75&    4.53&\\\\ 160118 & 8154&4931& S0&14.40& 1.76& 0.65&   96& Coma &10.22&10.45&11.15&13.35&14.22&  -  &16.95& 4&   -  &   11.22&    8.81&\\\\ 160120 & 8160&4934& S0&15.00& 1.33& 0.36&   96& Coma &  -  &11.58&  -  &14.36&15.11&15.55&17.40& 4&   -  &   10.76&    3.89&\\\\ 160122 &    -&   -& S0&15.60& 0.71& 0.53&   96& Coma &10.39&11.69&  -  &  -  &15.50&  -  &17.45& 4&   -  &   10.62&    4.05&\\\\ 160124 & 8167&4944& S0&13.30& 2.32& 0.84&   96& Coma &10.15&10.31&11.05&12.87&13.78&14.12&16.53& 4&   -  &   11.22&    5.76&\\\\ 160125 &    -&   -& S0&15.40& 0.89& 0.69&   96& Coma &  -  &11.52&  -  &  -  &15.46&  -  &17.75& 5&   -  &   10.73&    7.31&\\\\ 160129 & 8175&4952&  E&13.60& 1.74& 1.18& 78.2& ForC &  -  &10.08&  -  &12.94&13.79&  -  &16.68& 5& 0.290&   11.23&    8.46&\\\\ 160140 &    -&4971& S0&15.00& 0.99& 0.86& 85.3& ForC &10.80&10.96&  -  &14.01&14.94&  -  &17.32& 5& 0.284&   10.94&    9.25&\\\\ 160211 &    -&3947& S0&15.60& 0.58& 0.51&   96& Coma &11.86&12.03&12.79&14.91&15.77&  -  &17.38& 4& 0.282&   10.56&    4.14&\\\\ 160222 &    -&4867&  E&15.50& 0.64& 0.43&   96& Coma &11.29&11.68&12.33&14.53&15.45&15.92&16.73& 4& 0.294&   10.69&    3.92&\\\\ 160228 &    -&3973& S0&15.20& 0.89& 0.50&   96& Coma &11.02&11.34&12.17&14.30&15.28&15.81&18.12& 5& 0.318&   10.80&    7.40&\\\\ 160229 &    -&4873& S0&15.40& 0.57& 0.49&   96& Coma &11.48&11.62&12.53&14.44&15.41&15.94&18.13& 5& 0.287&   10.77&    5.11&\\\\ 160231 & 8103&4874&  E&13.70& 2.27& 1.93&   96& Coma & 8.91& 9.22&10.02&12.02&12.97&13.55&16.50& 4& 0.310&   11.81&    5.19&*\\\\ 160234 &    -&4876&  E&15.10& 0.61& 0.52&   96& Coma &11.30&11.57&12.37&14.52&15.47&16.03&18.36& 5& 0.260&   10.71&    4.11&\\\\ 160241 & 8110&4889&  E&13.00& 3.30& 2.23&   96& Coma & 8.29& 8.67& 9.41&11.48&12.44&13.06&15.58& 4& 0.342&   11.68&    4.35&*\\\\ 160248 &    -&4898&  E&14.70& 0.85& 0.62&   96& Coma &10.82&11.03&11.75&13.68&14.63&15.17&17.27& 5& 0.283&   10.89&    8.97&\\\\ 160249 & 8113&4895& S0&14.30& 2.00& 0.66&   96& Coma &10.27&10.53&11.31&13.19&14.12&14.64&16.74& 4& 0.301&   11.28&   12.39&*\\\\ 160256 &    -&4045&  E&15.10& 0.84& 0.64&   96& Coma &10.98&11.27&11.94&14.01&14.99&15.48&17.97& 5& 0.299&   10.82&    3.60&\\\\ 160258 &    -&4908&  E&14.90& 0.92& 0.68&   96& Coma &10.60&10.93&11.61&13.83&14.79&15.27&17.71& 5& 0.282&   11.05&    8.20&\\\\ 160259 & 8129&4051&  E&14.80& 1.49& 0.98&   96& Coma &10.36&10.78&11.51&13.46&14.40&14.96&17.10& 4& 0.331&   11.07&    4.48&\\\\  \\noalign{\\smallskip} \\hline \\end{longtable} \\end{landscape} \\footnotesize{ Comments to individual objects:  Virgo:  49: low-luminosity dwarf Seyfert nucleus (Ho et al. \\cite{ho})  731: $H\\alpha+[NII]$E.W.=2 \\AA~ (Kennicutt \\& Kent \\cite{ken0})  763: Low Excitation Radio Galaxy (NED); $H\\alpha+[NII]$E.W.=1 \\AA~ (Trinchieri \\& Di Serego Alighieri \\cite{tri}); also measured with FAUST (Deharveng et al. \\cite{deh})  781: interacting with NGC 4388? (Corbin et al. \\cite{cor})  881: M86; $H\\alpha+[NII]$E.W.=13 \\AA~ (Trinchieri \\& Di Serego Alighieri \\cite{tri}); also measured with FAUST (Deharveng et al. \\cite{deh})  944: classified `` SB0: sp'' in NED  1003: $H\\alpha+[NII]$E.W.=5 \\AA~ (Boselli \\& Gavazzi \\cite{bos2})  1030: interacting with NGC 4438; LINER  1226: M49; $H\\alpha+[NII]$E.W.=1 \\AA~ (Kennicutt \\& Kent \\cite{ken0}); also measured with FAUST (Deharveng et al. \\cite{deh})  1316: M87; radio galaxy; $H\\alpha+[NII]$E.W.=2 \\AA~ (Boselli \\& Gavazzi \\cite{bos2}); also measured with FAUST (Deharveng et al. \\cite{deh})  1327: bright star superposed (Prugniel et al. \\cite{pru})  1499: unresolved in UV from the nearby companion VCC 1491; given the extreme red color of VCC 1491 and the blue color of VCC 1499 (Gavazzi et al. \\cite{gav7}), the UV source should be identified to VCC 1499 (instead of VCC 1491 in Deharveng et al. \\cite{deh}). VCC 1499 is as blue as a dwarf irregular and has a spectrum with the characteristics of a post-starburst galaxy (PSB) (Gavazzi et al. \\cite{gav7}).  Coma/A1367/Cancer:  97125: $H\\alpha+[NII]$E.W.=26 \\AA~ (Moss et al. \\cite{mos}), $H\\alpha+[NII]$E.W.=21 \\AA~ (Gavazzi et al. \\cite{gav2})  97127: Low Excitation Radio Galaxy (NED)  98078: Mrk 758, $H\\alpha+[NII]$E.W.=82 \\AA~ (Gavazzi et al. \\cite{gav2})  119030: classified ``spiral'' in NED  119053: $H\\alpha+[NII]$E.W.=42 \\AA~ (Kennicutt et al. \\cite{ken})  127032: classified as (R)SAB(s)ab in NED  127045: $H\\alpha+[NII]$E.W.=13 \\AA~ (Moss et al. \\cite{mos})  160039: radio galaxy (NED)  160044: binary system  160068: AGN (NED); Balmer absorption lines (Sparke et al. \\cite{spa}); PSB (Caldwell et al. \\cite{cal})  160104: PSB (Caldwell et al. \\cite{cal1})  160231: cD galaxy  160241: cD galaxy  160249: classified as ``SA0 pec sp'' in NED   References to the UV data: 1: FAUST data (Deharveng et al. \\cite{deh}); 2: SCAP data (Donas et al. \\cite{don1}); 3: Donas et al. \\cite{don2} (reprocessed data, $\\Delta$zero-point=$+$0.30 mag, see text); 4: Donas et al. \\cite{don3} (reprocessed data, $\\Delta$zero-point=$-$0.12 mag, see text); 5--9: Donas et al. private communication (5, Coma-A1367), (6, Virgo center), (7, Cancer), (8, Virgo M100), (9, Virgo M49)} \\normalsize \\twocolumn ",
        "conclusions": "We have assembled a sample of 82 early-type galaxies with a flux measurement in the far-ultraviolet. In addition to more than doubling the number of objects, this sample has the advantage to be essentially UV-flux selected. The following has emerged from the analysis.  1) The large scatter of the $UV-V$ color in comparison with the colors in the optical is confirmed. As shown with a small number of objects studied previously in much detail, the color spread between 2 and 5 might be explained by changing proportions of stars along the ZAHB and the following post-HB evolutionary tracks.  2) The galaxies with red $UV-V$ ($\\sim 4$) colors (or weak UV upturn) outnumber those with blue $UV-V$ ($\\sim 2$) colors (or strong UV upturn) in our sample. If the current interpretation of the UV-upturn can be extended to our sample, the PAGB tracks would be the most common evolution path among elliptical galaxies. Only a minority of elliptical galaxies would need a fraction of their stars evolving from the blue part of the ZAHB. The GALEX survey should considerably refine this finding, including possible differences between the various categories of early-type galaxies.  3) Few blue objects ($UV-V$ $<$ 1.5)  may harbour some residual star formation as shown by the examples of NGC 205 and NGC 5102. The implication in terms of the cosmic density of the star formation rate in early-type galaxies should await for a more extended survey like GALEX.  4) For a fraction of the objects, the scatter of the $UV-V$ color is accompanied by a scatter in the $B-V$ color. The latter should be caused by the variety of morphological types and luminosities in the sample rather than the evolutionary features explaining the UV emission.  5) The correlation between the $UV-V$ color and the Mg$_2$ spectral index is not found. This is in line with the idea that the UV flux is not driven by the metallicity but by the mass loss along the giant branch that determines the envelope mass on the horizontal branch."
    },
    "0208/astro-ph0208052_arXiv.txt": {
        "abstract": "We present the first circular polarization measurements of circumstellar H$_2$O masers around a sample of late-type stars. These observations are used to obtain the magnetic field strength in the H$_2$O maser region with both an LTE and non-LTE analysis. We find fields from a few hundred milliGauss up to a few Gauss, indicating a solar-type $r^{-2}$ dependence of the magnetic field on the distance. No linear polarization is detected to less than 1\\%. ",
        "introduction": "We have observed 4 late-type stars with the VLBA, the supergiants S~Per, VY~CMa and NML~Cyg, and the Mira variable star U~Her. To get the highest spectral resolution, required for the circular polarization measurements, the data were correlated twice. Once with modest spectral resolution ($0.1$ km/s), to get all 4 polarization combinations (RR, LL, RL and LR), and once with high resolution ($0.027$ km/s), with only RR and LL. The calibration was mainly performed on the modest spectral resolution data and the solutions were copied and applied to the high resolution data. This data set was then used to produce circular polarization and total intensity image cubes.  \\begin{figure} \\resizebox{\\hsize}{!}{\\includegraphics{fig1.eps}} \\caption{(left) Total intensity image of the H$_2$O maser features around VY~CMa. (right) Total power (I) and V-spectra for selected maser features of VY~CMa. The bottom panel shows the best fitting LTE model (dashed), the middle panel shows the best fitting non-LTE model (dotted).} \\end{figure} ",
        "conclusions": ""
    },
    "0208/astro-ph0208264_arXiv.txt": {
        "abstract": "We present an analysis of low-resolution infrared spectra for 25 brown dwarf candidates in the NGC~1333 molecular cloud.  Candidates were chosen on the basis of their association with the high column density cloud core, and near-infrared fluxes and colors. We compare the depths of water vapor absorption bands in our candidate objects with a grid of dwarf, subgiant, and giant standards to determine spectral types which are independent of reddening.  These data are used to derive effective temperatures and bolometric luminosities which, when combined with theoretical tracks and isochrones for pre-main sequence objects, enable us to estimate masses and ages.  Depending on the models considered, a total of 9 to 20 brown dwarfs are identified with a median of age of $<$1 Myr. ",
        "introduction": "The NGC~1333 reflection nebula and its associated dark cloud L~1450 are part of a chain of molecular clouds in the Perseus region (Sargent 1979; Loren 1976).  Analysis of Hipparcos data suggests a distance of 300 pc (de Zeeuw, Hoogerwerf, \\& de Bruijne 1999; Belikov \\etal 2002).  The observations of emission-line stars and Herbig-Haro objects first established NGC~1333 as an active region of star formation (e.g., Herbig 1974). Surveys of the cloud at near-infrared wavelengths have revealed a large population of low mass stars (Strom, Vrba, \\& Strom 1976; Aspin, Sandell, \\& Russell 1994; Lada, Alves, \\& Lada 1996; Wilking \\etal 2002). The young stellar objects are concentrated into a northern and southern cluster, each with about 70 members. Identification of the lowest mass objects in the NGC~1333 cloud, as well as estimates for their ages and masses, requires infrared spectroscopy. Ultimately, the population of very low mass objects will be used to define the shape of the mass function and determine the number and total mass of brown dwarfs in the young cluster. ",
        "conclusions": ""
    },
    "0208/astro-ph0208114_arXiv.txt": {
        "abstract": "The paper is devoted to the methods of determination of the cosmological parameters and distortions of ionization history from recent CMB observations. We show that the more complex models of kinetics of recombination with a few \"missing\" parameters describing the recombination process provide better agreement between the measured and the expected characteristics of the CMB anisotropy. In particular, we consider the external sources of the resonance and ionizing radiation and the model with the strong clustering of the baryonic component. These factors can constrain the estimates of the cosmological parameters usually discussed. We demonstrate also that the measurements of the CMB polarization can improve these estimates and, for the precision expected from the PLANCK mission, allow to discriminate a wide class of models. ",
        "introduction": "Recent observations of the CMB anisotropy such as the BOOMERANG (de Bernardis et al. 2000), the  MAXIMA-1 (Hanany et al. 2000), the  DASI (Halverson et al. 2001), the VSA (Watson et al. 2002), the Archeops (Benoit et al. 2002) and especially the new CBI data (Mason et al. 2002) and  the DASI polarization measurements (Kovac et al. 2002) provide the good base for the progress in understanding the most general properties of the early Universe. In these experiments, major attention is on the statistical properties of the signal and the determination of the power spectrum of the CMB anisotropy and polarization. The obtained cosmological parameters for adiabatic cold dark matter model (CDM) are well consistent with the Big Bang nucleosynthesis predictions, SN Ia observational data and results obtained from investigations of galaxy surveys. Further progress will be achieved with more  sensitive experiments such as the launched MAP and the upcoming PLANCK missions. They will be able to measure the CMB anisotropy and polarization power spectra up to the multipole range $l\\simeq 3-4\\cdot 10^3$ where the lensing effect, Sunyaev-Zel'dovich effect in clusters, secondary anisotropies and some peculiarities of the ionization history of the Universe dominate.  The high precision of the future CMB measurements allows to reveal some distortions with respect to the standard model of the hydrogen recombination process occurred at redshifts $z\\simeq 10^3$. Here we show that polarization power spectrum contains important information about the kinetics of hydrogen recombination and allows to trace the ionization history of the cosmic plasma in wide range of redshifts  $10^3\\geq z\\geq 6-10$.  For the baryon--dominated Universe the classical theory of the hydrogen recombination was developed in Peebles (1968) and Zel'dovich, Kurt and Sunyaev (1968). For the dark matter dominated Universe, it was generalized in Zabotin and Naselsky (1982), Jones and Wyse (1985), Seager, Sasselov and Scott (1999) (see detailed review in White, Scott and Silk, 1994). Many distortions of the standard model of recombination have also been investigated. The delay of recombination due to evaporation of primordial black holes has been discussed in Naselsky (1978) and Naselsky and Polnarev (1987). Avelino et al. (2000), Battye et al. (2001), Landau et al. (2001) have noted that possible time variations of the fundamental constants could also influence the ionization history of the cosmic plasma. Sarkar and Cooper (1983), Scott et al. (1991), Ellis et al. (1992), Adams et al. (1998), Peebles, Seager and Hu (2000), Doroshkevich and Naselsky (2002) discussed distortions of the kinetics of hydrogen recombination caused by decays of hypothetical particles. It is worth noting that for such models the energy injection delays the recombination at $z\\simeq 10^3$ and distorts the ionization history of the Universe down to the period of galaxies formation at $z\\simeq 5-10$.  Recently Naselsky and Novikov (2002) have proposed a model of accelerated recombination with non--uniform spatial distribution of baryonic matter. In this model the recombination proceeds faster within the clumps and slowly in the intercloud medium. In such models at redshifts $z\\simeq 10^3$ the mean ionization fraction decreases faster than that in the standard model with the same cosmological parameters which mimics the acceleration of recombination. Here we show that, for suitable parameters of the clouds, such models can improve the best--fit of the observed power spectrum of the CMB anisotropy.  The non-standard models of hydrogen recombination are characterized by a few additional parameters which must be included in the list of the best-fit cosmological parameter determination using the CMB anisotropy and polarization data. We call these parameters as the ``missing parameters'' of the theory and show how important these parameters may be for the CMB physics.  In this paper we compare the observed CMB anisotropy power spectrum for the reference model of the standard recombination, with models of delayed and accelerated recombination. For the reference model we accept the cosmological parameters from the best-fit of the CBI observational data (Mason et al 2002). We show that the models considered in this paper can provide better fits for the already observed and the future MAP and PLANCK power spectra and can change usual estimates of the cosmological parameters. Moreover, the measurements of the polarization allow us to discriminate between some of such complex models of the Universe. These results can be important for the interpretation of the MAP and upcoming PLANCK data.  The paper is organized as follows. In section 2 we discuss the generic models of distorted recombination. In section 3 we demonstrate distortions of the CMB anisotropy power spectrum due to delay or acceleration of recombination. In section 4 we introduce the models of accelerated and delayed recombination and discuss the redshift variations of the hydrogen ionization fraction in these models. In section 5 we compare the observed CMB anisotropies and expected polarization power spectra with ones for the models under discussion. Section 6 is devoted to discussion of late reionization and of the polarization power spectrum for the delayed recombination models. In Conclusion we make some predictions for the MAP and the upcoming PLANCK experiments. ",
        "conclusions": "The already achieved precision of measurements of the CMB anisotropy and especially expected one for the MAP and PLANCK missions allows to discriminate wide set of cosmological models and to reveal noticeable distortions of the standard model of hydrogen recombination process at redshifts $z\\simeq 10^3$ and of the ionization history. Many physical models can be considered as a basis for such discussions and can be tested and restricted via such measurements.  In this paper we compare available observations of the CMB anisotropy with those expected in several models with different kinetics of recombination. The differences can be caused by possible external sources of resonance and ionized radiation, possible small scale clustering of baryonic component and other factors. We show that the models with more complicate ionization history provide better description of the available observational data. This means also that the accepted estimates of cosmological parameters can be changed when such models are incorporated in the fitting procedure. It is especially crucial for the estimates of baryonic density, $\\Omega_b h^2$, which is closely related to the Big-Bang nucleosynthesis predictions.  In this paper we do not consider the impact of Sunyaev--Zel'dovich (SZ) effect and the weak lensing by the large scale structure elements. However, the SZ effect is frequency dependent that allows to reduce its contribution for multifrequency measurements. As was shown in Schmalzing, Takada and Futamase (2000), the influence of lensing effect can really distort the measured characteristics of the CMB only at larger $l\\geq 10^3$ while the discussed deviations of ionizing history distort these characteristics at all $l$. This means that the measurements of the CMB spectra in wide range of $l$ allows to discriminate between these distortions. However, this problem requires more detailed discussions for various models of both lensing and recombination.  Near the first Doppler peak, the cosmic variance effect restricts an accuracy of estimates of $C^a(l)$ as $\\Delta C^a(l)/C^a(l)\\sim 10\\%$ for all experiments with the sky coverage $f_{sky}\\sim 65\\%$. For the recent balloon-borne and ground--based experiments the achieved accuracy in this range of $l$ is even worse than this limit. With such precision, we cannot discriminate the basic model and models with delayed recombination and another $\\Omega_b h^2$. To illustrate this statement we compare quantitatively the observational data with the power spectra for Model 1, 2\\,\\&\\,3 parameters which are given in Sec. 4.1\\,. Instead of the Model 4 we here present the results for Model 5 which differs from Model 4 by the values of parameters $\\Omega_bh^2=0.03$ and $\\Omega_m=0.256$ which are the same as in the Model 2.  The quality of models in comparison with the recent BOOMERANG, MAXIMA-1, CBI and VSA observational data can be characterized quantitatively by the $\\chi^2$-parameter listed in Table 2 for all models. Here the reference Model 1 provides the best-- fit for the CBI data for the standard ionization history while $\\chi^2_2\\geq \\chi^2_1$ for the Model 2 with larger $\\Omega_bh^2$\\,. However, the Model 3 demonstrates an excellent agreement with the CBI data, $\\chi^2_3=0.5\\chi^2_1$, and $\\chi^2_3\\leq\\chi^2_1$ for other observations. For the Model 5 with $\\Omega_bh^2=0.03$ we get also $\\chi^2_{5}\\leq \\chi^2_1$ for the CBI data and only for the BOOMERANG data $\\chi^2_{5}\\geq \\chi^2_1$.  Obviously,  such extension of the cosmological models increases the number of parameters using to fit observed power spectra of the anisotropy and, in perspective, of the polarization. We show that the influence of these \"missing\" parameters, namely $\\varepsilon_{i}(z)$ and $\\varepsilon_{\\alpha}(z)$ , can improve the fits used and, in particular, to obtain better agreement between the observed and expected positions and amplitudes of peaks for higher multipoles. At the same time, including of the \"missing\" parameters changes the measured values of cosmological parameters usually discussed.  We show also that the expected sensitivity of the MAP and the PLANK missions in respect of the measurements of the polarization will allow to discriminate between main families of such models and, in particular, between models with small and large optical depth at redshifts $10^3\\geq z\\geq$ 20-50. We would like to note that the realistic values of the cosmological parameters can not be obtained from the CMB data without the PLANCK observations of the polarization."
    },
    "0208/astro-ph0208322_arXiv.txt": {
        "abstract": "{ We present the results of a high spectral resolution ($\\lambda / \\Delta \\lambda $ = 49000) study of the circumstellar (CS) gas around the intermediate mass, pre-main sequence star \\object{UX Ori}. The results are based on a set of 10 \\'echelle spectra covering the spectral range 3800 -- 5900 \\AA, monitoring the star on time scales of months, days and hours. A large number of transient blueshifted and redshifted absorption features are detected in the Balmer and in many  metallic lines. A multigaussian fit is applied to determine for each transient absorption the velocity, $v$, dispersion velocity, $\\Delta v$, and the parameter $R$, which provides a measure of the absorption strength of the CS gas. The time evolution of those parameters is presented and discussed. A comparison of intensity ratios among the transient absorptions suggests a solar-like composition of the CS gas. This confirms previous results and excludes a very metal-rich environment as the cause of the transient features in UX Ori. The features can be grouped by their similar velocities into 24 groups, of which 17 are redshifted and 7 blueshifted. An analysis of the velocity of the groups allows us to identify them as signatures of the dynamical evolution of 7 clumps of gas, of which 4 represent accretion events and 3 outflow events. Most of the events decelerate at a rate of tenths of m\\,s$^{-2}$, while 2 events accelerate at approximately the same rate; one event is seen experiencing both an acceleration and a deceleration phase and lasts for a period of few days. This time scale seems to be the typical duration of outflowing and infalling events in UX Ori. The dispersion velocity and the relative aborption strength of the features do not show drastic changes during the lifetime of the events, which suggests they are gaseous blobs preserving their geometrical and physical identity. These data are a very useful tool for constraining and validating theoretical models of the chemical and physical conditions of CS gas around young stars; in particular, we suggest that the simultaneous presence of infalling and outflowing gas should be investigated in the context of detailed magnetospheric accretion models, similar to those proposed for the lower mass T Tauri stars. ",
        "introduction": "The detection of planetesimals is highly relevant for the study of the formation and evolution of planetary systems, since it is nowadays accepted that planets form from CS disks via the formation of planetesimals \\citep{beckwith2000}. Several lines of evidence suggest that the young main sequence A5V star $\\beta$~Pic \\citep[20 Myr, ][]{barrado1999} hosts planetesimals inside its large CS disk. The main argument is the presence of transient Redshifted Absorption Components (RACs) in high resolution spectra of strong metallic lines, like \\ion{Ca}{ii}~K 3934~\\AA. These spectroscopic events have been interpreted as being caused by the evaporation of comet-like highly hydrogen-depleted bodies. The interpretation is known as the Falling Evaporating Bodies (FEB) scenario \\citep[ and references therein]{lagrange2000}. However, the presence or absence of planetesimals during the pre-main sequence (PMS) phase is a controversial observational topic. The time scale for the formation of planetesimals \\citep[$\\sim$10$^4$~yr, ][]{beckwith2000} is shorter than the duration of the PMS phase ($\\sim$1--10~Myr), which suggests that they should exist during PMS stellar evolution.  UX~Ori-like PMS objects (UXORs) are characterized by a peculiar photo-polarimetric variability, which has been interpreted as the signature of massive, almost edge-on, CS disks \\citep{grinin1991}. Most UXORs have A spectral types and therefore are the PMS evolutionary precursors of $\\beta$~Pic. UXORs have been reported to show RACs \\citep{grinin1994,dewinter1999}; they also show Blueshifted Absorption Components (BACs) in their spectra. In this paper, the acronym TAC (Transient Absorption Component) will be used to denote both RACs and BACs. In analogy to $\\beta$~Pic, RACs observed in UXORs have been interpreted in terms of the FEB scenario \\citep{grady2000}. However, this interpretation has recently been questioned by \\citet{natta2000}, who used the spectra obtained by \\citet{grinin2001} to analyze the dynamics and chemical composition of a very strong, redshifted event in UX Ori itself, an A4 IV star \\citep{mora2001} $\\sim 2\\times 10^6$ year old \\citep{natta1999}. Gas accretion from a CS disk was suggested in \\citet{natta2000} as an alternative to the FEB scenario to explain the observed RACs in UX Ori. In addition, \\citet{beust2001} have found that the FEB hypothesis cannot produce detectable transient absorptions in typical HAe CS conditions, unless the stars are relatively old ($\\ge 10^7$ yr).  A detailed observational study of the kinematics and chemistry of TACs, e. g. by means of \\'echelle spectra which simultaneously record many metallic and hydrogen lines, can discriminate between the two scenarios. For instance, in a FEB event large metallicities are expected, while gas accreted from a CS disk would have approximately solar abundances. A strong observational requirement is set by the time scale of monitoring of the TACs. In \\citet{natta2000} spectra were taken 3 days apart. Since UX Ori is a highly variable star, there is some ambiguity in identifying transient spectral components of different velocities detected over this time interval as having the same physical origin. A better time resolution is needed in order to ensure that the TACs observed at different velocities  are due to the dynamical evolution of the same gas. The EXPORT collaboration \\citep{eiroa2000a} obtained high resolution \\'echelle spectra of a large sample of PMS stars \\citep{mora2001}. About 10 PMS stars showed TACs which were intensively monitored ($\\Delta$t~$\\leq$~1~day). The study of the kinematics and chemistry of these events provides important tools for identifying their origin, or at least to put severe constraints on it.  This paper presents an analysis of the TACs observed in UX Ori by EXPORT and shows that the results are not compatible with a FEB scenario. The layout of the paper is as follows: Section 2 presents a brief description of the EXPORT observations. Section 3 presents the procedures followed in the analysis of the spectra. Section 4 presents the kinematic and chemical results of the detected TACs. Section 5 gives a discussion of the dynamics and nature of the gas. In section 6, we present our conclusions. ",
        "conclusions": "The data presented in this paper allow us to analyse the spectroscopic behaviour of the CS gas disk around UX Ori on time scales of months, days and hours. Significant activity in the CS disk is always present, which manifests itself in the continuous appearance and disappearance of absorption components detected in hydrogen and in many metallic lines. This activity is not related to substantial variations of the stellar photosphere. Blobs of gas experiencing infalling and outflowing motions are the likely origin of the transient features. Blobs undergo accelerations/decelerations of the order of tenths of m\\,s$^{-2}$ and last for a few days. Detectable changes in the gas dynamics occur on a time scale of hours, but the intrinsic velocity dispersion of the blobs appears to remain rather constant. No noticeable differences are seen in the properties of the infalling and outflowing gas, although infalls might have larger velocity dispersion. The relative absorption strength of the transient absorptions suggests gas abundances similar to the solar metallicity, ruling out the evaporation of solid bodies as the physical origin of the spectroscopic features. We suggest that the data should be analysed in the context of detailed magnetospheric accretion models, similar to those used for T Tauri stars."
    },
    "0208/astro-ph0208379_arXiv.txt": {
        "abstract": "We present {\\em Chandra\\/} observations of the young elliptical galaxy NGC 1700.  The X-ray isophotes are highly flattened between semimajor axes of $30\\arcsec$ and $80\\arcsec$, reaching a maximum ellipticity $\\epsilon_X \\approx 0.65$ at $60\\arcsec$ ($15\\kpc$).  The surface brightness profile in the spectrally soft, flattened region is shallower than that of the starlight, indicating that the emission comes from hot gas rather than stellar sources. The flattening is so extreme that the gas cannot be in hydrostatic equilibrium in any plausible potential. A likely alternative is that the gas has significant rotational support. A simple model, representing isothermal gas distributed about a particular angular momentum, can reproduce the X-ray morphology while staying consistent with stellar kinematics. The specific angular momentum of the gas matches that of the stars in the most isophotally distorted outer part of the galaxy, and its cooling time matches the time since the last major merger. We infer that the gas was acquired in that merger, which involved a pre-existing elliptical galaxy with a hot ISM. The hot gas carried the angular momentum of the encounter, and has since gradually settled into a rotationally flattened, cooling disk. ",
        "introduction": "}  Much of what we know about the mass distributions in the outer parts of giant elliptical galaxies comes from X-ray observations of their hot interstellar media (ISM). Since the initial studies of M87 \\citep{Mat78,FLG80}, mass profiles of several dozen systems have been derived from X-ray data \\citep[and references therein, for example]{LW99}, strengthening the case for the ubiquity of dark halos. All these results rely on the assumption that the gas is in hydrostatic equilibrium, an assumption that is difficult to validate. If hydrostatic equilibrium holds, the X-ray surface brightness must trace the projected potential. Conversely, if there is no plausible potential consistent with the morphology, then the gas cannot be in equilibrium. But the distinction between an implausible potential and one that is merely strange is not clear-cut. The quintessential example is NGC 720, whose X-ray isophotes, while rounder than the optical isophotes, are flatter than the projected stellar potential. \\citet{BC94} were able to account for this difference by postulating a dark halo that was alarmingly flat, though not altogether implausible. Later studies making use of {\\em ASCA\\/} \\citep{BC98} and {\\em Chandra\\/} \\citep{Buo02} data have shown that removing the contribution from stellar sources makes the remaining diffuse emission rounder, relaxing the constraints somewhat, but still implying a substantial halo flattening.  In the {\\em Chandra\\/} era of high resolution, luminous ellipticals continue to appear round in X-rays, even at small radii. This comes as some surprise in light of arguments by Mathews and collaborators that at least some of the gas should be rapidly rotating, and thus not in true hydrostatic equilibrium. \\citet{KM95} pointed out that, since stellar evolution is a primary source of gas, the gas should retain the specific angular momentum of the stars, which is typically only a factor of a few below that required for rotational support. Cooling gas would therefore need to sink by only a modest factor in radius before becoming rotationally supported and forming a disk. These ideas were subsequently fleshed out in detailed simulations by \\citet[collectively BM]{BM96, BM97}. However, a careful examination of archival {\\em ROSAT\\/} data by \\citet{HB00} showed no indication of central X-ray disks, and in fact no significant difference between the distribution of X-ray and optical ellipticities, leading \\citet{BM00} to propose a combination of heat conduction and mass dropout to explain the absence of the predicted structures.  In this paper we report on {\\em Chandra\\/} observations of the young elliptical NGC 1700. We find that this galaxy's X-ray isophotes are not merely flatter than the stellar potential, but flatter than the starlight---so flat, in fact, that the gas cannot be in hydrostatic equilibrium in any plausible potential. We describe the observations and reductions in \\S\\ 2 below and the basic results in \\S\\ 3. We then argue in \\S\\ 4 that the gas is settling into a rotationally flattened cooling disk, similar to those predicted by BM, but much larger in scale. The high angular momentum and the long cooling time suggest that the gas was accreted in the last major merger. These points are reiterated, and their implications briefly discussed, in \\S\\ 5. ",
        "conclusions": "{\\em Chandra\\/} observations of NGC 1700 show extended X-ray emission that is dominated by a hot ISM. The gas is distributed in a flattened structure that is inconsistent with hydrostatic equilibrium in any plausible potential. A simple model in which gas at a single temperature, centered around a particular angular momentum, is dispersed in a logarithmic potential can account for the gross X-ray morphology and be consistent with stellar kinematics. The angular momentum of the gas matches that of the stars in the isophotally distorted outer part of the galaxy, and its cooling time matches the dynamical age of the last merger. We infer that the gas was acquired in that merger, carrying the angular momentum of the encounter, and is settling into a rotationally flattened, cooling disk.  The presence of a rotationally supported disk in this object, and the growing evidence for shocks and bubbles in the ISM of elliptical galaxies \\citep{Jon01,Fin01} casts doubt on the assumption that elliptical galaxies are simple, hydrostatic systems. Mass profiles derived from X-ray data and hydrostatic models may therefore be in error, even in those systems that appear symmetric. A modest degree of rotational support in the gas may significantly affect the shape of the inferred mass distribution, depending on how the angular momentum is distributed. In the models of \\S\\ \\ref{s.disk}, rotational support increases outward interior to $R_0$ and decreases outward beyond this radius. Applying a hydrostatic model would therefore lead to a mass profile that is too centrally concentrated. Interestingly, galaxy cluster mass profiles are found, in some cases \\citep{Dav01,Nev01}, to be unusually concentrated compared to the expected NFW \\citep{NFW} profile. Whether this discrepancy could also stem from gas rotation remains to be seen."
    },
    "0208/astro-ph0208465_arXiv.txt": {
        "abstract": "\\noindent Recent observational indications of an accelerating universe enhance the interest in studying models with a cosmological constant. We investigate cosmological expansion (FRW metric) with $\\Lambda>0$ for a general linear equation of state $p=w\\rho$, $w>-1$, so that the interplay between cosmological vacuum and quintessence is allowed, as well.  Four closed-form solutions (flat universe with any $w$, and $w=1/3$, $-1/3,\\, -2/3$) are given, in a proper compact representation. Various estimates of the expansion are presented in a general case when no closed-form solutions are available. For the open universe a simple relation between solutions with different parameters is established: it turns out that a solution with some $w$ and (properly scaled) $\\Lambda$ is expressed algebraically via another solution with special different values of these parameters.  The expansion becomes exponential at large times, and the amplitude at the exponent depends on the parameters. We study this dependence in detail, deriving various representations for the amplitude in terms of integrals and series. The closed-form solutions serve as benchmarks, and the solution transformation property noted above serves as a useful tool. Among the results obtained, one is that for the open universe with relatively small cosmological constant the amplitude is independent of the equation of state. Also, estimates of the cosmic age through the observable ratio $\\Omega_\\Lambda/\\Omega_M$ and parameter $w$ are given; when inverted, they provide an estimate of $w$, i. e., the state equation, through the known $\\Omega_\\Lambda/\\Omega_M$ and age of the universe. \\vskip5mm \\noindent {\\bf Key words:} {\\it cosmology---cosmological vacuum---quintessence---exact solutions} ",
        "introduction": "Recent data on the brightness of distant SN Ia ([1,2]; see [3] and the references therein), as well as evidence coming from the cosmic age, large scale structure, and cosmic microwave background anisotropy combined with the cluster dynamics, indicate, most probably, that the observed cosmological expansion is  accelerating. Perhaps the most natural, although definitely not unique, reason for this acceleration is the presence of a cosmic vacuum of nonzero energy and pressure; because of that, investigation of various cosmological models including a positive cosmological constant becomes interesting, once again. In this paper we study one such model involving matter with an arbitrary (linear) equation of state; in particular, an interplay between vacuum and quintessence [4] is studied.  We consider a Friedmann cosmology described by the Friedmann-Robertson-Walker metric , \\begin{equation}\\label{metr} ds^2= - dt^2 + a^2(t)\\left( \\frac{dr^2}{1-kr^2} + r^2d\\Omega^2\\right) \\; , \\end{equation} where $d\\Omega$ is the element of solid angle and $k= -1, 0$, or $1$, according to whether the universe is open, flat, or closed (we use the units with $G=c=1$). As usual, the energy-momentum tensor, consistent with homogeneity and isotropy of the universe, is the one corresponding to a perfect fluid, described by the energy density $\\rho$ and  pressure $p$. The latter are assumed to satisfy the linear equation of state \\begin{equation}\\label{eqst} p=w\\rho,\\qquad w={\\rm const}>-1\\; . \\end{equation} The range of values of $w$ is chosen on the grounds that $w=-1$ corresponds to the vacuum equation of state, and vacuum contributions are already included in our model with the cosmological constant $\\Lambda>0$. Moreover, no physical reasons seem to be known so far to justify the values of $w$ which are more negative that the vacuum one, $w=-1$, and the cosmological evolution in this range is very different. On the other hadn, it is not easy to imagine the kind of physics that would lead to matter with the pressure larger than its density, but the following analysis does not depend at all on whether $w$ is larger or smaller than one, so we do not limit this parameter from above. Condition (\\ref{eqst}) allows for quintessence , that is, for $-1<w<0$; investigation of its influence on cosmological expansion is one of the goals of this paper.  Under these assumptions, Einstein's equations reduce to \\begin{equation}\\label{goveq} 3\\left(\\frac{1}{a} \\frac{da}{dt}\\right)^2= 8 \\pi  \\rho + \\Lambda - 3\\, \\frac{k}{a^2},\\qquad \\frac{d\\rho}{dt} = -3\\,\\frac{\\rho + p}{a}\\,\\frac{da}{dt}\\; , \\end{equation} the second one expressing energy conservation. By (\\ref{eqst}), this equation is easily integrated giving the density in terms of the scale factor, \\begin{equation}\\label{rho} \\rho = \\frac{M}{a^{3(w+1)}}\\; , \\end{equation} where the constant $M>0$ characterizes the abundance of matter in the universe. A substituion of (\\ref{rho}) in the first of equations (\\ref{goveq}) produces the equation for the scale factor $a(t)$ only, \\begin{equation}\\label{sfeq} \\left(\\frac{da}{dt}\\right)^2= \\frac{8\\pi M}{3}\\,\\frac{1}{a^{3w+1}}  + \\frac{\\Lambda}{3}\\,a^2 - k \\; ; \\end{equation} we append it with the Big Bang initial condition \\begin{equation}\\label{initcon} a(0)=0\\; . \\end{equation} As seen from (\\ref{rho}), the initial value of the density, $\\rho(0)$, is then infinite for $w>-1/3$, and zero for $-1<w<-1/3$; it is finite and positive for $w=-1/3$, which value corresponds to the so called Einstein quintessence\\footnote{The term `Einstein quintessence' was coined recently in [5] on the grounds of the effective equation of state for Einstein's cosmological solution [6] of 1917.}.  Equation (\\ref{sfeq}) with the initial condition (\\ref{initcon}) constitute the problem studied below; its solution, together with formula (\\ref{rho}), specifies completely the corresponding cosmological expansion at all times. ",
        "conclusions": "\\subsection{Cosmological Applications}   We have determined various significant properties of the cosmological solution, particularly, those of the expansion amplitude, and provided different expressions for the latter. The plots of $ A_\\nu(k\\omega)$ as a function of  $k\\omega$ for several `popular' values of the parameter $\\nu$, including positive, zero, and negative ones, are given in Fig. 2.  An interesting question is what information on the universe can be gotten if the value of the expansion amplitude, say, $A_*$, is known? Even if no other information is available, but $A_*>1$, from (\\ref{A(0)}) and the monotonicity of the amplitude established in sec. 2, one immediately concludes that the universe is open, since  $A_\\nu(k\\omega)<1$ for $k=1$ (and, of course, $\\nu\\geq0$). If, on the other hand, $0<A_*<1$, but the equation of state, i. e., parameter $\\nu$, is more or less known, then equation \\[ A_\\nu(k\\omega)=A_* \\] provides, again due to the monotonic property, a unique solution for $k\\omega$, which gives the sign of the curvature, and a relation between $M$ and $\\Lambda$ [see (\\ref{par})].  One way of inferring the amplitude from observational data is as follows. Formula (\\ref{rho}) for the density combined with (\\ref{var}) reduces to $8\\pi\\rho=\\Lambda/y^{\\nu+2}$, therefore \\[ \\frac{\\Omega_M}{\\Omega_\\Lambda}=\\frac{8\\pi\\rho}{\\Lambda}=\\frac{1}{y^{\\nu+2}}\\; , \\] so that \\begin{equation} \\label{y} y=\\left(\\frac{\\Omega_\\Lambda}{\\Omega_M}\\right)^{1/(\\nu+2)}\\; . \\end{equation} This equality holds at all moments of time. If, however, the expansion is in the asymptotic regime, (\\ref{y}) and (\\ref{A_nu}) allow for \\begin{equation} \\label{Aobs} A_\\nu(k\\omega)\\approx \\left(\\frac{\\Omega_\\Lambda}{\\Omega_M}\\right)^{1/(\\nu+2)}\\exp(-\\tau),\\qquad y(\\tau)\\gg 1\\; . \\end{equation} So, if the matter equation of state, the ratio $\\Omega_\\Lambda/\\Omega_M$, and the properly scaled [see (\\ref{var})] cosmic age are known, the expansion amplitude is known as well, if the asymptotic stage of the expansion is reached. By (\\ref{y}), the latter requires \\begin{equation} \\label{asreg} \\left(\\frac{\\Omega_\\Lambda}{\\Omega_M}\\right)^{1/(\\nu+2)}\\gg1\\; , \\end{equation} which, for $\\nu$ not much smaller than negative unity, means just $\\Omega_\\Lambda/\\Omega_M\\gg1$. But in case the dominant form of matter in the universe is quitessence with the equation of state close to the vacuum one, condition (\\ref{asreg}) might hold, and the expansion be in the asymptotic regime, despite the fact that $\\Omega_\\Lambda/\\Omega_M>1$ is still of the order of unity.  For any moment of the expansion the estimates for the age of the universe come from combining expression (\\ref{y}) for the scale factor via observables with `geometrical' inequalities (\\ref{tauflat}): \\begin{equation} \\label{ages} \\tau\\,<\\,\\tau_{Fl}(\\Omega_\\Lambda/\\Omega_M,\\,\\nu),\\quad k=-1;\\qquad \\tau\\,>\\,\\tau_{Fl}(\\Omega_\\Lambda/\\Omega_M,\\,\\nu),\\quad k=1\\; . \\end{equation} Here \\begin{equation} \\label{taufl} \\tau_{Fl}(\\Omega_\\Lambda/\\Omega_M,\\,\\nu)\\equiv \\frac{2}{\\nu+2}\\,\\ln\\left[\\left(\\frac{\\Omega_\\Lambda}{\\Omega_M}\\right)^{1/2}+\\left(1+\\frac{\\Omega_\\Lambda}{\\Omega_M}\\right)^{1/2}\\right] \\end{equation} is the exact age of the flat universe with the same parameters. If $\\nu$, $\\Omega_\\Lambda/\\Omega_M$, and the cosmic age are known accurately enough, its comparison with the bound (\\ref{taufl}) allows for an immediate determination of the curvature of the universe. If, instead, only the last two parameters and the sign of the curvature are known, the state equation parameter can be estimated.  We now apply the above results to the currently favored cosmological model. It includes baryonic and dark matter, both with the same equation of state $w=0$ and total $\\Omega_M=0.3\\pm0.1$, and vacuum with $\\Omega_\\Lambda=0.7\\pm0.1$. The third component is relativistic particles ($w=1/3$), but its abundance $\\Omega_R<0.001$ is so small compared to the first two that it should be neglected. This results in exactly the studied model with $\\nu=1$ and $\\Omega_\\Lambda/\\Omega_M=1.5\\div4$. Thus \\begin{equation} \\label{Om/Om} \\left(\\frac{\\Omega_\\Lambda}{\\Omega_M}\\right)^{1/(\\nu+2)}=\\left(\\frac{\\Omega_\\Lambda}{\\Omega_M}\\right)^{1/3}=1.1\\,\\div\\,1.6 \\sim 1\\; , \\end{equation} so, by (\\ref{y}), we are still rather far from the large time regime.  According to (\\ref{taufl}) and (\\ref{Om/Om}), for our universe \\begin{equation} \\label{tauflour} \\tau_{Fl}=0.7\\,\\div\\,1.0\\; . \\end{equation} The dimensional time is related to $\\tau$ by [see (\\ref{var})] \\begin{equation} \\label{time} T=\\tau\\,\\sqrt{\\frac{3}{G\\Lambda}}=\\tau\\,(0.7\\,\\div\\,0.8)\\times 10^{18}\\,sec= \\tau\\,(22.2\\div25.4)\\,Gyr\\; , \\end{equation} with \\[ \\Lambda=8\\pi\\rho_V=8\\pi(0.7\\pm0.1)\\rho_c=(1.0\\,\\div\\,1.6)\\times 10^{-28}\\,g/cm^3\\; \\] based on \\[ \\rho_c=(0.7\\pm0.08)\\times 10^{-29}\\,g/cm^3\\;, \\] which corresponds to the range of the Hubble constant $H=70\\pm10\\,km/sec\\cdot Mpc$. By virtue of (\\ref{tauflour}) and (\\ref{time}), for our universe \\begin{equation} \\label{timefl} T_{Fl}=\\tau_{Fl}\\,(22.2\\,\\div\\,25.4)\\,Gyr=(15.5\\,\\div\\,25.4)\\,Gyr\\; , \\end{equation} that is, its age in case the universe is flat. Globular cluster data independently give the lower bound of the cosmic age as \\begin{equation} \\label{ageobs} T_u\\geq(12\\,\\div\\,16)\\,Gyr\\; . \\end{equation} According to these data, our universe is open if $T_u=(12\\,\\div\\,15.5-0)\\,Gyr$, it could be either open or closed if $T_u=(15.5\\,div\\,25.4)\\,Gyr$, and only closed if $T_u>\\,25.4)\\,Gyr$, which is rather improbable. Therefore one can conclude that our universe most probably is open.  Under this assumption we can now obtain a lower bound of the cosmic age, to check the consistency of our estimates. Indeed, using (\\ref{estyo}), we obtain an estimate of the parameter $\\omega$ for our universe as \\begin{equation} \\label{omour} \\omega\\,<\\,\\frac{y^2}{\\sinh^2\\tau_u}=\\frac{\\left({\\Omega_\\Lambda}/{\\Omega_M}\\right)^{2/3}}{\\sinh^2\\tau_u}< 2.7\\equiv\\omega_{est}\\;. \\end{equation} This and the basic expression (\\ref{tauy}) for the solution immediately produce \\[ \\tau\\,>\\, \\int_0^{\\left({\\Omega_\\Lambda}/{\\Omega_M}\\right)^{1/3}} \\frac{dx}{\\sqrt{x^{-1}+x^2+\\omega_{est}}}> \\int_0^{1.1} \\frac{dx}{\\sqrt{x^{-1}+x^2+2.7}}\\approx0.45 \\; , \\] or, by (\\ref{time}), \\[ T_u\\,>\\,(10.0\\,\\div\\,11.4)\\,Gyr\\; , \\] in a nice agreement with (\\ref{ageobs}).  Finally, to check the consistency of the current cosmological data, we can convert the estimate (\\ref{ages}) for an open universe into an upper bound for the state equation parameter $\\nu$, namely: \\begin{equation} \\label{estnuo} \\nu\\,<\\,2\\,\\left\\{\\tau_u^{-1} \\ln\\left[\\left(\\frac{\\Omega_\\Lambda}{\\Omega_M}\\right)^{1/2}+\\left(1+\\frac{\\Omega_\\Lambda}{\\Omega_M}\\right)^{1/2}\\right]-1\\right\\}\\; . \\end{equation} By (\\ref{ageobs}) taken as the real range of the cosmic age, (\\ref{time}) and (\\ref{Om/Om}), for our universe this provides \\[ \\nu\\,<\\,0.5\\,\\div\\,10\\; . \\] This is almost entirely consistent with the chosen $\\nu=1$ (pressureless dust), except for a small range of parameters, when the ratio $\\Omega_\\Lambda$ is close to its minimum observed value $0.6$ and simultaneously the cosmic age is close to its maximum of $16\\,Gyr$.  \\subsection{Some Generalizations}   In conclusion, we point out two possible generalizations of the present study. The first of them deals with the phase transition, that is, with a sudden change of the equation of state, when the value of $w$ changes abruptly at some moment of time. This case, as well as the one with a whole sequence of phase transitions, can be investigated in a fashion similar to the above.  The second way to generalize the study is to consider multicomponent matter, when \\[ \\rho=\\sum_{n=1}^N\\rho_n,\\qquad p=\\sum_{n=1}^Np_n,\\qquad p_n=w_n\\rho_n,\\quad w_n>-1,\\quad n=1,2,\\dots,N>1\\; , \\] and the components do not interact. That means that the conservation equation in (\\ref{goveq}) holds for every component separately, yielding $\\rho_n=M_n/a^{3(w_n+1)}$, $M_n={\\rm const>0}$, and thus  the analog of the governing problem (\\ref{ydot}) becomes \\[ \\dot{y}^2=y^{-\\nu_1}+y^2-k\\omega+\\sum_{n=2}^N\\,\\mu_ny^{-\\nu_n},\\qquad y(0)=0\\; , \\] with $y$, $\\tau$ and $\\omega$ normalized as in (\\ref{var}), (\\ref{par}) using $w_1$ and $M_1$, $\\nu_n=3w_n+1$, and $\\mu_n>0$ being the abundance of the $n$th component normalized appropriately. Writing $y(\\tau,\\nu_1,\\nu_1,\\dots,\\nu_N, k\\omega)$ for the solution, we have \\[ y(\\tau,\\nu_1,\\nu_1,\\dots,\\nu_N, k\\omega)>y(\\tau,\\nu_j, k\\omega) \\] for any pertinent $j$; generally, adding every new component enhances the expansion. Assuming $\\nu_1$ is the smallest of all the powers, we see that at very large times the expansion is described by the one--component equation (\\ref{ydot}) with $\\nu=\\nu_1$; however, the growth amplitude depends essentially on the whole expansion history, in other words, on all the equations of state.  As for the other results, all the estimates of sec. 4 remain true, including inequalities (\\ref{tauflat}) for the ages of open, flat, and closed universe (of course, the expression for $\\tau_{Fl}$ depends now on all the parameters involved). The properly generalized transformation property (\\ref{transol}) for the open universe solutions also holds; in fact, the number of such independent transformations  is equal to the number of matter components, $N$."
    },
    "0208/astro-ph0208186_arXiv.txt": {
        "abstract": "Of the light nuclides observed in the universe today, D, \\3he, \\4he, and \\7li are relics from its early evolution. The primordial abundances of these relics, produced via Big Bang Nucleosynthesis (BBN) during the first half hour of the evolution of the universe provide a unique window on Physics and Cosmology at redshifts $\\sim 10^{10}$.  Comparing the BBN-predicted abundances with those inferred from observational data tests the consistency of the standard cosmological model over ten orders of magnitude in redshift, constrains the baryon and other particle content of the universe, and probes both Physics and Cosmology beyond the current standard models.  These lectures are intended to introduce students, both of theory and observation, to those aspects of the evolution of the universe relevant to the production and evolution of the light nuclides from the Big Bang to the present.  The current observational data is reviewed and compared with the BBN predictions and the implications for cosmology (\\eg universal baryon density) and particle physics (\\eg relativistic energy density) are discussed. While this comparison reveals the stunning success of the standard model(s), there are currently some challenges which leave open the door for more theoretical and observational work with potential implications for astronomy, cosmology, and particle physics. ",
        "introduction": "Observations of the present universe establish that, on sufficiently large scales, galaxies and clusters of galaxies are distributed homogeneously and they are expanding isotropically.  On the assumption that this is true for the large scale universe throughout its evolution (at least back to redshifts $\\sim 10^{10}$, when the universe was a few hundred milliseconds old), the relation between space-time points may be described uniquely by the Robertson -- Walker Metric \\begin{equation} ds^{2} = c^{2}dt^{2} - a^{2}(t)({dr^{2} \\over 1 - {\\kappa}r^{2}} + r^{2}d\\Omega^{2})\\,, \\label{rwmetric} \\end{equation} where $r$ is a {\\it comoving} radial coordinate and $\\theta$ and $\\phi$ are {\\it comoving} spherical coordinates related by \\begin{equation} d{\\Omega}^{2} \\equiv d{\\theta}^{2} + sin^{2}{\\theta}d{\\phi}^{2}\\,. \\end{equation} A useful alternative to the comoving radial coordinate $r$ is $\\Theta$, defined by \\be d\\Theta \\equiv {dr \\over (1 - {\\kappa}r^{2})^{1/2}}\\,. \\ee The 3-space curvature is described by $\\kappa$, the curvature constant. For closed (bounded), or ``spherical\" universes, $\\kappa > 0$; for open (unbounded), or ``hyperbolic\" models, $\\kappa < 0$; when $\\kappa = 0$, the universe is spatially flat or ``Euclidean\".  It is the ``scale factor\", $a = a(t)$, which describes how physical distances between comoving locations change with time.  As the universe expands, $a$ increases while, for comoving observers, $r$, $\\theta$, and $\\phi$ remain fixed.  The growth of the separation between comoving observers is solely due to the growth of $a$.  Note that {\\bf neither} $a$ {\\bf nor} $\\kappa$ is observable since a rescaling of $\\kappa$ can always be compensated by a rescaling of $a$.  Photons and other massless particles travel on geodesics: $ds = 0$; for them (see eq.~\\ref{rwmetric}) $d\\Theta = \\pm ~cdt/a(t)$.  To illustrate the significance of this result consider a photon travelling from emission at time $t_{e}$ to observation at a later time $t_{o}$. In the course of its journey through the universe the photon traverses a comoving radial distance $\\Delta\\Theta$, where \\be \\Delta\\Theta = \\int^{t_{o}}_{t_{e}} ~{cdt \\over a(t)}\\,. \\ee  Some special choices of $t_{e}$ or $t_{o}$ are of particular interest. For $t_{e} \\rightarrow 0$, $\\Delta\\Theta \\equiv \\Theta_{\\rm H}(t_{o})$ is the comoving radial distance to the ``Particle Horizon\" at time $t_0$. It is the comoving distance a photon could have travelled (in the absence of scattering or absorption) from the beginning of the expansion of the universe until the time $t_{o}$.  The ``Event Horizon\", $\\Theta_{\\rm E} (t_{e})$, corresponds to the limit $t_{o} \\rightarrow \\infty$ (provided that $\\Theta_{\\rm E}$ is finite!).  It is the comoving radial distance a photon will travel for the entire future evolution of the universe, after it is emitted at time $t_{e}$.  \\subsection{Redshift}  Light emitted from a comoving galaxy located at $\\Theta_{g}$ at time $t_{e}$ will reach an observer situated at $\\Theta_{o} \\equiv 0$ at a later time $t_{o}$, where \\be \\Theta_{g}(t_{o}, t_{e}) = \\int^{t_{o}}_{t_{e}} ~{cdt \\over a(t)}\\,. \\label{thetag} \\ee Equation \\ref{thetag} provides the relation among $\\Theta_{g}$, $t_{o}$, and $t_{e}$.  For a comoving galaxy, $\\Theta_{g}$ is unchanged so that differentiating eq.~\\ref{thetag} leads to \\be {dt_{o} \\over dt_{e}} = {a_{o} \\over a_{e}} = {\\nu_{e} \\over \\nu_{o}} = {\\lambda_{o} \\over \\lambda_{e}}\\,. \\ee This result relates the evolution of the universe ($a_{o}/a_{e}$) as the photon travels from emission to observation, to the change in its frequency ($\\nu$) or wavelength ($\\lambda$).  As the universe expands (or contracts!), wavelengths expand (contract) and frequencies decrease (increase).  The redshift of a spectral line is defined by relating the wavelength at emission (the ``lab\" or ``rest-frame\" wavelength $\\lambda_{e}$) to the wavelength observed at a later time $t_{o}$, $\\lambda_{o}$. \\be z \\equiv {\\lambda_{o} - \\lambda_{e} \\over \\lambda_{e}} ~\\Longrightarrow ~1 + z = {a_{o} \\over a_{e}} = {\\nu_{e} \\over \\nu_{o}}\\,. \\label{redshift} \\ee  Since the energies of photons are directly proportional to their frequencies, as the universe expands photon energies redshift to smaller values: E$_{\\gamma} = h\\nu \\Longrightarrow ~$E$_{\\gamma} \\propto (1 + z)^{-1}$.  For {\\bf all} particles, massless or not, de Broglie told us that wavelength and momentum are inversely related, so that: p ~$\\propto ~\\lambda^{-1} \\Longrightarrow ~$p$ ~\\propto ~(1 + z)^{-1}$.  All momenta redshift; for non-relativistic particles (\\eg galaxies) this implies that their ``peculiar\" velocities redshift: v = p/M $\\propto (1 + z)^{-1}$.  \\subsection{Dynamics}  Everything discussed so far has been ``geometrical\", relying only on the form of the Robertson-Walker metric.  To make further progress in understanding the evolution of the universe, it is necessary to determine the time dependence of the scale factor $a(t)$.  Although the scale factor is not an observable, the expansion rate, the Hubble parameter, $H = H(t)$, is. \\be H(t) \\equiv {1 \\over a}({da \\over dt})\\,. \\ee The present value of the Hubble parameter, often referred to as the Hubble ``constant\", is $H_{0} \\equiv H(t_{0}) \\equiv 100~h$~km~s$^ {-1}$Mpc$^{-1}$ (throughout, unless explicitly stated otherwise, the subscript ``0\" indicates the present time).  The inverse of the Hubble parameter provides an expansion timescale, $H_{0}^{-1} = 9.78~h^{-1}$~Gyr.  For the HST Key Project (Freedman \\etal 2001) value of $H_{0} = 72$~km~s$^{-1}$Mpc$^{-1}$ ($h = 0.72$), $H_{0}^{-1} = 13.6$~Gyr.  The time-evolution of $H$ describes the evolution of the universe. Employing the Robertson-Walker metric in the Einstein equations of General Relativity (relating matter/energy content to geometry) leads to the Friedmann equation \\be H^{2} = {8\\pi \\over 3}G\\rho - {\\kappa c^{2} \\over a^{2}}\\,. \\label{friedmann} \\ee It is convenient to introduce a {\\it dimensionless} density parameter, $\\Omega$, defined by \\be \\Omega \\equiv {8\\pi G\\rho \\over 3H^{2}}\\,. \\label{omega} \\ee We may rearrange eq.~\\ref{friedmann} to highlight the relation between matter content and geometry \\be \\kappa c^{2} = (aH)^{2}(\\Omega - 1)\\,. \\label{kappa} \\ee Although, in general, $a$, $H$, and $\\Omega$ are all time-dependent, eq.~\\ref{kappa} reveals that if ever $\\Omega < 1$, then it will always be $< 1$ {\\bf and} in this case the universe is open ($\\kappa < 0$). Similarly, if ever $\\Omega > 1$, then it will always be $> 1$ {\\bf and} in this case the universe is closed ($\\kappa > 0$).  For the special case of $\\Omega = 1$, where the density is equal to the ``critical density\" $\\rho_{\\rm crit} \\equiv 3H^{2}/8\\pi G$, $\\Omega$ is always unity and the universe is flat (Euclidean 3-space sections; $\\kappa = 0$).  The Friedmann equation (eq.~\\ref{friedmann}) relates the time-dependence of the scale factor to that of the density.  The Einstein equations yield a second relation among these which may be thought of as the surrogate for energy conservation in an expanding universe. \\be {d\\rho \\over \\rho} + 3(1 + {p \\over \\rho}){da \\over a} = 0\\,. \\label{eqnofstate} \\ee For ``matter\" (non-relativistic matter; often called ``dust\"), $p \\ll \\rho$, so that $\\rho/\\rho_{0} = (a_{0}/a)^{3}$.  In contrast, for ``radiation\" (relativistic particles) $p = \\rho/3$, so that $\\rho/\\rho_{0} = (a_{0}/a)^{4}$. Another interesting case is that of the energy density and pressure associated with the vacuum (the quantum mechanical vacuum is not empty!).  In this case $p = -\\rho$, so that $\\rho = \\rho_{0}$. This provides a term in the Friedmann equation entirely equivalent to Einstein's ``cosmological constant\" $\\Lambda$.  More generally, for $p = w\\rho$, $\\rho/\\rho_{0} = (a_{0}/a)^{3(1+w)}$.  Allowing for these three contributions to the total energy density, eq.~\\ref{friedmann} may be rewritten in a convenient dimensionless form \\be ({H \\over H_{0}})^{2} = \\Omega_{\\rm M}({a_{0} \\over a})^{3} + \\Omega_{\\rm R}({a_{0} \\over a})^{4} + \\Omega_{\\rm \\Lambda} + (1 - \\Omega)({a_{0} \\over a})^{2}\\,, \\label{hsquared} \\ee where $\\Omega \\equiv \\Omega_{\\rm M} + \\Omega_{\\rm R} + \\Omega_{\\rm \\Lambda}$.  Since our universe is expanding, for the early universe ($t \\ll t_{0}$) $ a \\ll a_{0}$, so that it is the ``radiation\" term in eq.~\\ref{hsquared} which dominates; the early universe is radiation-dominated (RD).  In this case $a \\propto t^{1/2}$ and $\\rho \\propto t^{-2}$, so that the age of the universe or, equivalently, its expansion rate is fixed by the radiation density.  For thermal radiation, the energy density is only a function of the temperature ($\\rho_{\\rm R} \\propto T^{4}$).  \\subsubsection{Counting Relativistic Degrees of Freedom}  It is convenient to write the total (radiation) energy density in terms of that in the CMB photons \\be \\rho_{\\rm R} \\equiv ({g_{eff} \\over 2})\\rho_{\\gamma}\\,, \\ee where $g_{eff}$ counts the ``effective\" relativistic degrees of freedom.  Once $g_{eff}$ is known or specified, the time -- temperature relation is determined.  If the temperature is measured in energy units ($kT$), then \\be t({\\rm sec}) = ({2.4 \\over g_{eff}^{1/2}})T^{-2}_{\\rm MeV}\\,. \\ee If more relativistic particles are present, $g_{eff}$ increases and the universe would expand faster so that, at {\\bf fixed} $T$, the universe would be younger.  Since the synthesis of the elements in the expanding universe involves a competition between reaction rates and the universal expansion rate, $g_{eff}$ will play a key role in determining the BBN-predicted primordial abundances.  \\begin{itemize}  \\item{\\it Photons}  Photons are vector bosons.  Since they are massless, they have only two degress of freedom: $g_{eff} = 2$.  At temperature $T$ their number density is $n_{\\gamma} = 411(T/2.726K)^{3}~$cm$^{-3} = 10^{31.5}T^{3}_{\\rm MeV}~$cm$^{-3}$, while their contribution to the total radiation energy density is $\\rho_{\\gamma} = 0.261 (T/2.726K)^{4}$~eV~cm$^{-3}$.  Taking the ratio of the energy density to the number density leads to the average energy per photon $\\langle {\\rm E}_{\\gamma} \\rangle = \\rho_{\\gamma}/n_{\\gamma} = 2.70~kT$.  All other relativistic {\\bf bosons} may be simply related to photons by \\be {n_{\\rm B} \\over n_{\\gamma}} = {g_{\\rm B} \\over 2}({T_{\\rm B} \\over T_{\\gamma}})^{3} \\,, ~~~~~~{\\rho_{\\rm B} \\over \\rho_{\\gamma}} = {g_{\\rm B} \\over 2}({T_{\\rm B} \\over T_{\\gamma}})^{4}\\,, ~~~~~~\\langle {\\rm E}_{\\rm B} \\rangle = 2.70~kT_{\\rm B}\\,. \\ee The $g_{\\rm B}$ are the boson degrees of freedom (1 for a scalar, 2 for a vector, etc.).  In general, some bosons may have decoupled from the radiation background and, therefore, they will not necessarily have the same temperature as do the photons ($T_{\\rm B} \\neq T_{\\gamma}$).  \\item{Relativistic Fermions}  Accounting for the difference between the Fermi-Dirac and Bose-Einstein distributions, relativistic fermions may also be related to photons \\be {n_{\\rm F} \\over n_{\\gamma}} = {3 \\over 4}{g_{\\rm F} \\over 2}({T_{\\rm F} \\over T_{\\gamma}})^{3} \\,, ~~~~~~{\\rho_{\\rm F} \\over \\rho_{\\gamma}} = {7 \\over 8}{g_{\\rm F} \\over 2}({T_{\\rm F} \\over T_{\\gamma}})^{4}\\,, ~~~~~~\\langle {\\rm E}_{\\rm F} \\rangle = 3.15~kT_{\\rm F}\\,. \\ee $g_{\\rm F}$ counts the fermion degrees of freedom.  For example, for electrons (spin up, spin down, electron, positron) $g_{\\rm F} = 4$, while for neutrinos (lefthanded neutrino, righthanded antineutrino) $g_{\\rm F} = 2$.  \\end{itemize}  Accounting for all of the particles present at a given epoch in the early (RD) evolution of the universe, \\be g_{eff} ~= ~\\Sigma_{\\rm B}~g_{\\rm B}({T_{\\rm B} \\over T_{\\gamma}})^{4} ~+ ~{7 \\over 8}~\\Sigma_{\\rm F}~g_{\\rm F}({T_{\\rm F} \\over T_{\\gamma}})^{4}\\,. \\ee For example, for the standard model particles at temperatures $T_{\\gamma} \\approx $~few MeV there are photons, electron-positron pairs, and three ``flavors\" of lefthanded neutrinos (along with their righthanded antiparticles).  At this stage all these particles are in equilibrium so that $T_{\\gamma} = T_{e} = T_{\\nu}$ where $\\nu \\equiv \\nu_{e}$, $\\nu_{\\mu}$, $\\nu_{\\tau}$.  As a result \\be g_{eff} = 2 + {7 \\over 8}(4 + 3 \\times 2) = {43 \\over 4}\\,, \\ee leading to a time -- temperature relation: $t = 0.74~T^{-2} _{\\rm Mev}$~sec.  As the universe expands and cools below the electron rest mass energy, the \\epm pairs annihilate, heating the CMB photons, but {\\bf not} the neutrinos which have already decoupled.  The decoupled neutrinos continue to cool by the expansion of the universe ($T_{\\nu} \\propto a^{-1}$), as do the photons which now have a higher temperature $T_{\\gamma} = (11/4)^{1/3}T_{\\nu}$ ($n_{\\gamma}/n_{\\nu} = 11/3$). During these epochs \\be g_{eff} = 2 + {7 \\over 8} \\times 3 \\times 2({4 \\over 11})^{4/3} = 3.36\\,, \\ee leading to a modified time -- temperature relation: $t = 1.3~T^{-2}_{\\rm Mev}$~sec.  \\subsubsection{``Extra\" Relativistic Energy}  Suppose there is some new physics beyond the standard model of particle physics which leads to ``extra\" relativistic energy so that $\\rho_{\\rm R} \\rightarrow \\rho_{\\rm R}' \\equiv \\rho_{\\rm R} + \\rho_{X}$; hereafter, for convenience of notation, the subscript R will be dropped.  It is useful, and conventional, to account for this extra energy in terms of the equivalent number of extra neutrinos: $\\Delta N_{\\nu} \\equiv \\rho_{X}/\\rho_{\\nu}$ (Steigman, Schramm, \\& Gunn 1977 (SSG); see also Hoyle \\& Tayler 1964, Peebles 1966, Shvartsman 1969).  In the presence of this extra energy, prior to \\epm annihilation \\be {\\rho' \\over \\rho_{\\gamma}} = {43 \\over 8}~(1 + {7\\Delta N_{\\nu} \\over 43}) = 5.375~(1 + 0.1628~\\Delta N_{\\nu})\\,. \\ee In this case the early universe would expand faster than in the standard model.  The pre-\\epm annihilation speedup in the expansion rate is \\be S_{pre} \\equiv {t \\over t'} = ({\\rho' \\over \\rho})^{1/2} = (1 + 0.1628~\\Delta N_{\\nu})^{1/2}\\,. \\ee  After \\epm annihilation there are similar, but quantitatively different changes \\be {\\rho' \\over \\rho_{\\gamma}} = 1.681~(1 + 0.1351~\\Delta N_{\\nu})\\,, ~~~~~~~~S_{post} = (1 + 0.1351~\\Delta N_{\\nu})^{1/2}\\,. \\ee  Armed with an understanding of the evolution of the early universe and its particle content, we may now proceed to the main subject of these lectures, primordial nucleosynthesis. ",
        "conclusions": "As observations reveal, the present universe is filled with radiation and is expanding.  According to the standard, hot big bang cosmological model the early universe was hot and dense and, during the first few minutes in its evolution, was a primordial nuclear reactor, synthesizing in significant abundances the light nuclides D, \\3he, \\4he, and \\7li.  These relics from the Big Bang open a window on the early evolution of the universe and provide probes of the standard models of cosmology and of particle physics. Since the BBN-predicted abundances depend on the competition between the early universe expansion rate and the weak- and nuclear-interaction rates, they can be used to test the standard models as well as to constrain the universal abundances of baryons and neutrinos. This enterprise engages astronomers, astrophysicists, cosmologists, and particle physicists alike.  A wealth of new observational data has reinvigorated this subject and stimulated much recent activity.  Much has been learned, revealing many new avenues to be explored (a key message for the students at this school -- and for young researchers everywhere).  The current high level of activity ensures that many of the detailed, quantitative results presented in these lectures will need to be revised in the light of new data, new analyses of extant data, and new theoretical ideas.  Nonetheless, the underlying physics and the approaches to confronting the theoretical predictions with the observational data presented in these lectures should provide a firm foundation for future progress.  Within the context of the standard models of cosmology and of particle physics (SBBN) the relic abundances of the light nuclides depend on only one free parameter, the baryon-to-photon ratio (or, equivalently, the present-universe baryon density parameter).  With one adjustable parameter and three relic abundances (four if \\3he is included), SBBN is an overdetermined theory, potentially falsifiable.  The current status of the comparison between predictions and observations reviewed here illuminates the brilliant success of the standard models. Among the relic light nuclides, deuterium is the baryometer of choice.  For N$_{\\nu} = 3$, the SBBN-predicted deuterium abundance agrees with the primordial-D abundance derived from the current observational data for $\\eta_{10} = 5.6^ {+0.6}_{-1.2}$ ($\\Omega_{\\rm B} = 0.020^{+0.002}_{-0.004}$). This baryon abundance, from the first 20 minutes of the evolution of the universe, is in excellent agreement with independent determinations from the CMB ($\\sim $~few hundred thousand years) and in the present universe ($\\sim $~10 Gyr).  It is premature, however, to draw the conclusion that the present status of the comparison between theory and data closes the door on further interesting theoretical and/or observational work.  As discussed in these lectures, there is some tension between the SBBN-predicted abundances and the relic abundances derived from the observational data. For the deuterium-inferred SBBN baryon density, the expected relic abundances of \\4he and \\7li are somewhat higher than those derived from current data.  The ``problems'' may lie with the data (large enough data sets? underestimated errors?) or, with the path from the data to the relic abundances (systematic errors? evolutionary corrections?).  For example, has an overlooked correction to the \\hii region-derived \\4he abundances resulted in a value of \\Yp which is systematically too small (\\eg underlying stellar absorption)?  Are there systematic errors in the absolute level of the lithium abundance on the Spite Plateau or, has the correction for depletion/dilution been underestimated?   In these lectures the possibility that the fault may lie with the cosmology was also explored.  In one simple extension of SBBN, the early universe expansion rate is allowed to differ from that in the standard model.  It was noted that to reconcile D, and \\4he would require a {\\it slower} than standard expansion rate, difficult to reconcile with simple particle physics extensions beyond the standard model.  Furthermore, if this should be the resolution of the tension between D and \\4he, it would exacerbate that between the predicted and observed lithium abundances.  The three abundances could be reconciled in a further extension involving neutrino degeneracy (an asymmetry between electron neutrinos and their antiparticles).  But, three adjustable parameters to account for three relic abundances is far from satisfying. Clearly, this active and exciting area of current research still has some surprises in store, waiting to be discovered by astronomers, astrophysicists, cosmologists and particle physicists.  The message to the students at this school -- and those everywhere -- is that much interesting observational and theoretical work remains to be done.  I therefore conclude these lectures with a personal list of questions I would like to see addressed.  \\begin{itemize}  \\item Where (at what value of D/H) is the primordial deuterium plateau, and what is(are) the reason(s) for the currently observed spread among the high-$z$, low-Z QSOALS D-abundances?  \\item Are there stellar observations which could offer complementary insights to those from \\hii regions on the question of the primordial \\4he abundance, perhaps revealing unidentified or unquantified systematic errors in the latter approach?  Is \\Yp closer to 0.24 or 0.25?  \\item What is the level of the Spite Plateau lithium abundance? Which observations can pin down the systematic corrections due to model stellar atmospheres and temperature scales and which may reveal evidence for, and quantify, early-Galaxy production as well as stellar depletion/destruction?  \\item If further observational and associated theoretical work should confirm the current tension among the SBBN-predicted and observed primordial abundances of D, \\4he, \\7li, what physics beyond the standard models of cosmology and particle physics has the potential to resolve the apparent conflicts? Are those models which modify the early, radiation-dominated universe expansion rate consistent with observations of the CMB temperature fluctuation spectrum?  If neutrino degeneracy is invoked, is it consistent with the neutrino properties (masses and mixing angles) inferred from laboratory experiments as well as the solar and cosmic ray neutrino oscallation data?  \\end{itemize}   To paraphrase Spock, work long and prosper!"
    },
    "0208/astro-ph0208009_arXiv.txt": {
        "abstract": "We present {\\it Far Ultraviolet Spectroscopic Explorer} (\\FUSE) observations of the \\OVIll\\ absorption lines associated with gas in and near the Milky Way, as detected in the spectra of a sample of 100 extragalactic targets and 2 distant halo stars. We combine data from several \\FUSE\\ Science Team programs with guest observer data that were public before 2002 May 1. The sightlines cover most of the sky above galactic latitude $\\vert$$b$$\\vert$$>$25\\deg\\ -- at lower latitude the ultraviolet extinction is usually too large for extragalactic observations. We describe the details of the calibration, alignment in velocity, continuum fitting, and manner in which several contaminants were removed -- Galactic \\H2, absorption intrinsic to the background target and intergalactic \\Lyb\\ lines. This decontamination was done very carefully, and in several sightlines very subtle problems were found. We searched for \\OVI\\ absorption in the velocity range $-$1200 to 1200\\kms. With a few exceptions, we only find \\OVI\\ in the velocity range $-$400 to 400\\kms; the exceptions may be intergalactic \\OVI. In this paper we analyze the \\OVI\\ associated with the Milky Way (and possibly with the Local Group). We discuss the separation of the observed \\OVI\\ absorption into components associated with the Milky Way halo and componet at high-velocity, which are probably located in the neighborhood of the Milky Way. We describe the measurements of equivalent width and column density, and we analyze the different contributions to the errors. We conclude that low-velocity Galactic \\OVI\\ absorption occurs along all sightlines -- the few non-detections only occur in noisy spectra. We further show that high-velocity \\OVI\\ is very common, having equivalent width $>$65\\maa\\ in 50\\% of the sightlines and equivalent width $>$30\\maa\\ in 70\\% of the high-quality sightlines. The high-velocity \\OVI\\ absorption has velocities relative to the LSR of $\\pm$(100--330)\\kms; there is no correlation between velocity and absorption strength. We discuss the possibilities for studying \\OVI\\ absorption associated with Local Group galaxies, and conclude that \\OVI\\ is probably detected in M\\,31 and M\\,33. We limit the extent of an \\OVI\\ halo around M\\,33 to be $<$100\\kpc\\ (at a 3$\\sigma$ detection limit of log $N$(\\OVI)$\\sim$14.0). Using the measured column densities, we present 50\\kms\\ wide \\OVI\\ channel maps. These show evidence for the imprint of Galactic rotation. They also highlight two known \\HI\\ high-velocity clouds (complex~C and the Magellanic Stream). The channel maps further show that \\OVI\\ at velocities $<$$-$200\\kms\\ occurs along all sightlines in the region $l$=20\\deg--150\\deg, $b$$<$$-$30\\deg, while \\OVI\\ at velocities $>$200\\kms\\ occurs along all sightlines in the region $l$=180\\deg--300\\deg, $b$$>$20\\deg. ",
        "introduction": "The {\\it Far Ultraviolet Spectroscopic Explorer} (\\FUSE) provides high resolution spectra in the wavelength regime between 905 and 1187\\aa, making it one of the few observatories (past or present) that allows observations shortward of 1150\\aa\\ down to the Galactic Lyman edge. This spectral region contains the resonance absorption lines of the most abundant atoms and molecules including, for example, \\HI, \\H2, \\OI, \\OVI, \\CII, \\CIII, \\FeII\\ and \\FeIII. Interstellar absorption line observations with \\FUSE\\ enable studies of all phases of the interstellar gas, including the cold neutral and molecular medium, the warm neutral medium, the warm ionized medium and the hot ionized medium. \\FUSE\\ was launched in June 1999. The capabilities of \\FUSE\\ are described in detail by Moos et al.\\ (2000) and Sahnow et al.\\ (2000). \\par \\FUSE\\ data consist of 8 separate $\\sim$90\\aa\\ wide spectra, identified as LiF1A, LiF1B, LiF2A, LiF2B, SiC1A, SiC1B, SiC2A and SiC2B. The LiF1A, LiF2B, SiC1A and SiC2B spectra cover the region near 1030\\aa, which allows the study of absorption by five-times ionized oxygen (O$^{+5}$), a good diagnostic of gas at temperatures near 3\\tdex5~K. This ion has a high enough ionization potential (113.9~eV is required to convert O$^{+4}$ to O$^{+5}$) that it is not produced by photoionization caused by extreme ultraviolet radiation from normal stars, which is suppressed above energies of 54.4~eV because of the strong He$^+$ absorption edge in the stellar atmosphere. O$^{+5}$ has two strong resonance absorption lines at 1031.9261\\aa\\ and 1037.6167\\aa, with oscillator strengths of 0.133 and 0.066 (Morton 1991). \\par \\FUSE\\ is the first instrument with sufficient sensitivity {\\it and} spectral resolution to observe \\OVI\\ absorption using large numbers of extragalactic objects as background targets. The \\FUSE\\ Science Team designed a number of programs to systematically map the distribution and amount of Galactic \\OVI. Several programs concentrate on the Galactic disk, while others emphasize the Galactic halo. The halo program is divided into two parts -- a study of the vertical distribution of the \\OVI\\ using a set of stars located at different heights above the Galactic plane, and a study of the integrated column density using a sample of extragalactic targets. This paper describes the basic target information, the technical details of the data handling, and derived parameters for the extragalactic study. \\par We present a catalog and some general analyses of this dataset, with an emphasis on discriminating the different phenomena that are present. Detailed interpretations, with more emphasis on a physical understanding, are presented by Savage et al.\\ (2002) and Sembach et al.\\ (2002b). The former paper analyzes the \\OVI\\ absorption that is clearly associated with the Milky Way, which we also refer to as the ``thick disk'' \\OVI. It discusses in depth the angular and spatial distributions, the kinematics, the relation of \\OVI\\ to other components of the Galaxy and the implications for understanding the Galaxy as a whole. As we find in this catalog paper, high-velocity \\OVI\\ absorption is present along many sightlines. This turns out to sample varied phenomena, some of which may not be directly associated with the Milky Way. The physical interpretation of this aspect of the \\OVI\\ sky is discussed in detail by Sembach et al.\\ (2002b). \\par This paper is constructed as follows. First, we discuss the manner in which the sample of extragalactic objects was assembled (\\Sect\\Sdefsample). Table~\\Tbasic\\ lists the object and observation information. In \\Sect\\Sdata\\ we first summarize the reduction steps (\\Sect\\Sdatasum) and then describe the calibration (\\Sect\\Spipeline), the absolute alignment and relative alignment between different spectrograph segments (\\Sects\\SHI\\ and \\Svshift) and finally how different observations and segments were combined (\\Sect\\Scombine). Next we describe how the continuum was fit (\\Sect\\Scontinuum), how the final binning was determined (\\Sect\\SSoverN), and how contamination by Galactic molecular hydrogen (\\H2), intrinsic AGN lines, and intergalactic absorption was removed (\\Sects\\SHtcontam--\\Sgroupcontam). Section~\\Smeasure\\ details the process of measuring the \\OVI\\ lines. First, a velocity range of integration is determined (\\Sect\\Svrange), then we measure the equivalent width and study the distribution of the statistical and systematic errors in the equivalent width (\\Sect\\Seqvwidth). Next we discuss the column densities and compare the values derived from the \\OVIla\\ and \\OVIlb\\ lines to determine whether saturation could be important (\\Sect\\Scoldens). Scientific results are presented in \\Sect\\Sresults\\ -- the detection rate for Galactic and high-velocity \\OVI\\ (\\Sects\\SMWYdet--\\Sveldist), a check on the possibilities for detecting \\OVI\\ absorption from other Local Group galaxies using our sample (\\Sect\\SLocGrp) and \\OVI\\ channel maps (\\Sect\\Smaps). Finally, we add an Appendix, in which notes are presented concerning each sightline. ",
        "conclusions": " \\par 1) To align \\FUSE\\ spectra, it is necessary to use \\HI\\ 21-cm emission spectra in the target direction in order to determine the velocity of the peak absorption as the \\HI\\ does not always peak at 0\\kms\\ -- it may peak at any velocity between $\\sim$$-$60 and 20\\kms. Further, when using v1.8.7 of the calibration pipeline, the \\FUSE\\ wavelength scale appears to have shifts of up to $\\sim$10\\kms\\ between different regions of the same spectrum, so that it is necessary to align each absorption line individually. The wavelength calibration of v2.0.5 of the pipeline is much better, but we conclude that a comparison with \\HI\\ data is still necessary, and even then there may still be offsets up to 10\\kms, due to both the intrinsic accuracy and the possibility that the gas sampled by a broad \\HI\\ beam may differ from that in the narrow pencil beam toward the background target. \\par 2) For bright objects (flux $>$8\\tdex{-14}\\fu) \\FUSE\\ can obtain good spectra (Q$>$=3, S/N$>$9 per resolution element) in 15~ks. At flux levels of 4\\tdex{-14}\\fu\\ this requires 20~ks, while Q=2 (S/N=5) requires only 6~ks. For objects with a flux of 2\\tdex{-14}\\fu, a 10~ks integration time is needed to obtain a spectrum with Q=2 (S/N=5), from which reasonable \\OVI\\ information can be extracted after binning to 10 pixels, while good spectra (Q$>$=3, S/N$>$9) require an exposure time $>$30~ks. For faint objects (flux $\\sim$\\dex{-14}\\fu) the minimum exposure time for detecting \\OVI\\ at Q$>$=1 (S/N$>$3) is 10~ks, while Q$>$=2 (S/N$>$5) requires 25~ks or more and Q$>$=3 (S/N$>$9 per resolution element) requires $>$80~ks. For fainter objects even very long observations do not easily yield good spectra, because the background uncertainties start playing a role. The highest S/N ratio that has been achieved is $\\sim$30 per resolution element, both for a very bright object (vZ\\,1128, flux 60\\tdex{-14}\\fu, 31~ks) and a rather faint object (PG\\,1259+593, flux 1.8\\tdex{-14}\\fu, 633~ks). \\par 3) Galactic \\H2\\ in the $J$=0 and $J$=1 states is detected in 80\\% of the 102 directions with spectra of quality 1--4. The 19 non-detections cluster in the regions $l$=0--120\\deg, $b$$>$40\\deg, $l$=210--60\\deg, $b$$<$$-$40\\deg. The upper limit on $N$(\\H2) can be as low as \\dex{14}\\cmm2\\ in each rotational level. In 62 out of 102 sightlines, lines of $J$=3 or higher are detected. \\H2\\ turns out to be ubiquitous in intermediate-velocity clouds. We analyze this in a separate paper (Richter et al.\\ 2002). In 23 cases \\H2\\ contaminates the \\OVI\\ absorption profile, but we can correct for it in all but one case (NGC\\,3783). \\par 4) Contamination by intrinsic or intergalactic absorption occurs occasionally. In 16 sightlines this makes it impossible to measure the Galactic \\OVI\\ absorption. \\par 5) The Galactic \\OVI, which we also refer to as ``thick disk'' absorption, may extend as far $-$145 or 140\\kms\\ (Fig.~\\Fvlim), but in 88 sightlines it is confined to within $\\pm$120\\kms\\ and in 61 to within $\\pm$100\\kms. The average negative velocity limit is $-$90\\kms, while the average positive velocity limit is 90\\kms. \\par 6) We present measurements of the \\OVI\\ equivalent widths in the Galactic and high-velocity components, and calculate five contributions to the error: one associated with the random noise fluctuations in the spectrum, one associated with continuum fitting (placement), a fixed-pattern noise contribution, an error associated with the choice of integration range, and an error related to uncertainties in the determination of the parameters of contaminating \\H2\\ lines. The first two are combined into a statistical error, while the latter three give a systematic error. Comparing two methods of determining the error associated with the placement of the continuum (Fig.~\\Ferrcorr) shows that the $\\pm$rms/3 method of estimating continuum errors tends to overestimate these for flat continua, but gives reasonable answers for continua fit with a second-order polynomial. \\par 7) In the subsample of 20 sightlines where it is possible to compare the apparent optical depth profiles of the \\OVIla\\ and \\OVIlb\\ lines (Fig.~\\Foviratio), we find 17 components for which the \\OVI\\ absorption is not saturated, since the ratio of column densities is unity, to within the 1$\\sigma$ error. Of the remaining five components, minor saturation (ratio $\\sim$1.2) may occur in four, while clear saturation (ratio $\\sim$1.6) occurs in only one (Mrk\\,421). From this we conclude that integration of the apparent optical depth profile of the \\OVIla\\ line will yield a reliable column density in almost all cases, and will still be correct to within 30\\% in the maybe 5 total sightlines where some saturation occurs. \\par 8) We clearly detect Galactic \\OVIla\\ absorption in 91 out of 102 directions. In eleven directions we can only set an upper limit, and for five of these $N$(\\OVI) must be relatively low. With even slightly more sensitive data \\OVI\\ would probably have been found in the directions for which we now derive non-detections. The largest \\OVI\\ equivalent width found is 429$\\pm$12\\maa, toward PKS\\,2005$-$489, while the lowest detected equivalent width is about 80\\maa\\ (toward Mrk\\,1095 and NGC\\,7714). \\par 9) High-velocity \\OVIla\\ absorption stronger than 65\\maa\\ is detected in 48 of the 102 sightlines (47\\%) while high-velocity absorption stronger than 30\\maa\\ is found in 34 of the 49 sightlines with Q=3 or 4 (69\\%). In 13 sightlines a high-velocity feature as small as 20\\maa\\ could have been detected, and it is found in 12 of these (92\\%). \\par 10) We show that both weak and strong high-velocity \\OVI\\ absorption components occur at all velocities in the ranges $\\sim$$-$400 to $-$100 and 100 to 400\\kms\\ (Fig.~\\Fvelhist). The scatter plot of column density against velocity suggests that many absorbers similar to those seen at high-velocity may be blended with the Galactic absorption. \\par 11) We checked the sightlines that pass within a few degrees of Local Group galaxies. None of the dwarf irregulars, ellipticals and spheroidals appear to show associated \\OVI, nor was any expected because of the small sizes of these galaxies. Four sightlines are toward \\HII\\ regions in M\\,33. Since several M\\,33 OB stars have also been observed, it should be possible to study \\OVI\\ in M\\,33. Similarly, several M\\,31 OB stars have been observed. One QSO near M\\,31 is bright enough that an improved spectrum is possible. However, a study of \\OVI\\ in M\\,31 and M\\,33 is complicated by the fact that high-negative velocity \\OVI\\ absorption (at velocities expected for \\OVI\\ in those two galaxies) is present over a large part of the southern sky. \\par 12) For the case of M\\,33, its rotation curve predicts gas velocities near 0\\kms\\ or larger along the sightlines to Mrk\\,352, Mrk\\,357 and PG\\,0052+251, which pass 90, 96 and 133\\kpc\\ from M\\,33, respectively. No such absorption is seen, limiting the extent of an \\OVI\\ halo around M\\,33 to be $<$100\\kpc, at a 3$\\sigma$ detection limit of log $N$(\\OVI)$\\sim$14.0. \\par 13) A series of \\OVI\\ channel maps shows the imprint of differential Galactic rotation on the low-velocity absorption: in the $-$100 to $-$50\\kms\\ channel, $N$(\\OVI) is larger at longitudes $<$180\\deg, while in the 50 to 100\\kms\\ channel it is larger at longitudes $>$180\\deg. \\par 14) The $-$200 to $-$100\\kms\\ \\OVI\\ channel maps show that the HVC complex~C is clearly detected in \\OVI\\ absorption. The positive-velocity side of the Magellanic Stream ($l$$\\sim$270\\deg) is detected in the 150 to 300\\kms\\ channels although there is only one sightline where both \\HI\\ and \\OVI\\ are seen (Fairall\\,9). In all directions in the region $l$=20\\deg--150\\deg, $b$$<$$-$30\\deg\\ \\OVI\\ is detected at high negative velocities $<$$-$200\\kms, while elsewhere in the sky only upper limits can be set. Conversely, in the 150 to 300\\kms\\ channels high-positive velocity \\OVI\\ is common in the region $l$=180\\deg--300\\deg, $b$$>$20\\deg, and only upper limits are set elsewhere in the sky.   \\newpage"
    },
    "0208/astro-ph0208523_arXiv.txt": {
        "abstract": "Betelgeuse is an example of a nearby cool super-giant that displays temporal brightness fluctuations and irregular surface structures. Recent numerical simulations by Freytag and collaborators of the outer convective envelope comprising most of the entire star under realistic physical assumptions, have shown that the fluctuations in the star's apparent luminosity may be caused by giant cell convection, very dissimilar to solar convection. These detailed simulations bring forth the possibility of addressing another question regarding the nature of Betelgeuse and super-giants in general; namely whether these stars may harbor magnetic activity, which may contribute to their variability. Taking the detailed numerical simulations of the star at face value, we have applied a kinematic dynamo analysis to study whether or not the flow field of this super-giant may be able to amplify a weak seed magnetic field. We find that the giant cell convection does indeed allow a positive exponential growth rate of magnetic energy. The possible Betelgeusian dynamo can be characterized as belonging to the class of so-called ``local small-scale dynamos'' another often mentioned example of which is the dynamo action in the solar photosphere that may be responsible for the formation of small-scale flux tubes (magnetic bright points). However, in the case of Betelgeuse this designation is less meaningful since the generated magnetic field is both {\\em global} and {\\em large-scale}. ",
        "introduction": "While the cool super-giant star Betelgeuse ($\\alpha$ Orionis) is among the stars with the largest apparent diameters---corresponding to a radius in the range 600--800 ${\\rm R}_{\\odot}$---fundamental stellar parameters for this red M1--2 Ia--Iab star are by no means well-defined. Recently Freytag and collaborators (e.g.\\ Freytag, Steffen, \\& Dorch 2002) performed detailed numerical three-dimensional radiation-hydrodynamic simulations of the outer convective envelope and atmosphere of the star under realistic physical assumptions. They try to determine if its observed brightness variations may be understood by convective motions within the star's atmosphere: the resulting models are largely successful in explaining the observations as a consequence of giant-cell convection on the stellar surface, very dissimilar to solar convection. These detailed simulations bring forth the possibility of solving another question regarding the nature of Betelgeuse, and of super-giants in general; namely whether these stars may harbor magnetic activity, which in turn may also contribute to their variability. A possible astrophysical dynamo in Betelgeuse would most likely be very different from those thought to operate in solar type stars, both due to its slow rotation, and to the fact that only a few convection cells are present at its surface at any one time. ",
        "conclusions": "Based on the results presented here, we may not say conclusively if Betelgeuse {\\em does} have a magnetic field, of course. The results are tentative and should be used with caution. But we may say that it seems that it {\\em can} indeed have a presently unobserved magnetic field. The dynamo of Betelgeuse may be characterized as belonging to the class called ``local small-scale dynamos'' another example of which is the proposed dynamo action in the solar photosphere that may possibly be responsible for the formation of small-scale flux tubes (cf.\\ Cattaneo 1999, but also the discussion by Stein, ibid). However, in the case of Betelgeuse this designation is less meaningful since the generated magnetic field is both global and large-scale.  The future developments of this project will involve firstly using longer time sequences of the input flow field, to avoid having to rely on recycling. Secondly, simulations with higher numerical resolution is currently being carried out (see Freytag \\& Finnsson, 2002, and Freytag \\& Mizuno-Wiedner, 2002), which will allow larger runs with higher magnetic Reynolds numbers (i.e.\\ smaller magnetic structures can form). Thirdly, it will be a priority to apply a more realistic quenching expression to introduce the saturation (one that takes into account the relative inclinations of the field and the flow, e.g.\\ the cross-convergence). Lastly, more realistic boundary conditions, and a parameter study with increasing Re$_{\\rm m}$ will be performed. The final goal is to be able to identify the more of operation during the (non-linear) saturation phase of the dynamo at high Re$_{\\rm m}$ in order to predict the likely topology of the magnetic field that one might observe at stars such as Betelgeuse."
    },
    "0208/astro-ph0208245_arXiv.txt": {
        "abstract": "Under the assumptions that molecular clouds are nearly spatially and temporally isothermal and that the density peaks (``cores'') within them are formed by turbulent fluctuations, we argue that cores cannot reach a hydrostatic (or magneto-static) state as a consequence of their formation process. In the non-magnetic case, this is a consequence of the fact that, for cores at the same temperature of the clouds, the necessary Bonnor-Ebert truncation at a finite radius is not feasible, unless it amounts to a shock, which is clearly a dynamical feature, or the core is really embedded in hotter gas. Otherwise, quiescent cores must have non-discontinuous density profiles until they merge with their parent cloud, constituting extended structures. For these, we argue that any equilibrium configuration with non-vanishing central density is unstable. Since the cores' environment (the molecular cloud) is turbulent, no reason exists for them to settle into an unstable equilibrium. Instead, in this case, cores must be dynamical entities that can either be pushed into collapse, or else ``rebound'' towards the mean pressure and density as the parent cloud. Nevertheless, rebounding cores are delayed in their re-expansion by their own self-gravity. We give a crude estimate for the re-expansion time as a function of the closeness of the final compression state to the threshold of instability, finding typical values $\\sim 1$ Myr, i.e., of the order of a few free-fall times. Our results support the notion that not all cores observed in molecular clouds need to be on route to forming stars, but that instead a class of ``failed cores'' should exist, which must eventually re-expand and disperse, and which can be identified with observed starless cores. In the magnetic case, recent observational and theoretical work suggests that all cores are critical or supercritical, and are thus qualitatively equivalent to the non-magnetic case. This is, however, not a problem for the efficiency of star formation: within the turbulent scenario the low efficiency of star formation does not need to rely on magnetic support of the cores, but instead is a consequence of the low probability of forming collapsing cores in a medium that is globally supported by turbulence. Our results support the notion that the entire star formation process is dynamical, with no intermediate hydrostatic stages. ",
        "introduction": "\\label{sec:intro}  One of the most important goals in the study of star formation is to understand the state and physical conditions of the molecular cloud cores from which the stars form. The prevailing view concerning low-mass-star-forming cores is that they are quasi-static equilibrium configurations supported against gravitational collapse by a combination of magnetic, thermal and turbulent pressures (e.g., Mouschovias 1976a,b; Shu, Adams \\& Lizano 1987). When considering only thermal pressure, two variants of the equilibrium structures are usually discussed: either singular isothermal structures, with diverging central densities and smooth $r^{-2}$ density dependence extending to infinity (e.g., Shu et al.\\ 1987), or finite-central density structures, truncated at some finite radius and confined by the pressure of some external medium, generally assumed to be at higher temperatures and lower densities than the isothermal core (Ebert 1955; Bonnor 1956). More recently, the equilibria of non-axisymmetric configurations have also been studied (e.g., Fiege \\& Pudritz 2000; Curry 2000; Galli et al. 2001; Shadmehri \\& Ghanbari 2001; Lombardi \\& Bertin 2001; Curry \\& Stahler 2001).  The support from magnetic fields is generally included through the consideration of the mass-to-magnetic flux ratio of the core, since, assuming that the latter has a fixed mass, the flux freezing condition implies that its mass-to-flux ratio is constant (Chandrasekhar \\& Fermi 1953; Mestel \\& Spitzer 1956). Under isothermal conditions, the magnetic pressure and the gravitational energy scale as the same power of the core's volume; thus, self-gravity cannot overcome the magnetic support if the mass-to-flux ratio is smaller than some critical value, and collapse can only occur as the magnetic flux diffuses out of the cloud by ambipolar diffusion (see, e.g., Mestel \\& Spitzer 1956; Mouschovias \\& Spitzer 1976; Shu, Adams \\& Lizano 1987).  On the other hand, it is well established that the molecular clouds within which the cores form are turbulent, with linewidths that are supersonic for scales $\\gtrsim 0.1$ pc (e.g., Larson 1981), and with (magnetohydrodynamic) turbulent motions providing most of the support against gravity, with only a minor role of thermal pressure at all but the smallest ($\\lesssim 0.1$ pc) scales. Thus, there appears to be a conceptual gap between the turbulent nature of the clouds and the quasi-hydrostatic assumed nature of the cores. The cores in molecular clouds must be subject to global motions and distortions, as well as mass exchange with its surroundings (in general, to continuous ``morphing''), and, in fact, are likely to be themselves the turbulent density fluctuations within the clouds (von Weizs\\\"acker 1951; Bania \\& Lyon 1980; Scalo 1987; Elmegreen 1993; \\BP, \\VS\\ \\& Scalo 1999, hereafter BVS99; Padoan et al.\\ 2001). At present, one interpretation is that the cores are the dissipative end of the turbulent cascade, because the velocity dispersion within them becomes sonic or subsonic (e.g., Goodman et al.\\ 1998). However, in actuality, substructure is seen down to the smallest resolved scales (e.g., Falgarone, Puget \\& P\\'erault 1992), and appears even within what were previously considered to be ``smooth'' cores, as the resolution is improved (Wilner et al.\\ 2000). Also, inflow motions, themselves with substructure, are generally seen around these cores (e.g. Myers, Evans \\& Ohashi 2000). Moreover, if the transonic cores are part of a compressible cascade, they do not need to be the dissipative end of it, but may simply mark the transition to a regime of nearly incompressible turbulence (\\VS, \\BP \\& Klessen 2002, 2003).  This issue also poses a problem for the idea of confining clumps by turbulent pressure, since the latter is in general anisotropic and transient at large scales. In this regard, it is worth remarking that a frequent interpretation of the role of turbulent pressure in ``confining'' cores is that the total thermal-plus-turbulent pressure is larger outside a core than inside it, because the turbulent velocity dispersion increases with size. This is, however, an incorrect interpretation, as the dependence of turbulent pressure with size scale is a non-local property referring to statistical averages over domains of a given size, not to a gradient of the local value of the velocity dispersion as larger distances from the core's center are considered.  If the density peaks (clumps and cores) within molecular clouds have a dynamic origin, then an immediate question is whether they can ever reach hydrostatic equilibrium. Several pieces of evidence suggest that this is not possible. First, Tohline et al.\\ (1987) considered the potential energy curve of an initially gravitationally-stable fluid parcel in a radiative medium characterized by an effective adiabatic (or ``polytropic'') exponent, showing that it has a ``thermal energy barrier'' that must be overcome, say by an increase in the external turbulent ram pressure, in order to push the parcel into gravitational collapse. In particular, these authors estimated the Mach numbers required for this to occur. Although those authors did not discuss it, the production of a hydrostatic configuration within this framework would require hitting precisely the tip of such ``barrier'', the probability of which is vanishingly small, because the tips of potential barriers constitute unstable equilibria.  Second, although Shu (1977) has argued that the singular isothermal sphere is the state asymptotically approached by the flow as it seeks to establish detailed mechanical balance when its parts can communicate subsonically with one another, the maintenance of this configuration for long times seems highly unlikely, as this configuration constitutes an {\\it unstable} equilibrium, being the precursor of gravitational collapse. If the formation of the core is a dynamical process, no reason exists for the flow to relax onto an unstable equilibrium. Such a state can be used as an initial condition in simulations of gravitational collapse, but does not represent itself a realistic state that can be reached by a gas parcel in a turbulent medium.  Third, Clarke \\& Pringle (1997) have pointed out that cores cool mainly through optically thick lines, but are heated by cosmic rays, and therefore may be dynamically unstable, as velocity gradients may enhance local cooling.  Fourth, numerical simulations of self-gravitating, turbulent clouds (e.g., \\VS\\ et al.\\ 1996; Klessen, Heitsch \\& Mac Low 2000; Heitsch, Mac Low \\& Klessen 2001; Bate et al.\\ 2002) never show the production of hydrostatic objects. Instead, once a fluid parcel is compressed strongly enough to become gravitationally bound, it proceeds to collapse right away. Specifically, BVS99 suggested that hydrostatic structures cannot be formed by turbulent compressions in polytropic flows, in which the pressure is given by $P \\propto \\rho^\\gamma$, where $\\rho$ is the mass density and $\\gamma$ is the effective polytropic exponent. This is because the collapse of an initially{\\it  stable} gas parcel can only be induced (i.e, the parcel made unstable) by a (strong enough) mechanical compression if $\\gamma < \\gamma_{\\rm c}$, where the value of $\\gamma_{\\rm c}$ depends on the dimensionality of the compression and the specific heat ratio of the gas (see, e.g., \\VS, Passot \\& Pouquet 1996). However, once collapse has been initiated, it cannot be halted unless $\\gamma$ changes in the process, to become larger than $\\gamma_{\\rm c}$ again. In other words, for non-isothermal situations, with $\\gamma > \\gamma_{\\rm c}$, equilibria can be found even if the external pressure is time variable.  This is why stars can be formed as stable entities from highly anisotropic, dynamic, time-dependent accretion (Hartmann, \\BP\\ \\&  Bergin 2001).  For systems that are much closer to isothermal, such as molecular cloud cores, the boundary pressures are indispensible in establishing stable equilibria, which are therefore not expected to exist in an isothermal turbulent medium with a fluctuating ram pressure. In fact, an analysis of the energy contents of the clouds in numerical simulations shows that they are in near energy equipartition but nowhere near virial {\\it equilibirum} (VE) (see Ballesteros-Paredes \\& V\\'azquez-Semadeni 1995; 1997; Shadmheri, V\\'azquez-Semadeni \\& Ballesteros-Paredes 2003). This suggests that observations of rough energy equipartition (e.g., Myers \\& Goodman 1988) does not necessarily imply that clouds are in such detailed mechanical balance.  The only case when numerical studies show the formation of (magneto)static structures occurs in simulations of super-Jeans, yet subcritical clouds (e.g., Ostriker, Gammie \\& Stone 1999), in which the whole box is subcritical. However, as we discuss in \\S \\ref{sec:magn}, we believe that this is an artifact of the simulations being performed in closed boxes that do not allow further mass accretion until the system becomes supercritical.  In this paper, we provide further arguments against the possibility of molecular cloud cores being hydrostatic entities, and argue in favor of them being instead transients, although with low (subsonic) internal velocity dispersion. The plan of the paper is as follows: In \\S \\ref{sec:trunc_ext} we argue against the possibility of truncated Bonnor-Ebert-type configurations arising in nearly single-temperature molecular clouds, suggesting instead that cores must either be shock-confined or else have smooth(extended) density profiles, and then discuss the stability of extended structures, noting that unstable equilibria are not expected to arise in turbulent media. In \\S \\ref{sec:re-exp} we give a crude estimate of the re-expansion time of density peaks (cores) that are not sufficiently compressed to undergo gravitational collapse. In \\S \\ref{sec:magn} we then discuss the magnetic case, arguing that the subcritical case is also just a transient, on the basis of previous results existing in the literature. Then, in \\S \\ref{sec:dyn_scen} we discuss how the proposed dynamical nature of the cores is not inconsistent with observations, and finally, in \\S \\ref{sec:conclusions}, we summarize our results and give some conclusions. ",
        "conclusions": "\\label{sec:conclusions}  In this paper we have argued that the final state of isothermal fluid parcels compressed into ``cores'' by turbulent velocity fluctuations cannot remain in equilibrium. In the non-magnetic case, this is due to the isothermality of the flow, which implies that the a continuous pressure profile requires a continuous density profile, except if it is supplemented by ram pressure in a shock. Since shocks are already non-hydrostatic features, thus agreeing with our claim, we focus on continuous-profile (``extended'') structures. For these, we argued that all equilibrium configurations are unstable, contrary to the stability range found for truncated BE-type structures, and thus are not expected to arise in a dynamic, fluctuating medium. Thus, cores must in general collapse or re-expand, but cannot remain in equilibrium, unless they happen to enter (or be ``captured'' in) a hotter region, as is the case of the much-discussed B68 globule. In the magnetic case, we have recalled several recent results suggesting that all cores are critical or supercritical, thus being qualitatively equivalent to the non-magnetic case regarding their possibility of collapse.  Although our arguments are conceptually very simple, we believe they have been overlooked in the literature because the hydrostatic state is normally considered as an {\\it initial} condition, accepted without questioning how such state can be arrived at, and because the turbulent pressure is implicitly assumed to be ``microscopic'' (i.e., of characteristic scales much smaller than the core), neglecting the fact that molecular clouds are globally turbulent and that the bulk of the turbulent energy is at the largest scales, as clearly suggested by the observed velocity dispersion-size scaling relation (Larson 1981), implying that the cores themselves {\\it are} the turbulent density fluctuations.   Our results have the implication that many observed cores are not on route to forming stars, but instead ``fail'', and must re-expand and merge back into the general molecular cloud medium. For these, the re-expansion time is expected to be larger than the compression time due to the retarding action of self-gravity. A simple estimate based on virial balance suggests that the re-expansion time is of the order of a few free-fall times. This is consistent with the facts that molecular clouds typically contain more starless than star-forming cores (e.g., Taylor, Morata \\& Williams 1996; Lee \\& Myers 1999; see also Evans 1999 and references therein), and that most of the cores do not appear to be gravitationally bound (e.g., Blitz \\& Williams 1999).  It is worthwhile to note that these time scales are over one order of magnitude shorter than estimates based on ambipolar diffusion (see, e.g., McKee et al.\\ 1993).  Indeed, the long ambipolar diffusion time scales were necessary to explain the low efficiency of star formation in the old hydrostatic paradigm, but in the dynamic scenario of star formation, the low efficiency is a natural consequence that only a small fraction of the mass in a molecular cloud is deposited by the turbulence in collapsing cores (Padoan 1995; \\VS\\ et al.\\ 2002, 2003), and does not need to rely on magnetic support of the cores.   We thus suggest that hydrostatic configurations have no room in the process of star formation in turbulent, isothermal molecular clouds. Theories of core structure and star formation should consider the fact that core formation is a dynamical process. This probably implies that the density profile in cores is a function of time, and therefore {\\it not unique}. This may be in agreement with the fact that recent surveys find {\\it distributions} of the scaling exponent, rather than clearly defined unique values (e.g., Shirley et al.\\ 2002). Another implication is that fundamental properties like the star formation efficiency may be {\\it statistical} consequences of the turbulence in molecular clouds (Elmegreen 1993; Padoan 1995; V\\'azquez-Semadeni et al.\\ 2002), rather than depending on ambipolar diffusion to break the equilibrium state."
    },
    "0208/astro-ph0208073_arXiv.txt": {
        "abstract": "There has been a trend in the past decade to describe the large-scale structures in the Universe as a (multi)fractal set. However, one of the main objections raised by the opponents of this approach deals with the transition to homogeneity. Moreover, they claim there is not enough sampling space to determine a scaling index which characterizes a (multi)fractal set. In this work we propose an alternative solution to this problem, using the generalized thermostatistics formalism. We show that applying the idea of nonextensivity, intrinsic to this approach, it is possible to derive an expression for the correlation function, describing the scaling properties of large-scale structures in the Universe and the transition to homogeneity, which is in good agreement with observational data. ",
        "introduction": "One of the key problems in structure formation theories is the issue of the galaxy distribution and the transition to homogeneity and isotropy of the Universe in large enough scales. Historically, the hypothesis of homogeneity (the Cosmological Principle, CP) was introduced by Einstein to find simple solutions of the field equations for the case where the spatial hypersphere of the Universe is a maximally symmetric subspace of the space-time. That allows us to derive the Robertson-Walker metric and the Friedmann equations, the theoretical framework where cosmology has been developed. At the same time, the CP implies that all mass units should be statistically equivalent, with a Poisson distribution in space. In this case, correlations may appear only on average and should be the same when viewed from any point in the system. Actually, it has been shown that luminous matter in the Universe shows a quite structured distribution of galaxies and voids. This structure has led some authors [1] to tentatively describe galaxy clustering  using a fractal distribution with a dimension $\\sim 1.2$. The results of a number of redshift surveys [2,3] indicate that, indeed, there is a certain hierarchy in the Universe, with stars forming galaxies, galaxies grouping themselves in groups, clusters and superclusters. A rough limit to this clustering is seen at a distance of about 200 $h^{-1}$ Mpc and the present redshift surveys do not show evidence of large-scale structures beyond this scale. Particularly, in very large scales - those probed by cosmic microwave background (CMB) experiments, especially by the COBE-DMR experiment - there is no evidence of violation of local isotropy [4]. However, we should note that the existence of such a crossover towards homogenization, as well as the exact value of the fractal dimension, are questions of intense debate [5-7].  The fractal hypothesis is deeply connected with a topology theorem which states that homogeneity is implied by the condition of local isotropy plus the assumption of analicity (or regularity) for the distribution of matter. It is possible to prove that in a fractal the condition of local isotropy can be satisfied but, since a fractal is a non-analytic structure, the property of homogeneity is not implied [8]. This means that the fractal scenario is incompatible with the CP. Observations point out that galaxy structures indeed exhibit fractal properties up to some scale, although a pure fractal description does not seem to be favored at the moment [9]. Recently, some authors (e.g. [10-12]) have had some success in describing the clustering properties of visible matter in the Universe in terms of a multifractal phenomenon associated with density thresholds applied to multifractal sets. They show that both a hybrid fractal and a multifractal approaches can reasonably describe the matter distribution up to $\\approx~100h^{-1}$ Mpc.  In spite of the relative success of the multifractal approach to describe the distribution of matter in the Universe, it is important to understand the physics behind this framework. In general terms, the multifractal description of galaxies may represent a strange attractor that is the nonlinear outcome of the dynamical equations of gravitational galaxy clustering [13,14]. However, it is not simple to find a dynamical connection between fractal sets  and galaxy clustering. In this work we propose an alternative solution to this problem, using the generalized thermostatistics (GTS) formalism [15]. We show that applying the idea of nonextensivity, intrinsic to GTS, it is possible to derive an expression for the correlation function, describing the scaling properties of large-scale structures in the Universe.  The present approach is based on the assumption of a scale dependent correlation dimension ($D_2$), which leads to a reconciliation with observational data at various scales, showing  a smooth transition from a clustered, fractal Universe to large-scale homogeneity, with $D_2 = 3$.  The physical motivation behind our approach is the peculiar behavior of large-scale gravitational systems, dominated by the unshielded, long-range nature of gravity. In contrast, other many-body systems, like neutral gases and plasmas, are characterized by short-range interactions. Because of this fundamental difference, the standard Boltzmann-Gibbs statistical mechanics cannot be applied to gravitating systems, since the long-range nature of gravity strongly violates one of its basic premises (short-ranged effective interactions), and, thus, special techniques are needed [16]. One possibility to overcome this difficulty -- adopted by us -- is to use the GTS, a theory proposed to correct Boltzmann-Gibbs statistical mechanics in those cases where its prescriptions fail. ",
        "conclusions": "\\begin{figure}[ttt] \\begin{center} \\includegraphics[width=10cm]{fig1.ps} \\end{center} \\caption{The two-point correlation function versus scale, for the Stromlo-APM, the Las Campanas and the ESP redshift surveys [3], and for the present model, for $a=0.65$ and $\\beta=1.0$ (solid line); $a=1.60$ and $\\beta=0.8$ (dashed line); $a=0.28$ and $\\beta=2.0$ (long dashed line); with $L=100 h^{-1}$ Mpc for all cases. For comparison purposes, a purely fractal description ($D_2=2$) and a homogeneous scenario ($D_2=3$) are also displayed (dotted straight lines). } \\end{figure}  We compared our model against observational data from various redshift surveys [2,3]. Results are shown in Figs. 1 and 2, for $a=0.65$ and $\\beta=1.0$ (solid line); $a=1.60$ and $\\beta=0.8$ (dashed line); $a=0.28$ and $\\beta=2.0$ (long dashed line); with $L=100 h^{-1}$ Mpc for all cases. For comparison purposes, a purely fractal description ($D_2=2$) and a homogeneous scenario ($D_2=3$) are also depicted in Fig. 1 (dotted straight lines). We observe that for $\\beta = 1$, which corresponds to the simplest relation between $q$ and $r$, our results show a good agreement with observational data for both the two-point correlation function and the correlation dimension. Increasing $\\beta$, we may obtain a better fitting for $1+\\xi(r)$ at the expense of the correlation dimension data. The opposite is true for decreasing values of $\\beta$.   \\begin{figure}[ttt] \\begin{center} \\includegraphics[width=10cm]{fig2.ps} \\end{center} \\caption{The correlation dimension versus scale, for various surveys [2], and for the present model, for $a=0.65$ and $\\beta=1.0$ (solid line); $a=1.60$ and $\\beta=0.8$ (dashed line); $a=0.28$ and $\\beta=2.0$ (long dashed line); with $L=100 h^{-1}$ Mpc for all cases.} \\end{figure}  Equations (\\ref{corr2}) and (\\ref{ca3}) offer a quantitative description of matter clustering in the Universe, and of the smooth transition from small-scale nonextensive fractal behavior to large-scale extensive homogeneity. From a geometrical point of view, our  model shows a Universe displaying a clear hierarchy, with predominance of point-like ($D_2 \\sim 0$) and filamentary ($D_2 \\sim 1$) structures at small scales, and surface-like ones ($D_2 \\sim 2$) at intermediate scales. For sufficiently large scales ($r > 500 h^{-1} Mpc$), the homogeneity predicted by the Cosmological Principle is recovered. All these features of the present approach are in good agreement with observational data.  Summarizing, we may say that our primary motivation in this work was to investigate the prediction of multiscaling and nonextensivity of large-scale structures in the Universe within the context of the generalized thermostatistics formalism. The results presented above suggest that we cannot discard this theoretical framework as a viable way to explain the gravitational clustering in the Universe. However, we should be careful and keep in mind two important caveats: the poorness of observational data in some scale domains may be masking the results, and the fact that luminous galaxies may not perfectly trace the mass distribution (the biasing problem). From a theoretical point of view, what is missing at the moment is a detailed understanding of how $q$ vary with scale. In the present work, we adopted the simplest model that provided a good agreement with the data. We are currently running a number of COBE-normalized CDM simulations for some relevant cosmological models [25]. We believe that further study on the resulting correlations and velocity distributions, at different scales (from galaxies to superclusters), may reveal the connection between the entropic parameter and the dynamics of mass clustering in the Universe.  \\ack{This work was partially supported by FAPESP and CNPq-Brazil.}"
    },
    "0208/astro-ph0208590_arXiv.txt": {
        "abstract": "We propose that the internal energy of the GRB blast waves, thought to be stored in the form of relativistic protons co-moving with the blast wave, is converted explosively (i.e. on light crossing time scales) into relativistic electrons of the same Lorentz factor, which are responsible for the production of observed prompt \\g-ray emission of the burst. This conversion is the result of the combined effects of the reflection of photons produced within the flow  by upstream located matter, their re-interception by the blast wave and their eventual conversion into $e^+e^--$pairs in interactions with the relativistic protons of the blast wave (via the $p \\gamma \\rightarrow e^+e^-$ reaction). This entire procedure is contingent on two conditions on the relativistic protons: a kinematic one imposed by the threshold of the $p \\gamma \\rightarrow e^+e^-$ reaction and a dynamic one related to the column density of the post shock matter to the same process. This latter condition is in essence identical to that of the criticality of a nuclear pile, hence the terminology. It is argued that the properties of relativistic blast waves operating under these conditions are consistent with GRB phenomenology, including the recently found correlation between quiescence periods and subsequent flare fluence. Furthermore, it is shown that, when operating near threshold, the resulting GRB spectrum produces its peak luminosity at an energy (in the lab frame) $E \\simeq m_ec^2$, thereby providing an answer to this outstanding question of GRBs. ",
        "introduction": "\\label{sect:intro}  The longstanding issue of the distance and absolute luminosity of GRBs has been settled in the past decade through the observational evidence collected  by $BATSE$ (Meegan et al. 1992) and $BeppoSAX$ (Costa et al. 1997), while the theoretical work of M\\'esz\\'aros \\& Rees (1992) and Rees \\& M\\'esz\\'aros (1992) provided the broader physical framework into which these events seem to generally fit. This framework associates GRBs with radiation emitted by relativistic blast waves (hereafter RBWs), produced by an unspecified todate agent, presumably associated with the formation of a neutron star or black hole. While the source of the energy associated with GRBs has remained uncertain, there remains little doubt about the presence of the RBWs, which power also the later time emissions at X-ray, optical and radio frequencies, known collectively as GRB afterglows (see review of Piran 1999).  With the discovery of GRB afterglows, much of the theoretical activity has since shifted to the study of the physics of these later time emissions. Nonetheless, a number of issues associated with the prompt \\g-ray emission, besides the nature of their energy source, still remain open. Chief among them are the conversion of the RBW kinetic energy into radiation and the fact that the frequency at which the GRB luminosity peaks, $E_{\\rm p}$, is narrowly distributed around a value intriguingly close to the electron rest mass energy. The purpose of the present note is to describe a process that  provides a ``natural\" account of these generic, puzzling GRB features.  Following the work of Shemi \\& Piran (1990) it has been generally accepted that a certain amount of baryons must be carried off with the blast waves responsible for the GRBs. This baryon contamination has even been deemed necessary, else the entire blast wave internal energy would be converted into radiation on very short time scales, leading to events of very different temporal and spectral appearance (e.g. Paczy\\'nski 1986) than observed in  GRBs . While the low radiative efficiency of baryons is essential for the GRB energy transport to the requisite distances ($\\gsim 10^{16}$ cm), it becomes problematic when demanded that their internal energy be radiated away on the short time scales associated with the GRB prompt \\g-ray emission. Generally, this issue is sidestepped by appealing to an unknown process which transfers the proton energy into electrons (Dermer, B\\\"ottcher \\& Chiang 1999), whose radiative evolution could then be accurately computed.  The narrow range of the GRB $\\nu F_{\\nu}$ spectral peak energy, $E_{\\rm p}$, is another well documented systematic feature of these events, a result of the extensive data base accumulated by $BATSE$. The compilation of Malozzi et al. (1995) shows clearly a preference for an energy $E_{\\rm p} \\simeq 200$ keV at which the $\\nu F_{\\nu}$ GRB spectra exhibit a maximum. In fact, it is precisely this maximum in the spectral energy distribution that  qualifies GRBs as such. Furthermore, when corrected for the redshift ($z_{\\rm GRB} \\sim 1$), $E_{\\rm p}$ shifts close to the electron rest mass. While a compelling explanation of this fact is presently lacking, several accounts have occasionally been proposed. For example, Brainerd (1994) argues that this is the result of down-Comptonization of a power law photon distribution that extends to $E \\gg 1$ MeV by cold matter with Thompson depth $\\tau_T \\sim 10$, an explanation possibly in conflict with the timing properties of GRBs (see e.g. Kazanas, Titarchuk and Hua 1997). The association of the GRB emission with relativistically boosted synchrotron radiation from RBWs has made this particular issue far more acute, as the energy of the latter should scale like $E_{\\rm p} \\propto \\G^4$. Therefore, even very small variations in the values of \\G~ would lead to a very broad range in the values of $E_{\\rm p}$. Dermer et al. (1999) proposed that the observed distribution is the result of the time evolution of a blast wave with a specific baryon loading, which when convolved with the triggering criteria of existing detectors favors the detection of fireballs with $E_{\\rm p}$ in the observed range. On the other hand, on the basis of analysis of SMM data,  Harris \\& Share (1999) have argued that there is no apparent excess of GRBs with $E_{\\rm p} \\gg 1$ MeV, thus leaving this issue open.  The present paper is structured as follows: In \\S 2 we outline the fundamental notion behind our proposal for converting the RBW proton energy into radiation, we derive the associated necessary conditions and discuss its relation to GRB phenomenology. In \\S 3 we produce model spectra based on this proposal and indicate their relation to the particular value of $E_{\\rm p}$ observed. Finally, in \\S 4 the results are discussed and certain conclusions are drawn. ",
        "conclusions": "We have presented above a novel model for the prompt emission of GRBs. We believe that this model provides some of the missing physics between the RBW proposal, which describes successfully the GRB energetics and time scales and the prompt emission of radiation (\\g-ray as well as optical-UV), by a well defined mechanism for tapping the kinetic energy stored in the RBW baryonic component. Furthermore, the same physics employed in effecting the conversion of baryon kinetic energy into radiation is instrumental in producing a peak in the spectral energy distribution at $E_{\\rm p} \\simeq 1$ MeV, thus providing a ``natural\" account of this GRB feature, an issue that has actually gotten even more puzzling with the advent of the RBW model for GRB. In addition to this emission, the model implies the presence of \\g-ray emission at higher energies, namely $E \\simeq \\G^2 m_ec^2$, a fact supported by some observations already, but which will be explored in greater depth by SWIFT and GLAST, missions which will be able to test and put meaningful constraints on this specific model.  One of the fundamental features of this model is the presence of the combination of kinematic and dynamic thresholds; this combination, along with the presence or absence of the necessary  ``mirror\", make the model inherently time dependent. At the same time, the kinematic threshold provides (for the first time to our knowledge) a regulating mechanism that puts one of the peaks in the $\\nu F_{\\nu}$ distribution very close to the observed value, despite the motion of the emitting plasma with Lorentz factors of several hundreds. Concerning the nature of the ``mirror\" we are willing to speculate here that, because the main ``reflected\" component consists of the prompt synchrotron photons which are emitted at optical frequencies, one could use atomic cross-sections ($\\simeq 10^{-16}~ {\\rm cm}^2$) to estimate their reflected fraction $\\alpha$; values of $n$, $R$ typical to those associated with GRBs yield $\\alpha \\simeq 0.01 - 1$. This is a very rough estimate, because it ignores the ionization of the reflecting medium. A more detailed treatment including these effects is beyond the scope of the present work.  We would like to thank Jay Norris, Rob Preece and Brad Shaffer for stimulating discussions and much information on GRB phenomenology."
    },
    "0208/astro-ph0208303_arXiv.txt": {
        "abstract": "In this paper we present a model for the short ($< 1$ second) population of gamma-ray bursts.  In this model heated neutron stars in a close binary system near its last stable orbit emit a large amount of neutrinos ($\\sim 10^{53}$ ergs).  A fraction of these neutrinos will annihilate to form an $e^+e^-$ pair plasma wind which will, in turn, expand and recombine to photons which make the gamma-ray burst. We study neutrino annihilation and show that a substantial fraction ($\\sim 1/2$) of energy deposited into $e^+e^-$ pairs comes from inter-star neutrinos, where each member of the neutrino pair originates from each neutron star.  Thus, in addition to the annihilation of neutrinos blowing off of a single star, there is a new source of baryon-free plasma that is deposited between the stars.  To model the $e^+e^-$ pair plasma wind between stars, we do three-dimensional relativistic numerical hydrodynamic calculations. We find that the time scale for these bursts, deriving from the baryon-free plasma, is less than one second and they will have a hot spectrum $\\sim 5$ MeV.  The energy in bursts is the order of $10^{52}$ ergs. ",
        "introduction": "In \\citet{swm01} we investigated a model for gamma-ray bursts (GRBs) deriving from a neutrino burst of energy $\\sim 10^{53}$ ergs above a heated, collapsing neutron star in a binary.  Such conditions have been suggested by numerical relativistic hydrodynamic simulations \\citep{mw00} of compression, heating and collapse of binary neutron stars near their last stable orbit.  Such a thermal neutrino burst was found to partially recombine via $\\nu\\overline{\\nu} \\rightarrow e^+e^-$ into an $e^+e^-$ pair plasma which expands relativistically. This fireball then recombines into photons via $e^+e^- \\rightarrow \\gamma\\gamma$ and was found to give a gamma-ray burst of energy $\\sim 10^{51} - 10^{52}$ ergs with spectral and temporal characteristics consistent with observations.  This paper elaborates on the previous work in three distinct ways. 1) The latest general relativistic hydrodynamic calculations by Wilson and Mathews indicate that evolution and compression of the stars in a binary is faster than previously thought; $\\lesssim 1$ sec, thus making this process a more natural candidate for the generation of ``short'' GRBs ($\\lesssim 1$ sec). 2) Herein we use recent work by \\citet{sw01} which calculates the neutrino annihilation rate {\\it between} two neutron stars in addition to the annihilation rate from a single star \\citep{sw99}.  This `inter-star' annihilation will not drive a baryon wind from the stellar surface and thus we find no baryon loading in the $e^+e^-$ pair plasma that originates between the neutron stars.  3) We employ three dimensional (3D) relativistic hydrodynamic calculations to model the flow of the wind in the complex, rotating, strong gravity environment around the neutron stars.  It has been known for some time that the population of GRBs is bimodal, with approximately a third of bursts having durations less than 2 seconds, and these short bursts typically have harder spectra than those bursts lasting longer than 2 seconds \\citep{kmfb+93}.  This bimodality is thought to indicate that short and long bursts are separate populations and mechanisms.  This view is bolstered by evidence that spectral break energy \\citep{ppbm01} and pulse lags \\citep{nsb00} have a discontinous jump between short and long populations.  The lack of discovery and location of an afterglow associated with short bursts has severely impeded progress in determining the distance scale and thus the energetics of these bursts. In fact, comparisons of BeppoSAX and BATSE data archives indicates a dearth of X-ray afterglows \\citep{g+01}.  Analysis of BATSE data for short burst decay tails give conflicting results indicating both a lack of an underlying afterglow component \\citep{c00} and the existence of such a component \\citep{lrrg01}. ",
        "conclusions": "A key result of the 3D numerical simulations (Fig.~\\ref{fig:EoD}) is that the emission from this system is bimodal: about half of the total energy is deposited as a pure, baryon-free, $E/D \\rightarrow \\infty$, pair plasma along a `fan' of angular half-width $\\theta = 15^\\circ$ along the symmetry plane between the neutron stars, and the other half of the total energy is blown off of the neutron stars as a wind with $E/D \\approx 300$.  Very little of the energy is deposited in intermediate regimes of $E/D$.  As such we analyse the expected observations from this model.  The solid angle subtended by the fan of baryon-free plasma is $\\Omega_{fan} = 4\\pi\\sin \\theta$ and the wind blows off into the rest of space, $\\Omega_{wind} = 4\\pi(1-\\sin\\theta)$.  Therefore the fluence from each component at a distance $R$ for total energy $\\EuScript{E}$ is \\begin{equation} \\begin{split} F_{fan} &= \\frac{\\xi \\EuScript{E}/2}{\\Omega_{fan} R^2} = \\frac{\\xi \\EuScript{E}}{8\\pi\\sin\\theta R^2} \\\\ F_{wind} &= \\frac{\\xi \\EuScript{E}/2}{\\Omega_{wind} R^2} = \\frac{\\xi \\EuScript{E}}{8\\pi(1-\\sin\\theta) R^2} \\end{split} \\end{equation} where $\\xi$ is an efficiency factor.  Now the system is rotating so an observer with angle, $\\psi$, from the rotation axis will see a mixture of the fan and wind, unless he is located within an angle $\\psi < \\theta$.  Since the period of rotation of the system is much smaller than the energy deposition timescale, $\\Delta t = 0.1$ s, we average over the fluences from the wind and the fan to get a total average fluence.  The proportion of a rotation subtended by the fan is $2/\\pi \\arcsin(\\theta/\\sin\\psi)$ for $\\theta \\lesssim \\psi \\leqslant \\pi/2$ and is unity for $0 < \\psi < \\theta$.  So the average fluence contributed by the fan as a function of viewing angle is (Fig.~\\ref{fig:fanwind}) \\begin{equation} \\overline{F}_{fan}(\\psi) = \\frac{\\EuScript{E}}{8\\pi R^2} \\begin{cases} \\frac{2/\\pi \\arcsin(\\theta/\\sin(\\psi))}{\\sin\\theta}& \\text{for $\\theta \\lesssim \\psi \\leqslant \\pi/2$} ~,\\\\ \\frac{1}{\\sin\\theta}&  \\text{for $0 < \\psi < \\theta$} ~, \\end{cases} \\label{eqn:Ffan} \\end{equation} and that of the wind is \\begin{equation} \\overline{F}_{wind}(\\psi) = \\frac{\\EuScript{E}}{8\\pi R^2} \\begin{cases} \\frac{1 - 2/\\pi \\arcsin(\\theta/\\sin(\\psi))}{1 - \\sin\\theta}& \\text{for $\\theta \\lesssim \\psi \\leqslant \\pi/2$} ~,\\\\ 0& \\text{for $0 < \\psi < \\theta$} ~. \\end{cases} \\label{eqn:Fwind} \\end{equation} The total average fluence is \\begin{equation} \\overline{F}(\\psi) = \\overline{F}_{fan}(\\psi) + \\overline{F}_{wind}(\\psi) ~. \\end{equation} An observer within $0 < \\psi < \\theta$ sees only the fan emission. This is effectively a jet of opening half-angle $\\theta = 15^\\circ$. Such an observer sees a jet of strong thermal emission over a timescale of 0.1 s.  For $\\theta = 15^\\circ = \\pi/12$ we have \\begin{equation} \\begin{split} \\overline{F}(\\psi = 0) &= 1.6 \\times 10^{-4}~ \\frac{\\xi E_{52}}{R_{Gpc}^2} \\quad \\text{ergs cm}^{-2} \\\\ \\overline{F}(\\psi = \\pi/2) &= 7.3 \\times 10^{-5}~ \\frac{\\xi E_{52}}{R_{Gpc}^2} \\quad \\text{ergs cm}^{-2} \\end{split} \\end{equation} so an observer along the jet ($\\psi < \\theta$) will see roughly twice the fluence in a short, $0.1$ s, thermal, $T_{obs} \\approx 5$ MeV, gamma-ray burst, than an off-axis observer ($\\psi \\sim \\pi/2$) who will see a synchrotron burst of duration $\\sim 3$ s with characteristic photon energy 200 keV.   \\begin{figure} \\plotone{f6.eps} \\caption{ The top figure shows the average fluence $\\overline{F}$ of the fan (Eqn.~\\ref{eqn:Ffan}) and the wind (Eqn.~\\ref{eqn:Fwind}) as a function of viewing angle $\\psi$.  The bottom figure shows number fluence, $\\overline{F}$ divided by the characteristic photon energy for the fan, $\\overline{\\epsilon}_{fan} \\approx 5$ MeV (Fig.~\\ref{fig:Tobs}), and the wind, $\\overline{\\epsilon}_{wind} \\approx 200$ keV (Sec.~\\ref{sec:gammaschem}).  This demonstrates that most of the energy comes from the fan while most of the photons come from the wind.  A viewer within $\\psi < 15^\\circ = \\pi/12$ will see only the fan emission, which effectively constitutes a jet of half-opening angle $15^\\circ$.} \\label{fig:fanwind} \\end{figure}   In conclusion, when the binary system is viewed within $15^\\circ$ of the axis of rotation, we have a model for short gamma-ray bursts, $\\sim 0.1$ s, which have hard spectra, several MeV.  In addition, when the binary system is viewed at angles larger than $15^\\circ$ from the rotation axis, our model yields a second group of bursts, 30 times more common, with soft spectra, a few hundred keV, and intermediate time duration of 1 to 5 s.  The energy available depends on the neutron star equation of state and the masses of the neutron stars. The energy in a burst could range from $10^{50}$ ergs to a few times $10^{52}$ ergs.  From Fig.~\\ref{fig:fanwind} we see that if the detector is primarily sensitive to number counts then at intermediate viewing angles the fan contribution to the signal may be missed.  The pure fan signal, i.e.~a jet viewed within $15^\\circ$ of the rotation axis, will have no afterglow, but the fan plus wind would have an afterglow and a small contribution of high energy photons. \\citet{jfgh+01} have observed a 2 s burst with an afterglow. Unfortunately the high energy detector data was not obtained.   This work was performed under the auspices of the U.S. Department of Energy by University of California Lawrence Livermore National Laboratory under contract W-7405-ENG-48.\\\\  \\clearpage"
    },
    "0208/astro-ph0208135_arXiv.txt": {
        "abstract": "Very little attention has been paid to the properties of optical wavefronts and caustic surfaces due to gravitational lensing. Yet the wavefront-based point of view is natural and provides insights into the nature of the caustic surfaces on a gravitationally lensed lightcone. We derive analytically the basic equations governing the wavefronts, lightcones, caustics on wavefronts, and caustic surfaces on lightcones in the context of weak-field, thin-screen gravitational lensing. These equations are all related to the potential of the lens. In the process, we also show that the standard single-plane gravitational lensing map extends to a new mapping, which we call a wavefront lensing map. Unlike the standard lensing map, the Jacobian matrix of a wavefront lensing map is  not symmetric. Our formulas are then applied to caustic ``surfing.'' By surfing a caustic surface, a space-borne telescope can be fixed on a gravitationally lensed source to obtain an observation of the source at very high magnification over an extended time period, revealing structure about the source that could not otherwise be resolved.  Using our analytical expressions for caustic sheets, we present a scheme for surfing a caustic sheet of a lensed source in rectilinear motion. Detailed illustrations are also presented of the possible types of wavefronts and caustic sheets due to nonsingular and singular elliptical potentials, and singular isothermal spheres, including an example of caustic surfing  for a singular elliptical potential lens. ",
        "introduction": "Among relativistic concepts of direct application to gravitational lensing, the observer's past lightcone is perhaps the most fundamental. The lightcone concept unifies the temporal and spatial properties of lensing events in a geometrical manner that makes the multiplicity, magnification, and time delay  of the images arise naturally.  In this view, gravitational lensing by dark and luminous matter causes the observer's past lightcone to curve into a singular three-dimensional hypersurface that self-intersects and folds sharply. It is the singular ``folding'' of the lightcone that is responsible for most features of interest in gravitational lensing, such as image multiplicity, magnification, and time delay. Additionally, in this view, caustics --- so relevant to observational lensing --- literally acquire a new dimension, turning into caustic sheets (possibly multiple) contained within the lightcone itself, and carrying a sense of time as well as spatial location.  The lightcone concept is most naturally built by imagining an optical wavefront emanating from an observer and traveling out and towards the past or future.  At the start, the wavefront is convex and almost perfectly spherical. Upon encountering gravitational lenses,  the wavefront becomes non-uniformly retarded, acquiring dents. No matter how small and insignificant these dents in the wavefront are soon after leaving the neighborhood of the lens, in time the normal propagation enhances them, inevitably developing crossings and sharp ridges. The surface traced out in spacetime by a wavefront is a (past or future) lightcone.  The trace of the wavefront's sharp ridges and ``swallowtail'' points is called a ``comoving caustic surface'' if thought of as embedded in three-space, and a ``caustic surface'' if viewed in spacetime. In other words, a caustic surface is built out of caustic sheets with cusp ridges and higher-order singularities (e.g., swallowtails, elliptic umbilics, and hyperbolic umbilics). Remarkably, in spite of the naturalness of this concept, very little attention has been paid to the greater significance of the caustics as sheets or surfaces. As a result, to our knowledge, the properties of optical wavefronts and caustic surfaces in gravitational lensing have not progressed much beyond the generic local classification of caustic singularities of the general theory as developed by Thom, Arnold, and others. Our goal is to initiate a program  that investigates optical wavefronts and caustics as they relate to the matter distribution of the gravitational lenses in the spacetime.    The vast majority of gravitational lenses can be modeled extremely well using a small-angle approximation of weak-field perturbations of a Friedmann-Lema\\^{i}tre universe, and projecting the deflectors into planes (thin-screen approximation)  --- see the monographs by Schneider, Ehlers, and Falco \\cite{SEF}  and Petters, Levine, and Wambsganss \\cite{PLW} for a detailed treatment. An enormous advantage of weak-field, thin-screen gravitational lensing is that we can obtain explicit analytical expressions for the associated lightcones, optical wavefronts, and caustics on both the wavefronts and lightcones. Indeed, one of the reasons there has been limited progress in the study of quantitative properties of the optical caustic surfaces and wavefronts due to gravitating matter is the lack of computable physical models with observational relevance. Thin-screen, weak-field gravitational lensing provides us with such a setting.  Gravitational lensing from a wavefront perspective was used by Refsdal \\cite{R64} to relate the Hubble constant to the time delay between lensed images (see \\cite{R64}, Kayser and Refsdal \\cite{KR,B} for more). Arnold's singularity theory was used by Petters \\cite{Ptt93}  to treat the local qualitative properties of caustic surfaces (big caustics) in gravitational lensing. Friedrich and Stewart \\cite{FS}, Stewart \\cite{Stw90}, Hasse, Kriele and Perlick \\cite{HKP96}, Low \\cite{L97}, and Ehlers and Newman \\cite{EN00} also used Arnold's theory to treat wavefronts and caustics in general relativity with varied motivations. In the case of a strong gravitational field, Rauch and Blandford \\cite{RB94} numerically computed the caustic sheets due to the Kerr metric. Wavefronts in the context of liquid droplet lenses have been studied extensively by Berry, Hannay, Nye, Upstill, and others --- see Berry and Upstill \\cite{BU80}, Nye \\cite{Ny99}, and references therein. Readers may also benefit from the popular article by Nityananda \\cite{Nt90} on wavefronts in gravitational lensing.  Our paper calculates explicitly the equations governing the wavefronts and caustic surfaces in gravitational lensing with the standard approximations, and analytically relates the wavefronts and caustic surfaces to the gravitational potential of the lens. This allows us to give a first quantitative treatment of {\\it caustic surfing,\\/} a futuristic notion suggested by Blandford \\cite{Bl01} in his millennium essay. The motivation for caustic surfing is to lock a satellite on a gravitationally lensed moving source so as to observe the source at a high magnification over an extended time period. We derive the starting equations for the trajectory to be followed in order to surf the caustic surface.  The appeal of caustic surfing lies in its potential as a tool to obtain information about a distant moving source that could not otherwise be resolved.  Our article is organized as follows.  Section II derives an explicit expression for the present optical path length in units of time of light signals that reach the observer in the thin-screen approximation. In Section III, this expression is used to define a new lensing map in which the sources lie on an instantaneous wavefront, rather than on a lens plane. The caustic surfaces and sheets are related to the wavefront lensing map in Section IV. Section V develops some fundamentals of the concept of caustic surfing. Applications to the case of nonsingular and singular elliptical potential models, as well of singular isothermal spheres, are given in Section VI.  The latter section also gives full classifications of the singularities of the wavefronts and caustic surfaces  of the noted lenses, and explicitly illustrates the caustic surfing concept.  We conclude in Section VII with some remarks and an outlook of future applications. The research reported in this paper extends preliminary results on wavefronts in gravitational lensing first reported by Frittelli and Petters \\cite{FP01} at the Ninth Marcel Grossmann Meeting. ",
        "conclusions": "We have shown how to construct a lensing map that takes points on the lens plane into points on a wavefront surface, as opposed to a source plane. This represents a sort of ``instantaneous'' lens map, carrying a sense of constant time. By contrast, the standard lens map carries no sense of time at all. Additionally, the wavefront lensing map has an  asymmetric Jacobian matrix. Notice that, barring multiple sheets, the wavefront surface lies very close to a plane in the weak-field case, as illustrated in most of our figures. In the figures the optical axis is magnified several times in order to appreciate the distance between the different sheets in the folded wavefront. Our intial motivation for explicitly constructing a wavefront-lens mapping was inspired by several works (e.g., \\cite{BU80,Nt90,Ptt93,exactuniv,EFN01}). In this respect,  it appears that an extension of our constructions in Secs. III and IV beyond the weak field domain is feasible~\\cite{ehlers00,perlick}.    We took advantage of the conformally flat nature of the flat FL universe in order to make use of a comoving lens equation, which in addition to being comoving is also a conformal lens equation. A conformally related flat spacetime exists, of course, in all three types of FL universes, so in principle, our wavefront map could be adapted for interpretation in the open or closed universes. However, in such universes the conformal factor relating the FL universes to the corresponding flat spacetime depends on the space point, as well as the time, and the translation of our conformal present proper length function  to a cosmological present proper length  is not at all as direct as we found it in this work. Some of the subtleties that would be involved in the translation in the open and closed cases are treated in detail by Frittelli, Kling and Newman \\cite{FKN02}, where the conformally flat lens map and time delay are transformed into the cosmological lensing map and time delay. Nonetheless,  we do not need to rely on the conformally related flat space to calculate our present proper length function.      The individual wavefronts for the potentials illustrated proved useful in visualizing the relationship between the location of the source in reference to the caustic, and the number of images observed. Additionally, the wavefronts of the singular potentials helped explain the anomalous counting of images observed by Petters, Levine, and Wambsganss \\cite{PLW}, p. 188.  The latter showed that in the case of the singular potentials there is a simple closed curve, that is not a caustic, but that separates regions in the source plane where the number of images differs by one, rather than two.  We have here shown that such a curve is the boundary of the wavefront, it is not a caustic, and the number of images differs by one less than in the regular case because one whole sheet of the wavefront is missing due to an obstruction in the lens plane.  Lastly, we have taken a step towards a preliminary scheme  for caustic surfing in a meaningful and consistent manner. The form of the subsequent iterations remains an open problem, as does the implementation of the procedure, particularly the measurement of the transverse velocity of the the moving source that one intends to follow.  Clearly the caustic surfing proposal depends only on the caustic surface, and not on the method used to obtain it. But the point is that optimal caustic surfing (with the least effort) is achieved only by allowing the telescope to surf the caustic sheet, rather than the planar caustics at fixed distance from the source. One could also image scenarios where a telescope may not ride a caustic sheet, but move so as to stay inside certain chambers of the caustic surface (compare with Gaudi and Gould \\cite{GGa97}), possibly near higher order singularities.  Future studies of the aforementioned issues are clearly warranted."
    },
    "0208/astro-ph0208121_arXiv.txt": {
        "abstract": " ",
        "introduction": "\\label{sec:intro}  Magnetic reconnection is the most probable candidate of the energy conversion process for solar flares. Such an explosive event in plasma systems is very common in astrophysical problems, e.g., geomagnetospheric substorms, YSO flares and origins of galactic ridge X-ray emission. However, entire macroscopic structure of evolutionary reconnection is still unclear.  Actual magnetic reconnection in astrophysical systems usually grows in a huge dynamic range in its spatial dimension. The initial scale of the reconnection system can be defined by the initial current sheet thickness. Let us assume it to be of the order of the ion Larmor radius ($\\sim 10^0$[m] for solar corona). Eventually, the reconnection system will develop into the scale comparable to the initial curvature radius of the magnetic field lines ($\\sim 10^7$[m] for typical magnetic loop of solar corona). The dynamic range of the spatial scale in evolution is therefore extremely large ($\\sim 10^7$ for solar flares). For the geomagnetospheric substorms, their dynamic range of growth is also large ($\\sim 10^4$ for substorms). Such a very wide dynamic range of growth suggests that the evolution of the magnetic reconnection should be treated as a development in free space, and individual conditions of the outer circumstances of evolution do not affect the evolution of magnetic reconnection, at least in the initial stage.  The numerical study of magnetic reconnection in free space has been performed by our group (see Nitta, Tanuma, Shibata and Maezawa 2001), and we have discovered a self-similar evolutionary process of the fast reconnection. In our work, a stable self-similar growth of the reconnection system was shown. We here note the definition of the self-similar growth as following: The solution at arbitrary time is the same with the solution at any other time if we change the scale of physical quantities (see section \\ref{sec:zoom-out}). The spatial scale is expanding in proportion to the time. Such self-similar evolution is expected to continue unlimitedly in our case. However, simulated dynamic range of self-similar growth was restricted ($\\sim 10^3$) owing to the restriction of computer memory and CPU time.  The reconnection model with a pair of slow-mode shocks was first presented by Petschek (1964). In this model, the magnetic energy is mainly converted at the slow-mode shocks. Therefore, the time scale of energy conversion is determined by that of MHD wave propagation, and, as such, is much shorter than that of the simple diffusion model (Tajima \\& Shibata 1997) and the Sweet-Parker model (Sweet 1958, Parker 1963). According to the Petschek model, the reconnection rate is almost independent of the magnetic Reynolds number ($\\propto \\log R_m \\sim {R_m}^0$). Hence, we call this quick magnetic energy conversion ``fast reconnection''.  We should note, however, that the question of what controls the reconnection rate is still open. Concerning this point, there have been two different general ways of producing reconnection. One assumes that external boundary conditions should control the reconnection, so that the resistivity has no effective influence on the energy conversion (see, e.g., Petschek 1964, Priest \\& Forbes 1986 for theoretical studies, and Sato \\& Hayashi 1979 for numerical simulations). Another way has been proposed in a series of numerical works originated by Ugai \\& Tsuda (Ugai \\& Tsuda 1977, 1979, Tsuda \\& Ugai 1977, and recently, Ugai 1999). In their simulation, they put a localized finite resistivity in the current sheet to represent an anomalous resistivity that may exist in plasmas. This resistivity acts as a trigger to start the magnetic reconnection. Once this resistivity is put in, the reconnection starts and evolves self-consistently, and forms a fast-reconnection system with the Petschek-type structure (a pair of slow-mode shocks is formed along the current sheet).  However, even in the previous numerical studies aimed at clarification of the time evolution of fast reconnection, the evolution could not be followed for a long time. This is mainly owing to the finite size of the simulation box, hence the application of the results has been limited to spatial scales typically, say, hundred times the spatial scale of the diffusion region. We are interested in evolutionary process in a free space without any influence of the outer circumstance. In such a system, the evolution and resultant structure would be quite different from these previous models. The key process of the self-similar reconnection is summarized as following.  We suppose a two-dimensional equilibrium state with anti-parallel magnetic field distribution, as in the Harris solution. When magnetic diffusion takes place in the current sheet by some localized resistivity, magnetic reconnection will occur, and a pair of reconnection jet is ejected along the current sheet. This causes a decrease of the total pressure near the reconnection point. Such information propagates outward as the rarefaction wave. In a low-$\\beta$ plasma ($\\beta \\ll 1$ in the region very distant from the current sheet [asymptotic region]; as typically encountered in astrophysical problems), the propagation speed of the fast-magnetosonic wave is isotropic, and is much larger than that of other wave modes. Thus, the information about the decreasing total-pressure propagates almost isotropically as a fast-mode rarefaction wave (hereafter we call it FRW) with a speed almost equal to the Alfv\\'{e}n speed $V_{A0}$ in the asymptotic region. Hence, the wave front of FRW (hereafter we call it FRWF) has a cylindrical shape except near the point where FRWF intersect the current sheet. When FRWF sufficiently expands, the initial thickness of the current sheet becomes negligible comparing with the system size $V_{A0} t$, where $t$ is the time from the onset of reconnection. In such a case, there is only one characteristic scale, i.e., the radius of FRWF ($V_{A0} t$), which increases linearly with proceeding of the time. This is just the condition for the self-similar growth.  In this paper, our attention is focused on the analytic study of self-similar stage of magnetic reconnection in free space. An analytic approach has the following importance.  1) By finding the analytic solution for the self-similar growth, we can verify the possibility of the self-similar evolution more rigorously. We set the initial thickness of the current sheet $D$. The self-similar stage is realized in the limit of $V_{A0} t \\gg D$. However, in the computer simulation, we can only perform calculation for a finite duration owing to technical reasons. Hence, we will never reach the exact self-similar stage. But, if we can verify that the result of the numerical simulation is very similar to the analytic self-similar solution, we can accept that the result presented by our numerical simulation is truly a self-similar evolution.  2) Existence of the analytic solution ensures that the self-similar growth will continue unlimitedly. This is important because, in the computer simulation, we cannot continue the calculation more than the case we performed in the previous paper owing to technical reasons. Of course, we should note that when the system sufficiently grows to a scale similar to the initially imposed system size (e.g., the scale of the flux tube for the case of solar flares), this self-similar evolution must be modified by its circumstance. We should note that such analytic treatment cannot ensure the stability of the self-similar evolution, while it is ensured by our numerical simulation. Thus the analytic treatment and numerical simulation are complementary to each other.  This paper is organized as follows. In section 2, we introduce a special coordinate system called ``the zoom-out coordinate''. In this coordinate, the real spatial scale is shrinking linearly as the time proceeds. If we choose the origin of time appropriately, self-similar expansion can be a stationary solution in this coordinate. We should modify the MHD equations in a relevant form in the zoom-out coordinate. In section 3, we review basic assumptions which will be made in this work. We study magnetic reconnection in the limit of very small reconnection rate, and adopt the method of perturbative expansion by using the reconnection rate as a small perturbation parameter. The initial equilibrium is treated as the zeroth order (unperturbed) situation. We solve the linearized problem for the first order quantities, which show the variation from the initial state. We can rearrange the first order equations to a single 2nd order partial differential equation for magnetic vector potential. This procedure is called the Grad-Shafranov approach. We will see that this equation is elliptic in our problem. Using the SOR routine for solving the Grad-Shafranov equation, we obtain a self-similar solution. The boundary condition used in this work is discussed at the end of section 3. In section 4, properties of this self-similar solution are discussed. In section 5, we summarize our results, and discuss the properties of this self-similar evolution model. ",
        "conclusions": "\\subsection{Summary}  Together with our time-dependent numerical simulation (see Nitta et al. 2001), the linearized perturbation solution discussed in this paper has ensured the existence of self-similar growth of fast magnetic reconnection.  Thus we propose here the new model describing ``self-similar evolution of fast reconnection''. The time dependent simulation directly solving the MHD equations numerically is effective to check the stability of the evolving system. However, the duration of simulation is restricted owing to the restriction of computer memory and run time, or stability of the simulation code itself. Hence, even if we find the self-similar (-like) behavior in the result of simulation, we cannot be convinced that this behavior is a true one that may continue indefinitely.  On the other hand, an analytical study has the following properties. If one solves the MHD equation in the zoom-out coordinate under the stationary assumption and obtain a (semi-) analytic solution which is identical to the solution of numerical simulation, we can be convinced that the numerically obtained self-similar-like solution is truly the self-similar one. On the other hand, an analytical study of the stationary equation does not give information on the stability of the solution.  From these arguments, we can reach the following conclusion. Our time-dependent numerical study and analytical study are complimentary to each other in establishing the model of ``self-similar evolution of fast reconnection''.   \\subsection{Consistency between semi-analytic and numerical studies} In section 4, we compared our semi-analytic result with numerical result. We can notice significant similarity between semi-analytic and numerical results.  If we compare them quantitatively in detail, we can notice that these two results are consistent in the region in which $v_{1y}'<0$. However, in the region in which $v_{1y}'>0$, the value of semi-analytic result is somewhat different from the numerical result. In this region, strong fast-mode compression (``piston effect'' of the reconnection jet) takes place, and, strictly speaking, our linearized treatment might not be suitable to be applied here.  Thus, we can conclude that the inflow region is well described by our linearized theory. This is obviously due to the very small reconnection rate ($\\sim 10^{-2}$, see Nitta et al. 2001). However, the mechanism leading to such a small reconnection rate is still unclear. In this self-similar reconnection model, the reconnection rate should be self consistently determined by the dynamical evolution process, and this point will be discussed in our forthcoming paper.     \\subsection{Condition for self-similar reconnection} \\label{sec:cond}  The above type of evolutionary reconnection will be realized in the systems in which 1) the initial spatial scale of the disturbance ( e.g., scale of microscopic instabilities which will lead to the anomalous transport phenomena $\\sim 10^0$ [m] if we estimate it by ion Larmor radius in the typical case of solar flares) is much smaller than the entire spatial scale of the system ( e.g., the curvature radius of the loop of magnetic flux tube $\\sim 10^7-10^8$ [m]  in the typical case of solar flares), and 2) there is no proper spatial scale except the one which expands as time proceeds (e.g., the radius of FRWF in our case). In such a system, the magnetic reconnection triggered by the initial disturbance can evolve as the FRW propagates (see section 2 of Nitta et al. 2001). The spatial scale of FRW propagation is the only proper scale of the system if it is much larger than the initial current sheet thickness. The spatial dynamic range of the evolution is determined by the ratio of the entire system scale to the initial disturbance scale. In the above mentioned case of typical solar flares, it will be of the order of $10^7$. In such a system, the external boundary does not affect the evolution for a certain amount of time after the onset of reconnection. This means that the system can freely evolve independent of the outer boundary condition. Such a very wide dynamic range and only one evolving spatial scale strongly suggest the possibility of self-similar solution. In fact, we have obtained the self-similar solution both numerically and analytically.    \\subsection{Near FRWF} The region near the FRWF ($r \\sim 1$) actually has a complicated nature. The region near the spearhead of reconnection jet (roughly to say, $r \\sim 1, \\ 0<\\theta<\\pi/8$) is influenced  not only by the fast-mode rarefaction wave, but also by the fast-mode compression wave induced by the ``piston effect'' of the jet. Hence one might think that the word ``FRWF'' (fast-mode {\\it rarefaction} wave front) is somewhat misleading.  However, we should notice that, in our linearized treatment, any non-linear interaction between waves is completely neglected. Thus the fast-mode rarefaction wave emitted from the vicinity of reconnection point keeps a circular shape which is truly located at $r=1$ in the zoom-out coordinate even if the compressional mode is superimposed. From this reason, we may continue to call it ``FRWF'' as we used this term in our previous paper.    \\subsection{Relation to reconnection jet}  We have treated only the inflow region in this paper in spite of the importance of the reconnection jet. This is owing to the mathematical difficulty to treat the reconnection jet where deviation from the initial state is very large and the linear analysis we have used here breaks down. The complete solution which covers the entire region has not been obtained yet.  However, we should note the importance to study the property of the inflow region. As Priest \\& Forbes (1986) discussed, the property of the inflow region crucially shows the relation to previous classical models of reconnection. We can see that our new model is an extension of the Petschek model to the evolutionary model in free space (see section \\ref{sec:Com}). The fast-mode rarefaction dominated inflow restricts the reconnection rate as in the Petschek model (see Priest \\& Forbes 1986 or Vasyliunas 1975). In our case, the region near the X-point is filled with the fast-mode rarefaction dominated inflow. Hence we may expect that our model has a maximum value of the reconnection rate. This point will be clarified by our forthcoming study of the reconnection jet which induces the fast-mode rarefaction wave.  In our analytic work, the boundary condition at $y=0$, which corresponds to the junction condition to the reconnection jet, is artificially imposed approximating the result of our numerical simulation. Needless to say, this boundary condition is more important than other boundary conditions (e.g., conditions on $r=1$ or $\\theta=\\pi/2$) because it represents physical information about the reconnection jet, and it crucially influences the solution of inflow region.  One might think that the similarity of the boundary condition at $y=0$ is not trivial, and the above boundary condition is imposed {\\it ad hoc}. However, we must note that everything evolves self-similarly in this situation. In the self-similar stage $V_{A0} t \\gg D$, the entire reconnection system including reconnection jet has no proper length other than the scale of FRWF $V_{A0} t$. Thus, it naturally leads that the reconnection jet itself will grow self-similarly, and hence, the similarity growth of reconnection jet is realized.    \\subsection{Relation to diffusion region}  The self-similar stage realizes after the system size $V_{A0} t$ sufficiently exceeds any finite fixed spatial scale, i.e., the initial thickness of the current sheet and size of the diffusion region. Since both of these scales are estimated as $D$, the self-similar stage can be realized for $V_{A0} t \\gg D$ as noted in section \\ref{sec:intro}. There is no need to go into details about the structures of them because any fixed scale is negligible in the self-similar stage. In other words, this shows the universality of the self-similar evolution. The self-similar evolution does not depend on detailed structure of the current sheet. We can adopt any current sheet model if it satisfies the initial MHD equilibrium. The self-similar evolution also does not depend on detailed resistivity model and structure of the diffusion region. The only function of the diffusion region which we need is to reconnect the field lines in a desired speed. We may expect that this is possible by dynamical change of the thickness of diffusion region. In fact, our numerical simulation showed that our new model is insensitive to the resistivity model (this is an important property of the fast reconnection).    \\subsection{Comparison with previous steady and unsteady models} \\label{sec:Com}  There are several theoretical models for steady state and time-depending magnetic reconnection. We compare our self-similar evolution model with these previous models. Our discussion is focused only on fast reconnection, because very quick energy conversion frequently observed in astrophysical phenomena suggests that fast reconnection should be considered as the responsible mechanism.  The Petschek model (Petschek 1964) is characterized by a pair of slow shocks and a fast-mode rarefaction wave in the inflow region. This fast-mode rarefaction wave propagates outward from the reconnection point. As a result of the fast-mode rarefaction, the gradient of the magnetic field strength near the neutral point decreases due to the bending of magnetic field lines (Vasyliunas 1975). This process suppresses and limits the diffusion speed, and hence the reconnection rate.  The Sonnerup model (Sonnerup 1970) is developed from the Petschek model. This model is characterized by a pair of slow shocks and the hybrid of fast-mode and slow-mode rarefaction. The fast-mode rarefaction wave is produced at the central region as in the Petschek model, but the slow-mode rarefaction wave is injected from the boundary. Because of the hybrid nature of rarefaction modes, the gradient of the magnetic field strength near the neutral point does not decrease as in the Petschek model. Therefore the Sonnerup model attains the maximum reconnection rate possible for magnetic energy converters (Priest \\& Forbes 1986). A modified model of the Sonnerup solution was presented by Priest \\& Forbes (1986).  They found a solution with a diffuse slow-mode rarefaction waves spread throughout the inflow region, while in the original Sonnerup model the rarefaction waves were treated as discontinuities.  The property of the central region of our model can be categorized into the ``almost-uniform reconnection'' which means that magnetic field lines in the inflow region are almost straight and uniform (see Priest \\& Forbes 2000). Priest \\& Forbes (1986) presented a unification scheme of 2-dimensional steady almost-uniform reconnection in a finite space filled with incompressible plasma. They found a family of continuous solutions characterized by a non-dimensional parameter ``$b_0$''. The Petschek ($b_0=0$) and Sonnerup ($b_0=1$) types are particular cases. Their result includes other types, e.g., Sweet-Parker type, flux-pile-up type and stagnation-flow type according to the value of $b_0$. These solutions ($b_0>0$) other than the Petschek type are characterized by the hybrid of slow-mode and fast-mode rarefaction. Our self-similar evolution is never influenced by the outer conditions, so the hybrid rarefaction never takes place in our case because there are no boundary-imposed waves in our case. We can conclude that the unique candidate which is worth comparing with the central region of our model is the Petschek model.  Figures \\ref{fig:A1} and \\ref{fig:rho1} clearly show decreases of the magnetic field and gas pressure, thus the fast-mode rarefaction dominates in the vicinity of the diffusion region. Obviously the central region of this self-similarly evolving system is of the Petschek-type. Therefore, the reconnection rate might be limited in a way similar to the original Petschek model. This point will be clarified in our forthcoming paper.  Far from the central region, there is a region in which the fast-mode compression takes place by the ``piston effect'' of the reconnection jet. This fast-mode compression causes the vortex-like return flow (see figure 13 of Nitta et al. 2001). The combination of the inner structure which resembles the original Petschek model and the outer structure involving the flow vortices characterizes the evolutionary process, and the entire system unlimitedly expands self-similarly.  Biernat, Heyn and Semenov (1987) and Semenov et al. (1992) analytically studied the evolutionary process of fast magnetic reconnection in a similar situation to our case. Their analysis gives a general formalism of spontaneous time-varying reconnection. They succeeded to obtain the solution for Petschek-type reconnection. However, in their works, plasma in the inflow region is assumed to be incompressible for analytical convenience. This means that the sound speed is infinitely large even when the Alfv\\'{e}n speed is finite. This is equivalent to assuming that the inflow region is filled with extremely high $\\beta$ plasmas (note that $\\beta \\sim$ (sound speed)$^2$/ (Alfv\\'{e}n speed)$^2$). Needless to say, this assumption is unsuitable for most cases of astrophysical problems. Contrary to our case, fast-mode waves emitted from the central region instantly propagates to infinity. Although the proper spatial scale determined by the fast-mode wave propagation does not exist in their case, there is a possibility that the structure formed by slow-mode wave can be self similar, because the propagation speed of the slow-mode wave is finite in their case. Our work is understood as an extension of their works to realistic astrophysical systems filled with low $\\beta$ compressible plasma.    \\subsection{Astrophysical applications} In the present level of our observational technology, only one object in which we can obtain enough spatial and time resolution of the distributions of plasma parameters is solar flares. We expect the application of our self-similar reconnection model to the solar flares.  \\subsubsection{Dimming and inflow structure} The propagation speed of FRW is estimated to be $\\sim 10^6$ [m s$^{-1}$] for solar corona. Hence, the duration of the self-similar evolution is typically $10^1-10^2$[s] for the flux tubes having the spatial dimension $\\sim 10^7-10^8$ [m]. The evolution having such a time scale is able to be resolved by the {\\it Solar-B} project (required cadence of {\\it Solar-B} X-ray telescope [XRT] is 2 sec., which is sufficient to resolve the expected time evolution of the self-similar reconnection, see Golub 2000.). Especially we expect that the ``dimming'' will be detected by {\\it Solar-B}. Expected X-ray image of the dimming is argued in Nitta et al. 2001 (see section 4.5). We consider that the dimming is a naturally expected phenomenon at the inflow region of our self-similar reconnection model. If we obtain the information of velocity field and other plasma parameters at the reconnection site by observation, we can inspect our model more directly.  The {\\it Solar-B} satellite will clarify also the velocity field, plasma density and temperature near the reconnection point by the extreme ultraviolet imaging spectrometer (EIS). The combination of XRT and EIS will reveal a detailed structure of the reconnection system in solar flares. Yokoyama et al. (2001) found the evidence for reconnection inflow in a flare on 1999 March 18. The estimated inflow speed is 5 km s$^{-1}$ which corresponds to the reconnection rate 0.001-0.03 (this result is not far from the expected value 0.05 from our model). Their result encourages our theoretical effort of applying our self-similar model to the inflow region of real solar flares. We hope that the expected feature of the inflow region by virtue of our self-similar reconnection model will be inspected by the {\\it Solar-B} project.    \\subsubsection{Application to realistic cases} Roussev et al. (2001 a, b, c) predicted emission from explosive events in the solar transition region induced by magnetic reconnection. They performed MHD numerical simulations including thermal conduction, radiative losses and volumetric heating. They treated spontaneous evolution of magnetic reconnection with the initial conditions similar to our model (anti-parallel magnetic field and asymptotically uniform plasma distribution) and with the resistivity (artificially localized and kept as a constant) that is also very similar to ours.  The resultant dynamical properties are similar to our study, indicating that the essence of our model is applicable to more complicated situations with more realistic thermal and radiative properties.   \\subsection{Expectation of 3D self-similar reconnection} Our present discussion is restricted to 2D reconnection in this as well as in the previous paper (Nitta et al. 2001) for simplicity. However, actual current sheet systems have finite depth in the direction perpendicular to the figures of this paper (for example, radius of the flux tube or length of the arcade structure of a bunch of flux tubes). We believe the essence of the self-similar growth has already been understood by our 2D approximation. When we discuss the phenomenology of the evolutionary reconnection by virtue of the self-similar reconnection model in near future, we need a more precise 3D model.  We can expect that the self-similar growth also takes place in a 3D system. As discussed in section \\ref{sec:cond}, a sufficiently evolved system has only one proper spatial scale, i.e., the scale of FRWF $V_{A0} t$ which is increasing in proportion to time (note that the propagation speed of the fast-mode is almost isotropic in a 3D current sheet system filled with a low $\\beta$ plasma, hence the shape of FRWF will be spherical). This is just the situation that leads to realize the self-similar expansion. Of course, detailed structure of the 3D reconnection system will be different from our 2D model, but essential properties will be common between them."
    },
    "0208/astro-ph0208317_arXiv.txt": {
        "abstract": "The circumstellar masers around evolved stars offer an interesting possibility to measure stellar parameters through VLBI astrometry. In this paper the application of this technique is discussed, including the accuracy and the uncertainties of the method. The different maser species (OH, H$_2$O, SiO) have slightly different characteristics and applications. This paper does not concern astrometry of maser spots to study the kinematics of the envelope, but concentrates on attempting to measure the motion of the underlying star. ",
        "introduction": "The most straightforward way of doing VLBI astrometry is to use ``phase referencing'' to a bright and close calibrator in order to overcome the spatial and temporal fluctuations of the ionosphere and\\slash or troposphere.  The calibrator is usually an extragalactic source with a fixed and known position, which is sufficiently bright to calibrate the atmospheric and instrumental effects within a fraction of the coherence time.  Because the modern VLBI correlators have incorporated the best geodetic models, phase referencing results obtained in this way can be accurate to a few mas.  This is already sufficient for a number of applications. In other cases, higher accuracy is required and some adjustments are necessary to improve the model. In such cases one probably needs to return to the \"totals\" in order to use astrometric/geodetic software. To get accuracies below 1 mas is not trivial; better models of the atmospheric behavior and structure of the reference source are required.  For maser astrometry an additional complication can be that one needs to observe in mixed bandwidth mode, applying a narrow band for the maser detection and simultaneous wide bands for the reference source. Obviously the maser dictates the observing frequency, which rules out some calibration schemes used in continuum and geodetic VLBI.  Ideally the masers should be bright and persistent. The brightness relates to the size of individual maser features, which sets an upper limit to the resolution one can obtain. A fundamental issue remains to link the motion of the maser spot to the stellar properties. One way forward is to assume that the masers are on a linear path with respect to the star (e.g.\\ a constant outflow). In this case the parallax of individual spots equals the parallax of the star, and the average motion may be assumed to be the stellar motion. In other cases one may be able to deduce the position of the star from the distribution of the masers, for example when they form a ring. If we can determine the position of the star, one can in principle measure the motion and the parallax.  Note that only when one can consistently determine the stellar position with (much) better than 1 AU accuracy, a useful parallax can be determined. In some cases there is evidence that special maser spots exists that correspond to the stellar continuum amplified by the maser shell. Such spots are then tied very accurately to the stellar position. In other cases one has to worry about the systematic and turbulent motions in the maser shell. In any case, there may be additional motions involved, for example when masers occur in binary stars. ",
        "conclusions": ""
    },
    "0208/astro-ph0208444.txt": {
        "abstract": "I show that the relativistic winds of newly born magnetars (neutron stars with petagauss surface magnetic fields) with initial spin rates close to the centrifugal breakup limit, occurring in all normal galaxies with massive star formation, can provide a source of ultrarelativistic light ions with an $E^{-1}$ injection spectrum, steepening to $E^{-2}$ at higher energies, with an upper cutoff at $10^{21}-10^{22}$ eV. Interactions with the CMB yield a spectrum at the Earth which compares favorably with the spectrum of Ultra-High Energy Cosmic Rays (UHECR) observed at energies up to a few$\\times 10^{20}$ eV. The fit to the observations suggests that $\\sim 5-10$\\% of the magnetars are born with rotation rates and voltages sufficiently high to allow the acceleration of the UHECR. The form the spectrum incident on the Earth takes depends sensitively on the mechanism and the magnitude of gravitational wave losses during the early spindown of these neutron stars - pure electromagnetic spindown (the $E^{-1}$ injection spectrum) yields a GZK feature (a flattening of the $E^3 J(E)$ spectrum) below $10^{20}$ eV, rather than a cutoff, while a moderate GZK cutoff appears if gravitational wave losses are strong enough to steepen the injection spectrum above $10^{20}$ eV. The flux above $10^{20}$ eV comes from magnetars in relatively nearby galaxies ($D < 50$ Mpc.)  I outline the probable physics of acceleration of such particles in a magnetar's wind - it is a form of ``surf-riding'' in the approximately force free fields of the wind. I also show how the high energy particles can escape with small energy losses from the magnetars' natal supernovae. In particular, I show that the electromagnetic energy emitted by the magnetar ``shreds'' the supernova envelope in times short enough to allow most of the relativistic energy to escape largely unmimpeded into the surrounding interstellar medium, where it drives a relativistic blast wave that expands to parsec scale before slowing down to nonrelativistic speeds. I also show that since the ions are accelerated in a region where the magnetic field has the structure of a strong electromagnetic wave but propagate at larger radii through a region of weaker magnetic field near the rotational equator of the outflow, the ultrahigh energy particles escape with negligible adiabatic and radiation losses.  The requirement that the magnetars' relativistic winds not overproduce interstellar supershells and unusually large supernova remnants suggests that most of the initial spindown energy is radiated in khz gravitational waves for several hours after each supernova. For typical distances to events which contribute to $E > 100 $ EeV air showers, the model predicts gravitational wave strains $\\sim 3 \\times 10^{-21} $. Such bursts of gravitational radiation should correlate with bursts of ultra-high energy particles. The Auger experiment should see bursts of particles with energy above 100 EeV every few years.  %\\vspace{0.15in} ",
        "introduction": "I study the possibility that relativistic winds from rapidly rotating {\\it magnetars}, neutron stars with surface dipole fields on the order of $10^{15}$ Gauss (Duncan and Thompson 1992, Paczynski 1992, Kouveliotou {\\it et al.} 1998, 1999) create the highest energy cosmic rays (ultra-high energy cosmic rays, a.k.a. UHECR). I assume magnetars occur in all normal galaxies which form massive stars, with the UHECR ariving from outside our own Galaxy, except on the occasions (perhaps once per $10^5 $ years) when a rapidly rotating magnetar is born in our own galaxy. This  model has the virtue of having little difficulty in accelerating protons to energies in excess of $10^{21}$ eV, with the sources being in all normal (star forming) galaxies and a luminosity density entirely acceptable from the point of view of the (very approximately known) rate of formation of magnetars in our own galaxy, thus offering an explanation of the puzzling air showers with energies  above $10^{19.6} $ eV without having to introduce major extrapolations of known or suspected energetics.  In \\S \\ref{sec:background}, I summarize the data on UHECR, the loss processes that affect their transport through intergalactic space, the energetics of UHECR and aspects of the astronomy and physics of magnetized compact objects as known from studies of rotation powered pulsars that are relevant to the present investigation. \\S \\ref{sec:spectrum} outlines the calculation of the particle injection spectrum from a newly born magnetar and the effects intergalactic transport have in altering the injected spectrum to the spectrum received at the Earth. In \\S \\ref{sec:escape}, I discuss the escape of the relativistic wind from a magnetar's natal supernova, and in \\S \\ref{sec:blast} I outline how the wind drives a relativistic blast wave containing a Magnetar Wind Nebula (MWN) into the surrounding interstellar medium. The accceleration mechanism of the UHE ions in the magnetar's wind is addressed in \\S \\ref{sec:accel}, and their escape from the expanding MWN nebula is outlined in \\S \\ref{sec:MWNescape}. The effect the electromagnetic energy lost from a magnetar has in creating HI supershells in the interstellar medium is discussed in \\S \\ref{sec:supershells}, with results that are used to suggest that more than 90\\% of a newly born, rapidly rotating magnetar's $5 \\times 10^{52} $ ergs of rotational energy gets lost as gravitational radiation. The typical strains of such gravity waves, from relatively nearby events that could contribute particles in the UHECR spectrum with energies above $10^{20}$ eV, are estimated in \\S \\ref{sec:gravity_waves}. The possibility of observing  bursts of UHECR associated with the birth of individual magnetars gets attention in \\S \\ref{sec:beaming}. I discuss the relation of this study to other work in \\S \\ref{sec:other_models}.  I draw my conclusions in \\S \\ref{sec:conclusions}.  The most prominent difference between this model for UHECR and most others which appeal to sources more or less uniformly distributed throughout the Universe is that the underlying objects create an extremely flat injection spectrum, with particles per unit energy range injected at a rate $\\propto E^{-1}[1+(E/E_g)^s]^{-1}$, with $E_g$ reflecting the strength of gravity wave emission on the magnetars' early spin down; for the model developed in detail here, $s=1$.  As a result, the usual Greisen-Zatsepin-Kuzmin (GZK) cutoff disappears. This cutoff, apparent in models that assume power injection spectra $\\propto E^{-s}, s\\geq 2$ (see Berezinsky {\\it et al.} 2002 for a recent example of such ``conservative'' models), becomes a flattening of the observed $E^3 J(E)$ spectrum between $10^{19}$ and $10^{20}$ eV; the flatttening  becomes a moderate replica of a GZK cutoff if the gravitational wave losses are strong enough. ",
        "conclusions": "}  The model's basic results and predictions are shown in Figure \\ref{fig:theory_data}.  The upper cutoff of the spectrum, at $E = 10^{21.5} Z\\eta_1 \\mu_{33} \\Omega_4^2 $ eV, is likely to be reduced by dissipation of the wind as it first breaks free of the magnetar's natal supernova.  The estimates of \\S \\ref{sec:escape} suggest that escape from the supernova consumes a fraction $1 - W_{blowout} \\sim 0.3 - 0.4$ of the magnetar's intial energy loss, in turn suggesting that $\\sim 80$\\% of the highest voltages are available for cosmic ray acceleration.  If so, then the highest energies in the spectra shown in Figure \\ref{fig:theory_data} should be be taken seriously.  Agreement with the well determined UHE cosmic ray flux below $5 \\times 10^{19} $ eV constrains the combination $W_{geom} n_{g2} \\nu_{m4}/\\mu_{33}$ to be approximately 0.02-0.06, where $\\mu_{33}$ is the magnetic moment in units of $10^{33}$ cgs, $\\nu_{m4}$ is the magnetar birth rate in units of $10^{-4}$ yr$^{-1}$, $n_{g2} = n_{galaxy}/0.02 \\; {\\rm Mpc}^{-3}$ and $W_{geom} = 0.5$ is the inefficiency factor due to the fact that only 50\\% of the magnetic geometries are appropriate to the ion current being emitted into the rotational equator, where radiation and adiabatic losses on the UHECR would be weak.  The small value of this normalization required to get a fit, appropriate to the ``no GR'' and ``moderate'' GR cases considered in Figure 1, implies that the birth rate of magnetars with voltages high enough to create UHECR is 5-10\\% of the overall magnetar birth rate, a conclusion consistent with not overpopulating our and other galaxies with too many supershells and overly large supernova remnants.  The high energy ions suffer negligible adiabatic and radiation losses in escaping the magnetars' winds and their surrounding, relativistically expanding nebulae. The existing observations require voltages $\\Phi > 2 \\times 10^{21} /Z\\eta_1 $ Volts, or $\\mu_{33} \\Omega_4^2 > 0.06 /\\eta_1 Z \\; (P_{initial} < 2.6 (Z\\eta_1 \\mu_{33})^{1/2}$ ms), with $\\eta_1 = \\eta /0.1 $ the fraction of the voltage actually sampled by each ion. I suggested in \\S \\ref{sec:accel} that electromagnetic surf-riding in the relativistically strong electromagnetic waves in the wind, generated by the oblique rotator, is responsible for the ion acceleration, occuring at or within the radius ($\\ll 0.1$ AU) where the wave structures no longer are frozen into the expanding plasma. That acceleration site is consistent with small radiative losses.  This model works only if the ions travel from their source galaxies on more or less straight lines. That requires intergalactic magnetic turbulence to be small amplitude, or the intergalactic magnetic field to be weak, or both - $\\delta B /B  < 10^{-2} (10^{-9} \\; {\\rm Gauss}/ Z B_{IGM})^{1/2}$ (\\S \\ref{sec:scatter}). Such weak scattering introduces negligible time delays into the particles' transport from source to observer.  If the current data are interpreted as providing evidence for the conventional GZK cutoff, {\\it i.e.} the AGASA results are disregarded (Bahcall and Waxman 2003) or are renormalized to bring them into accord with the Hi-Res results (Abu-Zayyad {\\it et al.} 2002b), agreement of this model with existing observations requires rather strong losses of rotational energy due to gravitational radiation. The ``strong GR'' curve in Figure (\\ref{fig:theory_data}) shows the effect of the rapid spindown depopulating the highest energy end of the injection spectrum.  The model can replicate the high energy end of the spectrum observed by AGASA only if gravitational wave losses are small. The metagalactic magnetar model has no trouble with a spectrum that compromises between the highest energy AGASA results and the Hi-Res (without the 1995 Bird {\\it et al.} event) results - including that data point favors such a compromise.  Even when one assumes a birth rate of fast magnetars $\\sim$ 5-10\\% of the overall magnetar birth rate, the requirement that the model not overproduce supershells and large supernova remnants in our and other galaxies' interstellar media suggests gravitational wave losses are an important drain of rapidly rotating magnetars' initial rotational energy. Therefore, one expects bursts of almost coherent gravitational waves with millisecond periods and strains with magnitude $\\geq 10^{-21}$ if observed in an experiment designed to find almost coherent signals, lasting for several hours and overlapping the arrival of the higher energy particles.   The model suggests the UHE cosmic rays come from sources whose distribution should mimic that of luminous baryons. More specifically, the sources should follow luminous matter with stars that are progenitors of core collapse SNe, thus should anticorrelate with large galaxy clusters. Events with energy above $10^{20}$ eV come from small distances ($D < 50$ Mpc), which may allow some imprint of galaxies' large scale structure to appear on the isotropy of such events. Starburst galaxies are obvious suspects for especially luminous UHECR sources.  In conclusion, I have shown that bare magnetars in normal galaxies provide a possible source for the origin of UHECR, both in total flux and in form of the spectrum. I have also given plausible arguments for how the particles accelerated by a magnetar embedded in its supernova and its magnetar wind nebula can escape without catastrophic losses, to contribute to the particle spectrum observed at the Earth.  This subject will remain observationally driven, most prominently by the AGASA ({\\it e.g.} Takeda {\\it et al.} 1998), Hi-Res ({\\it e.g.} Sokolosky 1998, Abu-Zayyad {\\it et al.} 2002b) and Auger ({\\it e.g.} Boratov 1997, Cronin 2001a,b) experiments, which provide a window into the spectrum at energies above $10^{20.5} $ eV. The theory described here suggests that there will be something to see at these super-GZK energies. Also, Auger, with several thousand km$^2$ collecting area and with fluorescence and particle detectors recording events simultaneously at the same site, will have the opportunity to resolve individual UHE particle bursts, with enough statistics to measure a spectrum for one event, at least up to $10^{20}$ eV - see \\S \\ref{sec:beaming} for bookeeping on this model's predictions. On the theoretical front, the theory of how the acceleration actually occurs, in the model presented here and in other schemes, urgently needs substantial improvements."
    },
    "0208/astro-ph0208584_arXiv.txt": {
        "abstract": "{Based on detailed 2D and 3D numerical radiation-hydrodynamics (RHD) simulations of time-dependent compressible convection, we have studied the dynamics and thermal structure of the convective surface layers of a prototypical late-type M-dwarf ($\\Teff\\approx 2800\\pun{K}$, $\\logg=5.0$, solar chemical composition).  The RHD models predict stellar granulation qualitatively similar to the familiar solar pattern. Quantitatively, the granular cells show a convective turn-over time scale of $\\approx 100\\pun{s}$, and a horizontal scale of $80\\pun{km}$; the relative intensity contrast of the granular pattern amounts to 1.1\\pun{\\%}, and root-mean-square vertical velocities reach 240\\pun{m/s} at maximum. Deviations from radiative equilibrium in the higher, formally convectively stable atmospheric layers are found to be insignificant allowing a reliable modeling of the atmosphere with 1D standard model atmospheres. A mixing-length parameter of \\mbox{\\mlp=2.1} provides the best representation of the average thermal structure of the RHD model atmosphere while alternative values are found when fitting the asymptotic entropy encountered in deeper layers of the stellar envelope \\mbox{(\\mlp=1.5)}, or when matching the vertical velocity field \\mbox{(\\mlp=3.5)}. The close correspondence between RHD and standard model atmospheres implies that presently existing discrepancies between observed and predicted stellar colors in the M-dwarf regime cannot be traced back to an inadequate treatment of convection in the 1D standard models. The RHD models predict a modest extension of the convectively mixed region beyond the formal Schwarzschild stability boundary which provides hints for the distribution of dust grains in cooler (brown dwarf) atmospheres. ",
        "introduction": "Late-type M-dwarfs are fully convective stars where the convective flows penetrate far into the atmospheres reaching optical depths as low as $10^{-3}$ \\citep{Allard+Hauschildt95}. \\citet{Allard+al97} have reviewed the physical, spectroscopic, and photometric properties of these objects.  In the past, model atmospheres have typically failed to reproduce their spectroscopic and photometric properties in two respects: i) the near-IR spectral distribution ($JHK$ colors) where, independent of the source of water vapor line data used, models all agree to predict an underluminous $K$-band (relative to $J$), and ii) the optical M$_V$ vs $V-I$ color-magnitude relation, where all models systematically predict bluer colors (i.e. being overluminous in $V$) than observed.  \\citet{Brett95} raised  the possibility that this  near-IR problem was due to models being ``too cool in the upper photospheric layers'', and suggested  two  possible  causes:  chromospheric  heating  and/or  the treatment of convection based on mixing-length theory (MLT).  Hydrodynamical simulations of solar and stellar granulation including a realistic description of radiative transfer have become an increasingly powerful and handy instrument for studying the influence of convective flows on the the structure of late-type stellar atmospheres as well as on the formation of their spectra \\citep[e.g.][]{Nordlund+Dravins90, Steffen+Freytag91, Ludwig+al94, Freytag+al96, Stein+Nordlund98, Asplund+al00}. Hitherto, model calculations have been exclusively performed for atmospheres where atomic lines are dominating the line blanketing. A possible next step in the development of hydrodynamical models is towards cooler atmospheres where {\\em molecular absorption\\/} dominates the atmospheric energy balance. Constructing hydrodynamical model atmospheres for cooler stars can shed light on the presently pressing shortcomings of the classical models mentioned above. Regarding the considerable improvements in the quality of the molecular opacities and related atmospheric models, it becomes more and more important to determine whether the treatment of convection by MLT is at the origin of the observed discrepancies.  The basic questions we want to answer in this theoretical investigation are: Is mixing-length theory adequate to handle convection in the atmospheres of M-dwarfs?  And if so, which mixing-length parameter~\\mlp\\ is necessary to reproduce the various thermal and dynamical properties of an atmosphere (temperature profile in the line forming region, surface boundary condition connecting to stellar evolution models, convective velocities)?  We start by describing some methodological aspects and the applied computer codes, in particular discuss the critical question of how accurately we can describe the complex radiative transfer within the hydrodynamical simulations. We continue by presenting our results which give some insight in what granulation looks like on the surface of an M-dwarf. We proceed with quantitative estimates of the mixing-length parameter, and discuss the consequences for conventional atmosphere modeling. Finally, we extrapolate slightly beyond the existing hydrodynamical models proper, and suggest a scenario for the transport of dust grains in brown dwarf atmospheres due to convective overshoot which is motivated from our present simulations at hotter temperatures. Often we refer to the Sun as our benchmark for comparison and assume some familiarity with its atmospheric properties. ",
        "conclusions": "We used elaborate 2D and 3D radiation-hydrodynamics simulations to study properties of convection on the surface of a prototypical late M-dwarf ($\\Teff\\approx 2800\\pun{K}$, $\\logg=5.0$, solar chemical composition). Despite the significant differences in the physical conditions encountered in the solar and an M-dwarf atmosphere we obtained the striking result that M-dwarf granulation does not look qualitatively different from what is familiar from the Sun (see Fig.~\\ref{f:flow}). Quantitative differences (intensity contrast 1.1\\pun{\\%}, horizontal scales $\\approx 80\\pun{km}$, maximum RMS velocities $\\approx 240\\pun{m/s}$, convective turn-over time scale $\\approx 100\\pun{s}$) remain within the expectations derived from mixing-length theory.  Connected to this basic finding is the --- for practical purposes --- perhaps most important result that the temperature structure of the higher atmospheric layers is determined by the condition of radiative equilibrium, and is not very much affected by processes usually not accounted for in standard stellar atmosphere models. Convective overshoot as well as energy transport by waves do {\\em not\\/} significantly affect the temperature structure outside of the region of convective instability. We expect that this finding also holds for main-sequence objects of higher effective temperature where radiation becomes relatively more important.  Answering the questions posed in the introduction we confirm that in late M-dwarfs mixing-length theory can be applied to obtain a realistic description of the convective energy transport in a 1D stellar atmosphere code, provided one can remove uncertainties related to the choice of the mixing-length parameter. If an independent calibration is available we expect that 1D atmosphere models allow reasonable accurate predictions (on a level as displayed in Fig.~\\ref{f:salpha}) of the atmospheric temperature structure and ultimately the stellar spectrum to be made. Effects on the stellar spectrum related to {\\em horizontal\\/} temperature inhomogeneities are expected to be very small in the M-dwarf studied here, primarily due to the small horizontal fluctuations of the thermodynamic variables. Even in the Sun --- with the much higher temperature contrast present at its surface --- effects related to horizontal temperature inhomogeneities are rather subtle \\citep[cf.][]{Steffen+Ludwig99}. We expect that 1D stellar atmosphere models provide an acceptable overall approximation to the spectrum of main-sequence objects between the Sun and late M-dwarfs. Only if one demands for a precision exceeding commonly adopted levels or desires to study effects in principle not included in standard model atmospheres (e.g. spectral line shifts and asymmetries), one has to go to more sophisticated modeling. We emphasize that this refers to objects at solar metallicity. For metal poor objects the situation is clearly different \\citep[see][]{Asplund+al99b}.  A downside of our findings is that we cannot trace back the remaining differences between theoretical and observed colors for M-dwarfs to shortcomings in the treatment of the convective energy transport. The resolution of the problem has to be found elsewhere, as pointed out earlier, deficiencies in the molecular opacities are still a possible option.  We are left with the problem of finding an adequate value of the mixing-length parameter to obtain a description of the vertical temperature run in the superadiabatic layers. In the present case we find a mixing-length parameter \\mbox{\\mlp($\\Delta s$)=2.1} which gives a match to the entropy jump and the temperature gradient of the deeper atmospheric layers (see Fig.~\\ref{f:salpha}). Despite perhaps the best value to be employed in stellar atmosphere calculations, for global stellar structure models a value of \\mbox{\\mlp($\\Delta s$)=1.5} would be more appropriate since it ensures to find the correct asymptotic entropy in the convective envelope which is important for obtaining the correct stellar radius. To complete the ``zoo'' of mixing-length parameters, we get a value of \\mbox{\\mlp($v$)=3.5} when matching the convective velocities predicted in our hydrodynamical models. We reiterate that all values are given with reference to the formulation of MLT by \\citet{Mihalas78}.  The various values of the mixing-length parameter point towards the deeper rooted problem that MLT can give a reasonable but not exact description of the average convective properties. Even calibrating one aspect does not ensure the overall correct functional form of, say, the temperature profile. We have seen that different formulations of MLT can give quite different functional dependencies. They offer the possibility to improve fits beyond the quality limited by fitting the mixing-length parameter only.  For late M-dwarfs all this does not matter much since the differences of the atmospheric structure for various values of \\mlp\\ are small. However, we stress that this statement refers to cooler M-dwarfs on or close to the main-sequence. For pre-main-sequence (PMS) objects the situation is markedly different \\citep{Baraffe+al02}. There the specific choice of \\mlp\\ has a large impact on the resulting atmospheric structure. Work is underway to extend the present study into this regime which may also allow us to find the most suitable MLT formulation.  A result beyond the scope of classical model atmospheres is the derivation of a proxy of atmospheric mixing-time scales (see Fig~\\ref{f:massex}) due to convective overshoot. We find an exponential ``leaking'' of the convective velocity field into the formally stably stratified layers. Depending on the exact criterion, overshooting extends the efficiently mixed regions about 2 pressure scale heights beyond the Schwarzschild boundary. We suggest that the mixing found in the $\\Teff\\approx 2800\\pun{K}$ model studied here, takes place in an analogous fashion in brown dwarfs, and provides the mixing which counteracts dust sedimentation. Hydrodynamical models can be used to address this problem more directly by performing simulations including the formation and transport of dust which we consider as an interesting challenge for the future.  Last but not least we would like to point out the two weakest points of our investigation. While the precision of the radiative transfer in the hydrodynamical calculations is sufficient to address the questions discussed here there is certainly room for improvement of the OBM to get an even closer agreement with detailed spectral synthesis calculations. Secondly, the models presented here are rather shallow. Improvements in the formulation of the lower boundary condition would perhaps allow the use of deeper computational domains, and would reduce the influence of the specific formulation of the lower boundary conditions."
    },
    "0208/astro-ph0208067_arXiv.txt": {
        "abstract": "{The X-ray afterglows of GRB\\,001025A and GRB\\,010220 were detected by \\emph{XMM-Newton} with an average 0.2--10.0\\,keV flux of 4.4 and $3.3\\times10^{-14}$\\,erg\\,cm$^{-2}$\\,s$^{-1}$ respectively; the afterglow of GRB\\,001025A is observed to decay.  Afterglows at other wavelengths were not detected for either burst.  A set of broadened soft X-ray emission lines are detected in the afterglow of GRB\\,001025A, at $5.0\\,\\sigma$ significance above a Galactic-absorbed power-law continuum.  The spectra of both afterglows are significantly better fit by a variable abundance thermal plasma model than by an absorbed power-law and are consistent with the observations of GRB\\,011211, indicating that thermal emission from light elements may be common in the early X-ray afterglows of GRBs. ",
        "introduction": "} Much of the recent progress in the understanding of gamma-ray bursts (GRBs) has come from bursts detected with good spatial accuracy with \\emph{BeppoSAX} and it is particularly at X-ray wavelengths that GRB afterglows are detected \\citep{2001grba.conf...97P}, about half producing no detectable optical afterglow emission \\citep{2001A&A...369..373F}.  It is also only at X-ray wavelengths that emission lines are detected in afterglows, allowing firm estimates to be made of the cosmological redshifts and the outflow velocities of the afterglow material \\citep{2000Sci...290..955P,2002Natur.416..512R}. \\emph{XMM-Newton} \\citep{2001A&A...365L...1J}, with its large effective area, is particularly suited to this work.  Previous detections of emission lines in GRB afterglows with \\emph{BeppoSAX} and \\emph{Chandra} have concentrated on emission from highly-ionised iron \\citep{1998A&A...331L..41P,2000Sci...290..955P,2000ApJ...545L..39A,2001ApJ...557L..27Y}; however recent observations with \\emph{XMM-Newton} have revealed several emission lines at lower energies \\citep{2002Natur.416..512R}.  Most plausible mechanisms for the production of a GRB involve a newly-formed black hole surrounded by a short-lived accretion disk regardless of the progenitor \\citep{1998ApJ...494L..45P,% 1999ApJ...524..262M,1999A&A...344..573R,2001Sci...291...79M}. Recent evidence suggests that the progenitors of long-duration GRBs are massive stars \\citep{1998ApJ...494L..45P,1999Natur.401..453B,% 1999ApJ...524..262M,2002Natur.416..512R}.  Various models have been proposed to account for the emission spectra (in particular the claim of a high equivalent width Fe emission line \\citep{2000ApJ...545L..39A}) and lightcurves of the afterglow.  For instance, the nearby reprocessor model \\citep{2000ApJ...545L..73R,2001ApJ...559L..83B,astro-ph/0110654} involving reflection of synchrotron emission from the walls of a cone tunnelled out of a massive star, yielding large equivalent width Fe emission lines; or a `supranova' model \\citep{1998ApJ...507L..45V} which invokes a time delay ($\\gtrsim30$ days) between an initial supernova (SN) explosion and the GRB, giving a spectrum briefly dominated either by the recombination of Fe in a photoionised plasma \\citep{1998ApJ...507L..45V} or reflection of synchrotron emission off the walls of a wide funnel excavated in the SN remnant \\citep{2001ApJ...550L..43V}.  Recently, \\citet{2002Natur.416..512R} have suggested that the early X-ray afterglow spectrum of GRB\\,011211 is dominated by thermal emission from a metal-enriched, but notably Fe-poor collisionally-ionised plasma ejected in a recent SN explosion and heated by the GRB.  In Sect.~\\ref{observations} we report on observations of two GRB afterglows with \\emph{XMM-Newton}, presenting the spectra in Sect.~\\ref{results}.  In Sect.~\\ref{discussion} these results are discussed and their implications for other X-ray observations of afterglows examined. Our conclusions are in Sect.~\\ref{conclusions}.  Unless otherwise stated, all errors quoted are 90\\% confidence limits for one parameter of interest.  A cosmology where H$_0=75$\\,km\\,s$^{-1}$\\,Mpc$^{-1}$ and q$_0=0.5$ is assumed throughout. ",
        "conclusions": "} \\emph{XMM-Newton} detected the afterglows of GRB\\,001025A and GRB\\,010220 the former showing a decaying lightcurve.  Positions accurate to\\ $\\sim1$\\arcsec\\ were determined; this makes it feasible for optical spectroscopy to be used to confirm the X-ray redshift reported in this paper for the host galaxies of these GRBs, in particular for the host of GRB\\,001025A, where the extinction is relatively low.  In both cases, the X-ray spectra are significantly better fit by the thermal plasma model proposed by \\citet{2002Natur.416..512R} to explain the line features in GRB\\,011211 than by an absorbed power-law and in the case of GRB\\,001025A the thermal plasma model is a significantly better fit than an ionised reflection model.  The parameters determined from these thermal fits are consistent with their suggested scenario.  It seems likely that a large contribution from highly-ionised light metals is a common feature in the X-ray spectra of GRB afterglows hours to days after the burst and the detection of thermal emission in three of the four \\emph{XMM-Newton}--detected afterglows implies that this may be a significant component in the total afterglow luminosity of all long GRBs."
    },
    "0208/astro-ph0208251_arXiv.txt": {
        "abstract": "{For the first time the large-scale clustering and the mean abundance of galaxy clusters are analysed simultaneously to get precise constraints on the normalized cosmic matter density $\\Omega_m$ and the linear theory RMS fluctuations in mass $\\sigma_8$. A self-consistent likelihood analysis is described which combines, in a natural and optimal manner, a battery of sensitive cosmological tests where observational data are represented by the (Karhunen-Lo\\'{e}ve) eigenvectors of the sample correlation matrix. This method breaks the degeneracy between $\\Omega_m$ and $\\sigma_8$. The cosmological tests are performed with the ROSAT ESO Flux-Limited X-ray (REFLEX) cluster sample. The computations assume cosmologically flat geometries and a non-evolving cluster population mainly over the redshift range $0<z<0.3$. The REFLEX sample gives the cosmological constraints and their $1\\sigma$ random errors of $\\Omega_m\\,=\\,0.341\\,^{+0.031}_{-0.029}$ and $\\sigma_8\\,=\\,0.711\\,^{+0.039}_{-0.031}$. Possible systematic errors are evaluated by estimating the effects of uncertainties in the value of the Hubble constant, the baryon density, the spectral slope of the initial scalar fluctuations, the mass/X-ray luminosity relation and its intrinsic scatter, the biasing scheme, and the cluster mass density profile. All these contributions sum up to total systematic errors of $\\sigma_{\\Omega_m}=^{+0.087}_{-0.071}$ and $\\sigma_{\\sigma_8}=^{+0.120}_{-0.162}$. ",
        "introduction": "Several observational arguments suggest that we live in a geometrically flat Universe in a phase of accelerated cosmic expansion (Riess et al. 1998, Perlmutter et al. 1999, Stompor et al. 2001, Netterfield et al. 2002, Pryke et al. 2002, Sievers et al. 2002). Strong indications for a low cosmic matter density come from, e.g., the abundance of galaxy clusters (e.g., Carlberg et al. 1996, Viana \\& Liddle 1996, Bahcall \\& Fan 1998, Eke et al. 1998, Henry 2000, Borgani et al. 2001, Reiprich \\& B\\\"ohringer 2002, hereafter RB02), and from the large-scale distribution of clusters (e.g. Collins et al. 2000, Schuecker et al. 2001, hereafter Paper\\,I) and galaxies (e.g., Percival et al. 2001, Szalay et al. 2001). A recent overview including also new results on gravitational dynamics, weak lensing and the baryon mass fraction in clusters of galaxies can be found in Peebles \\& Ratra (2002).  With the present investigation we are improving the constraints on the matter density $\\Omega_m$ and the linear theory RMS matter fluctuations $\\sigma_8$ in comoving spheres with $8\\,h^{-1}\\,{\\rm Mpc}$ radius. We base our approach on observational estimates of spatial clustering and abundance of galaxy clusters. Simultaneous constraints on $\\Omega_m$ and $\\sigma_8$ are derived under the assumption of a flat geometry of the Universe. Cluster abundances although almost insensitive to the cosmological constant $\\Omega_\\Lambda$ are known to be quite sensitive to $\\Omega_m$ and actually one of the best ways of measuring $\\sigma_8$ (White et al. 1993, Eke et al. 1996, Kitayama \\& Suto 1997, Borgani et al. 1997, Mathiesen \\& Evrard 1998). Our method uses the statistical information of the present epoch large-scale structure to characterize the cosmological model. In comparison to other methods of the assessment of the large-scale structure e.g. through the galaxy distribution, the analysis of the galaxy cluster distribution has the advantage that it relies on astrophysics mostly governed by gravitation, so that cluster masses and biasing factors can be obtained from first principles, especially when the clusters are selected in X-rays (e.g., Borgani \\& Guzzo 2001).  We use the ROSAT ESO Flux-Limited X-ray (REFLEX) galaxy cluster sample. In the REFLEX survey special care was taken to get a well-controlled sample selection and a high completeness (B\\\"ohringer et al. 2001). The sample consists of the 452 X-ray brightest southern clusters with redshifts mainly below $z=0.3$, selected in X-rays from the ROSAT All-Sky Survey (RASS, Voges et al. 1999) and is confirmed by extensive optical follow-up observations within a large ESO Key Programme (B\\\"ohringer et al. 1998, Guzzo et al. 1999). The sample has been used to measure with unrivalled accuracy the cluster X-ray luminosity function (B\\\"ohringer et al. 2002), the spatial cluster-cluster correlation function (Collins et al. 2000), and its power spectrum (Paper\\,I).  In order to get accurate constraints on $\\Omega_m$ and $\\sigma_8$ we use several independent cosmological tests based on the dependence of the cosmic matter power spectrum and the volume on the values of the cosmological parameters. On the one side we measure the spatial fluctuations of galaxy clusters. This test is known to be especially sensitive to the shape of the matter power spectrum on large scales. On the other side we measure the average abundance of clusters. This test is known to be especially sensitive to the normalization of the matter power spectrum on small scales. Finally and independent from the matter power spectrum, we count clusters to measure their abundance as a function of redshift and thus to measure the volume which again strongly depends on cosmology (redshift-volume test, e.g., Robertson \\& Noonan 1968).  A natural way to combine the different tests is offered by the Karhunen-Lo\\'{e}ve (KL) eigenvector analysis. The method was first applied by Bond (1995) to analyse cosmic microwave background (CMB) temperature maps and translated by Vogeley \\& Szalay (1996) to the case of the spatial analysis of galaxies. The KL eigenvectors $\\Psi$ are constructed in a way to obey orthogonality properties which allow unbiased studies of fluctuations up to the largest scales covered by a survey. The method avoids artifical correlations introduced by the survey window between different fluctuation modes, affecting under realistic conditions all power spectral analyses based on plane-wave expansions on large scales. Applications of the KL method to galaxy surveys can be found in Matsubara, Szalay \\& Landy (2000) and Szalay et al. (2001). Their KL tests are, however, still restricted to the analysis of the clustering properties of galaxies.  To determine a precise eigenvector base $\\Psi$ which does not bias the final results, a more refined fiducial cosmological model has to be assumed which is already close to our best model fits (the construction of the fiducial model is described in Sect.\\,\\ref{PRAC}): a pressure-less spatially flat Friedmann-Lema\\^{\\i}\\-tre model, the cosmic matter density $\\Omega_m=0.3$, the cosmological constant $\\Omega_\\Lambda=0.7$, the Hubble constant in the form $h=H_0/100\\,{\\rm km}\\,{\\rm s}^{-1}\\,{\\rm Mpc}^{-1}=0.7$, the linear RMS normalization $\\sigma_8=0.75$, the spectral index of initial scalar fluctuations $n_{\\rm S}=1.0$, the baryon density $\\Omega_bh^2=0.020$, and the CMB temperature $T_{\\rm CMB}=2.728$\\,K.  In our first KL study of the REFLEX sample we followed the approach of A.\\,Szalay and collaborators, applying the KL method to the analysis of three-dimensional spatial fluctuations (Schuecker et al. 2002, hereafter Paper\\,II).  The resulting constraints on the matter density, $0.03<\\Omega_mh^2<0.19$ (95\\% confidence interval), were found to be quite robust against major changes in the model assumptions.  The present investigation completes our previous KL study by utilizing now both the spatial fluctuations of galaxy clusters and their mean abundance to get constraints on $\\Omega_m$ and $\\sigma_8$.  Some general features of the KL method relevant for the present work are outlined in Sect.\\,\\ref{METHOD}. In Sect.\\,\\ref{MODEL} we briefly describe a model for the matter and cluster power spectrum already developed in Paper\\,II. We further introduce a new theoretical model for the average cluster abundance which replaces the more phenomenological prescription used in Paper\\,II. The REFLEX sample and the values of important model parameters are discussed in Sect.\\,\\ref{PRAC}. Sect.\\,\\ref{RESULTS} summarizes the cosmological constraints obtained with the REFLEX sample. In Sect.\\,\\ref{CONCLUSION} we discuss the results and draw our basic conclusions. In the Appendix we derive equations used to relate cluster masses defined for different density contrast and background levels. ",
        "conclusions": "In the present investigation cluster number counts and their spatial fluctuations are analysed for the first time simultaneously in order to constrain cosmological models. The test utilizes the complementarity of clustering and abundance of galaxy clusters and allows us to fully exploit the cosmological potential of the REFLEX survey of X-ray clusters (but see below).  For spatially flat cosmologies as suggested by all recent CMB measurements (Sect.\\,\\ref{INTRO}) and for a non-evolving cluster population as suggested by the $z$-independent comoving REFLEX cluster number densities (Paper\\,I), we obtain $\\Omega_m=0.341$ for the cosmic matter density and $\\sigma_8=0.711$ for the normalization of the matter power spectrum. The random errors range between 8-9\\% for $\\Omega_m$ and 4-5\\% for $\\sigma_8$, not including cosmic variance. Systematic errors can be identified and quantified so that the whole process appears to be well-controlled. Under the assumption that the systematics add quadratically the total systematic errors are found to be about 2.5 and four times larger than the random errors of $\\Omega_m$ and $\\sigma_8$, respectively.  Compared to the results obtained with the fluctuations only (Paper\\,II), the combination of abundance and fluctuation measurements reduces the random errors of $\\Omega_m$ by a factor of about two and of $\\sigma_8$ by a factor of about nine. The main driver of the accuracy of the present results thus appears to be the cluster abundance, especially for $\\sigma_8$ (but see below).  In contrast to many other cosmological parameter estimations we tested the assumed functional form of the sample likelihood function. As in Paper\\,II we could verify Gaussianity on large scales -- a fundamental property of the cosmic matter field. In this respect the present analysis appears to be fully consistent and self-contained.  \\begin{figure*} \\vspace{-0.0cm} \\centerline{\\hspace{-9.5cm} \\psfig{figure=h3921f8a.ps,height=8.5cm,width=8.5cm}} \\vspace{-8.5cm} \\centerline{\\hspace{ 8.2cm} \\psfig{figure=h3921f8b.ps,height=8.5cm,width=8.5cm}} \\vspace{-0.65cm} \\caption{\\small Comparison of the reference KL solution ($\\Omega_{\\rm tot}=1$, $\\Omega_m=0.341$, $\\Omega_bh^2=0.020$, $h=0.70$, $n_{\\rm S}=1.0$, $\\sigma_8=0.711$, $\\sigma_{\\rm eff}=25\\%$, ST biasing) with REFLEX observations. {\\bf Left:} Power spectrum obtained with the flux-limited REFLEX sample (Paper\\,I, points with $1\\sigma$ error bars including cosmic variance) and the KL solution (continuous line). The theoretical model takes into account the effects of the different volumes covered by the present KL and the former Fourier analysis, includes the effects of the baryons, and is transformed into redshift space using the nonlinear model described in Paper\\,I. In order to make the measured power spectrum less crowded adjacent power spectral densities and their errors were averaged. {\\bf Right:} Redshift histogram of the REFLEX subsample used here (steps) and the KL solution (continuous line). } \\label{FIG_COMP} \\end{figure*}  The KL likelihood contours (Figs.\\,\\ref{FIG_LOG1} to \\ref{FIG_LOG5}) show some correlations between $\\Omega_m$ and $\\sigma_8$, probably indicating that the primordial part of the power spectrum is still less constrained by the REFLEX data. The correlations are, however, restricted to comparatively small $(\\Omega_m,\\sigma_8)$ ranges so that it does not make much sense to determine an $\\Omega_m$-$\\sigma_8$ relation from the data. Therefore, the degeneracy between $\\Omega_m$ and $\\sigma_8$ can be regarded as broken. The breaking of the degeneracy by the KL method using both cluster abundance and fluctuations is nicely seen when the KL reference results (Fig.\\,\\ref{FIG_LOG4}, left) are compared with the likelihood contours obtained with the REFLEX cluster abundance only (Fig.\\,\\ref{FIG_ABUN}).  Generally, likelihood analyses which optimize simultaneously two or more almost independent observational quantities like the power spectrum and the cluster abundance, try to find the best compromise between the individual observables and not necessarily the best fit of the different components. However, Fig.\\,\\,\\ref{FIG_COMP} shows that the best KL solution obtained with the REFLEX data is consistent with both the redshift histogram and the redshift-space power spectrum obtained with a standard Fourier method (Paper\\,I). Note that the power spectrum was measured within a comparatively small volume to reduce correlations between different fluctuation modes. A further argument supporting the reliability of the present cosmological constraints is given by numerical fits of the amplitude of the observed power spectrum of three volume-limited REFLEX subsamples using $\\Lambda$CDM models which yield $\\sigma_8=0.70$ (Paper\\,I). This result is at variance to the standard normalization of the $\\Lambda$CDM model, but in very good agreement with the results of the present KL analysis.  The value of the cosmic matter density obtained with REFLEX is in good agreement with other recent independent measurements (see also Turner 2002). For example, Pryke et al. (2002) measured with DASI a matter density of $\\Omega_m=0.40\\pm0.15$ for $h=0.72\\pm0.08$ and Sievers et al. (2002) measured with CBI $\\Omega_m=0.62\\pm0.22$ (weak priors). Netterfield et al. (2002) obtained for flat geometries $\\Omega_m=0.33\\pm0.05$, combining COBE-DMR, BOOMERANG, large-scale structure, and supernovae type Ia data.  Sievers et al. (2002) found from the combination of COBE, BOOMERANG, MAXIMA-1, DASI, CBI, large-scale structure surveys, Hubble parameter determinations, and supernovae type Ia the value $\\Omega_m=0.32\\pm0.06$. With the galaxy power spectrum of the 2dF Galaxy Redshift Survey and a new compilation of CMB data (COBE, BOOMERANG, MAXIMA-1, DASI, VSA see Scott et al. 2002, CBI see Mason et al. 2002), Percival et al. (2002) obtained for a flat cosmology $\\Omega_m=0.313\\pm0.055$ (all $1\\sigma$ errors without cosmic variance). Therefore, a significant number of independent applications appear to converge to a low matter density.  The linear theory $\\sigma_8$ obtained with REFLEX is lower than the COBE normalization and most estimates from cluster abundance as summarized in, e.g., Table\\,6 in RB02. Note that values of the effective scatter $\\sigma_{\\rm eff}$ of the mass/X-ray luminosity relation larger than assumed here would increase this descrepancy even more. However, small $\\sigma_8$ values were also obtained by Markevitch (1998) using a local cluster sample with ASCA X-ray temperatures and ROSAT X-ray luminosities, by Borgani et al. (2001) using the ROSAT Deep Cluster Survey, by Seljak (2001) from the empirical mass/X-ray temperature relation of nearby clusters, by Viana, Nichol \\& Liddle (2002) using SDSS/RASS data and the X-ray luminosity function of REFLEX clusters of galaxies, and by Bahcall et al. (2002) using clusters selected from SDSS data. In this context it is also interesting to note that the KL results are in very good agreement with the new constraints obtained by Lahav et al. (2002) from their simultaneous analysis of the amplitudes of the fluctuations probed by the 2dF Galaxy Redshift Survey and COBE, BOOMERANG, MAXIMA-1, DASI data, and with the constraints obtained with the KL analysis of the early Sloan Digital Sky Survey by Szalay et al. (2001) obtained under similar priors (as described in Lahav et al.). There is thus increasing observational evidence for a $\\sigma_8$ value lower than previously, although some recent weak lensing results show a tendency for a higher $\\sigma_8$ for given $\\Omega_m$ (e.g., Hoekstra et al. 2002, van Waerbeke et al. 2002). Small $\\sigma_8$ values help (James S. Bullock, private communication) to reduce the apparent problem of the $\\Lambda$CDM model with the standard normalization to overpredict the number of low-mass satellite galaxies (Klypin et al. 1999, Moore et al. 1999).  In all the cosmological studies mentioned above including the present, various model corrections might change the final results by 10\\% to 30\\% so that more detailed analyses of possible sources of systematic errors are needed to get below this error level. Our tests for systematic errors show that the largest systematic errors are introduced by the mass/X-ray luminosity relation (as expected). In order to use the full potential of REFLEX-like cluster surveys it is thus important to have more precise measurement of this function, i.e., its shape, intrinsic scatter, redshift-dependency, etc. which are expected to be provided by detailed observations with the Chandra and XMM satellites. Future REFLEX papers will test carefully the residual effects of possible evolution of nearby clusters, the cosmological constant and related quantities (H. B\\\"ohringer et al., P. Schuecker et al., in preparation).  Regardless of these interim restrictions, the present investigation illustrates that large and fair samples of X-ray clusters of galaxies give quite clean measurements of the cosmological parameters. In addition to the abovementioned quality of $\\sigma_8$ constraints traditionally obtained with clusters of galaxies, the $1\\sigma$ random error for the matter density of $\\sigma_{\\Omega_m}=0.030$ obtained with the REFLEX data is already close to the random error of $\\sigma_{\\Omega_m}=0.010$ expected to be attainable with the CMB Planck Surveyor satellite and similar to $\\sigma_{\\Omega_m}=0.015$ attainable with the SNfactory plus SNAP supernovae satellite projects (see, e.g., Hannestad \\& M\\\"ortsell 2002). Moreover, the cosmological constraints obtained with cluster data have degeneracies different from high-$z$ supernovae and CMB anisotropies (Holder, Haiman \\& Mohr 2001). Therefore, galaxy clusters can play a significant r\\^{o}le in high precision measurements of cosmological parameters."
    },
    "0208/astro-ph0208298_arXiv.txt": {
        "abstract": "Abundances for 5 molecules (C$^{18}$O, CS, NH$_{3}$, H$_{2}$CO, and C$_{3}$H$_{2}$) and 1 molecular ion (N$_{2}$H$^{+}$) and upper limits for the abundances of 1 molecule ($^{13}$CO) and 1 molecular ion (HCO$^{+}$) are derived for gas within the Bok globule Barnard 68 (B68).  The abundances were determined using our own BIMA millimeter interferometer data and single-dish data gathered from the literature, in conjunction with a Monte Carlo radiative transfer model.  Since B68 is the only starless core to have its density structure strongly constrained via extinction mapping, a major uncertainty has been removed from these determinations.  All abundances for B68 are lower than those derived for translucent and cold dense clouds, but perhaps only significantly for N$_{2}$H$^{+}$, NH$_{3}$, and C$_{3}$H$_{2}$. Depletion of CS toward the extinction peak of B68 is hinted at by the large offset between the extinction peak and the position of maximum CS line brightness.  Abundances derived here for C$^{18}$O and N$_{2}$H$^{+}$ are consistent with other, recently determined values at positions observed in common. ",
        "introduction": "Stars form out of cores of dense molecular gas.  Such cores can be embedded deeply within molecular clouds, but numerous examples of isolated cores, e.g., Bok globules (Bok \\& Reilly 1947; Clemens \\& Barvainis 1988), have been also identified as dark patches of high visual extinction against rich stellar backgrounds.  As isolated cores, Bok globules can be relatively free from the influences of nearby star-forming events, and so can provide very simple and pristine examples of the physical and chemical structures of cores prior to the formation of low-mass stars.  In this paper, we examine the abundances of several species of molecules or molecular ions in the globule Barnard 68 (B68; Barnard 1919).  B68 (also known as CB82 or LDN57) is an Ophiuchus dark cloud, located at $\\sim$125~pc (de Geus, de Zeeuw, \\& Lub 1989)\\footnote{A distance range of 60-205 pc for the Ophiuchus dark clouds was formally estimated by de Geus et al. with 125 pc $\\pm$ 25 pc defined as their center.}, within the Loop I superbubble.  Figure 1 shows an $R$-band image of B68 obtained from the CDS/Aladin sky atlas, revealing its compact, roundish morphology.  B68 has no evidence of star formation within its interior, e.g., from detections of outflow wings in lines of $^{12}$CO (Avery et al. 1987) or from an IRAS point source (Parker 1988)\\footnote{The 24\\as\\ $\\times$ 5\\as\\ (P.A. of 94\\deg) error ellipse of the IRAS point source 17194-2351, associated with B68 by Clemens \\& Barvainis, is centered $\\sim$3\\am\\ beyond the opaque edge of B68 seen in Figure 1.}.  Furthermore, continuum emission at $\\lambda$ = 1.3 mm toward B68 has not been detected (Reipurth, Nyman, \\& Chini 1996; Launhardt \\& Henning 1997).  B68 is unique in that it is the only starless Bok globule to have its density structure well defined.  Using the extinction of the background Galactic bulge K-giant population, Alves, Lada, \\& Lada (2001; ALL01) found an azimuthally-averaged radial column density profile for B68 that was well matched with that of an isothermal, self-gravitating Bonnor-Ebert sphere, specifically one with a near-critical center-to-edge density contrast of 16.5 (Bonnor 1956; Ebert 1955).  The robust definition of density within B68 removes a major uncertainty towards determining its molecular abundances.  Recently, Bergin et al.\\/ (2002) used single-dish maps of B68 in conjunction with models specifically incorporating the ALL01 Bonnor-Ebert density structure to ascertain radial variations in C$^{18}$O and N$_{2}$H$^{+}$ abundance, with the lowest values occurring at the highest extinctions within B68.  Hotzel et al.\\/ (2002) also found evidence for depletion of CO isotopes with extinction from their own single-dish maps and a re-analysis of the ALL01 data.  Each group speculated these molecules sublimate from the gas phase onto dust grains at the high densities and low temperatures in the B68 interior.  A similar interpretation was made by Tafalla et al.\\/ (2002) to explain the radial depletion of CS and C$^{18}$O in the interiors of 5 other starless cores, although N$_{2}$H$^{+}$ was not depleted and NH$_{3}$ was enhanced at these locations.  We explore further the abundances of several molecules or molecular ions within B68, incorporating the ALL01 Bonnor-Ebert density structure into our own models.  New observations of B68 made with the BIMA millimeter array are used to obtain upper limits on abundances of $^{13}$CO and HCO$^{+}$ near the extinction peak found by ALL01, i.e., at R.A./decl. (2000) = 17$^{h}$22$^{m}$38.6$^{s}$, -23$^{\\circ}$49$^{\\prime}$46.0$^{\\prime\\prime}$. Upper limits to C$^{18}$O and N$_{2}$H$^{+}$ abundances are also derived at this position, consistent with values derived by Bergin et al.  In addition, single-dish detections of lines reported in the literature are used to derive abundance values for C$^{18}$O, CS, N$_{2}$H$^{+}$, NH$_{3}$, H$_{2}$CO, and C$_{3}$H$_{2}$ toward positions in B68 offset from the extinction peak.  We describe the BIMA observations and single-dish data in \\S 2, and our method of determining abundances with a Monte Carlo radiative transfer code in \\S 3. The derived results are discussed in \\S 4.  A concluding summary is found in \\S 5. ",
        "conclusions": "Using our own interferometer line data and single-dish line data culled from the literature of the Bok globule B68 and a Monte-Carlo radiative transfer code, we derive abundances for C$^{18}$O, CS, N$_{2}$H$^{+}$, NH$_{3}$, H$_{2}$CO, and C$_{3}$H$_{2}$ and upper limits for abundances of $^{13}$CO and HCO$^{+}$.  Foreknowledge of the Bonnor-Ebert density configuration of B68, as found from extinction mapping by ALL01, has removed a major uncertainty that affected previous estimations of abundances in other clouds.  We find molecular abundances in B68 are lower than those estimated previously for clouds similar to B68.  N$_{2}$H$^{+}$, NH$_{3}$, and C$_{3}$H$_{2}$ have abundances lower than the lowest values determined for other clouds by 1.4, 1.5, and 2.2 orders of magnitude respectively, but abundances of the other species are only lower by one order of magnitude or less.  Depletion of CS is suggested by the large offset between the position of maximum CS line brightness and the extinction peak.  Furthermore, abundances of C$^{18}$O and N$_{2}$H$^{+}$ derived using data from the literature at locations where line characteristics were reported are consistent with those derived by Bergin et al.  As shown by Bergin et al., fully-sampled maps from single-dish telescopes can more effectively determine abundances within B68.  Further maps of B68 can be now easily obtained using the new multi-beam focal-plane arrays on current single-dish telescopes.  Furthermore, the upcoming generation of millimeter interferometers will have the sensitivity to provide additional data for combined maps of even higher resolution.  With these data, more accurate molecular abundances will be determined for B68.  For example, it will be interesting to determine if abundances of CS, HCO$^{+}$, NH$_{3}$, H$_{2}$CO, and C$_{3}$H$_{2}$ also vary with extinction, like $^{13}$CO, C$^{18}$O, or N$_{2}$H$^{+}$.  Similar abundance determinations across other isolated dense cores may also be possible, provided their density structures can be similarly well-defined."
    },
    "0208/astro-ph0208192_arXiv.txt": {
        "abstract": "The Sunyaev-Zel'dovich effect (SZE) provides a unique way to map the large-scale structure of the universe as traced by massive clusters of galaxies. As a spectral distortion of the cosmic microwave background, the SZE is insensitive to the redshift of the galaxy cluster, making it well-suited for studies of clusters at all redshifts, and especially at reasonably high redshifts ($z > 1$) where the abundance of clusters is critically dependent on the underlying cosmology. Recent high signal-to-noise detections of the SZE have enabled interesting constraints on the Hubble constant and the matter density of the universe using small samples of galaxy clusters. Upcoming SZE surveys are expected to find hundreds to thousands of new galaxy clusters, with a mass selection function that is remarkably uniform with redshift.  In this review we provide an overview of the SZE and its use for cosmological studies with emphasis on the cosmology that can, in principle, be extracted from SZE survey yields. We discuss the observational and theoretical challenges that must be met before precise cosmological constraints can be extracted from the survey yields. ",
        "introduction": "\\label{sec:intro}  The Sunyaev-Zel'dovich Effect (SZE) offers a unique and powerful observational tool for cosmology.  Recently, there has been considerable progress in detecting and imaging the SZE.  Efforts over the first two decades after the SZE was first proposed in 1970 \\citep{sunyaev70,sunyaev72} yielded few reliable detections.  Over the last decade, new detectors and observing techniques have allowed high quality detections and images of the effect for more than 50 clusters with redshifts as high as one.  The next generation of SZE instruments that are now being built or planned will be orders of magnitude more efficient. Entering the fourth decade of SZE observations, we are now in position to exploit fully the power of the SZE, by obtaining detailed images of a set of clusters to understand the intra-cluster medium (ICM), by obtaining large SZE samples of clusters to determine statistically robust estimates of the cosmological parameters and, most importantly, by conducting large untargeted SZE surveys to probe the high redshift universe. These surveys will provide a direct view of the growth of large scale structure and will provide large catalogs of clusters that extend past $z \\sim 2$ with remarkably uniform selection functions.  The physics of the SZE has been covered well in previous reviews \\citep{birkinshaw99, rephaeli95,sunyaev80}, with \\citet{birkinshaw99} and \\citet{carlstrom00} providing recent reviews of the observations. In this review, we look to the near future, using recent observations as a guide to what we can expect.  The SZE is best known for allowing the determination of cosmological parameters when combined with other observational diagnostics of clusters of galaxies such as X-ray emission from the intracluster gas, weak and strong lensing by the cluster potential, and optical galaxy velocity dispersion measurements.  For example, cluster distances have been determined from the analysis of SZE and X-ray data, providing independent estimates of the Hubble constant. A large homogeneous sample of galaxy clusters extending to high redshift should allow a precise measure of this number, as well as a measure of the angular diameter distance relation to high redshift where it is highly sensitive to cosmological parameters.  Similarly, the SZE and X-ray measurements will allow tight constraints on cluster gas mass fractions which can be used to estimate $\\Omega_M$ assuming the composition of clusters represents a fair sample of the universal composition. The observed redshift dependence of the gas mass fraction can also be used to constrain cosmological parameters as well as test speculative theories of dark matter decay.   The most unique and powerful cosmological tool provided by the exploitation of the SZE will likely be the direct measurement of the evolution of the number density of galaxy clusters by deep, large scale SZE surveys.  The redshift evolution of the cluster density is critically dependent on the underlying cosmology, and in principle can be used to determine the equation of state of the dark energy. SZE observations are particularly well suited for deep surveys because the important parameter that sets the detection limit for such a survey is the mass of the cluster; SZE surveys will be able to detect all clusters above a mass limit independent of the redshift of the clusters.  This remarkable property of SZE surveys is due to the fact that the SZE is a distortion of the cosmic microwave background (CMB) spectrum. While the CMB suffers cosmological dimming with redshift, the ratio of the magnitude of the SZE to the CMB does not; it is a direct, redshift independent measurement of the ICM column density weighted by temperature, i.e., the pressure integrated along the line of sight. The total SZE flux detected will be proportional to the total temperature-weighted mass (total integrated pressure) and, of course, inversely proportional to the square of the angular diameter distance. Adopting a reasonable cosmology and accounting for the increase in the universal matter density with redshift, the mass limit for a given SZE survey flux sensitivity is not expected to change more than a factor of $\\sim 2-3$ for any clusters with $z>0.05$.   SZE surveys therefore offer an ideal tool for determining the cluster density evolution. Analyses of even a modest survey covering $\\sim$~10 square degrees will provide interesting constraints on the matter density of the universe. The precision with which cosmological constraints can be extracted from much larger surveys, however, will be limited by systematics due to our insufficient understanding of the structure of clusters, their gas properties and evolution.  Insights into the structure of clusters will be provided by high resolution SZE observations, especially when combined with other measurements of the clusters.  Fortunately, many of the cluster properties derived directly from observational data can be determined in several different ways.  For example, the gas mass fraction can be determined by various combinations of SZE, X-ray, and lensing observations. The electron temperature, a direct measure of a cluster's mass, can be measured directly through X-ray spectroscopy, or determined through the analysis of various combinations of X-ray, SZE, and lensing observations.  Several of the desired properties of clusters are therefore over-constrained by observation, providing critical insights to our understanding of clusters, and critical tests of current models for the formation and evolution of galaxy clusters. With improved sensitivity, better angular resolution, and sources out to $z \\sim 2$, the next generation of SZE observations will provide a good view of galaxy cluster structure and evolution. This will allow, in principle, the dependence of the cluster yields from large SZE surveys on the underlying cosmology to be separated from the dependence of the yields on cluster structure and evolution.   We outline the properties of the SZE in the next section and provide an overview of the current state of the observations in \\S\\ref{sec:obs_status}.  This is followed in \\S\\ref{sec:survey_yields} by predictions for the expected yields of upcoming SZE surveys. In \\S\\ref{sec:sze_cosmo}, we provide an overview of the cosmological tests which will be possible with catalogs of SZE-selected clusters. This is followed by a discussion of backgrounds, foregrounds, contaminants, and theoretical uncertainties that could adversely affect cosmological studies with the SZE and a discussion of observations which could reduce or eliminate these concerns. Throughout the paper, $h$ is used to parametrize the Hubble constant by $H_0 = 100 h$ km s$^{-1}$ Mpc$^{-1}$, and $\\OmM$ and $\\OmL$ are the matter density and vacuum energy density, respectively, in units of the critical density. ",
        "conclusions": "\\label{sec:summary}  The \\sze\\ is emerging as a powerful tool for cosmology. Over the last several years, detection of the \\sze\\ toward massive galaxy clusters has become routine,  as has high quality imaging at moderate angular resolution of order an arcminute. Measurements of the effect already have been used to place interesting constraints on the Hubble constant and, through measurements of cluster gas mass fractions, the matter density of the universe, $\\Om$.  The next step is to exploit the redshift independence of the \\sze\\ signal to conduct blind surveys for galaxy clusters.  The limit for such a survey is essentially a mass limit that is remarkably uniform with redshift. The cluster catalog from such a unbiased survey could be used to greatly increase the precision and redshift range of present \\sze\\ constraints on the Hubble constant and $\\Om$, and could, for example, allow $\\Da(z)$ to be determined to high redshift ($z \\sim 2$).  The most powerful use of the SZE for cosmology will be the measurement of the evolution of the abundance of galaxy clusters. SZE surveys are ideally suited for this since they are able to probe the abundance at high redshift as easily as the local universe. The evolution of the abundance of galaxy clusters is a sensitive probe of cosmology. For example, the yields from a deep \\sze\\ survey covering only ten square degrees would be able to place interesting constraints on $\\Om$, $\\Ol$, and $\\sigma_8$.  A generic prediction of inflation is that the primordial density fluctuations should be Gaussian.  Non-Gaussianity in the form of an excess of high mass clusters should be readily apparent, especially at high redshift, from \\sze\\ survey yields. SZE cluster surveys will therefore probe both the structure formation history of the universe and the nature of the primordial fluctuations. In this way, cluster surveys are emerging as the next serious test of the cold dark matter paradigm.  Current \\sze\\ observations, while routine, require substantial integration time to secure a detection; a prohibitively long time would be required to conduct blind surveys over a large region of sky with the instruments now available. However, the next generation of instruments now being built or planned will be substantially faster. Dedicated interferometric arrays being built will be able to conduct deep SZE surveys over tens of square degrees. Heterogeneous arrays, such as the SZA combined with the OVRO array, will also allow detailed high resolution follow up \\sze\\ observations of the resulting cluster catalog.  A dedicated, low noise, single dish telescope with $\\sim 1'$ resolution, equipped with a next generation, large format bolometric array receiver ($\\sim 1000$ elements) and operating from a superb site would be able to conduct a deep \\sze\\ survey over thousands of square degrees. The statistics provided by the yields from such a large survey ($\\sim 10^4$ clusters) in the absence of systematic effects and assuming redshifts are known would be sufficient to determine precise constraints on $\\Om$, $\\Ol$, $\\sigma_8$, and even set meaningful constraints on the equation of state of the dark energy.  The possible systematics that could affect the yields of \\sze\\ surveys are presently too large to realize the full potential of a deep \\sze\\ survey covering thousands of square degrees. The systematics include, for example, the uncertainties on the survey mass detection limit due to unknown cluster structure and cluster gas evolution, as well as the uncertainties in the theoretical mapping between the initial density field and the number density of clusters of a given mass as a function of redshift, i.e., the mass function.  These systematics can begin to be addressed through detailed follow-up observations of a moderate area \\sze\\ survey (tens of square degrees). High resolution \\sze, X-ray, and weak lensing observations will provide insights into evolution and structure of the cluster gas. Numerical simulations directly compared and normalized to the \\sze\\ yields should provide the necessary improvement in our understanding of the mass function.  It is not unreasonable to consider the possibility of a space-based telescope operating at centimeter through submillimeter wavelengths with high angular resolution ($<1$ arcminute) and good spectral coverage.  For studies of the SZE, this would allow simultaneous determinations of electron column densities, temperatures, and peculiar velocities of galaxy clusters. Such a satellite would make detailed images of the cosmic microwave background, while also providing important information on the high frequency behavior of radio point sources and the low frequency behavior of dusty extragalactic submillimeter sources.  The upcoming {\\it Planck Surveyor} satellite is a first step in this direction; it should provide an SZE all-sky survey although at moderate, $\\sim 5$ arcminute, resolution. Such a survey should find on the order of $10^4-10^5$ clusters, most of them at redshift $z<1$.  We can look forward to the \\sze\\ emerging further as a unique and powerful tool in cosmology over the next several years as the next generation of \\sze\\ instruments come online and \\sze\\ surveys become a reality."
    },
    "0208/astro-ph0208471_arXiv.txt": {
        "abstract": "{ We analysed the light curves of a large sample of long period variables in the LMC from the AGAPEROS catalogue. The (non)regularity of the light change is discussed in detail showing that the majority of the light curves cannot be described properly by a single period. We show that semiregular and small amplitude variability do not necessarily correlate as has been assumed in several previous studies. Using near-infrared data from the DENIS survey we correlate the light change with colours and luminosities of the objects. These results are used to compare long period variables in the LMC with LPVs in the Galactic Bulge and in the solar neighborhood. ",
        "introduction": "\\label{introduction} The late stages of stellar evolution are characterized by regular and irregular light variability, a well-known signature of the stellar pulsation of Asymptotic Giant Branch stars (hereafter AGB).  These light changes allow to identify AGB stars over large distances and to derive the pulsation characteristics (periodicity, etc.), which are key parameters for understanding the fundamental properties of the highly extended atmospheres of these stars.  The pulsational properties have a strong impact on the structure of AGB stars. Pulsation, as the driving mechanism for the stellar winds, plays a key role for the high mass loss rates reached during the AGB phase.  A new era in the study of variable red giants started, when microlensing surveys produced a large amount of light curves of these stars, especially for objects in the LMC. The pioneering work by Wood (\\cite{Wood2000}), using data from the MACHO survey, showed that the red giant variables form four roughly parallel sequences in a period-magnitude diagram.  Three of these sequences could be associated with fundamental, first and second overtone pulsation. The explanation of the fourth sequence is not clear yet (Wood \\cite{Wood2000}, Hinkle et al.~\\cite{Hinkle2002}). Recently, Cioni et al.~(\\cite{Cioni2001}) presented a survey of variable red giants in the LMC based on data from the EROS-2 microlensing survey (Lasserre et al. \\cite{Lasserre2000}). The work of Cioni et al.~focused on the logP$-$K relation confirming three of the relations found by Wood, and they discussed the behaviour of different groups of variables in near infrared colour-magnitude diagrams.  The present paper relies on the variability information contained in the AGAPEROS variable star catalogue (Melchior et al. \\cite{Melchior2000}). Here, we extend and analyse the corresponding light curves, relying on the EROS-1 microlensing survey data set (Ansari et al. \\cite{Ansari95}, Aubourg et al. \\cite{Aubourg95}). These data have been obtained between December 1991 and April 1994. The second half of the data set therefore overlaps with the MACHO survey. We combine the EROS data with IJ$\\rm K_{S}$ photometry of the DENIS survey (Epchtein et al. \\cite{Epchtein97}), with an approach similar to the work of Cioni et al. \\cite{Cioni2001}. The intention of our work is to discuss the light change of red variables on a large and homogeneous sample and to compare the results for LMC red variables with the corresponding objects in the Galactic Disk and Bulge. ",
        "conclusions": "\\subsection{Variability on the AGB} According to Fig.\\,\\ref{CMD} most of the variables in our sample are on the AGB. Therefore, we can use our results to discuss the variability during the AGB phase. Our classification system for the type of variability aims to measure the regularity of the light change. Even not taking into account variations in the amplitude of the light change, we show that most stars have light curves that cannot be fitted by the simple combination of one or two excited periods. Regular variations are found with a wide range in period, while semiregular variability typically occurs mainly on time scales below 150 days (see Fig.\\,\\ref{Perioddistribution}).  In Fig.\\,\\ref{agapgcvs}, we compare the period distribution of the semiregular variables in our sample with the Milky Way SRVs listed in the GCVS. While in both cases the maximum of the distribution is at short periods, the GCVS distribution shows a significantly larger fraction of stars with periods longer than 150 days. These long periods may have been missed by our rather short time window. One would also expect a bias of the GCVS sample towards large amplitude variables as most of the data used there are based on photographic measurements. Furthermore, the period distribution from the GCVS given in Fig.\\,\\ref{Perioddistribution} includes only one (main) period per object, while for the AGAPEROS data we give also secondary periods found for these stars.  \\begin{figure} \\resizebox{\\hsize}{!}{\\includegraphics{lebzfig14.eps}} \\caption{Period distribution of semiregular variables in our sample and in the GCVS.} \\label{agapgcvs} \\end{figure}  Due to the separation of amplitude and regularity in our classification system, we can explore the relation between these two quantities. We find that large amplitude variation occurs almost exclusively among the {\\it regular} variables (see Fig.\\,\\ref{ampldist}). However, there exist {\\it regular} pulsators with small amplitudes.  It is therefore not correct to classify all red variables below a certain amplitude limit as semiregular.  A division into large and small amplitude variables seems to be more meaningful.  Large and small amplitude variables are both found all along the AGB. This is illustrated in Fig.\\,\\ref{KsAmpl} where the $\\rm R_{EROS}$ light amplitude is plotted against the DENIS K band measurement. Towards the tip of the AGB the fraction of regular as well as large amplitude variables increases. Below the RGB-tip, amplitudes become on the average smaller.  The occurrence of regular and semiregular as well as small and large amplitude variables on the AGB indicates that AGB stars have to be seen as a highly inhomogeneous group. One reason for this may be a difference in stellar mass as noted above.  Summarizing, large amplitudes are well correlated with regular pulsations, but we find no correlation between large amplitude and stellar luminosity nor between small amplitude variability and semiregularity of the light change.  This result is in agreement with Wood et al.\\,(\\cite{Wood99}).  \\begin{figure} \\resizebox{\\hsize}{!}{\\includegraphics{lebzfig15.ps}} \\caption{$\\rm R_{EROS}$ amplitude versus $\\rm K_S$. Open boxes denote {\\it regular} variables, filled triangles {\\it semiregular} stars.} \\label{KsAmpl} \\end{figure}  \\subsection{PL-relation} In the literature, the observed PL-relation of long-period variables is considered to be the same in different environments such as the LMC, the Galactic Bulge or globular clusters (see e.~g. Glass et al. \\cite{Glass95}, Feast et al. \\cite{Feast2002}). It is therefore independent of metallicity, contrary to the predictions of pulsation theory (see e.~g. Wood\\,\\&\\,Sebo \\cite{Wood96}).  However, previous studies were restricted to Mira variables mainly due to limitation of sensitivity.  Thanks to the microlensing surveys such as EROS, MACHO or OGLE, we can study {\\em{systematically}} small amplitude variations over a (still rather small) time interval. Wood (\\cite{Wood2000}) found for the SRVs in the LMC different PL-relations for different pulsational modes. However, these separations cannot be reproduced in the Galactic Bulge (Schultheis \\& Glass 2001). In addition, the PL-relation of the solar neighborhood (Bedding \\& Zijlstra \\cite{Bedding98}) looks different. Why is the PL-relation the same for Miras in different galactic environments, but {\\it not} for SRVs?  Fig.\\,\\ref{KlogP} shows the PL-relation for the AGAPEROS sample.  On the one hand, the Mira variables, classically defined as long period and large amplitude stars, concentrate along Wood's sequence C.  {\\it Regular} variables at shorter periods would not have been classified as Miras. On the other hand, the semiregular variables (both according to the classical and to our definition), are spread all over the K-log\\,P-plane.  Making one fit with all {\\it semiregular} stars would not result in a K vs.~log\\,P relation.  In the solar neighborhood, Bedding \\& Zijlstra (\\cite{Bedding98}) note that the SRVs are actually found on two sequences: the first one corresponds to the LMC Mira PL-relation (Wood's sequence C); the second one is located close to a PL-relation derived from Galactic globular cluster LPVs shifted 0.8\\,mag from the Whitelock globular cluster sequence (Whitelock \\cite{Whitelock86}), as shown in Fig.\\,\\ref{KlogP}. The Bedding \\& Zijlstra sequence, defined for SRVs, obviously mixes objects from Wood's sequence B and C, as shown in Fig.\\,\\ref{KlogP}. The increase towards longer periods is consistent with the larger fraction of long period SRVs in the GCVS (Fig.\\,\\ref{agapgcvs}) assuming that the detection of long periodic small amplitude variations is biased towards bright objects. Therefore, three PL-sequences seem to be more appropriate for semiregular variables. Multiperiodic stars are found on all three sequences A, B and C (see Fig.\\,\\ref{KlogP}). Sequence D is almost exclusively occupied by stars with two periods in agreement with the suggestion from Wood (\\cite{Wood2000}) that these long periodic variations are either due to binarity or a pulsation mode resulting from an interaction of pulsation and convection.  However, there are also a few {\\it regular} pulsating variables on this sequence with only one period.  These stars would be definitely worth further investigation.  Schultheis \\& Glass (\\cite{Schultheis2001}) showed that the interpretation of the PL-relation of Bulge SRVs is rather complex due to the depth of the Bulge ($\\sim$ $\\rm \\pm 0.35^{mag}$, see Glass et al. \\cite{Glass95}) and the variable interstellar extinction. There is no clear separation of the four sequences. We also showed that the LMC variables are much more homogeneous in their metallicity than the Bulge AGB stars (Fig.\\,\\ref{logPIJ}).  This would explain part of the scatter in the K vs.~log\\,P plot for the Bulge.  \\subsection{Number densities} The number of semiregular variables in comparison to the regular variables is about a factor of 3.  If we use the selection criterion of Cioni et al.\\,(\\cite{Cioni2001}), i.e.~all stars with $\\rm R_{EROS}$ amplitudes smaller than 0.9 mag are SRVs, we end up with a ratio of almost 37 between SRVs and Miras in our sample. This value is much higher than what was found by Cioni et al.~($\\sim$5), so we assume that our sample is more complete at smaller amplitudes.  In the Galactic Bulge, Alard et al. (\\cite{Alard2000}) found that the proportion of SRVs with respect to Miras is about a factor of 20. Most recently, Derue et al.\\,(\\cite{Derue02}) found a similarly large ratio between semiregulars and miras in the Galactic spiral arms. However, this ratio is of course very sensitive to the classification of SRVs (see above).  For the Galactic disk, Kerschbaum \\& Hron (\\cite{KH92}) found equal number densities for Miras and semiregular variables. However, they note that their sample of semiregular variables is probably not complete due to the difficulties in detecting small amplitude variables.  Do we see in different environments the same ratio of SRVs to Mira variables or does it depend on metallicity?  Vassiliadis \\& Wood (\\cite{Vassiliadis93}) calculated lifetimes of the major evolutionary phases for different initial masses and different metallicities. They found that higher metallicity will increase the lifetime of the early-AGB but decrease the lifetime on the TP-AGB. Miras stars populate the TP-AGB, therefore in environments with higher metallicities, such as the Galactic Bulge the lifetime of the TP-AGB is shorter and thus the number densities should decrease. This might explain the correlation between the ratio of SRVs to Miras and metallicity. However, while a large fraction of our variables on the TP-AGB are {\\it regular} variables\\footnote{In this case {\\it regular} variables and Miras can be assumed to be identical.} also {\\it semiregular} variables are found. Lebzelter \\& Hron (\\cite{LH99}) have shown that for stars in the solar neighborhood stellar evolution goes from SRVs to Miras. The {\\it semiregular} stars found at a similar luminosity as the Miras (see Fig.\\,\\ref{KlogP}) are therefore probably not in the same evolutionary state or they have different masses. Comparison of the number densities with expected lifetime is therefore problematic.  A lower metallicity leads also to a shift of the AGB towards higher temperatures in the HR diagram. The visual light change of these cool variables is dominated by highly temperature sensitive molecules like TiO (e.g.~Reid \\& Goldston \\cite{RG02}). If the stellar temperature is higher, these molecules will play a minor role. Lower metallicity will also make the TiO bands weaker. Therefore one would expect that the visual amplitudes will in general be smaller for lower metallicity. This would favour small amplitude variability in metal poor environments and would explain the smaller fraction of large amplitude objects in the LMC compared to the Bulge. It would also be consistent with the complete lack of Miras in metal poor globular clusters (Frogel \\& Whitelock \\cite{FW98}).  However, one has to be extremely careful concerning possible selection effects, in particular for small amplitude variables. A homogeneous survey of variable stars in different Galactic environments is therefore needed."
    },
    "0208/astro-ph0208082_arXiv.txt": {
        "abstract": "{ We present new orbits for sixteen Ap spectroscopic binaries, four of which might in fact be Am stars, and give their orbital elements. Four of them are SB2 systems: HD~5550, HD~22128, HD~56495 and HD~98088. The twelve other stars are : HD~9996, HD~12288, HD~40711, HD~54908, HD~65339, HD~73709, HD~105680, HD~138426, HD~184471, HD~188854, HD~200405 and HD~216533. Rough estimates of the individual masses of the components of HD~65339 ($53$ Cam) are given, combining our radial velocities with the results of speckle interferometry and with Hipparcos parallaxes. Considering the mass functions of 74 spectroscopic binaries from this work and from the literature, we conclude that the distribution of the mass ratio is the same for cool Ap stars as for normal G dwarfs. Therefore, the only differences between binaries with normal stars and those hosting an Ap star lie in the period distribution: except for the case of HD 200405, all orbital periods are longer than (or equal to) 3 days. A consequence of this peculiar distribution is a deficit of null eccentricities. There is no indication that the secondary has a special nature, like e.g. a white dwarf. ",
        "introduction": "Ap stars are conspicuous not only because of their strong chemical anomalies, but also because of their strong, large-scale magnetic field (at least in the Si and SrCrEu subtypes) and slow rotation. The latter characteristic is associated with a complete lack of Ap stars in binaries with very short orbital periods (i.e. 1.5 days or less), contrarily to normal stars, probably because such systems are synchronized and their components have to rotate fast, which does not seem compatible with the development of chemical peculiarities. One could also think of an observational bias as another possible cause, the line widening erasing mild peculiarities; the fact that some Bp or Ap stars do rotate fast (200~km\\,s$^{-1}$) does not support this explanation, however. But, in addition to the fact that tidal synchronisation will preclude  the existence of Ap stars in short period systems, one may reasonably expect that some special conditions are needed to form an Ap star, and that these conditions might leave their blueprint not only in the magnetic field and slow rotation, but also in the frequency and orbital elements of binaries. One important purpose of this paper is precisely to explore this possibility.  The first, systematic search for binaries among Ap stars has been done by Abt \\& Snowden (\\cite{AS73}), who examined 62 bright northern stars and concluded to a low rate of binaries (20 percent), except for HgMn stars (43 percent). Aikman (\\cite{A76}) increased the sample of HgMn stars from 15 to 80 and confirmed the rate found by Abt \\& Snowden, since he found 49 percent, which is very close to the result of Jaschek \\& Gomez (\\cite{JG70}) for normal B0 to M stars of the main sequence ($47\\pm 5 \\% $).  There has been no systematic review of multiplicity among Ap stars since the work of Gerbaldi et al. (\\cite{GFH85}, hereafter GFH85), apart from some attempts by Budaj (\\cite{B95}, \\cite{B96}, \\cite{B97}) to interpret the role that a binary companion might have in the appearance of chemical peculiarities of both Am and Ap stars. According to GFH85, the rate of binaries tends to be rather small among the He-weak and Si stars. For the coolest Ap stars, as well as for the HgMn stars, this rate behaves in the same way as for normal stars. Moreover, the magnetic Ap stars show a strong deficit of SB2 binaries: Only two SB2's containing a magnetic Ap star had been well studied before the CORAVEL observations: HD~55719 (Bonsack \\cite{B76}) and HD~98088 (Abt et al. \\cite{A68}; Wolff \\cite{W74}). On the other hand, HgMn stars, which generally have no significant magnetic field (see, however, Mathys \\& Hubrig \\cite{MH95}), are often found in SB2 systems, and their companion seems to be always an Am star when its effective temperature is below 10000~K (Ryabchikova \\cite{Ry98}). Am stars are also known to be frequently associated with SB2 systems (Abt \\& Levy \\cite{A85}).  A radial-velocity survey of a small number of cool, well-known magnetic Ap stars has been initiated in 1980 using the CORAVEL scanner (Baranne et al. \\cite{BMP79}), and the sample has been extended in 1985 to all stars brighter than $V=8.6$, visible from the northern hemisphere and having Geneva photometry. The purpose was to increase the relatively poor statistics and obtain a better understanding of the role multiplicity might play in the context of chemically peculiar stars. Preliminary results have been published by North (\\cite{N94}), especially the discovery of a long-period SB2 system which has been studied in more details later (HD 59435, Wade et al. \\cite{WN96}, \\cite{WM99}) and the discovery of an SB1 system with a period as short as 1.6 days (HD 200405).  This paper is the second one in a series dedicated to multiplicity among Ap and Am stars. Since Drs. Nicole Ginestet and Jean-Marie Carquillat in Toulouse had independantly measured a few of our programme stars with the same instrument, we had dedicated the first paper of this series to four common stars (North et al. \\cite{NC98} Paper I), especially the Ap stars HD~8441 and $\\beta$ CrB. Here we present the results for all Ap stars measured to date with CORAVEL, with a few additional data from the ELODIE spectrograph (Baranne et al. \\cite{BQ96}) which are by-products of a survey of magnetic fields (Babel \\& North, in preparation). ",
        "conclusions": "We have determined a dozen of new spectroscopic orbits of Ap stars. Moreover, as shown in the Appendix, we have confirmed the 273-day period of HD~9996 and found a new value for the period of HD~216533 (P~=~1414.73 days) in complete disagreement with the old values (16 days). We have computed the mass of both components of the Ap star $53$~Cam, thanks to our homogeneous radial velocities which could be combined with the published speckle orbit. We have also shown that no significant apsidal motion has occurred in the HD 98088 system for the last 40 years.  The main result of this study is that statistically, the orbital parameters of Ap stars do not differ from those of normal stars, except for an almost complete lack of orbital periods shorter than 3 days. This cut-off is accompanied by a parallel lack of circular and low eccentricity orbits, the latter being due to the former. But in spite of this general rule, there is the interesting exception of HD 200405, an SB1 system with $P_{\\mbox{rot}}=1.6$~days. This system would merit further investigation.  It is important to mention that the anomalous eccentricity distribution found by GFH85 is certainly not an independant fact, but is tightly linked with the lack of orbital periods shorter than 3 days. So short periods always correspond to circular orbits; therefore, removing them will result in an apparent excess of high eccentricities. There is nothing abnormal about the orbital parameters of binary systems hosting an Ap star, except for the lack of short periods.  The distribution of the mass ratios of Ap binaries is found to be compatible with the mass ratios of normal binaries with smaller masses (G-dwarfs). However, the sample of Ap stars with well-determined orbits is not sufficient to explore possible differences between the distributions of orbital parameters of each of the four categories of Ap stars on the one hand, and of the normal stars on the other hand.   \\appendix"
    },
    "0208/astro-ph0208427_arXiv.txt": {
        "abstract": "We investigate the feasibility of determining the pairwise velocity dispersion (PVD) for Lyman Break Galaxies(LBGs), and of using this quantity as a discriminator among theoretical models. We find that different schemes of galaxy formation lead to significant changes of the PVD. We propose a simple phenomenological model for the formation of Lyman break galaxies, determined by the formation interval parameter $\\Delta_z$ and the halo mass threshold $M_h$. With a reasonable choice for these two parameters, our model predicts an occupation number distribution of galaxies in halos which agrees very well with the predictions of semi-analytical models. We also consider a range of galaxy formation models by adjusting the two model parameters. We find that model LBGs can have the same Two Point Correlation Function (TPCF) over the range of observable separations even though the cosmology and/or galaxy formation model are different. Moreover, with similar galaxy formation models, different currently popular cosmologies can result in both the same TPCF and the same PVD. However, with the same cosmology, different galaxy formation models may show quite different PVDs even though the TPCF is the same. Our test with mock samples shows furthermore that one can discriminate among such models already with currently available observational samples (if the measurement error of the redshift is negligible) which have a typical error of $80\\kms$. The error will be reduced by a factor of 2 if the samples are increased four times. We also show that an erroneous assumption about the geometry of the universe and different infall models only slightly change the results. Therefore the PVD will become another promising statistic to test galaxy formation models with redshift samples of LBGs. ",
        "introduction": "The Lyman break technique developed by Steidel and his coworkers has opened a window to uncover high-redshift galaxies at redshift $z\\approx 3$ based on ground-based photometric observations only (Steidel et al. 1996; Steidel et al. 1999). The technique has proved to be very efficient, since most of the candidates identified with photometric colors have been confirmed as high-redshift galaxies in subsequent spectroscopic observations. About 1000 Lyman-break galaxies with redshifts have already been compiled by Steidel's group. The sample of high-redshift galaxies is likely to be enlarged significantly in the next years with more 10-meter telescopes being used in the observations (e.g. Ouchi et al. 2001).  Lyman-break galaxies have been found to be strongly clustered. The correlation length of the galaxies is $3$ to $6\\mpc$ in comoving coordinates, similar to that of local normal galaxies (Steidel et al. 1998; Adelberger et al. 1998; Giavalisco et al. 1998; Connolly, Szalay \\& Brunner 1998; Arnouts et al.1999; Adelberger \\& Steidel 2000; Giavalisco \\& Dickinson 2001, hereafter GD01). The strong clustering is generally expected in cold dark matter (CDM) models if the galaxies are hosted by massive halos with mass $M\\gsim 5\\times 10^{11}\\himsun$ (Mo \\& Fukugita 1996; Steidel et al. 1998; Jing \\& Suto 1998; Giavalisco et al. 1998; Adelberger et al. 1998; Connolly, Szalay \\& Brunner 1998).  Combined with other observations of Lyman Break galaxies, such as the star formation rate, kinematics, metal abundances, it is hoped that the clustering properties of the high-redshift galaxies can set significant constraints on galaxy formation models.  Current theoretical models (e.g. the standard CDM model and the $\\lambda$-dominated CDM model) have been fine-tuned to fit the local observations. The degeneracy found in the model parameters might be broken with the help of observations at high redshift, like those of Lyman-break galaxies.   The clustering of Lyman-break galaxies is found to be consistent with the predictions of most currently interesting models, partly because these models have been tested by other observations and partly because LBGs with different host halo mass can exhibit the same clustering property as long as galaxy formation recipes are tuned correspondingly.  In a recent study by Wechsler et al. (2001, hereafter W01), five scenarios were examined for associating dark matter halos in the standard LCDM model with Lyman-break galaxies, and the clustering length of the model galaxies was compared with the observations of Adelberger \\& Steidel (2000). They conclude that the strong clustering of Lyman-break galaxies can generally be reproduced in different cosmogonic models (including the mixed dark matter model) by adjusting the galaxy formation recipes (e.g. introducing plausible physical processes like starbursts, but in an average way by adjusting various global parameters). Although the host halo mass is rather different in different cosmogonic models, these differences could not be used to discriminate among models, mainly because reliable determinations of the mass of halos hosting Lyman break galaxies are lacking. Early spectroscopic observations indicate that the LBGs are hosted by massive halos (Steidel et al. 1996), but recent observations reveal that the disk rotation within Lyman-break galaxies could be much slower than was thought before (Pettini et al. 2001). Thus between models of different hosting galaxies a choice could not be made yet.  The pairwise velocity dispersion (PVD) of galaxies, which probes the dark matter potential, could provide information about the dark matter distribution which is independent of and complementary to the spatial distribution of galaxies (e.g. the two-point correlation function). The PVD has already been widely applied to the local redshift surveys of galaxies, and has yielded very valuable information about the dynamics of local galaxies which has been widely taken as important input for the cosmological models. The PVD of galaxies determined (Davis \\& Peebles 1983) from the first wide angle redshift survey of galaxies, the Center for Astrophysics (CfA) redshift survey, was one principal reason to argue for biased galaxy formation, i.e. for a difference in the distribution of the galaxies and the dark matter particles.  This survey was later found to be too small for robustly measuring the PVD (Mo et al. 1993). The publicly available Las Campanas Redshift survey of galaxies is about ten times larger than the CfA, and Jing, Mo, \\& B\\\"orner(1998, hereafter JMB98) have measured the PVD for this survey. The accurate PVD measurement of the LCRS survey not only constrains the $\\beta$ parameter of the CDM models to around $0.4$ (where $\\beta=\\Omega_0^{0.6}\\sigma_8$, $\\Omega_0$ is the current density parameter and $\\sigma_8$ the rms linear density contrast within a sphere of $8\\mpc$), but also suggests an anti-bias for the local galaxy distribution on very small scales. The statistical results of JMB98 are confirmed by a recent analysis of the early data release of the Sloan Digital Sky Survey (Zehavi et al. 2002).  The PVD could also be a very important observable quantity for high redshift galaxies, when redshift surveys, like those of the Lyman Break galaxies and the DEEP2 survey become available. The local observations, like the cluster abundance, the PVD of galaxies, and the peculiar velocity field of galaxies, require that the $\\beta$ is around $0.4\\sim 0.5$ almost independently of whether the cosmos is open, flat or dominated by vacuum energy. The local degeneracy may be broken by the PVD observation at high redshift. Because the density fluctuation grows quite differently in different cosmological models, the $\\beta$ value is very different at high redshifts in these models. The $\\beta$ value is much smaller in the SCDM than that in the LCDM model, as is the pairwise velocity dispersion of the dark matter. Since the Lyman-break galaxies are known from their high spatial correlation to be a biased tracer of the dark matter they do not uniformly sample the dark matter distribution, so the pairwise velocity dispersion of LBGs could differ significantly from that of the dark matter.  Thus a measurement of their PVD may be useful to put constraints on the galaxy formation models, but it is unknown if it is feasible to measure this important quantity with observations available now or in the near future.  In this paper, we investigate this important issue within two CDM models: the SCDM and LCDM.  We will attempt to address the following four questions: 1) how does the bias of the LBGs relative to the dark matter show up in the PVD; 2) how does the PVD of Lyman break galaxies depend on the cosmological model; 3) how does the PVD of LBGs depend on the recipes for galaxy formation; 4) how accurately can one measure the PVD of Lyman break galaxies with currently available samples and with samples available in the near future.  Although this work is focused on the Lyman Break galaxies, the approach is readily extended to other high-redshift galaxy surveys, like the DEEP2 survey, where an important goal is the measurement of the PVD of galaxies at redshift about one (Coil et al. 2001). As galaxies at $z=1$ may be more closely connected to the galaxies at $z=0$, the PVD study of the DEEP2 survey can probably yield interesting constraints on theoretical models.  The paper is arranged as follows: We will use a plausible phenomenological model to identify Lyman break galaxies from high-resolution N-body simulations, as described in \\S 2.  The model predicts the occupation number of galaxies within halos.  This agrees very well with the prediction of a physically motivated semi-analytical model of galaxy formation. Because the PVD and the correlation function of galaxies mainly depend on the occupation number of galaxies in halos, we believe that our results represent the prediction of a class of physically motivated galaxy formation models. We also investigate how the PVD depends on formation models of galaxies and the difference in PVD between galaxies and dark matter. We will discuss the possibility of discriminating between cosmological models with the PVD measurement. In \\S 3, we consider a set of mock samples generated according to the observational strategy of Steidel et al. (1998), and assess the accuracy of the PVD measurement with currently available Lyman break galaxy samples and with samples available in the near future. With these mock samples we will also discuss how the measurement depends on the assumption of the world model. Our results will be further discussed in the final section \\S 4. ",
        "conclusions": "We have investigated the feasibility of determining the pairwise velocity dispersion for the Lyman Break Galaxies, and of using this quantity as a discriminator among theoretical models. Our central conclusion is that the PVDs change significantly with different schemes of galaxy formation within the same cosmogony model. On the other hand, with similar galaxy formation models, the same two-point correlation function and pairwise velocity dispersion can be obtained for several currently popular cosmogony models. Thus, the PVD of high-redshift objects can be used to set constraints on the way galaxies form (even through cosmogony keeps unknown).  We have proposed a simple phenomenological model for the formation of Lyman break galaxies which is determined by the formation interval parameter $\\Delta_z$ and the halo mass threshold $M_h$. With a reasonable choice for these two parameters, our model predicts an occupation number distribution of galaxies in halos which agrees very well with the predictions of semi-analytical models (W01). This implies that our model incorporates the essential physical processes involved in the formation of the Lyman Break galaxies. Since in our scheme the properties of LBGs are mainly determined by the occupation number distributions of galaxies, we can allow for uncertainties in the current understandings (e.g. the semi-analytical model) of galaxy formation by adjusting the two model parameters. It is likely that the massive halo model and the recent model proposed by Shu et al. (2001) can predict a PVD a few hundred $\\kms$ smaller than our fiducial models at the same separation $r_p$. Our tests with mock samples show that such models can already be constrained with currently available observed samples (if the measurement error of the redshift is negligible; see discussion below), where the PVD has a typical error of $80\\kms$. This error will be reduced by a factor of 2 if the samples are increased four times.  Therefore the PVD will become another promising statistic to test galaxy formation models with redshift samples of LBGs.  The determination of the PVD at small separation $r_p$ is insensitive to the assumptions about the world model for computing the comoving separations of galaxies. Measuring redshift distortions at small scales therefore cannot be used to measure the cosmological parameters, though the cosmological parameters might be effectively constrained by the distortion measurement on larger scales (Matsubra \\& Suto 1996; Ballinger et al. 1996).  The typical error of the PVD for a 10-beam sample is $80\\kms$ only, and this is really the accuracy one can achieve with currently available redshift samples of LBGs. As Mo et al. (1993) and later other authors (Zurek et al. 1994; Marzke et al. 1995) found, the value of PVD is very sensitive to the presence or absence of rich clusters in a sample, and a redshift sample much larger than the CfA survey which has about 2000 galaxies is needed to achieve an accuracy of $\\sim 100\\kms$ in the determination of the PVD for local galaxies. Indeed the accuracy of measuring the PVD for the Las Campanas Redshift Survey which contains 25,000 galaxies is $75\\kms$ (JMB98), very similar to our expected error for a 10-beam sample of LBGs. The reason is simply that the PVD is better determined at high redshift, because there rich clusters are much rarer, and the effective volume of 10 beams is large enough to include many massive halos formed at that time.   So far we have not taken into account the measurement error in observing the redshift. The measurement error may be treated just like a random motion which can contribute to the measured result of the pairwise peculiar velocity. Thus, in order that the PVD of the LBGs can be measured, the redshift error is required to be much smaller than $400\\kms/\\sqrt{2}/c\\approx 0.001$. Correspondingly, the spectral resolution in the measurement should be better than $7{\\rm \\stackrel{o}{A}}$ at $7000{\\rm \\stackrel{o}{A}}$, if just one line is used. Steidel et al. (1998) measured the redshifts from $\\sim 7$ strong absorption features in the rest-frame far-UV and, when possible, the Ly$\\alpha$ emission line, so the random error may not be a problem for measuring the PVD. It is also noted that there is a systematic difference between redshifts determined from emission lines and from absorption lines (Steidel et al. 1998; Pettini et al. 2001). This systematic difference should be taken into account, and its effect on the PVD measurement may be reduced if redshifts either from absorption lines or from emission lines only are adopted. Furthermore, with the help of near-IR spectroscopy of some of the high redshift galaxies, which measures the true redshift (Steidel et al. 1998), we can model the distribution of the redshift error, and thus it is possible to lessen its effect by adding one term to the right hand side of Equation (7) to correct the random error."
    },
    "0208/cond-mat0208230_arXiv.txt": {
        "abstract": "We study the classical Antonov problem (of retrieving the statistical equilibrium properties of a self-gravitating gas of classical particles obeying Boltzmann statistics in space and confined in a spherical box) for a rotating system. It is shown that a critical angular momentum $\\lambda_c$ (or, in the canonical language, a critical angular velocity $\\omega_c$) exists, such that for $\\lambda<\\lambda_c$ the system's behaviour is qualitatively similar to that of a non-rotating gas, with a high energy disordered phase and a low energy collapsed phase ending with Antonov's limit, below which there is no equilibrium state. For $\\lambda>\\lambda_c$, instead, the low-energy phase is characterized by the formation of two dense clusters (a ``binary star''). Remarkably, no Antonov limit is found for $\\lambda>\\lambda_c$. The thermodynamics of the system (phase diagram, caloric curves, local stability) is analyzed and compared with the recently-obtained picture emerging from a different type of statistics which forbids particle overlapping. ",
        "introduction": "The name ``Antonov problem'' is usually referred to the study of the mean-field equilibrium state of self-gravitating $N$-body systems of classical particles in a finite volume. One considers the system with Hamiltonian \\begin{gather} H_N\\equiv H_N(\\{\\boldsymbol{r}_i\\},\\{\\boldsymbol{p}_i\\})=\\frac{1}{2} \\sum_{i=1}^N p_i^2+\\Phi(\\{\\boldsymbol{r}_i\\})\\label{ham}\\\\ \\Phi(\\{\\boldsymbol{r}_i\\})=-G\\sum_{1\\leq i<j\\leq N} \\frac{1}{|\\boldsymbol{r}_i-\\boldsymbol{r}_j|}\\label{pot} \\end{gather} enclosed in a finite three-dimensional spherical box of radius $R$, which is necessary in order to prevent particles from evaporating. In (\\ref{ham},\\ref{pot}): $\\boldsymbol{r}_i\\in V\\subset\\mathbb{R}^3$ and $\\boldsymbol{p}_i\\in\\mathbb{R}^3$ denote, respectively, the position and momentum of the $i$-th particle, $G$ is the gravitational constant, and $V$ stands for the (volume of the) box containing the particles; we set the particles' masses to one. The main problems are: (a) find the equilibrium configurations, i.e. the particle density profiles $\\rho(\\boldsymbol{r})$ ($\\int_V\\rho(\\boldsymbol{r})d\\boldsymbol{r}=N$) that maximize the microcanonical entropy ($\\alpha=$ constant) \\begin{equation} S_N\\equiv S_N(E)=\\log\\l[\\frac{\\alpha}{N!}\\int \\delta(H_N-E) D\\boldsymbol{r}D\\boldsymbol{p}\\r] \\end{equation} ($D\\boldsymbol{x}=\\prod_{i=1}^N d\\boldsymbol{x}_i$); and (b) analyze the corresponding thermodynamics. Clearly, the existence of such profiles and their actual structure will depend on the value of the total energy $E$. Intuitively, when $E$ is large and positive (that is when the kinetic term dominates) a gaseous-type of structure will be preferred, with particles uniformly spread over $V$, whereas when $E$ is sufficiently negative particles are expected to attract each other in a dense core embedded in a vapor. The situation presents however several subtleties. \\emph{Assuming that particles are point masses obeying Boltzmann statistics in space}, i.e. that they can come arbitrarily close to each other, it is found within the mean-field approximation \\cite{antonov62,lyndenbell68,padmanabhan89,padmanabhan90,chavanis01} that \\begin{enumerate} \\item[(a)] there is no global entropy maximum, therefore all maxima are local and correspond to metastable situations; \\item[(b)] if $\\frac{RE}{GN^2}\\equiv\\epsilon>\\epsilon_c\\simeq -0.335$ (i.e. at sufficiently high energies, for we take $R$ to be fixed) there exist local entropy extrema where the particle density satisfies the condition \\begin{equation}\\label{is} \\rho(\\boldsymbol{r})=A\\exp\\l[\\beta \\int\\frac{\\rho(\\boldsymbol{r'})}{ |\\boldsymbol{r}-\\boldsymbol{r'}|}d\\boldsymbol{r'}\\r] \\qquad\\beta=\\frac{3/2}{\\l[E-\\Phi(\\{\\boldsymbol{r}_i\\})\\r]} \\end{equation} where $A$ is a constant. Notice that $\\beta$ is itself a function of $\\rho$. The physically admissible solutions of (\\ref{is}) are spherically symmetric (in fact, they can be mapped onto the solutions of the Emden equation $y''+2y'/x=e^{-y}$, with $y(0)=y'(0)=0$) and are called microcanonical ``isothermal spheres''; \\item[(c)] in general, local extrema in the space of $\\rho$'s can be maxima (metastable states) or saddle points (unstable). Microcanonical isothermal spheres are local maxima if the ``density contrast'' $\\rho(0)/\\rho(R)\\lesssim 709$, otherwise they are at most saddle points of the entropy surface; \\item[(d)] if $\\epsilon$ lies in the range between $-0.2$ and $\\epsilon_c$ isothermal spheres have negative specific heat\\footnote{As first pointed out by Thirring \\cite{thirring70}, in the microcanonical ensemble self-gravitating systems can have negative specific heat, at odds with the canonical setup, in which it is related to energy fluctuations and positive definite.}: in such states the system heats up and shrinks upon decreasing energy; \\item[(e)] finally, if $\\epsilon<\\epsilon_c$ (i.e. at sufficiently low energies) there are not even local entropy extrema. \\end{enumerate} The standard interpretation \\cite{lyndenbell68} is that, while at high energies an equilibrium configuration is achievable in the form of an isothermal sphere, below $\\epsilon_c$ the system collapses to a dense core and overheats, following the negative-specific-heat trend to its extreme consequences. Such a ``transition'', occurring at the Antonov \\emph{point} $\\epsilon=\\epsilon_c$, is known as ``gravothermal collapse'' or ``catastrophe'' (see also \\cite{lblb,katz} for a more formal treatment of this transition).  In this work we extend the above theory to rotating systems, by including the effects of the angular momentum $\\boldsymbol{L}=\\sum_{i=1}^N\\boldsymbol{r}_i\\times\\boldsymbol{p}_i$ as a second conserved quantity besides $E$, and discuss the thermodynamics and the corresponding equilibrium states (which will now depend on both $E$ and $\\boldsymbol{L}$). Our work is motivated mainly by the crucial role rotation is expected to play in astrophysical systems on different scales such as globular clusters and the cores of elliptic galaxies \\cite{chandrasekhar39,binney&tremaine87}, both of which can be modeled by (\\ref{ham},\\ref{pot}). (Hence our ``particles'' can be interpreted equally well as atoms and as stars.) A gravity-driven collapse can be seen as the prime cause for the equilibrium shaping of self-gravitating systems, both rotating and not. But the nature and the result of this collapse are expected to change significantly in the presence of rotation, and in particular for systems rotating sufficiently fast. Rotating systems were tackled in the past by different models (see e.g. \\cite{hachisu1,hachisu2} for a discussion of the stability of rotating gaseous cylinders, and the more recent works \\cite{gross181,slowlyrot} and references therein), but a complete analysis of the Antonov problem with angular momentum is still lacking. Some progress was achieved recently for another class of systems where particles are assumed to obey a statistics that prohibits overlapping, the so-called Lynden-Bell statistics \\cite{prl,epjb}.  Indeed, we will show that the above picture (a)-(e) holds with minor modifications for slowly rotating systems, that is when $|\\boldsymbol{L}|=L$ is sufficiently small. In particular, defining the reduced angular momentum $\\lambda=L/\\sqrt{RGN^3}$, we identify an Antonov \\emph{line} $\\epsilon_c(\\lambda)$, with $\\epsilon_c(0)=\\epsilon_c\\simeq -0.335$ which plays in rotating systems the same role as the Antonov point plays in non-rotating ones. For $\\lambda$ larger than a critical value $\\lambda_c\\simeq 0.45$, however, the collapse occurs not to a single dense cluster but to a double cluster (a ``binary star''), signaling a breaking of the (axial-)rotational symmetry of (\\ref{ham}). It is remarkable that, strictly speaking, for $\\lambda>\\lambda_c$ there is no Antonov limit: double cluster solutions exist at all energies, albeit being locally unstable (they are saddle points of entropy in the space of $\\rho$'s). The lower the energy, the more dense and point-like the two clusters. We calculate the global phase diagram of the model in the $(\\epsilon,\\lambda)$ plane and analyze the thermodynamics by studying the phase transition that take place and the caloric curves. The equilibrium density profiles obtained in the different regions of the phase diagram show that, depending on the values of $\\epsilon$ and $\\lambda$, the formation of ``rings'', ``single stars'', ``double stars'' and ``disk''-like structures is possible.  From a broad statistical mechanics viewpoint, it is well known that self-gravitating systems, and more generally systems with long-range interaction potentials\\footnote{By which we mean potentials decaying with the interparticle distance $r$ more slowly than $r^{-D-\\delta}$ with $\\delta>0$ in $D$ dimensions when $r\\to\\infty$.}, represent a subtle and delicate issue \\cite{gross174}. For a recent critical discussion of the general statistical mechanics of non-trivial systems see \\cite{leshouches}. The main problem is that the usual thermodynamic limit $(N\\to\\infty,V\\to\\infty,N/V\\text{ finite})$ for systems with potentials such as (\\ref{pot}) does not exist \\cite{gallavotti99}. Their thermodynamics is evidently not extensive. Now extensive systems necessarily have positive specific heat, but this need not be true for non-extensive ones \\cite{thirring70,gross82,gross158,gross124,gross140,lyndenbell99}. Hence the canonical and microcanonical description of non-extensive systems do not necessarily coincide (they do only in ``regular'' phases with positive specific heat, and in the dilute limit $(N\\to\\infty,V\\to\\infty,N/V^{1/3}$ finite$)$ discussed at length in \\cite{devega02a,devega02b}). In the most general and physically interesting cases, self-gravitating systems should be studied in the microcanonical ensemble, a circumstance that obviously introduces several technical difficulties which are amplified by the introduction of rotation.  The choice of Boltzmann statistics, which is made by most authors \\cite{padmanabhan90}, to describe the behaviour of particles contributes to complicate the situation further. The basic assumption is that all particles can eventually occupy the same point/cell in space, i.e. overlapping between particles is admitted. With Boltzmann statistics, \\emph{the system (\\ref{ham},\\ref{pot}) has no ground state}, hence the entropy is not bounded. This is the ultimate origin of Antonov's catastrophe, even (as we shall see) in rotating systems. Clearly, if one wants to describe a system where Newtonian gravity is the dominant macroscopic interaction then configurations with high particle density should be avoided, for at such short distances quantum effects and nuclear interactions become more important than gravity. One way around this problem is to introduce hard cores for particles. Another substantially equivalent way is to use Lynden-Bell statistics \\cite{lyndenbell67} instead of Boltzmann's.  We shall proceed as follows. In Sec. 2, we will expound our microcanonical mean-field theory, deriving the analogous of Eq. (\\ref{is}) for rotating systems. In Sec. 3 we will show results from the numerical solution of the new equation, and discuss in particular the phase diagram in the $(\\epsilon,\\lambda)$ plane and the caloric curves. A comparison between the Lynden-Bell scenario discussed in \\cite{prl,epjb} and the Boltzmann theory is presented in Sec. 4. Finally, we formulate our conclusions with some final remarks . ",
        "conclusions": "In summary, we have studied the equilibrium properties of a self-gravitating and rotating gas assuming that particles obey Boltzmann statistics in space. The physical properties of this system can be understood easily as deriving from the balance of (a) kinetic and gravitational energy at low angular momentum, and (b) gravitational and rotational energy at high angular momenta. In case (a), the effects of rotation are basically negligible except for trivial distortions of the equilibrium shapes. If the total energy is sufficiently high so that the kinetic term dominates the gravitational one, the equilibrium state of the system is axially-symmetric and gas-like. At low energies, instead, gravitation induces the system to collapse into a dense core in a low-density gas. In case (b), the high energy regime is similar but at low energy rotation is strong enough to hinder the collapse into a single cluster and drive the system to organize into a double cluster structure (a ``binary star''). This simple, intuitive picture is confirmed entirely by our mean-field theory, which shows that in case (b) at least one main bifurcation (from axially symmetric to axially asymmetric) in the solutions of the basic Eq. (\\ref{equa}) occurs, and possibly many others can be found upon varying the conserved quantities $\\epsilon$ and $\\lambda$.  Our work leaves several directions for further research. The main question concerns probably the dynamics of these systems, which is an equally old \\cite{henon} and widely-discussed theme in astrophysics, see e.g. \\cite{lbe,what}. Among the many open issues and possible extensions for the equilibrium case (such as the effect of different masses, see \\cite{inagaki,devega3}), we want to put forward the following few. First is the behaviour of the system left of the marginal stability line where double cluster solutions become unstable. (We mentioned in Sec. 3 that the location of this line turns out to depend on the maximum order of harmonics included in the calculation of the gravitational potential, $l_{\\text{max}}$. Also the fact that no Antonov limit exists for $\\lambda>\\lambda_c$ could be an effect of truncation.) At low angular momenta, namely below Antonov limit, no statistical equilibrium state exists and dynamical methods should be used to grasp the physics of the system (see e.g. \\cite{chavanis01} and references therein). But for $\\lambda>\\lambda_c$ it is still possible to obtain (unstable) equilibrium states as saddle point of the entropy surface. This means that certain perturbations applied to such ``singular'' double-cluster structures (highly concentrated) will lead to an increase of entropy. It is not clear to us whether local entropy maxima (metastable states) exist in this region. Definitely, there are no global maxima for $S$ is everywhere unbounded. This issue certainly deserves to be studied further. Another important question that we didn't explore concerns the relevance of mass as a further conserved quantity. If $N$ is fixed, it should be treated on the same level as $\\epsilon$ and $\\lambda$, and this could lead to some qualitative changes in the picture presented here. As a final consideration, we stress the mean-field character of our solution. In spite of several claims being made of mean-field theory being exact for self-gravitating systems, we believe that it would be important to investigate the effects of particle-particle correlations, also at the merely quantitative level.    \\medskip  \\textbf{Acknowledgments.} Useful discussions with O. Fliegans are gratefully acknowledged."
    },
    "0208/hep-th0208156_arXiv.txt": {
        "abstract": "We propose a description of dark energy and acceleration of the universe in  extended supergravities with de Sitter (dS) solutions. Some of them are related to  M-theory with non-compact internal  spaces. Masses of ultra-light scalars in these models are quantized in units of the Hubble constant: $m^2 = n\\, H^2$. If dS solution corresponds to a minimum of the effective potential, the universe eventually becomes dS space. If dS solution corresponds to a maximum or a saddle point, which is the case in all known models based on $N=8$ supergravity, the flat universe  eventually stops accelerating and collapses to a singularity. We show that in these models, as well as in the simplest models of dark energy based on  $N=1$ supergravity, {\\it the  typical time remaining  before the global collapse  is comparable to the present age of the universe}, $t = O(10^{10})$ years. We discuss the possibility of distinguishing between various models and finding our destiny using cosmological observations. ",
        "introduction": "\\subsection{Dark energy and our future}  Recent observations of Type Ia supernovae and the CMB show that the Universe is accelerating, and it is spatially flat ($\\Omega_{\\rm tot}= \\Omega_{M}+\\Omega_D = 1$). Approximately $0.3$ of the total energy density of the universe $\\rho_0 \\sim 10^{-120} M_p^4 \\sim 10^{-29}$ g/cm$^{3}$ consists of ordinary  matter ($\\Omega_{M} \\approx 0.3$), and  $0.7$ of the energy density corresponds to dark energy ($\\Omega_D \\approx 0.7$), see \\cite{supernova,Bond} and references therein.  One can interpret  the dark energy $\\rho_D$ either as the vacuum energy (cosmological constant) $\\Lambda \\sim 0.7\\rho_0$, or as the  slowly changing energy of a scalar field $\\phi$ with a vacuum-like equation of state $p_D = w\\, \\rho_D$,\\, $w  \\approx -1$.  In either case our universe is supposed to be accelerating and rapidly approaching de Sitter (dS) regime. Therefore it would be tempting to conclude that our universe in the future is going to expand exponentially for an indefinitely long time,  $a\\sim e^{Ht}$, even if it is closed.  However, dS regime may be transient, and the future of the universe may be quite different. For example, in most of the models of dark energy it is assumed that the cosmological constant is equal to zero, and the potential energy $V(\\phi)$ of the scalar field  driving the present stage of acceleration, slowly decreases and eventually vanishes as the field rolls to $\\phi =\\infty$, see e.g.  \\cite{Dolgov:gh,Wett,Ratra:1987rm,Armendariz-Picon:2000dh,Albrecht:2001xp}. In this case, after a transient dS-like stage, the speed of expansion of the universe decreases, and the universe reaches  Minkowski regime. In both cases (dS space and Minkowski space) life may survive for an extremely long time until all protons decay and the energy resources of the nearby part of the universe are depleted.  However, there is another possibility, which for a long time did not attract much attention. It is quite possible that $V(\\phi)$ has a minimum at $V(\\phi) <0$, or that it does not have any minimum at all and the field $\\phi$ is free to fall to $V(\\phi) = -\\infty$. In this case {\\it the universe eventually collapses,  even if it is flat} \\cite{MTW,Banks:1995dt,Krauss:1999br,Kaloper:1999tt,Kallosh:2001du,Kallosh:2001gr,Linde:2001ae,Steinhardt:2001st,Felder:2002jk,Heard:2002dr}.  Thus, depending on the choice of the model describing dark energy, the flat  universe with $\\Omega=1$ may eventually become dS space, or Minkowski space, or collapse. (It can never become AdS space with energy density dominated by  a negative cosmological constant \\cite{Kallosh:2001gr,Linde:2001ae,Felder:2002jk}.)  \\subsection{Dark energy in extended supergravity}  The main goal of this paper is to investigate the possibility to describe dS space and dark energy in supergravity. We will mainly concentrate on the extended gauged supergravities, although we will also discuss some models based on $N=1$ supergravity. The extended supergravities are most interesting for two reasons. First of all, they  have much closer relation to M/string theory than the ordinary N=1 supergravity. In particular, the maximal $d=4$ $N=8$ supergravity has the same amount of supersymmetries as  M-theory, $8\\times 4=32$ (there are   32  supersymmetries in $d=11$ M-theory). Also,   in the context of theories with extra  dimensions the standard N=1 supergravity (with 4 supercharges) simply does not exist in the bulk; the smallest possible supersymmetry in $d=5$ has 8 supercharges, $2\\times 4=8$ and therefore $N\\geq 2$ in $d\\geq 5$.   In $N=1$ supergravity it is  easy to  construct a nonvanishing scalar potential (the $F$-term potential) by choosing a proper superpotential $W$. The freedom of choice is almost unlimited, which does not provide us with strict guiding principles that could help us to construct a realistic theory. Meanwhile, in the extended supergravity one does not have this freedom. Nonvanishing potentials appear only after  gauging of some of the global symmetries of the theory. In a certain sense, the total potential  is similar to the $D$-term potential in $N=1$ theory: it disappears for vanishing gauge coupling $g$.  Recently it was found that one can describe dark energy in some $d=4$ extended gauged supergravities     that have  dS solutions \\cite{Kallosh:2001gr,Kallosh:2002wj}.  Some of these supergravities solve the equations of motion of M-theory with non-compact internal spaces \\cite{Hull:1988jw}.   These dS solutions correspond to the extrema of the effective potentials $V(\\phi)$ for some scalar fields $\\phi$. An interesting and very unusual feature of  these scalars in all known theories with $N\\geq 2$ is that their mass squared is quantized in units of the Hubble constant $H_0$ corresponding to dS solutions: ${m^2\\over H_0^2}=n $, where $n$ are some integers of the order 1. This property was first observed in \\cite{Kallosh:2001gr} for a large class of extended supergravities with unstable dS vacua, and confirmed and discussed in detail more recently  in \\cite{Kallosh:2002wj} with respect to a new class of $N=2$ gauged supergravities with stable dS vacua \\cite{Fre:2002pd}. The universality of the relation ${m^2\\over H_0^2}=n $ may be attributed to the fact that  ${m^2\\over H_0^2}$ is an eigenvalue of the  Casimir operator of  dS group,  \\cite{Kallosh:2002wj}.  The meaning of this result can be explained in the following way. Usually  the effective potential near its extremum can be represented as $V(\\phi) = \\Lambda + m^2\\phi^2/2$, where $\\Lambda$ and $m^2$ are two free independent parameters. However, in extended supergravities with $\\Lambda > 0$ one always has $m^2 = n \\Lambda/3$, where $n$ are integers (we are using units $M_p = 1$). Taking into account that in dS space $H_0^2 = \\Lambda/3$, one has, for $|\\phi| \\ll 1$, $ V(\\phi) = \\Lambda(1 + n \\phi^2/6) = 3H^2_0(1 + n \\phi^2/6)$. In particular, in all known versions of $N=8$ supergravity dS vacuum corresponds to an unstable maximum, $m^2 = -6H_0^2$ \\cite{Kallosh:2001gr,Kallosh:2002wj}, {\\it i.e.} at $|\\phi| \\ll 1$ one has \\be \\label{simplepot} V(\\phi) = \\Lambda(1 -\\phi^2) = 3H^2_0(1 -\\phi^2) \\ . \\ee Meanwhile, for the $N=2$ gauged supergravity with stable dS vacuum found in \\cite{Fre:2002pd} one has $m^2 = 6H_0^2$ for one of the scalars,  and for $|\\phi| \\ll 1$ one has \\be V(\\phi) = \\Lambda(1 +\\phi^2) = 3H^2_0(1 +\\phi^2) \\ . \\ee   In \\cite{Kallosh:2001gr,Linde:2001ae} it was explained that the `fast-roll' regime with $ {m^2\\over H^2_0}=O(1)$ is suitable for the late inflation describing the present stage of acceleration of the universe. If one takes  $\\Lambda \\sim \\rho_0 \\sim 10^{-120}$  in units $M_p = 1$ and $H_0 \\sim 10^{-33}$ eV,  corresponding to the present stage of expansion of the universe, then the quantization rule implies that there are ultra-light scalars with the mass of the order $ |m ^2|\\sim H^2_0 \\sim (10^{-33}eV)^2. $ Here one should distinguish between the time-dependent Hubble constant $H^2(t) = (\\rho_M + \\rho_D)/3$ and its value $H_0^2 = \\Lambda/3$ in  dS regime where $\\rho_M=0$ and the dark energy field $\\phi$ stays at the minimum/maximun of its potential. However, at the present time, with $\\Omega_D \\approx 0.7$, one has $H(t) = O(H_0)$.    In the early universe the ultra-light scalar fields may stay away from the extrema of their potentials; typically they `sit and wait' and start moving only when the Hubble constant $H(t)$ determined by cold dark matter, decreases and becomes comparable to $|m|$. We will see  that this could result in noticeable changes of the effective cosmological constant during the last 10 billion years. The existence of this effect can be verified by  observational studies of the acceleration of the universe.  Extended supergravity may lead not only to  potentials with dS extrema, but also to  exponential potentials. We will show that a particularly interesting potential  $V \\sim e^{\\sqrt 2\\phi}$ can be derived in several models based on extended supergravity.  Despite the current lore \\cite{Albrecht:2001xp}, the theory with this potential can describe dark energy and the present  stage of acceleration of the universe, and it does not suffer from the problems with the existence of the event horizon discussed in  \\cite{Hellerman:2001yi}.   An important feature of extended gauged supergravities is the fact that {\\it quantum corrections to the cosmological constant as well as to the ultra-light  masses} are related to the  value  of the  cosmological constant $\\Lambda$ defining the scale of SUSY breaking. For $\\Lambda \\sim 10^{-120} M_p^4$  these quantum corrections are very small. Thus extended supergravities provide an example of a model where ultra-light scalars $|m^2| \\sim  \\Lambda$ naturally appear  and are protected against quantum corrections.   \\subsection{Supergravity and the fate of the universe}     The relation $|m^2| \\sim H^2$  appears not only in extended supergravity. It is  rather common in $N=1$ supergravity for the moduli fields that have vanishing mass at $M_p\\rightarrow \\infty$ \\cite{Dine:1983ys}. In contrast to the  extended supergravities, in the $N=1$ case it is possible to avoid this relation: one can take a non-minimal K\\\"{a}hler potential, fine-tune the superpotential, and/or introduce non-trivial D-terms, what will modify  the mass/Hubble  ratio in a significant way.  Indeed, in the supersymmetry breaking hidden sector one should avoid the relation $|m^2| \\sim H_0^2$, in order to avoid huge cosmological constant $ \\Lambda > (10^3\\, {\\rm GeV})^4$. However, there is no need to make this fine-tuning in the dark energy hidden sector\\footnote{A hidden sector for quintessence models in $N=1$ supergravity, different from the supersymmetry breaking hidden sector, was proposed in \\cite{Binetruy:1998rz,Brax:2001ah}.}.  In particular, the relation $|m^2| \\sim H^2$ occurs in the so-called supergravity  quintessence model  with dS minimum \\cite{Brax:2001ah}. Although the potential of a truncated model in this case is simple,  the supergravity model \\cite{Brax:2001ah} is rather complicated, involving many fields with non-minimal K\\\"{a}hler potentials and a set of additional assumptions. In order to study a much simpler toy model for dark energy in $N=1$ supergravity we will consider a Pol\\'{o}nyi-type model \\cite{Polonyi:1977pj} of the dark energy hidden  sector.  It has a minimal K\\\"{a}hler potential and the simplest superpotential $W(z) = \\mu^2 (z+\\beta)$. The parameter $\\mu$ should be taken extremely small,  $\\mu^4 \\sim \\rho_0 \\sim 10^{-120} M_p^4$. In this case there is no need to make the standard fine-tuning  $\\beta = 2-\\sqrt 3$ in order to avoid the large cosmological constant. For $|\\beta| < 2-\\sqrt 3$ the potential has dS minimum with $\\Lambda \\sim +10^{-120} M_p^4$. For larger values of $\\beta = O(1)$ the potential has a minimum with $\\Lambda \\sim -10^{-120} M_p^4$. As we will see, in both cases this model, just like the models based on $N=8$ and $N=2$ supergravity, can describe the present state of acceleration of the universe. However, the  future of the universe does depend on the choice of the model.  Another interesting model  is the axion quintessence  \\cite{Frieman:1995pm,Choi:1999xn}. In the M-theory motivated version of this model proposed in \\cite{Choi:1999xn} one has $V(\\phi) \\sim \\Lambda \\cos(\\phi/f)$; for $f = O(M_p)$ one finds $m^2 = V''(0) = -O(H^2_0)$. As we will show, this version of the axion quintessence model can successfully describe the stage of acceleration of the universe, but, just like the $N=8$ models, it leads to a global collapse of the universe in the future.  In order to obtain a fully realistic model of dark energy, one would need to construct theories involving the observable sector, the hidden sector responsible for supersymmetry breaking, and the hidden dark energy sector. This is a complicated and as yet unsolved problem. The supersymmetry breaking scale in the  dark energy hidden  sector is of the order $ M_{susy}^{\\rm D }\\sim 10^{-12}\\, \\rm GeV \\ $. Meanwhile, in the supersymmetry breaking hidden sector the corresponding scale  is $M_{susy}\\sim 10^{3} \\, \\rm GeV$ or even much greater. This means that  the difference between these two types of  supersymmetry  breaking  is more than $15$ orders of magnitude. Even though the fields from each of these sectors may interact with each other only gravitationally, this interaction may be strong enough to alter the important relation $m = O(H_0)$ for the ultra-light scalars. This problem and various ways to address it were discussed in \\cite{Binetruy:1998rz,Brax:2001ah,Kolda:1998wq,Choi:1999xn}.  In this respect, extended supergravity may be particularly interesting as  the dark energy hidden sector if the mysterious mass quantization rule $m^2 =  n\\,H_0^2$ has some fundamental meaning and remains stable with respect to the interaction of the ultra-light scalars with the fields from the observable sector. One may even argue that the reason for using extended $N\\geq 2$ supergravities is due to the nature of gravitational and vector fields that may live in five dimensions or higher, where $N=2$ is the smallest supersymmetry available. However,  realistic models of supersymmetry breaking in the context of supergravity, branes and extra dimensions are  yet to be developed.   For the time being, one may consider the simple models of dark energy based on supergravity as toy models with some interesting and very unusual  features that could be studied by cosmological observations.  In particular, if dS vacuum corresponds to a minimum of the effective potential, as in the $N=2$ model of Ref. \\cite{Fre:2002pd} and in the Pol\\'{o}nyi model with $|\\beta| < 2-\\sqrt 3$, the universe asymptotically approaches a stable dS regime.  On the other hand, if the effective potential is negative at the minimum, $V(\\phi_0) <0$, as in the Pol\\'{o}nyi model with $|\\beta| > 2-\\sqrt 3$ and in the axion quintessence model with  $V(\\phi) \\sim \\Lambda \\cos(\\phi/f)$, or if it is unbounded from below, as in all known versions of $N=8$ supergravity admitting dS solutions, the flat $\\Omega=1$ universe eventually  collapses.  The typical time remaining before the collapse in these models is $O(m^{-1})\\sim H_0^{-1} $. Since the total age of the universe now is also given by $O(H_0^{-1})$, in this class of models {\\it  the time remaining before the global collapse  is of the same order as the present age of the universe,} $t_{{}_{\\rm Big Crunch}} \\sim 10^{10}$ years.     Thus, in this paper we will study the supergravity  models  which are able to describe nicely the present and the past, and may predict  an  ultimate collapse  that may occur within the next 10-20 billion years. We will show that some of the models predicting gravitational collapse within the next 5 billion years can be ruled out by observational data. The investigation of the stability of the universe on a greater time scale will require a much more detailed observational study of the present stage of acceleration of the universe. ",
        "conclusions": "In this paper we studied various possibilities to describe dark energy in supergravities and the future of the universe in such models. Most of the previous works on this subject  concentrated on the attempts to obtain  inverse power law tracker potentials  in $N=1$ supergravity \\cite{Binetruy:1998rz,Brax:2001ah}. These models required introduction of rather complicated superpotentials and K\\\"{a}hler potentials of several superfields and many assumptions.  In this paper we have chosen a different strategy. First of all, we considered several toy models based on extended supergravity which may have closer relation the M-theory. In addition we considered  the simplest dark energy models in $N=1$ supergravity. All of these models share a very interesting feature: The absolute value of the effective mass squared of the scalar field responsible for the dark energy of the universe  is of the same order as the effective potential of this field, which implies that $|m^2| = O(H^2)$. Whereas in the phenomenological models of quintessence this property usually is {\\it required} for their consistency, in supergravity this feature is rather common and sometimes it is even unavoidable. In particular, in all known versions of extended supergravity with de Sitter solutions with the Hubble constant H, scalar field masses are always quantized: $m^2 = n\\, H^2_0$, where $n$ are some integers which can be either positive or negative \\cite{Kallosh:2001gr,Fre:2002pd,Kallosh:2002wj}. For example, in all versions of $N=8$ supergravity of this type one always has a tachyon field with mass $m^2 = V''_0 = -6 H^2_0 = -2V_0$, where $V_0$ is the value of the effective potential in its extremum corresponding to de Sitter solution, and $V''_0$ is the curvature of the potential at that point \\cite{Kallosh:2001gr,Kallosh:2002wj}. This makes all models with $|m^2| = O(V)$ interesting candidates for the role of the dark energy.  In all models with de Sitter vacuum state with $m^2 \\sim V_0 > 0$ the field $\\phi$ slowly rolls to the minimum of the effective potential and the universe eventually reaches de Sitter state with $H^2_0 = V_0/3$.  On the other hand,  all $N=8 ,\\,  N=4 ,\\,  N=2$ models with $m^2 = V''_0 = -6 H^2_0 = -2V_0$, as well as  $N=1$ Pol\\'{o}nyi models with $\\beta >2-\\sqrt 3$,  describing the present state with $\\Omega_D \\approx 0.7$, lead to the following generic prediction concerning the future of the universe: Our universe is going to collapse within the time comparable to its present age $t \\sim 14$ billion years. Similar results are valid for the axion model of  quintessence with the potential $V =\\Lambda [\\cos \\left(\\phi/f\\right)+C]$ for $C<1$.   The possibility that a flat universe may collapse in the future was known for quite a while \\cite{MTW} -\\cite{Heard:2002dr}. However, this possibility seemed to be rather extravagant, especially in view of the observations suggesting that the universe is accelerating. And even though we knew that the universe may collapse  in a distant future, there was no reason to expect that this may happen relatively soon.  Now the situation becomes quite different. Among all models of dark energy based on extended supergravity only the $N=2$ model of Ref. \\cite{Fre:2002pd} leads to a stable de Sitter space in the future; all other models lead to a collapse. Among the simplest $N=1$ Pol\\'{o}nyi models only the models with $\\beta < 0.268$ lead to a stable de Sitter space. Similarly, among all axion models with the potential $V =\\Lambda [\\cos \\left(\\phi/f\\right)+C]$ only the models with $C \\geq 1$ lead to a stable dS space. All other models predict that our universe should collapse within the next $O(10)$ billion years.  Of course, all of the models considered in our paper are just toy models. We assumed that the dark energy hidden sector can be successfully incorporated into the theory of elementary particles and that the cosmological constant problem in the observable sector can somehow be solved. But this is a general issue with all models of dark energy. On the other hand,   the new class of models may provide an unusual solution to the coincidence problem.  Indeed, the total lifetime of the universe in $N=8$ theories with de Sitter solutions \\cite{Hull:1988jw,Kallosh:2001gr,Kallosh:2002wj}  is $O(H^{-1}) \\sim O(\\Lambda^{-1/2})$. This lifetime can be few times greater, but only if the initial value $\\phi_0$ of the field $\\phi$ is exponentially close to $\\phi=0$. Thus, large values of $\\Lambda$ are forbidden since they do not allow the long-living universes.  If one considers all combinations of $\\phi_0$ and $\\Lambda$ compatible with the total lifetime of the universe $t_{\\rm tot} > 14$ billion  years and  assume that all such combinations are equally probable,  one finds that the  value of $\\Lambda$ should very close to $\\rho_0 \\sim 10^{-120}$,  and the most probable time before the future collapse of the universe is $O(10)$ billion  years.   This may provide a solution to the fine-tuning/coincidence problem for $\\Lambda$ and $\\Omega_D$  simultaneously predicting  the typical time-scale of the global collapse of the universe in the models based on $N=8$ supergravity  \\cite{KallLin}. Similar conclusion is valid for the simplest $N=1$ Pol\\'{o}nyi models with $\\beta > 0.268$.   Interestingly, all models that lead to the ``doomsday prediction'' have some  features that may allow us either to rule them out or to estimate the time remaining until the global collapse. For example, the $N=8$ model with $\\phi_0 < 0.3$ and $\\Omega_D = 0.7$ would lead to a collapse within the next 10 billion years. Meanwhile, the model with $\\phi_0 = 0.35$ would lead to a collapse within the next 7 billion years, see Fig. \\ref{ScalefactorColl}. However, this model leads to a maximal value of $\\Omega_D \\approx 0.65$, which is possible but not particularly favoured by the recent observations \\cite{Bond}. The models with $\\phi_0 > 0.4$  would place the doomsday  even much closer to the present moment, but they lead to $\\Omega_D < 0.56$, which is incompatible with the present cosmological measurements of $\\Omega_D$.   An additional information can be obtained by the measurements of $w$. In all models predicting the doomsday in the near future, the value of $w$ tends to grow significantly at small $z$, i.e. at the present epoch, see Figs. \\ref{wColl}, \\ref{Polw}. It is difficult to determine the time dependence of $w$ using CMB experiments alone \\cite{Bond}, but one can combine them with the supernova observations,  counts of galaxies and of clusters of galaxies, and with investigation of weak gravitational lensing \\cite{WellAl,Copeland,Weller:2001gf}. Even in this case it is difficult to find  the equation of state of dark energy $w(z)$. However, if one takes the cosmological models based on supergravity seriously and realizes that our future is at stake, one gets an additional strong incentive to develop observational and theoretical cosmology. It was never easy to look into the future, but it is possible to do so and we should not miss our chance.    \\   It is a pleasure to thank A.~Albrecht, T.~Banks, T.~Dent, M.~Dine, G.~Felder, J.~ Frieman, N. Kaloper, A. Klypin, L. Kofman, D. Lyth, S.~Perlmutter, A.~ Starobinsky, L. Susskind, P. Townsend, A. Vilenkin and C.~Wetterich for useful discussions. This work was supported by NSF grant PHY-9870115. The work by A.L. was also supported by the Templeton Foundation grant No. 938-COS273.  The work of S.P. was also supported by Stanford Graduate Fellowship foundation. M.S. work was also supported by DOE grant number DE-AC03-76SF00515."
    },
    "0208/nucl-th0208055_arXiv.txt": {
        "abstract": "Spurred by the recent complete determination of the weak currents in two-nucleon systems up to ${\\cal O}(Q^3)$ in heavy-baryon chiral perturbation theory, we carry out a parameter-free calculation of the threshold $S$-factors for the solar $pp$ (proton-fusion) and $hep$ processes in an effective field theory that {\\it combines} the merits of the standard nuclear physics method and systematic chiral expansion. The power of the EFT adopted here is that one can correlate in a unified formalism the weak-current matrix elements of two-, three- and four-nucleon systems. Using the tritium $\\beta$-decay rate as an input to fix the only unknown parameter in the theory, we can evaluate the threshold $S$ factors with drastically improved precision; the results are $S_{pp}(0) = 3.94\\!\\times\\!(1 \\pm 0.004) \\!\\times\\!10^{-25}\\ \\mbox{MeV-b}$ and $S_{hep}(0) = (8.6 \\pm 1.3)\\!\\times\\! 10^{-20} \\ \\mbox{keV-b}$. The dependence of the calculated $S$-factors on the momentum cutoff parameter $\\Lambda$ has been examined for a physically reasonable range of $\\Lambda$. This dependence is found to be extremely small for the $pp$ process, and to be within acceptable levels for the $hep$ process, substantiating the consistency of our calculational scheme. ",
        "introduction": "The standard approach to nuclear physics~\\cite{snpa} anchored on wavefunctions obtained from the Schr\\\"odinger (or Lippman-Schwinger) equation with ``realistic\" phenomenological potentials has scored an impressive quantitative success in describing systems with two or more nucleons, achieving in some cases accuracy that defies the existing experimental precision. We refer to this approach as SNPA (standard nuclear physics approach). The advent of quantum chromodynamics (QCD) as $the$ theory of strong interactions raises a logical question: What is the status of SNPA in the context of the fundamental theory, QCD ? Put more bluntly, is SNPA (despite its great success) just a model-dependent approach unrelated to the fundamental theory ? In our view this is one of the most important issues in nuclear physics today. In this paper, we propose that SNPA can be properly identified as a $legitimate$ component in the general edifice of QCD. The next important question is: If SNPA is indeed a bona-fide element of QCD, how can we establish an expansion scheme which includes SNPA as an approximation and which allows a systematic calculation of correction terms with error estimation ? This systematic correction with an error estimation is not feasible with SNPA alone.  Broadly speaking, formulating nuclear physics calculations starting from effective field theories (EFTs) based on QCD calls for ``double-step decimation\"~\\cite{doubledecimation}. To start with, the ``bare\" EFT Lagrangian for strong interaction physics in the nonperturbative sector is defined at the chiral scale $\\Lambda_\\chi\\sim 4\\pi f_\\pi\\sim 1$ GeV, with the parameters in the Lagrangian determined by suitably matching them to QCD at that scale~\\cite{harada-yamawaki}. This Lagrangian cannot be directly used for studying complex nuclei. In order to render it applicable to low-energy nuclear physics, it is desirable, if not indispensable, that a ``decimation\" be made from the chiral scale $\\Lambda_\\chi$ down to what might be called the ``Fermi-sea scale\" $\\Lambda_{FS}\\sim k_F$, where $k_F$ the nuclear Fermi momentum. An effective Lagrangian resulting from the decimation to this Fermi-sea scale is expected to contain parameters that reflect ``intrinsic\" density-dependence (as suggested in, {\\it e.g.},Ref.~\\cite{BR}) {\\it in addition} to the usual dense loop effects. The next stage of decimation consists of integrating out the effective degrees of freedom and modes from the scale $\\Lambda_{FS}$ to $\\Lambda_0\\sim 0$ MeV. This stage corresponds to doing shell-model type calculations in finite nuclei~\\cite{kuo} and to formulating Fermi-liquid theory in nuclear matter~\\cite{schwenk}. The physics of heavy nuclei or nuclear matter will involve both decimations but, for light nuclei, one can bypass the second decimation and work directly with the chiral Lagrangian.   The aim of this article is to describe a formalism that combines the high accuracy of SNPA and the power of EFT to make totally parameter-free {\\it predictions} for electroweak transitions in light nuclei. To be concrete, we shall consider the following two solar nuclear fusion processes: \\be pp:&&\\ \\ \\ p+p\\rightarrow d + e^+ +\\nu_e\\,, \\label{pp} \\\\ hep:&&\\ \\ \\ p+\\He3 \\rightarrow \\He4 + e^+ + \\nu_e\\,. \\label{hep} \\ee We stress that in our EFT approach these processes involving different numbers of nucleons can be treated on the same footing. A concise account of the present study was previously given in \\cite{PMetal2001} for the $pp$ process, and in \\cite{PMetal2} for the $hep$ process.  The reactions (\\ref{pp}) and (\\ref{hep}) figure importantly in astrophysics and particle physics; they have much bearing upon issues of great current interest such as, for example, the solar neutrino problem and non-standard physics in the neutrino sector. Since the thermal energy of the interior of the Sun is of the order of keV, and since no experimental data is available for such low-energy regimes, one must rely on theory for determining the astrophysical $S$-factors of the solar nuclear processes. Here we concentrate on the threshold $S$-factor, $S(0)$, for the reactions (\\ref{pp}) and (\\ref{hep}). The necessity of a very accurate estimate of the threshold $S$-factor for the $pp$ process, $S_{pp}(0)$, comes from the fact that $pp$ fusion essentially governs the solar burning rate and the vast majority of the solar neutrinos come from this reaction. Meanwhile, the $hep$ process is important in a different context. The $hep$ reaction can produce the highest-energy solar neutrinos with their spectrum extending beyond the maximum energy of the ${}^8\\mbox{B}$ neutrinos. Therefore, even though the flux of the $hep$ neutrinos is small, there can be, at some level, a significant distortion of the higher end of the ${}^8\\mbox{B}$ neutrino spectrum due to the $hep$ neutrinos. This change can influence the interpretation of the results of a recent Super-Kamiokande experiment that have generated many controversies related to neutrino oscillations~\\cite{controversy,monderen}. To address these issues quantitatively, a reliable estimate of $S_{hep}(0)$ is indispensable.  The primary amplitudes for both the $pp$ and $hep$ processes are of the Gamow-Teller (GT) type ($\\Delta J=1$, no parity change). Since the single-particle GT operator is well known at low energy, a major theoretical task is the accurate estimation of the meson-exchange current (MEC) contributions.  The nature of the specific challenge involved here can be elucidated in terms of the {\\it chiral filter} picture. If the MEC in a given transition receives an unsuppressed contribution from a one-soft-pion exchange diagram, then we can take advantage of the fact that the soft-pion-exchange MEC is uniquely dictated by chiral symmetry~\\footnote{A more modern and complete discussion on this observation has recently been given by Ananyan, Serot and Walecka~\\cite{ananyan}.} and that there is a mechanism (called the chiral filter mechanism) that suppresses higher chiral-order terms~\\cite{KDR,BR2001}. We refer to a transition amplitude to which the chiral filter mechanism is applicable (not applicable) as a chiral-protected (chiral-unprotected) case. It is known that the space component of the vector current and the time component of the axial current are chiral-protected, whereas the time component of the vector current and the space component of the axial current are chiral-unprotected (see Appendix A). This implies among other things that the isovector M1 and axial-charge transitions are chiral-protected~\\cite{M1,axialch}, but that the GT transition is chiral-unprotected. This feature renders the estimation of the GT amplitude a more subtle problem; the physics behind it is that MEC here receives significant short-ranged contributions the strength of which cannot be determined by chiral symmetry alone.  The difficulty becomes particularly pronounced for the $hep$ process for the following reasons.  First, the single-particle GT matrix element for the $hep$ process is strongly suppressed due to the symmetries of the initial and final state wave functions. Secondly, as pointed out in Refs.~\\cite{CRSW91} (referred to as ``CRSW91\") and \\cite{SWPC92} (referred to as ``SWPC92\"), the main two-body corrections to the ``leading\" one-body GT term tend to come with the opposite sign causing a large cancellation. A recent detailed SNPA calculation by Marcucci {\\etal}~\\cite{MSVKRB}, hereafter referred to as MSVKRB, has re-confirmed the substantial cancellation between the one-body and two-body terms for the $hep$ GT transition. The two-body terms therefore need to be calculated with great precision, which is a highly non-trivial task. Indeed, an accurate evaluation of the $hep$ rate has been a long-standing challenge in nuclear physics~\\cite{challenge}. The degree of this difficulty may be appreciated by noting that theoretical estimates of the $hep$ $S$-factor have varied by orders of magnitude in the literature.   As mentioned, in obtaining accurate estimates of the GT transition amplitudes, it is imperative to have good theoretical control of short-distance physics. We expect that a ``first-principle\" approach based on effective field theory (EFT) will provide a valuable insight into this issue.  We therefore adopt here the approach developed in Refs.~\\cite{BR2001,PKMR}, which purports to combine the highly sophisticated SNPA with an EFT based on chiral dynamics of QCD. Our starting point is the observation that, to high accuracy, the leading-order single-particle operators in SNPA and EFT are identical, and that their matrix elements can be reliably estimated with the use of realistic SNPA wave functions for the initial and final nuclear states. Next, we note that in EFT the operators representing two-body corrections\\footnote{The argument made here should apply generally to $n$-body currents ($n\\geq 2$) but since the 2-body terms are dominant, we shall continue to restrict our discussion to the latter.} to the leading-order one-body term can be controlled by systematic chiral expansion in heavy-baryon chiral perturbation theory (HB$\\chi$PT)~\\cite{PKMR}. Then, since the ratio of a two-body matrix element to the leading-order one-body matrix element can be evaluated with sufficient accuracy with the use of the realistic SNPA wave functions\\footnote{This statement holds only for the finite-range part of two-body operators, with the zero-range part requiring a regularization to be specified below.}, we are in a position to obtain a reliable estimate of the total (one-body + two-body) contribution. This approach takes full advantage of the extreme high accuracy of the wave functions achieved in SNPA while securing a good control of the transition operators via systematic chiral expansion. For convenience, we will refer to this method, which exploits the powers of {\\em both} SNPA and EFT, as ``MEEFT'' (short for {\\em more effective} EFT). MEEFT -- which is close in spirit to Weinberg's original scheme~\\cite{weinberg} based on the chiral expansion of ``irreducible terms\" -- has been found to have an amazing predictive power for the $n + p\\rightarrow d+\\gamma$ process~\\cite{M1,npp} and several other processes~\\cite{beane}. An alternative approach, which however is in line with our reasoning, has been discussed by Ananyan, Serot and Walecka~\\cite{ananyan}.  An early HB$\\chi$PT study of the $pp$ process was made in Ref.~\\cite{pp} (hereafter referred to as PKMR98) by four of the authors. The calculation in PKMR98 was carried out up to next-to-next-to-next-to-leading order (\\nlo3) in chiral counting (see below). At \\nlo3, two-body meson-exchange currents (MEC) begin to contribute, and there appears one unknown parameter in the chiral Lagrangian contributing to the MEC. This unknown constant, called $\\dR$ in Ref.~\\cite{pp}, represents the strength of a four-nucleon-axial-current contact interaction. In Ref.~\\cite{pp}, since no method was known to fix the value of $\\dR$, the $\\dR$-term was simply ignored by invoking a qualitative argument that the short-range repulsive core would strongly suppress its contribution. Due to uncertainties associated with this argument, Ref.~\\cite{pp} was unable to corroborate or exclude the result of the latest SNPA calculation \\cite{TBDexp}, $\\delta_{\\rm 2B}=0.5 \\sim 0.8$ \\%, where $\\delta_{\\rm 2B}$ is the ratio of the contribution of the two-body MEC to that of the one-body current (see below).   The situation can be greatly improved by using MEEFT. As first discussed in Refs.\\cite{PMetal2001,PMetal2} and as will be expounded here, the crucial point is that exactly the same combination of counter terms that defines the constant $\\dR$ enters into the Gamow-Teller (GT) matrix elements that feature in $pp$ fusion, tritium $\\beta$-decay, the $hep$ process, $\\mu$-capture on a deuteron, and $\\nu$--$d$ scattering and that the short-range interaction involving the constant $\\dR$ is expected to be ``universal,\" that is, $A$-independent. Therefore, assuming that three- and four-body currents can be ignored (which we will justify a posteriori), if the value of $\\dR$ can be fixed using one of the above processes, then we can make a totally parameter-free prediction for the GT matrix elements of the other processes.  Indeed, the existence of accurate experimental data for the tritium $\\beta$-decay rate, $\\Gamma_\\beta^t$, and the availability of extremely well tested realistic wave functions for the $A$=3 nuclear systems allow us to carry out this program. In the present work we determine the value of $\\dR$ from $\\Gamma_\\beta^t$ and perform parameter-free EFT-based calculations of $S_{pp}(0)$ and $S_{hep}(0)$.   As described below, our scheme has a cutoff parameter $\\Lambda$, which defines the energy/momentum cutoff scale of EFT below which reside the chosen explicit degrees of freedom\\footnote{The cutoff specifies not just the relevant degrees of freedom but also their momentum/energy content. This should be understood in what follows although we do not always mention it.}. Obviously, in order for our result to be physically acceptable, its cutoff dependence must be under control. In our scheme, for a given value of $\\Lambda$ in a physically reasonable range (to be discussed later), $\\dR$ is determined to reproduce $\\Gamma_\\beta^t$; thus $\\dR$ is a function of $\\Lambda$. According to the premise of EFT, even if $\\dR$ itself is $\\Lambda$-dependent, physical observables (in our case the $S$-factors) should be independent of $\\Lambda$ as required by renormalization-group invariance. We shall show that our results meet this requirement to a satisfactory degree. The robustness of our calculational results against changes in $\\Lambda$ allows us to make predictions on $S_{pp}(0)$ and $S_{hep}(0)$ with much higher precision than hitherto achieved. Thus we predict: $S_{pp}(0) = 3.94\\!\\times\\!(1 \\pm 0.004) \\!\\times\\!10^{-25}\\ \\mbox{MeV-b}$ and $S_{hep}(0) = (8.6 \\pm 1.3)\\!\\times\\! 10^{-20} \\ \\mbox{keV-b}$.  The remainder of this article is organized as follows. In Section II we briefly explain our formalism; in particular, we describe the relevant transition operators derived in HB$\\chi$PT. Section III presents the calculation of $S_{pp}(0)$, while Section IV is concerned with the estimation of $S_{hep}(0)$. Section V is devoted to discussion and conclusions. We have made the explanation of the formalism in the text as brief and focused as possible, relegating most technical details to the Appendices. ",
        "conclusions": "It is worth emphasizing that the above MEEFT prediction for $\\delta_{\\rm 2B}$ for the $pp$ process is in line with the latest SNPA results obtained in Ref.~\\cite{TBDexp} (and mentioned earlier). There too, the short range behavior of the axial MEC was constrained by reproducing $\\Gamma^t_\\beta$. The inherent model dependence of such a procedure within the SNPA context was shown to be very weak simply because at small inter-particle separations, where MEC contributions are largest, the pair wave functions in different nuclei are similar in shape and differ only by a scale factor~\\cite{forest96}. As a consequence, the ratios of GT and $p$$p$-capture matrix elements of different two-body current terms are nearly the same, and therefore a knowledge of their sum in the GT matrix element is sufficient to predict their sum in the $p$$p$-capture matrix element~\\cite{TBDexp}.   In order to better understand how the present scheme works, it is helpful to compare the $hep$ reaction with the radiative $np$-capture. The polarization observables in $\\vec{n}+\\vec{p}\\rightarrow d+\\gamma$ are known to be sensitive to the isoscalar M1 matrix element, $M1S$, and this amplitude has been extensively studied in EFT \\cite{npp,crs99}. The similar features of the $hep$ GT amplitude and $M1S$ are: (i) the leading one-body contribution is suppressed by the symmetries of the wave functions; (ii) there is no soft-pion exchange contribution; (iii) nonetheless, short-range physics can be reliably subsumed into a single contact term. In the $\\vec{n}\\vec{p}$ case the strength of this term can be determined from the deuteron magnetic moment (for a given value of $\\Lambda$). The calculation in Ref.~\\cite{npp} demonstrates that the $\\Lambda$-dependence in the contact term and that of the remaining terms compensate each other so that the total $M1S$ is stable against changes in $\\Lambda$. This suggests that, if we go to higher orders, the coefficient of the contact term in question will be modified, with part of its strength shifted to higher order terms; however, the total physical amplitude will remain essentially unchanged. These features are quite similar to what we have found here for the $hep$ GT amplitude.  Evaluating the matrix element of the leading-order one-body operator in EFT with the use of realistic nuclear wave functions is analogous to fixing parameters in an EFT Lagrangian (at a given order) using empirical inputs~\\cite{PKMR98}; the realistic wave functions in SNPA can be regarded as a theoretical input that fits certain sets of observables. In the present MEEFT scheme, we take the view that the same realistic wave functions also provide a framework for reliably calculating (finite-range) many-body corrections to the leading-order one-body matrix element. The short-ranged part inherited from the integrated out degrees of freedom is regulated by the $\\dR$ term. This way of handling ``short-range correlation\" is analogous to Bogner {\\it et al.}'s derivation \\cite{kuo} of ``$V_{low-k}$\" based on renormalization-group theory. While our approach here is, in certain cases, not in strict accordance with the systematic power-counting scheme of EFT proper, nevertheless the severity of this potential shortcoming should vary from one case to another (see discussion in Ref.~\\cite{beaneetal2}). For the $pp$ and $hep$ amplitudes under consideration, the degree of $\\Lambda$-dependence exhibited by the numerical results does suggest that deviations from rigorous power-counting cannot be too significative. Indeed, this type of ``resilience\" may also explain why the SNPA calculation in Ref.~\\cite{MSVKRB} gives a result very similar to the present one. It is true that the two-body terms in MSVKRB are not entirely in conformity with the chiral counting scheme we are using here; some terms corresponding to chiral orders higher than \\nlo3 are included, while some other terms which are \\nlo3 in EFT are missing (see Appendix B.3). Most importantly the $\\dR$-term -- that plays a crucial role here -- is omitted in MKSVRB although heavy-meson exchange graphs may account for some part of it. This formal problem, however, seems to be largely overcome by the fact that also in MSVKRB a parameter (the axial $N\\Delta$ coupling strength) is adjusted to reproduce $\\Gamma_\\beta^t$.  Not unrelated to the above issue of power-counting is the question of consistency of embedding ``realistic\" wave functions obtained from ``realistic\" potentials that are fitted $accurately$ to experiments into an EFT framework with the currents obtained to a given order of chiral perturbation theory. It is a well-known fact that potentials that fit experiments are not necessarily unique. The non-uniqueness resides however in the short-range part of the potential, with the long-range part primarily governed by the pion exchange. Let us suppose that one can calculate potentials to a very high order in a consistent expansion (that is, consistent with symmetries etc.). The structure of the potential would depend on various aspects of the calculation. For instance, although they all may fit equally well various experimental data such as e.g., nucleon-nucleon scattering, different regularizations would lead to different potentials, the difference residing mainly in the short-range part. One might worry that this non-uniqueness would upset the basic premise of an EFT, rendering the predictions untrustworthy.  Another intricate issue, which is also connected to short-range physics, is the off-shell ambiguity. This problem should be absent in a formally consistent EFT. In MEEFT, however, we insert the current operators derived from irreducible diagrams up to a given chiral order between phenomenological (albeit realistic) wave functions. Since the inserted current involves off-shell particles, there can in principle be terms other than those that have been included in our approach. While those additional terms that may be required to eliminate the off-shell dependence are expected to be of higher order than \\nlo3, this issue warrants a further examination.  To answer the above question with full rigor, much more work is needed. However, partial and yet reasonably satsifactory answers can be obtained from this work. For chiral-filter protected processes, we have presented a clear argument that the above-mentioned ambiguity does not matter at the level of accuracy in question. The results listed in Table \\ref{TabS} for the $P$-wave capture (to which the chiral-protected time component of the axial current contributes) demonstrate this point. The question of short-distance ambiguity arises only for chiral-unprotected processes like the GT transition. As already explained, however, the $\\dR$ regularization essentially removes this ambiguity. The point is that the physics of the degrees of freedom above the cutoff scale $\\Lambda$ gets lodged in the short-range $\\dR$ term. In fixing this term as a function of $\\Lambda$ via the experimental value of $\\Gamma_\\beta^t$, one is essentially incorporating the short-range correlations that render low-energy physics insensitive to short-distance physics.  As for the off-shell problem, we note that for processes involving a long-wavelength external current -- such as the solar $pp$ and $hep$ reactions -- the off-shell ambiguity should be small, so long as one uses high-quality phenomenological wave functions that accurately describe processes without the external current. The wave functions used here describe with high accuracy a rich ensemble of data available for the systems in question; they describe very well the three-nucleon scattering states, and furthermore, the $n^3$He elastic scattering cross section as well as the coherent scattering length calculated with these wave functions are in excellent agreement with the experiments. What is involved here seems to be a generic feature. A similar stabilizing mechanism is at work when Bogner {\\it et al}.~\\cite{kuo} arrive at a unique effective force $V_{low-k}$ by integrating out the high-energy/momentum components contained in various ``realistic\" potentials. Nuclear physics calculations done with this effective force~\\cite{kuo2} have much in common with the MEEFT calculation described here. Furthermore, we remark that different off-shell properties reflect different choices of the field variables and that, for each choice, the LECs need to be readjusted. It is in principle possible to choose the field variables in such a manner that off-shell contributions become highly suppressed. We are essentially adopting this particular choice by using the forms of the transition operators described above and adjusting the corresponding LEC, $\\dR$, to reproduce $\\Gamma^t_\\beta$.   A possible approach that is formally consistent with systematic power counting is the pionless EFT based on the power divergence subtraction (PDS) scheme \\cite{pds} (for a recent review, see Ref.~\\cite{seattle}), which has been applied to the $pp$ fusion \\cite{bc01}. Due to the fact that this scheme also involves one unknown low-energy constant, PDS has not so far led to a definite prediction on the $pp$ fusion rate. The problem is that this approach cannot be readily extended to systems with $A\\ge3$, in particular to electroweak transition amplitudes in these systems. What is lacking presently is a method to correlate in a unified framework the observables in different nuclei (different mass numbers). This limitation keeps one from exploiting the experimental data available for the $A\\ge3$ nuclei to fix unknown LEC.  Apart from the basic problem of organizing chiral expansion for complex nuclei from ``first-principles\", a plethora of parameters involved would present a major obstacle. (For recent efforts in this approach, see Ref.~\\cite{seattle,bedaque02} and references given therein.) This difficulty is expected to be particularly pronounced for the $hep$ reaction.  There has been a series of intensive studies by the J\\\"ulich Group to extend EFT calculations in the Weinberg scheme to systems with three or more nucleons~\\cite{epeetal00}. The relationship between this approach and the phenomenological potential approach has been examined in great detail. This line of study, however, has been so far limited to nuclear observables that do {\\it not} involve the electroweak currents. An extension of the formalism developed in Ref.~\\cite{epeetal00} to electroweak transitions should be extremely useful."
    },
    "0208/astro-ph0208177_arXiv.txt": {
        "abstract": "White Dwarf Research Corporation is a 501(c)(3) non-profit organization dedicated to scientific research and public education on topics relevant to white dwarf stars. It was founded in 1999 in Austin, Texas to help fulfill the need for an alternative research center where scarce funding dollars could be used more efficiently, and to provide a direct link between astronomers who study white dwarf stars and the general public. Due to its administrative simplicity, WDRC can facilitate the funding of multi-institutional and international collaborations, provide seamless grant portability, minimize overhead rates, and actively seek non-governmental funding sources. I describe the motivation for, and current status of, one of the long-term goals of WDRC: to establish a permanent endowment for the operation of the Whole Earth Telescope. I pay particular attention to fund-raising efforts through the website at {\\tt http://WhiteDwarf.org/donate/} ",
        "introduction": " ",
        "conclusions": ""
    },
    "0208/astro-ph0208494_arXiv.txt": {
        "abstract": "{ Stephan's Quintet (SQ) is a system consisting of at least four interacting galaxies  which is well known for its complex dynamical and star formation history. It possesses a rich intergalactic medium (IGM), where  hydrogen clouds, both atomic and molecular, associated with two starbursts  (refered to as SQ~A and B) have  been found. In order to study the extent, origin and fate of the intergalactic molecular gas and its relation to the formation of stars outside galaxies and Tidal Dwarf Galaxies (TDGs), we mapped with the IRAM 30m antenna the carbon monoxide (CO) towards several regions of the IGM in SQ. In both SQ~A and B, we detected unusually large amounts of molecular gas ($3.1 \\times 10^9$ \\msun \\ and $7 \\times 10^8$ \\msun, respectively). In contrast, no significant CO detection was achieved towards HII regions south of the pair NGC~7318a/b despite their high \\halpha \\ luminosities. The molecular gas is very extended in both SQ~A and SQ~B, over areas of between 15 and 25 kpc. The CO clouds seem to have otherwise different properties and may be of a different nature. The integrated CO line of SQ~A is in particular much wider than in SQ~B. Its CO spectrum shows emission  at two velocities (6000 and 6700 \\kms) that are coincident with two HI lines. The strongest emission at 6000 \\kms \\ is however spatially offset from the HI emission and situated on a ridge south-east of the starburst region. In SQ~B the CO emission coincides with that of tracers of star formation (\\halpha, 15 $\\mu$m and radio continuum). The CO peak lies slightly offset from the HI peak towards a steep HI gradient. This is indicating that the molecular gas is forming in-situ, possibly in a region of compressed HI, with subsequent star formation. The star forming region at SQ~B is the object in SQ that most resembles a TDG. ",
        "introduction": "\\begin{figure*} \\resizebox{\\hsize}{!}{\\rotatebox{270}{\\includegraphics{sq_abc2.ps}}} \\caption{ An archival V-band image from CFHT   of Stephan's Quintet, showing the positions observed by us. NGC~7320 is a foreground galaxy. The fourth member of the group, NGC~7320c lies about 4 arcmin east of NGC~7319. The velocities are taken from \\citet{Sulentic01}. The positions observed in CO are indicated by circles. The large  circle shows the central (i.e. offset 0,0) position in each region and  gives the size of the CO(1--0) beam. } \\label{opt_map} \\end{figure*}  Stephan's Quintet (Hickson Compact Group 92; hereafter SQ) is one of the best studied examples of a Hickson Compact Group. It contains four interacting galaxies, NGC~7319, NGC~7318a, NGC~7318b, and NGC~7317. A fifth galaxy, NGC~7320, happens to be a foreground object.  An optical image of the group is presented in Fig.~\\ref{opt_map}, \\citep[see also the excellent image in][]{Arp76}. One of its most striking properties is that the major part of the gas is in the intragroup medium, most likely the result of interactions in the past and present.  A plausible scenario for the dynamical history of SQ is presented by \\citet{Moles97}. They suggest that a few times 10$^8$ yr ago the group experienced a collision with NGC~7320c, a galaxy $\\sim$4\\,arcmin to the east of NGC~7319 but with a very similar recession velocity \\citep[6583 \\kms,][]{Sulentic01} to the other galaxies in SQ. This collision removed most of the gas of NGC~7319 towards the west and east,  and produced the eastern tidal tail which connects to NGC~7319. Presently, the group is experiencing another collision with the ``intruder'' galaxy NGC~7318b which strongly affects the interstellar medium (ISM) removed during the first collision. NGC~7318b  has a recession velocity of 5765\\,\\kms, in contrast to the other members of the group which have velocities close to 6600\\,\\kms. \\citet{Sulentic01}, in a multi-wavelength study of the group, confirm this scenario and suggest that the group has been visited twice by NGC~7320c. The first collision created the very faint tidal arm east of the interloper NGC~7320, whereas the second interaction produced the tidal arm which stretches from NGC~7319 eastwards.  This violent dynamical history has induced star formation at various places outside the individual galaxies.  ISOCAM mid-infrared and H$\\alpha$ observations have revealed a starburst region \\citep[object A in][hereafter called SQ~A]{Xu99} at the intersection of two faint optical arms stemming north from NGC~7318a/b. Several knots of star formation are visible in the tidal arm extending from NGC~7319 to the east. At this position, identified as B by \\citet{Xu99} and hereafter called SQ~B, there is also mid-infrared and H$\\alpha$ emission, although much weaker than in SQ~A.  Using HST observations, \\citet{Gallagher01} found 115 candidate star clusters, most of them distributed among the tidal debris of  SQ. From color-color diagrams they estimated their ages ranging from 2--3 Myr up to several Gyr. The distribution of ages sheds light on the star formation history in SQ: The youngest star clusters (with ages of less than 10 Myr) are found in SQ~A and south of NGC~7318a/b, while somewhat older star clusters, with ages between 10 and 500 Myr are in the young tidal tail  and around NGC~7319. This is consistent with the picture that the eastern tidal tail was  produced in a previous interaction whereas the collision and star formation around SQ~A is ongoing. \\citet{Mendes01} have carried out Fabry-Perot observations of the \\halpha \\, emission in SQ~A and south of NGC~7318a/b, from which they determined the velocity curves for several \\halpha \\ emitting regions.  SQ~B was not covered by their observations. They identified seven Tidal Dwarf Galaxy (TDG) candidates, selected as objects exhibiting a velocity gradient compatible with rotation and possessing an $L_{\\rm B}/L_{\\rm H\\alpha}$ ratio consistent with that of a dwarf galaxy.  Abundant atomic hydrogen is present to the east of the 3 central galaxies, around SQ~A and south of NGC~7318a \\citep[][see their Fig. 5] {Shostak84,Williams02}. \\citet{Williams02} suggest that the atomic gas towards the east of SQ consists of two tidal features, each connected to one of the optical tidal arms (the old and the new one, see Fig.~\\ref{opt_map}). Molecular gas has been found at the position of SQ~A by \\citet{Gao00} with BIMA and by \\citet{Smith01} with the NRAO 12m radio telescope.  Both the molecular and the atomic gas in this location present two velocity components, centered at about 6000 and 6700\\,\\kms, which implies that they originate from different galaxies.  \\citet{Braine01} in a study of the CO emission of TDGs detected 2.9$\\times$10$^8$\\,\\msun \\  of molecular gas at SQ~B. In order to follow up on these detections and to elucidate the extent, origin and fate of the CO in the intergalactic medium (IGM) in SQ, we embarked on a single dish survey, the results of which we report in this paper.  Details of our observations can be found in Section 2. In Section 3 we present the results of these observations, as well as a comparison between the molecular gas and emission at other wavelengths. We discuss our findings in Section 4 and summarize our conclusions in Section 5. Following \\citet{Williams02}, who assumed a recession velocity of 6400 \\kms \\ and $H_0$=75\\,km\\,s$^{-1}$\\,Mpc$^{-1}$, we adopt a distance of 85\\,Mpc, in which case 10\\arcsec \\ correspond to 4.1\\,kpc. ",
        "conclusions": "We have observed CO in several regions in SQ. Our main results are:  \\begin{enumerate}  \\item In both SQ~A and SQ~B we detected large amounts of molecular gas of $3.1 \\times 10^9$ \\msun\\ and  $7.0 \\times 10^8$ \\msun, respectively. The molecular-to-atomic gas mass ratios are high (1.2, respectively 0.5), much larger than what has been  found in typical TDGs \\citep{Braine01}. The molecular gas is very extended, over regions of about 15 -- 25 kpc.  \\item The CO spectrum in SQ~A consists of lines at  different velocities, centered at 6030 and 6700 \\kms, with most of the emission coming from the low velocity component, whereas in SQ~B the CO is at 6625 \\kms. The kinematical agreement (velocity, width and spectral shape) between the CO and HI lines is very good in both regions, except for a small offset in the low-velocity components of the CO and HI lines in SQ~A (the HI line peaks at 6000 \\kms).  \\item In SQ~A, the CO emission at 6000 \\kms \\  is found south of the starburst region, covering the southern part of the HI distribution at this velocity and coinciding with X--ray and radio continuum emission which are indicative of shocks. The much weaker CO emission at 6700 \\kms \\ is more spatially concentrated. We conclude that the presence of large amounts of extended molecular gas at 6000 \\kms \\ implies that the shocks traced by X--ray and radio continuum emission only affect the gas at 6700 \\kms \\ and that the apparent spatial coincidence is a projection.   \\item In SQ~B there is good spatial agreement between the tracers of star formation (\\halpha, 15 $\\mu$m and radio continuum) and the CO emission. The CO peak is slightly offset from the HI peak towards a steep HI gradient. We interpret this as indicating that the molecular gas is forming in-situ, possibly in a region of compressed HI and that the star formation is taking place subsequently. This, together with the fact that the emission at different wavelengths is similar to that of normal star forming galaxies and that SQ~B has a clear tidal origin, makes it the object in SQ that most resembles a TDG.    \\item We have searched without success for CO in several objects around NGC~7318a/b in which \\citet{Mendes01} found \\halpha \\ emission. The upper limits for the ratios between molecular gas mass and SFR (derived from the \\halpha \\ emission) are  lower than those found in spiral galaxies and TDGs. The molecular-to-atomic gas mass ratio is also low, though not unusual. The low atomic and molecular gas content of these regions makes them different from TDGs and will not allow them to sustain star formation for a long time.  \\end{enumerate}  \\begin{acknowledgement} We would like to thank J. Iglesias-P\\'aramo and J. V\\'\\i lchez for making available their \\halpha \\ data to us, L. Verdes-Montenegro for the HI data, C. Xu for the ISOCAM and \\halpha \\,data and S. Gallagher for useful information about the star clusters in SQ. Thanks are also due to the referee, G. Petitpas, for the careful reading of the manuscript and detailed suggestions for improvements. VC would like to acknowledge the partial support of JPL contract 960803 and EB acknowledges financial support from CONACyT via project 27607-E. We made use of data from the Canadian Astronomy Data Center, which is operated by the Dominion Astrophysical Observatory for the National Research Council of Canada's Herzberg Institute of Astrophysics.  \\end{acknowledgement}"
    },
    "0208/astro-ph0208388_arXiv.txt": {
        "abstract": "We show that accretion disks with magnetic fields in them ought to make jets provided that their electrical conductivity prevents slippage and there is an ambient pressure in their surroundings.  We study {\\bf equilibria} of highly wound magnetic structures. General Energy theorems demonstrate that they form tall magnetic towers whose height grows with every turn at a velocity related to the circular velocity in the accretion disk.  The pinch effect amplifies the magnetic pressures toward the axis of the towers whose stability is briefly considered.  We give solutions for all twist profiles $\\Phi(P)=\\Omega(P)t$ and for any external pressure distribution $p(z)$.  The force--free currents are given by ${\\bf j} = \\widetilde{\\alpha}(P){\\bf B}$ and we show that the constant pressure case gives $\\widetilde{\\alpha} \\propto P^{-{\\scriptscriptstyle{1\\over 2}}}$ which leads to analytic solutions for the fields. ",
        "introduction": "\\label{sec1} To take some of the most dynamic objects in the universe in which `apparently superluminal' motions have been seen and to work on such problems using statics may seem an indication of a seriously deranged scientist.  Nevertheless, the potential energies involved in any problem can be studied statically and it is those potential energies that drive the motion.  Thus, without studies of the way the potential energies operate, the basic understanding of {\\bf why} the motions occur in the way they do may be lost.  As \\citet{ed26} said `the chief aim of the physicist in discussing a theoretical problem is to obtain ``{\\bf insight}'' -- to see which of the numerous factors are particularly concerned in any effect and how they work together to give it'. \\begin{quote} {\\bf Even a perfect model of a phenomenon, that gives all the observables correctly, is not good science until it is analysed to show which aspects are essential for the phenomenon.} \\end{quote} An unnecessarily detailed model, which reproduces the phenomenon, can actually be a barrier to understanding.  The lack of real understanding of what makes red--giants is a case in point!  It may require the insight of an Eddington rather than the calculations of a Chandrasekhar to simplify the model to the bare essentials.  Within an accretion disk any radial magnetic field will be sheared and stretched by the differential rotation, so the resultant toroidal magnetic field will grow until it is strong enough to arch up out of the disk with the gas flowing back down.  \\citet{hoy60} were the first to describe this instability but the conditions for its occurrence were more precisely calculated by \\citet{par65}.  In the resulting configuration the mass of gas in the accretion disk continues to anchor and to twist the feet of the flux tube.  We show here that if there is an ambient pressure in the tenuous gas above the accretion disk then the flux tube will grow in height with every twist of its feet but will not expand laterally. Thus a tall tower of magnetic field is formed whose collimation or aspect ratio $Z/R$ increases linearly with the number of twists.  My thesis is that an important pre-requisite to understanding the dynamics of the jets above accretion disks is a serious study of the magneto-statics of the magnetic fields that they twist into their coronae.  My first studies in this direction ended in total failure. In 1979 I had a mechanism that would give a tall tower of magnetic field that grew more collimated with every twist.  With much enthusiasm I made a more detailed and exact calculation but to my amazement and chagrin the field managed to expand to infinity and disconnect itself after just over half a turn!  The hoped for collimation that got better with every turn ended with only half a turn when the degree of collimation was not a needle--like, one degree, but a full 120 degrees! See Figure 1.  \\begin{figure} \\begin{center} \\includegraphics[scale=.65]{dlbfig1_new.eps} \\caption{After half a turn a field stretches to very large distances in the absence of a confining coronal pressure.  After ${2\\pi \\over {\\sqrt 3}}\\ {\\rm turns}\\ = 207\\deg$ the field stretches to infinity and further turning does no work.  For more recent work on field opening see \\citet{uzab} and \\citet{uz02}.} \\end{center} \\end{figure}  \\noindent Years later I wrote up this problem \\citep{dlb94} for a conference celebrating Mestel's work. It proved to be a fine example of a phenomenon long advocated by \\citet{aly8495} in solar MHD, see also \\citet{stu91}.  This rekindled my interest, but it still took two years for me to realise an ambient coronal pressure external to the field would prevent it expanding to infinity.  To do so would take too much energy.  When the field cannot expand to infinity, the continual turning of the accretion disk does wind many turns into the corona, provided the conductivities are great enough for flux freezing at the feet.  Then the original argument comes into operation and tall towers will be generated whose heights grow with each successive turn.  Basic theorems on magneto-statics and much simplified models of such magnetic towers in a constant external pressure were given in paper II \\citep{dlb96} which used a rigidly rotating inner disk and a fixed outer disk to which the field returned.  Simplified models with realistically rotating accretion disks were derived in a conference paper \\citet{dlb01}.  The present paper is a development of those calculations to allow for a pressure that decreases with height.  We also find a better approximation for the distribution of twist with height along each field line.  This improves on the assumption of a uniform distribution made earlier.  We derive the {\\bf magneto-statics} of force--free magnetic fields whose feet have been twisted by an accretion disk.  We study the {\\bf equilibria} as a function of the twist angles.  The magnetic field lines labelled $P$ being twisted by an angle $\\Phi(P)=\\Omega(P)t$ at their feet which are anchored in the disk.  Here $\\Omega(P)t$ is the differential rotation of the feet of the flux tube and $t$ is a parameter that gives the amount of that twist.  If we increase $t$ we change our equilibrium along a Poincar\\'{e} sequence.  Thus, in the language of his catastrophe theory, $t$ is our control parameter.  Of course, if all inertia were unimportant and if a real accretion disk were dragging the feet of the flux tubes at relative angular velocity $\\Omega(P)$, then this sequence of equilibria would give us a film of how the field would evolve in the presence of the external pressure field $p(z)$.  This would not give a true picture just because the inertial terms are not always negligible, nevertheless such sequences are very instructive in that they show us how the field would like to change if inertia did not slow its accelerations.  We find it possible to get a reasonable understanding of this problem for any specified $\\Omega(P)$ and any chosen external pressure field $p(z)$.  We emphasise that we have the simplest magnetically dominated model. Above the disk, magnetism so dominates that the field is force--free everywhere within the towers, but at their surfaces the magnetic forces balance the ambient hot gas pressure $p(z)$.  Where there is gas there is no field, where there is magnetic field there is no gas pressure but we nevertheless assume perfect electrical conductivity. When we view our sequences of equilibria parameterised by $t$ they are so reminiscent of what is seen in radio galaxies, quasars, star--forming disks with Herbig--Haro objects, etc., that it is hard to resist the temptation of talking in the terminology of dynamics rather than statics.  Please remember that the velocities we talk of are only  the velocities that would occur in the absence of inertia when accelerations can become arbitrarily large without penalty.  We consider some effects of inertia in paper IV \\citep{dlb02}.  We believe that the very simple calculation given in section \\ref{sec2.3} contains the answer to the question `Why do flat accretion disks produce needle like magnetic jets?', but further refinements, given in paper IV on detailed field structure, give greater understanding of why jets act as giant linear accelerators and why their electric currents are so concentrated to the tower's axis.  In the above, I have given a personal account of the evolution of my thoughts and calculations but there are many other ideas in this field, which started with the Jet in M87 observed by \\citet{cur18}. Although \\citet{sch12} wrote a book which detailed the theory of synchrotron radiations in 1912, it was not applied to M87 until \\citet{baa56}.  Even though the first quasar 3C273 had a prominent jet these remained an enigma emphasised by \\citet{whe71} in 1970, who rightly % drew attention to a paper by \\citet{leb70}; \\citet{ree71} provided one of the first explanations which evolved into \\citet{bla74} and \\citet{sch74}.  \\citet{bar72} and \\citet{lov76} were among the first to treat magnetically dominated models, while for black holes \\citet{bla77} gave such ideas a new and most interesting twist by producing a magnetic mechanism for extracting the spin energy from a black hole.  \\citet{bla82} have a nice mechanism for driving winds centrifugally.  Extra-galactic radio jets were reviewed with a wonderful series of radio pictures by \\citet{beg84}.   \\citet{sak8587} showed a slow asymptotic collimation of such winds and the wind equations have been much studied \\citep{hey89, app93, pud87, lov91}. \\citet{oka99} criticised the collimation claims of most workers. Computational simulations of jets have been made by \\citet{bel9596} and \\citet{ouy97, ouy99}.  Many of these studies start with a uniform magnetic field at infinity as did \\citet{lov76} and \\citet{shi8586} who produced very convincing jets this way.  However, a uniform field at infinity puts in a collimation at the start, whereas some wish to see both the field and the collimation to emerge as a consequence of the persistent rotation rather than having them inserted as a boundary condition. Jets are by no means confined to relativistic objects but are common in the accretion discs that occur around young stars.  In fact, not long after accretion disks were applied to quasars and mini--quasars \\citep{dlb6971}, and the galactic centre \\citep{eke71, dlb72}, they were applied to star formation \\citep{dlb74}.  The jets that emerge from star--forming disks do not travel at relativistic speeds but at speeds of one or two hundred km/s.  There is a correlation between the maximum circular rotational velocity within the object and the speed of the jet and we give an explanation of this in sections \\ref{sec2.3} and \\ref{sec3.3}.  First numerical calculations of the force--free structures we are advocating, were made by \\citet{li01} but to date they have only calculated the first two turns which give indications of the initial growth of magnetic towers.  \\citet{kra99} studied the launching of jets loaded with plasma, as did \\citet{rom98}, \\citet{con95a} has - like us - looked for force--free self--similar solutions for jets and has considered the possibility of purely toroidal fields \\citep{con95b}, Winds and Jets are considered by \\citet{lov91} and Poynting jets from loaded winds  have been computed by \\citet{lov02}. ",
        "conclusions": "A sequence of static models can be more illuminating than detailed dynamical simulations.  The statics should be understood before the dynamics is attempted.  The continued winding of the magnetic field of an accretion disk will build up tall towers which make magnetic cavities, towering bubbles of magnetic field in the surrounding medium.  Such a conclusion does not depend on axial symmetry.  Non--axysymmetric behaviour is observed in the pretty plasma experiments of \\citet{hsu02}.  When that symmetry is imposed we can calculate the tall tower shapes of the magnetic cavities for any prescribed winding angles $\\Phi(P) =\\Omega(P)t$ and for any prescribed pressure distribution $p(z)$. Examples of these towers are shown in figures 3 and 4.  The heights of these towers grow at a velocity closely related to the maximum circular velocity in the accretion disk.  Our primary results are encapsulated in equations (\\ref{eq44}) and (\\ref{eq45}).  As stellar--mass black--holes form, the winding of the magnetic field of the collapsing core may cause jets to emerge from supernovae as first predicted by \\citet{leb70}.  Such ideas may have application to some $\\gamma$--ray bursts, to the micro--quasars \\citep{mir99} and to SS433 \\citep{mar84}.  The possibility that the elongated hour--glass planetary nebulae, some of the most delicate objects in the sky, may be magnetic towers arising from the accretion disks of their central binaries is particularly appealing."
    },
    "0208/astro-ph0208031_arXiv.txt": {
        "abstract": "We look in detail at the process of mapping an astrophysical initial model from a stellar evolution code onto the computational grid of an explicit, Godunov type code while maintaining hydrostatic equilibrium. This mapping process is common in astrophysical simulations, when it is necessary to follow short-timescale dynamics after a period of long timescale buildup.  We look at the effects of spatial resolution, boundary conditions, the treatment of the gravitational source terms in the hydrodynamics solver, and the initialization process itself. We conclude with a summary detailing the mapping process that yields the lowest ambient velocities in the mapped model. ",
        "introduction": "Many astrophysical phenomena involve a dramatic change between timescales of interest --- the slow convection and simmering in the interior of a white dwarf followed by ignition of a Type Ia supernova, for example, or the accretion of a layer of fuel onto a white dwarf or neutron star leading to ignition and runaway at the base of the layer, producing a nova or an X-ray burst.  These two regimes are difficult to follow with a single hydrodynamic algorithm because of the disparity of the relevant timescales.  Modeling these long timescale events requires an implicit or anelastic hydrodynamic method; short timescale events require explicit hydrodynamic methods that can capture the transient phenomena.  Often, a one-dimensional stellar hydrodynamics code is used to follow the accretion process until just before the fuel reaches the ignition temperature.  Simulations of the flash following the ignition requires a multidimensional hydrodynamics code.  Matching the two regimes is a difficult process, and can introduce numerous errors into a calculation.  Simulations in the atmosphere or interior of a star or compact object often begin with an initial model that was generated by a 1-d implicit code (see for example \\citealt{kepler,glasner,sugimoto,tycho,nova}), that evolves the system through the long timescale processes (accretion, slow convection, simmering of nuclear fuel) until just before the short timescale dynamics begin.  This 1-d initial model is then used as input to a multidimensional, explicit hydrodynamics code (see for example \\citealt{glasner, kercek, zingale, bazan, kane}).  The mapping of a 1-d hydrostatic initial model onto a multidimensional grid is the focus of the present paper.  In the absence of any perturbations or external forces, the system should remain in hydrostatic equilibrium after the mapping.  The mapping process can introduce a variety of errors.  It is common for the two codes to use different equations of state (EOS), which can have a large effect on the structure of the atmospheres.  Even if the basic physical components are the same, the treatment of physical details (for example, Coulomb corrections or ionization) may differ, leading to differences in the pressure of a fluid element between the two codes, even with the same density, temperature, and composition. Simply updating the thermodynamics of the initial model with the new EOS will most likely push it out of hydrostatic equilibrium. One-dimensional initial models are almost always created using mixing length theory to describe the convection, implicitly assuming a velocity field necessary to transport the energy as required.  If the 1-d model was convectively unstable, it is unclear how to define the 2- or 3-d velocity field consistent with the 1-d input velocity field. When mapping into higher dimensions, there is not enough information to set the velocities properly.  Typically the velocities are zeroed during this transition.  Finally, it is unusual for the number of points and the grid spacing in the initial model to match that in the multidimensional grid.  The initial model may have come from a Lagrangian code and will need to be converted into an Eulerian coordinate system via \\begin{equation} dm = 4 \\pi r^2 \\rho(r) dr. \\end{equation} The initial model will then need to be interpolated onto the new grid, which will introduce even more hydrostatic equilibrium errors.  Once the model is on the new grid, differences in the hydrodynamical algorithms can cause problems.  Although the two codes are solving the same equations, it is also very likely that the discretization used in the two codes is different.  The very definition of the quantities on the grid may also differ; the mapping may proceed from a 1-d implicit finite-difference code, where the values on the grid may represent nodal points, to a multidimensional finite-volume code, where the values represent cell averages.  If everything else were constant between the codes, the differences in the discretization and the definition of the variables ({\\em{e.g.}} pointwise values vs.\\ cell averages) is enough to upset the hydrostatic equilibrium.  Poor boundary conditions can drive velocities on the grid, pushing our initial model out of hydrostatic equilibrium.  Ideally, the boundaries should match the physics of the initial model and present a smooth state to the hydrodynamics solver.  In this article, we look at bringing an initial model to hydrostatic equilibrium by studying the effects of the initialization method, the boundary conditions, and the solver itself.  An initial model atmosphere is mapped onto the computational grid, transverse to the direction of gravity.  Since we are not concerned with perturbing the model after the mapping, all the calculations presented here will be one-dimensional for computational efficiency.  We consider three different initial models---a simple analytic model, a simple hydrostatic equilibrium model with a complicated EOS, and an initial model from an implicit stellar hydrodynamics code.  These three different models will allow us to isolate the importance of the different parts of the code (initialization, boundary conditions, and solver itself).  Our goal is to find an optimal configuration that allows us to hold an initial model from a different hydrodynamics code in equilibrium in the present code until a perturbation or external force that we impose disturbs it.  Furthermore, we want the resulting mapped hydrostatic model to be as close to the original initial model as possible.  Regions where disturbances have not yet propagated should remain quiescent for tens or hundreds of dynamical timescales. The absolute magnitude of the velocity that a simulation can tolerate will be problem dependent; however, it must be considerably less than the velocity of any dynamics we wish to study (e.g., burning front speed, convective speed, and certainly the sound speed.)  For the hydrodynamics, we chose to use PPM \\citep{ppm}, a widely used Godunov method \\citep{godunov}.  PPM solves the Euler equations in conservative form, using a finite-volume discretization, guaranteeing conservation.  PPM is a shock-capturing scheme, which is desirable for the modeling of the rapid transients in the short-timescale regimes we wish to study.  The implementation of PPM we use is contained in the FLASH code \\citep{flash}, and is based on PROMETHEUS \\citep{prometheus}.  While we use PPM to demonstrate results in this paper, only the discussion in \\S \\ref{sec:ppmhse}, where we discuss extensions to the method to better treat the gravitational source term,  is particular to PPM. Discussion of initialization and boundary conditions should be relevant to any Godunov-type method, and perhaps other finite-volume methods. Although FLASH can use an adaptive mesh, in this study we run all the simulations on a uniform grid.  This paper is organized as follows: \\S 2 discusses the hydrodynamics algorithm employed in this study and improvements made to better follow a hydrostatic atmosphere.  In \\S 3 we look at the initial models that will form the basis of our tests.  \\S 4 discusses the different boundary conditions considered.  In \\S 5, we show the results of a grid of calculations of each of the different initial models, varying spatial resolution, boundary conditions, and the details of the hydrodynamics.  Finally, we conclude in \\S6. ",
        "conclusions": "We have studied the effects of the initialization process, boundary conditions, resolution, and the hydrodynamic algorithm on the process of mapping an initial model onto a Eulerian grid in a Godunov-type code. We saw that depending on how one goes about the process, the result can be a nicely relaxed, quiet mapping, or one dominated by high velocities, swamping out any physical processes that one may be interested in studying.  Boundary conditions have the greatest impact on maintaining the stability of the hydrostatic atmosphere.  The standard reflecting boundary is poorly suited for maintaining a hydrostatic envelope---the inability of minor pressure disturbances to escape off the bottom of the grid as the model relaxes results in high velocities.  The actual choice of boundary condition should reflect the physics of the atmosphere being supported.  If the initial model is isothermal, then an isothermal/HSE boundary condition provides with best match.  The constant entropy boundary may be a better match for an atmosphere that is generated by a stellar evolution code.  For any model, a hydrostatic boundary condition, using differencing that matches that of the initialization, with a secondary constraint that matches the physics of the model atmosphere is the best solution.  Assuming that the density is constant in the boundary proved to be the worst assumption. The degeneracy of the EOS puts strong demands on the temperature profile to counteract the weight of the atmosphere.  This assumption may fare better with a gamma-law EOS, but this was not considered in the present paper.  We also demonstrated that, once an appropriate boundary condition is chosen, the treatment of source terms in the hydrodynamic solver can have an impact on the stability of the atmosphere.  Our modified-states PPM was effective in reducing the magnitude of the spurious velocities generated as our model relaxed onto our grid. This method, or that proposed by \\citet{leveque}, can be adopted to most Godunov type codes to increase the accuracy of the hydrodynamics in the presence of gravity.  As expected, the resolution is important in reducing the errors in maintaining a hydrostatic atmosphere.  The resolution should be chosen to be fine enough to resolve the scale height of the atmosphere well, and to keep ambient velocities smaller in magnitude than whatever physical processes under study would yield.   In an AMR calculation, the coarsest resolution used in the simulation must be this fine, even in regions of the atmosphere with no features.  Boundary conditions, source terms, and poor resolution can all be sources of spurious velocities throughout a calculation.   By contrast, initialization methods can cause at most a transient while the simulation settles into hydrostatic equilibrium.   This transient may take a while to settle, however, or may have dynamical consequences later in the simulation, so care must be taken.  The initialization methods described in this paper produce profiles which generate very small transients.  We note that the methods that we describe in this paper can be applied to any initial hydrostatic atmosphere.  As a final example, we show the havoc a poorly initialized model can wreak in a simulation.  Figure \\ref{fig:beforeafter} represents two-dimensional simulations evolved from a one dimensional model provided by S.~A.~Glasner (2002, private communication) of a classical nova precursor; it is a model from the same simulation that produced the 1d model used in \\cite{glasner,kercek}, but at an earlier time, before convection begins.  Thus, one could hope to examine the multidimensional onset of convection in these nova precursors \\citep{novapaper}.  The convection is driven by nuclear reactions `simmering' near the interface between the C/O white dwarf and the accreted layer of stellar material.  Since the vertical scale of the simulation is significant compared to the radius of the white dwarf, plane-parallel inverse-square gravity is used rather than constant gravity.  The model is perturbed in the highest-temperature region of the accreted atmosphere with a 10\\% temperature increase at time $t = 0$.  The white contours in the figures enclose the region in the simulation with a temperature greater than this perturbed temperature.   The black contour marks the white dwarf/accreted material interface.  Without taking any of the precautions outlined in this paper, an unphysical `settling' occurs, caused by poor boundary conditions and differences in the EOS.  This generates large velocities ($v \\sim 5\\times 10^6 {\\ \\mathrm{cm\\,s^{-1}}}$) and compressional heating, as shown in the figure.  The heating, combined with unphysical mixing across the interface caused by the large velocities, then cause a completely spurious layer of increased burning, which then dominates the long-term evolution of the simulation.    By contrast, the simulation using constant-temperature second-order boundary conditions, second-order initialization, and the modified-states PPM shows the beginning of the formation of convective rolls.  Both simulations were done on a $1920 \\times 640$ uniform mesh, with a computational domain of $1 \\times 10^7 {\\ \\mathrm{cm}} \\times 3 \\times 10^7 {\\ \\mathrm{cm}}$; Figure \\ref{fig:beforeafter} shows only the domain near the interface of the white dwarf and the accreted material."
    },
    "0208/astro-ph0208207_arXiv.txt": {
        "abstract": "{We  present  Giant   Meterwave  Radio   Telescope  (GMRT) observations of  the \\ion{H}{i}~21~cm line and Very  Large Array (VLA) observations  of the OH~18~cm  line from  the Seyfert~2  galaxy Mrk~1. \\ion{H}{i}  emission is  detected from  both Mrk~1  and  its companion NGC~451. The \\ion{H}{i} emission  morphology and the velocity field of Mrk~1  are disturbed.   We speculate  that the  nuclear  activities of Mrk~1 are triggered by tidal interactions.  We estimate the \\ion{H}{i} masses of Mrk~1  and NGC~451 to be $8.0(\\pm0.6)\\times10^{8}$~M$_\\odot$ and  $1.3(\\pm0.1)\\times10^{9}$~M$_\\odot$ respectively.   We  have also detected the \\ion{H}{i}~21~cm line and the OH~18~cm line in absorption toward the nucleus of Mrk~1  at a blueshifted velocity with respect to its systemic  velocity indicating an  outflow of atomic  and molecular gas. Two  OH lines, at 1665  and 1667~MHz, are detected.   Each of the profiles  of  the  \\ion{H}{i}   and  OH  absorption  consists  of  two components  that are  separated  by $\\sim$~125~km~s$^{-1}$.   Gaussian fitting  gave   dispersions  of  $\\sim44$~km~s$^{-1}$   for  both  the components  of  the \\ion{H}{i}  absorption.   The  profile  of the  OH absorption  is  qualitatively  similar   to  that  of  the  \\ion{H}{i} absorption.   Both  components  of  the OH  absorption  are  thermally excited.   The  peak optical  depths  of  the  two components  of  the \\ion{H}{i}     absorption    are     $(7.3\\pm0.4)\\times10^{-2}$    and $(3.2\\pm0.4)\\times10^{-2}$.  The corresponding  peak optical depths of the   1667~MHz  OH   absorption  are   $(2.3\\pm0.3)\\times10^{-2}$  and $(1.1\\pm0.3)\\times10^{-2}$.   The higher  velocity  components of  the \\ion{H}{i} and OH (1667~MHz) absorption lines are blueshifted from the [\\ion{O}{iii}]$\\lambda5007$,    [\\ion{O}{i}]$\\lambda6300$,   and   the systemic velocity  by $\\sim$~100~km~s$^{-1}$, but  are consistent with the  [\\ion{O}{ii}]$\\lambda3727$ velocity.   We explain  these velocity discrepancies as due  to shock ionization of a  region which is pushed forward due  to shocks  in front of  the radio nucleus  thereby giving apparent    blueshift   to    \\ion{H}{i},   OH,    and   [\\ion{O}{ii}] velocities.          The         optical          depth         ratios $\\tau_\\mathrm{\\ion{H}{i}}/\\tau_\\mathrm{OH}^{1667}$    of    both   the components  of   the  \\ion{H}{i}  and  OH   absorption  are  $\\sim$~3, indicating   their   origin   in   dense  molecular   clouds.    Using OH/A$_\\mathrm{v}$ values for the  Galactic molecular clouds, we obtain 9  $<$  A$_\\mathrm{v}<$  90  toward   the  line  of  sight  of  Mrk~1. ",
        "introduction": "Both the AGN  and the nuclear starburst activities  in galaxies which require  inflow of  material  toward  the centre  either  to fuel  the central black hole  or to cause rapid burst  of nuclear star formation can be  accomplished by tidal interactions (\\cite{her95}).   It is not clear,  however,  in the  case  of  Seyfert  galaxies whether  nuclear activities in  these low luminosity  active galactic nuclei  (AGN) are due to interactions as found in QSOs, radio galaxies, and BL Lacs (see \\cite{der98} for  a review on  the subject). It is  generally accepted that interactions  leading to mergers (bound interactions)  may play a significant  role in  triggering  nuclear activities  than unbound  or hyperbolic encounters (\\cite{der98}).  Interactions can be effectively traced via \\ion{H}{i} 21~cm  line emission from galaxies as \\ion{H}{i} disks often extend well beyond the optical radii of galaxies where the disks  respond  quickly  to  gravitational  perturbations.  \\ion{H}{i} emission  studies may  be  particularly useful  since  most often  the \\ion{H}{i}  morphology provides  evidences of  interactions  which are undetectable at optical wavelengths (e.g., \\cite{sim87}).  \\ion{H}{i}  in absorption  can  trace kinematics  and distribution  of atomic gas near  the centres of active galaxies on  the size scales of their background radio sources. The advantage of absorption studies is that it can detect relatively  small quantities of gas irrespective of the  redshift  of  the  object.   Recently,  \\cite{gal99}  have  found \\ion{H}{i} rich absorbing disks on the scales of a few hundred parsecs in  several  Seyfert  galaxies.    As  a  result  of  intense  nuclear activities,  gas in  the central  regions  of active  galaxies may  be perturbed due to interactions of the radio plasma with the surrounding ISM which may result in bulk outflows of material (e.g., \\cite{tad01}, \\cite{mor98}). The  molecular gas near the centres  of active galaxies can  be  traced  via  18~cm~OH  line  in absorption.   The  18  cm  OH absorption line is sensitive to  molecular gas in both the diffuse ISM and in the dark clouds with  OH to H$_{2}$ ratio being almost constant over a  large variety of Galactic clouds  (\\cite{lis96}). Studies have shown that chances  of detecting OH absorption are  higher in infrared luminous galaxies (Schmelz et al. 1986).  In   this   paper,   we   present  synthesis   observations   of   the \\ion{H}{i}~21~cm  line obtained with  the GMRT  and the  OH~18~cm line obtained with the VLA of the infrared luminous active galaxy Mrk~1 and its companion  NGC~451. The global properties of  Mrk~1 are summarized in the next section. The details of observations and data analyses are given  in  Sect.~3. The  results  are  presented  in Sect.~4.  Sect.~5 discusses  the radio  continuum properties,  \\ion{H}{i}  emission, and \\ion{H}{i} and OH absorption. The conclusions are in the last section. ",
        "conclusions": "We have  presented the observations  of the Seyfert~2 galaxy  Mrk~1 in the \\ion{H}{i}  21 cm line  using the  GMRT and in  the OH 18  cm line using the VLA.  Unlike the optical morphology, the \\ion{H}{i} emission morphology of Mrk~1  indicates that this galaxy is  disturbed which we interpret  as due  to  tidal interactions  with  the nearby  companion NGC~451.   We also showed  based on  the dynamical  study of  Mrk~1 -- NGC~451  system that  the interaction  is  bound leading  to a  merger within  a  small fraction  of  their  orbital  time period.   This  is consistent with  the hypothesis that the bound  interactions should be more   efficient  in  triggering   nuclear  activities   than  unbound interactions.  The  \\ion{H}{i} and  OH absorption detected  toward the nucleus of  Mrk~1 indicates  an outflow of  both atomic  and molecular gas.   The  column  densities   of  the  detected  \\ion{H}{i}  and  OH absorption indicate that the line of sight toward the nucleus of Mrk~1 is  rich  in both  atomic  and molecular  gas.   The  gas detected  in absorption is kinematically different than that traced via CO emission and  water megamaser  emission from  Mrk~1.  We  found  evidences that shocks  (presumably   due  to  nuclear  activities)   can  affect  the kinematics of gas  near the nucleus. The \\ion{H}{i}  and OH absorption being    blueshifted   from    the   systemic    velocity    and   the [\\ion{O}{iii}]$\\lambda5007$   velocity  while   consistent   with  the [\\ion{O}{ii}]$\\lambda3727$  velocity  is understood  in  terms of  the shock   ionization  of   gas  (which   predicts  enhancement   of  the [{\\ion{O}{ii}] line intensity)  and an outflow of ISM  in front of the shock. Based  on the optical depth  ratios and the line  widths of the \\ion{H}{i}  and OH  absorption, we  speculate that  the  absorption is arising in turbulent molecular clouds  of similar types as those found near the Galactic centre.  These observations also imply that the line of sight toward the nucleus of Mrk~1 is heavily obscured."
    },
    "0208/astro-ph0208213_arXiv.txt": {
        "abstract": "We analyze Ly$\\alpha$ fluoresced H$_{2}$ lines observed in the UV spectrum of Mira~B.  We identify 13 different sequences fluoresced by 13 different H$_{2}$ transitions within the Ly$\\alpha$ line.  The observed H$_{2}$ line ratios within these sequences imply significant line opacity, so we use a Monte Carlo radiative transfer code to model the line ratios, correcting for opacity effects.  We find the observed line ratios can best be reproduced by assuming that the H$_{2}$ is fluoresced in a layer between the observer and Mira~B with a temperature and column density of $T=3600$~K and $\\log {\\rm N(H_{2})}=17.3$, respectively.  The strengths of H$_{2}$ absorption features within the Ly$\\alpha$ line are roughly consistent with this temperature and column.  We use the total flux fluoresced within the 13 sequences to infer the Ly$\\alpha$ profile seen by the H$_{2}$.  In order to explain differences between the shape of this and the observed profile, we have to assume that the observed profile suffers additional interstellar (or circumstellar) H~I Ly$\\alpha$ absorption with a column density of about $\\log {\\rm N(H~I)}=20.35$.  We also have to assume that the observed profile is about a factor of 2.5 lower in flux than the profile seen by the H$_{2}$, and a couple possible explanations for this behavior are presented. Several lines of evidence lead us to tentatively attribute the fluoresced emission to H$_{2}$ that is heated in a photodissociation front within Mira~A's wind a few AU from Mira~B, although it is possible that interaction between the winds of Mira A and B may also play a role in heating the H$_{2}$.  We estimate a Mira~B mass loss rate of $\\dot{M}=5\\times 10^{-13}$ M$_{\\odot}$ yr$^{-1}$ and a terminal velocity of $V_{\\infty}=250$ km~s$^{-1}$, based on wind absorption features in the Mg~II h \\& k lines.  We note, however, that the wind is variable and IUE Mg~II spectra suggest significantly higher mass loss rates during the IUE era. ",
        "introduction": "Mira (o Cet, HD~14386) is one of the most well studied variable stars in the sky, representing the prototype for the Mira class of pulsating variables.  Optical spectra of Mira at its minimum brightness reveal the presence of a hot companion star, Mira~B, which is not easily resolvable from the ground \\citep{ahj26,yy77}.  Nevertheless, both speckle interferometric techniques applied to ground-based observations and observations from the {\\em Hubble Space Telescope} (HST) have proven effective at resolving the 2 members of the Mira system, and HST observations from 1995 show the companion $0.578^{\\prime\\prime}$ from the primary at a position angle of $108.3^{\\circ}$ \\citep{mk91,mk97,mk93}.  The {\\em Hipparcos} distance to the Mira system is $128\\pm 18$~pc \\citep{macp97}.  This is a significant increase from a previous parallax estimate of $d=77$~pc \\citep{lfj52}.  The {\\em Hipparcos} distances for Miras (including Mira itself) have been used to define a new period-luminosity relation for Miras that leads to a distance measurement to the Large Magellanic Cloud consistent with the accepted value \\citep{fvl97,paw00}.  This provides support for the {\\em Hipparcos} distance to Mira and the other variables in its class.  Although the optical spectrum of Mira~B is difficult to separate from that of Mira~A even at Mira~A minimum, \\citet{ahj26} estimated a spectral type of B8 for Mira~B.  The high temperature but low overall luminosity of Mira~B have led to the assumption that Mira~B is probably a white dwarf.  However, this conclusion is complicated by the fact that Mira~B is accreting material from Mira~A's massive cool wind, and the accretion process clearly affects the appearance of Mira~B's optical spectrum, resulting in variability on timescales ranging from minutes to years, with some suggestion of a 14 year periodicity \\citep{bw72,yy77}.  The existence of accretion onto Mira~B makes Mira rather unique in being a wind accretion system in which one can actually spatially resolve the accretor from the star whose wind is feeding the accretion. However, the accretion makes it difficult to be certain whether the continuum emission we see is from Mira~B or from the accretion onto the star.  \\citet{mj84} argue that the dearth of X-rays from Mira~B implies that Mira~B cannot be a compact object like a white dwarf and must instead be a faint main sequence star, in which case the optical and UV emission is entirely from the accretion rather than the star itself \\citep[see also][]{mk96}.  The clearest indicators of accretion onto Mira~B are provided by UV spectra from the {\\em International Ultraviolet Explorer} (IUE).  Spectra taken with IUE show lines such as C~IV $\\lambda$1550 and Si~III] $\\lambda$1892, whose broad widths and high temperatures of formation are most naturally explained by their formation within a hot, rapidly rotating accretion disk \\citep{dr85}.  Mira~B was observed numerous times during the 18 year lifespan of IUE, the archive consisting of a total of 94 usable IUE spectra, including both low and high resolution spectra of both the long and short wavelength regions covered by the IUE spectrographs.  These data show that the UV continuum and emission line fluxes vary by a factor of two or so within the data set \\citep{dr85}.  Ultraviolet spectra taken in 1995 with the Faint Object Camera (FOC) instrument on HST showed fluxes near the low end of the range observed by IUE, but still consistent with the behavior seen within the older IUE data set \\citep{mk97}.  However, when the Space Telescope Imaging Spectrometer (STIS) instrument on HST observed Mira~B on 1999 August 2, the UV spectrum was dramatically different \\citep[][hereafter Paper 1]{bew01}. Continuum fluxes in these most recent observations are uniformly more than 10 times lower than ever observed by IUE or HST/FOC. The STIS observations cover the spectral ranges $1140-1735$~\\AA\\ and $2303-3111$~\\AA, so this tremendous drop in continuum flux presumably extends into the optical regime above 3000~\\AA.  The temperature of the gas responsible for the continuum emission apparently did not change, given that the shape of the continuum did not change.  Thus, the tremendous change in flux is very hard to explain if the emission is from a stellar photosphere, but perhaps not so hard to explain if the emission is from the accreting material itself, in which case a large drop in accretion rate could account for a drop in flux that would not necessarily be accompanied by a temperature change.  The continuum is not the only aspect of the UV spectrum to show remarkable differences between the IUE era and the HST/STIS observations. The UV emission lines also varied.  Many lines showed flux decreases similar to that of the continuum, including the Mg~II h \\& k lines near 2800~\\AA.  The factor of $\\sim 100$ decrease in C~IV $\\lambda$1550 flux was even more extreme.  The P Cygni-like profiles of the Mg~II lines suggest that accretion onto Mira~B drives a warm, fast outflow.  Comparing the Mg~II profiles observed by IUE and STIS suggests that the mass loss rate from Mira~B was significantly lower during the STIS observations than any time IUE observed these lines.  This is consistent with the idea that the accretion rate must have dropped substantially, thereby leading to less mass outflow from Mira~B (see Paper 1).  However, the most dramatic change in the appearance of the far-UV (FUV) region of the STIS spectra was not in the continuum or high temperature line emission but in the appearance of a very large number of narrow H$_{2}$ lines, which in fact dominate the FUV spectrum observed by STIS despite not being detected at all in any of the IUE spectra.  In Paper 1, we announced the discovery of these lines in Mira~B's spectrum, and we noted that all are Lyman band H$_{2}$ lines fluoresced by the H~I Ly$\\alpha$ line at 1216~\\AA.  The Ly$\\alpha$ fluorescence mechanism for exciting UV H$_{2}$ lines has been observed for many astrophysical objects. The mechanism was first described by \\citet{cj77,cj78} to explain H$_{2}$ lines in the solar spectrum.  Molecular hydrogen lines excited by Ly$\\alpha$ have subsequently been observed in red giant stars \\citep{adm98,adm99}, T Tauri stars \\citep{ab81,jav00,dra02a,gjh02}, and Herbig-Haro objects \\citep{rds83,sc95}.  The first astrophysical detections of Ly$\\alpha$ fluoresced H$_{2}$ (excluding the Sun) were from IUE data, but the low sensitivity of IUE meant that only a few of the strongest H$_{2}$ lines could be detected, and even these could generally only be detected with low resolution gratings \\citep[e.g.,][]{ab81}.  The Goddard High Resolution Spectrograph (GHRS) instrument that preceded STIS on board HST equals STIS in sensitivity and could easily detect the H$_{2}$ lines at moderate resolution.  However, GHRS only provided limited wavelength coverage for each exposure, which effectively limited the number of H$_{2}$ lines that could be studied with GHRS \\citep[e.g.,][]{sc95,dra02a}. With STIS, however, one can observe the entire $1140-1735$~\\AA\\ region in a single exposure.  Thus, STIS promises to dramatically increase our understanding of the H$_{2}$ fluorescence process by allowing detection of many, many more lines than either IUE or GHRS.  In this paper, we perform a detailed analysis of the H$_{2}$ lines from Mira~B in order to try to determine where they are coming from, and to see if the lines can shed light on the nature of the accretion process onto Mira~B. ",
        "conclusions": " \\begin{description} \\item[1.] The average centroid velocity of the H$_{2}$ lines is $56.9\\pm 0.2$ km~s$^{-1}$, consistent with the $\\sim 56$ km~s$^{-1}$ radial velocity of the Mira binary.  The average width of the H$_{2}$ lines is $FWHM=19.7\\pm 0.4$ km~s$^{-1}$. \\item[2.] The observed flux ratios within the H$_{2}$ fluorescence sequences show evidence for opacity effects that cause the low wavelength lines to systematically have lower fluxes than the line branching ratios predict, due to the higher opacity of those lines.  We use a Monte Carlo radiative transfer code to model the H$_{2}$ line ratios that emerge from a plane parallel slab.  We find that the observed line ratios are best reproduced by a slab with temperature $T=3600$~K and column density $\\log {\\rm N(H_{2})}=17.3$. \\item[3.] Even for our best fit to the data, there are still significant discrepancies between the observed H$_{2}$ line ratios and those predicted by our radiative transfer code.  The discrepancies are largest for lines in sequences fluoresced from lower energy levels, suggesting that the populations of these lower levels may not be entirely determined by thermal processes. \\item[4.] We find that we can use the total flux fluoresced by each of the 13 H$_{2}$ pumping transitions within Ly$\\alpha$ to map out a self-consistent Ly$\\alpha$ profile seen by the H$_{2}$, assuming the temperature of $T=3600$~K derived from the line ratio analysis.  This means that the H$_{2}$ energy levels can at least to first order be described by a thermal population appropriate for $T=3600$~K.  When we change the assumed temperature, the analysis is not as successful in deriving a smooth, self-consistent Ly$\\alpha$ profile, providing support for the $T=3600$~K measurement. \\item[5.] The observed Ly$\\alpha$ profile is completely redshifted from the rest frame of the star, presumably due to wind absorption eating away the blue side of the line.  The model Ly$\\alpha$ profile seen by the H$_{2}$ that we derive is also completely redshifted relative to the stellar rest frame, implying that the wind absorption must occur before the Ly$\\alpha$ emission encounters the H$_{2}$. \\item[6.] The fluxes of the model Ly$\\alpha$ profile seen by the H$_{2}$ are significantly higher than those of the observed Ly$\\alpha$ profile. Interstellar and/or circumstellar H~I absorption of Ly$\\alpha$ can account for much, but not all, of this discrepancy.  We find that we can best reconcile the modeled and observed Ly$\\alpha$ profiles if we assume an interstellar (or circumstellar) H~I column density of $\\log {\\rm N(H~I)}=20.35$, and if we also assume that Ly$\\alpha$ fluxes along our line of sight are suppressed by a factor of 2.5 compared to the average fluxes seen by the H$_{2}$.  We speculate that perhaps this suppression factor of 2.5 is due to a near edge-on orientation of Mira~B's accretion disk relative to our line of sight --- either dust extinction from the disk or higher H~I wind densities near the disk could act to suppress the observed Ly$\\alpha$ fluxes for our line of sight. \\item[7.] We observe H$_{2}$ absorption features within the Ly$\\alpha$ line at the location of the transitions where the H$_{2}$ is being fluoresced. The strength of the absorption is roughly consistent with the amount of absorption predicted by an H$_{2}$ slab with $T=3600$~K and $\\log {\\rm N(H_{2})}=17.3$, providing further support for the accuracy of these measurements. \\item[8.] We propose that the most likely location for the H$_{2}$ that is being fluoresced is in the wind of Mira~A. The H$_{2}$ absorption visible within the Ly$\\alpha$ line, our finding that the H$_{2}$ most likely completely surrounds the star, and the lack of rotational broadening signatures in the H$_{2}$ line profiles argue against the H$_{2}$ being within the accretion disk. \\item[9.] One possibility is that the H$_{2}$ that we observe is within the interaction region of the winds of Mira~A and Mira~B.  Thus, we estimate the properties of Mira~B's wind from wind absorption observed within the Mg~II h \\& k lines.  Typical IUE Mg~II profiles suggest mass loss rates of $\\sim 1\\times 10^{-11}$ M$_{\\odot}$ yr$^{-1}$ and terminal speeds of $V_{\\infty}=400$ km~s$^{-1}$, while for our more recent STIS data we find $\\dot{M}=5\\times 10^{-13}$ M$_{\\odot}$ yr$^{-1}$ and a terminal velocity $V_{\\infty}=250$ km~s$^{-1}$.  The lower STIS mass loss rate is consistent with the dramatically lower UV line and continuum fluxes seen by STIS compared with IUE, corresponding to a lower accretion rate and naturally a lower mass loss rate. \\item[10.] Based on our estimates for Mira~B's mass loss rate from the Mg~II analysis, and based on previously published estimates for Mira~A's wind properties, we estimate that the ram pressures of the two winds will balance about 3.7~AU from Mira~B.  This is one possible location for the warm H$_{2}$ that we observe, but simple hydrodynamic shock calculations have difficulty reproducing the observed H$_{2}$ temperature and column density. \\item[11.] We use crude models to demonstrate that a photodissociation front has more success reproducing the properties of the H$_{2}$ emission that we observe than a wind interaction shock, since photodissociation fronts naturally tend to yield temperatures of $\\sim 3600$~K and warm H$_{2}$ columns of $\\log N(H_{2})\\sim 17.3$, consistent with observations.  It is possible, however, that the H$_{2}$ could be produced in a combination of a wind interaction shock and a photodissociation front. \\item[12.] One possible explanation for the order of magnitude decrease in Mira~B's accretion luminosity from the IUE era to that of our STIS observations is a decrease in Mira~A's wind density near Mira~B. However, the H$_{2}$ lines suggest a large H$_{2}$ dissociation rate of $1.0\\times 10^{-9}$ M$_{\\odot}$ yr$^{-1}$, emphasizing the need for a constant supply of H$_{2}$.  Since it is doubtful that a weaker Mira~A wind could supply this amount of H$_{2}$, we propose that accretion instabilities are a more likely cause of the drop in accretion luminosity. \\item[13.] The Ly$\\alpha$ flux observed from Mira~B in our STIS data is {\\em not} lower than the Ly$\\alpha$ flux observed in the IUE era, in contrast with the UV continuum and all other non-H$_{2}$ lines, which are all lower by over an order of magnitude.  We believe this is the primary reason why the H$_{2}$ lines fluoresced by Ly$\\alpha$ are far more prominent in the STIS data, while being undetectable in the IUE spectra.  It is uncertain why the Ly$\\alpha$ fluxes did not change with the other UV line and continuum fluxes, but we speculate that perhaps the stronger wind present in the IUE era somehow suppressed the Ly$\\alpha$ flux. \\end{description}"
    },
    "0208/astro-ph0208355_arXiv.txt": {
        "abstract": "Searching for transits provides a very promising technique for finding close-in extra-solar planets. Transiting planets present the advantage of allowing one to determine physical properties such as mass and radius unambiguously. The EXPLORE (EXtra-solar PLanet Occultation REsearch) project is a transit search project carried out using wide-field CCD imaging cameras on 4-m class telescopes, and 8--10m class telescopes for radial velocity verification of the photometric candidates. We describe some of the considerations that go into the design of the EXPLORE transit search to maximize the discovery rate and minimize contaminating objects that mimic transiting planets. We show that high precision photometry (2 to 10 millimag) and high time sampling (few minutes) are crucial for sifting out contaminating signatures, such as grazing binaries. We have an efficient data reduction pipeline  which allows us to completely reduce the data and search for transit candidates in less than one month after the imaging observations, allowing us to conduct same-semester radial velocity follow-up observations, reducing the phase uncertainty.  We have completed two searches using the 8k MOSAIC camera at the CTIO4m and the CFH12k camera at CFHT, with runs covering 11 and 16 nights, respectively. Using the 4400 images from the two fields, we obtained preliminary light curves for approximately 47,000 stars with better than $\\sim1$\\% photometric precision.  A number of light curves with flat-bottomed eclipses consistent with being produced by transiting planets has been discovered. Preliminary results from follow-up spectroscopic observations using  the VLT UVES spectrograph and the Keck HIRES spectrograph obtained for a number of the candidates are presented. Data from four of these can be interpreted consistently as possible planet candidates, although further data are still required for definitive confirmations. ",
        "introduction": "\\label{sect:intro}  %  The discovery of giant extra-solar planets, such as 51 Peg b \\cite{mayor95}, with orbital periods of a few days and orbital radius $<0.1$AU, was completely unexpected. Currently, 13 of these close-in extra-solar giant planets (CEGPs) are known, representing about 15\\% of the planets discovered by the radial velocity (RV) technique.\\footnote{ See, e.g., Extrasolar Planets Encyclopaedia, http://www.obspm.fr/encycl/catalog.html.} The existence of close-in giant planets shows that planetary systems can be radically different from our own, and sparked much theoretical work on planet formation and migration scenarios to explain the proximity of giant planets to the parent stars (e.g, Refs. \\citenum{lin96}, \\citenum{holman97}, \\citenum{murray98}, and \\citenum{rasio96}). This new class of planets also makes the method of finding planets via the transiting of their parent star very promising. In the situation of a Jupiter-sized planet transiting a solar-sized parent star, an eclipse with a flat-bottom light curve (due to the planet being completely superimposed on the parent star) of depth $\\sim$1\\% is expected, an easily measurable effect with modern CCD photometry. The probability that a given planet will show transits is inversely proportional to its orbital distance; for CEGPs, this is typically $\\sim10$\\%. Moreover, the typical periods of a few days for CEGPs also makes monitoring for transits relatively easy.  Transiting planets offer a number of advantages over those discovered by the RV technique alone. They are currently the only ones for which a radius can be measured. Furthermore, absolute masses of transiting planets can be measured using the RV technique without  the usual sin $i$ ambiguity. Transiting planets are also the most suitable for many kinds of follow-up studies, e.g., atmosphere transmission spectroscopy\\cite{seager00}\\cite{char02}, searches for moons and rings\\cite{brown01}, and others.  Over the past few years, a large number of transit searches for extra-solar planets have been started (e.g., Refs. \\citenum{vulcan}, \\citenum{stare}, \\citenum{ogle}). However, currently only one unambiguous transiting planet is known, HD 209458b, which was discovered originally by the RV technique and later  found to be a transiting planet\\cite{char00}\\cite{henry00}. In this paper we present preliminary results from the EXPLORE (EXtra-solar PLanet Occultation REsearch) project, which is the first 4m-class telescope transit search with 8-10m class spectroscopic radial velocity follow-up observations as an integral component of the search strategy. In Section 2 we outline some of the considerations that went into the design of the searches. Section 3 describes the two EXPLORE searches conducted so far. Preliminary results and possible transiting planet candidates are presented in Section 4. Finally, in Section 5 we provide a summary and briefly outline the future prospects for the EXPLORE project. ",
        "conclusions": "Planet transit surveys have the promise of providing the next breakthrough in extra-solar planet detection and characterization. Transit searches can potentially yield a large number of close-in planets, and such a sample will allow extra-solar planets to be characterized in much greater detail than is possible with non-transiting planets discovered by RV techniques. However, up to now, the transit search technique has not been successful in producing new detections of planets. We examined the various aspects in the design of transit searches, focusing on optimizing searches which  use telescopes with limited time availability (e.g., 4m-class national facilities) and the requirements for producing high-yield samples for spectroscopic follow-up RV observations using 8-10m class telescopes.  We showed that high-quality light curves with high precision photometry and frequent time sampling can provide sufficient information to winnow out most of the contaminating objects which may mimic transiting planets. We demonstrated that 8-10m class telescopes can provide useful mass limits for these possible planet candidates even for relatively faint stars of $I\\sim18$ mag. The advent of wide-field mosaic CCD cameras has made transit searches very attractive. We have conducted two searches using 4-m class telescopes as initial surveys.  These two searches, with $\\sim47,000$ preliminary light curves examined, have produced four possible planet candidates which still require additional spectroscopic or photometry data for definitive confirmation.  New imaging capabilities at various observatories will further improve the efficiency of transit searches in the near future. One such example is the MegaCam at CFHT which will become operational in 2003. With a 1 square degree field, small pixels, excellent image quality and a very short readout time, it will be able to gather data for transit searches at a much greater rate than currently possible. For transit searches for fields at the Galactic Plane, the MegaCam can routinely monitor as many as 120,000 stars simultaneously with photometric precision of better than 1\\%. Currently there are a large number of on-going transit surveys, and large samples of transiting planets are expected to be discovered in the near future. The EXPLORE project is continuing with additional searches in 2002 and 2003 using NOAO 4m telescopes under their Survey Program. As MegaCam becomes available, we also plan to propose to conduct additional searches using CFHT."
    },
    "0208/astro-ph0208480_arXiv.txt": {
        "abstract": "We present well-sampled light curves of Nereid for 1999 and 2000 which show symmetric shapes centered on opposition as is characteristic of a large opposition surge.  Surprisingly, the phase functions for 1999 and 2000 are significantly different from the well-measured phase function for the year 1998, with the 1999 and 2000 curves being more peaked at low phase angles and substantially brighter at high phase angles.  We know of neither precedent nor explanation for this mystery on Nereid. ",
        "introduction": "Nereid has an unusual orbit around Neptune ($e=0.75$, $P=360$ days, and $i=28 \\degr$) which suggests that it might be a captured Kuiper Belt Object \\citep{scs95,scs00}.  Nereid also has the most unusual known photometric history of all objects in the Solar System \\citep{scs88,scs00,sct01}.  From 1987 to 1991, several groups found that Nereid had fast photometric variations with amplitudes up to around one magnitude \\citep{scs88,scs00,wjt91,bul89}.  From 1991 to 1997, Nereid's amplitude was smaller at $\\sim 0.4$ mag \\citep{scs00}, with highly significant evidence that it showed fast variations \\citep{bgm97,brw98}.  In 1998, we observed Nereid on 52 nights and found that Nereid displayed no significant variability other than a very large opposition surge \\citep{sct01}.  This unique behavior of large-and-small variations that change from year to year could be caused by chaotic rotation \\citep{dob95}, much like Hyperion \\citep{kla89}.  This possibility is appealing because Nereid has the necessary ingredients of a $>1\\%$ out-of-round shape (due to its small size) and a highly eccentric orbit around a planet with a large J2 component in its gravity.  Unfortunately, Dobrovolskis has shown that Nereid's rotation will be chaotic only if its rotational period is longer than roughly two weeks, and this requirement is not consistent with the observed brightness changes on time scales of one day.  Thus, the cause of Nereid's variations is not known \\citep{sct01}.  The 1998 light curve \\citep[see Figure 1]{sct01} displayed no significant brightness changes other than those associated with the normal changing of the solar phase angle, $\\alpha$, and the opposition surge.  The phase function (see Figure 2) shows an opposition surge of 0.52 magnitudes over a range of 2$\\degr$ in phase and this is among the largest known in our Solar System. The shape of the phase function is definitely not linear, while a broken line fits well.  Given Nereid's long history of changes in its variability, it is prudent to continue photometric monitoring.  The Yale 1-m telescope on Cerro Tololo is operated in a queue mode by a resident operator, and hence is perfect for the long-term synoptic study of Nereid.  So again we have used this telescope to monitor Nereid, this time during its 1999 and 2000 oppositions.  We find that Nereid's phase function has significantly changed its shape from 1998 to 1999 and 2000. ",
        "conclusions": ""
    },
    "0208/astro-ph0208163_arXiv.txt": {
        "abstract": "We report on VLBI observations of supernova 1986J in the spiral galaxy NGC~891 at two new epochs, 1990 July and 1999 February, $t=7.4$ and 15.9 yr after the explosion, and on a comprehensive analysis of these and earlier observations from $t\\sim 4$ yr after the explosion date, which we estimate to be $1983.2\\pm 1.1$. The source is a shell or composite, and continues to show a complex morphology with large brightness modulations along the ridge and with protrusions.  The supernova is moderately to strongly decelerated.  The average outer radius expands as $t\\,^{0.71 \\pm 0.11}$, and the expansion velocity has slowed to 6000~\\kms\\ at $t=15.9$~yr from an extrapolated 20,000~\\kms\\ at $t=0.25$~yr.  The structure changes significantly with time, showing that the evolution is not self-similar.  The shell structure is best visible at the latest epoch, when the protrusions have diminished somewhat in prominence and a new, compact component has appeared. The radio spectrum shows a clear inversion above 10~GHz.  This might be related to a pulsar nebula becoming visible through the debris of the explosion.  The radio flux density between 1.5 and 23~GHz decreases strongly with time, with the flux density $\\propto t^{-2.94\\pm0.24}$ between $t \\sim 15$ to 19~yr.  This decrease is much more rapid than that found in earlier measurements up to $t \\sim 6$~yr. ",
        "introduction": "SN~1986J was discovered at a radio frequency of 1.4 GHz in the edge-on spiral galaxy NGC~891 south south-west of the galaxy's center by van Gorkom et~al.\\ (1986; see also Rupen et~al.\\ 1987) on 1986 August 21/22.  The distance to NGC~891 was determined to be $\\sim 10$~Mpc \\citep[see e.g.][]{Tonry+2001, Ferrarese+2000, Tully1988, KraanKorteweg1986, Aaronson1982}, and we will adopt a round value of 10~Mpc throughout this paper.  With a peak flux density at 5~GHz of 128 mJy \\citep[][WPS90 hereafter]{WeilerPS1990}, SN~1986J is one of the radio-brightest supernovae ever detected. Very-long-baseline interferometry (VLBI) observations were made soon after the discovery and revealed an elongated brightness distribution \\citep{BartelSR1989}. Shortly after, in 1988 September 29 (1988.7), further observations using a more sensitive VLBI array allowed an image to be made, the first one of any optical supernova \\citep[ B91 hereafter]{RupenBF1991, Bartel+1991}. The image showed a complex source, perhaps with a composite structure, and a marginal indication of a shell with a highly modulated brightness distribution along the ridge and with at least one protrusion directed to the south-east and another one directed to the north-west.  Optical observations, partly made even before the radio discovery, showed a faint object of $\\sim 18$ mag that decayed unusually slowly \\citep{Rupen+1987}. Prominent H$\\alpha$ lines in the spectrum led to the classification as a type~II supernova. The spectral lines, however, were surprisingly narrow, with a full-width at half-maximum (FWHM) of $\\lesssim 1000$~\\kms\\ \\citep{Rupen+1987, Leibundgut+1991}, at least an order of magnitude smaller than the expansion velocity expected for the shock front.  Interestingly, similarly narrow lines were found for SN~1988Z \\citep{StathakisS1991}, but in addition dim and much broader lines were also present at early times for that supernova.  Perhaps such broad lines also existed for SN~1986J, but had already become too dim at the time of discovery. A large Balmer decrement and a small extinction together with ``metallic'' lines were also found for SN~1986J, which led \\citet{Rupen+1987} to propose a very high electron density of $n_e > 10^9$~cm$^{-3}$ in the emitting region, along with regions of much lower density. \\citet{Chevalier1987} interpreted the slow decay of the optical emission as being due to energy input from a central pulsar and the narrow H$\\alpha$ emission-lines as originating in the central region of the supernova.  A different interpretation of the narrow H$\\alpha$ emission-lines is offered by \\citet{Chugai1993}, who does not relate them to an inner emission zone, but rather to shock-excited dense clouds of gas in the circumstellar material, which move much more slowly than the shock front of the supernova.  The explosion date of SN~1986J is not well known.  The earliest pre-discovery radio detection was on 1984 May 1 \\citep{vdHulstdBA1986}.  Considering this detection along with subsequent measurements, \\citet{Rupen+1987} concluded that SN~1986J probably exploded around 1982 to 1983.  \\citet{Chevalier1987} used a more extended set of flux density measurements and fit his widely-used circumstellar interaction or mini-shell model \\citep{Chevalier1982a, Chevalier1982b} to the data and obtained an explosion date of $1983.0\\pm 0.5$.  Using a complete set of radio flux density measurements at five frequencies up to 1988 December 28, WPS90 obtained an explosion date of $1982.7^{+0.8}_{-0.5}$.  The progenitor was believed to have been a red supergiant with a mass of $\\sim 20$ to 60~\\Msol\\ \\citep{Rupen+1987} or $\\sim 20$ to 30~\\Msol\\ (WPS90) and estimated to have rapidly lost mass into a clumpy circumstellar medium (CSM) with a mass-loss to wind-velocity ratio of ${\\dot M / w} \\sim 2.4\\times 10^{-4}$~\\Msol~yr$^{-1}$ per 10~\\kms\\ (WPS90; note that this value is somewhat dependent on assumptions made by those authors).  The canonical interpretation of the radio emission from supernovae is that it is synchrotron radiation produced in the region around the contact discontinuity where the supernova ejecta interact with the CSM\\@.  The evolution of the expanding radio shell provides an observational window on the structure of both the CSM and the supernova ejecta, and is therefore of considerable interest.  The interaction region is bounded on the outside by the forward shock that travels outward from the contact discontinuity into the CSM, and on the inside by the reverse shock that travels, in co-moving coordinates, back into the ejecta. In his mini-shell model of an expanding supernova, Chevalier approximated the mass density profiles of the CSM and of the ejecta by power laws in radius, $r$, with the density of the CSM being $\\propto r^{-s}$ and that of the ejecta being $\\propto r^{-n}$.  In this case, self-similar solutions can be found, and the supernova expands with $r \\propto t^m$, with the deceleration parameter $m$ given by $m= (n-3) / (n-s)$.  The ratio between the radius of the forward shock, or the outer radius of the radio shell, and the radius of the reverse shock, or the inner radius of the radio shell, defines the shell thickness.  In the case of self-similar expansion as given by the Chevalier model, this ratio remains constant, and is thought be 1.21 to 1.29 for $20 > n > 7$.  At early times after shock breakout, most of the radio radiation from the interaction region is free-free absorbed by the photo-ionized CSM, with the optical depth, $\\tau$, being a function of $\\dot M/w$ of the progenitor.  As the shock front expands, it sweeps away the obscuring layers of the CSM, decreasing the external absorption and causing the flux density to rise quickly.  After the absorption has become small and the radio shell optically thin, the flux density, $S_\\nu$, at frequency $\\nu$, decays as $\\propto t^\\beta$, where $\\beta$ is related to the radio spectral index, $\\alpha$ ($S_{\\nu}\\propto \\nu^{\\alpha}$), and to $m$ by $\\beta = \\alpha-3+3m$ \\citep{Chevalier1982b}.  In the case of SN~1986J, the flux densities rose more slowly than predicted by this model.  This was accounted for by assuming additional absorption within the radio shell itself.  With this extension to the model, the rise of the radio light curves could be well fit (Chevalier 1987; WPS90).  An unstable shock front or filamentation in the ejecta (WPS90) might well produce such mixed absorption.  This model and its extensions have generally been successful in describing the early evolution of radio supernovae.  However, caution is probably warranted in applying it to SN~1986J, which is a complex source (B91).  Furthermore, deviations from a self-similar evolution have been found for SN~1993J \\citep{Bartel+2002, Bartel+2000}, and could also be expected for SN~1986J\\@. Thus a hydrodynamic model \\citep[e.g.,][]{MioduszewskiDB2001} for the evolution of a supernova shell may be needed. High-resolution measurements of the evolution of the radio supernova morphology and the deceleration of the expansion are therefore of particular interest to study the dynamics of the interaction between the ejecta and the CSM, and to guide the development of such more elaborate models.  Furthermore, and perhaps most importantly, such measurements may reveal a young pulsar nebula surrounding the compact remnant of the explosion.  Here we report on new VLBI observations of SN~1986J at two further epochs, 1990 July 21 (1990.6) and 1999 February 22 (1999.1), and on a comprehensive analysis of these and earlier observations from 1987 February 23 onward. In \\S\\ref{obss} we describe our VLBI and VLA observations and data reduction.  We then give our VLBI results. In \\S\\ref{imagess} we present a high-resolution image from our latest epoch of observations, and give a comprehensive analysis of the images from three epochs from 1988.7 onward. In \\S\\ref{lightcurves} we give flux density measurements from 1998 to 2002, and compare them with the radio light curve deduced from earlier measurements.  We also derive the radio spectrum of SN~1986J, which exhibits a high-frequency inversion. In \\S\\ref{astroms} we give our astrometric results, which will likely be of importance for future detailed studies of the dynamical evolution of SN~1986J\\@. In \\S\\ref{sizes} we determine the size of SN~1986J at different epochs, and in \\S\\ref{decels} the deceleration of the expansion. In \\S\\ref{bfields} we infer the evolution of the magnetic field, and finally, in \\S\\ref{discuss} we discuss our results and in \\S\\ref{concs} give our conclusions. ",
        "conclusions": "}   \\noindent Here we give a summary of our main conclusions.  \\begin{trivlist}  \\item{1.} The sequence of images of SN~1986J from 5.5 to 15.9 yr after the explosion is only the second such sequence, after that of SN~1993J, where the expansion and evolution is so clearly visible.  \\item{2.} The sequence shows a complex source with a shell or composite morphology and protrusions that distort the perimeter of the supernova.  \\item{3.} The structure of the supernova changes with time. The protrusions diminish somewhat in prominence, the shell structure becomes more visible, and a compact source with a flux density of $\\sim 1$~mJy emerges half way between the perimeter and geometric center.  \\item{4.}  The spectrum shows a significant inversion above 10~GHz, which was not present before 1989.  If the spectrum is composed of two power laws, the component with an inverted spectrum has a spectral index of $\\alpha = +1.4^{+0.6}_{-0.4}$ below 23~GHz.  The normal component of the spectrum has $\\alpha = -0.55^{+0.09}_{-0.16}$, which is consistent with the value measured in 1989. Between 1998 to 2002, the spectrum remains relatively unchanged.  \\item{5.}  The inverted component of the spectrum or the image component C1 may be related to a pulsar nebula, but the current evidence is not conclusive.  \\item{6.} The explosion date is estimated to be $1983.2 \\pm 1.1$, derived from both our expansion measurements and radio light curve modeling of others.  \\item{7.} The expansion is found to be moderately to strongly decelerated, with the average outer radius being $\\propto t^{0.71\\pm0.11}$ between $t = 0$ and 15.9~yr.  The expansion velocity in 1999.1 was 6000~\\kms, only about one third of the extrapolated velocity after three months of $20,000$~\\kms.  \\item{8.} With the assumptions given, the average magnetic field is 70~mG at $t=5.5$~yr and declines $\\propto t^{-1.2}$ between $t = 5.5$ and 19.2~yr.  \\item{9.} The flux density decreases $\\propto t^\\beta$ with $\\beta = -2.94 \\pm 0.24$ from $t = 15.9$ to 19.2~yr after the explosion.  The rate of flux density decay increased substantially since $t = 5.5$~yr, when $\\beta$ was $-1.18^{+0.02}_{-0.04}$.  \\item{10.} The changing structure, the emergence of an inverted part of the spectrum, and the change in the rate of flux density decay all clearly point out that the evolution of SN~1986J is not self-similar. Its evolution cannot be adequately described by self-similar solutions.  \\end{trivlist}"
    },
    "0208/astro-ph0208449_arXiv.txt": {
        "abstract": "This is the unofficial proceeding of my invited review at the Ringberg Castle Workshop on ``The Chemical Evolution of Dwarf Galaxies'', July 28th - August 2nd, 2002, and presents an update of a similar review I gave in 1999.  \\\\  \\noindent The current status of carbon stars in the Local Group and beyond, is discussed. Although many carbon stars and late M-stars have been identified in Local Group galaxies, a coherent understanding in terms of the chemical evolution- and star formation history of a galaxy is still largely lacking. Although a few major new surveys have been carried out over the last three years, the observational data is still incomplete in many respects: 1) for some of the larger galaxies only a small fraction in area has been surveyed so far, 2) surveys have been conducted using different techniques, and some of the older surveys are incomplete in bolometric magnitude, 3) only for some galaxies is there information about the late M-star population, or it is sometimes unpublished even when the data is available, 4) not all galaxies in the Local Group have been surveyed, 5) especially for some of the older work insufficient data is available to determine bolometric magnitudes.  I have correlated carbon star positions with the 2MASS 2nd incremental data release to obtain $JHK$, and bolometric magnitudes, to remedy this situation in some cases.  From the existing observations one can derive the following: the formation of carbon stars is both a function of metallicity and star-formation history. In galaxies with a similar star formation history, there will be relatively more carbon stars formed in the system with the lower metallicity. On the other hand, the scarcity of AGB type carbon stars in some galaxies with the lowest metallicity indicates that these systems have had a low, if any, star-formation over the last few Gyrs. ",
        "introduction": "Carbon stars are tracers of the intermediate age population in galaxies. Either they are currently undergoing third dredge-up on the (Thermal-Pulsing) Asymptotic Giant Branch (TP-AGB) -- the cool and luminous N-type carbon stars --, or have been enriched with carbon-rich material in a binary system when the present-day white dwarf was on the AGB (the carbon dwarfs and CH-stars. The R-stars may be the result of a coalescing binary [McClure 1997]).  Since their spectral signature is very different from oxygen-rich and S-type stars, it is relatively easy to identify carbon stars even at large distances.  In Sect.~2 the main technique to identify carbon stars is briefly discussed, and in Sect.~3 the various surveys for carbon stars in external galaxies are summarised. That section is in fact, an update of a review I gave at IAU Symposium 191 (Groenewegen 1999; hereafter G99), and here I will only refer to new results published since then, or older literature when I use it to obtain new results. Please refer to G99, and the similar review by Azzopardi (1999) for the full story. The results and some recent theoretical predictions are discussed in Sect.~4. ",
        "conclusions": "In principle, the overall carbon star LF and C/M ratio contains information about the star-formation rate history from, say, 10 Gyr ago (the low-luminosity C-stars in binaries) to a few-hundred Myr ago (the high luminosity tail of the LF). Its a challenge to theoretical models to use these constraints together with other data to derive the chemical evolution and star formation history of these galaxies. The models of Mouhcine \\& Lan\\c{c}on represent a first successful step in this direction. \\\\"
    },
    "0208/astro-ph0208119_arXiv.txt": {
        "abstract": "We have carried out a comprehensive multiwavelength study of Bright-Rimmed Globule TC2 in the Trifid Nebula, using the IRAM~30m telescope, the VLA centimeter array and the Infrared Space Observatory (ISO). TC2 is one of the very few globules to exhibit signs of active ongoing star formation while being photoevaporated by the Ly-c flux of the exciting star of the nebula ($\\sim 10^{10}\\cmmd\\smu$). The globule consists of a cold dense core of mass $27\\msol$ surrounded by a lower density envelope of molecular gas. The impinging Ly-c photons induce the propagation of an ionization front into the globule. The evaporation of the ionized gas forms a thin layer of density $n_e= 1-2\\times 10^3\\cmmt$ around the globule, which could be mapped with the VLA. The globule is illuminated mainly on its rear side, by a FUV field of intensity $\\rm G_0\\simeq 1000$. It creates a Photon-Dominated Region (PDR) below the surface, which was mapped and characterized with the ISOCAM Circular Variable Filter and the Short Wavelength Spectrometer (SWS) onboard the Infrared Space Observatory. The physical conditions derived from the analysis of the far-infrared lines $\\Oi$~$\\rm 63\\mu m$, $\\rm 145\\mu m$ and $\\Cp$~$\\rm 158\\mu m$, and the continuum emission are in good agreement with some recent PDR models. The emission of the PAHs band at 6.2, 7.7, 8.6 and $\\rm 11.3\\mu m$ is detected over the whole globule. The relative intensity variations observed across the globule, in the PDR and the photoionized envelope, are consistent with the changes in the ionization fraction. In the head of TC2, we find a second  kinematic component which is the signature of the radiatively-driven collapse undergone by the globule. This component indicates that the PDR propagates at low velocity inside the body of TC2. The molecular emission suggests that the star formation process was probably initiated a few $10^5\\yr$ ago, in the large burst which led to the formation of the nebula. The globule has already evaporated half the mass of its envelope. However, the ionization timescale of the globule is long enough ($\\sim 2\\Myr$) to let the newly born object(s) reach smoothly the ultimate stages of protostellar evolution. The impact of photoionization on the star formation process appears limited. ",
        "introduction": "It is well established that the bright-rimmed globules found in \\Hp\\ regions are often sites of star formation. Reipurth (1983) first showed that these objects do form stars and subsequent work based on IRAS data by Sugitani et al. (1991) confirmed that they are indeed active ``stellar factories'' which produce intermediate-mass (Herbig AeBe) stars. These condensations are local clumps which emerge from the expanding nebula or form from the fragmentation of the dense molecular layer surrounding the ionized gas. The various theoretical works (Bertoldi, 1989; Bertold \\& McKee, 1990; Lefloch \\& Lazareff, 1994) and and the numerous observational studies on bright-rimmed globules (see e.g. Cernicharo et al. 1992; Lefloch \\& Lazareff, 1995) have enabled to draw the following evolutionary picture, summarized in Fig.~1.  As a globule of neutral gas is exposed to the ionizing field of the exciting star(s) of the nebula (region I in Fig.~1), an ionization front (IF) forms at the surface of the condensation. For standard ionization conditions, the pressure of the surface ionized gas is much higher than in the nebular gas and in the molecular globule. As a consequence, the photoionized gas expands into the H{\\small II} region, inducing the formation of a photoionized envelope around the globule. It is the ionization front and the photoionized envelope which are detected in the optical as a bright rim (region II). The incident FUV field drives the formation of a Photon-Dominated Region (hereafter PDR, region III) while a shock front, driven by the surface overpressure, propagates towards the dense molecular core (region IV). Observational evidence of this mechanism, also called Radiatively-Driven Implosion (or RDI), has been reported in a few objects by Cernicharo et al. (1992) and Lefloch \\& Lazareff (1995). Progressively, a dense core forms behind the surface, and the globule adjusts its internal structure to balance the pressure of the ionized gas, while the bulk of its mass is photoevaporated. Eventually, the globule reaches a quasi steady-state, the ``cometary phase'', in which the shock front has disappeared. The globule now consists of a small dense ``head'' prolonged by a long tail of diffuse gas.  It has long been suggested that the shock front inside the globule could trigger the star formation inside the bright-rimmed globules (see e.g. Reipurth 1983; Lefloch et al. 1997) . These objects appear therefore as ideal laboratories to test the scenarios of star formation triggered by an external compression wave. Most of the studies led until now were focused onto the molecular core of bright-rimmed globules (see e.g. Lefloch et al. 1997). Therefore, the physical conditions reigning in the PDR and in the shocked molecular gas are not well characterized and the impact of photoionization on the gravitational collapse is therefore difficult to evaluate. Moreover, all the bright-rimmed condensations studied until now are found in relatively old \\Hp\\ regions, with ages of a few Myr. Because the condensations are usually found at rather large distances from the ionizing stars, they experience a reduced UV field and it is difficult to discriminate between a star formation induced by Radiatively-Driven Implosion and a spontaneously evolving globule which has already started to form stars by the time it is hit by the ionization front.   This is why we have started a systematic multiwavelength study of a young H{\\small II} region~: the Trifid nebula. It appears as a small dusty nebula of $10\\arcmin$ diameter at an heliocentric distance of $1.68\\kpc$ (Lynds et al. 1985), with a dynamical age of $0.3-0.4\\Myr$ . The nebula is excited by the O star HD~164492A. In a preliminary work (Cernicharo et al. 1998, CL98), we reported on the mapping of the thermal dust emission of the nebula at millimeter wavelengths. This mapping revealed the presence of several protostellar cores (dubbed TC1 to TC4) in the shell of dense molecular gas surrounding the \\Hp\\ region. It was not clear however if the birth of the protostars was triggered or not. For this reason, we undertook a more detailed analysis of the protostellar cores. In a subsequent paper (Lefloch \\& Cernicharo, 2000), we reported on the continuum and molecular line emission around the two most massive protostellar cores~: TC3-TC4. Their masses are high, between 60 and $90\\msol$. They harbour a Class I source and one of the few high-mass Class~0 candidates known until now. Comparison of their properties with the models of Elmegreen \\& Lada (1977) and Whitworth et al. (1994)  allowed to conclude that the formation of TC4 had probably been triggered in the fragmentation of the dense shell surrounding the ionized gas. The molecular properties of TC3 and TC4 are similar to those of the protostellar cores discovered in Orion, though at an earlier, ``pre-Orion'', evolutionary stage.   We report here on the TC2, protostellar core, which is associated with a bright-rimmed globule on the Southern border of the Trifid. TC2 appears to be in a more advanced stage of photoionization than TC3-4~: unlike TC3-4 which are still embedded cores, TC2 has already emerged from the diffuse molecular gas layers  and it exhibits the optical bright rim and the cometary shape typical of photoionized globules. As discussed in this work, TC2 is exposed to a rather strong ionizing field, which drives a shock into the condensation, in agreement with the evolutionary scheme presented above. Like TC3 and TC4, TC2 displays signs of protostellar activity. CL98 reported the presence of the Herbig-Haro jet HH399 coming out of the head of the globule and propagating into the ionized nebula. It is the best example of globules undergoing at the same time strong photoevaporation and active star formation. It offers a good opportunity to study the quantify the relation both phenomenons hold to each other, and better constrain the role that Radiatively-Driven Implosion could play in the star formation process. In this purpose, we have led a detailed study of the structure and the physical conditions of the globule (density, temperature, velocity). Because TC2 lies almost in the plane of the sky, it provides also a good opportunity to study the structure of a typical low-density photon-dominated region, and to confront its properties against the existing models. This work provides the first comprehensive study of the whole gas structure of a bright-rimmed globule, from the ionized surface layers to the cold dense molecular core.   The paper is organized as follows. We first derive the structure of the globule~: The HII region (Sect.~3); the bright rim and the photo-evaporated envelope (Sect.~4); the PDR (Sect.~5); the dust continuum emission (Sect.~6) and the molecular core (Sect.~7). We then discuss the observational evidences of Radiatively-Driven Implosion in TC2 and the implications on the past history of the globule (Sect.~8). In the following section, we first study the star forming conditions in TC2 and attempt to characterize the protostellar source and the outflowing material, before studying the impact of photo-ionization both on the protostar and the evolution of the globule (Sect.~9). The conclusions are presented in Sect.~10. ",
        "conclusions": "We have carried out a multiwavelength study of bright-rimmed globule TC2 in the Trifid nebula. The globule lies almost in the plane of the sky, immersed in  the low-density gas of the $\\Hp$ region ($\\rm n_e\\sim 50-100\\cmmt$). It is illuminated by the O star HD~164492A, mainly on the rear side. The globule consists of a very dense core of cold gas and dust (T= 22~K, $\\rm n(\\htwo)= 3\\times 10^5\\cmmt$), of small dimensions surrounded by a lower-density envelope ($\\rm n(\\htwo)= 3\\times 10^4\\cmmt$). Its mass ($\\simeq 63\\msol$) is typical of bright-rimmed globules, equally distributed between the core and the envelope (27 and $36\\msol$ respectively). The ionization conditions at the surface of the globule were determined from VLA and ISO (LWS and SWS) observations. The impinging ionizing flux is $1.4\\times 10^{10}\\cmmd\\smu$; it creates an ionization front and a photoevaporated envelope of density $1-2\\times 10^3\\cmmt$ at the surface of the globule.  The PDR is traced by the emission of the PAHs bands at 6.2, 7.7, 8.6 and $\\rm 11.3\\mu m$. The observed variations of the relative intensities can be accounted for by the change in the excitation conditions. We find that the intensity of the $\\rm 11.3\\mu m$ band drops outside of the PDR in the photoionized layers, as a consequence of the ionization of the PAHs. The $\\rm 7.7\\mu m$ band is still detected in the photoionized envelope of the globule, though at a weaker level. Despite the  high visual extinction of the globule, the $\\rm 7.7\\mu m$ PAH band, excited in the PDR on the rear side is detected in the body of the globule thanks to a minimum in the absorption of the ices and the dust in this wavelength range.  Millimeter line observations reveal the kinematical signature of the PDR which precedes the ionization front as a shock moving into the globule. The relative projected velocity is weak,  $\\approx 0.7\\kms$. Comparison with models of photoionized globules (LL94) indicates that TC2 has been  exposed to the ionizing radiation for $3\\times 10^5\\yr$, almost as soon as the exciting star of the nebula turned on. The structure of the photon-dominated layer has been derived from the ISO far-infrared line and continuum observations. The agreement between all the data and the model of K99 is good. The intensity of the radiation field at the surface is $\\rm G_0\\simeq 1000$. The PDR has a typical column density of $2\\times 10^{21}\\cmmd$ and a molecular hydrogen density $\\simeq 10^4\\cmmt$. The FUV field heats the dust to a temperature $\\simeq 46\\K$ whereas the average gas temperature is found close to $300\\K$. The emission of the $\\Sip$~$\\rm 34.8\\mu m$ line in the PDR can be accounted for by assuming a silicon gas phase abundance of $6\\times 10^{-6}$, i.e. $\\simeq 17\\%$ the solar value.  The globule is currently undergoing star formation. Our observations show that the globule is forming a low- or intermediate-mass star of luminosity $\\leq 500 \\lsol$. The protostar powers a Herbig-Haro jet, which appears fully ionized outside of the globule. No molecular counterpart (outflow) to the optical jet has been found. In particular, no SiO emission was found. This implies that the source is already in an intermediate stage between Class 0 and Class I, or even a full Class I member. The evolutionary age of the source is therefore typically a few $10^5\\yr$, which suggests that star formation in TC2 probably started in the large burst which accompanied the birth of HD~164492A. As a result of the photoevaporation of the surface layers, the globule has evacuated about half the mass in its envelope until now but we have not found any evidence that the birth process itself has been perturbed. On the contrary, the photoevaporation rate is low enough to leave ample time for protostars to reach safely the ultimate stages of star formation."
    },
    "0208/astro-ph0208433_arXiv.txt": {
        "abstract": "The SuperCOSMOS Sky Surveys provide a complete coverage of the Southern sky in three passbands (photographic $B_J, R$ and $I$) and at different epochs \\citep{hambly01a,hambly01b,hambly01c}. These data are the basis for a new high proper motion survey which aims at finding extremely red nearby dwarf stars and brown dwarfs. One of the first candidates, which is relatively bright ($I=16$) but very red ($R-I=2.8, B_J-R=3.6$), was detected in the equatorial zone by its large proper motion of 0.56~arcsec/yr. Spectroscopic follow-up observations with the 2.2m telescope of the Calar Alto Observatory classified this object as L2 dwarf very similar to the first free-floating L dwarf Kelu~1 also discovered in a proper motion survey by \\citet{ruiz97}. If we assume SSSPM~J0829$-$1309 to have the same luminosity as Kelu~1, we get a distance estimate for the new L dwarf of about 12~pc since it is about one magnitude brighter than Kelu~1 in the SSS $I$ and $R$ bands. This makes SSSPM~J0829$-$1309 one of the nearest objects of its class, well suited for detailed investigations. We present a brief overview of all known nearby ($d<20$~pc) southern L dwarfs and give first proper motion values for DENIS-P~J0255$-$47 and SDSSp~J1326$-$00 and an improved proper motion for LHS~102B. ",
        "introduction": "Only five years ago, the first field L dwarfs were discovered in a proper motion survey using ESO Schmidt plates \\citep{ruiz97} and in the DEep Near-Infrared Survey (DENIS) \\citep{delfosse97}. In the following years, the new spectral type ''L'' was defined using the rapidly growing number of discoveries from the 2-Micron All-Sky Survey (2MASS) \\citep{kirkpatrick99} and from DENIS \\citep{martin99}. Most of the additional L dwarfs found since then are from 2MASS \\citep{kirkpatrick00,reid00,gizis00} and from the Sloan Digital Sky Survey (SDSS) \\citep{fan00,schneider02,hawley02}.  The colour-based search for L dwarfs with DENIS, 2MASS and SDSS data has increased the sample size to more than 200, where the majority of the objects lie at distances larger than 25~pc, the limit of the Nearby Star Catalogue \\citep{gliese91}.  Nearby L dwarfs, suitable for detailed follow-up observations, have also been found (see the list of \\citealt{kirkpatrick00} and references therein). However, many nearby L stars are still waiting to be discovered according to the predictions on their number densities from observations \\citep{reid99,gizis00} and theory \\citep{chabrier02}. The number density of early L dwarfs (L0.0-L4.5) alone is expected to be around 0.002 per cubic parsec \\citep{gizis00}, corresponding to about 70 objects within 20~pc. The completion of the census of L dwarfs in the Solar neighbourhood is important for investigations of the star formation process, the stellar and substellar luminosity function and the initial mass function.  As demonstrated with the recent discovery of three nearby L dwarfs in the Southern sky \\citep{lodieu02}, the SuperCOSMOS Sky Surveys (SSS) \\citep{hambly01a,hambly01b,hambly01c} provide an effective tool to find nearby extremely red and low-luminosity objects via their proper motion and photographic colour. The SSS cover the whole Southern sky with scans of UK Schmidt Telescope (UKST) plates in three different passbands ($B_J, R, I$) and at different epochs with additional scans of ESO Schmidt plates. Since practically all very nearby stars ($d<10$~pc) in the Nearby Star Catalogue \\citep{gliese91} have large proper motions ($\\mu>0.18$~arcsec/yr, i.e. the limit of the New Luyten Two Tenths (NLTT) catalogue, \\citealt{luyten7980}), we can expect large proper motions for all nearby L dwarfs as well. The reason for the non-detection of L dwarfs among the NLTT stars is the lower limiting magnitude of the first Palomar Observatory Sky Survey (POSS-1) used by Luyten, compared to later Schmidt telescope surveys (POSS-2, UKST, ESO). UKST and ESO Schmidt plates are much deeper with limiting magnitudes of $B_J\\sim23, R\\sim22, I\\sim19$ \\citep{hambly01a}. It was thus possible to find the first free-floating L dwarf in the Solar neighbourhood, Kelu~1 with $R=19.7, I=17.1$ (SSS magnitudes) as high proper motion object on Schmidt plates \\citep{ruiz97}. Trigonometric parallax measurements \\citep{dahn02} have shown this benchmark L2 dwarf to lie at a distance of 19~pc.  In this paper, we announce the detection of another nearby L dwarf using the proper motion search technique. ",
        "conclusions": "We have discovered one of the nearest field L dwarfs, SSSPM~J0829$-$1309, which we classify as L2 dwarf similar to Kelu~1. Based on the comparison of SSS magnitudes of these two objects we estimate SSSPM~J0829$-$1309 to lie at a distance of about 12~pc. The newly discovered L dwarf is a potential target for both Northern and Southern parallax programmes."
    },
    "0208/astro-ph0208269_arXiv.txt": {
        "abstract": "{  We report the analysis of optical spectra of the extreme helium star LSS~3184 (BX~Cir) to determine its effective temperature and gravity throughout its pulsation cycle. The spectra were also used to measure its chemical abundances.  We report rest gravity, $\\log g = 3.38 \\pm 0.02$, and a chemical abundance mixture consistent with those reported earlier in a study using an optical spectrum with lower spectral resolution and a lower signal to noise ratio. Our analysis decreases the upper limit for the H abundance to ${\\rm H < 6.0}$ (mass fraction $< 7.1 \\times 10^{-7} $).  Our gravity corresponds to stellar mass $M = 0.47 \\pm 0.03 M_\\odot$.  We find that the effective $\\log g$ varies through the pulsation cycle with an amplitude of 0.28~dex.  The effective gravity is smaller than the rest gravity except when the star is very near its minimum radius.  The change in effective gravity is primarily caused by acceleration of the stellar surface.  Based on the optical spectra, we find the temperature varies with an amplitude of 3450~K. We find a time averaged mean temperature, $23390 \\pm 90$~K, consistent with that found in the earlier optical spectrum study.  The mean temperature is 1750~K hotter than that found using combined ultraviolet spectra and V and R photometry and the variation amplitude is larger.  This discrepancy is similar to that found for the extreme helium star \\object{V652~Her}. ",
        "introduction": "Extreme Helium stars (EHes) are a class of low-gravity, hot, evolved stars with very large helium abundances ($\\ga 99$ per cent) and weak or non-existent hydrogen lines.  They appear to be rapidly evolving to become white dwarfs. The short time the stars spend as EHes explains why few EHes are known, despite their brightness ($L \\ga 900 L_\\odot$).  The evolutionary history of EHes is still uncertain. The more popular proposals have been that they are formed by the merger of two white dwarfs (Webbink \\cite{w84}, Iben \\& Tutukov \\cite{it84}) or that they are the result of a `late thermal pulse' when the helium shell near the surface of a white dwarf ignites, causing the star to expand (Iben et al. \\cite{i83}). Saio \\& Jeffery (\\cite{sj02}) have recently shown that the white dwarf merger model is the most likely explanation.  Determining the history of EHes will provide clues to the evolutionary history of possibly related stars (RCrB stars, He-rich subdwarfs, carbon stars, etc.).  As the chemical abundance mixture in EHes appears to be the end result of a combination of CNO and triple-$\\alpha$ processing, their study may provide information about physical processes occurring in a large fraction of stars.  Some EHes pulsate. The pulsations are driven through the $\\kappa$ mechanism, with iron group (Z-bump) elements providing the needed opacity (Saio \\cite{s95}). Pulsation provides additional methods to study the physical properties of stars. Radial velocities, ultraviolet spectra, and optical photometry have been used to determine the radius, gravity, temperature, mass, and absolute magnitude of two pulsating EHes: LSS~3184 and V652~Her (Lynas-Gray et~al. \\cite{lg84}, Kilkenny et al. \\cite{k99}, Woolf \\& Jeffery \\cite{wj00}, Jeffery et~al. \\cite{jwp01}). Mass and chemical composition are the parameters which best constrain evolutionary models for EHes. In previous studies of LSS~3184 the relatively large uncertainty for the gravity produced most of the uncertainty for the mass determination.  In a recent study of pulsation models for EHes, Monta\\~n\\'ez Rodr\\'{\\i}guez \\& Jeffery (\\cite{mj02}) showed that the pulsation period and radial velocity curve of LSS~3184 require a mass between 0.38 and 0.5 $M_\\odot$ if the temperature is assumed to be between 22\\,400 and 24\\,000~K. However, the models with those parameters also require a luminosity smaller than half of that previously measured.  In this paper we report the determination of chemical abundances and time resolved temperature and effective gravity of LSS~3184 using spectral analysis of the optical spectra used to find radial velocities in Woolf \\& Jeffery (\\cite{wj00}). ",
        "conclusions": "We have used time resolved spectra with high signal to noise ratios and high spectral resolution to determine the temperature, gravity, and chemical composition of LSS~3184.  Our results are consistent with those of previous analyses based on optical spectra and photometry. We have reduced the uncertainties on some of these quantities.  We have reduced the upper limit on the hydrogen abundance by a factor of 52 to $\\rm H < 6.0$.  Our gravity and mass estimates are 12 per cent larger than previous estimates.  The time averaged mean temperature estimate we obtain with our spectral analysis is 1750~K hotter than that obtained by fitting UV and visible flux levels.  A similar temperature discrepancy was found for V652~Her.  The problem is probably the result of incomplete line opacity in our model stellar atmospheres.  The mass and temperature we find for LSS~3184 is consistent with those expected based on the pulsation models of Monta\\~n\\'ez Rodr\\'{\\i}guez \\& Jeffery (\\cite{mj02}) and with those found using optical photometry. The discrepancy between the luminosity expected from the models and that calculated based on UV observations remains a problem.  The chemical composition of LSS~3184 corresponds well to that expected from a white dwarf merger. However, its mass is smaller than the merger models currently allow."
    },
    "0208/astro-ph0208095.txt": {
        "abstract": "{ We consider the gravitational magnification of light for binary systems containing two compact objects: white dwarfs, a white dwarf and a neutron star or a white dwarf and a black hole. Light curves of the flares of the white dwarf caused by this effect were built in analytical approximations and by means of numerical calculations. We estimate the probability of the detection of these events in our Galaxy for different types of binaries and show that gravitational lensing provides a tool for detecting such systems. We propose to use the facilities of the Sloan Digital Sky Survey (SDSS) to search for these flares. It is possible to detect several dozen compact object pairs in such a programme over 5 years. This programme is apparently the best way to detect stellar mass black holes with open event horizons. ",
        "introduction": "One of the most important manifestations of gravitational lensing is the visible variability of the astrophysical objects whose emission is affected by the gravitation field of the lens. This effect becomes observable over a reasonable time period, when the velocities of relative motions of the observer, the lens and the source are great enough, i.e. their mutual location changes rapidly. Many cases of this kind of variability have been examined: from light variation of distant quasars to outburst of stars caused by the influence of planets (see, for instance, Zakharov 1997). The brightness variation of the binary system components caused by gravitational lensing was been first considered by Ingel (1972, 1974) and Maeder (1973). They noted that repetition and small characteristic times of the effect make it one of the most accessible to detection and detailed study. It has also been shown (Maeder 1973) that the large amplitudes of brightness variations ($0\\fm 1-0\\fm 5$) can be expected in binary systems consisting of compact objects -- white dwarfs, neutron stars, black holes. In fact, only in these cases does the Einstein-Hvol'son radius turn out to be smaller than the lens star radius, and the image of the radiation source is practically not overlapped by it. The latter circumstance makes the gravitational lensing influence in binary systems an effective means for the detection and detailed investigation of compact objects with the aid of photometric methods alone. Their permeability is by $2^{\\rm m}-3^{\\rm m}$ higher (as compared to spectroscopy), and, therefore, the number of stars amenable to study is an order of magnitude larger. Here the light curve analysis determines all the parameters of the binary system in full analogy to this task for eclipsing binary systems (with allowance made for its singularity) (see, for example, Goncharsky et al. 1985).  A complete set of data for compact object binaries and their statistical analysis gives a possibility to test binary evolution theories and to investigate the last stages of single star evolution as well. We note that only 10 double white dwarfs were detected during the last 10 years (Maxted~et~al. 2000) and about 40 pairs of a white dwarf and a neutron star -- during 25 years (Thorsett \\& Chakrabarty 1999). We further note that by means of gravitational lensing it is possible to detect compact object binaries at the same rate (see Section 5). Moreover, this may help to discover binaries containing neutron stars with a low magnetic field (not pulsars). Study of these systems in combination with binary star evolution theories gives a possibility to test the detailed cooling models for white dwarfs and neutron stars and the equation of state of the latter (Nelemans~et~al. 2001; Yakovlev~et~al. 1999).  Studying compact object pairs is very important to solve a number of astrophysical problems. We present below several such examples.  The \"Standard candle\" of modern cosmology, type Ia supernovae, apparently arises from merging of double CO white dwarfs (Webbink 1984; Iben \\& Tutukov 1984), which have not been detected at this time.  Close double compact objects have to contribute a significant part of the gravitational wave signal at low frequencies. Thus, white dwarf pairs may be a source of the unresolved noise (Evans~et~al. 1987; Grischuk~et~al. 2001). Statistical properties of the closest double compact objects in principle determine parameters of the gravitational wave signal.  Binary systems consisting of a white dwarf and a neutron star (a pulsar) are unique laboratories for high-precision tests of general relativity. To date, post-Keplerian general relativistic parameters have been measured in four such pairs (Thorsett \\& Chakrabarty 1999). This problem may be solved very  easily for binaries with gravitational self-lensing since they are observed nearly edge-on.  It seems that the detection of black holes forming pairs with white dwarfs might be an extraordinarily important result of the search for gravitational lensing in compact object binaries. Despite the opinion of most researchers, in a certain sense black holes have not been discovered so far. Only observational data on the behaviour of matter close to the event horizon showing its presence may testify that a black hole is identified (Damour 2000). There is indirect evidence of the presence of black holes in X-ray binaries and galactic cores based on the \"mass-size\" relation expected for them (Cherepashchuk 2001). The horizon neighbourhood is seen neither in X-ray binaries nor in the cores of active galaxies because they are screened out by the accreted gas (the accretion rates are very high). This means that only black holes accreting usual interstellar plasma at low rates of $10^{-14}-10^{-15} M_{\\odot}$/year can be recognized as objects without a surface and with an event horizon (Shvartsman 1971). The black hole companions of white dwarfs in the binaries detected by means of gravitational lensing could be the best objects for horizon study and tests of general relativity in the strong field limit (Damour 2000). It is very easy to measure the mass and size of such a black hole due to binary edge-on orientation and investigate the radiation of gas near the horizon. This matter is accreted from interstellar medium only because its transfer from the white dwarf is absent.  The gravitational lensing effects in binary systems consisting of compact components were studied previously in several papers (Maeder 1973; Gould 1995; Qin~et~al. 1997). However, the authors did not go further than estimation of the probability of detecting the effect of orientation of the binary system (often underestimating it by a factor of 2-3), which turned out to be very low. Their general conclusion is that the effect cannot be actually recorded. Nevertheless, the data accumulated by now on the evolution of binary systems and their parameters make it possible to define with higher accuracy the probability of detection of the brightness enhancement in binary systems, to find the expected number of such flares and to propose in the final analysis the strategy for their search based on present-day facilities. Our paper is devoted to performing these tasks.  We examine in Section 2 the distinguishing features of light magnification in the systems consisting of two white dwarfs, a white dwarf and a neutron star, and also a white dwarf and a black hole. In Section 3 we derive probabilities of recording the effect for its different amplitudes in all three cases. In Section 4 the numbers of systems of different types which may be detected by means of gravitational lensing are estimated. In Section 5 we discuss the possibilities of the quest for brightness magnification in such systems with the aid of the telescope and equipment used in the survey SDSS (York et al. 2000). ",
        "conclusions": "To determine the number of objects that may be detected using a certain telescope one has to know not only the overall number but also their distribution in the Galaxy, their luminosity function and the distribution of absorbing matter as well.  We used the star distribution in the Galaxy of the following form: \\begin{equation} \\rho=\\rho_0e^{-\\frac{R}{H}}e^{-\\frac{|z|}{h}} , \\end{equation} where $R$ and $z$ are the point cylindrical coordinates in the Galaxy, $H$ and $h$ are radial and vertical distance scales, respectively, for which we used the following values: $H=8$~kpc and $h=250$~pc (see Dehnen \\& Binney 1998; Nelemans et al. 2001). Adopting the Sun's coordinates $R_{\\odot}=8.5$~kpc and $z_{\\odot}=30$~pc we have: $$ \\rho_0=\\rho_{loc}e^{\\frac{R_{\\odot}}{H}}e^{\\frac{z_{\\odot}}{h}}. $$  The overall number of objects in the Galaxy is expressed by the integral $$ N=\\int\\limits_{-\\infty}^{\\infty}dz\\int\\limits_0^{\\infty} 2\\pi R \\rho_0e^ {-\\frac{R}{H}}e^{-\\frac{|z|}{h}} dR, $$ which yields \\begin{equation} \\rho_{loc}=\\frac{Ne^{-\\frac{R_{\\odot}}{H}}e^{\\frac{|z_{\\odot}|}{h}}} {4\\pi H^2h} =\\frac{N}{6.56\\cdot 10^{11}} {\\rm pc^{-3}} \\end{equation} and \\begin{equation} \\rho(R,z)=\\frac{N}{6.56\\cdot 10^{11}}e^{-\\frac{R-R_{\\odot}}{H}}e^ {-\\frac{|z|-z_{\\odot}}{h}} {\\rm pc^{-3}}. \\end{equation} We used distributions for WD-NS and WD-BH in the same form with the only difference in normalization coming from their fewer overall number in the Milky Way. The local densities corresponding to the overall numbers given in Table 6 are presented in Table 7.  Note that the part of WD-WD pairs among the complete number of WDs is (7-25)~\\% (Iben et al. 1997; Fryer et al. 1999) and the WD density in the Solar neighbourhood is $(4-20)\\cdot 10^{-3}$ pc$^{-3}$ (see Nelemans et al. 2001 and references therein). This gives us a range of local density of $3\\cdot 10^{-4} - 5\\cdot 10^{-3}$~pc$^{-3}$. It is very close to our estimate shown in Table 7 and the most optimistic values of density may be $2-2.5$ times more than those.  Unfortunately, there is no data to compare with for binaries other than WD-WD. We suggest only that our estimation range may be $2-3$ times narrower. Therefore, the real number of objects may be a few times higher. \\begin{table}[h] \\centering \\caption[]{The local densities of objects derived from Table 6 data and distirbution (27) (in pc$^{-3}$).} %vspace*{0.3cm} \\begin{tabular}{cccc} \\hline \\noalign{\\smallskip} $k$     &WD--WD     &WD--NS&WD--BH\\\\ &            &       &  \\\\ \\hline \\noalign{\\smallskip} 2.35 &$2.0\\cdot 10^{-3}$&$4.1\\cdot 10^{-5}$ &$1.2\\cdot 10^{-5}$\\\\ 2.5  &$1.3\\cdot 10^{-3}$&$2.4\\cdot 10^{-5}$ &$6.1\\cdot 10^{-6}$\\\\ 2.7  &$8.4\\cdot 10^{-4}$&$9.1\\cdot 10^{-6}$ &$1.8\\cdot 10^{-6}$\\\\ \\noalign{\\smallskip} \\hline \\end{tabular} \\end{table}  The insterstellar absorption is proportional to the interstellar medium density which is distributed in the Galaxy by the same~ (27) law with $h=80$~pc (Dehnen \\& Binney 1998). The normalization was done according to the local average value of absorption $A_{g^{\\prime}}=1\\fm 16 {\\rm kpc^{-1}}$ in the $g^{\\prime}$ band of the SDSS equipment (Schlegel et al. 1998).  The WD luminosity function is well known and is explained fairly well in the frame of their cooling theory. We used the luminosity function of Oswalt et al. (1996). It is readily seen from this that more than half of the WDs are brighter than $13\\fm 5$ and only less than 20 percent are fainter than $15^{\\rm m}$, where the function reaches its maximum. Thus, the magnitudes of the majority of WDs do not suffer much from the bolometrical corrections which is a few tenth for the stars of these spectral classes (see, e.g. Allen 1973).  In other binary systems, the situation is more complicated. It is only clear that the optical emission from both a NS and a BH can be associated with interstellar gas accretion. In the former case it is likely to be negligibly small because of the high orbital velocity of the NS ($> 100$ km/s) (Shapiro \\& Teukolsky 1983). At the same time the orbital velocity of a BH in a pair with a WD is about 50 km/s, and the luminosity of accretion plasma can reach $10^{30}$ erg/s by moderate-optimistic estimates (Shvartsman 1971; Beskin \\& Karpov 2002) and will be by an order of magnitude less by a pessimistic one (Ipser \\& Price 1982). Apparently, its contribution to the total luminosity of the system is insufficient to correct the depth of the space of detection. However, this level of optical emission makes possible searching for its fast variations with a microsecond time resolution for investigation of accretion processes and strong gravitational fields near the horizon of events (Shvartsman 1971; Beskin et al. 1997; Beskin \\& Karpov 2002).  Now we have everything we need to calculate the on-sky density of objects in a given direction ($l,b$): \\begin{equation} \\begin{array}{ll} \\sigma(l,b)&=\\frac{d^2N}{\\cos{b} dl db}\\\\ %        &\\\\ &=\\int r^2 \\rho(R,z) I(M)dr\\; , \\end{array} \\end{equation} where $I(M)$ is a cumulative luminosity function from Oswald et al. (1996), $M=m_{lim}-5\\lg(r/10)-A(r,l,b$), $m_{lim}$ is an equipment limiting apparent stellar magnitude, $A(r, l, b)$ is a total extinction in the $(l,b)$ direction up to the distance $r$, and $$ \\begin{array}{l} z=z_{\\odot}+r\\sin{b},\\\\ R=\\sqrt{R^2_{\\odot}+r^2\\cos^2{l}-2R_{\\odot}r\\cos{b}\\cos{l}} \\end{array} $$ are the point cylindrical coordinates expressed in terms of its galactic coordinates $(l,b)$ and distance $r$.  Unfortunately, this integration cannot be done analytically and so we will turn at once to a particular version of the equipment being discussed. In our view, an ideal tool for carrying out the proposed programme is currently the  2.5-metre telescope at APO (New Mexico). It has been used to accomplish one of the most promising projects -- the Sloan Digital Sky Survey (SDSS). The telescope has a field of $3^{\\circ}$, in which 30 CCD chips of $2048\\times 2048$ pixels are mounted (York et al. 2000). Sky scanning at a speed of its movement along a strip $2\\fdg 5$ wide is performed within the framework of the project. During an exposure of 55 s, the limiting stellar magnitude in the $g^{\\prime}$ filter (it is close to the $B$ filter of the Johnson system) is $23\\fm 2$ at a signal-to-noise ratio of $\\sim 5$ (Oke \\& Gunn 1983; York et al. 2000). We have constructed a relationship between amplitude of flares recorded at a $5\\sigma$ level and stellar magnitude of a quiet object in the $g^{\\prime}$ band (Fig.~8). In so doing, we used the parameters corresponding to the  SDSS project: the seeing -- 0.8 arcsec, the quantum efficiency of the CCD matrix -- 80~\\%.  By comparing the data of Fig.~8 with the mean parameters of flares (Table~5) and their light curves for different binary systems (Fig.~5), it is clear that with a particular exposure of time $t_e\\sim 10-30$ s, the flares with $\\Delta m\\ge 0\\fm 2$ of any pairs that we have examined can be recorded at a significance level of $5\\cdot 10^{-3} - 10^{-5}$ (duration of any flare is a few $t_e$). In fact, the level of the flare detection will be better due to registration in the SDSS of the field in several colour bands ($u^{\\prime}, g^{\\prime}, r^{\\prime}, z^{\\prime}$) and the limiting $r^{\\prime}$ stellar magnitude is also $23\\fm$ Moreover, the use and comparison of the data in the $g^{\\prime} $ and $r^{\\prime}$ filters give the possibility of easy separation of cosmic ray traces.  \\begin{figure}[h] \\centering \\centering\\includegraphics[width=8.5cm,%height=7.5cm ]{MS2132f8.eps} \\caption{Magnitudes of detectable flares in the SDSS versus object brightness} \\end{figure}  On the basis of (30) we have built a map of the expected on-sky densities of WD-WD pairs up to $m_{lim}=23^{\\rm m}$ which is presented in Fig. 9. The value of the local density corresponding to the Salpeter mass-spectrum index of $2.35$ is used. One can see from the map that the maximal density regions have galactic latitudes of $5-10$ degrees and a question might arise if it is possible to observe in such star-rich regions of the sky. However, even in these latitudes the average distance between stars of up to $23^{\\rm m}$ is about 3-4 arcseconds (see e.g., Zombeck, 1982, p.34) and thus, they are resolved with the SDSS seeing and this crowding therefore  cannot have much negative impact on observations with this equipment.  \\begin{figure}[h] \\centering \\centering\\includegraphics[width=8.5cm]{MS2132f9.eps} \\caption{Simulated on-sky distribution of WD-WD pairs, $m_{lim}=23^{\\rm m}$. Maximum density value is 490 pairs per square degree.} \\end{figure}  To derive the expected number of events of the discussed type  $N_{exp}$ one multiplies the probability of detecting it from a randomly chosen pair $F(t)$ for $\\Delta m=0\\fm 2$ (see (15), (20)-(22$^{\\prime}$) and Fig.7) by the number $D(\\Omega)$ of such pairs observed during a certain observational programme that covered the celestial sphere part $\\Omega$: \\begin{equation} N_{exp}=F(t)\\times D(\\Omega) , \\end{equation} where \\begin{equation} D(\\Omega)=\\int\\limits_{\\Omega} \\sigma(l,b) \\cos{b} dl db\\; . \\end{equation}  It is clear, however, that when applying to reality one deals with a telescope of limited field of view and the observational time provided for a programme is limited as well, so increasing the total exposure of a field to increase $F(t)$ leads to decreasing the number of such fields in the programme. It is also evident that one has to monitor at first the best, i.e. maximal density $\\sigma(l,b)$ fields first. Fig. 10 represents the number of pairs of each type $D(\\Omega)$, numerically calculated from (32), in $\\Omega$ fields with a size of $2\\fdg 5\\times 2\\fdg 5$ (field of view of SDSS) after sorting them in decending order of pair density.  \\begin{figure}[h] \\centering \\centering\\includegraphics[width=8.5cm,%height=7.5cm ]{MS2132f10.eps} \\caption{Expected number of objects $D(\\Omega)$ in $\\Omega$ the best (with the highest density of pairs) $2\\fdg 5\\times 2\\fdg 5$ fields.} \\end{figure}  Now let the programme last for 5 years. Considering that the matter in question is detection of rather faint objects, observations have to be made on dark nights only at moon phases within $\\pm 5^d$ from the new moon, which gives approximately $36 \\%$ of night time. Assuming the average observational night of $9^h$ the 5-year resource of time then will make $n_{obs}=657$ nights or $T_{obs}=5913$ hours. Thus, as every field is observed for $t$ hours or $n$ nights (and $\\Omega = T_{obs}/t$, $n_{obs}/n$) the expected number of detected pairs is \\begin{equation} N_{exp}=D\\left(\\frac{T_{obs}}{t}\\right) F(t), \\end{equation} or $$ N_{exp}=D\\left(\\frac{n_{obs}}{n}\\right) F(n). \\hspace*{4.6cm}(33^{\\prime}) $$  These quantities derived from Figs.~7 and 10 are presented in Fig.~11. One can see from this figure that the optimal total exposure per field for WD-WD pairs is $6-7$ nights, and it is about 1 night when searching for WD-NS or WD-BH systems.  \\begin{figure}[t] \\centering \\centering\\includegraphics[width=8.5cm,%height=7.5cm ]{MS2132f11.eps} \\caption{Expected detection numbers in the 5-year programme as a function of total exposure per field.} \\end{figure}  The expected numbers of detections with optimal strategies along with the optimal total exposures per field are given at Table 8. \\begin{table}[h] \\centering \\caption[]{The number of pairs of compact objects detectable over 5 years and the optimal total exposures per field} %\\vspace*{0.3cm} \\begin{tabular}{lccc} \\hline \\noalign{\\smallskip} $k$     &WD--WD     &WD--NS&WD--BH\\\\ &            &       &  \\\\ \\hline \\noalign{\\smallskip} 2.35 &22&9&16\\\\ 2.5  &15&5&8\\\\ 2.7  &9&2 &3 \\\\ \\hline Optimal & & & \\\\ exposure time&6-7 nights &1-2 nights &6-9 hours \\\\ per $2\\fdg 5\\times 2\\fdg 5$ field&&&\\\\ \\noalign{\\smallskip} \\hline \\end{tabular} \\end{table}  A new project entitled \"The Dark Matter Telescope\" ({\\tt http://www.dmtelescope.org/index.htm}) being developed involved 8.4 (effectively 6.9) metre telescope with a field of view slightly wider than that of the SDSS. The use of it in a manner similar to that described above increases the number of detectable objects by a factor of 2.2. Other wide field telescopes of 4 metres in diameter being built now, LAMOST ({\\tt http://www.greenwich-observatory.co.uk/lamost.\\\\html}) and VISTA ({\\tt http://www.vista.ac.uk}), could detect 1.5 times more binaries than the SDSS.  Thus, for searching for the flares caused by gravitation lensing in binary systems with compact companions, the facilities of the SDSS can be used, having reduced the time of an individual exposure to 10-20 s. It is obvious that after recording a flare in a certain object, another telescope of similar class must be involved in observations. This object should be monitored to prove the effect and for detailed investigation of the detected system.  Thus, within the frame of the proposed programme, it will be possible to find several dozen binary systems comprised of compact objects. It will be recalled that in numerous surveys of microlensing, the number of reliably detected events amounts to about 350 (Alcock 1999), only in two cases the matter in question being of massive lenses (black holes?) (Bennett et al. 2001). When applying the technique suggested, there is a possibility of detecting with absolute assurance more than 10 black holes with open event horizons. An investigation of these objects with high time resolution within the MANIA experiment (Beskin et al. 1997) may permit at last the detection of observational evidence of extreme gravitational fields."
    },
    "0208/astro-ph0208094.txt": {
        "abstract": "and % In this paper we present a new method for obtaining the optical wavelength-dependent reddening function of planetary nebulae, using the nebular and stellar continuum. The data used was a spectrum of NGC~6302 obtained using the Double Beam Spectrograph on the 2.3m telescope at Siding Springs Observatory over three nights. This resulted in a spectrum covering a wavelength range $3300-8600$~\\AA\\, with a large dynamical range and a mean signal to noise of $>10^2$~\\AA$^{-1}$ in the nebular continuum. With such a high S/N the continuum can be accurately compared with a theoretical model nebular plus stellar continuum. The nebular electron temperature and density used in the model are determined using ratios of prominent emission lines. The reddening function can then be obtained from the ratio of the theoretical and the observed continuum. In the case of NGC 6302, it is known that much of the reddening arises from dust within or around the nebula, so that any differences between the measured reddening law and the `standard' interstellar reddening law will reflect differences in the nebular grain size distribution or composition. We find that for NGC 6302, the visible to IR extinction  law is indistinguishable from `standard' interstellar reddening, but that the UV extinction curve is much steeper than normal, suggesting that more small dust grains had been ejected into the nebula by the PN central star. We have detected the continuum from the central star and determined its Zanstra Temperature to be of order 150,000K. Finally, using the extinction law that we have determined, we present a complete de--reddened line list of nearly 600 emission lines, and report on the detection of the He(2-10) and He(2-8) Raman Features at $\\lambda4331$~\\AA\\ and $\\lambda4852$~\\AA, and the detection of Raman-Scattered \\OVI features at 6830 and 7087 \\AA. We believe this to be the first detection of this process in a PN. ",
        "introduction": "Because interstellar dust grains are very small, typically less than a micron in diameter, their absorption and scattering properties %at optical wavelengths are not only composition--dependent but also wavelength--dependent. Blue and UV light is usually preferentially scattered compared to that of longer wavelengths. Dust grains can not only absorb and scatter light from objects, they can also re-emit in the thermal infra-red, polarize light through grain alignment mechanisms or be accelerated, heated and photoelectrically charged by the electromagnetic radiation which impinges upon them.  All of these processes are known to occur in the planetary nebula (PN) environment. In particular, during the Asymptotic Giant Branch (AGB) phase of evolution, mass--loss releases material into the circum--stellar environment which has undergone partial nuclear processing in the central star. Since this environment is fairly cool, dust may be formed by direct condensation out of the gaseous phase whenever the kinetic temperature of the gas falls below a critical value which allows solids to form. In this case, we have a gas which is slowly cooling from higher temperatures and in which the pressure and supersaturation are high enough to allow both nucleation and grain growth. However, it is unlikely that there exists a state of thermodynamic equilibrium in the dust-forming gas, and shock heating and cooling are often both important. Therefore, a complex and detailed time-dependent description of the chemical reactions, usually referred to as a kinetic model, is needed to describe this situation.  Because of the physics of the condensation process, and the interaction between the grains formed in the flow and the radiation field of the star, there is a complex relationship between the nature of the grains, their size distribution and the terminal velocity of the dusty outflow. Kozasa \\& Sogawa (1997) showed that the grain size increases as the mass--loss rate increases, since the size of the grain produced by condensation depends upon the gas density in the wind where a strong supersaturation exists in the gaseous phase and upon the period during which the condensation timescale is much shorter than the dynamical expansion timescale. On the other hand, radiation pressure acting upon the grains accelerates the stellar mass-loss flow (thereby arresting the condensation process). This has been seen observationally by Loup {\\it et al.} (1993) and explained theoretically by Habing {\\it et al.} (1994). The expansion velocities of the carbon rich objects are larger than those of the oxygen rich AGB stars, and radiation pressure induced expansion of the atmosphere may limit the size of the typical carbon-bearing grain to $\\sim 50$~\\AA, similar to that which is needed to explain the 2175~\\AA\\ bump in the interstellar extinction curve. During the PN phase of evolution, we expect the grain size distribution to be further modified by radiative destruction processes (photoevaporation and coulomb destruction by excessive photoelectric charging) and by mechanical processes (grain coagulation and shattering).  Taking all of these considerations into account, it is clear that we should expect that the dust formed in the gas ejected during the AGB, and later observed in PN phase, would be quite unlike like that seen in the interstellar medium as a whole. It is therefore of great interest to either observe this dust directly through IR observations, or else through the extinction produced by it in the optical and UV.  As far as direct observations are concerned, enormous progress has recently been made using the ISO satellite to obtain spectroscopy of the far--IR emission features characteristic of different grain materials (Waters {\\it et al.} 1996). The bright southern PN NGC 6302 is an ideal object for such studies, as it is known to have within it a dense circumstellar torus containing the bulk of the dust mass (Lester \\& Dinerstein, 1984), and within this, a dense ring of ionised gas, inclined at about 45 degrees to the plane of the sky (Rodriguez {\\it et al.} 1985). Recently, Kemper {\\it et al.} 2002 have reported the detection of features in the far--IR spectrum of this object which may be ascribed to the silicates amorphous olivine, forsterite, clino-enstatite, and diopside. In addition features due to water ice and metallic iron are seen. Remarkably, the carbonates calcite and dolomite were also detected.  At optical wavelengths, the lack of strong spectral features renders such exquisite mineralogy impossible. However, because dust grain dimensions are often comparable to or smaller than the wavelength of light, the dust extinction curve can in principle be used as a powerful constraint on the grain size distribution in the nebula.  For PNe we usually characterise the reddening by a single `reddening constant', $c$, and then assume that the absorption through the optical wavelength region can be fit by a `standard' Whitford (1958) reddening law. This curve, f($\\lambda$), can then be used to deredden the observed emission line fluxes. The relationship between the corrected flux, F$_c$, and that observed, F$_o$ is:   \\begin{equation} \\text{F}_c=\\text{F}_o\\times10^{cf(\\lambda)} \\end{equation}   The reddening constant is usually determined from a comparison of the ratio of the intensities of the Balmer lines, since this `Balmer decrement' is only slightly dependent upon the temperature and density of the nebula, and the theoretical values are well--determined. Alternatively, we can compare the radio continuum flux density and the H$\\beta$ flux. The radio emission is basically free from interstellar reddening and the ratio between the radio continuum flux and the H$\\beta$ flux is determined only by the electron temperature and the relative helium abundance. A third technique is to measure the ratio of two emission lines which share a common upper energy level, such as H$\\beta$ and Br$\\gamma$ (Ashley, 1990).  All of these methods have their problems. In the first case, the reddening is determined at only a few discrete wavelengths, and over a restricted wavelength range. In the other two cases, we may be seeing regions of ionized gas in the radio or at IR wavelengths which are entirely dust--obscured in the optical, and therefore we can neither correctly evaluate the effective total obscuration nor the differential extinction at different optical wavelengths in the nebular gas.  The motivation behind the work described in this paper, is to obtain an intrinsic reddening function which does not depend on the Whitford curve, which is continuous in its wavelength coverage, and which can be used to place constraints on the grain size distribution in a planetary nebula.  To do this, we have obtained very high signal to noise observations of NGC 6302 covering the wavelength range $3300--8600$~\\AA, allowing observations of both Paschen and Balmer lines, and of both the Balmer and the Paschen discontinuities of Hydrogen.  We have then compared the observed continuum spectral energy distribution to a theoretical (nebular $+$ stellar) spectral energy template to derive the reddening function. As far as we are aware, this represents the first practical application of this novel technique in the literature. ",
        "conclusions": "A high signal-to-noise ratio, high resolution spectrum of the bright planetary nebula NGC 6302 was obtained with a wavelength range covering the visible spectrum and its continuum has been used to provide the first detection of the central star of NGC 6302, and to determine the reddening function of the dust in the nebula.  As far as the authors know this is the first time the continuum of a planetary nebula has been measured to such accuracy over such a wide range, and the first time the intrinsic reddening curve of a nebula been determined from the form of the nebular continuum. Certainly, the continuum distribution of planetary nebulae have been used before, but mainly to measure the electron temperature of the nebulae (Liu \\& Danziger 1993).  The UV steepening of the reddening curve of NGC 6302 is taken to mean that there is a higher abundance of small dust grains in the nebula than is found in the interstellar medium. However, with only one example, it is not known whether this property is common to all planetary nebulae or just to those of Type I composition."
    },
    "0208/astro-ph0208004_arXiv.txt": {
        "abstract": "Using a reduced proper motion discriminator, I obtain a sample of 4588 subdwarfs from the Revised NLTT Catalog of Salim \\& Gould. The ample statistics and low contamination permit much more precise determinations of halo parameters than has previously been possible.  The stellar halo is not moving with respect to the Local Standard of Rest (LSR) in either the vertical or radial direction, up to uncertainties of $2\\,\\kms$.  This indicates that either the LSR is on a circular orbit or the Sun happens to lie very close to an extremum of the LSR's elliptical orbit.  Similarly, tentative detections of vertical proper motion of Sgr A* relative to the LSR are either incorrect or they reflect real physical motion of the central black hole relative to the Galactic potential.  The correlation coefficients of the halo velocity ellipsoid, which would reflect any possible misalignment between its principal axes and the cardinal directions of the Galaxy, vanish to within 2\\%.  The halo subdwarf luminosity function peaks at $M_V\\sim 10.5$ with a full width half maximum of about 2.5 mag. ",
        "introduction": "}  Samples of nearby halo stars can be analyzed to find the bulk properties of the population: their velocity, spatial, and metallicity distributions, as well as their luminosity function.  The principal difficulty is obtaining a sample that is large enough to draw statistically significant conclusions while still not being contaminated with disk and thick disk stars, which locally outnumber halo stars by a factor $\\sim 10^3$.  The most secure method to construct such a sample would be to obtain parallaxes, proper motions, and radial velocities for a larger, unbiased sample of stars and then select halo stars based on their space motions. The GAIA satellite would be able to do this, but even under the most optimistic projections, its data will not be available for well over a decade.  In the absence of such ideal datasets, most nearby halo samples have been culled from catalogs of high proper-motion stars (although there are a few notable exceptions to this rule).  For example, \\citet{dahn} obtained trig parallaxes for about 100 proper-motion selected stars, thereby determining their transverse velocities, distances, and absolute magnitudes.  By rigorous selection on transverse velocity ($v_\\perp>260\\,\\kms$), they obtained a sample that was virtually free of disk and thick-disk contamination. The distances and absolute magnitudes then allowed them to measure the luminosity function (LF).  Of course, to do so they had to correct for the halo stars that were eliminated from their sample (along with the unwanted disk stars) by their stringent velocity criterion, and this in turn required a model of the halo velocity distribution.  The best such model up to that date was constructed by \\citet{crb}, who used maximum likelihood to decompose two proper-motion selected samples into disk, thick disk, and halo components making use of both photometric and proper-motion data.  They thereby identified different populations within the data, even though individual stars could not generally be unambiguously associated with a specific population. In particular, \\citet{crb} showed that the likelihood fit was significantly improved by allowing for a third ``intermediate'' or thick disk population rather than just two.  The kinematics of the halo when so fit were more extreme than in the 2-component fit earlier obtained by \\citet{bc} because thick-disk contamination was drastically reduced.   RR Lyrae stars are halo tracers selected on variability rather than proper motion.  Estimates of the RR Lyrae absolute magnitude from statistical parallax automatically yield the velocity ellipsoid. While this technique has been applied for almost a century, only in the last decade or so has it been realized that the RR Lyrae samples are actually mixtures of thick-disk and halo stars.  Since statistical parallax uses both radial velocities (whose spectra also yield metallicity information) and proper motions, and since it derives distances for all stars, full kinematic as well as metallicity information is generally available. In a series of papers, \\citet{layden94,layden95,layden97} both systematized pre-existing data and obtained substantial new data, thereby laying the basis for a new statistical parallax solution that clearly separated the thick-disk and halo populations using a combination of kinematic and metallicity criteria \\citep{layden}.  \\citet{pg1,pg2} and \\citet{gp} (collectively PG$^3$) introduced new mathematical methods and on this basis conducted a thorough overhaul of the \\citet{layden} sample, recalibrating much of the old photographic photometry, incorporating more modern extinctions, identifying suspicious astrometry, and developing a new method to incorporate non-RR-Lyrae radial velocities into the analysis.  Of particular note in the present context, PG$^3$ were the first to measure five of the nine components of the halo velocity ellipsoid: all previous analyses had measured the three diagonal components of the velocity dispersion tensor and the component of bulk motion in the tangential direction (the asymmetric drift), but had assumed that the off-diagonal components as well as the bulk motion in the radial and vertical directions were zero.  While the PG$^3$ measurements all turned out to be consistent with zero (in the frame of the Local Standard of Rest -- LSR), the error bars were tantalizingly close to being able to probe some interesting scientific questions.   The bulk motion of the halo in the radial and vertical directions is more likely to coincide with the rest frame of the Galaxy than is the motion of the LSR.  The LSR could well be on an elliptical orbit, in which case it would be moving towards or away from the Galacitic center unless the Sun happened to lie at an extremum of this orbit.  Indeed, \\citet{bs} claimed that the LSR is moving outward at $14\\,\\kms$ based on radial-velocity measurements of gas in the outer galaxy (assumed to be on circular orbits). On the other hand, \\cite{ms} concluded that the LSR was moving inward at $6.6\\pm 1.7\\,\\kms$ based on radial velocities of carbon stars in the outer Galaxy.  Similarly, if the Milky Way disk is warped, then one would expect the LSR to be moving either up or down relative to the Galactic rest frame, unless the Sun happened to be at an extremum of the warp.  \\citet{backer} found that Sgr A* is moving down at $17\\pm 6\\,\\kms$ relative to the LSR.  If the supermassive black hole associated with Sgr A* is assumed to be at rest with respect to the Galaxy, then this apparent motion would actually be a reflex of the warped motion of the LSR.  On the other hand, \\citet{reid} find that Sgr A* is moving in the opposite direction (although with much larger errors) at $15\\pm 11\\,\\kms$.  New more precisements measurements are expected soon (M.\\ Reid, private communication 2001).  For a roughly isotropic ensemble of $N_s$ stars, the bulk motion $U_i$ can be measured with a precision, \\begin{equation} \\sigma(U_i) \\sim \\sqrt{3\\,c_{ii}\\over n_d N_s} \\label{eqn:sigmaui} \\end{equation} where $c_{ii}$ is the dispersion in the $i$th directin and $n_d$ is the number of components of the velocity measured for each star. For the PG$^3$ sample, $N_s\\sim 170$ and $n_d=3$, while $c_{11}\\sim (160\\,\\kms)^2$ and $c_{33}\\sim (90\\,\\kms)^2$.  Hence $\\sigma(U_1) \\sim 13\\,\\kms$ and $\\sigma(U_3)\\sim 8\\,\\kms$.  The PG$^3$ measurement errors were therefore not quite small enough to probe these interesting questions.  Of course, measurement of a difference between the bulk-halo and LSR velocities would not be unambiguous evidence of LSR motion \\citep{oresme}. For example, the angular momentum vector of material infalling onto the Milky Way could have radically changed between the time of the formation of the halo and disk.  The former therefore could in principle be rotating in a basically polar orbit (albeit with low Mach number) relative to the latter.  Hence any sort of relative motion would be intriguing evidence of a non-simple Galaxy whose exact origins would have to be sorted out making use of other data and arguments.  Similarly, the off-diagonal elements of the velocity dispersion tensor (normalized to the diagonal elements) $\\tilde c_{ij}$ could potentially provide evidence of asymmetries of the Galaxy that would reflect on its origins.  The errors in these quantities are $\\sim (n_d N_s/3)^{-1/2}$, or about 8\\% for the RR Lyrae sample.  To the best of my knowledge, no one has investigated what might cause these quantities to differ from zero, so I do not know whether their consistency with zero at the 8\\% level challenges or confirms any theory.  Nevertheless, it seems interesting to try to probe the off-diagonal elements at higher precision.  The status of the LF and local density of the stellar halo are also somewhat controversial.  \\citet{dahn} find that the LF peaks at around $M_V\\sim 12$, in qualitative agreement with the shape of the LF seen in undisturbed globular clusters \\citep{piotto}.  However, \\citet{bc} and \\citet{gfb} find a roughly flat LF over the interval $9\\la M_V \\la 13$.  While \\citet{dahn} and \\citet{bc} both studied local stars drawn from proper-motion catalogs, \\citet{gfb} adopted a radically different approach: they located stars in {\\it Hubble Space Telescope} images that were too faint at their observed color to be in the disk and so were assigned absolute magnitudes and distances based on a halo color-magnitude relation.  These stars were generally quite distant ($\\ga 3\\,$kpc) and therefore perhaps not directly comparable to the local samples.  \\citet{sz} had earlier suggested that the stellar halo actually has two components, one roughly spheroidal and one highly flattened.  (The highly flattened component is not to be confused with the thick disk: it is not rotating significantly.)\\ \\ In their model, the two components have roughly equal densities at the solar circle.  Such a model predicts that the halo density should be roughly twice as great in the solar neighborhood as it is at a similar Galactocentric radius but a few kpc above the Galactic plane.  Indeed, \\citet{gfb} found a halo density that was lower than the \\citet{dahn} measurements by just this fraction. Recently \\citet{siegel} have argued on the basis of a sample of 70,000 stars along multiple pencil beams that even a 2-component halo model is inadequate to explain their star counts.  The newly released Revised NLTT Catalog \\citep{bright,faint} allows one to obtain a very large and very clean sample of halo stars.  The reduced proper-motion (RPM) diagram using the newly obtained $V-J$ colors clearly separates main-sequence stars, subdwarfs, and white dwarfs into different tracks \\citep{rpm} in sharp contrast to RPM diagram constructed from the original NLTT \\citep{luy}. Although the first release of this catalog covers only 44\\% of the sky, it contains more than 5000 local halo stars, well over an order of magnitude more than have ever been cleanly distinguished from disk stars on a star-by-star basis.  This sample therefore opens the way to a much more detailed study of the local halo population than has previously been possible. In its present form, the sample does have some limitations. Since most of its stars lack radial velocities and parallaxes, it is not possible to establish the absolute distances or the amplitude of the velocity ellipsoid based on the Revised NLTT Catalog alone. Nevertheless, the amplitude of the velocity ellipsoid is already known with a precision of about 10\\% from previous studies, and by incorporating this external information one can obtain much more precise measurements of the five components of the ellipsoid that are currently poorly measured: $U_1$, $U_3$ and $\\tilde c_{ij}$.  Once the velocity scale is set, the mean distances to the stars are also determined, which permits one to measure the LF.  The large number of stars in the sample therefore offers the hope of probing the bottom of the subdwarf sequence which, because of its dimness, is poorly represented in magnitude limited samples. Finally, the catalog contains a large number of stars from the Galactic plane to about $z\\sim \\bar V_\\perp/\\mu_{\\rm lim}\\sim 350\\,$pc above the plane, where $\\bar V_\\perp\\sim 300\\,\\kms$ is the typical transverse speed seen toward the Galactic poles and $\\mu_{\\rm lim}= 180\\,\\masyr$ is the proper-motion limit of the catalog. While this distance is short compared to the several kpc's hypothesized as the height of the flattened halo component, the large number of stars in the sample may yield a statistically significant statement about the presence of a density gradient on these larger scales.  Maximum-likelihood (ML) analysis is absolutely critical for extracting halo parameters from this catalog.  For example, since the mean tangential velocity of stars seen toward the Galactic poles is $\\sim 200\\,\\kms$, one might naively expect that the stars selected toward the poles would have, on average, this velocity.  However, given the fact that the sample is proper-motion limited, for all but the dimmest absolute magnitudes the number of stars seen with velocities that are $1\\,\\sigma$ higher than average ($300\\,\\kms$) is $(300/100)^3=27$ times higher than the number with velocities that are $1\\,\\sigma$ lower ($100\\,\\kms$).  This severe selection bias does not {\\it directly} affect any other parameters.  However, it couples through the highly uneven (but perfectly known) sky coverage from 2MASS, to {\\it indirectly} affect essentially all other parameters.  These effects can only be removed by comparing the predictions of models with the observations, as ML does automatically.  Hence, I begin in \\S~\\ref{sec:maxlike} by giving a careful summary of the ML modeling procedure.  In \\S~\\ref{sec:results}, I present my results and comment on various aspects of these whose interpretation requires caution.  Finally, in \\S~\\ref{sec:discuss}, I compare my results to previous work and briefly discuss the implications of this comparison.  I reserve to the Appendix a somewhat technical discussion of the problems in determining the completeness of the Revised NLTT Catalog and the impact of this completeness on parameter estimation. ",
        "conclusions": "}  To high precision (roughly $2\\,\\kms$) the LSR is not moving with respect to the halo in either the radial or vertical directions.  If the halo itself has no radial motion, the first result sharply contradicts the conclusion of \\citet{bs} based on gas motions that the LSR is moving outward at $14\\,\\kms$.  On the other hand, it is reasonably consistent with the radial-motion estimate of \\citet{ms} based on carbon stars.  More specifically, I find that the halo is moving at $11.4\\pm 2.2\\,\\kms$ relative to the Sun, and they find that the outer-Galaxy carbon stars are moving at $15.6\\pm 1.7\\,\\kms$.  I find that all three off-diagonal components of the velocity dispersion tensor are small, within $\\sim 2\\%$ of zero. The only previous measurements of these quantities (PG$^3$) were consistent with zero, but with errors that were about 4 times larger.  To date, I am not aware of any effort to predict the off-diagonal terms from theory.  The measurement presented here of the LF confirms the basic peaked shape found by \\citet{dahn}, but with about 40 times more stars and therefore covering a magnitude interval that is roughly twice as large.  It is inconsistent with the flat LF found by \\citet{bc}, and \\citet{gfb} (although in principle, since the latter determination was based on stars away from the solar neighborhood, it cannot be rigorously ruled out by my measurement). The present measurement is in rough agreement with that of \\cite{bc} at brighter magnitudes, $M_V<9$.  A shortcoming of the present approach is that there is no information about distances within the data set, so the scale of the velocity ellipsoid must be set by external information.  The distance scale could be set by obtaining either radial velocities (RVs) or trig parallaxes for a {\\it representative} (i.e., random) subset of the stars in the sample. The former would yield a statistical parallax solution. I stress ``representative'' because if the subsample is biased, for example is weighted toward stars with extreme kinematics and/or low metallicities, then the scale of the velocity ellipsoid will be overestimated by statistical parallax because the stars with RVs move faster than the those in the sample as a whole. It would be misestimated by trig parallax both because the selected stars would be faster and subluminous compared to the sample as a whole. Hence, one must choose a fair sample, and then make use of archival data only for stars within that sample.  From the standpoint of maximizing the precision of the distance-scale measurement with the minimum effort, statistical parallax is to be much preferred over trig parallax.  Even velocity errors of $\\sim 20\\,\\kms$ are quite adequate for a statistical parallax measurement with the limiting precision $\\sigma(\\eta)/\\eta = 0.65 N^{-1/2}$ \\citep{pg1}.  Here $N$ is the number of stars in the statistical parallax sample and $\\eta$ is the distance-scale parameter.  Even assuming perfect parallaxes, the limit for the trig parallax technique is $\\sigma(\\eta)/\\eta=0.2\\ln 10\\sigma(M_V)N^{-1/2}$. Given the large number of halo stars in the Revised NLTT Catalog and the relative ease  of making RV compared to trig parallax measurements, the modest per-star advantage of trig parallax will be overwhelmed by the mass-production techniques available for RVs.  However, good trig parallaxes would provide information on the luminosity of individual stars, which cannot be obtained from statistical parallax techniques.  Thus, the two approaches are complementary."
    },
    "0208/astro-ph0208232_arXiv.txt": {
        "abstract": "We present the first results of the Chandra and optical follow-up observations of hard X-ray sources detected in the ASCA Medium Sensitivity Survey (AMSS). Optical identifications are made for five objects. Three of them show either weak or absent optical narrow emission lines and are at low redshift $<$z$>$~$\\sim$~0.06. One of them is a broad line object at z~=~0.910 and one is a z~=~0.460 object with only narrow lines. All the narrow line objects show strong evidence for absorption in their X-ray spectra. Their line ratios are consistent with a Seyfert II/LINER identification as are the line widths. The three low redshift objects have the colors of normal galaxies and apparently the light is dominated by stars. This could be due to the extinction of the underlying nuclear continuum by the same matter that absorbs X-rays and/or due to the dilution of the central source by starlight. These results suggest that X-ray sources that appear as ``normal'' galaxies in optical and near-IR bands significantly contribute to the hard X-ray background. This population of objects has a high space density and probably dominates the entire population of active galaxies. ",
        "introduction": "The origin of the Cosmic X-ray Background (CXB) has been one of great mysteries in X-ray astronomy. At soft X-ray energies (below 2~keV), the ROSAT surveys resolved 70--80~\\% of the CXB into discrete sources (e.g., \\cite{Hasinger1998}). The majority of the optical identifications are unobscured active galactic nuclei (AGNs), mostly Seyfert Is and quasars (\\cite{Lehmann2001}). At higher energies, an additional population of absorbed or flat spectrum objects are needed to make up the CXB, which has a flatter spectrum in the 2--10~keV band than the unobscured AGNs.  Hard X-ray surveys performed with ASCA resolved about 30~\\% of the CXB in the 2--10~keV band, revealing the presence of a population of hard sources emerging at flux levels of several 10$^{-13}$~\\ergs\\ (\\cite{Ueda1999a}; \\cite{Ueda1999b}). Recently, almost all of the hard X-ray background has been resolved into point sources by the Chandra deep surveys (e.g., \\cite{Mushotzky2000}; \\cite{Brandt2000}; \\cite{Giacconi2001}). The faint sources detected with Chandra have, on average, rather hard spectra (e.g., \\cite{Rosati2002}), which indeed account for the spectrum of the CXB. However, due to the faintness of the Chandra sources in the optical band, their nature still remains, in general unclear (\\cite{Barger2001}).  In order to understand this hard X-ray population over a broad range in  luminosity and redshift , it is necessary to reveal the nature of the ``bright'' sources (i.e., with fluxes above $10^{-13}$~\\ergs\\ in the 2--10~keV band) that make up about 30~\\% of the CXB. In addition, because of their brightness, these sources should be much easier to study over a  wide range of wavelengths than the ultra-faint Chandra sources, and serve as  a prototype of the key populations that comprise the CXB. The results of the optical identification of a complete hard-band selected sample in the ASCA Large Sky Survey (ALSS, \\cite{Akiyama2000}) indicates that the majority of the hard sources are nearby $(z\\simlt 0.5)$ type-2 AGNs. However, in particular when sources with hard X-ray spectra are concerned, our knowledge is still quite limited due to the small sample size.  Thus, we have started to study the optical counterparts of the ``bright'' and ``hard'' sources selected from the ASCA Medium Sensitivity Survey (AMSS; \\cite{Ueda2001}), a serendipitous source survey of the extragalactic sky based on the ASCA GIS archival data at flux limits of 10$^{-13}$~\\ergs\\ (2--10~keV). The advantage of the AMSS is its large solid angle of sky coverage, which is indispensable for obtaining a reasonable sample of ``bright'' (hence rare) sources.  In this paper we report the first results of the Chandra observations and optical identifications for hard AMSS sources selected in Chandra AO-1. The excellent positional accuracy of $\\sim$~1~arcsec obtained with Chandra has enabled us to make unambiguous optical identifications of these sources. We present the results from X-ray, optical, and near infrared observations, in \\S~2, 3 and 4, respectively, followed by detailed description of individual sources in \\S~5. We discuss the results in \\S~5 and our conclusions are given in \\S~6. Throughout the paper, we assume a Hubble constant of $H_0$~=~50~km~s$^{-1}$~Mpc$^{-1}$ and a deceleration parameter of $q_0$~=~0.5. Optical identification for a complete flux-limited sample using a part of the AMSS catalog is reported by \\citet{Akiyama2002}. ",
        "conclusions": "\\subsection{X-ray Time Variability}  Two out of the seven ASCA sources observed by Chandra are at least ten times weaker than seen in the ASCA data. To check if this is attributable to selection effects in the ASCA data and/or the result of source confusion, we re-examined the results of Monte Carlo simulations as performed in \\S~3 of \\citet{Ueda2001}. With the same selection criteria for significance (above 4.5$\\sigma$ in the 2--10 keV band and below 4.5$\\sigma$ in the 0.7--2~keV band) and flux ($<3\\times10^{-13}$ \\ergs\\ in the 2--10~keV band) as for the present sample, we find that the probability that the flux measured with the ASCA GIS is more than 10 times larger than the real flux to be 6\\%, assuming that 20\\% of sources in the sky are highly absorbed at these flux levels (\\cite{Akiyama2000}). Hence, the expected number of false detections in our sample is 0.4. Note that the requirement of ``no detection in the soft band'' works to increase the fraction of false detections in the sample if the majority of sources have non-absorbed spectra, because unlike non-absorbed sources, the statistical fluctuation is independent of the energy bands. Given the areal density of Chandra sources at a flux level of 10$^{-14}$~\\ergs\\ of $\\sim$~200~deg$^{-2}$ the probability that an unrelated source will fall in the $\\sim$~1~arcmin$^{2}$ AMSS error circle is only 0.05. Thus it is unlikely that these Chandra sources are completely unrelated to the AMSS objects.  Thus, it is likely that these objects truly have a large variability range, although we do not rule out the possibility that either of the two is not real. Besides the two sources, AXJ~0223$+$4212 also shows a similar amplitude of variability. While detailed variability studies on times scales of years are only available for the brightest 25 AGNs, it is not unusual for objects to vary by factors of five over time scales of years and factors of ten are not unknown (e.g., \\citet{Peterson2000} for NGC4051 and \\citet{Weaver1996} for NGC2992). Thus it is not clear if the pattern of variability of these objects is unusual. However the large amplitude of variability confirms that these objects are indeed AGNs.  \\subsection{Summary of Identification and Comparison with Other Surveys}  Here we summarize the results of identification for the five Chandra sources. Three show either weak or absent optical narrow emission lines and are at very low redshifts of $<$z$>$~$\\sim$~0.06 (AXJ~0223$+$4212, AXJ~0431$-$0526, AXJ~1025$+$4714), one is a narrow line object at z~=~0.460 (AXJ~1510+0742), and the other is a broad line object at z~=~0.910 (AXJ~1951+5609). The optical line ratios, line widths, and the X-ray luminosity of the four narrow line objects indicate that they are consistent with their identifications as Seyfert II or LINER. Their X-ray spectra show strong evidence for high column densities.  The hard X-ray flux ($f_{\\rm X}$) and optical magnitudes ($f_R$) of the identified objects are plotted in Fig.~\\ref{opt_x1_fig}. In the same figure, we plot hard X-ray selected AGNs from HEAO1-A2 (triangles), the AMSS (circles), the ALSS (squares), and Chandra Deep Field North (asterisks and crosses, crosses means upper limit on the optical magnitude). Note that the HELLAS sample (\\cite{LaFranca2002}) falls on a similar range to the whole AMSS sample (see \\cite{Akiyama2002}). As seen from the figure, there is a wide variety of the $f_X/f_R$ ratio in our sample: AXJ~1510+0742 and AXJ~1951+5609 have hard X-ray to optical flux ratio close to that of the optically-faint X-ray sources seen by Chandra and XMM ($\\log f_X/f_R \\simgt +1$), while the three low redshift sources show a much smaller ratio ($\\log f_X/f_R < 0$). We defer to a later paper a detailed discussion of the origin of the wide range in the X-ray to optical flux ratios.   Fig.~\\ref{z_pi_fig} the redshift versus apparent photon index plot of our sample together with the ALSS sample, a flux limited complete sample at similar flux levels (\\cite{Akiyama2000}). As seen in this figure, three sources in our Chandra sample are even harder than the hardest X-ray source in the ALSS (AXJ~131501+314) in terms of the best-fit photon index. For comparison, Table~\\ref{lss_tab} summarizes the optical and X-ray properties of the five ``hardest'' sources in the above ALSS sample. These ALSS hard sources are all located at low redshifts of $z<0.4$ and are identified as four narrow line AGNs and one broad line AGN (AXJ~130926+2952). Equivalent widths of a narrow emission line of H$_\\alpha$ and [OIII]$\\lambda5007$ are also given in Table~\\ref{lss_tab} (for line ratios see Table~3 of \\cite{Akiyama2000}). Except for AXJ~130926+2952, which shows a strong [OIII]$\\lambda5007$ line, their equivalent widths are similar to those of the three nearby sources in our Chandra sample. Similar to the Chandra sample (see next section) most of the ALSS hard sources have the morphology of elliptical or early type spiral galaxies and no clear nucleus is visible in the optical images.   \\subsection{Hard X-ray Population of ``Normal'' Galaxies}  In this subsection, we discuss the nature of the three nearby sources in our AMSS-Chandra sample (AXJ~0223$+$4212, AXJ~0431$-$0526, and AXJ~1025$+$4714), which are identified as  Seyfert IIs or a LINERs. As we describe below, in many respects these sources are apparently ``normal'' galaxies, and could be considered to represent the major population of very hard sources at fluxes of $10^{-13}$ \\ergs\\ (2--10 keV). They are all in the nearby universe with an average redshift of 0.06 and have a very low hard X-ray to optical flux ratio ($\\log f_X/f_R < 0$). As shown in Fig.~\\ref{opt_image_fig}, these objects are clearly resolved and show no signs of a luminous nucleus. Interestingly, as opposed to virtually all optically selected Seyfert IIs, 80\\% of which are Sa or later in the sample of \\citet{Ho1997} and \\citet{McLeod1995}, the host galaxies of AXJ~0431$-$0526 and AXJ~1025$+$4714 have the morphology of E or S0. This may be related to the fact that these X-ray selected sources have larger luminosities compared with the optically selected Seyfert IIs.  They all have similar $J-K$ colors, consistent with that of old elliptical galaxies at z~$\\sim$~0.05 ($J-K=1$), and bluer than optically-selected QSOs at z~$\\simlt$~0.2 ($J-K=1.5-2$; \\cite{Elvis1994}).  Additionally, the existence of NaD and Mgb absorption lines in the optical continuum spectra of the objects suggests that their continuum in the optical wavelength is dominated by emission from host galaxy.  The X-ray to $K-$band flux ratio and the X-ray to optical flux ratio of these objects are smaller than normal QSOs. Even if the nuclei of AXJ~0223$+$4212, AXJ~0431$-$0526, and AXJ~1025$+$4714 are not affected by optical extinction (have similar $f_X/f_R$ or $f_X/f_K$ ratio to normal QSOs), the host galaxy components of these objects are brighter than the expected optical flux from the nucleus and may explain the relative weakness of the optical emission lines and the absence of a blue continuum by dilution of the nuclear component in the beam by starlight.  The optical line widths and equivalent widths themselves do not distinguish these objects from normal (i.e., non-AGN) galaxies. The upper limits of $\\sim$~500~km~s$^{-1}$ on the line widths are consistent with the distribution of line widths in Seyfert~II galaxies (Figure~4 of \\cite{Koski1978}) which has a average value of 600~km~s$^{-1}$ with a 250~km~s$^{-1}$ variance (see also \\citet{Whittle1992} who shows that OIII is a bit narrower with a median value of 325~km~s$^{-1}$ with a variance of 200~km~s$^{-1}$) and thus is not definitive in the classification of the objects. The low equivalent width of the lines and the fact that their intensity is similar to that found in a field galaxy sample (\\cite{Tresse1999}, \\cite{Carter2001}) again does not distinguish these objects. Similarly the luminosity of the lines is consistent with that seen in field galaxy samples. On the other hand, as we have shown above, the line ratios of two objects lie in the ionization range seen for many Seyfert~IIs and thus are indicative of AGN, and one has line ratios indicative of LINER activity. Such objects have been already detected in optical spectroscopic surveys of field galaxies: \\citet{Carter2001} find $\\sim$~15~\\% of all field galaxies have \"AGN-like\" emission line ratios.   The colors of the three nearby objects are completely consistent with that of stars with no sign of a non-thermal continuum. However, these galaxies are optically luminous with Mv~$\\sim$~$-$(21--22) and the relatively low luminosity active nuclei in these objects may be hidden by the glare of the starlight despite the low redshifts. Using the \\citet{Elvis1994} spectral energy distribution one predicts that the effective optical magnitude of the active nucleus would be $\\sim$~17--19~mag assuming that the nucleus is not absorbed and thus rather difficult to detect from ground based observations. The fact that these rather bright nearby galaxies harbor ``optically invisible'' AGN shows the incompleteness of optical active galaxy searches. Since the colors of these objects also show no abnormalities they would be missed by color selected surveys like the Sloan digital sky survey. It is also possible that the nuclei are absorbed in the optical and UV bands by the same matter that absorbs the X-ray continuum, reducing the observed optical flux. The effective reddening corresponding to the X-ray column densities of (0.5--15)~$\\times$~10$^{22}$~cm$^{-2}$ is Av~$\\sim$~ 2--70 mags thus effectively eliminating the optical-UV flux.   \\subsection{Contribution to the X-ray background}  Our results indicate that ``very hard'' X-ray sources at fluxes of $\\sim~10^{-13}$~\\ergs\\ (2--10~keV) are mostly nearby $z < 0.5$ sources with a column density of $\\sim~10^{22-23}$~cm$^{-2}$, rather than high redshift (hence high luminosity) objects with a higher column density of $\\sim~10^{23-24}$~cm$^{-2}$. This fully supports the result of the ALSS and is a strong constraint on  the unified models of the X-ray background. According to the calculation of \\citet{Comastri2001} and \\citet{Gilli2001}, about half of sources with a column density $>$~10$^{22}$~cm$^{-2}$ should have redshifts greater than 1 at a flux of $\\sim~1~\\times$~10$^{-13}$~\\ergs . We have to remember, however, that there could be bias against high redshift objects having moderate column densities in our Chandra sample because our selection criteria were based on the ``apparent'' hardness: due to K-correction absorbed sources at higher redshifts show softer spectra in the observer frame. Precise measurement of the column densities of slightly less hard sources detected in the ALSS and AMSS (\\cite{Akiyama2002}) is important for a definitive test of the unified model.   The optical counterparts of the hard, faint ASCA sources discussed in this paper have rather different properties from those of soft X-ray and optically selected AGN in the sense that they have earlier morphological types, do not possess a luminous optical or IR nucleus, have near-infrared colors similar to that of normal galaxies, and show only weak narrow emission lines. These unusual properties have already seen in the ALSS hard sample (Table~\\ref{lss_tab}) and in the HELLAS survey at a similar X-ray flux level (\\cite{Maiolino2000}) as well as in the deeper Chandra fields (e.g., \\cite{Barger2001})."
    },
    "0208/astro-ph0208468_arXiv.txt": {
        "abstract": "We analyzed the luminosity-temperature (L-T) relation for 2 samples of galaxy clusters which have all been observed by the {\\em ASCA} satellite.  We used 32 high redshift clusters (0.3$<$z$<$0.6), 53 low redshift clusters (z$<$0.3), and also the combination of the low and high redshift datasets.  This is one of two surveys to use only {\\em ASCA} data, and has the largest number of high redshift clusters.  We assumed a power law relation between the bolometric luminosity of the galaxy cluster and its integrated temperature (uncorrected for cooling flows) and redshift (L$_{bol,44}$=CT$^{\\alpha}$(1+z)$^A$).  We found that for an $\\Omega_{M}$=1.0 $\\Omega_{\\Lambda}$=0.0 universe, A = 1.134$^{+1.057}_{-1.073}$$\\pm1.66$, $\\alpha$ = 2.815$^{+0.322}_{-0.316}$$\\pm0.42$, and log(C) = $-$1.167$^{+0.216}_{-0.221}$$\\pm0.25$, and for $\\Omega_{M}$=0.3 $\\Omega_{\\Lambda}$=0.7, A = 2.052$^{+1.073}_{-1.058}$$\\pm1.63$, $\\alpha$ = 2.822$^{+0.320}_{-0.323}$$\\pm0.43$, and log(C) = $-$1.126$^{+0.223}_{-0.219}$$\\pm0.26$ (all errors at 68\\% confidence for one and two interesting parameters, respectively).  We found the dispersion at constant kT in this relation to be $\\Delta$logL=0.282 for $\\Omega_{M}$=1.0 $\\Omega_{\\Lambda}$=0.0, and $\\Delta$logL=0.283 for $\\Omega_{M}$=0.3 $\\Omega_{\\Lambda}$=0.7.  The results for the combined dataset and those found using the low and high redshift clusters are consistent, independent of cosmology, with previous estimates of L$\\sim$T$^3$ found by other authors.  The observed weak or zero evolution agrees with the predictions of models that produce L$\\sim$T$^3$ incorporating an initial source of non-gravitational energy before cluster collapse. ",
        "introduction": "Galaxy clusters have three major mass components; the galaxies themselves comprise about 1\\% of the cluster mass, about 10\\% of the mass is contained in the hot ($\\sim$10$^7$K) X-ray emitting gas of the intracluster medium (ICM), and the rest of the mass is contained in dark matter.  The dark matter itself probably exhibits self-similar scaling of its properties down to at the least the sizes of groups of galaxies due to its collisionless properties, and has been modeled with precision (e.g. Navarro, Frenk, \\& White 1997).  The ICM does not have the same scaling properties as the dark matter.  Instead it has a core of hot matter which is more extended with respect to self-similar scaling in groups of galaxies and small galaxy clusters than in rich clusters (Ponman \\etal 1999).  This break in the self-similar scaling between the hot intracluster medium and the dark matter raises questions about the formation of large structures.  The leading explanation for the break is the presence of additional non-gravitational energy in the early stages of galaxy cluster formation, proposed by Kaiser (1991) and Evrard \\& Henry (1991).  This energy produces an initial excess entropy that has a stronger effect on smaller and cooler galaxy clusters than high-mass clusters if the excess entropy is relatively constant for all clusters.  There are five reasons to suspect an entropy excess.  The first is the difference between the chemical properties and the spatial distribution of the ICM in groups and clusters with ICM temperatures less than about 1keV and those with ICM temperatures greater than 1keV (Renzini 1999).  Second, the observed mass-temperature (M-T) relation is steeper than the relation predicted by self-similar scaling for clusters of ICM temperatures less than about 3keV (Horner, Mushotzky, \\& Scharf 1999, Finoguenov \\etal 2001).  Third, Ponman \\etal (1999) found direct evidence of an entropy floor at low cluster temperature by looking at the entropy of the cluster gas at a fiducial radius as a function of the cluster temperature.  Fourth, entropy injection at early epochs is necessary to explain the correlation function and level of the X-ray background (Pen 1999; Wu, Fabian \\& Nulsen 2000).  The fifth and final reason to suspect a substantial initial entropy is the relation between the bolometric luminosity of galaxy clusters and their temperature (L-T).  If the simple scaling laws were applicable, the luminosity would scale as $\\sim$T$^2$ (Kaiser 1986).  The observed L-T relation is actually closer to L$\\sim$T$^3$ (Henry \\& Arnaud 1991, David \\etal 1993, Henry 1997, Arnaud \\& Evrard 1999, Borgani \\etal 1999, Riechart \\etal 1999, Fairley \\etal 2000, Henry 2000).  Including an initial entropy in numerical simulations of galaxy cluster formation can produce an L-T relation closer to the observational value (Bialek \\etal 2001, Tozzi \\& Norman 2001).  Initial entropy can be introduced into cluster formation scenarios at different times and from different sources such as star formation or active galactic nuclei (AGN).  If the entropy is not present when the gas is accreted onto the cluster, a much higher level is required to overcome the effects of cooling of high density gas in the cluster core.  Wu, Fabian \\& Nulsen (2000) concluded that the energies required for the initial entropy may be provided by either supernovae or AGN, however, the conditions for which the supernova energy injection is large enough to produce the observed relations are highly contrived.  They found that AGN easily produce sufficient energy although the mechanism for using this energy to heat the hot gas is not known.  Fujita (2001) found that blast waves driven by quasars are a possible heating mechanism for the intragroup gas prior to falling into the galaxy clusters.  Not all studies have concluded that additional entropy is required to produce the observed scaling laws of L-T and M-T.  Muanwong \\etal (2001) simulated galaxy cluster formation including radiative cooling with cool gas dropout and were able to reproduce L$\\sim$T$^3$ without adding any entropy to the gas.  Voit \\& Bryan (2001) and Voit \\etal (2002) propose a similar mechanism in which radiative cooling with subsequent supernova heating eliminate low-entropy gas.  In these models the self-similar scaling is broken because cooling is more efficient at lower temperatures.  The L-T relation has been well studied at low redshift, however, the situation at higher redshift is less clear. Self-similar scaling laws predict that L$\\sim$T$^2$(1+z)$^{1.5}$ for $\\Omega_M$=1 (where $\\Omega_{M}$ is the present matter density in units of the critical density), which would indicate a strong positive evolution, but these simple scaling laws are contra-indicated by current observational evidence.  Henry \\etal (1994), using three redshift bins, found that there is moderate to no evolution out to a redshift of 0.33.  Mushotzky \\& Scharf (1997), using galaxy clusters that had been observed with {\\em ASCA}, found evidence for no evolution out to a redshift of 0.4.  Fairley \\etal (2000) found that L$\\sim$T$^{3.15}$(1+z)$^{0.60\\pm0.38}$ for an $\\Omega_{M}$=0.3 universe, which is consistent with no evolution in the L-T relation out to a redshift of 0.8.  Sadat \\etal (1999) found some evidence for positive evolution, although their analysis estimates $\\Omega_{M}\\sim$0.85, which is in disagreement with the majority of recent results which indicate an $\\Omega_{M}$ of 0.2 to 0.4.   Arnaud, Aghanim, \\& Neumann (2001) used {\\em ASCA} and {\\em ROSAT} data for 25 hot (kT$>3.5$keV) clusters, and found that A is positive for a flat universe with $\\Omega_{M}$=0.4.  The L-T relation not only plays an important role in discerning the physics behind the formation of galaxy clusters, as summarized above, but it provides a link between observations of clusters and estimation of cosmological parameters. Cluster number abundance evolution can constrain the shape and amplitude of the mass fluctuation power spectrum as well as the matter density of the universe.  When using luminosity to constrain cosmology, it is necessary to modify the mass-luminosity (M-L) relation using the observed L-T relation to incorporate the fact that the luminosities are not given by the self-similar scaling laws.  Also, galaxy cluster samples are selected via their X-ray luminosity, but the X-ray temperature is more directly linked to the cluster mass and hence models of cluster abundance.  The L-T relation is used to convert the luminosity selection function to the temperature selection function.  These topics are beyond the scope of this paper, but for a complete discussion of the methods, see Henry (2000).  We constructed the L-T relation for a sample of galaxy clusters which have all been observed by {\\em ASCA}.  The sample is comprised of all of the clusters of which we are aware with 0.3$<$z$<$0.6 that have been observed with {\\em ASCA}, in combination with the clusters in the sample in David \\etal 1993 for which there are {\\em ASCA} data available, giving us a total sample of 85 clusters.  This sample is not flux limited but is assumed to be representative of galaxy clusters.  While data from {\\em Chandra} or {\\em XMM} might be better suited for our purposes, {\\em ASCA} data are the only ones available for a sufficiently large number of clusters to conduct a statistically relevant uniform study.  The {\\em ASCA} satellite has two high-quality spectroscopic instruments covering an energy range similar to the typical temperature range of galaxy clusters (kT$\\sim$2$-$10keV).  Many of the previous studies have used {\\em ROSAT} data alone or in combination with {\\em ASCA} data.  {\\em ROSAT} is only sensitive to a small range of energies (0.1$-$2.4keV) which are less than the average cluster temperature.  Using only {\\em ASCA} data gave us a sample which is uniform and allows us to do high precision statistical analysis.  The sample most similar to ours is the {\\em ASCA} Cluster Catalogue, which contains 273 galaxy clusters and groups all observed with {\\em ASCA} (Horner 2001).  This work used both pointed and serendipitous observations of clusters and explored the L-T, M-T, and optical velocity dispersion ($\\sigma$)-T relation out to a redshift of about 0.5.  We examined the slope of the L-T relation at low redshift ($<$0.3, 53 clusters), high redshift (0.3$<$z$<$0.6, 32 clusters), and for the dataset as a whole.  We investigated whether the L-T relation evolves with redshift for two different cosmologies thereby testing the prediction of little to no evolution out to z$\\sim$1 if there was a significant source of non-gravitational energy available durning the early stages of cluster formation (Bialek \\etal 2001, Tozzi \\& Norman 2001, Bower \\etal 2001).  We assumed a Hubble constant of 50 km s$^{-1}$Mpc$^{-1}$ throughout. ",
        "conclusions": "Previous studies at both low and high redshift have found $\\alpha\\simeq$3 (see earlier references).  We found $\\alpha$=3 within 1.52$\\sigma$ for all of our data, independent of cosmology.  Borgani \\etal (1999), using the X-Ray Luminosity Function (XLF) from the Rosat Deep Cluster Survey (Rosati \\etal 1998) and the Brightest Cluster Survey (Ebeling \\etal 1997), constrained $\\alpha$ to be between 3 and 4.  Using their XLF and constraints on $\\alpha$, they found 1$\\le$A$\\le$3 for an $\\Omega_{M}$=1.0 universe and A=0 implied a low density universe.  Mushotzky \\& Scharf (1997) found A=0.  Donahue \\etal (1999) found a slightly negative value of A (for $\\Omega_{M}$=1.0, A=$-$1.4$^{+0.8}_{-1.6}$, and for $\\Omega_{M}$=0.3 $\\Omega_{\\Lambda}$=0.7, A=$-$0.8$^{+0.9}_{-1.1}$, as read from their figure 7), and rule out A=1.5; their results were also consistent with A=0.  Reichart, Castander and Nichol (1999) determined the L-T relation for cooling flow corrected luminosities and temperatures. They found $\\alpha$ = 2.80$\\pm$0.15 and A = 0.35$^{+0.54}_{-1.22}$ or 1.53$^{+0.54}_{-1.22}$ for $\\Omega_M$=1, $\\Omega_{\\Lambda}$=0 or $\\Omega_M$=0.3, $\\Omega_{\\Lambda}$=0.7 respectively. These results agree with ours.  Horner (2001) found for L$_{bol}$$>$2x10$^{44}$erg s$^{-1}$ that $\\alpha$=2.98$\\pm$0.14 and A=0.02$\\pm$0.16 for no cooling flow corretions and $\\Omega_M$=1.0 (his equation 5.7), again consistent with our results.  The A parameter is closely related to the duration and heating epoch of the ICM in the non-gravitational heating models (Cavaliere \\etal 1997).  Bialek \\etal (2001) found that a model using nonzero initial entropy yielded an $\\alpha$ consistent with observations.  A consequence of this model is that A is approximately zero out to a redshift of 0.5.  Tozzi \\& Norman (2001) included cooling and shocks in their model with an initial entropy, and they predicted A to be zero or only slightly positive out to a redshift of one.  They found that changing cosmology does not have a strong effect on the value of A.  The evolution expected in the cooling/drop out models has not yet been determined.  Obviously if the evolution is different from the heating models then we may be able to discriminate between them.  Our results indicate that the value of A for the matter-dominated $\\Omega_{M}$=1.0 cosmology is smaller than for the $\\Omega_{M}$=0.3 $\\Omega_{\\Lambda}$=0.7 case, but that both results are consistent with zero.  We conclude that our results are consistent with weak or no evolution of the L-T relation, and agree with the heating models mentioned above."
    },
    "0208/astro-ph0208142_arXiv.txt": {
        "abstract": "We present an analysis, using Yale stellar evolution models, of the color-magnitude diagrams (CMDs) of three intermediate-age LMC clusters, namely NGC~2173, SL~556 and NGC~2155, obtained with the VLT. The main goal of our project is to investigate the amount of convective core overshoot necessary to reproduce the CMDs of relatively metal-poor, intermediate age stellar populations, to check whether the extrapolation that is usually made from solar metallicity is valid. In the process, we obtained values for the binary fraction of each cluster, together with refined age estimates. Our method involved the comparison of the observed CMDs with synthetic CMDs computed using various values of the overshoot parameter and binary fraction. We conclude that a moderate amount of overshoot and some fraction of binary stars are essential for reproducing the observed shapes around the turnoff in the CMD's of all three clusters: unresolved binary stars fill in the expected core contraction gap, and make a unique sequence near the gap, which cannot be reproduced by single stars alone, even with a larger amount of overshoot. We utilize ratios of the number of stars in different areas around the core contraction gap to constrain the binary fraction, which is around 10-20\\% (for primary-to-secondary mass ratio $\\ge$ 0.7) in all three clusters. Even if binary stars contaminate the core contraction gap, it is shown that the overshoot parameter can be inferred from the color dispersion of the stars around the contraction gap, regardless of the assumed binary fraction. From our overall analysis such as, shape of isochrones, star counts, color distribution, and synthetic CMD comparisons, we conclude that overshoot $\\sim 20\\%$ of the local pressure scale height best reproduces the CMD properties of all three clusters. The best age estimates are 1.5, 2.1 and 2.9\\,Gyr for NGC~2173, SL~556 and NGC~2155, respectively. ",
        "introduction": "Stellar evolution theory is routinely tested using Galactic clusters, either globular (in general old, and both metal-poor and metal-rich) or open (intermediate-age or young, and generally more metal-rich than globular clusters). These clusters reflect the particular star formation and chemical enrichment history of the Milky Way, and certainly do not represent all possible stellar populations. For example, intermediate-age, metal-poor populations, which are not well represented in the Galaxy, are conspicuous in the Magellanic Clouds. This is an important part of the parameter space, which is relevant, for example, to the study of stellar populations of dwarf galaxies in the Local Group, or to that of galaxies at high redshifts, and therefore, in early evolutionary stages.  There are uncertainties that are critical to modeling intermediate-mass stellar models. Most notable of them are the treatments of convection (e.g., convective core overshoot) and mass loss. In this paper, we will address them, using intermediate-age Magellanic Cloud clusters as comparison templates.  Convective core overshoot (OS) is one of the physical parameters which must be taken into account during the core hydrogen burning phase of the evolution of intermediate-mass stars. Stellar evolution theory predicts that the CMD of young and intermediate-age star clusters is greatly affected by the amount of OS. In the early days of stellar structure theory, the convective core size was determined by the classic Schwarzschild (1906) criterion, ignoring the so-called OS, which is due to the inertial motion of materials beyond the formal core edge. Initially, the extent of core OS was thought to be negligible (Saslaw \\& Schwarzschild 1965). Subsequent theoretical studies, however, have emphasized the importance of convective core OS in stellar evolutionary models (Shaviv \\& Salpeter 1973).  Since then, many observational and theoretical studies have been devoted to investigating the effects of core OS (Prather \\& Demarque 1974; Bressan, Chiosi \\& Bertelli 1981; Bertelli, Bressan \\& Chiosi 1985; Stothers 1991; Meynet, Mermilliod \\& Maeder 1993). It is now generally believed that a moderate amount of core OS is necessary in order to reproduce the shape of the core contraction gap observed in CMDs of galactic open clusters (Maeder \\& Mermilliod 1981; Stothers \\& Chin 1991; Carraro et al. 1993; Daniel et al. 1994; Demarque, Sarajedini, \\& Guo 1994; Kozhurina-Platais et al. 1997; Rosvick \\& VandenBerg 1998), but the dependence of core OS on stellar mass is still poorly understood and the subject of on-going investigations (e.g. Ribas, Jordi, \\& Gimenez 2000).  The extent of OS is not currently derivable through direct theoretical calculations but can be determined empirically by comparing synthetic CMDs to observed ones, with OS in the models often being parameterized as a fraction of the pressure scale height at the formal edge of the convective core. In the CMDs of intermediate-age star clusters ($2 \\lesssim$ Age $\\lesssim 5$\\,Gyr), which have a distinctive gap near the turnoff (TO) caused by the rapid contraction of a hydrogen-exhausted convective core, the TO topology can be used to evaluate the extent of core OS (Demarque et al. 1994; Kozhurina-Platais et al. 1997). In the case of younger clusters, the importance of core OS can be estimated from the observed luminosity function of main sequence (MS) stars (Chiosi et al. 1989; Barmina, Girardi \\& Chiosi 2002) and from the ratio of main sequence to post-main sequence stars. Studies of detached eclipsing binaries can also provide empirical estimates of OS by direct comparison with stellar tracks in the log $T_{\\rm eff}$ - log $g$ plane (Schroder, Pols, \\& Eggleton 1997; Guinan et al. 2000; Ribas et al. 2000). However, the number of binary systems available for OS studies is very limited.  Many galactic open cluster studies have suggested a moderate amount of OS, i.e. $\\sim 20 \\%$ of the local pressure scale height at the edge of a convective core. However, this has been tested only for near solar metallicity clusters in the Galaxy. So far, the presence of OS in metal-poor clusters in the Magellanic Clouds has only been tested in a limited sample of young (age $\\le$ 500 Myr) clusters (Lattanzio et al. 1991; Vallenari et al. 1991, 1994; Stothers \\& Chin 1992; Brocato, Castellani \\& Piersimoni 1994; Chiosi et al. 1995; Testa et al. 1999; Keller, Da Costa \\& Bessell 2001; Barmina et al. 2002). For Magellanic Clouds intermediate-age clusters, the necessary observations are more challenging due to the fainter magnitude of the TO, and HST or 8-m class telescopes under excellent seeing conditions are required.  The characteristics of mass loss during the RGB phase are still unclear in many aspects. Among the early papers on the subject, that of Reimers (1975) was most influential and his mass loss formula has been widely used to date:  \\begin{equation} \\dot{M} = -4 \\times 10^{-13} \\eta \\frac{L}{gR} \\end{equation} where L, g, and R are luminosity, gravity and radius, respectively. The mass loss efficiency parameter $\\eta$, which was later introduced, has been reported to vary from 0.25 to 2 -- 3 for metal-rich stars \\cite{d86,kr78,r81}. In contrast, the estimated range of $\\eta$ for metal-poor stars appeared to be narrow, as most studies on metal-poor stars suggest $\\eta$ = 0.3 -- 0.7 \\cite{am82,ma82,r81,ldz94}, and these values successfully reproduce the horizontal branch morphology of the old Galactic globular clusters \\cite{yi99}. While the application of Reimers' formula for low mass stars predicts that the total amount of mass loss gets smaller as mass increases, it is not clearly demonstrated how the amount of mass loss depends on the mass of the stars in intermediate-age clusters (Tripicco, Dorman \\& Bell 1993; Liebert, Saffer \\& Green 1994; Carraro et al. 1996).  In this context, we started a project to investigate several aspects in the evolution of intermediate-age, metal-poor stellar populations, in particular, the amount of OS and the mass loss characteristics. To reach this goal, we observed three intermediate-age clusters using the Very Large Telescope (VLT) under excellent seeing conditions. The project and the observations are presented in Gallart et al. (2002; Paper I). In the current paper, we will compare the observations with the recent set of Yonsei-Yale ($Y^{2}$) stellar evolutionary models, and with models specifically calculated for this project using the same main input physics (but differing in the OS parameter), to constrain the amount of OS and the mass loss.  The three clusters under the study seem to present the signature of a core contraction gap in their TO region, which is predicted by the models as well for this age range. A possible obstacle in studies of OS using the shape of the core contraction gap is binary contamination. Castellani, Degl'Innocenti \\& Marconi (1999) pointed out that the contraction gap enhanced by core OS can be filled in by unresolved binary stars.  In fact, some intermediate-age clusters, which are expected to have a core contraction gap, show a continuous sequence rather than a distinctive gap (Carraro et al. 1994; Paper I). Although this could be considered as a limiting factor for the topology method, we will show that it is possible, using synthetic CMDs, to put some constraints on the binary star population. It may be also expected that an increase of binary fraction in synthetic CMD simulations tends to decrease the value of OS required for a good fit. Testa et al. (1999) have claimed that a no-OS model with $\\sim 30\\%$ of unresolved binaries allows a fair fit to the observed MS luminosity function of NGC~1866 in the LMC, while Barmina et al. (2002), using a revised analysis of the same data, have concluded that the inclusion of unresolved binaries would not alter the need for OS (see also for other clusters, Vallenari et al. 1992; Carraro et al. 1994)  In this paper, we present a short description of the observational data (\\S~2). Stellar models including OS treatment and synthetic CMD simulations are described in section 3. Detailed analysis on the steps of deriving ages, binary fraction, the amount of OS and mass loss is described in section 4, followed by a short discussion (\\S~5) and conclusions (\\S~6). ",
        "conclusions": "In this paper, we have performed detailed synthetic CMD simulations on the effect of OS and binary stars, compared with observed CMDs of three LMC clusters, NGC~2173, SL~556, and NGC~2155.  Unresolved binary stars fill in the core contraction gap and make a unique sequence near the gap, which cannot be reproduced by single stars alone, even with a larger OS amount.  We utilize the number ratios of stars around the core contraction gap to derive a first order estimate of the BF. Based on our star count analysis, all three clusters have a BF of 10\\% to 20\\% (with $q \\ge 0.7$).  Using the color distribution of stars around the core contraction gap, which does not significantly depend on the BF, we find that the best OS parameter for each cluster is close to 20\\% of the local pressure scale height.  With the best solutions of the BF and OS from star count and color distribution analysis, we present synthetic CMDs for each cluster with the observed CMD in Figure 11, 12, and 13, respectively.  From the overall analysis, such as isochrone fitting, star counts, color distributions, and synthetic CMD comparison, we conclude that a moderate amount of OS, $\\sim 20\\%$ of the local pressure scale height, is essential to reproduce the observed shape around the core contraction gap in the CMDs of all three clusters, which are more metal-poor than the intermediate-age open clusters observed in the Galaxy, implying OS does not depend on metallicity at least for this metallicity range. With OS=0.2, our best age estimates are 1.5, 2.1 and 2.9\\,Gyr for NGC~2173, SL~556 and NGC~2155, respectively.  We constrainted mass loss along the RGB phase from the observed RC luminosity of each cluster, utilizing the RC luminosity -- mass relation derived from synthetic CMDs. We found that the mas loss estimate for NGC~2155 is consistent with the typical mass loss parameterization that works for the Galactic globular clusters. However, the SL~556 data, and perhaps the data for the other young cluster NGC~2173 as well, indicate a substantially larger amount of mass loss than suggested by Reimers' formula. This may indicate once again the complexity of mass loss processes."
    },
    "0208/astro-ph0208374_arXiv.txt": {
        "abstract": "We combine near-UV spectra obtained with the {\\it Hubble Space Telescope} GHRS echelle with far-UV spectra obtained with IMAPS and {\\it Copernicus} to study the abundances and physical conditions in the predominantly ionized gas seen at high velocity ($-$105 km s$^{-1}$ $\\la$ $v_{\\odot}$ $\\la$ $-$65 km s$^{-1}$) and at intermediate velocity ($-$60 km s$^{-1}$ $\\la$ $v_{\\odot}$ $\\la$ $-$10 km s$^{-1}$) along the line of sight to the star $\\zeta$ Ori. We have high resolution (FWHM $\\sim$ 3.3--4.5 km s$^{-1}$) and/or high S/N spectra for at least two significant ions of C, N, Al, Si, S, and Fe --- enabling accurate estimates for both the total $N$(\\ion{H}{2}) and the elemental depletions. C, N, and S have essentially solar relative abundances; Al, Si, and Fe appear to be depleted by about 0.8, 0.3--0.4, and 0.95 dex, respectively, relative to C, N, and S. While various ion ratios would be consistent with collisional ionization equilibrium (CIE) at temperatures of 25,000--80,000 K, the widths of individual high-velocity absorption components indicate that $T$ $\\sim$ 9000$\\pm$2000 K --- so the gas is not in CIE. Analysis of the \\ion{C}{2} fine-structure excitation equilibrium, at that temperature, yields estimates for the densities ($n_e$ $\\sim$ $n_{\\rm H}$ $\\sim$ 0.1--0.2 cm$^{-3}$), thermal pressures ($2 n_{\\rm H} T$ $\\sim$ 2000--4000 cm$^{-3}$K), and thicknesses (0.5--2.7 pc) characterizing the individual clouds. We compare the abundances and physical properties derived for these clouds with those found for gas at similar velocities toward 23 Ori and $\\tau$ CMa, and also with several different models for shocked gas. While the shock models can reproduce some features of the observed line profiles and some of the observed ion ratios, there are also significant differences between the models and the data. The measured depletions suggest that roughly 10\\% of the Al, Si, and Fe originally locked in dust in the pre-shock medium may have been returned to the gas phase, consistent with recent predictions for the destruction of silicate dust in a 100 km s$^{-1}$ shock. The observed near-solar gas phase abundance of carbon, however, appears to be inconsistent with the predicted longer time scales for the destruction of graphite grains. ",
        "introduction": "\\label{sec-intro}  Interstellar gas found at high and intermediate velocities in the Galactic plane near associations of young, massive stars presumably reflects the combined effects of stellar winds and supernovae arising in those associations. Observations of UV absorption lines can reveal the kinematics, elemental abundances (and depletions), ionization state, and physical properties (density, temperature, etc.) characterizing such high- and intermediate-velocity gas --- and can thus provide significant empirical tests for models of (1) the interaction of supernovae with the surrounding interstellar medium, (2) shocks and the non-equilibrium evolution of shocked gas, and (3) the survival of dust grains in such environments.  UV spectra obtained with {\\it Copernicus} revealed absorption from various singly and doubly ionized species at both high ($-$100 to $-$80 km s$^{-1}$) and intermediate ($-$80 to $-$15 km s$^{-1}$) heliocentric velocities toward a number of stars in Orion (Cohn \\& York 1977; Cowie, Songaila, \\& York 1979). Cowie et al. ascribed the widespread high-velocity (HV) gas to a radiative shock produced by one or more recent supernovae and called it ``Orion's Cloak''. Subsequent UV spectra of additional stars in Orion obtained with {\\it IUE} provided further information on the distribution of the HV and intermediate-velocity (IV) gas throughout that region (Shore 1982; Sonneborn, Shore, \\& Brown 1988). More recently, higher resolution UV spectra of 23 Ori obtained with the {\\it Hubble Space Telescope} Goddard High Resolution Spectrograph ({\\it HST}/GHRS) have enabled more precise characterization of the HV and IV gas in that line of sight (Trapero et al. 1996; Welty et al. 1999). Toward 23 Ori, the HV gas is essentially fully ionized, and is characterized by a density $n_e$ $\\sim$ $n_{\\rm H}$ $\\sim$ 0.4--0.5 cm$^{-3}$, a temperature $T$ $\\sim$ 8000 K, and a thermal pressure 2$n_{\\rm H}T$ $\\sim$ 5--10 $\\times$ 10$^3$ cm$^{-3}$K. As the ion ratios $N$(\\ion{X}{3})/$N$(\\ion{X}{2}) for aluminum and silicon would be consistent with collisional ionization equilibrium (CIE) at much higher temperatures --- of order 25000 K --- the HV gas is clearly not in CIE. No data (or only upper limits) were available, however, for a number of potentially significant ions (e.g., \\ion{C}{3}, \\ion{N}{3}, \\ion{S}{3}, \\ion{Fe}{2}, \\ion{Fe}{3}), and some of the spectra were obtained at only moderate spectral resolution.  In this paper, we discuss the predominantly ionized HV and IV gas toward a second Orion star ($\\zeta$ Ori), using near-UV spectra obtained with GHRS and far-UV spectra obtained with the Interstellar Medium Absorption Profile Spectrograph (IMAPS) and {\\it Copernicus}. In Sections~\\ref{sec-obsred} and~\\ref{sec-prof}, we describe the spectra and our analysis of the observed absorption-line profiles. In Section~\\ref{sec-res}, we discuss the abundances and physical properties derived for the individual components in the HV and IV gas. The unprecedented combination of high spectral resolution (FWHM $\\sim$ 3.3--4.5 km s$^{-1}$), high signal-to-noise ratio (S/N), and broad spectral coverage has enabled the determination of accurate abundances for both singly and doubly ionized C, N, Al, Si, S, and Fe --- which then provide strong constraints on the overall $N$(H), the depletions, and the ionization in the HV and IV gas. Detailed multicomponent fits to the high-resolution line profiles and analyses of the \\ion{C}{2} fine-structure excitation equilibrium yield values for the temperature and density of the gas, respectively. In Section~\\ref{sec-disc}, we compare the elemental abundances and derived physical properties with those found for gas observed at similar velocities in several other lines of sight, we compare the observed line profiles and ion ratios [$N$(\\ion{X}{3})/$N$(\\ion{X}{2}), $N$(\\ion{X}{4})/$N$(\\ion{X}{2})] with those predicted by various models for ionized gas (including newly-computed models for shocked gas), and we discuss the observed depletions in relation to recent models for dust processing in shocks. In an appendix, we present similar (though more limited) data for the high-velocity gas toward $\\tau$ CMa. ",
        "conclusions": "\\label{sec-sum}  We have used near-UV spectra obtained with the {\\it Hubble Space Telescope} GHRS together with far-UV spectra from IMAPS and {\\it Copernicus} to study the abundances and physical conditions in the ionized high- and intermediate-velocity gas seen along the line of sight to the star $\\zeta$ Ori A. The combination of high spectral resolution (FWHM $\\sim$ 3.3--4.5 km s$^{-1}$), high S/N (90--550 for GHRS echelle; 30--80 for IMAPS), and broad spectral coverage has enabled a much more detailed and accurate view of the HV and IV gas than was possible using spectra from {\\it Copernicus} alone. With 2$\\sigma$ detection limits better than 1 m\\AA\\ in many cases, we have obtained HV column densities for both singly and doubly ionized C, N, Al, Si, S, and Fe, and also for \\ion{Mg}{2} and \\ion{C}{2}*; we also have significantly restrictive limits for a number of neutral and triply ionized species. In the IV gas, we also detect absorption from \\ion{O}{1}, \\ion{C}{4}, and \\ion{Si}{4}. The column density sums $N$(\\ion{X}{2}) + $N$(\\ion{X}{3}) for C, N, and S yield estimates for the total hydrogen column density $N$(\\ion{H}{1}) + $N$(\\ion{H}{2}); by comparison, the corresponding sums for Al, Si, and Fe then provide estimates for the depletions. Limits (or values) for $N$(\\ion{O}{1}) and $N$(\\ion{N}{1}) yield estimates for the amount of neutral gas $N$(\\ion{H}{1}). Fits to the high-resolution line profiles suggest (at least) 5 components for the HV gas and 8 components for the IV gas; comparisons of line widths, for species of different atomic weight, allow estimates for the temperature and turbulent velocity in each component. Once $T$ is known, analysis of the \\ion{C}{2} fine-structure excitation provides estimates for the local densities ($n_e$ $\\sim$ $n_{\\rm H}$) in several of the components.  The HV gas ($-$105 km s$^{-1}$ $\\la$ $v$ $\\la$ $-$65 km s$^{-1}$) is predominantly ionized, with \\\\ $N$(\\ion{H}{1})/$N$(\\ion{H}{2}) $<$ 0.001 and $N$(\\ion{X}{3})/$N$(\\ion{X}{2}) ranging from 1.4 (nitrogen) to 11 (iron). The total hydrogen column density $N$(\\ion{H}{2}) is about 7.6 $\\times$ 10$^{17}$ cm$^{-2}$. The depletions [$-$0.8 dex (Al), $-$0.4 dex (Si), and $-$0.95 dex (Fe), relative to C, N, and S] appear to be intermediate between those found in Galactic halo clouds and in warm, diffuse disk clouds --- though the depletions for aluminum and iron would be less severe if significant amounts of \\ion{Al}{4} and \\ion{Fe}{4} are present. While the various $N$(\\ion{X}{3})/$N$(\\ion{X}{2}) ratios would be consistent with collisional ionization equilibrium at temperatures of 25,000--80,000 K, the component line widths imply $T$ $\\sim$ 9000 K --- i.e., the HV gas is not in CIE. Characteristic densities for the HV components are $n_e$ $\\sim$ $n_{\\rm H}$ $\\sim$ 0.1--0.2 cm$^{-3}$; typical thermal pressures are 2$n_{\\rm H}T$ $\\sim$ 2000--4000 cm$^{-3}$K.  For the IV gas ($-$60 km s$^{-1}$ $\\la$ $v$ $\\la$ $-$10 km s$^{-1}$), the overall column densities are all roughly 3 times higher than for the HV gas, but the ion ratios, depletions, and inferred physical conditions are all similar to those found for the HV gas.  We have considered two sets of shock models to try to understand the origins of the HV and IV gas toward $\\zeta$ Ori: (1) for the HV gas, a shock moving toward us at a velocity of order $-100$ km s$^{-1}$, and (2) for the IV gas, a shock moving away from us, produced when a flow of high-velocity gas encounters lower velocity material. The models are 1-D, assume steady flow, and include the effects of magnetic fields, stellar ionizing radiation, and radiation emitted as the gas behind the shock cools. The inclusion of a stellar radiation field seems necessary to reproduce the observed ion ratios $N$(\\ion{X}{3})/$N$(\\ion{X}{2}) and the weakness of \\ion{N}{1}. While the models can yield ion abundances and absorption-line profiles that resemble those observed, there are some significant differences. For example, the models have difficulty reproducing the full set of observed ion ratios; the broad, complex tails in the observed profiles; and the general weakness of the observed \\ion{C}{4} and \\ion{Si}{4}. Inhomogeneities in the density and/or magnetic field may produce the more complex structure seen in the observed profiles. More complex shock models (e.g., for slow mode MHD shocks and/or switch-on, switch-off shocks) may be required to more accurately reproduce the observed line profiles and ion ratios.  We have also briefly examined other models for ionized gas --- diffuse photoionized gas, denser \\ion{H}{2} regions, collisionally ionized gas in equilibrium, and isochorically cooling gas. Those diverse models predict very different values for the various ion ratios [e.g., $N$(\\ion{X}{3})/$N$(\\ion{X}{2})]. While photoionization by an O9 star radiation field might yield a reasonable match with a number of the HV gas ion ratios, none of the various models appears able to reproduce the full set of ratios observed in the HV or IV gas toward $\\zeta$ Ori.  If we assume that the depletions in the pre-shock gas were similar to those found in warm, diffuse disk clouds --- such as some of those seen at lower velocities toward $\\zeta$ Ori and 23 Ori --- then the residual depletions of Al, Si, and Fe found in the HV and IV gas suggest that of order 10\\% of the dust might have been destroyed in the shocks that produced the HV and IV gas. While the $\\sim$ 10\\% destruction for Al, Si, and Fe is comparable to the percentages predicted by the theoretical models for a 100 km s$^{-1}$ shock, the similar values for silicon and iron disagree with expectations derived from comparisons of cold cloud and warm cloud depletions --- which suggest that silicon is returned to the gas faster than iron. Moreover, the observed near-solar gas phase abundance of carbon appears inconsistent with the predicted longer time scales for the destruction of graphite grains."
    },
    "0208/astro-ph0208412_arXiv.txt": {
        "abstract": "We present $J$ and $K$-band near-infrared photometry of a sample of mid-infrared sources detected by the {\\em Infrared Space Observatory} ({\\em ISO}) as part of the European Large Area {\\em ISO}-Survey (ELAIS) and study their classification and star-forming properties. We have used the Preliminary ELAIS Catalogue for the 6.7 $\\umu$m (LW2) and 15 $\\umu$m (LW3) fluxes. All of the high-reliability LW2 sources and 80 per cent of the LW3 sources are identified in the near-IR survey reaching $K \\approx 17.5$ mag. The near- to mid-IR flux ratios can effectively be used to separate stars from galaxies in mid-IR surveys. The stars detected in our survey region are used to derive a new accurate calibration for the ELAIS ISOCAM data in both the LW2 and LW3 filters.   We show that near to mid-IR colour-colour diagrams can be used to further classify galaxies, as well as study star-formation.  The ISOCAM ELAIS survey is found to mostly detect strongly star-forming late-type galaxies, possibly starburst powered galaxies, and it also picks out obscured AGN. The ELAIS galaxies yield an average mid-IR flux ratio LW2/LW3 $=0.67 \\pm 0.27$. We discuss the $f_{\\nu}(6.7\\umu{\\rm m}) / f_{\\nu}(15\\umu{\\rm m})$ ratio as a star formation tracer using {\\em ISO} and {\\em IRAS} data of a local comparison sample.  We find that the $f_{\\nu}(2.2\\umu{\\rm m}) / f_{\\nu}(15\\umu{\\rm m})$ ratio is also a good indicator of activity level in galaxies and conclude that the drop in the $f_{\\nu}(6.7\\umu{\\rm m}) / f_{\\nu}(15\\umu{\\rm m})$ ratio seen in strongly star-forming galaxies is a result of both an increase of $15\\umu$m emission and an apparent depletion of $6.7\\umu$m emission. Near-IR data together with the mid-IR give the possibility to estimate the relative amount of interstellar matter in the galaxies. ",
        "introduction": "There has been determined effort over the past several years to understand the history of luminous matter in the Universe. Ultimately, one wishes to have a consistent understanding which would tie together the detailed physical processes at work in stars and ISM in the Milky Way and local galaxies with the integrated properties of more distant systems.  The spectral properties and energy budget of the distant galaxies in turn are crucial in understanding the universal history of star formation, the very faintest source counts, and the extragalactic background radiation.  In particular, the infrared and sub-mm regimes have become the focal point of interest in studies of galaxies, both normal and extreme objects. The near-infrared is an important region for galaxy evolution studies for several reasons.   Dust extinction is significantly less hampering here than in the optical, and the light mostly comes from a relatively stable old population of late-type stars making galaxy colours, counts, and K-corrections easier to predict and interpret. It is also in the near-IR that the energy output of a galaxy starts to shift from normal starlight to emission re-radiated by interstellar matter. By $5 \\umu$m the dust emission has taken over from radiation from stellar photospheres, except in most ellipticals.  Apart from the $[12/25] \\equiv f_{\\nu}(12\\umu{\\rm m}) / f_{\\nu}(25\\umu{\\rm m})$ colours of {\\em IRAS} galaxies, the mid-infrared truly opened up for study only with the {\\em ISO}-mission (see reviews by Genzel \\& Cesarsky 2000, Helou 1999). Many studies (\\eg Mattila, Lehtinen \\& Lemke 1999, Helou \\ea 2000) have confirmed the complex nature of spectral energy distributions of disk galaxies in the $3 - 20 \\umu$m range. In addition to a continuum due to hot (or warm) dust there are bright IR-bands at 3.3, 6.2, 7.7, 8.6, 11.3, and $12.7 \\umu$m -- these are often called the Unidentified Infrared Bands (UIBs), due to the lack of understanding of their carriers.  These broadband aromatic features are proposed to be the signature of Polycyclic Aromatic Hydrocarbons (PAH; L\\'eger \\& Puget 1984).  The PAHs are an essential component in forming the mid-infrared $[6.7/15]$ colour ratio which is emerging as a tracer of star forming activity in galaxies (Vigroux \\ea 1996, 1999, Sauvage \\ea 1996, Dale \\ea 2000, Roussel \\ea 2001a, Helou 2000). The value $[6.7/15] \\approx 1$ is expected in quiescent medium and PDRs, while HII regions have $[6.7/15] < 0.5$ (\\eg Cesarsky \\ea 1996). The $[6.7/15]$ ratio thus remains close to unity for quiescent and mildly star forming galaxies, while it starts to drop for those with more vigorous star formation activity. This mid-IR flux ratio has been also shown to correlate with the IRAS $[60/100]$ colour ratio, which is a well known indicator of activity level in galaxies (Helou 2000, Dale \\ea 2000, Vigroux \\ea 1999). A two-component model of a galaxy as a linear combination of differing amounts of cold dust in cirrus clouds and warmer dust in HII regions has been seen as the explanation for the IRAS and IRAS-ISO colour-colour diagrams (Helou 1986, Dale \\ea 1999, 2000). On the other hand, the situation might be more complicated (see \\eg Sauvage \\& Thuan 1994), and for example it is possible that the proportion of star formation in the disk relative to the central region of a galaxy plays a dominant role (\\eg Vigroux \\ea 1999, Roussel \\ea 2001b).  On another front, deep {\\em ISO} galaxy counts (\\eg Oliver \\ea 1997, Taniguchi \\ea 1997, Elbaz \\ea 1999a, Aussel \\ea 1999, Flores \\ea 1999, for ISOCAM counts) have produced surprising results.  The differential $15\\umu$m counts show a remarkable upturn below flux densities of 3 mJy and then a rapid convergence at approximately 0.4 mJy. This peak clearly requires strong (luminosity) evolution and can be a result of strong mid-IR emission features, a new population of sources, or some combination of these (Xu 2000, Elbaz 1999b, Genzel \\& Cesarsky 2000). To understand these results and to develop a coherent picture of early galaxy evolution, it is imperative to learn as much as possible about the more local galaxies.  The ELAIS project (Rowan-Robinson \\ea 1999, Oliver \\ea 2000) stands as a bridge between the very deep galaxy surveys in the infrared mentioned above, and nearby galaxy surveys (\\eg Boselli \\ea 1998, Dale \\ea 2000, Roussel \\ea 2001a). ELAIS was the largest open time {\\em ISO}-project with the driving ambitious goal to study the un-obscured star formation out to redshifts of $z \\sim 1$.  Source counts in the mid-IR have been published in Serjeant et al.\\ (2000) and the far-IR counts in Efstathiou et al.\\ (2000).  The aims of this paper are as follows: In Section~\\ref{obs} we present a subset of the ISOCAM ELAIS survey with near-IR follow-up observations. A central new result of this paper, the calibration of the ELAIS data is performed in Section~\\ref{calib} and in the Appendix using the stars detected in our fields. In Section~\\ref{colcol} various near- to mid-IR colour-colour diagrams of the ELAIS galaxies are constructed and compared to evolutionary models including the UIB features in the mid-IR, and to a local ISO galaxy sample. In Section~\\ref{class} we attempt to classify sources based on their NIR-MIR colours. Classifications such as this are expected to be helpful in the future, eg.\\ with {\\em SIRTF} and {\\em ASTRO-F} data, when large numbers of galaxies with near-IR and mid-IR fluxes become available without high-resolution spectra accompanying them at least in the first instance.  In Section~\\ref{tracer} we discuss star formation properties of the ELAIS galaxies, and the mid-IR and near-to-mid-IR colours as tracers of star formation. Finally, active galaxies and extreme objects are discussed in Section~\\ref{qsos}. ",
        "conclusions": "1. We have presented photometry of a subsample of the ISOCAM ELAIS survey from the N1 and N2 fields.  Our near-IR survey reaches down to $J \\approx 19$ and $K \\approx 17.5$.  All of the $6.7 \\umu$m (LW2) REL=2 sources are identified to these limits, as well as 84 per cent of $15 \\umu$m (LW3) REL=2 sources. The detection efficiencies for REL=3 sources are 88 and 35 per cent at LW2 and LW3 bands, respectively.  2. The near- and mid-IR stars were used, along with stellar models, to perform an accurate new calibration of the ELAIS ISOCAM data at both 6.7 and 15 $\\umu$m.   3. Stars were separated from galaxies using near- to mid-IR colours. At 6.7 $\\umu$m, 80 per cent of the identified ELAIS objects are stars. In contrast, at 15 $\\umu$m, 80 per cent of the near-IR identified ELAIS sources are galaxies.   4. Only one third of LW3 galaxies are also detected in LW2, while two thirds of LW2 galaxies are seen in LW3. The mid-IR survey as a whole mainly detects late type spiral galaxies and starbursts. The faintest population of these is missed by the LW2 filter. The few objects missed by the longer mid-IR filter are most probably early type galaxies.  Simple arguments indicate that typical redshifts of the sample seen with both mid-IR bands are $z \\leqslant 0.2$.   5. We have presented several colour-colour plots useful in studying the relative emission strengths of stellar, PAH, and warm dust components in galaxies and we discuss galaxy classification and star formation properties using the diagrams.  In a $[15/2.2]$ vs.\\ $[6.7/2.2]$ plot the Hubble type of a galaxy can be roughly estimated from its position along the diagonal ($[6.7/15] = 1$), which is a measure of the proportion of ISM in the galaxy.   Of the near-IR-identified galaxies detected with both mid-IR filters, 75 per cent fall in the Scd-group. However, some of these might be earlier morphological types with significant nuclear star formation.  6. In the same $[15/2.2]$ vs.\\ $[6.7/2.2]$ plot the quiescent galaxies fall on the diagonal (where $[15/6.7] \\approx 1$) with increasing star formation activity raising the galaxies above the one-to-one curve.  The ELAIS galaxies are found to have significant star-formation, as indicated by the $[6.7/15]$ tracer ($f_{\\nu}(6.7\\umu{\\rm m}) / f_{\\nu}(15\\umu{\\rm m}) = 0.67 \\pm 0.27$) as well as by estimates from published relations between mid-IR luminosity and SFR. Redshift information and resolved imaging is however needed to better quantify SFRs and to decide whether the ELAIS galaxies are powered by strong nuclear starbursts or otherwise high star formation activity in the disk.  7. In quiescent galaxies, as indicated by their $[60/100]$ IRAS colour, $[6.7/15]$  remains very constant.  These are also the galaxies where the classification of galaxies using NIR/MIR ratios works the best. The MIR ratio starts to drop at hotter $[60/100]$. Using NIR/MIR colours we find support for the view that both the increase of $15\\umu$m emission and an apparent depletion of emission at $6.7\\umu$m are responsible for the effect.  At these higher $[60/100]$ levels, both $[6.7/15]$ and $[2.2/15]$ ratios (anti)correlate well with the $[60/100]$ activity level indicator, thus making them useful tracers of star-formation.  8. The ELAIS survey covered here detects several active galactic nuclei. By selecting objects using a `KX-method' (considering optical to near-IR properties only) we pick out sources from our catalogue, whose mid-IR fluxes are consistent with the objects being AGN/QSOs.    \\begin{table*} \\begin{center} \\small % \\caption{The near-IR sample of ELAIS galaxies detected with both ISOCAM filters.  The galaxies are ordered with decreasing $15 \\umu$m flux. Columns 7 and 9 labeled `R' refer to the REL parameter. The coordinates (J2000) are from the NIR data. The bright galaxies A to E from Fig.~\\ref{nir-early} are indicated; `fir' and `vla' indicate that the objects has been detected in the 90 $\\umu$m ELAIS survey (Efstathiou \\ea 2000) and a VLA followup survey (Ciliegi \\ea 1999); `q1' and `q2' indicate a confirmed and potential quasar, respectively, as discussed in Section~\\ref{qsos}. } \\label{table1} \\begin{tabular}{lccccccccccl} \\hline &  RA & DEC & $J$ mag & $J$-err& $K$ mag & $K$-err & $S_{15}$ (mJy) & R & $S_{6.7}$ (mJy) & R & {Notes} \\\\ \\hline 1  &  16 37 34.4 & +40 52 08 & 13.22  &  0.02 &  12.20  &  0.02  &  34.3  &   2  &  37.3  &   2 & `C', vla  \\\\ 2  &  16 35 07.9 & +40 59 29 & 12.65  &  0.01 &  11.29  &  0.01  &  23.0  &   2  &  22.2  &   2 & `B', fir, vla \\\\ 3  &  16 34 01.8 & +41 20 52 & 13.06  &  0.01 &  12.03  &  0.02  &  18.5  &   2  &  11.8  &   2 & `E', fir, vla\\\\ 4  &  16 37 29.3 & +40 52 49 & 12.66  &  0.01 &  11.56  &  0.01  &  12.9  &   2  &  18.9  &   2 & `D', fir, vla  \\\\ 5  &  16 35 25.2 & +40 55 43 & 14.14  &  0.03 &  12.99  &  0.03  &   8.8  &   2  &   9.0  &   2 & fir, vla \\\\ 6  &  16 37 05.1 & +41 31 55 & 15.26  &  0.04 &  14.02  &  0.04  &   8.5  &   2  &   3.8  &   2 & fir, vla \\\\ 7  &  16 33 59.1 & +40 53 04 & 14.80  &  0.04 &  13.57  &  0.04  &   7.6  &   2  &   5.3  &   2 & vla\\\\ 8  &  16 36 08.1 & +41 05 08 & 15.51  &  0.03 &  13.85  &  0.04  &   7.6  &   2  &   2.7  &   2 & vla \\\\ 9  &  16 35 06.1 & +41 10 38 & 15.48  &  0.04 &  14.10  &  0.03  &   6.2  &   2  &   4.9  &   2 & fir, vla \\\\ 10  &   16 35 46.9 & +40 39 01 & 15.35  &  0.04 &  14.18  &  0.05  &   5.6  &   2  &   3.0  &   2 & vla \\\\ 11  &   16 36 45.0 & +41 51 32 & 14.23  &  0.04 &  13.15  &  0.05  &   4.9  &   2  &   1.7  &   2 & \\\\ 12  &   16 36 13.6 & +40 42 30 & 13.44  &  0.02 &  12.42  &  0.02  &   4.5  &   2  &   3.1  &   2 & vla \\\\ 13  &   16 36 07.6 & +40 55 48 & 15.83  &  0.05 &  14.30  &  0.05  &   4.1  &   2  &   1.3  &   2 & fir, vla \\\\ 14  &   16 37 08.1 & +41 28 56 & 15.26  &  0.04 &  13.91  &  0.04  &   3.6  &   2  &   2.1  &   3 & vla \\\\ 15  &   16 37 31.4 & +40 51 56 & 15.47  &  0.04 &  14.13  &  0.06  &   3.3  &   2  &   1.7  &   2 & fir, vla \\\\ 16  &   16 34 14.2 & +41 03 19 & 14.79  &  0.03 &  13.98  &  0.04  &   2.9  &   2  &   1.5  &   3 & vla \\\\ 17  &   16 34 12.0 & +40 56 53 & 17.59  &  0.11 &  15.43  &  0.08  &   2.9  &   2  &   1.4  &   2 & vla\\\\ 18  &   16 37 20.5 & +41 11 21 & 12.92  &  0.01 &  11.67  &  0.01  &   2.9  &   2  &   2.7  &   2 & `A' \\\\ 19  &   16 36 09.7 & +41 00 18 & 15.89  &  0.05 &  14.45  &  0.07  &   2.9  &   2  &   1.5  &   2 & vla \\\\ 20  &   16 35 19.2 & +40 55 57 & 16.27  &  0.07 &  14.90  &  0.07  &   2.7  &   3  &   0.9  &   3 & vla \\\\ 21  &   16 38 51.9 & +41 10 53 & 15.80  &  0.08 &  14.27  &  0.07  &   2.7  &   2  &   2.1  &   3 & \\\\ 22  &   16 34 23.9 & +40 54 10 & 15.94  &  0.06 &  14.27  &  0.07  &   2.6  &   2  &   1.6  &   2 & vla \\\\ 23  &   16 37 16.8 & +40 48 26 & 14.99  &  0.03 &  13.97  &  0.04  &   2.6  &   2  &   2.1  &   2 & vla \\\\ 24  &   16 35 34.0 & +40 40 25 & 18.02  &  0.15 &  16.06  &  0.15  &   2.3  &   2  &   2.1  &   2 & q2 \\\\ 25  &   16 34 49.6 & +41 20 50 & 15.62  &  0.04 &  ...    &  ...   &   2.2  &   2  &   1.6  &   2 & \\\\ 26  &   16 35 31.2 & +41 00 28 & 17.87  &  0.17 &  16.34  &  0.17  &   2.2  &   3  &   0.8  &   3 & q1 \\\\ 27  &   16 36 40.0 & +40 55 38 & 16.30  &  0.06 &  14.96  &  0.06  &   1.9  &   2  &   1.3  &   2 & \\\\ 28  &   16 34 43.0 & +41 09 49 & 15.60  &  0.05 &  14.30  &  0.06  &   1.9  &   3  &   1.7  &   2 & \\\\ 29  &   16 36 15.2 & +41 19 12 & 16.52  &  0.08 &  15.07  &  0.07  &   1.7  &   3  &   0.9  &   3 & \\\\ \\hline \\end{tabular} \\end{center} \\end{table*}"
    },
    "0208/astro-ph0208138_arXiv.txt": {
        "abstract": "The accelerating expansion of the universe suggests that an unknown component with strongly negative pressure, called dark energy, currently dominates the dynamics of the universe.  Such a component makes up $\\sim70\\%$ of the energy density of the universe yet has not been predicted by the standard model of particle physics.  The best method for exploring the nature of this dark energy is to map the recent expansion history, at which Type Ia supernovae have proved adept.  We examine here the depth of survey necessary to provide a precise and qualitatively complete description of dark energy. Realistic analysis of parameter degeneracies, allowance for natural time variation of the dark energy equation of state, and systematic errors in astrophysical observations all demonstrate the importance of a survey covering the full range $0<z\\la2$ for revealing the nature of dark energy. ",
        "introduction": "The discovery of the acceleration of the expansion of the universe through the Type Ia supernova distance-redshift relation is a major development in cosmology~\\cite{SCP,HIGHZ}. Exploring the expansion history of the universe is a key aim of cosmology, producing literally a textbook picture of the universe. Furthermore, such a map provides key clues to the underlying physics, independent of whether this is dark energy, higher dimensions, or an altered theory of gravitation \\cite{Li02}.  In its interpretation as arising from a universal vacuum, or dark, energy, such a component would comprise some 70\\% of the critical density, be unclustered on subhorizon scales, and possess a substantially negative equation of state (EOS) $w=p/\\rho\\lesssim -0.6$ \\cite{w6}.  While these properties are unexpected from the standard model of particle physics, it has been suggested that they can be motivated by a number of fundamental theories \\cite{WA,RMP}. Dark energy thus poses a crucial mystery to unravel for the fields of high energy physics, cosmology, and gravitation.  \\begin{figure}[!ht] \\begin{center} \\epsfig{file=zeq.ps, height=3.3in, width=3.4in} \\caption{The epochs of equality between the dark energy density and matter and of transition from acceleration to deceleration are plotted vs.~dark energy equation of state.  The positively slanted hatching denotes the accelerating phase; the negatively slanted hatching shows when the dark energy density dominates over the matter density. Despite these both occurring below redshift $z\\approx0.7$, dark energy can be probed to much higher redshift. } \\label{zeq} \\end{center} \\end{figure}  Supernovae studies, which first provided the evidence for the acceleration, are well suited for elucidating the nature of the dark energy~\\cite{triangle,HT}. One experiment being designed specifically to probe the accelerating universe using supernovae is the Supernova/Acceleration Probe (SNAP \\cite{snap}). At an initial theoretical glance, the redshift range over which this exploration is most easily done seems simple to understand: the energy density dominance and dynamical influence (accelerating power) of dark energy enters at redshifts $z\\lesssim 0.7$ (see Fig.~\\ref{zeq}).  Moreover, an idealized perturbative, or Fisher matrix, calculation shows that the ``sweet spot'' of sensitivity to the equation of state $w$ lies at $z\\approx0.3$~\\cite{HT,WA,HS}. So why are observations at $z>1$ necessary for characterizing the dark energy?  The answer lies in the breakdown of the ideal case: \\begin{itemize} \\item Cosmological degeneracies \\item Dark energy model degeneracies \\item Systematic errors \\end{itemize} The required survey depth depends on the rigor of our scientific investigation, how much we are willing to assume about the other parameters entering into the determination of the dark energy equation of state.  One could estimate a false precision without knowing how accurate, i.e.~biased, the result is.  We label this blind trust by three heresies\\footnote{The authors in no way advocate burning at the stake.}, and here aim to demonstrate their insidious effects through simple illustrations rather than mathematical arguments. ",
        "conclusions": "The discussions and illustrations presented here show that expectations based on oversimplified cosmology, physics, and astrophysics prove insufficient and misleading for understanding how to probe dark energy. Could we detect dark energy with measurements at $z<1$?  Assuredly -- we already have through the supernova method.  Could we reliably distinguish its equation of state from that of a cosmological constant?  Possibly -- wide field ground based surveys, possibly together with higher redshift Hubble Space Telescope observations, could well give indications of this, though not necessarily definitive ones.  Could we see the critical evidence of time variation in the equation of state that sets us on the path of a fundamental theory?  No.  For that we required detailed observations out to $z\\approx1.5-2$ and control of systematics.  In the realistic view, one clearly appreciates the need for a precision survey reaching out to $z_{max}\\approx 1.5-2$. More rigorous Monte Carlo simulations \\cite{sys} implementing a variety of systematic error, cosmology, and dark energy models bear out this conclusion.  \\vspace{0.2in}"
    },
    "0208/astro-ph0208554_arXiv.txt": {
        "abstract": "To understand the formation of stellar groups, one must first document carefully the birth pattern within real clusters and associations. In this study of Taurus-Auriga, we combine pre-main-sequence ages from our own evolutionary tracks with stellar positions from observational surveys. Aided by the extensive, millimeter data on the molecular clouds, we develop a picture of the region's history. Star formation began, at a relatively low level and in a spatially diffuse manner, at least $1\\times 10^7$~yr in the past. Within the last few million years, new stars have been produced at an accelerating rate, almost exclusively within a confined group of striated cloud filaments.  The gas both inside and around the filaments appears to be in force balance. Thus, the appearance of the filaments is due to global, quasi-static contraction of the parent cloud material. Gravity drives this contraction and shock dissipation mediates it, but the internal motion of the gas does not appear to be turbulent. The accelerating nature of recent star formation means that the condensation of cloud cores is a threshold phenomenon, requiring a minimum background density. Other, nearby cloud regions, including Lupus and Chamaeleon, contain some locales that have attained this density, and others that have not. In the latter, we find extensive and sometimes massive molecular gas that is still devoid of young stars. ",
        "introduction": "One basic characteristic of stars is that they form not as isolated objects, but in populous groups within molecular clouds and cloud complexes. This observational fact raises other basic issues of a theoretical nature. What supports a massive cloud against its self-gravity {\\it before} the production of an interior stellar group? How do individual dense cores, each capable of forming single and binary stars, arise within this kinetic, and perhaps turbulent, medium? Which properties of the large parent cloud determine whether it will spawn a bound cluster, T association, or expanding OB association?  Answering all these questions will require a broader and deeper understanding than is currently at hand, one that connects the birth of individual stars to the growth and evolution of clouds on a multi-parsec scale. Any such theory, if it is to be quantitative, must be based on detailed information concerning the earliest stages of existing groups. That is, we should first ascertain the actual pattern of stellar births in observed clusters and associations. To address this issue, we began by constructing a set of pre-main-sequence evolutionary tracks (Palla \\& Stahler 1999; hereafter Paper I). These covered masses from 0.1 to 8 $\\Msun$, \\ie, from close to the brown-dwarf regime to the upper mass limit of the pre-main-sequence phase itself (Palla \\& Stahler 1990). The tracks can be used to assign contraction ages to any young star that has reliable values of effective temperature and luminosity. A compilation of ages then represents the star formation history of the region of interest.  Paper I applied this method to the Orion Nebula Cluster. Utilizing the extensive database of Hillenbrand (1997), we found that star formation began at a relatively low rate at least $10^7$~yr ago, then increased markedly within the last $2\\times 10^6$~yr. In a subsequent contribution (Palla \\& Stahler 2000; hereafter Paper II), we examined seven additional, nearby groups. Most exhibited a similar pattern of activity as Orion: a slow rise followed by rapid acceleration through the current epoch.  This result has recently been questioned by Hartmann (2001), who claims that observational errors in effective temperatures and luminosities are so large that they preclude the inference of any detailed history at all. Hartmann further contends that the data from T associations are consistent with a rapid burst of star formation in the recent past, and that apparently older stars have simply been assigned erroneously low luminosities. This view is contradicted, however, by the HR diagrams in Papers I and II. If the observational errors were indeed large enough to ``age\" a substantial portion of stars, they would also give spuriously {\\it high} luminosities to an equal number of other objects. Many of these would appear above the birthline in the HR diagram, which is not the case. In his statistical analysis, Hartmann avoids this difficulty by ignoring the birthline and allowing arbitrarily high luminosities for his sample population (see his equation (2) and Figure 2).  Assuming that our findings are robust, many systems are increasing their production of stars to the present time. This intriguing result naturally prompts additional questions. Where are the clouds that are not yet actively forming stars? Within a given cloud, what ends the acceleration?  A partial answer to the second point is available from the earlier studies. In the volume surveyed by Hillenbrand, \\ie, within several pc of the Trapezium and in front of the massive, Orion~A cloud, molecular-line studies show that there are no dense cores (Bergin 1996). Hence, new stars are no longer forming in this region, and their production rate must overturn sharply. We could not resolve this rapid decline with our technique. However, one system from Paper II did show a decline, albeit over a longer time scale. This was Upper~Scorpius, an OB association with very low gas content (de Geus, Bronfman, \\& Thaddeus 1990).  In an interesting, related study, Dolan \\& Mathieu (2001) have documented the star formation history of the $\\lambda$~Orionis region. They found the star formation rate to be accelerating in unperturbed cloud material well removed from the massive stars, but to have fallen to a low level closer to these objects, where the gas has largely been dispersed.  The manner in which accelerating star formation ends within a low-mass T~association is less clear. The results concerning $\\lambda$~Orionis illustrate, in any case, the value of probing both the temporal and {\\it spatial} pattern of activity. In the present paper, we apply this philosophy to the Taurus-Auriga association. This region has, of course, been thoroughly studied over many years. Stars have been catalogued through a number of deep surveys covering the optical, infrared, and X-ray regimes (Kenyon \\& Hartmann 1995; Brice\\~no et al. 1998; Luhman 2000; K\\\"onig, Neuhauser, \\& Stelzer 2001). The stellar population is located within a cloud complex that has itself been mapped systematically in a number of CO isotopes (Kawamura et al. 1998; Onishi et al. 1996; Dame, Hartmann, \\& Thaddeus 2001), and more sparsely in other, high-density tracers (Benson \\& Myers 1989; Pound \\& Bally 1991; Saito et al. 2001). Thus, we inherit a rich trove of data for assessing the star formation pattern.  We describe this pattern in \\S 2, below, and show how activity has become centrally concentrated with time. We then focus on those stars {\\it outside} the central region, to confirm that they are bona fide members of the association. Section 3 discusses the implications of our findings with regard to the basic questions posed earlier.  We also compare our results with recently proposed dynamical models of stellar group formation. Finally, in \\S 4, we indicate fruitful directions for research in the near future. ",
        "conclusions": "Our investigation of Taurus-Auriga has yielded a much more detailed picture of star formation in this region than was previously available. We find that stellar births occur over a broad area until the cloud's own contraction yields a region of sufficiently high density to induce much more rapid formation. This connection between spatial and temporal trends adds further credence to our pre-main-sequence ages. It also encourages us to apply the same technique to other sites, in the hope of elucidating fundamental issues concerning stellar groups.  The Taurus-Auriga findings already give us further insight concerning the origin of such associations. Since global contraction has led to the current, active phase of stellar production, the cloud gas must have been more rarefied in the past. At that time, the surface density was everywhere below the star formation threshold. However, with only a modest decrease from the current $N_H$-value, self-shielding would not have been effective, and the bulk of the gas would have been HI. An older study by Lucke (1978) may be pertinent in this regard. Lucke mapped the $B-V$ color excess toward many stars over a large portion of the Northern sky. He surmised that the Taurus-Auriga molecular clouds are actually linked to the Perseus complex through a ``supercloud\" of HI gas. Whether or not such a coherent structure exists, Lucke's study reminds us that the Taurus-Auriga clouds undoubtedly appeared very different prior to the onset of star formation. The contours in the leftmost panels of our own Figure 2, which implicitly assume constancy of the cloud morphology, should therefore be viewed only as a crude schematic.  On the other major question, that of cloud dispersal and the truncation of stellar births, our study has less to contribute. Most of the 60 stars for which we have no estimates are deeply embedded and located in the central filaments. Thus, accelerating star formation will undoubtedly continue for several Myr. Further progress in understanding the general issue of the cutoff will come by observing associations slightly more evolved than Taurus-Auriga, to witness firsthand expansion of the parent cloud. The discovery of TW~Hydra, $\\eta$~Chamaeleon, and related groups is of considerable interest, but these associations are already gas-free. We note finally that expansion will cause the H$_2$ gas to become HI, in the reverse process as the prior contraction. The sighting of discrete HI clouds with embedded T~Tauri and post-T~Tauri stars would therefore constitute an important advance."
    },
    "0208/astro-ph0208281_arXiv.txt": {
        "abstract": "Orbital, spin and astrometric parameters of the millisecond pulsar PSR~J0621+1002 have been determined through six years of timing observations at three radio telescopes. The chief result is a measurement of the rate of periastron advance, $\\dot{\\omega}=0\\fdg 0116 \\pm 0\\fdg 0008$ yr$^{-1}$. Interpreted as a general relativistic effect, this implies the sum of the pulsar mass, $m_1$, and the companion mass, $m_2$, to be $M=m_1+m_2=2.81\\pm 0.30$\\,M$_\\odot$. The Keplerian parameters rule out certain combinations of $m_1$ and $m_2$, as does the non-detection of Shapiro delay in the pulse arrival times. These constraints, together with the assumption that the companion is a white dwarf, lead to the 68\\% confidence maximum likelihood values of $m_1=1.70^{+0.32}_{-0.29}$\\,M$_\\odot$ and $m_2=0.97^{+0.27}_{-0.15}$\\,M$_\\odot$ and to the 95\\% confidence maximum likelihood values of $m_1=1.70^{+0.59}_{-0.63}$\\,M$_\\odot$ and $m_2=0.97^{+0.43}_{-0.24}$\\,M$_\\odot$. The other major finding is that the pulsar experiences dramatic variability in its dispersion measure (DM), with gradients as steep as 0.013 pc\\,cm$^{-3}$\\,yr$^{-1}$. A structure function analysis of the DM variations uncovers spatial fluctuations in the interstellar electron density that cannot be fit to a single power law, unlike the Kolmogorov turbulent spectrum that has been seen in the direction of other pulsars. Other results from the timing analysis include the first measurements of the pulsar's proper motion, $\\mu = 3.5\\pm 0.3$\\,mas\\,yr$^{-1}$, and of its spin-down rate, $dP/dt=4.7\\times10^{-20}$, which, when corrected for kinematic biases and combined with the pulse period, $P=28.8$\\,ms, gives a characteristic age of $1.1\\times10^{10}$\\,yr and a surface magnetic field strength of $1.2\\times10^{9}$\\,G. ",
        "introduction": "Recent pulsar searches have uncovered a new class of binary pulsars. Most millisecond pulsars are in nearly circular binary orbits with low-mass He white dwarfs, comprising the class of so-called low-mass binary pulsars (LMBPs). The new category of systems, the intermediate-mass binary pulsars (IMBPs), have companion stars with masses above 0.45\\,M$_\\odot$. Since helium flash occurs at a core mass of 0.45\\,M$_\\odot$, the stars must be CO or ONeMg white dwarfs.  About eight IMBPs are currently known. Observationally, they are distinguished by high mass functions, $f_1 > 0.015$\\,M$_\\odot$; by moderately spun-up pulse periods, 10\\,ms $< P <$ 200\\,ms; and by orbital eccentricities that are small, $e < 10^{-2}$, but somewhat higher than those of the LMBPs \\cite{cnst96, clm+01, eb01b}.  This paper reports on timing measurements of PSR~J0621+1002, an IMBP with a 28.8-ms spin period in an 8.3-day orbit. The main goal of our observations was to determine the pulsar and companion masses through measurement of post-Newtonian orbital parameters, particularly the rate of apsidal motion. Measuring apsidal motion is challenging in white dwarf-pulsar binaries because their often extremely small eccentricities---values of $10^{-5}$ are typical---make it difficult to determine the angle of periastron, and hence to measure periastron advance. A comparatively high eccentricity---still only $e = 0.0025$---made the detection of apsidal motion feasible for PSR~J0621+1002.  \\begin{table*}[t] \\caption{Pulse Timing Parameters of PSR~J0621+1002\\label{tab:param}\\tablenotemark{a}} \\centerline{ \\begin{tabular}{ll} \\hline \\hline\\omit\\vrule height3pt width0pt\\\\ \\multicolumn{2}{c}{Measured Parameters} \\\\[3pt] \\hline\\omit\\vrule height3pt width0pt\\\\ Right ascension, $\\alpha$ (J2000)\\dotfill       & $06^{\\rm h}21^{\\rm m}22\\fs11108(3)$\\\\ Declination, $\\delta$ (J2000)\\dotfill           & $+10\\arcdeg02\\arcmin38\\farcs741(2)$\\\\ Proper motion in R.A., $\\mu_{\\alpha}=\\dot{\\alpha}\\cos\\delta$ (mas yr$^{-1}$)\\dots & 3.5(3)\\\\ Proper motion in Dec., $\\mu_\\delta=\\dot{\\delta}$ (mas yr$^{-1}$)\\dotfill & $-$0.3(9)\\\\ Pulse period, ${P}$ (ms)\\dotfill                        & 28.853860730049(1)\\\\ Period derivative, $\\dot{P}_{\\rm obs}$ ($10^{-20}$)\\dotfill & 4.732(2)\\\\ Epoch (MJD [TDB])\\dotfill                       & 50944.0\\\\ Orbital period, $P_b$ (days)\\dotfill            & 8.3186813(4)\\\\ Projected semi-major axis, $x$ (lt-s)\\dotfill   & 12.0320744(4)\\\\ Eccentricity, $e$\\dotfill                       & 0.00245744(5)\\\\ Epoch of periastron\\tablenotemark{b}, $T_0$ (MJD [TDB])\\dotfill & 50944.75683(4)\\\\ Longitude of periastron\\tablenotemark{b}, $\\omega$ (deg)\\dotfill & 188.816(2)\\\\ Periastron rate of change, $\\dot{\\omega}$ (deg yr$^{-1}$)\\dotfill & 0.0116(8)\\\\ Dispersion measure\\tablenotemark{c}, DM (pc\\,cm$^{-3}$)\\dotfill & 36.6010(6)\\\\ \\hline\\omit\\vrule height3pt width0pt\\\\ \\multicolumn{2}{c}{Measured Upper Limits} \\\\[3pt] \\hline\\omit\\vrule height3pt width0pt\\\\ Parallax (mas)\\dotfill                          & $<2.7$ \\\\ Pulse period second derivative, $\\ddot{P}$ (s$^{-1}$) \\dotfill                        & $<4\\times 10^{-30}$ \\\\ Orbital period rate of change, $\\dot{P_b}$  \\dotfill  &  $<5\\times 10^{-12}$  \\\\ Orbital axis rate of change, $\\dot{x}$ \\dotfill &  $<1.5\\times 10^{-14}$   \\\\ \\hline\\omit\\vrule height3pt width0pt\\\\ \\multicolumn{2}{c}{Derived Parameters} \\\\[3pt] \\hline\\omit\\vrule height3pt width0pt\\\\ Mass function, $f_1$ (M$_\\odot$)\\dotfill        & 0.027026841(4)\\\\ Total mass, $M$ (M$_\\odot$)\\dotfill             & $2.81\\pm0.30$\\\\ Pulsar mass, $m_{1}$ (M$_\\odot$)\\dotfill        & $1.70^{+0.32}_{-0.29}$\\\\ Companion mass, $m_{2}$ (M$_\\odot$)\\dotfill     & $0.97^{+0.27}_{-0.15}$\\\\ Characteristic age (yr)\\dotfill & $1.1\\times10^{10}$ \\\\ Surface magnetic field strength (Gauss)\\dotfill & $1.2\\times10^{9}$ \\\\ Total proper motion, $\\mu$ (mas yr$^{-1}$)\\dotfill  &  $3.5(3)$ \\\\ \\hline\\omit\\vrule height3pt width0pt\\\\ \\multicolumn{2}{l}{\\scriptsize $^a$Values in parentheses are $1\\sigma$ uncertainties (68\\% confidence) in the last digit quoted.}\\\\ \\multicolumn{2}{l}{\\scriptsize $^b\\omega$ and $T_0$ are highly covariant. Observers should use $\\omega = 188{\\fdg} 815781$ and $T_0 = 50944.756830176$.}\\\\ \\multicolumn{2}{l}{\\scriptsize $^c$The DM varies. The value here is the constant term in an 18-term polynomial expansion (see \\S\\ref{subsec:dm}).}\\\\ \\end{tabular} } \\end{table*}  Mass measurements in white dwarf-pulsar systems can be used to constrain theories of binary evolution. The LMBPs and the IMBPs share roughly similar histories. They both originate when a giant star transfers mass onto a neutron star companion, resulting in a spun-up millisecond pulsar and a white dwarf. The histories of LMBPs and IMBPs differ, however, in the details of mass exchange. For LMBPs, it is a stable process that occurs when the giant swells past its Roche lobe \\cite{pk94}. For IMBPs, it is thought to be an unstable transfer that operates via common envelope  evolution \\cite{vdh94} or super-Eddington accretion (Taam, King, \\& Ritter 2000)\\nocite{tkr00}. Other pulsar binaries, such as double neutron star systems, evolve in still other ways. One way to test binary evolution scenarios is by comparing the masses of neutron stars in different classes of binary systems to infer differences in amounts of mass transferred. (For a review of pulsar mass measurements, see Thorsett \\& Chakrabarty 1999\\nocite{tc99}.)  The first year of timing observations of PSR~J0621+1002 was discussed in Camilo et al. (1996)\\nocite{cnst96}, which reported the pulsar's Keplerian orbital elements, position, and spin period.  With five additional years of timing data, we have measured the apsidal motion, spin-down rate, and proper motion of the pulsar, and we have derived significant constraints on Shapiro delay. We also have found sharp variations in the dispersion measure (DM), which we use to analyze turbulence in the interstellar medium (ISM) in the direction of PSR~J0621+1002. ",
        "conclusions": "We have found substantial constraints on the masses of PSR~J0621+1002 and its orbital companion. The pulsar mass is found to be $m_1=1.70^{+0.32}_{-0.29}$\\,M$_\\odot$ (68\\% confidence). The lower end of this uncertainty range is near the canonical pulsar mass of 1.35\\,M$_\\odot$, but the mass may also be several tenths of a solar mass higher, allowing the possibility that a substantial amount of material accreted onto the neutron star during the evolution of the system. The mass of the secondary, $m_2=0.97^{+0.27}_{-0.15}$\\,M$_\\odot$ (68\\% confidence), makes it one of the heaviest known white dwarfs orbiting a pulsar.  Can the mass measurements be improved by continued timing observations? Unfortunately, post-Keplerian effects beyond those considered in this paper, such as orbital period decay, and gravitational redshift and time dilation, will not be detectable in the timing data for PSR~J0621+1002 in the foreseeable future, so any improvement must come about through tighter measurements of $\\dot{\\omega}$ and Shapiro delay.  The uncertainty in the measurement of the total mass, $M=2.81\\pm0.30$\\,M$_\\odot$, scales linearly with the uncertainty of $\\dot{\\omega}$, which in turn is inversely proportional to the time span over which data are collected. This has a simple explanation: the longer the time span of the observations, the more $\\omega$ shifts, and so the easier it is to measure $\\dot{\\omega}$. The highest precision data used in this work---the annual Arecibo campaigns---were collected over two years. A similar campaign carried out several years in the future would shrink the uncertainty in $M$ by a factor of a few.  The pulsar and companion masses were further constrained by Shapiro delay. The precision of the Shapiro delay measurement (or limit) has no dependence on data span length, so its uncertainty is reduced only as $n^{-1/2}$, where $n$ is the number of observations.  At best, a modest improvement could be made with existing radio telescope resources."
    },
    "0208/astro-ph0208248_arXiv.txt": {
        "abstract": "X-ray absorption and emission lines now serve as powerful diagnostics of the outflows from active galaxies. Detailed X-ray line studies of outflows have recently been enabled for a significant number of active galaxies via the grating spectrometers on \\chandra\\ and \\xmm. We will review some of the recent X-ray findings on active galaxy outflows from an observational perspective. We also describe some future prospects.  X-ray absorption lines from H-like and He-like ions of C, N, O, Ne, Mg, Al, Si, and S are often seen. A wide range of ionization parameter appears to be present in the absorbing material, and inner-shell absorption lines from lower ionization ions, Fe~L-shell lines, and Fe~M-shell lines have also been seen. The X-ray absorption lines are typically blueshifted relative to the systemic velocity by a few hundred km~s$^{-1}$, and they often appear kinematically consistent with UV absorption lines of C~{\\sc iv}, N~{\\sc v}, and H~{\\sc i}. The X-ray absorption lines can have complex profiles with multiple kinematic components present as well as filling of the absorption lines by emission-line photons. A key remaining uncertainty is the characteristic radial location of the outflowing gas; only after this quantity is determined will it be possible to calculate reliably the amount of outflowing gas and the kinetic luminosity of the outflow. ",
        "introduction": "Outflows are observed to be ubiquitous in active galactic nuclei (AGN), being seen in objects spanning a range of $\\sim 10,000$ in luminosity. They have been studied in the most detail via observations of ultraviolet (UV) resonance lines from moderately ionized gas. In luminous Broad Absorption Line quasars (BALQSOs), outflows are observed to reach velocities up to a few $10^4$~km~s$^{-1}$, and they subtend $\\approx$~10--30\\% of the sky as viewed from the central source. In lower luminosity Seyfert galaxies, outflows are observed $\\simgt 50$\\% of the time although they have velocities up to only $\\approx 10^3$~km~s$^{-1}$. These outflows are a major component of the nuclear environment, and they may carry a significant fraction of the accretion power. They may also be important in regulating the growth of the black hole and its host galaxy (e.g., Silk \\& Rees 1998; Fabian 1999) as well as in injecting matter, energy, and magnetic fields into the intergalactic medium (e.g., Turnshek 1988; Wu, Fabian, \\& Nulsen 2000; Furlanetto \\& Loeb 2001; Elvis et~al. 2002). The observed outflows are photoionized by the radiation from the central source, and they are probably driven by radiation pressure. Despite their ubiquity and importance, their physical location and origin in the AGN system remain unclear; outflows may arise from winds driven off the surface of an accretion disk (e.g., Murray et~al. 1995; Proga, Stone, \\& Kallman 2000; Elvis 2000), a dusty torus (e.g., Voit, Weymann, \\& Korista 1993; Krolik \\& Kriss 1995), or perhaps stars in the nucleus (e.g., Scoville \\& Norman 1995; Netzer 1996).  \\begin{figure}[t!] \\includegraphics[height=5 in,width=4.5 in]{brandt_fig01.eps} \\caption{Part of the 10.4-day \\chandra\\ HETGS spectrum of the Seyfert galaxy NGC~3783. Marked are the large number of detected absorption lines as well as several emission lines. The lines are marked at their expected wavelengths in the rest frame of NGC~3783; the blueshifts of the absorption lines are noticeable. In total, more than 140 spectral features are detected in the X-ray spectrum of NGC~3783. Adapted from Kaspi et~al. (2002).} \\end{figure}  Ionized absorption in the X-ray band has been intensively studied in bright, low-redshift AGN for over a decade (e.g., the ``warm absorbers'' in Seyfert galaxies; Reynolds 1997; George et~al. 1998). The luminous X-ray source in the nucleus acts as a ``flashlight'' allowing observers to ``X-ray'' material along the line of sight. However, prior to the launches of \\chandra\\ and \\xmm, such investigations were limited by a lack of spectral resolution. Over the past three years, the X-ray grating spectrometers on these two missions\\footnote{The relevant instruments are the \\chandra\\ High-Energy Transmission Grating Spectrometer (HETGS; C.R. Canizares et~al., in preparation), the \\chandra\\ Low-Energy Transmission Grating Spectrometer (LETGS; Brinkman et~al. 1997), and the \\xmm\\ Reflection Grating Spectrometer (RGS; den~Herder et~al. 2001).} have enlarged the number of spectral features available for study by a factor of $\\sim 50$ (see Fig.~1; from 2--3 to more than 140). They have improved the velocity resolution available to observers from $\\sim 15,000$~km~s$^{-1}$ to $\\sim 400$~km~s$^{-1}$. They have thereby provided qualitatively new information on the physical conditions, kinematics, and geometry/location of the absorbing material.  At present, efficient grating X-ray spectroscopy is possible only for $\\approx 20$ bright, low-redshift (mainly $z<0.1$) AGN, mainly Seyfert galaxies. Even for these, the required observation lengths are typically $\\simgt$~1--2~days; the spectrum of NGC~3783 shown in Fig.~1 required a 10.4-day exposure with \\chandra. Of necessity, the discussion below applies predominantly to this fairly small sample of objects. Highly luminous and distant quasars, for example, may have significantly different X-ray absorption properties. Furthermore, the discussion below will be closely tied to the X-ray observations, without detailed descriptions of theoretical models. For further information on theoretical models, the reader should consult one of the current reviews (e.g., Netzer 2001; Krolik 2002; and references therein).   \\begin{figure}[t!] \\includegraphics[height=2.5 in,width=5.7 in]{brandt_fig02.eps} \\caption{The 10.4-day \\chandra\\ HETGS spectrum of the Seyfert galaxy NGC~3783 shown in $EF_{\\rm E}$ vs. energy format. The insert focuses on the spectrum below 1~keV. The different types of observed spectral features are labeled. Adapted from Kaspi et~al. (2002).} \\end{figure} ",
        "conclusions": ""
    },
    "0208/astro-ph0208295_arXiv.txt": {
        "abstract": "Primordial fluctuations in the cosmic density are usually assumed to take the form of a Gaussian random field that evolves under the action of gravitational instability. In the early stages, while they have low amplitude, the fluctuations grow linearly. During this phase the Gaussian character of the fluctuations is preserved. Later on, when the fluctuations have amplitude of order the mean density or larger, non-linear effects cause departures from Gaussianity. In Fourier space, non-linearity is responsible for coupling Fourier modes and altering the initially random distribution of phases that characterizes Gaussian initial conditions. In this paper we investigate some of the effects of {\\em quadratic} non-linearity on basic statistical properties of cosmological fluctuations. We demonstrate how this form of non-linearity can affect asymptotic properties of density fields such as homogeneity, ergodicity, and behaviour under smoothing. We also show how quadratic density fluctuations give rise to a particular relationship between the phases of different Fourier modes which, in turn, leads to the generation of a non-vanishing bispectrum. We thus elucidate the relationship between higher--order power spectra and phase distributions. ",
        "introduction": "In most popular versions of the gravitational instability model for the origin of cosmic structure, particularly those involving cosmic inflation (Guth 1981; Guth \\& Pi 1982), the initial fluctuations that seeded the structure formation process form a Gaussian random field (Adler 1981; Bardeen et al. 1986).  Gaussian random fields are the simplest fully-defined stochastic processes, which makes analysis of them relatively straightforward. Robust and powerful statistical descriptors can be constructed that have a firm mathematical underpinning and are relatively simple to implement. Second-order statistics such as the ubiquitous power-spectrum (e.g. Peacock \\& Dodds 1996) furnish a complete description of Gaussian fields. They have consequently yielded invaluable insights into the behaviour of large-scale structure in the latest generation of redshift surveys, such as the 2dFGRS (Percival et al. 2001). Important though these methods undoubtedly are, the era of precision cosmology we are now entering requires more thought to be given to methods for both detecting and exploiting departures from Gaussian behaviour.  The pressing need for statistics appropriate to the analysis of non-linear stochastic processes also suggests a need to revisit some of the fundamental properties cosmologists usually assume when studying samples of the Universe. Gaussian random fields have many useful properties. It is straightforward to impose constraints that result in statistically homogeneous fields, for example. Perhaps more relevantly one can understand the conditions under which averages over a single spatial domain are well-defined, the constraint of sample-homogeneity. The conditions under which such fields can be ergodic are also well established. It is known that smoothing Gaussian fields preserves Gaussianity, and so on. These properties are all somewhat related, but not identical. Indeed, as we shall see in Section 4, looking at the corresponding properties of non-linear fields turns up some interesting results and delivers warnings to be careful. Exploring these properties is the first aim of this paper.  Even if the primordial density fluctuations were indeed Gaussian, the later stages of gravitational clustering must induce some form of non-linearity. One particular way of looking at this issue is to study the behaviour of Fourier modes of the cosmological density field. If the hypothesis of primordial Gaussianity is correct then these modes began with random spatial phases. In the early stages of evolution, the plane-wave components of the density evolve independently like linear waves on the surface of deep water. As the structures grow in mass, they interact with other in non-linear ways, more like waves breaking in shallow water. These mode-mode interactions lead to the generation of coupled phases. While the Fourier phases of a Gaussian field contain no information (they are random), non-linearity generates non-random phases that contain much information about the spatial pattern of the fluctuations. Although the  significance of phase information in cosmology is still not fully understood, there have been a number of attempts to gain quantitative insight into the behaviour of phases in gravitational systems. Ryden \\& Gramann (1991), Soda \\& Suto (1992) and Jain \\& Bertschinger (1998) concentrated on the evolution of phase shifts for individual modes using perturbation theory and numerical simulations. An alternative approach was adopted by Scherrer, Melott \\& Shandarin (1991), who developed a practical method for measuring the phase coupling in random fields that could be applied to real data. Most recently Chiang \\& Coles (2000), Coles \\& Chiang (2000), Chiang (2001) and Chiang, Naselsky \\& Coles (2002) have explored the evolution of phase information in some detail.   Despite this recent progress, there is still no clear understanding of how the behaviour of the Fourier phases manifests itself in more orthodox statistical descriptors.  In particular there is much interest in the usefulness of the simplest possible generalisation of the (second-order) power-spectrum, i.e. the (third-order) bispectrum (Peebles 1980; Scoccimarro et al. 1998; Scoccimarro, Couchman \\& Frieman 1999; Verde et al. 2000; Verde et al. 2001; Verde et al. 2002). Since the bispectrum is identically zero for a Gaussian random field, it is generally accepted that the bispectrum encodes some form of phase information but it has never been elucidated exactly what form of correlation it measures. Further possible generalisations of the bispectrum are usually called polyspectra; they include the (fourth-order) trispectrum (Verde \\& Heavens 2001) or a related but simpler statistic called the second-spectrum (Stirling \\& Peacock 1996). Exploring the connection between polyspectra and non-linearly induced phase association is the second aim of this paper.  We investigate both these issues by reference to a particular form of non-Gaussian field that serves both as a kind of phenomenological paradigm and as a reasonably realistic model of non-linear evolution from Gaussian initial conditions. The model involves a field which is generated by a simple quadratic transformation of a Gaussian distribution, hence the term {\\em quadratic} non-linearity. Quadratic fields have been discussed before from a number of contexts (e.g. Coles \\& Barrow 1987; Moscardini et al. 1991; Falk, Rangarajan \\& Srednicki 1993; Luo \\& Schramm 1993; Luo 1994; Gangui et al. 1994; Koyoma, Soda \\& Taruya 1999; Peebles 1999a,b; Matarrese, Verde \\& Jimenez 2000; Verde et al. 2000; Verde et al. 2001; Komatsu \\& Spergel 2001; Shandarin 2002; Bartolo, Matarrese \\& Riotto 2002); for further discussion see Section 3. Our motivation is therefore very similar to that of Coles \\& Jones (1991), which introduced the lognormal density field as an illustration of some of the consequences of a more extreme form of non-linearity involving an exponential transformation of the linear density field.  The plan is as follows. In the following section we introduce some fundamental concepts underlying statistical cosmology, more-or-less from first principles. We do this in order to allow the reader to see explicitly what assumptions underlie standard statistical practise. In Section 3 we then look at some of the contexts in which quadratic non-linearity may arise, either primordially or during the non-linear growth of structure from Gaussian fields. In Section 4 we revisit some of the basic properties used in Section 2 from the viewpoint of quadratic non-linearity and show how some basic implicit assumptions are violated. We then, in Section 5, explore how phase correlations arise in quadratic fields and relate these to higher-order statistics of quadratic fields. We discuss the lessons learned from this study in Section 6. ",
        "conclusions": "In this paper we have addressed two main issues, using the quadratic model as an illustrative example. First we showed explicitly how this  non-Gaussian model has properties that contradict standard folklore based on the assumption of Gaussian fluctuations. We used this model to distinguish carefully between various inter-related concepts such as sample homogeneity, statistical homogeneity, asymptotic independence, ergodicity, and so on. We showed the conditions under which each of these is relevant and deployed the quadratic model for particular examples in which they are violated.  We then used the quadratic model to show, for the first time, how phase association arises in non-linear processes which has exactly the correct form to generate non-zero bispectra and three--point covariance functions. The magnitude of these statistical descriptors is of course related to the magnitude of the Fourier modes, but the factor that determines whether they are zero or non-zero is the arrangement of the phases of these modes.  The connection between polyspectra and phase information is an important one and it opens up many lines of future research, such as how phase correlations relate to redshift distortion and bias. Also, we assumed throughout this study that we could straightforwardly take averages over a large spatial domain to be equal to ensemble averages. Using small volumes of course leads to sampling uncertainties which are  quite straightforward to deal with in the case of the power-spectra but more problematic for higher-order spectra like the bispectrum. Understanding the fluctuations about ensemble averages in terms of phases could also lead to important insights."
    },
    "0208/astro-ph0208540_arXiv.txt": {
        "abstract": "We present methods to compute maps of CMB fluctuations from high resolution cosmic string networks using a full Boltzmann code, on both large and small angular scales.  The accuracy and efficency of these methods are discussed. ",
        "introduction": "The potential role of cosmic strings and other topological defects in cosmology has been the subject of considerable interest for well over two decades (for a review see ref.~\\cite{vilenkin94}).  Perhaps the most exciting prospect would be the detection of their distinct observational signatures in the cosmic microwave sky.  Cosmic strings, for example, are expected to create line-like discontinuities in the CMB temperature pattern, whereas other defects such as global monopoles or textures create `hot spots'.  Their discovery would provide unprecedented information about the nature of unification in the early universe, while their absence from the CMB would significantly strengthen constraints on a wide range of models.  To many, the publication of the BOOMERanG results~\\cite{bernardis2000}, in particular, signaled the demise of topological defects in cosmology.  Indeed, the detection of an acoustic peak around $\\ell\\simeq 200$ was seen as evidence that primordial adiabatic perturbations were the seeds for large-scale structure formation, a view that has been strengthened with the apparent resolution of further peaks (see also~\\cite{hanany2000,scott2002,pearson2002}).  However, the presence of defects is not incompatible with inflation and post-BOOMERanG analyses, such as refs.~\\cite{bouchet2002,contaldi2000}, concluded that they could not be ruled out. Current data allows defects to play a significant (but subdominant) role in large-scale structure formation.  In this sense, it is of great importance to accurately characterize nonGaussian signals from strings, as they are likely to provide the only direct method of detection.  In this paper we will detail the methods that we have developed to create full-sky and high resolution CMB maps generated by cosmic defects or any other `causal' or `active' sources.  First, in \\S\\ref{sec_pert} we detail the large set of perturbation equations that have to be solved, following this in \\S\\ref{sec-sources} with a discussion of the treatment of the source terms which distinguish this analysis from that for inflationary fluctuations. In \\S\\ref{sec-numerics} we then discuss efficient numerical implementation of CMB map-making using the analogue of Green's function techniques, without which the problem would not be tractable computationally.  However, we complete this introduction by discussing previous work on cosmic defects and the CMB, pointing out its relationship to this paper.  Some of the earliest work featured analytic results obtained for simple string configurations \\cite{kaiser84b,gott85,veeraraghavan92}. Such exact solutions are important for testing computational methods. However, although these analytic results are interesting, numerical simulations are essential to obtain accurate quantitative predictions in more general contexts.  The main drawback of numerical results in this context is their limited dynamic range, restricted by the light-crossing times of cosmic defect simulations.  All-sky (large angle) CMB maps can be generated and have been used to obtain the normalization of the power spectrum to COBE, but their angular resolution has been poor.  On the other hand, small angle maps permit the very important characterization of nonGaussian signals due to defects.  Given the prospect of high resolution all-sky observations from the MAP and Planck satellites, ideally one would aim to compute all-sky defect maps of corresponding resolution, but computational resources remain insufficient for this task at present.  Probably the earliest attempt at computing realistic CMB patterns generated by defects was that of Bouchet, Bennett \\& Stebbins~\\cite{bouchet88}.  They employed a flat-space formalism to calculate the CMB temperature $\\Delta T/T$ in the direction $\\hat{\\bf n}$, solving the metric perturbation equations using Green's functions $G({\\bf k},t,t')$ schematically as: \\begin{equation} \\frac{\\Delta T}{T}(\\mathbf{\\hat{n}},{\\bf k},t) \\propto \\int G_{\\mu\\nu}(\\hat{\\bf n},{\\bf k},t,t^{\\prime})\\Theta_{\\mu\\nu}({\\bf k},t') dt^{\\prime}\\, , \\end{equation} where $\\Theta_{\\mu\\nu}$ is the energy-momentum tensor of cosmic strings.  Their methods neglected many effects, notably the presence of baryons and the expansion of the Universe, concentrating solely on the integrated Sachs-Wolfe effect.  Pen, Spergel \\& Turok~\\cite{pen94} computed all-sky maps (at COBE resolution) produced by different global defects including an approximate treatment of CDM, baryons and radiation and their work was extended on intermediate angular scales in ref.~\\cite{coulson94}.  COBE resolution maps generated by local cosmic strings were presented in~\\cite{allen96} using the Allen-Shellard (AS) string code~\\cite{allen90}.  The power spectrum for this map was evaluated for $\\ell\\leq 20$ (using an ensemble of 192 realisations) and they inferred the string linear energy density to be $G\\mu/c^2 = 1.05^{+0.35}_{-0.20}\\times 10^{-6}$.  The most recent work on CMB fluctuations in the presence of causal seeds have made use of full Boltzmann codes (see section~\\ref{sec_pert}), thus including all the relevant physics (to first order).  The AS string code was employed again in the full Boltzmann analysis in ref.~\\cite{allen97}, in which power spectra were computed from the brightness distribution, thus bypassing the maps. Power spectra were computed from simulations of different cosmological epochs and provided clear evidence of the importance of vector and tensor modes in these models, as well as the apparent absence of strongly defined acoustic peaks.  An alternative line-of-sight approach was used in ref.~\\cite{pen97}, and also later in refs.~\\cite{durrer99,contaldi99}, to calculate power spectra for global defects.  Here, the idea was to use unequal time correlators (UETCs) of the defect energy-momentum tensor (approximated by an expansion in eigenvectors) as the source for the perturbation power spectra.  In principle, the method greatly extends the available dynamic range by exploiting the scalability of the correlators during defect evolution.  However, while scalability is approached asymptotically in the radiation and matter eras, during the important radiation-matter transition the UETCs must still be calculated from large simulations bridging this time period.  The line-of-sight method has also been used by~\\cite{albrecht99,pogosian99} who employed an ensemble of toy model realisations of a string network and averaged the power spectra.  The line-of-sight method can be used to compute maps as well: Simatos \\& Perivaropoulos~\\cite{simatos2001} modified it using a more general expansion of plane waves to accommodate for phase differences in a toy model for wiggly strings.  However, while the method is phenomenologically interesting it was necessary to make a number of assumptions about the string perturbation phases.  It is important to note that none of these methods is perfect, and in some sense, they are complimentary: The direct approach developed further here, solving the full Boltzmann equations on a three-dimensional grid, provides reliable high resolution CMB maps. However, the UETC method with a greater dynamic range provides a more extensive view of the angular power spectrum. ",
        "conclusions": "We have presented the full set of sourced evolution equations with the Boltzmann hierarchy necessary to study the gravitational effects of cosmic defects on the CMB.  We have focussed attention on using cosmic string simulations as the evolving causal source terms in these equations.  We have developed efficient numerical techniques for the Green's function computation of high accuracy maps from these string network simulations.  We have also presented numerical tests of the full pipeline.  More quantitative results from extensive supercomputer simulations will be presented in~\\cite{lacmb,hrcmb}, featuring large-angle and small-angle maps respectively."
    },
    "0208/astro-ph0208589_arXiv.txt": {
        "abstract": "We present archival ROSAT data for three recently identified, nearby ($D<70$ pc), young ($\\sim10-40$ Myr) stellar associations: the TW Hydrae Association, the Tucana-Horologium Association, and the $\\beta$ Pic Moving Group. The distributions of ROSAT X-ray hardness ratios (HR1, HR2) for these three groups, whose membership is dominated by low-mass, weak-lined T Tauri stars, are tightly clustered and very similar to one another. The value of HR1 for TW Hya itself --- the only {\\it bona fide} classical T Tauri star in any of the nearby groups --- is clearly anomalous among these nearby young stars. We compare the hardness ratio distributions of stars in the three nearby groups with those of T Tauri stars, the Hyades, and main sequence dwarfs in the field. This comparison demonstrates that the X-ray spectra of F through M stars soften with age, and that F and G stars evolve more rapidly in X-ray spectral hardness than do K and M stars. It is as yet unclear whether this trend can be attributed to age-dependent changes in the intrinsic X-ray spectra of stars of type F and later, to a decrease in the column density of circumstellar gas (e.g., in residual protoplanetary disks), or to the diminishing contributions of star-disk interactions to X-ray emission. Regardless, these results demonstrate that analysis of archival ROSAT X-ray spectral data can help both to identify nearby, young associations and to ascertain the X-ray emission properties of members of known associations. ",
        "introduction": "The recent discovery of several groups of young ($\\sim5-40$ Myr) stars within 100 pc of the Sun has given new direction to the field of star and planet formation (Jayawardhana \\& Greene 2001). The seminal nearby, young stellar group is the TW Hydrae Association (TWA), which lies only $\\sim50$ pc from Earth (Kastner et al.\\ 1997), is $\\sim5-10$ Myr old (Weintraub et al.\\ 2000), and has $\\sim20$ known member star systems (Webb et al.\\ 1999; Zuckerman et al.\\ 2001c [hereafter Zc]). Several dozen additional, recently identified TWA candidate systems are listed in Makarov \\& Fabricius (2001) and Webb (2001). Other examples are the very nearby $\\beta$ Pictoris Moving Group (bPMG; $D \\sim 36$ pc, age $\\sim12$ Myr; Zuckerman et al.\\ 2001a [hereafter Za]; Ortega et al.\\ 2002), and the Tucana and Horologium Associations (each $D \\sim 40$ pc, age $\\sim30$ Myr; Torres et al.\\ 2000; Zuckerman \\& Webb 2000; Zuckerman et al.\\ 2001b [hereafter Zb]). Each of these groups consists of $\\sim20$ known and candidate member star systems. Based on their adjacent positions and their similar distances, space motions, and ages, Zb and de la Reza et al.\\ (2001) have proposed that the Tucana and Horologium Associations constitute a single, young stellar group. We adopt this suggestion in the remainder of this paper, and we designate the combined group as the Tucana-Horologium Association (T-HA).  In many respects systems such as the TWA, T-HA, and bPMG, though only recently identified and still in a rapid state of flux, are better suited to detailed studies of star and planet formation than more distant, well-studied star-forming regions like the Orion and Taurus molecular clouds.  Unlike T Tauri stars in regions of active star formation, the nearby groups typically are not readily associated with parent molecular clouds and, as a result, there remains considerable uncertainty concerning their origin and evolutionary status (see reviews in Jayawardhana \\& Greene 2001).  Indeed, the approximate age range of the TWA stars, and consequently the identification of this group as a nearby association, was initially ascertained by Kastner et al.\\ (1997) largely through the strength of the stars' X-ray emission.  The TWA's estimated age of about 10 Myr makes the association especially intriguing, since this age corresponds to the epoch of Jovian planet formation, according to present theory, and is a defining characteristic of post-T Tauri stars (Herbig 1978). The stars in the TWA and other nearby young associations, therefore, likely represent a long-sought missing link between the very young and well-studied T Tauri stage and the stellar main sequence (Jensen 2001).  As a consequence of the proximity of the TWA, X-ray observations of its members (and of TW Hya in particular) have yielded new insight into the origin of high-energy emission from young stars. Archival data from the Roentgen Satellite (ROSAT) demonstrate that the $\\sim10$ Myr age of the TWA represents an especially X-ray-luminous epoch in the early evolution of solar-mass stars (Kastner et al.\\ 1997), and X-ray (ASCA and ROSAT) spectral monitoring suggests the optical and X-ray variability of some classical T Tauri stars is due to a combination of short-term flaring and long-term variations in absorbing column (Kastner et al.\\ 1999).  Chandra/HETGS (High Energy Transmission Gratings Spectrograph) observations of TW Hya itself produced the first high-resolution X-ray spectrum of a T Tauri star (Kastner et al.\\ 2002).  The HETG spectra yield unexpected results for plasma density, temperature, and elemental abundances which, taken together, suggest that the X-ray emission from TW Hya may arise from accretion rather than from coronal activity. If so, this would suggest that many other X-ray luminous young stars derive part or most of their X-ray luminosity from accretion or other star-disk interactions, although coronal activity remains the widely accepted mechanism for such emission (e.g., Feigelson \\& Montmerle 1999).  Motivated by these results, we are conducting an archival study of ROSAT data that focuses on the TWA and other nearby stellar groups. Our goal is to establish the gross X-ray spectral properties of nearby post-T Tauri stars. Such data, when compared with similar results already available for T Tauri stars embedded in molecular clouds (e.g., Neuhauser et al.\\ 1995) and for main-sequence field stars (e.g., Fleming et al.\\ 1995), should offer clues to the evolutionary status of stars in nearby associations and, more generally, to the mechanism(s) responsible for the bright X-ray emission that appears ubiquitous among young (age $\\le 100$ Myr), solar-mass stars. Here, we present an analysis of ROSAT Position-Sensitive Proportional Counter data available for the known members of the TWA, the TA, and the bPMG. These results strengthen the interpretation that the stars in these and other nearby, dispersed young associations constitute a transition stage between cloud-embedded T Tauri stars and main sequence stars. ",
        "conclusions": "Our analysis of the ROSAT PSPC hardness ratios of local associations, and of cloud T Tauri stars and field main sequence stars, indicates that X-ray spectral hardness decreases monotonically with increasing stellar age (Figs.\\ 2 and 3 and Table 1), with the trend stronger for F and G stars than for K and M stars. There are three alternative interpretations of this trend: \\begin{enumerate} \\item The intrinsic X-ray spectrum of a star of spectral type F or later softens with age (Fleming et al.\\ 1995; Neuhauser 1997). Such a trend could be due to decreasing X-ray emission temperature and/or to age-dependent changes in X-ray emitting region abundances that modulate the strengths of the brightest emission lines in the ROSAT band. \\item The column density of X-ray absorbing material associated with the environments of young stars declines monotonically as these objects evolve. For Taurus TTS, the X-ray absorbing gas and dust may consist of both circumstellar and molecular cloud material, whereas for the older, nearby stellar groups like the TWA, T-HA, and bPMG, we expect residual circumstellar material to dominate. This would suggest that PSPC hardness ratios of the members of local associations can be used to examine the evolution of gaseous circumstellar disks around young stars. \\item Contributions to X-ray emission from star-disk interactions in general, and accretion in particular, decline as stars evolve from the T Tauri phase toward the main sequence. Such contributions may arise in star-disk magnetic field reconnection events (e.g., Shu et al.\\ 1997) or in energetic shocks along accretion columns (Kastner et al.\\ 2002). In either model, the emitting region temperature would be in excess of $\\sim10^6$ K, and therefore should contribute a relatively hard excess X-ray emission component that diminishes with age as the disk disperses. \\end{enumerate}  For F and G stars --- which continue to evolve in HR from the T Tauri through post-T Tauri through early main sequence stages --- we cannot distinguish, at present, between these alternative explanations. Given that TW Hya is surrounded by a disk viewed nearly pole-on (e.g., Kastner et al.\\ 2002 and references therein), the anomalous position of this K7 star in Fig.\\ 1 offers support for the third interpretation (declining star-disk interactions) in the case of young ($\\stackrel{<}{\\sim} 10$ Myr), low-mass stars. The first interpretation (changing intrinsic stellar X-ray emission properties) best explains the continuing X-ray spectral evolution of K and M stars beyond the age of the Hyades, however. We are analyzing archival PSPC spectra of individual stars to determine whether further progress can be made from the available ROSAT data (Kastner \\& Crigger, in preparation). However, high-resolution Chandra and XMM X-ray spectroscopy of representative objects --- from which precise X-ray emitting region temperatures, densities, and elemental abundances can be determined (e.g., Kastner et al.) --- will no doubt be required to establish the mechanism(s) responsible for the softening of X-ray emission with stellar age.  Regardless of the correct interpretation of the observed trend, Figs.\\ 2 and 3 demonstrate that analysis of ROSAT HRs can help discriminate between X-ray-emitting T Tauri stars, post-T Tauri stars, and both young (Hyades- and Pleiades-age) and older (age $\\stackrel{>}{\\sim} 1 $ Gyr) main sequence stars in the field. ROSAT HRs evidently are a particularly useful tool for assessing the ages of samples that consist mostly of F and G stars, while for spectral types of K and later, ROSAT HRs can effectively isolate peculiar cases (e.g., TW Hya). The results presented here suggest, therefore, that candidate members of young associations can be identified partly on the basis of ROSAT PSPC hardness ratios, particularly if other age-related X-ray emission discriminants are applied (such as the ratio of X-ray luminosity to bolometric luminosity; Kastner et al.\\ 1997)."
    },
    "0208/astro-ph0208406_arXiv.txt": {
        "abstract": " ",
        "introduction": "Gravitational wave (GW) radiation is expected to be detected within 10 years or so, thereby opening a new window to probe extremely high-density objects and the early universe (see, e.g. Thorne 1987, 1997; Blair 1991; Tsubono et al. 1997). When detected, the impact will be enormous and the understanding of astrophysical objects should inevitably undergo revolutionary development. It is thus of great importance to discuss at this moment what kinds of objects can emit GW radiation.  In the present study, we consider if it is feasible to detect GW radiation from accretion disks in general contexts.  At first glance, it is unreasonable to expect GW radiation from accretion disks, since, while the generation of GW requires the presence of quadruple moment, the usual disk models are constructed under the assumption of axisymmetry. Actually, the axisymmetric assumption is thought to be quite generally satisfied (see, e.g., Kato et al. 1998 for a review of various disk models). Then, how and under what circumstances can non-axisymmetry arise?  A non-axisymmetric disk structure is often discussed in the context of accretion disks in close binary systems. Since the shape of the Roche lobe is not totally spherically symmetric, we expect a disk there to lose axial symmetry, especially when the disk size is large, close to the size of the Roche lobe (Paczy\\'nski 1977).  Another good indication of non-axisymmetric structure is frequently discussed based on hydrodynamical simulations; e.g. Sawada et al. (1986). Note that the presence of spiral patterns on a disk was discovered through the observing technique of Doppler tomography (Steeghs et al. 1997). As we will see later, however, because disks in close binary systems have tiny mass, the production of strong GW radiation is unlikely.  Alternatively, strong magnetic fields, if they ever exist, could greatly modify the disk structure.  This is the subject of the present study. As was first discussed by Shakura and Sunyaev (1973) and later demonstrated by many authors through MHD simulations (Matsumoto 1999; Stone et al. 1999; Machida et al. 2000; Hawley et al. 2001), the magnetic energy can be amplified by the number of MHD instabilities together with differential rotation up to the value of $p_{\\rm mag}/p_{\\rm gas} \\simeq$ 0.01--1 (e.g. Machida et al. 2000). The corresponding viscosity parameter is $\\alpha \\simeq 0.01 - 0.1$. This is just in the range that we require to account for the observations of dwarf-nova outbursts (Cannizzo 1993) and X-ray nova eruptions (Mineshige, Wheeler 1989).  In standard-type disks, however, it is hard to believe a large influence of magnetic fields on the shape of the disk, since the magnetic energy is, at most, comparable to the internal energy of the gas, which is much less than the kinetic (rotational) energy of the gas as a consequence of efficient radiative cooling of the disk material. In other words, the magnetic pressure cannot overcome the gravitational force in such radiation-dominated accretion flow. In advection-dominated regimes (see Kato et al. 1998 for a review), in contrast, the internal energy of gas can be comparable to its kinetic energy and potential energy because of a low radiation efficiency. We can then expect a large influence of magnetic fields on the dynamics of accretion disks.  Another argument to support this idea comes from the observed complex variability commonly observed in many black-hole candidates during their hard state.  In that state, the spectra are hard, and a hot accretion flow model (ADAF model) can fit the observations (Ichimaru 1977; Narayan et al. 1996). Although its origin is not yet established, many authors suggest that the variability could be caused by a sporadic release of magnetic energy triggered by magnetic reconnection and flares (Takahara 1979; Galeev et al. 1979). Without large magnetic field energy, it is difficult to account for substantial variations. This idea is consistent with the absence of large fluctuations during the soft state, when a standard-type disk seems to be present. It then follows that a non-axisymmetric disk structure may be generated by magnetic fields in cooling-inefficient regimes (i.e. adiabatic regimes) of disk accretion. In fact,  the global MHD simulation of a disk exhibits a spatial inhomogeneous structure (Kawaguchi et al. 2000). Finally, note that large fluctuations are also observed in the high-luminosity state.  Magnetic activities also seem to be enhanced in such a state (e.g. Mineshige et al. 2000).  We, here, consider GW radiation from accretion disks, the shapes of which are possibly influenced by large magnetic fields or other effects. In section 2 we calculate the moments of inertia by using the MHD simulations data and see to what extent a deviation from an axisymmetric disk can be expected. We then discuss the detectability of GW radiation in various astrophysical contexts. The final section is devoted to discussion. ",
        "conclusions": "We have discussed the detectability of GW radiation from accretion disks based on the assumption that disks are magnetically dominated, and thus their structure is inhomogeneous, leading to the emergence of non-axial symmetry. In the case of close binaries, however, GW radiation is totally negligible because of their very tiny disk masses. In the case of AGN, GW radiation is not significant, either, because of slow rotation velocities. Instead, we may concentrate on the innermost part of the AGN disks, which has much faster rotation velocities. In fact, the inner hot part of AGN disks seems to be in a non-radiative state, thus producing hard X-ray emission. Yet, the detection of GW radiation is not very feasible with the currently planned GW detectors, unless additional mechanisms work to enhance the disk deformation.  The most probable case may be found in gamma-ray bursts and supernovae (or hypernovae), in which a massive disk (with solar mass or so) is suggested to be formed around a stellar-mass compact component. Note, however, that the time variations of GW radiation are not predictable; they are highly chaotic, although there is a hint of quasi-periodicities. Also note that much more intense GW radiation is expected before the formation of a massive torus.  Nevertheless, the detection of GW radiation from such a massive disk is of great importance to probe the central engines of the GRBs and SNe (HNe).  The accretion disks can emit electro-magnetic (EM) wave radiation as well as GW radiation.  The former might be even more important than the latter, when considering the long-term evolution of the disk, since whereas the emissivity of GW radiation is proportional to $\\Omega^4$, that of the EM wave radiation is proportional to $\\Omega^2$. One might think that the disk rotation may then be efficiently decelerated by the EM wave radiation so that significant GW radiation cannot be expected. However, this does never occur, since the removal of angular momentum of a disk gas results in faster rotation, rather than slower rotation, unlike the case of rotating stars. This is because when the disk material loses angular momentum, $\\ell$ ($ \\propto \\sqrt{r}$), the disk gas moves inward because of the reduced centrifugal force (which is $\\propto \\ell^2$), leading to more rapid rotation with higher angular frequency (note that $\\Omega \\propto r^{-3/2}$). This makes a good contrast with the case of rotating neutron stars, in which the removal of angular momentum results in a spin-down of the star.  Further, we find it unlikely that the evolution of the disk is totally modified by the EM wave radiation, since the magnetic energy, which produces EM wave radiation, is only $\\sim$ 1\\% of the gravitational energy (which is equal to the rotational energy by the Virial theorem). To be more precise, the Alfv\\'en wave is more important than EM wave radiation in terms of the energy loss from the disk. The Poynting flux due to Alfv\\'en waves propagating along the poloidal magnetic field is \\begin{equation} {\\dot E}_{\\rm Alf} \\sim B^2 v r^2, \\end{equation} where $v$ is on the order of the rotation velocity of the footpoint of the poloidal field anchored by the disk; that is $v = r\\Omega$.  On the other hand, the energy loss due to the EM wave radiation is at most \\begin{equation} {\\dot E}_{\\rm EM} \\sim (\\Omega^2 B r^3)^2/c^3, \\end{equation} since the magnetic moment is of the order of $Br^3$ at maximum. Thus, we have \\begin{equation} {\\dot E}_{\\rm EM} \\sim {\\dot E}_{\\rm Alf} (r\\Omega/c)^3 < {\\dot E}_{\\rm Alf}. \\end{equation} The characteristic time of the disk evolution due to the emission of Alfv\\'en wave is then \\begin{equation} t_{\\rm Alf} \\sim (GM_*\\rho/r) B^{-2} \\Omega^{-1} \\gg \\Omega^{-1}, \\end{equation} [recall $B^2/(GM_* \\rho/r) \\sim 0.01$]. Note that this seems to be highly underestimated, since the magnetic fields have bisymmetric spiral patterns (Machida, Matsumoto 2002) so that the magnetic moment should be much less than what we assumed above. We can thus safely conclude that the EM effects are not very significant, and, thus, our argument concerning the GW radiation from the disk is not basically altered.  Since different parts of the disk material rotate with different angular frequencies ($\\Omega$),  the actual GW radiation could be the superpositions of many modes which vary on a variety of periods.  This leads to a reduction in the effective GW amplitudes estimated in the previous section. Whether this is the case or not strongly depends on the magnetic-field distribution in the disk. (Such a problem does not occur in the GRB case, since only a compact disk has been considered there.) If magnetic fields are dominant in the innermost region (within, say, several hundreds of Schwarzschild radii), clear spiral patterns are only appreciable in a relatively narrow region. Unfortunately, however, it is still difficult to calculate the long-term evolution of the magnetized disk with sufficiently large spatial dimension.  Further, it is currently impossible to simulate two-phased disk corona structure with proper treatments of magnetic fields and radiation transfer. We need to await until it becomes possible to calculate the coupling between the radiation, magnetic fields, and matter in accretion flow.  Strong GW radiation from the gamma-ray burst was discussed by van Putten (2001) in a slightly different context. He conjectured that magnetic fields with poloidal topology equivalent to pulsar magnetospheres provide tight coupling between a Kerr hole and suspended accretion torus, thereby radiating GW by extracting the spin-energy of the black hole in the presence of non-axisymmetry. In the present study, in contrast, we do not assume the presence of rapidly spinning black holes, nor tight coupling between the black hole and the torus, which makes a distinction between his model and ours.  Our model should work in any disks with hot gas, in general, and does not require any special condition.  We wish to finally note that for a massive disk whose mass is comparable to, or exceeds, that of the central object, the torus itself becomes gravitationally unstable, thus giving rise to non-axisymmetry, even without magnetic fields (e.g. Woodward et al. 1994).  Then, the corresponding $\\epsilon$ value will be much increased: say, $\\epsilon \\sim 1$ in the extreme case of fragmentation of the disk matter into several pieces. This may be relevant in the case of a compact massive disk as in GRBs and HNe and strengthens our conclusion.  \\bigskip \\bigskip We greatly appreciate a number of valuable comments made by an anonymous referee. We also acknowledge the Yukawa Institute for Theoretical Physics at Kyoto University, where this work was initiated during the YITP workshop YITP-W-01-16 on {\\lq\\lq}gravitational waves.{\\rq\\rq} This work was partially supported by Japan Science and Technology Corporation (ACT-JST) and by Grants-in Aid of the Ministry of Education, Culture, Sports, Science and Technology of Japan (13640238, SM). Numerical computations were carried out by using Fujitsu VPP300/16R at National Astronomical Observatory, Japan and Yukawa Institute Computer Facility."
    },
    "0208/hep-th0208207_arXiv.txt": {
        "abstract": "\\nk We extent our previous study on spherically symmetric braneworld solutions with induced gravity, including non-local bulk effects. We find the most general static four-dimensional black hole solutions with $g_{tt}=-g_{rr}^{-1}$. They satisfy a closed system of equations on the brane and represent the strong-gravity corrections to the Schwarzschild-$(A)dS_{4}$ spacetime. These new solutions have extra terms which give extra attraction relative to the Newtonian-$(A)dS_{4}$ force; however, the conventional limits are easily obtained. These terms, when defined asymptotically, behave like $AdS_{4}$ in this regime, while when defined at infinitely short distances predict either an additional attractive Newtonian potential or an attractive potential which scales approximately as $\\sqrt{r}$. One of the solutions found gives extra deflection of light compared to Newtonian deflection. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0208/astro-ph0208156_arXiv.txt": {
        "abstract": "We consider the possibility that at least some GRB explosions take place inside pulsar wind bubbles (PWBs), in the context of the supranova model, where initially a supernova explosion takes place, leaving behind a supra-massive neutron star (SMNS), which loses its rotational energy on a time scale of months to tens of years and collapses to a black hole, triggering a GRB explosion. The most natural mechanism by which the SMNS can lose its rotational energy is through a strong pulsar type wind, between the supernova and the GRB events, which is expected to create a PWB. We analyze in some detail the observational implications of such a plerionic environment on the afterglow and prompt GRB emissions, as well as the prospect for direct detection of the plerion emission. We find that for a simple spherical model, GRBs with iron lines detected in their X-ray afterglow should not have a detectable radio afterglow, and should have small jet break times and non-relativistic transition times, in disagreement with observations for some of the GRBs with X-ray lines. These discrepancies with the observations may be reconciled by resorting to a non-spherical geometry, where the PWB is elongated along the polar axis. We find that the emission from the PWB should persist well into the afterglow, and the lack of detection of such a component provides interesting constraints on the model parameters. Finally, we predict that the inverse Compton upscattering of the PWB photons by the relativistic electrons of the afterglow (external Compton, EC) should lead to high energy emission during the early afterglow that might explain the GeV photons detected by EGRET for a few GRBs, and should be detectable by future missions such as GLAST. ",
        "introduction": "\\label{sec:intro}  The leading models for Gamma-Ray Bursts (GRBs) involve a relativistic wind emanating from a compact central source. The prompt gamma-ray emission is usually attributed to energy dissipation within the outflow itself, due to internal shocks within the flow that arise from variability in its Lorentz factor, while the afterglow emission arises from an external shock that is driven into the ambient medium, as it decelerates the ejected matter (Rees \\& M\\'esz\\'aros 1994; Sari \\& Piran 1997). In this so called `internal-external' shock model, the duration of the prompt GRB is directly related to the time during which the central source is active. The most popular emission mechanism is synchrotron radiation from relativistic electrons accelerated in the shocks, that radiate in the strong magnetic fields (close to equipartition values) within the shocked plasma. An additional radiation mechanism that may also play some role is synchrotron self-Compton (SSC), which is the upscattering of the synchrotron photons by the relativistic electrons, to much higher energies.   Progenitor models of GRBs are divided into two main categories. The first category involves the merger of a binary system of compact objects, such as a double neutron star (NS-NS, Eichler et al. 1989), a neutron star and a black hole (NS-BH, Narayan, Pacy\\'nski \\& Piran 1992) or a black hole and a Helium star or a white dwarf (BH-He, BH-WD, Fryer \\& Woosley 1998; Fryer, Woosley \\& Hartmann 1999). The second category involves the death of a massive star. It includes the failed supernova (Woosley 1993) or hypernova (Pacy\\'nski 1998) models, where a black hole is created promptly, and a large accretion rate from a surrounding accretion disk (or torus) feeds a strong relativistic jet in the polar regions. This type of model is known as the collapsar model. An alternative model within this second category is the supranova model (Vietri \\& Stella 1998), where a massive star explodes in a supernova and leaves behind a supra-massive neutron star (SMNS) which on a time scale of a few years loses its rotational energy and collapses to a black hole, triggering the GRB event. Long GRBs (with a duration $\\gtrsim 2\\;{\\rm s}$) are usually attributed to the second category of progenitors, while short GRBs are attributed to the first category. In all the different scenarios mentioned above, the final stage of the process consists of a newly formed black hole with a large accretion rate from a surrounding torus, and involve a similar energy budget ($\\lesssim 10^{54}\\;{\\rm ergs}$).  In this work we perform a detailed analysis of the supranova model, focusing on its possible observational signatures. This aims towards establishing tools that would enable us to distinguish between the supranova model and other progenitor models through observations, and to constrain the model parameter using current observations. The original motivation for the supranova model was to provide a relatively baryon clean environment for the GRB jet. As it turned out, it also seemed to naturally accommodate the later detection of iron lines in several X-ray afterglows (Lazzati, Campana, \\& Ghisellini 1999; Piro et al. 2000; Vietri et al. 2001).  It was later suggested that the most natural mechanism by which the SMNS can lose its rotational energy is through a strong pulsar type wind, between the supernova and the GRB events, which typically creates a pulsar wind bubble (PWB), also referred to as a plerion (K\\\"onigl \\& Granot 2002, KG hereafter; Inoue, Guetta \\& Pacini 2002, IGP hereafter). KG suggested that the shocked pulsar wind into which the afterglow shock propagates in this picture may naturally account for the large inferred values of $\\epsilon_e\\sim 0.1$ and $\\epsilon_B\\sim 10^{-3}-0.1$ (the  fractions of the internal energy in the electrons and in the magnetic field, respectively) that are inferred from fits to afterglow observations (Wijers \\& Galama 1999; Granot, Piran \\& Sari 1999; Chevalier \\& Li 2000; Panaitescu \\& Kumar 2002). This is attributed to the fact that pulsar winds are believed to largely consist of electron-positron pairs, and have magnetization parameters in the right range. This relaxes the need of generating strong magnetic fields in the shock itself, as is required in other models, where the magnetic field in the external medium (assumed to be either the ISM or a stellar wind of a massive star progenitor) is typically too small to account for the values of $\\epsilon_B$ that are inferred from observations. Another attractive feature of this model, pointed out by IGP is the possible high energy emission, in the GeV-TeV range, that may result from the upscattering of photons from the plerion by the relativistic electrons in the afterglow shock (external Compton, EC hereafter), and may be detected by GLAST. They have shown that the EC emission can provide a viable explanation for the extended GeV emission seen by EGRET in GRB 940217 (Hurley et al. 1994).   We use a simple spherical model for the PWB. We find that a spherical model cannot accommodate the typical afterglow emission together with the iron line features observed in the X-ray afterglow of some GRBs. However, it was mentioned early on that in order to have a long lived afterglow emission together with the iron line features, a deviation from spherical symmetry is needed, where the line of sight is relatively devoid of the material producing the iron lines (Lazzati et al. 1999; Vietri et al. 2001). This is required in order to avoid a direct collision of the afterglow shock with the line producing material on an observed time of the order of a day or so. It was later pointed out that a PWB is expected to exist inside the SNR shell, which decelerates the afterglow shock at a smaller radius, so that in order for the afterglow to remain relativistic up to a month or more, and produce the iron lines, we need the PWB to be elongated along its rotational axis (KG). In this paper we strengthen this conclusion, and show that in order to produce iron lines with a spherical PWB, its radius must be sufficiently small, resulting in a large density inside the PWB and a high self absorption frequency implying no radio afterglow, in contrast with observations. We leave the detailed treatment of an elongated PWB to a future work, while in the present work we briefly comment about the expected effects of an elongated geometry compared to a spherical one.  In this work we extend the analysis of KG and IGP, and perform detailed calculations of the radiation from the PWB, the prompt GRB and from the afterglow that occur inside the PWB. We now give a short overview of the structure of the paper, where in each section we stress the original features, new results and the observational constraints on the model. In \\S \\ref{PWB} we present our ``PWB'' model, introduce the relevant parameterization and model the acceleration of the supernova remnant (SNR) shell by the shocked pulsar wind. We use a simple spherical geometry and the pulsar wind is assumed to consist of proton and $e^\\pm$ components with roughly equal energies, as well as a magnetic field. The conditions under which the iron line features that were observed in several X-ray afterglows may be reproduced within the PWB model, are investigated in \\S \\ref{lines}. We find that this requires a time delay of $\\lesssim 1\\;$yr between the supernova and the GRB events. In \\S \\ref{PlerEmis} we perform a detailed study of the plerion emission, including the synchrotron and SSC components, and provide an elaborate description of the relevant Klein-Nishina effect. We also discuss the upper cutoff that is imposed on high energy photons due to pair production with the radiation field of the PWB, go over the prospect for direct detection of the plerion emission, and derive observational constraints on the parameters of our model. The effects of the PWB environment on the prompt GRB emission are analyzed in \\S \\ref{prompt_GRB}, and we find that the EC from the prompt GRB should typically be very small, but might be detectable for extreme parameters. In \\S \\ref{Afterglow} we discuss the implications of a plerionic environment on the afterglow emission, and introduce the appropriate parameterization. The radial density profile of the PWB is approximated as a power law in radius (KG), $\\propto r^{-k}$, where $k$ typically ranges between $0$ (similar to an ISM) and $1$ (intermediate between an ISM and a stellar wind). The synchrotron, SSC and EC components are calculated and we provide detailed expressions for the break frequencies and flux normalization, for $k=0,\\,1$. We also calculate the high energy emission that is predicted in this model. The results are discussed in \\S \\ref{discussion} and in \\S \\ref{conclusions} we give our conclusions. ",
        "conclusions": "\\label{conclusions}  Our main conclusion is that existing afterglow observations put interesting constraints on the model parameters, the most important of which being the time delay $t_{\\rm sd}$ between the supernova and GRB events, which is constrained to be $\\gtrsim 20\\;{\\rm yr}$, in order to explain typical afterglow observations and the lack of detection of the plerion emission in the radio during the afterglow. Another important conclusion is that iron line features that have been observed in a few X-ray afterglows cannot be naturally explained within the simplest spherical version of the PWB model, that has been considered in this work. This is because the production of these lines requires $t_{\\rm sd}\\lesssim t_{\\rm Fe}\\sim 1\\;{\\rm yr}$ which implies a very large density for the PWB and effects the afterglow emission in a number of different ways: i) The self absorption frequency of the afterglow is typically above the radio, implying no detectable radio afterglow, while radio afterglows were detected for GRBs 970508, 970828, and 991216, where the iron line feature for the latest of these three is the most significant detection to date ($\\sim 4\\sigma$). We also expect the self absorption frequency of the plerion emission to be above the radio in this case, so that the radio emission from the plerion should not be detectable, and possibly confused with that of the afterglow. ii) A short jet break time $t_j$ and a relatively short non-relativistic transition time $t_{\\rm NR}$ are implied, as both scale linearly with $t_{\\rm sd}$ and are in the right range inferred from observations for $t_{\\rm sd}\\sim 30\\;{\\rm yr}$ (see Eqs. \\ref{t_j}, \\ref{t_NR}). iii) The electrons are always in the fast cooling regime during the entire afterglow.  The above constraints regarding the iron lines may be relaxed if we allow for deviations from the simple spherical geometry we have assumed for the PWB. A natural variant is when the PWB becomes elongated along its rotational axis (KG). This may occur if the surface mass density of the SNR shell is smaller at the poles compared to the equator, so that during the acceleration of the SNR shell by the pressure of the shocked pulsar wind (that is expected to be roughly the same at the poles and at the equator) its radius will become larger at the poles, as the acceleration there will be larger. A large-scale toroidal magnetic field within the PWB may also contribute to the elongation of the SNR shell along its polar axis (KG). It is also likely that the progenitor star that gave rise to a SMNS had an anisotropic mass loss, which results in a density contrast between the equators (where the density is higher) and the poles (where the density is lower). A sufficiently large density contrast between the equator and the poles can also contribute to the elongation of the shell, for sufficiently large $t_{\\rm sd}$, as the SNR shell will begin to be decelerated due to the interaction with the external medium, at a smaller radius near the equator, compared to the poles. A similar non-spherical variant of the model is if we allow for holes in the SNR shell, that extend over a small angle around the polar axis, where all the wind is decelerated in a termination shock within the SNR shell ($R_s<R_b$), but most of the shocked pulsar wind can get out through the holes near the poles and reach a radius considerably larger than $R_b$. This variant may be viewed as a limiting case of the previous variant, when the surface density contrast of the SNR shell, between the equator and the poles is very large. This implies that most of the mass in the SNR shell is concentrated near the equator, while only a small fraction of it is near the poles, so that the radius near the poles can be as large as $\\sim ct_{\\rm sd}$, while the equatorial radius is $\\lesssim R_b$. In both variants, the total volume of the PWB is close to that of a sphere with the polar radius, and much larger than that of a sphere with the equatorial radius. This would allow for a smaller density with the same small value of the equatorial radius that is required to produce the iron lines. We see that in principle, variants of the simple model are capable of reconciling between the iron line detections and the afterglow observations.  An important advantage of the PWB model is that it can naturally explain the large values of $\\epsilon_B$ and $\\epsilon_e$ that are inferred from fits to afterglow data (KG), thanks to the large relative number of electron-positron pairs and large magnetic fields in the PWB. This is in contrast with standard environment that is usually assumed to be either an ISM or the stellar wind of a massive progenitor, that consists of protons and electrons, and where the magnetic field is too small to explain the values inferred from afterglow observations, and magnetic field generation at the shock itself is required. Additional advantages of the PWB model are its ability to naturally account for the range of external densities inferred from afterglow fits, and allowing for a homogeneous external medium ($k\\sim 0$), as inferred for most afterglows, with a massive star progenitor.  Another advantage of the PWB model is its capability of explaining the high energy emission observed in some GRBs (Schneid et al. 1992; Sommer et al. 1994; Hurley et al. 19994; Schneid et al. 1995). We find that the high energy emission during the early afterglow at photon energies $\\gtrsim 100\\;$keV is dominated by the external Compton (EC) component, that is due to the upscattering of photons from the plerion radiation field to higher energies by the relativistic electrons behind the afterglow shock. We predict that such a high energy emission may be detected in a large fraction of GRBs with the upcoming mission GLAST. However, we find an upper cutoff at a photon energy of $\\sim 1\\,t_{\\rm sd,0}^2\\;$GeV, due to opacity to pair production with the photons of the PWB. This implies no high energy emission above $\\sim 1\\;$GeV for afterglows with X-ray line features, but allows photons up to an energy of $\\sim 1\\;$TeV for afterglows with an external density typical of the ISM ($t_{\\rm sd}\\sim t_{\\rm ISM}$)."
    },
    "0208/astro-ph0208010_arXiv.txt": {
        "abstract": "We compare the sensitivity of WIMP detection via direct separation of possible signal vs.\\ background to WIMP detection via detection of an annual modulation, in which signal and background cannot be separated on an event-by-event basis.  In order to determine how the constraints from the two different types of experiments might be combined an adequate incorporation of uncertainties due to galactic halo models must be made.  This issue is particularly timely in light of recent direct detection limits from Edelweiss and CDMS, which we now demonstrate cannot be made consistent with a recent claimed DAMA annual modulation observation, even by including halo uncertainties.  On the other hand, we demonstrate that a combination of these two techniques, in the event of any positive direct detection signal, could ultimately allow significant constraints on anisotropic halo models even without directional sensitivity in these detectors. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0208/astro-ph0208057_arXiv.txt": {
        "abstract": "We present a detailed analysis of the behaviour of the Red Clump $K$-band absolute magnitude ($M_K^{\\rm RC}$) in simple and composite stellar populations, in light of its use as standard candle for distance determinations. The advantage of using $M_K^{\\rm RC}$, following recent empirical calibrations of its value for the solar neighbourhood, arises from its very low sensitivity to the extinction by interstellar dust. We show that, as in the case of the $V$- and $I$-band results, $M_K^{\\rm RC}$ is a complicated function of the stellar metallicity $Z$ and age $t$. In general, $M_K^{\\rm RC}$ is more sensitive to $t$ and $Z$ than $M_I^{\\rm RC}$, for high $t$ and low $Z$. Morever, for ages above $\\sim$1.5 Gyr, $M_K^{\\rm RC}$ decreases with increasing $Z$, the opposite behaviour with respect to $M_V^{\\rm RC}$ and $M_I^{\\rm RC}$.  We provide data and equations which allow the determination of the $K$-band population correction $\\Delta M_K^{\\rm RC}$ (difference between the Red Clump brightness in the solar neighbourhood and in the population under scrutiny) for any generic stellar population. These data complement the results presented in Girardi \\& Salaris~(2001) for the $V$- and $I$-band. We show how data from galactic open clusters consistently support our predicted $\\Delta M_V^{\\rm RC}$, $\\Delta M_I^{\\rm RC}$ and $\\Delta M_K^{\\rm RC}$ values.  Multiband $VIK$ population corrections for various galaxy systems are provided. They can be used in conjunction with the method devised by Alves et al.~(2002), in order to derive simultaneously reddening and distance from the use of $VIK$ observations of Red Clump stars. We have positively tested this technique on the Galactic globular cluster 47~Tuc, for which both an empirical parallax-based main sequence fitting distance and reddening estimates exist. We have also studied the case of using only $V$ and $I$ photometry, recovering consistent results for both reddening and distance. Application of this method to an OGLE-II field, and the results by Alves et al.~(2002), confirm a LMC distance modulus of about 18.50, in agreement with the $HST$ extragalactic distance scale zero-point. ",
        "introduction": "\\label{sec_intro}  During the last few years a large body of work has been focused on the use of helium burning red clump (RC) stars as distance indicator; this has been mainly due to the fact that $Hipparcos$ parallaxes allows an accurate calibration of the average RC brightness in the solar neighbourhood (as a comparison, $Hipparcos$ parallaxes of the closest Cepheid and RR~Lyrae stars have much larger errors, and the calibration of their absolute magnitude has provided conflicting results, see e.g. Feast \\& Catchpole~1997; Luri et al.~1998), and that the RC is easily recognizable in the colour-magnitude-diagram (CMD) of intermediate-old stellar populations.  The determination of the absolute magnitude of the local RC in a given pass-band $\\lambda$, $M_{\\lambda, {\\rm local}}^{\\rm RC}$, and the apparent magnitude $m_\\lambda^{\\rm RC}$ of the RC in a given stellar population is not difficult, since in both the {\\em Hipparcos} database of nearby stars, and in CMDs covering even a small fraction of a nearby galaxy, one finds several hundreds of clump stars, easily identifiable from their CMD location. As proposed by Stanek \\& Garnavich~(1998), a non-linear least-square fit of the function \\begin{equation} N(m_\\lambda) = a + b m_\\lambda + c m_\\lambda^2 + d \\exp\\left[-\\frac{(m_\\lambda^{\\rm RC}-m_\\lambda)^2}{2\\sigma_{m_\\lambda}^2}\\right] \\label{eq_fit} \\end{equation} to the histogram of stars in the clump region per magnitude bin provides the value of $m_\\lambda^{\\rm RC}$ and its associated standard error. By applying this procedure to the {\\em Hipparcos} database of nearby stars, the RC absolute brightness in the $I$-band (Cousins) $M_I^{\\rm RC}$ has been determined with accuracy of hundredths of magnitude (Paczy\\'nski \\& Stanek 1998; Stanek, Zaritsky \\& Harris 1998).  Once the mean apparent magnitude of the RC in a given photometric band, $m_{\\lambda}^{\\rm RC}$, is measured in a nearby galaxy, its absolute distance modulus $\\mu_0=(m-M)_0$ is easily derived by means of \\begin{equation} \\mu_0 = m_\\lambda^{\\rm RC} - M_{\\lambda, {\\rm local}}^{\\rm RC} - A_{m_\\lambda} + \\Delta M_\\lambda^{\\rm RC}. \\label{eq_mu} \\end{equation} In this equation, $A_{m_\\lambda}$ is the interstellar extinction to the RC population of an external galaxy, and $\\Delta M_\\lambda^{\\rm RC} = M_{\\lambda, {\\rm local}}^{\\rm RC} - M_{\\lambda, {\\rm galaxy}}^{\\rm RC}$ is the population effect, i.e.\\ the difference of the mean RC absolute magnitude between the local and external samples of stars.  After early claims (Paczy\\'nski \\& Stanek 1998; Udalski et al.\\ 1998; Stanek et al.\\ 1998) based mainly on the constancy of  $M_I^{\\rm RC}$ with respect to the colour $(V-I)$, that  the population correction ($\\Delta M_I^{\\rm RC}$) is negligible in the $I$-band, Cole~(1998), Girardi et al.\\ (1998, hereafter Paper~I) and Girardi \\& Salaris~(2001, hereafter Paper~II) have conclusively demonstrated that this is not the case. The in-depth analysis we presented in Paper~II has shown how theoretical stellar evolution models are able to reproduce the morphology and properties of the $I$-band RC brightness in Galactic open clusters, of the local $Hipparcos$ RC as well as the RC in a sample of external galaxies, when using current estimates of the Star-Formation-History (SFR) and Age-Metallicity-Relation (AMR) of their stellar populations. The models predict a non-negligible dependence of $M_I^{\\rm RC}$ on both metallicity and age, so that the population correction $\\Delta M_I^{\\rm RC}$ must be taken into account. In particular, the dependence on age is complex and non-monotonic. This implies that accurate RC distances can be determined only if one can estimate SFR and AMR for the stellar population under scrutiny. If there are no determinations of these two key parameters, errors up to $\\sim$0.3 mag in the derived distance modulus are to be accepted.  In the specific case of the Large Magellanic Cloud (LMC) distance, that sets the zero point of the extragalactic distance scale, we obtained a population correction in the $I$-band $\\Delta M_I^{\\rm RC}$=+0.20. Unfortunately, the evaluation of the RC distance to the LMC is plagued by what possibly are uncertainties in the estimate of the interstellar extinction. Romaniello et al.\\ (2000) and Sakai, Zaritsky and Kennicutt~(2000) obtain, respectively, $I_0^{\\rm RC}=18.12\\pm0.02$ and $I_0^{\\rm RC}=18.06\\pm0.02$ from $HST$ data, after deriving the extinction to individual stars from multiband colour fitting using theoretical model atmospheres results; these values, used in conjunction with the appropriate evolutionary correction and $M_I^{\\rm RC}=-0.23\\pm$0.03 for the local RC, provide $\\mu_0=18.55\\pm0.04$ and $\\mu_0=18.49\\pm0.04$, respectively. On the other hand Udalski (2000) has obtained $I_0^{\\rm RC}=17.94\\pm0.05$, using different photometric data and determining the average stellar reddenings from the $COBE/DIRBE$ reddening maps (Schlegel, Finkbeiner \\& Davis~1998), which gives a much shorter value, $\\mu_0=18.37\\pm0.06$.  To overcome uncertainties related to the extinction correction, Alves~(2000) and Grocholski \\& Sarajedini~(2002) have recently discussed the use of the RC in the $K$-band; the main advantage of working in this wavelength range is a very much reduced sensitivity to the extinction ($A_K \\sim 0.2~A_I$). Alves~(2000) has derived $M_K^{\\rm RC}=-1.61\\pm$0.03 for the local RC stars with $Hipparcos$ parallaxes and spectroscopic metallicities, and obtained a distance to the Galactic centre ($\\mu_0=14.58\\pm0.11$, or $8.24\\pm0.42$ Kpc) using the observed RC $K$-band magnitude for a sample of Baade's window RC stars with the same mean metallicity of the local sample; the underlying assumption is that there is no potential age-effect on the $K$-band RC level. Grocholski \\& Sarajedini~(2002) have tested the behaviour of $M_K^{\\rm RC}$ on a sample of single-age, single-metallicity stellar populations, constituted by 14 Galactic open clusters and 2 globular clusters with independent estimates of reddening and distances, finding a good agreement with the behaviour predicted by the theoretical models by Girardi et al.~(2000), which are the same used in Paper~I and Paper~II. Moreover, Alves et al.~(2002) have used simultaneously $V$, $I$ and $K$ photometry of LMC RC stars to determine both the  LMC distance and the extinction for the observed RC population. They have demonstrated the need to apply population corrections to the RC absolute brightness determined from the local clump, otherwise a negative extinction is obtained; their final result $\\mu_0=18.506\\pm0.033$ for the LMC ($\\mu_0=18.493\\pm0.033\\pm0.03$ random plus systematic error, after correcting for the location of the observed fields with respect to the LMC center, and including an estimate of the probable systematic errors involved) and $E(B-V)=0.089\\pm0.015$ was obtained making use of evolutionary corrections derived with the techniques and models presented in paper~II.   In light of these results we consider important to investigate theoretically the properties of the RC in the $K$-band, as well as to assess the simultaneous reliability of the population corrections to the $V$-, $I$- and $K$-band for a given stellar population.  In Sect.~\\ref{sec_agemet} we discuss in detail the properties of the RC in the $K$-band and its dependence on age and metallicity, while in Sect.~\\ref{sec_popcorr} we perform further tests to assess the reliability of the population corrections predicted by theory.  In Sect.~\\ref{sec_galaxies} we present multiband population corrections for the galaxy systems discussed in Paper~II; in addition, we will discuss the technique used by Alves et al.~(2002) for their LMC distance determination, and apply it to the case of the Galactic globular cluster 47~Tuc (for which reddening estimates and an accurate empirical main-sequence fitting distance do exist) and to a field of the LMC with independent reddening determinations.  Our final conclusions are presented in Sect.~\\ref{sec_conclu}. ",
        "conclusions": ""
    },
    "0208/astro-ph0208261_arXiv.txt": {
        "abstract": "Kuiper Belt Objects (KBOs) are generally very faint and cannot in practice be monitored with a well-sampled long-term light curve; so our discovery of the bright KBO 2000 $EB_{173}$ offers an excellent opportunity for synoptic studies.  We present a well-sampled photometric time series (77 R magnitudes and 29 V magnitudes on 78 nights) over a 225-day time span centered on the 2001 opposition.  The light curve (corrected to the year 2001 opposition distance) varies from 19.11 to 19.39 mag with a single peak that is smooth, time symmetric, and coincident with opposition.  All variations in the light curve are consistent with a linear opposition surge ($R_{OPP} = 19.083 + 0.125 \\alpha$, where $\\alpha$ is the solar phase angle), while any rotational modulation must have a peak-to-peak amplitude of less than 0.097 mag. This is the first measured opposition surge for any KBO (other than Pluto). The \\vr ~ color is $0.63 \\pm 0.02$, with no apparent variation with phase at the few percent level. With $R=19.11$ at opposition, 2000 $EB_{173}$ remains the brightest known KBO and a prime target for future photometric and spectroscopic studies. ",
        "introduction": "Most searches for Kuiper Belt Objects (KBOs) scan over small areas of the sky to deep limits, with little chance of discovering the rare-bright objects. Only in the last two years have a number of studies searched wide areas \\citep{fer01,lar01,she00,jlt98}. As part of the QUEST KBO Survey ($66.8 deg^2$ to $R=20.1$), we discovered the very bright KBO 2000 $EB_{173}$ \\citep{fer01}.  This body is one of the largest known KBOs, and the brightest after Pluto.  As a bright source, 2000 $EB_{173}$ can provide the best signal-to-noise ratio for a variety of physical observations, including visible and infrared spectroscopy.  Our own discovery and initial follow-up photometry \\citep{fer01} established the red color of 2000 $EB_{173}$, typical of many KBOs and attributed to the presence on the surface of organic solids \\citep{cru98}. \\citet{bbk00} give a spectrum from 1.4-2.4 {$\\mu$}m, \\citet{lod01} present a spectrum from 0.9-2.4 {$\\mu$}m, and \\citet{jel01} give a reflectance spectrum from 1.0-2.4 {$\\mu$}m.  The infrared spectrum of 2000 $EB_{173}$ is generally featureless although Licandro et al. note strong absorption in the K band.  This might be related to highly carbonized materials with a deficiency of hydrogen bonds.  Here, we report the first long-term photometric light curve for 2000 $EB_{173}$ measured on 78 nights from December 2000 to July 2001 with the Yale 1-meter telescope at Cerro Tololo Observatory. The data reveal a prominent opposition surge, which is a significant increase in brightness within a few degrees of zero phase. The effect is commonly observed for the moon, asteroids, and satellites of planets in the outer solar system \\citep{geh56}, and has been attributed both to the sudden disappearance of shadows cast by regolith particles and to coherent backscattering by particle aggregates \\citep{shh01,hap93}. For the high-albedo icy satellites in the outer solar system, laboratory investigations suggest that coherent backscattering may play a dominant role \\citep{nel98}. For the well-studied asteroids in the main belt, the shape of the phase curve is known to correlate with spectral type and albedo \\citep{bes00,hev89,bol79}. In this paper we compare the phase curve we observe for 2000 $EB_{173}$ with model predictions and with other solar system bodies to place constraints on the likely composition and surface properties. ",
        "conclusions": "We measured 77 R-band and 28 V-band magnitudes of 2000 $EB_{173}$ on 78 nights over a 225 day interval centered on its 2001 opposition.  We find that the light curve varies from $R_{OPP}=19.11$ to $19.39$, with all the changes fully described by a linear phase function with $R_{OPP}=19.083+0.125 \\alpha$. This is the first opposition surge measured for any KBO (other than Pluto).  The amplitude of any rotational modulation is small, with a peak-to-peak amplitude of less than 0.097 mag. The \\vr ~ color is constant at $0.63 \\pm 0.02$ mag.  At opposition (with $\\alpha=0.28 \\degr$), 2000 $EB_{173}$ is the brightest known KBO with $R=19.11$ mag.  We are able to fit the R-band phase curve assuming a Hapke-type opposition  surge with width  $h=0.050$ and amplitude $B(0)>1.7$. This is relatively broad compared to Nereid, and does not offer confirmation that Nereid could be a captured KBO. The large amplitude of the surge, however, indicates the coherent backscattering may be important.  Our measurements of the phase curve in the V-band are not accurate enough to distinguish between shadow hiding and coherent backscattering.  Future measures of the $B-I$ or $B-R$ color near opposition could determine the physical cause of the opposition surge."
    },
    "0208/astro-ph0208111_arXiv.txt": {
        "abstract": "We present the discovery of OB type absorption lines superimposed to the emission line spectrum and the first double-lined orbital elements for the massive Wolf-Rayet binary HDE~318016 (=WR~98), a spectroscopic binary in a circular orbit with a period of 47.825 days. The semiamplitudes of the orbital motion of the emission lines differ from line to line, indicating mass ratios between 1 and 1.7 for $\\mathcal{M}_{WR}/\\mathcal{M}_{OB}$. ",
        "introduction": "HDE~318016 was discovered to have an emission line spectrum of Wolf-Rayet type by \\citet{can38}. Because both N and C emission lines appear strong in the spectrum it was classified as WC7-N6 by \\citet{smi68}. This star was included in the Sixth Catalogue of Galactic Wolf-Rayet Stars \\citep{huc81} as WR~98 and classified as WN7+WC7. However WR~98 was confirmed to be a single-lined binary with a period of 47.8 days with N and C emission lines moving in phase \\citep{nie91}.  The optical spectrum of WR~98 has been described by \\citet{lun84}. \\citet{con89} proposed the nomeclature WN/WC for WR~98 suggesting that this star, along with others with similar emission line spectra where both N and C lines are observed, has a transition composition between the WN and WC subclasses.  \\citet{smi96} classified WR~98 as WN8o/C7, using a new three-dimensio\\-nal classification scheme, where the ``o'' means that no hydrogen is observed in the spectrum. Relevant parameters of WR~98 can be found in the recent VIIth Catalogue of galactic WR stars \\citep{huc01}.  WR~98 is a probable member of the open cluster Trumpler 27 \\citep[e.g.][]{fei00}, and it has also been detected as a non-thermal radio source by \\citet{abb86}. Furthermore, WR~98 exhibits random relatively large amplitude optical light variations ($\\sim$ 0.1 magnitude in the Johnson V filter), typical of stars with WN8 type spectrum, but a periodicity for these variations has not been found \\citep{mar98}.   In this work we present a detailed radial velocity analysis of optical spectral lines of WR~98 showing it to be a double-lined binary with high minimum masses. ",
        "conclusions": "\\subsection{The spectrum}   The blue optical spectrum of WR~98 is illustrated in fig~\\ref{spect}, with identifications for main spectral features. As seen in fig~\\ref{spect}, the spectrum of WR~98 is dominated by emission lines of N{\\sc iii}, N{\\sc iv}, He{\\sc ii}, He{\\sc i}, with a strong C{\\sc iii} feature at $\\lambda$ 4650\\AA\\,, consistent with the WN-WC classification originally proposed by \\citet{smi68}. All of He{\\sc i} emission lines in our blue spectra show P-Cygni profiles, as well as He{\\sc ii} 5411\\AA\\, and N{\\sc iv} 5203\\AA\\, (See Fig.~\\ref{spect}). Relative intensities of N{\\sc iii}, He{\\sc ii}, N{\\sc iv}, and C{\\sc iii} lines in our spectra of WR~98 indicate a spectral type WN7-8/C.  No hydrogen is detected in the spectrum. Relative intensities of emission lines in the spectrum of WR~98 show no appreciable changes in the whole of our dataset spanning 20 years of observations.  We have compared WN the spectrum of WR~98 with that of stars classified as WN7o and WN8o \\citep{smi96}, namely \\astrobj{WR 55} and \\astrobj{WR 123}, for which we also have digital spectra observed with the same instrumental configuration as WR~98. This comparison indicates that WR~98 has a higher ionization degree than the WN8o star, because N{\\sc iv} emission lines in the spectrum of WR~98 are stronger and N{\\sc iii} emission lines are weaker. Actually, the WN spectrum of WR~98 mostly resembles that of WR~55, classified as WN7 by \\citet{smi96}. Thus our blue optical spectra of WR~98 would be best described as WN7o/WC. Fig.~\\ref{spect3} shows the spectra of WR\\,123, WR\\,98, and WR\\,55 for comparison.  Several faint absorption lines were detected upon the WN emission lines in our spectra of WR~98. These absorptions are H$\\gamma$ and H$\\beta$, He{\\sc i} $\\lambda\\lambda$ 4026, 4471, and 5015 \\AA\\, and He{\\sc ii} $\\lambda\\lambda$ 4200, and 4540\\AA\\,. As will be shown below, these lines belong to an OB companion. Because He{\\sc ii} absorptions appear fainter than He{\\sc i} in the OB spectrum, we presume that the companion probably is of spectral type O8-9. The luminosity class is not possible to determine from our data.  \\begin{figure} \\resizebox{\\hsize}{!}{\\includegraphics{gamenrfig1.eps}} \\caption {Continuum normalized spectrum of WR\\,98 obtained at CASLEO in 2000, September. Main emission lines are identified above, and companion OB absorptions below the spectrum} \\label{spect} \\end{figure}  \\begin{figure} \\resizebox{\\hsize}{!}{\\includegraphics{gamenrfig2.eps}} \\caption {Continuum normalized spectra of WR\\,123 (WN8o), WR\\,98, and WR\\,55 (WN7o). Note the similarity of the WN spectrum of WR~98 with that of WR~55. } \\label{spect3} \\end{figure}  \\subsection{The orbital period}  Previous results \\citep{nie91} already have shown the variability of radial velocities of the emission lines in the spectrum of WR~98. We searched for periodicities of the radial velocity variations in the strongest emission lines in our spectra, namely He{\\sc} $\\lambda$ 4686\\AA\\, and C{\\sc iii} $\\lambda$ 4648\\AA\\, using algorithms published by \\citet{mar80} and \\citet{cin95}. The most probable period obtained by both codes is 47.8 days.  Using this period as initial value, we then calculated orbital solutions for the radial velocity variations of both emission lines. We found that the solutions tend to circular orbits, with the most probable value of the period as $47.825 days \\pm 0.005$. However, other periods close to this value can not be discarded.  \\subsection{The radial velocity orbit}  Orbital elements for each emission line and the H$\\gamma$ and H$\\beta$ absorptions were determined with an improved version of the program originally published by \\citet{ber68}. With the present data, the orbits of the four emission lines, N{\\sc iv}, N{\\sc v}, C{\\sc iii}, and He{\\sc ii} have negligible eccentricity, thus we have fitted circular orbits for all our radial velocities. Circular orbital elements for the individual lines are listed in Table~\\ref{elements}, where $V_{0}$ refers to the center-of-mass velocity, $K$, to the semi-amplitude of the radial velocity variations, and $T_0$ is the time when the WR star is in the front of the system.  In our orbital fits the three best defined emission lines gave equal $T_0$ values within errors, therefore we adopt as ephemeris for the WR~98 binary system:  \\begin{centerline}{T$_0$ = 2,445,676.4+47.825E}\\end{centerline}  N{\\sc iv}, N{\\sc v}, C{\\sc iii}, and He{\\sc ii} emission lines move in phase, indicating that they are formed in the same stellar envelope. This is also the case in the two other known WN/C binaries, namely WR~145 (MR~111) and WR~153 (GP Cep) \\citep{mas89}. No detectable phase delays among emission lines are present within the errors of the orbital fits.  Semi-amplitudes of the orbital motion of the He{\\sc ii} and C{\\sc iii} emission lines appear lower than those of ionized Nitrogen emission lines. This effect is also observed in other WR binaries, e.g. WR~29 \\citep{nie00}, and may arise if  the He{\\sc ii} and C{\\sc iii} lines are partly formed in the interaction region of the binary components.  We measured H$\\gamma$ and H$\\beta$ absorptions in those spectra of WR~98 were this was possible. Radial velocities of these hydrogen lines phased with the binary period move anti-phased with the emission lines, thus indicating that they belong to an O type companion of the WR component in the binary system. Fig.~\\ref{comp} depicts the behaviour of the H$\\beta$ absorption line upon the He{\\sc ii} 4859 \\AA\\, emission in four different binary phases, illustrating the antiphased movement of the absorption and emission lines.  The circular orbital elements for the hydrogen absorption lines are included in Table~\\ref{elements}, and depicted in Fig~\\ref{f2}.   \\begin{figure} \\resizebox{\\hsize}{!}{\\includegraphics{gamenrfig3.eps}} \\caption {Continuum rectified spectra of the He{\\sc ii} 4859\\AA\\, emission with the superimposed H$\\beta$ absorption observed during four different phases of the binary system. Note the antiphased movement of absorption and emission. The spectra are intensity-shifted for a better presentation.} \\label{comp} \\end{figure}  The orbital semi-amplitudes of the radial velocity variations of the absorption lines and of N{\\sc iv}, N{\\sc v}, He{\\sc ii} and C{\\sc iii} emission lines indicate mass-ratios ($\\mathcal{M}_{WR}/\\mathcal{M}_{O}=q$) between 1  and 1.8. If a monotonic outward decreasing temperature gradient exists in the expanding WR envelope, we expect that the highest ionization emission, namely N{\\sc v}, represents better the orbital motion of the WN/C component, thus a value of $q \\sim 1$ seems more plausible. However, taking into account the rather high uncertainties in the radial velocity values of the absorption lines, these mass--ratios should not be overinterpreted. Values of minimum masses and $q$ for both components are tabulated for each emission line in Table~\\ref{elements}.   We also estimated the radii of the critical Roche lobes using the expression given by \\citet{pac71}, which resulted $r_{RL} sin~i = 79.5 R_\\odot$ for both components. Each component of the WR~98 binary system seems to be well inside its critical Roche radius.  \\begin{table*} \\label{elements} \\setlength{\\tabcolsep}{0.3mm} \\leavevmode \\caption[]{Circular Orbital Elements of WR~98} \\begin{tabular}{rl  c rcl c rcl c rcl c rcl c rcl} \\noalign{\\smallskip}\\hline\\noalign{\\smallskip} &&&\\multicolumn{15}{c}{WN/C}&&\\multicolumn{3}{c}{OB}\\\\ \\hhline{~~~--------------~~---} \\noalign{\\smallskip} \\multicolumn{2}{c}{Parameter}     \t   & & \\multicolumn{3}{c}{ ~N\\,{\\sc  iv} em.~~~}& & \\multicolumn{3}{c}{ ~N\\,{\\sc   v} em.~~~}& & \\multicolumn{3}{c}{ ~C\\,{\\sc iii} em.~~~}& & \\multicolumn{3}{c}{~He\\,{\\sc  ii} em.~~~}& & \\multicolumn{3}{c}{~absorptions~~~} \\\\ \\noalign{\\smallskip} \\hline \\noalign{\\smallskip} $P$      &  [days] & \\multicolumn{19}{c}{47.825 $\\pm$ 0.005}\\\\  $V_{0}$ &  [km\\,s$^{-1}$] &  & -41 & $\\pm$ & 4 &  & -15 & $\\pm$ & 3 &  & -47 & $\\pm$ & 2 &  & 4 & $\\pm$ & 2 &  & 6 & $\\pm$ & 15 \\\\ $K$ &  [km\\,s$^{-1}$] &  & 106 & $\\pm$ & 6 &  & 109 & $\\pm$ & 5 &  & 72 & $\\pm$ & 3 &  & 65 & $\\pm$ & 3 &  & 112 & $\\pm$ & 16 \\\\ $T_{0}$ &  [HJD]$^{(\\ast)}$ &  & 6.4 & $\\pm$ & 0.5 &  & 5.5 & $\\pm$ & 0.4 &  & 6.5 & $\\pm$ & 0.3 &  & 6.3 & $\\pm$ & 0.3 &  &   &   &   \\\\ $a\\,\\sin i$ &  [R$_\\odot$] &  & 100 & $\\pm$ & 5 &  & 103 & $\\pm$ & 5 &  & 68 & $\\pm$ & 3 &  & 61 & $\\pm$ & 3 &  & 106 & $\\pm$ & 15 \\\\ $\\mathcal{M}_{WN}~sin^{3}i$ & $[\\mathcal{M}_\\odot]$ &  & 27 & $\\pm$ & 10 &  & 28 & $\\pm$ & 10 &  & 19 & $\\pm$ & 8 &  & 18 & $\\pm$ & 7 &  &  &   &  \\\\ $\\mathcal{M}_{OB}~sin^{3}i$ & $[\\mathcal{M}_\\odot]$ &  & 25 & $\\pm$ & 7 &  & 27 & $\\pm$ & 7 &  & 12 & $\\pm$ & 3 &  & 10 & $\\pm$ & 3 &  &  &   &  \\\\ $q$ &   &  & 1.06 &   &  &  & 1.03 &   &  &  & 1.56 &   &  &  & \\multicolumn{1}{c}{1.72} &   &  &  &  &   &  \\\\ \\hline \\multicolumn{22}{l}{$\\ast$ HJD 2,445,670+: Time of conjunction, the WN star in front of the system.}\\\\ \\end{tabular} \\end{table*}  \\begin{figure*} \\resizebox{\\hsize}{!}{\\includegraphics{gamenrfig4.eps}} \\caption {Radial Velocities of He{\\sc ii}, C{\\sc iii}, N{\\sc iv}, and N{\\sc v} emission lines, and H$\\gamma$ (circles) and H$\\beta$ (triangles) absorption lines phased with the ephemeris 2,445,676.4+47.825E. Open symbols represent the photographic data, and solid symbols, the digital CCD data. Curves are the orbital solutions from Table~\\ref{elements}.} \\label{f2} \\end{figure*}  \\subsection{The radial velocity analysis of P-Cygni absorption lines}  We analysed the radial velocities of the three strongest P-Cygni absorption lines in our spectra of WR~98, namely He{\\sc i} 3888\\AA, 4471\\AA, and 5015\\AA. As the He{\\sc i} 3888\\AA\\, absorption presents a rather asymmetric profile, we measured the barycenter of this line. Because He{\\sc i} 5015\\AA\\, was not observed in the wavelengh range of the photografic spectra, the radial velocities of this line  were only determined in the digital spectra. The three P-Cygni absorption lines follow the orbital motion of the WN/C component of the binary, as illustrated in Fig~\\ref{pcyg}.  Circular orbits were fitted to the radial velocity variations of the P--Cyg absorptions. We obtained similar systemic velocities for the three absorption lines: $v_0 \\sim -1130 \\pm 15 km s^{-1}$. This systemic velocity could be considered as a lower limit of the terminal velocity of the WR stellar wind, and in fact the value agrees with the determination of terminal wind velocity of WR~98 by \\citet{een94}.  Semi-amplitudes of the radial velocity variations of the P--Cyg absorptions gave different values, namely 65 and 104 $km s^{-1}$ for He{\\sc i} $\\lambda\\lambda$ 3888, and 4471 and 5015\\AA\\, respectively. The lower semi-amplitude of the orbital motion of He{\\sc i} 3888\\AA\\, could be indicating that this line has another component, possibly originating in a common expanding envelope surrounding the binary, and which is not posible to deblend in our spectra. Similar behavior of He{\\sc i} $\\lambda$ 3888\\AA\\, low-energy metastable absorption line is observed in the spectra of other WR binary stars, e.g. \\astrobj{HD 214419} (= WR~155) \\citep[See][]{leu83}.  \\begin{figure} \\resizebox{\\hsize}{!}{\\includegraphics{gamenrfig5.eps}} \\caption {Variations of the radial velocities of two He{\\sc i} P-cygni absorption lines in the spectrum of WR~98, symbolized and phased with the same ephemeris as emission lines in Fig.~\\ref{f2}. Dashed lines represent circular orbital solutions.} \\label{pcyg} \\end{figure}"
    },
    "0208/astro-ph0208327_arXiv.txt": {
        "abstract": "{ A series of $N$-body simulations aimed to study  the dynamics of small groups of galaxies are presented. In particular, our results are compared with the dynamical properties of Hickson's compact groups (HCG). `Maximum expansion' and virial initial conditions are tested, and no primordial common dark halo is considered. The properties of small galaxy groups are very well reproduced, and those of Hickson's groups are well reproduced by the most advanced stage of collapsing  groups. We find no overmerging problem in our simulations. An important fraction of groups ($\\sim 40$\\%) initially in virial equilibrium can last for $\\sim 10$ Gyr without complete merging. These results  provide  an alternative solution to the overmerging expected in Hickson's compact groups. Also, the mass-to-light ratio of HCG are probably similar to those found in clusters, suggesting that both kinds of systems have about the same fraction of barionic to total mass. } ",
        "introduction": "Small groups are the most common galaxy associations and contain about $50$\\% of all galaxies in the universe \\cite{huchra-geller,nolthenius}. Important work has been done to compile catalogs with relevant kinematical data. Thus Nolthenius \\& White (1987) and Nolthenius (1993) found that small groups from the CfA catalog (hereafter NCfA) have the following median values: a one-dimensional velocity dispersion of $\\sigma \\approx 116$ km s$^{-1}$, a mean harmonic radius of $R_{\\rm H}= 480$ kpc, a deprojected median radius of $R_{\\rm S}=720$ kpc and a crossing time of $H_0\\tau_{\\rm c} = 0.44$.\\footnote{Here, $R_{\\rm S}\\equiv 4 R/\\pi$,   $R_{\\rm H}^{-1}\\equiv 4 \\Sigma_i \\Sigma_{i<j}\\,R_{ij}^{-1}/(\\pi N_{\\rm g}(N_{\\rm g}-1))$, and $\\tau_{\\rm c}\\equiv  2R_{\\rm H}/(\\sqrt{3}\\sigma)$, where $R$, $R_{ij}$ and $N_{\\rm g}$ are the average projected separation, the projected separation between galaxy pairs and the number of galaxies, respectively.} Hereafter a Hubble constant of $H_0=75$ km s$^{-1}$Mpc$^{-1}$ is assumed. Gourgoulhon, Chamaraux \\& Fouqu\\'e (1992, hereafter GCF) found for nearby small groups ($<80$ Mpc) a median $\\sigma\\approx 75$ km s$^{-1}$, a virial radius of about $770$ kpc and a crossing time-scale of $\\sim 4$ Gyr,  these quantities corrected for the effect of the Hubble flow. More recently, Makarov \\& Karachentsev (2000, hereafter MK) derived from a sample of 839 groups, in particular for those consisting of five galaxies, the following mean values, with quartiles: $\\sigma = 57^{+26}_{-14}$ km s$^{-1}$, a non-deprojected harmonic radius of $230^{+92}_{-99}$ kpc and a dimensionless crossing time of about $0.1$.  A special and very important case is Hickson's compact groups (HCG) which are characterized for being composed of at least four galaxies with a median projected separation of about three times the diameter of their luminous component and densities similar to the central region of rich clusters \\cite{hickson82,hickson97}. Hickson's initial catalog contained 100 groups with a median projected separation of $R=52_{2.1}^{180}$ kpc, a velocity dispersion of $\\sigma =200_{13}^{617}$ km s$^{-1}$ and a dimensionless crossing time of $H_0 t_{\\rm c}=0.016_{0.001}^{8.7}$; where $t_{\\rm c} \\equiv 4R/(\\pi \\sqrt{3}\\sigma)$, and subscripts and superscripts indicate the range of observed values. However, Sulentic (1997) pointed out that only $61$ of them show according redshifts.  It is noteworthy that HCG provide an ideal site to study the evolution of galaxies, via interactions and/or mergers, in a high density environment. One of the most striking challenges of the compact groups is how to reconcile their apparent longevity  with their quite small crossing time of about $1$\\% of the Hubble time. This short time-scale suggests a rapid evolution of these systems toward a complete merging: the so-called {\\it overmerging} problem (Hickson, Richstone \\& Turner 1977; Hickson 1982; White 1990; Hickson 1997). However, Hickson et al.~(1984) and Rubin, Hunter \\& Ford (1991) have suggested that compact groups may have formed relatively recently.  In general, two complementary approaches have been proposed to address this issue; see Athanassoula (2000) for review on this topic. On the one hand, compact groups are assumed to be formed in the early universe and survive somehow until the present epoch (Ishizawa 1986; Diaferio, Geller \\& Ramella 1994; Governato, Tozzi \\& Cavaliere 1996). Diaferio et al. (1994) introduced a scenario where a {\\it continuous} replenishment of the compact group occurs within a {\\it single} collapsing rich loose group. Governato et al. (1996) proposed an alternative model that relies in the evolution of a compact group as an ongoing and frequent process through  secondary infall of galaxies in a critical universe. However, in these scenarios merging is not suppressed so one would expect more compact groups and merger remnants than those observed today \\cite{mamon2000}.  On the other hand, there is a dynamical view where some kind of tuning of the initial conditions is imposed in order to guarantee the survival of the compact group. In this sense, Barnes (1985) was the first to investigate the effects of different kinds of initial conditions on the evolution of compact groups. He found that (1) a compact group immersed in a {\\it massive common halo} would, in general, delay its merging process if more mass is initially placed in this common halo. This merging delay results from the fact that the halo common mass is obtained from the galaxy haloes, thus reducing their masses and increasing the dynamical friction time-scale. This result was also corroborated by Bode, Cohen and Lugger (1993). And, (2), groups whose initial conditions begin in `turnaround' or `expansion' merge more rapidly than their virialized counterparts.   More recently, Athanassoula et al.~(1997) carried out a more systematic study of the parameter space for small groups (e.g.~concentrations of the common and individual haloes, common-halo-mass to total-group-mass ratios, etc.).~Their main findings are: (1) centrally concentrated groups merge faster for high common-halo to total-mass ratios and slower for low common-halo to total-mass ratios. (2) The overmerging issue of compact groups may not be a problem if {\\it appropriate} initial conditions are chosen. To support this last point, they built up a virialized group with a large common-halo to total-group and an almost homogeneous central concentration that was able to survive for as long as $\\sim \\, 20$ Gyr. According to these results the merger rates obtained from group simulations without including a common envelope, as the ones to be considered here, would be an upper limit.   In this work the hypothesis of Barnes (1989) and White (1990) regarding that diffuse groups might be the progenitors of compact ones, but criticized by Diaferio~et~al.~(1994) for lacking quantitative support, is tested; similarly, the hypothesis  advanced on observational grounds that HCG might be relatively young systems (Hickson~et~al.~1984, Rubin~et~al.~1991). The dynamical approach is used, and for this end a series of N-body simulations are performed {\\it without} including a common halo. These simulations allow us to compare with the observed kinematical properties of small groups. The rest of this paper has been organized as follows: in section $\\S 2$ the numerical methods to set up the galaxy models and the initial condition for the group are described. In section $\\S 3$ the merging histories of our simulations and the dynamical properties are given in section $\\S 4$. In section $\\S 5$ the mass estimation of our groups is addressed. A constrained representation of the `phase-space' is shown in section $\\S 6$. A more realistic N-body simulation involving spiral galaxies is given in section $\\S 7$, and section $\\S 8$ contains a general discussion of the results. Finally, the main conclusions are summarized in section $\\S 9$. ",
        "conclusions": "The main results of this work are as follows:  \\begin{enumerate}  \\item  Groups starting from `maximum expansion' with a diffuse configuration have, at the present epoch, dynamical properties very similar to the medians obtained from several catalogs of small groups. In particular, about $10$\\% of them show values close to the observed medians of Hickson's groups.  \\item The suggestion that Hickson's compacts groups originated from diffuse systems, and are relatively young systems, finds quantitative support in our numerical experiments.  \\item It is found that overmerging is not an important problem in our simulations, either  with virialized or cold initial conditions. For the case of virialized groups about $40$\\% survive merging during $\\approx 10$ Gyr. This also indicates that the existence of a large, massive common dark halo does not appear to be a strong requirement to explain the present day properties of small groups.  \\item The median mass estimator (MME) appears to be a better estimator of mass than the virial mass estimator (VME) for groups at the present epoch. It is estimated that the mass-to-light ratio of HCG is probably $\\Upsilon \\sim 250 \\Upsilon_\\odot$ suggesting that clusters and compact groups have about the same fraction of baryonic to dark matter.  \\end{enumerate}"
    },
    "0208/astro-ph0208441_arXiv.txt": {
        "abstract": "We study the mixing of chemical species in the interstellar medium (ISM).  Recent observations suggest that the distribution of species such as deuterium in the ISM may be far from homogeneous. This raises the question of how long it takes for inhomogeneities to be erased in the ISM, and how this depends on the length scale of the inhomogeneities. We added a tracer field to the three-dimensional, supernova-driven ISM model of Avillez (2000) to study mixing and dispersal in kiloparsec-scale simulations of the ISM with different supernova (SN) rates and different inhomogeneity length scales.  We find several surprising results.  Classical mixing length theory fails to predict the very weak dependence of mixing time on length scale that we find on scales of 25--500~pc.  Derived diffusion coefficients increase exponentially with time, rather than remaining constant.  The variance of composition declines exponentially, with a time constant of tens of Myr, so that large differences fade faster than small ones. The time constant depends on the inverse square root of the supernova rate.  One major reason for these results is that even with numerical diffusion exceeding physical values, gas does not mix quickly between hot and cold regions. ",
        "introduction": "Measurements of the O/H ratio in the interstellar medium (ISM) along lines of sight reaching altitudes above the galactic disk of up to 1500 pc by Meyer et al.\\ (1998) and Cartledge et al.\\ (2001) suggest that the ISM is well-mixed up to these heights. This suggestion is also supported by the O/H ratios observed in 41 H{\\rm II} regions in M101 (Kennicutt \\& Garnett, 1996), which show only a 0.1--0.2 dex dispersion about the radial gradient. In NGC 1569 and 4214 the dispersion is extremely low (Kobulnicky \\& Skillman 1996, 1997). Based on these observations it seems that there exists strong evidence pointing to a fairly well-mixed ISM in galaxies including the Milky Way.  On the other hand, recent observations using the Interstellar Medium Absorption Profile Spectrograph (IMAPS) suggest that there are local variations in the D/H ratio of up to a factor of three among the studied lines of sight (Jenkins et al.\\ 1999; Sonneborn et al.\\ 2000; for an up to date discussion on deuterium abundances see York 2002). The detection of deuterium in the ISM indicates that it was not processed into stars and the local variations reported in the D/H ratio suggest incomplete mixing in the ISM. However, such an inhomogeneous picture contradicts the almost homogeneous distribution seen in O/H ratioes.  Deuterium was produced during big bang nucleosynthesis and is destroyed during the pre-main sequence phase of stars. Thus, its ratio to hydrogen is initially constant in primordial gas and decreases as it is processed in stars. Deuterium may also be added into galactic interstellar gas by accretion of primordial gas, as required by some chemical evolution models in order to explain its present abundance in the Galaxy (see Tosi et al.\\ 1998 for details.). Mulan \\& Linsky (1999) proposed that deuterium could be formed in stellar flares from M dwarf stars. However, the production of deuterium by this mechanism should be accompanied by a significant production of lithium, which has not been observed.  Contrary to oxygen, deuterium has a primordial origin and limited destruction mechanisms, making this species a good tracer of the mixing in the ISM. Abundances of deuterium are probably not influenced by depletions onto grains, as is the case with most other species. Although there has been some suggestion that deuterium may be preferentially bonded onto the surfaces of dust grains, this effect is not sufficient to explain the large variations in D/H seen in the ISM. Furthermore, deuterium and oxygen are the only species that have an ionization fraction that closely follows that of H. Deuterium and hydrogen have almost the same ionization potentials, photoionization cross sections and recombination coefficients, making it difficult for the D/H variations to be a result of any unusual ionization effects. Other effects such as fractionation by molecular chemistry have been considered by different authors, but there is no strong evidence that such effects can be responsible for the D/H variations seen in the ISM.  We face a set of observations that point towards contradicting conclusions. O/H observations support the idea that complete mixing occurs in the ISM on timescales comparable to those for chemistry to reach equilibrium, while D/H measurements indicate a not so well mixed ISM. Solving this contradiction is an important step towards the understanding of the mixing and variation of these elements in the ISM, and therefore, of the the interpretation of local measurements as constraints of its cosmic abundance.  Roy \\& Kunth (1995) discussed the problem of mixing in the ISM and concluded that the ISM should appear more homogeneous than it does. These authors suggested that mixing in the ISM is driven by three main mechanisms acting on different length scales. On the Galactic scale, they argued that turbulent diffusion of interstellar clouds in the shear flow of galactic differential rotation could remove the inhomogeneities in the ISM in less than $10^{9}$ yr, in particular the fluctuations seen in O/H. At scales of $100\\le \\ell \\le 1000$ pc, cloud collisions and expanding supershells driven by supernovas (SNs) from evolving massive star associations, differential rotation and triggered star formation (by stellar winds and SN explosions) could re-distribute and mix inhomogeneities efficiently in about $10^{8}$~yr; and at smaller scales of the order of $1 \\le \\ell\\le 100$ pc, turbulent diffusion may be the dominant mechanism in cold clouds, while Rayleigh-Taylor and Kelvin-Helmoltz instabilities could develop in regions of gas ionized by massive stars, leading to complete mixing in less than 2~Myr. (This analysis does not take into account that during the mixing process new inhomogeneities are added into the ISM.)  Bateman \\& Larson (1993) already pointed out that turbulent mixing should be important in the ISM. However, they based their analysis on a picture of cloud collisions that ignores the likely origin of clouds as density enhancements in regions of converging flow in the turbulent ISM (e.g. V\\'azquez-Semadeni et al.\\ 1995).  Tenorio-Tagle (1996) noted that classical diffusion theory on the molecular level predicts mixing timescales in the ISM for gas with $T<10^{5}$~K several orders of magnitude longer than that proposed by Roy \\& Kunth (1995). As a consequence, Tenorio-Tagle (1996) concluded that turbulent mixing as well as hydrodynamical instabilities, as proposed by Roy \\& Kunth (1995), are important mixing mechanisms that allow the reduction of the long mixing time scales obtained from classical diffusion theory.  The observations showing that the ISM has an inhomogeneous distribution pose the question of how it is possible in a SN-driven ISM for different species not to be mixed well enough to suppress local variations of their ratio to hydrogen along and among lines of sight. The ISM is regulated by SN explosions, and therefore, by well-structured and explosive flows rather than just by diffuse turbulence that acts on the smaller scales. To understand the mixing process in such a medium we carry out direct numerical simulations of the evolution of the ISM polluted with inhomogeneities of different length scales.  We aim to understand if classical mixing length theory can be used to predict the mixing time scales as a function of size scale.  In \\S~2 we review the existing theoretical ideas on dispersion, diffusion and turbulent mixing and apply them to the ISM. Section 3 describes the modified version of the three-dimensional model of de Avillez (2000), how the runs used in the current study were set up, and the global evolution of the ISM as seen in these simulations. In \\S~4, the evolution of the tracer fields and the mixing process in the ISM are described. Section~5 discusses the results and their implications. Finally, \\S~6 presents a summary of the main results and final remarks. ",
        "conclusions": "In this paper we try to understand how long it takes to erase inhomogeneities in the ISM and if this process depends on the inhomogeneity length scale. That is, does it take longer to mix big regions than small ones? We modified the three-dimensional model of Avillez (2000) to incorporate tracer fields and ran simulations with different SN rates, distributions of SNe in the field, and numerical resolutions. (Clustered SNe from OB associations are explicitly set up, with the first SN in an association occurring in a location where the local density is greater than 1 cm$^{-3}$.)  The three-dimensional adaptive mesh refinement code used in our simulations, unlike 1-zone chemical evolution models (Matteucci et al.\\ 1999; Tosi et al.\\ 1998) and chemodynamical models (Samland et al.\\ 1997; Samland 1998; Argast et al. 2000, to name a few), enables us to resolve local inhomogeneities in the ISM with a spatial resolution as good as 1.25~pc. This allows us to test the usual assumption of Galactic chemical evolution models that assume instantaneous mixing of the ISM at all times.  Mixing in the real world takes finite time, and ultimately occurs on the molecular scale, through physical diffusion. We set a lower limit on the mixing time scale by assuming that it is set at the far larger grid scale by faster numerical diffusion.  We find that the actual value of the mixing scale does not strongly influence the mixing timescale.  Matter can enter and leave the section of the Galaxy under study through the top and bottom outflow boundary conditions, as well as moving to neighboring periodic boxes (eg passing through a periodic side boundary condition), although we do not yet include Galactic rotation in our model.  Chemical evolution models do not normally allow radial transport of matter between different rings, but it occurs in our models.  Because the domain extends to 10 kpc on both sides of the midplane, gas can cycle between the disk and the halo in a Galactic fountain as a result of the dynamics of the flow in the gravitational field of the Galaxy (see Kahn 1981; Avillez 1999).  The mixing time scales of the disk gas are regulated mainly by the SNe, because the disk gas that enters the disk-halo cycle takes some 200 Myr to return to the disk (Kahn 1981). Furthermore before reaching the thin gas disk, the descending gas hits the thick gas disk and decelerates due to drag (Kahn 1981; Benjamin \\& Danly 1997). Therefore, most of the descending gas will not contribute strongly to the mixing process of the ISM with $\\left|z\\right|\\leq 250$ pc.  This work only studies the mixing process as a way of erasing inhomogeneities in the ISM. It does not take into account continuing pollution of the ISM by SNe, planetary nebulae, and stellar winds, which we will study in future work.  The actual enrichment of the disk gas depends on the sources and the efficiency of mixing.  If the ejected metals are quickly distributed over a large volume, spatially homogeneous enrichment takes place. If the mixing volume is small, the ISM in the vicinity of a SN will be highly enriched, while large parts of the disk remain metal-poor until the gas mixes out on time scales of hundreds of Myr or more. In this case the ISM may remain chemically inhomogeneous, with newly formed stars having different compositions depending on where they form.  Over gigayear timescales, mixing does appear to be quite effective over scales of many kiloparsecs.  This work shows that inhomogeneities that are present in the ISM, assuming that no further inhomogeneities are introduced into the system, can take a long time to be erased. If inhomogeneities are introduced at a time larger than the time scale for mixing, the system will show temporarily an homogeneous state, which ends when inhomogeneities are introduced. On the other hand, if inhomogeneities are introduced on a time scale smaller than the mixing time, then the system will show an inhomogeneous distribution during its evolution. Therefore, from the simulations we conclude that for the ISM to show a homogeneous distribution, no inhomogeneities should be introduced over periods of roughly a hundred megayears.  This is longer than typical star formation time scales, so new SNe and pre-main sequence stars can continue to maintain an inhomogeneous composition so long as star formation continues.  The rate at which inhomogeneities are erased increases with the SN rate to values about ten times greater than the Galactic rate.  Above that rate, no cold medium remains, mixing occurs quite quickly, but above this threshold, further increasing the SN rate does not markedly further increase the mixing efficiency.  The time for complete mixing is already longer than any time scale for chemical equilibrium to occur and therefore the ISM will show an uneven distribution. Even if the major fraction of the ISM gas is completely mixed, lower level inhomogeneities will still be seen as abundances are measured more carefully. Furthermore, as the mixing time scales are longer than the time interval for SNe type II to occur and enrich the ISM with metals, low-level inhomogeneities are continuously maintained, leading to a poorly mixed ISM, except in the hot regions where the mixing process is fast (of order 50~Myr).  For a galactic SN rate, the ISM is likely fairly inhomogeneous, since the SNe are spatially well separated and erasing these inhomogeneities takes a long time (of the order of 350 Myr). A single SN event can thus substantially influence its surroundings.  If we consider, for example, a 20~pc sphere with number density $n=1$ around a SN with 1~M$_{\\odot}$ of metals in its ejecta, the ejecta represent a fraction $\\sim 0.1 (Z/Z_{\\odot})$ of the metals in the region.  Thus, individual SNe can drastically change the local metallicity in low metallicity gas.  Stars formed before mixing has finished can have a wide range of metallicities; further work will be needed to make quantitative predictions of the expected dispersion for comparison with observations, however.  The smallest mixing time scale that we find in our study is $\\sim 120$ Myr and occurs for $\\sigma/\\sigma_{\\rm gal}\\ge 20$.  After some $80\\%$ of the gas has been mixed, the simulations show that the ISM remains inhomogeneous for at least twice as long on scales smaller than a kpc. Even when the rate of SNe is increased to ten times the Galactic rate, this uneven abundance distribution structure takes tens of Myr to disappear.  At the finest grid scale we suppress the turbulent diffusivity, as numerical viscosity becomes the dominant diffusion mechanism.  At these scales there would physically be a continuing turbulent mixing that we do not resolve. The resolution study shown in Figures~\\ref{mixtimesnrate} and~\\ref{mixtimeresolution} demonstrates, however, that the mixing time is practically independent of the actual value of the diffusion scale.  Changing the linear diffusion scale by factors of as much as four produced less than 20\\% changes in the diffusion time.  This means that we can effectively assume that the mixing is dominated by an inertial scale decoupled from the diffusion scale, so that the details of how the diffusion occurs do not affect the larger-scale behavior."
    },
    "0208/astro-ph0208488_arXiv.txt": {
        "abstract": "We model an accretion disk atmosphere and corona photoionized by a central X-ray continuum source.  We calculate the opacity and one-dimensional radiation transfer for an array of disk radii, to obtain the two-dimensional structure of the disk and its X-ray recombination emission. The atmospheric structure is extremely insensitive to the viscosity $\\alpha$.  We find a feedback mechanism between the disk structure and the central illumination, which expands the disk and increases the solid angle subtended by the atmosphere. We apply the model to the disk of a neutron star X-ray binary. The model is in agreement with the $\\sim 12 \\degree$ disk half-angle measured from optical light curves. We map the temperature, density, and ionization structure of the disk, and we simulate high resolution spectra expected from the {\\it Chandra} and {\\it XMM-Newton} grating spectrometers. X-ray emission lines from the disk atmosphere are detectable, especially for high-inclination binary systems. The grating observations of two classes of X-ray binary systems already reveal important spectral similarities with our models. The model spectrum is dominated by double-peaked lines of H-like and He-like ions, plus weak Fe L. The line flux is proportional to the luminosity and is dominated by the outer radii.  Species with a broad range of ionization levels coexist at each radius: from \\ion{Fe}{26} in the hot corona, to \\ion{C}{6} at the base of the atmosphere. The line spectrum is very sensitive to the temperature, ionization, and emission measure of each atmospheric layer, and it probes the heating mechanisms in the disk. We assume a hydrostatic disk dominated by gas pressure, in thermal balance, and in ionization equilibrium. As boundary conditions, we take a Compton-temperature corona and an underlying Shakura-Sunyaev disk. The choice of thermally stable solutions strongly affects the spectrum, since a thermal instability is present in the regime where X-ray recombination emission is most intense. ",
        "introduction": "When the infall of matter into a deep gravitational potential is mediated by an accretion disk, gravitational energy is converted to the thermal radiation which powers both low mass X-ray binary (LMXB) systems and active galactic nuclei (AGN). Accretion disks present unique problems involving magnetized plasma dynamics, photoionization, atomic kinetics, thermal and ionization equilibria, general relativity, and radiation transfer. The accretion disks in LMXBs and AGN are expected to have many common properties. The compactness of the accretor in LMXBs and AGN and their inferred accretion rates imply temperatures of $T> 10^7$~K and intense X-ray emission in the inner disk region. The inner radii of these disks, as well as the disk atmosphere as a whole, is substantially more ionized than the case in which the accretor is a white dwarf. In both LMXBs and AGN, the vast energy emitted in the inner disk region is reprocessed in the outer disk, where the external radiative heating can dominate the local thermal emission. The subsequent photoionization of the disk plasma radically alters its equilibrium state, structure, and spectrum, especially in the atmospheric and coronal disk layers, which are the subject of this study.  High resolution X-ray spectroscopy is an essential tool to study the physics of this \"hot class\" of accretion disks and the conditions near black hole event horizons. In this paper, we concentrate on the outer radii of disks in neutron star LMXBs, since current observational constraints provide more stringent tests for LMXBs than for AGN. The following points support these assertions:  \\begin{itemize}  \\item High resolution spectra can reveal discrete emission or absorption from atomic transitions within the accretion disk plasma, providing information on the accretion disk structure, dynamics, and physics. These spectra open a window into photoionized gases and their phase equilibria.  \\item X-rays, and in particular discrete atomic transitions of hydrogen- and helium-like ions, probe the regions in the disk with the highest levels of ionization.  Regions closer to the compact object will have the highest ionization levels, although vertical stratification is also expected.  \\item The knowledge of the accretion disk physics directly impacts our ability to probe the physical conditions around the compact object. For example, the Fe K emission originating in the innermost regions of an accretion disk has been proposed as a direct probe of the general relativistic effects near a black hole event horizon in AGN, by virtue of the observed characteristic line shape \\citep{tanaka1995}. However, very little is known about the physical conditions in the Fe K emission region, and it is still unclear how our ignorance of the physical processes within the accretion disk affect the modeled Fe K line profile and flux. It is also unclear whether the soft X-ray line features reported by \\citet{brandu} are feasible.  \\item While in neutron star LMXBs the photoionizing source must be near the neutron star surface, in AGN and galactic black hole candidates (BHC) the location of the ionizing source is unknown. In AGN, various authors have assumed the ionizing source to be located in the rotation axis of the black hole, above the disk midplane, possibly close to the base of a jet. Alternatively, an ionizing source might be present on the upper layers of the disk, perhaps due to disk flares, or to Comptonization of thermal UV photons in the accretion disk corona (ADC).  \\item LMXB systems are observed in less crowded regions than AGN.  \\item In contrast to AGN, LMXBs often have measured orbital parameters which constrain the geometry of the system, such as the maximum disk radius.  LMXBs may also have a measured value of the disk inclination, while orbital phase and eclipse phase variations provide tomographic information.  For example, measurements of the optical light curve amplitudes of LMXBs have yielded estimates of the angle subtended by the disk of $\\sim$ 12$\\degree$ \\citep[]{dejong}.  \\item To our knowledge, the inner disk radii can only be studied in the X-ray band. Theoretically, the thermal emission of the inner disk radii must peak in the X-ray band for both LMXBs and AGN. The optical and UV emission originates in the outer regions of the accretion disk and further away from the compact object, and the radio and gamma ray emission is likely dominated by emission from jets.  \\end{itemize} Fewer physical ingredients are needed to model the outer radii of disks than the inner radii, so the logical progression is to successfully model the outer radii first. The X-ray spectroscopy of neutron star LMXBs will allow us to construct a physical picture of their accretion disks, which we can then use to investigate the inner disk in BHC and AGN.  To fully exploit the high energy-resolution X-ray spectra of accretion disks, we created physical disk models and calculated synthetic spectra. The disk plasma, at $10^5$ to $10^7$~K, cools through atomic line emission that can be detected with space-borne X-ray observatories such as {\\it Chandra} and {\\it XMM-Newton}. Modeling the equilibrium state of the plasma and the radiation transfer within the disk allows a calculation of the disk structure and its X-ray spectrum. A synthetic spectrum can be compared to the data.  The model spectrum is unique in that it is calculated purely on physical, and not just phenomenological, grounds.  We describe four fiducial disk models. Two of these models were introduced in \\citet[]{jimenez}. We use a newly developed adaptive-mesh disk structure calculation, the \\citet{ray93} photoionized plasma code, and a new X-ray emission code which uses $HULLAC$ data (Hebrew University/Lawrence Livermore Atomic Code, \\citet[]{hullac}). The models consist of a disk illuminated by a pure neutron star continuum, and they contain as boundary conditions a Compton-temperature ADC at the top of the disk and a modified \\citet[hereafter abbreviated as SS73]{ss73} disk at the bottom \\citep{vrtilek}. Thus, the region of interest has a temperature and ionization which is intermediate of these two regions, and it emits copious X-ray radiation.  In section \\ref{sub:line-lmxb}, we introduce the X-ray line observations prior to {\\it Chandra} and {\\it XMM-Newton}; in section \\ref{sub:raddisk}, we introduce theoretical work on the structure of X-ray illuminated accretion disks; in section \\ref{sec:modelatm}, we describe the disk structure calculations and the assumptions of hydrostatic, thermal and ionization equilibrium; in section \\ref{sec:instab}, we detail the effects of a thermal instability on a layer of the disk atmosphere throughout the disk; in section \\ref{sec:adamodel}, we discuss the calculation of the high resolution spectrum, which is done {\\it a posteriori} from the structure calculation; in section \\ref{sec:diskstruct}, the disk density, temperature and ionization structure are presented; in section \\ref{sec:spec}, the model spectra are shown, assuming a full, partial or obstructed view of the neutron star region, and we show simulated spectra utilizing the response of the {\\it XMM-Newton} reflection grating spectrometer (RGS) and the {\\it Chandra} medium energy gratings (MEG); in section \\ref{sec:disc}, comparisons of the model to the observed X-ray spectra of LMXBs are discussed briefly, and we discuss the limitations of the model. In section \\ref{sec:conclusions}, concluding remarks are presented. ",
        "conclusions": "\\label{sec:conclusions}  We have calculated the hydrostatic structure of a photoionized accretion disk atmosphere which is in thermal equilibrium and ionization balance. We also determined the atmosphere's thermal stability and its observable high resolution X-ray recombination emission spectrum. \\begin{itemize} \\item {\\it A feedback mechanism between illumination and atmospheric structure enlarges the atmosphere}. The disk atmosphere is orders of magnitude less dense than the disk midplane. The atmosphere extends for a few tens of disk pressure scale heights (if the pressure scale height is calculated using the disk midplane temperature). Illumination heats and expands the disk atmosphere, increasing the number of absorbed photons in the atmosphere and heating it further, producing further expansion of the atmosphere, and so on. The expansion stops because the atmosphere becomes optically thin, cooling and contracting. The inclusion of the feedback mechanism increases the size and the line emission flux of the atmosphere by an order of magnitude. The atmospheric thickness is much larger than the standard $\\alpha$-disk model thickness, and it is consistent with the $\\sim 12 \\degree$ subtended semi-angle deduced from optical modulations in LMXBs. The disk atmosphere thickness also explains the under-abundance of eclipsing LMXBs.  \\item {\\it The disk atmosphere subtends a large solid angle $0.07 \\lesssim \\Omega/4\\pi \\lesssim 0.2$}. If the inclination is $i \\gtrsim 80 \\degree$, the disk photosphere (which subtends $0.04 \\lesssim \\Omega/4\\pi \\lesssim 0.08$) may shield the neutron star flux, producing an ADC source with partial eclipses or without eclipses altogether. The disk $\\Omega$ depends weakly on the neutron star luminosity, but $\\Omega$ scales linearly with disk radius. The disk recombination luminosity scales linearly with $\\Omega$.  \\item {\\it The atmospheric structure is independent of the viscosity parameter $\\alpha$}. The viscosity changes the density in the optically-thick part of the disk, producing a small shift in atmospheric height, but this has no effect on the X-ray spectrum.  \\item {\\it The X-ray spectra are dominated by lines from H-like and He-like ions} of abundant elements from C to Fe, as well as RRC and weak Fe L lines. The line ratios are a sensitive probe of the atmospheric and coronal structure.  \\item {\\it Clear spectral signatures of photoionization are present, as well as temperature, density, and radiation field diagnostics.} An intercombination to resonance line ratio of $\\sim 4$ is modeled for low-Z He-like ion line triplets. RRC are unequivocal signposts of photoionization. The density diagnostics from He$\\alpha$ lines of low-Z and intermediate-Z elements are degenerate with the presence of an intense UV radiation field from the disk itself, so the $R$ ratio may not give conclusive signatures of high-density in LMXBs. Much of the disk atmosphere is close to the photosphere, such that the dilution factor of the UV field is small. The He$\\alpha$ density diagnostics could operate at the densities predicted by the disk atmosphere model in LMXBs, of $10^{13} \\lesssim n_e \\lesssim 10^{15}$ cm$^{-3}$.  \\item {\\it The line fluxes are nearly proportional to the X-ray continuum luminosity}. The disk line fluxes decreased by a factor of 20 when the system luminosity was decreased by a factor of 10 (to $L=10^{37.3}$~erg~s$^{-1}$). The atmospheric density was reduced by a factor of $\\sim 5$, its optical depth was reduced, and the atmosphere was $\\sim 2$ times less extended than in the high-luminosity ($L=10^{38.3}$~erg~s$^{-1}$) case.  \\item {\\it The line equivalent widths depend strongly on inclination}.  The relative obscuration of the neutron star affects the equivalent width and detectability of the disk X-ray emission. The modeled disk emission is almost undetectable when the neutron star continuum is also in the line of sight. As such, high inclination systems, or systems with dips or ADC, are more likely to show X-ray lines due to enhanced contrast. This expected trend has been largely confirmed by {\\it Chandra} and {\\it XMM-Newton} observations. We have demonstrated that for a highly absorbed neutron star continuum in our fiducial system, the disk X-ray lines are detectable with both the {\\it Chandra} and {\\it XMM-Newton} grating spectrometers.  \\item {\\it Double-peaked X-ray lines can be detected for $r=10^{10}$~cm disks, but larger $r=10^{11}$~cm disks may appear blended in a single peak in the grating spectra.} The emission line region spans several orders of magnitude in disk radius.  The modeled line profiles are needed to deduce the outer disk radius.  Line emission from the outer regions of the disk dominates.  The emission increases with disk radius, and the Doppler broadening of the lines decreases for larger $r$. The wings of the broadest lines are lost in the continuum, decreasing their apparent equivalent width.  \\item {\\it The resonance line optical depths can be measured.} If the $r$ line in He-like ions has a value which differs from the model calculations, it may be due to resonant scattering of continuum photons. The line ratios in the Lyman series can also work as optical depth diagnostics. We have not included these effects on the current version of the model, but our results for ionic column densities show that appreciable line optical depths are present, and hence this process will be included in future versions of the code.  \\item {\\it The continuum optical depth of the atmosphere is generally small ($\\tau \\ll 1$), except for photons which propagate nearly parallel to the disk plane.} The atmosphere is optically thick to X-ray continuum photons from the neutron star. However, most of the recombination line emission is not appreciably affected by the continuum opacity.  \\item {\\it The spectrum is sensitive to a thermal instability present in photoionized gases.} By forcing all the chosen solutions to be thermally stable, a break in the temperature, density, and ionization structure is created. Measurably different X-ray spectra are obtained depending on the resolution of this instability. The shape of RRC profiles, which in the models show multiple temperature components, and the relative intensity of lines such as \\ion{Mg}{11}, are useful diagnostics of the stable temperature regime.  \\end{itemize}  The spectra obtained with the {\\it Chandra} HETG and the {\\it XMM-Newton} RGS show that the plasmas in LMXBs are photoionized, as our model assumes. Two LMXBs (4U1626-67 and EXO0748-67) show kinematic signatures of accretion disk emission \\cite[]{schulz2001,exo0748}. The line fluxes, line profiles, the ionization distribution, density, and RRC temperatures, provide a wealth of diagnostic capability for the identification of accretion disk atmospheres and their properties. The spectral comparisons with the data are promising, and they will be addressed in future work."
    },
    "0208/astro-ph0208394_arXiv.txt": {
        "abstract": "We present precise constraints on the normalization of the power spectrum of mass fluctuations in the nearby universe, $\\sigma_8$,  as a function of the mean local matter density, $\\Omega_{\\rm m}$. Using the observed local X-ray luminosity function of galaxy clusters from the extended BCS and REFLEX studies, a mass-luminosity relation determined from Chandra and ROSAT X-ray data and weak gravitational lensing observations, and the mass function predicted by the Hubble Volume simulations of Evrard \\etal,  we obtain $\\sigma_8 = (0.508\\pm0.019) \\,\\Omega_{\\rm m}^{-(0.253\\pm0.024)}$,  with $\\Omega_{\\rm m} < 0.34$ at 68 per cent confidence.  The degeneracy between $\\sigma_8$ and $\\Omega_{\\rm m}$ can be broken using Chandra measurements of the X-ray gas mass fractions in dynamically relaxed clusters. Using this information and including Gaussian priors on the mean baryon density of the universe and the Hubble constant,  we obtain $\\sigma_8=0.695\\pm0.042$ and $\\Omega_{\\rm m}=0.287\\pm0.036$, for an assumed flat $\\Lambda$CDM cosmology (marginalized 68 per cent confidence limits). Our results are in good agreement with some recent studies based on the local X-ray temperature function of clusters, the redshift evolution of the X-ray luminosity and temperature functions of clusters, early results from the Sloan Digitized Sky Survey, the most recent results from studies of cosmic shear, and combined analyses of the 2dF galaxy redshift survey and cosmic microwave background anisotropies. ",
        "introduction": "The X-ray luminosity function of galaxy clusters in the nearby universe  provides a powerful cosmological probe. The observed luminosity function,  $n(L)$, can be combined with a relation linking the observed X-ray luminosity and mass, and the mass function, $n(M)$, predicted by simulations, to obtain  tight constraints on the combination of cosmological parameters $\\Omega_{\\rm m}$ and $\\sigma_8$, where $\\Omega_{\\rm m}$ is the mean matter density of the local universe and $\\sigma_8$ is the root-mean-square (rms) variation of the density field smoothed by a top hat window function of size 8$h^{-1}$Mpc.  Observationally, the keys to such studies are precise determinations of the local X-ray luminosity function of clusters and the relation linking the observed X-ray luminosities and total masses. X-ray selection currently offers the best way to identify massive galaxy clusters, and the local X-ray luminosity function has now been precisely determined by the BCS (Ebeling \\etal 1997; Ebeling \\etal 2000)  and REFLEX (B\\\"ohringer \\etal 2002) studies. The flux-limited BCS and REFLEX  samples, which  are based on data from the ROSAT All-Sky Survey (RASS; Tr\\\"umper 1993), together include $\\sim 750$ clusters and cover approximately two thirds of the sky.  Recently, significant effort has also been invested into measuring the local temperature function of clusters (\\eg Markevitch 1998; Pierpaoli \\etal 2001; Ikebe \\etal 2002), which offers a complementary  method for determining cosmological parameters. At present, however, the temperature function samples are significantly smaller than the combined BCS-plus-REFLEX luminosity function data set, and the analysis of the temperature function is complicated by the fact that these samples are selected according to X-ray flux as well as temperature, requiring the use of both mass-temperature and temperature-luminosity relations in the analysis.  Recent years have also seen significant efforts directed towards a precise calibration of the `virial' relations linking the observed luminosities, temperatures and masses of galaxy clusters (\\eg Horner, Mushotzky \\& Scharf 1999; Nevalainen, Markevitch \\&  Forman 2000; Finoguenov, Reiprich  \\& B\\\"ohringer 2001; Allen, Schmidt \\&  Fabian 2001b; Sanderson \\etal 2003). In particular, the launch of the Chandra X-ray Observatory has permitted the first precise measurements of the temperature and mass profiles of relaxed clusters from X-ray data.  Using a combination of Chandra  and gravitational lensing data, Allen \\etal  (2001b) confirmed that luminous, relaxed galaxy clusters follow the simple scaling relations predicted by theory, but that the normalization  of the observed mass-temperature relation measured within $r_{2500}$ (where the mean enclosed mass density is 2500 times the critical density of the universe at the redshifts of the clusters) is approximately 40 per cent lower than predicted by standard adiabatic simulations. This highlights the likely importance of additional physics such as cooling and pre-heating in the intracluster gas (see also  Pearce \\etal 2000; Thomas \\etal 2002; Voit \\etal 2002; Muanwong \\etal 2002).  Theoretically, the primary requirement for cosmological studies using the observed luminosity and/or temperature functions  of clusters is a precise prediction of the mass function. This has now been achieved for flat $\\Lambda$CDM (and $\\tau$CDM) cosmologies using the Hubble Volume simulations  of Jenkins \\etal (2001) and Evrard \\etal (2002).  In this paper we present precise constraints on $\\sigma_8$  and $\\Omega_{\\rm m}$ based on the observed local luminosity  function of the most X-ray luminous clusters in the extended BCS  (Ebeling \\etal 2000) and REFLEX samples, and a new calibration, using pointed Chandra and ROSAT X-ray observations and weak gravitational lensing results, of the  mass-luminosity  relation linking the masses of clusters measured within $r_{200}$ to their total $0.1-2.4$ keV ROSAT luminosities.  Having determined our combined constraint on $\\sigma_8$ and $\\Omega_{\\rm m}$, we show that the degeneracy between these parameters can be broken using Chandra results on the X-ray gas mass fractions in the most dynamically relaxed clusters. Including Gaussian priors  on the  mean baryon density of the universe ($\\Omega_{\\rm b}h^{2} = 0.0205\\pm0.0018$; O'Meara \\etal 2001), the Hubble constant ($h=0.72\\pm0.08$; Freedman \\etal 2001),  and a theoretical bias factor ($b=0.93\\pm0.05$; Bialek, Evrard \\& Mohr 2001) relating the asymptotic baryon fraction in the most X-ray luminous clusters to the mean value for the universe as a whole, we obtain  $\\sigma_8=0.695\\pm0.042$ and $\\Omega_{\\rm m}=0.287\\pm0.036$ (marginalized 68 per cent confidence limits for an assumed flat $\\Lambda$CDM cosmology). We compare our results to other measurements based on the local number density of clusters, evolution of the X-ray luminosity and temperature functions, the 2dF galaxy redshift survey, cosmic microwave background anisotropies, and measurements of cosmic shear.  Throughout this paper, a flat $\\Lambda$CDM cosmology with $\\Omega_{\\Lambda} = 1 - \\Omega_{\\rm m}$ is assumed.  In order to facilitate a direct comparison with previous X-ray studies, results on the masses, X-ray luminosities and X-ray gas mass fractions of individual clusters are quoted for a Hubble parameter $h =H_0/100$\\kmpspMpc $= 0.5$ or, equivalently, $h_{50} =H_0/50$\\kmpspMpc $= 1.0$. ",
        "conclusions": ""
    },
    "0208/astro-ph0208513_arXiv.txt": {
        "abstract": "The usual interpretation of the spectrum of a BALQSO is that a broad--band continuum from the central engine plus broad emission lines from a surrounding region (the BELR) emerges from near the center of the QSO, and broad absorption lines are superimposed in a separate outlying region (the BALR).  For \\1214, designated as an FeLoBAL QSO because its spectrum contains numerous absorption features from excited states of Fe~II, we explore an alternative interpretation based on resonance scattering.  In this model line emission and absorption occur in the same line--forming region (the LFR).  A resonance--scattering synthetic spectrum computed with the parameterized supernova synthetic--spectrum code SYNOW fits the spectrum of \\1214 rather well, so the resonance--scattering model merits further study.  Some implications of the model and its possible applicability to other QSOs are briefly discussed. ",
        "introduction": "BALQSOs --- quasars that have broad absorption lines in their spectra --- can be divided into LoBALs and HiBALs, which do and do not have strong absorption lines produced by low--ionization species such as Mg~II, Al~III, and Fe~II.  LoBALs that have numerous lines from excited states of Fe~II are called FeLoBALs (Becker et~al. 1997, 2000).  The usual interpretation of the spectrum of an FeLoBAL QSO is similar to that of the spectra of other BALQSOs.  The effective continuous spectrum consists of a true continuum from the accretion disk, plus broad emission lines (BELs) that form in a surrounding region --- the BELR.  Broad absorption lines (BALs) are superimposed on the effective continuum as it passes through a separate region --- the BALR.   The distance of the BALR from the QSO center may be large compared to the size of the BELR, and the global covering factor --- the fraction of the sky covered by the BALR as viewed from the center of the QSO --- may be small.  In this {\\sl Letter} we explore an alternative interpretation of the spectrum of a particular FeLoBAL QSO.  In our simple model, the rest--frame UV spectrum consists of P~Cygni features formed by resonance scattering, superimposed on a continuum.  The line emission and absorption components come from the same spherically--symmetric line--forming region --- the LFR.  To illustrate this interpretation we concentrate on an FeLoBAL that was discovered in the First Bright Quasar Survey (White et~al. 2000) and is designated FIRST J121442.3+280329, or \\1214 for short.  We choose this particular FeLoBAL because it has recently been analyzed in detail, in the context of the usual BAL model, by de~Kool et~al. (2002; hereafter dK02), and because its spectrum, although rich in lines, appears to be amenable to an analysis based on a single LFR.  In \\S2 we describe the modeling of \\1214 by dK02.  The alternative resonance--scattering interpretation is presented in \\S3. Some of the implications of the resonance--scattering model, and its possible applicability to other QSOs, are briefly discussed in \\S4. ",
        "conclusions": ""
    },
    "0208/astro-ph0208039_arXiv.txt": {
        "abstract": "{Black holes;AGN;quasars;galaxy evolution} Violent activity in the nuclei of galaxies has long been considered a curiosity in its own right; manifestations of this phenomenon include distant quasars in the early Universe and comparatively nearby Seyfert galaxies, both thought to be powered by the release of gravitational potential energy as material from the host galaxy accretes onto a central supermassive black hole (SMBH).  Traditionally, the broader study of the formation, structure and evolution of galaxies has largely excluded active galactic nuclei. Recently however, this situation has changed dramatically, both observationally and theoretically, with the realisation that the growth and influence of the SMBH, the origin and development of galaxies and nuclear activity at different epochs in the Universe may be intimately related.  The most spectacular fireworks seen in distant quasars, may be relatively easy to explain since the era of greatest quasar activity seems to coincide with turbulent dynamics at the epoch of galaxy formation in the young, gas-rich Universe.  Ubiquitous black holes are believed to be a legacy of this violent birth.  Alternatively, black holes may be the seeds which drive galaxy formation in the first place. Closer to home, and hence more recently in the history of the Universe, a fraction of comparatively ordinary galaxies, similar to our own, have re-ignited their central engines, albeit at a lower level of activity. Since these galaxies are more established than their younger and more distant counterparts, the activity here is all the more puzzling. Whatever the mechanisms involved, they are likely to play an important role in galaxy evolution.  I review the intriguing evidence for causal links between supermassive black holes, nuclear activity and the formation and evolution of galaxies, and describe opportunities for testing these relationships using the next generation of earth-bound and space-borne astronomical facilities. ",
        "introduction": "Over the last 50 years, astronomers have been intrigued by enormously energetic objects called Active Galactic Nuclei (AGN), a violent phenomenon occurring in the nuclei, or central regions, of some galaxies with intensities and durations which cannot easily be explained by stars, thus providing some of the first circumstantial evidence for theoretically-predicted supermassive black holes. Despite their intriguing properties they were largely viewed as interesting but unimportant freaks in the broader study of galaxy formation and evolution, leading astronomers studying the properties of galaxies to exclude the small fraction of galaxies with active centres as irritating aberrations.  Here I describe the discovery of AGN and the variety of classifications that followed; I describe some features of unifying models of the central engine that attempt to explain the varied properties of different AGN classes that give rise to the classification.  The search for supermassive black holes in AGN and non-active galaxies is discussed along with the developing realisation that all galaxies with significant bulge components might harbour dormant supermassive black holes as remnants of a past adolescent period of quasar activity and therefore posses the potential to be re-triggered into activity under the right conditions, making nuclear activity an integral part of galaxy formation and evolution. ",
        "conclusions": ""
    },
    "0208/astro-ph0208380_arXiv.txt": {
        "abstract": "We investigate how galactic disks react to external tidal torques.  We calculate the strength and radial dependence of torques on disks that arise from a misalignment between the disk and the main axis system of a flattened dark matter halo. Density profile, misalignment and flattening of the halo are chosen to match the corresponding values typically found for dark matter halos in large cosmological N-body simulations. We find that except for in the very inner regions, the torques are well-described by a power law of the form $\\tau \\propto r^{-2.5}$. For torques as they arise in typical cosmological settings, the magnitude of the torque is large enough for the entire disk to react to the torque in less than the Hubble time.  We demonstrate analytically that disks which are originally located in the $xy$-plane and which are subjected to a torque around the $x$-axis tilt around the $y$-axis, as also found in fully non-linear N-body simulations. We further demonstrate that that the torque causes the radius of a chosen particle to increase with time. Investigations of tilting disks which treat the disk as a set of solid rings thus may systematically overestimate the effects of the torque by a factor of two.  For torques of the form we investigate, the inner regions of the disk react to the torque faster than the outer regions, resulting in a trailing warp. We then study the effect of the self-gravity of the disk in such a scenario using numerical N-body models.  Self-gravity flattens out the inner regions of the disk, but these regions are tilted with respect to their initial plane followed by a non-flat outer region whose tilt decreases with radius. The ``warp radius,'' which marks the end of the inner flat disk, grows throughout the disk at a rate that depends only on the strength of the torque and the local surface density of the disk. ",
        "introduction": "Many edge-on disk galaxies show integral-sign or S-shaped warps, where the majority of the disk is planar but where the outer region of the disk lies above that plane on one side of the galaxy and below the plane on the other \\citep{binney92}. The Milky Way is warped both in neutral hydrogen \\citep[e.g.][]{diplas and savage91} and in the stellar distribution \\citep{reed96,lopez-corredoira et al02b}. \\citet{reshetnikov and combes98} estimate that half of all disk galaxies have optical warps, and most HI disks which extend beyond the optical disks appear to be warped \\citep{bosma81,briggs90,christodoulou et al93}.  A number of methods have been proposed for creating and maintaining warps. Several authors \\citep{lynden-bell65,sparke and casertano88} have suggested that the system of discrete particles which make up the disk may have normal bending modes that could be excited. \\citet{battaner et al} have investigated magnetic fields as a cause of warps, while \\citet{toomre83} and \\citet{dekel and shlosman83}  suggested that disks askew in flattened dark matter halos could develop long-lived warps, assuming the radial profile of the halo were appropriately fine-tuned.  Following on the idea that infalling material will shift the angular momentum of a galaxy \\citep{ryden88,quinn and binney92}, \\citet{ostriker and binney89} studied how a disk of massive rings reacts to a slewing disk potential. They found that the reaction of the disk depends largely on its surface density, with warps appearing in regions of low surface density. Also motivated by the cosmic infall of material with differing angular momentum, \\citet{debattista and sellwood99} found that when the angular momentum of a halo and disk are misaligned, dynamical friction transfers angular momentum between them in such a way as to produce a long-lived warp.  More recently, \\citet{lopez-corredoira et al02a} examined the torques produced by the transfer of angular momentum from gas falling onto a galactic disk, and the warping of the disk in response to these torques as well as the internal torques due to the interaction of different rings within the disk. They find a disk response which mirrors the Milky Way warp, though the effects of embedding the disk in a dark halo are as yet insufficiently modelled.  In a cosmological setting, a galactic disk experiences gravitational torques from three different sources: external galaxies, non-spherically-symmetric substructure in the halo, and misalignment between the disk and halo; if the dark matter halo is not spherical, as suggested both by simulations \\citep[e.g.][]{fwed85,katz91,dubinski and carlberg91,cole and lacey96} and observations \\citep{sackett99}, then it will exert a gravitational torque on a disk not in its symmetry plane.  In this paper, we evaluate the effect of the gravitational tidal torques a typical galactic disk experiences from its misalignment with the halo, and study whether these torques provide a possible origin for warped disks. We first develop a framework for how orbits in a disk react to torques. We then investigate the form and magnitude of torques due to flattened misaligned halos. The effect of these torques on massless disks is examined by analytic and numerical techniques. Finally, we perform N-body simulations of massive disks resembling the Milky Way and investigate how the self-gravity of the disk may affect its reaction to cosmological torques. ",
        "conclusions": ""
    },
    "0208/astro-ph0208349_arXiv.txt": {
        "abstract": "{The cumulative light curves of Gamma Ray Bursts (GRBs) smooth the spiky nature of the running light curve.  The cumulative count increases in an approximately linear way with time t for most bursts.  In 19 out of 398 GRBs with T$_{90} >$ 2 s, the cumulative light curve was found to increase with time as $\\sim$ t$^{2}$ implying a linear increase in the running light curve . The non-linear sections last for a substantial fraction of the GRB duration, have a large proportion of the cumulative count and many resolved pulses that usually end with the highest pulse in the burst.  The reverse behaviour was found in 11 GRBs where the running light curve decreased with time and some bursts are good mirror images of the increases. These GRBs are among the spectrally hardest bursts observed by BATSE. The most likely interpretation is that these effects are signatures of black holes that are either being spun up or down in the accretion process. In the spin up case, the increasing Kerr parameter of the black hole allows additional rotational and accretion energy to become available for extraction. The process is reversed if the black hole is spun down by magnetic field torques. The luminosity changes in GRBs are consistent with the predictions of the BZ process and neutrino annihilation and thus provide the link to spinning black holes. GRBs provide a new window for studying the general relativistic effects of Kerr black holes. ",
        "introduction": "It is suspected that spinning black holes reside in a variety of astrophysical sources.  Frame dragging creates a special region called the ergosphere in which any material or energy must rotate in the same direction as the black hole.  The energetic reactions near the black hole maybe responsible for relativistic jets in active galactic nuclei (AGN), microquasars and gamma ray bursts \\citep{koide:2002,frail:2001}. In GRBs the source of the enormous energy in gamma rays maybe the cataclysmic formation of a spinning black hole involving mergers of compact objects such as neutron star (NS) binaries or NS and black holes \\citep{piran:1999,ruffjan:1999} and also during or after the collapse of massive stars \\citep{macfad:1999,pacy:1998,vietri:1998,reeves:2002}. The central engine is hidden from view and only gravitational radiation and neutrinos may escape and reach the observer directly from the engine.  A key feature of the internal shock model is that the observed gamma rays reflect the variability of the central engine and the GRB duration may be determined by the engine \\citep{reemes:1994,piran:1999}. The cumulative output in gamma rays of a burst indirectly reflects the output of the central engine via a relativistic jet.  The advantage of using the cumulative light curve is that it reveals the trends by smoothing the spiky nature of the running light curve.  The cumulative light curves of most bursts can be approximated by a linear function of time and GRBs may be regarded as relaxation systems that continuously accumulate energy in the reservoir and discontinuously release it \\citep{mcbreenb:2002}. In a relatively small number of GRBs, the cumulative light curves depart from linearity in a consistent way.  The selection and properties of these GRBs are presented here in sections 2 and 3, discussed in section 4 and summarised in section 5. ",
        "conclusions": "The results of this study of BATSE light curve of GRBs are presented.  The analysis revealed that the running light curve of a minority of bursts increase or decrease approximately linearly with time.  These results are interpreted as possible evidence for spin-up and spin-down of black holes during the burst.  The BZ process and neutrino annihilation are consistent with these results.  The brute force appearance of Kerr black holes being spun up an down by hyperaccretion in GRBs contrasts with the more sedate and longer lasting indications of black holes in MCG-6-30-15 and other sources \\citep{wilms:2001,miller:2002,woo:2002,koide:2002}. \\small"
    },
    "0208/astro-ph0208455_arXiv.txt": {
        "abstract": "A review of basic properties of Be/X-ray binaries is presented. These systems (called also hard X-ray transients), which form the most numerous class of massive X-ray binaries in the Galaxy, are composed of Be stars and neutron stars (X-ray pulsars) on wide (P$_{orb} \\sim 17 - 263$ d), usually eccentric (e $\\sim 0.1 - 0.9$) orbits. The systems contain two quasi-Keplerian ($\\mid v_r \\mid /v_{orb} \\la 10^{-2}$)  discs: decretion disc around Be star and accretion disc around neutron star. Both discs are temporary: decretion disc disperses and refills on time scales $\\sim$ years (dynamical evolution of the disc, formerly known as the \"activity of a Be star\"), while accretion disc disperses and refills on time scales $\\sim$ weeks to months (which is related to the orbital motion on an eccentric orbit and, on some occasions, also to the major instabilities of the other disc). Accretion disc might be absent over a longer period of time ($\\sim$ years), if the other disc is very weak or absent. The X-ray emission of Be/X-ray binaries has distinctly transient nature and is controlled by the centrifugal gate mechanism, which, in turn, is operated both by the periastron passages (Type I bursts) and by the dynamical evolution of the decretion disc (both types of bursts). The X-ray pulsars in these systems rotate at equilibrium periods (with the possible exception of the slowest pulsars). Be/X-ray binaries are excellent laboratories for investigation of both the evolutionary processes in the two discs and of the evolution of the neutron star rotation. ",
        "introduction": "At present, there are no doubts, that Be/X-ray systems (or hard X-ray transients)  dominate the population of the massive X-ray binaries. These systems, whose primaries are Be stars and secondaries are neutron stars were initially believed to be just atypical cases of massive X-ray binaries (typical cases were supergiant systems). The final Uhuru catalogue (Forman et al., 1977) listed just 1 such system (as opposed to 6 supergiant systems). Catalogue of Bradt et al. (1978), which was compilation of Uhuru, SAS-3, Ariel, Copernicus and HEAO-A sources, listed 5 Be/X-ray systems (and 8 supergiant systems). During my previous Vulcano talk on Be/X-ray binaries, 9 years ago (Zi\\'o{\\l}kowski, 1992), I listed 13 such binaries. Van Paradijs' (1995) catalogue contained 14 Be/X-ray systems (and 14 supergiant systems). In this review, I will present a list of 63 Be/X-ray binaries (while the number of the presently known supergiant systems is 20).  As one can see, the number of the known Be/X-ray systems is growing fast and, due to the transient nature of their emission, it is likely to continue its fast growth in the future. At the same time, the number of the known supergiant systems (in our Galaxy) is already saturated and is not expected to increase substantially,  One may notice that the situation in the field of the low mass X-ray binaries is similar. The fastest growing class of these systems is the group of soft X-ray transients (or X-ray Novae). These systems, composed typically of a black hole and a low mass optical companion, are being discovered at a high rate (again, due to the transient nature of their X-ray emission). In near future, they will, probably, dominate the population of the low mass X-ray binaries. It seems, that we have already detected almost all strong permanent X-ray sources in the Galaxy. In the future, we will detect mostly transient sources (both massive hard X-ray transients and low mass soft X-ray transients). ",
        "conclusions": ""
    },
    "0208/astro-ph0208105_arXiv.txt": {
        "abstract": "{ The possible association of the supernova remnant (SNR) \\object{G\\,343.1$-$2.3} with the pulsar \\object{PSR B\\,1706$-$44} (superposed on the arclike ``shell\" of the SNR) has been questioned by some authors on the basis of an inconsistency between the implied and measured (scintillation) transverse velocities of the pulsar, the absence of any apparent interaction between the pulsar and the SNR's ``shell\", and some other indirect arguments.  We suggest, however, that this association could be real if both objects are the remnants of a supernova (SN) which exploded within a mushroom-like cavity (created by the SN progenitor wind breaking out of the parent molecular cloud). This suggestion implies that the actual shape of the SNR's shell is similar to that of the well-known SNR \\object{VRO\\,42.05.01} and that the observed bright arc corresponds to the ``half\" of the SNR located inside the cloud. We report the discovery in archival radio data of an extended ragged radio arc to the southeast of the bright arc which we interpret as the ``half'' of the SN blast wave expanding in the intercloud medium. ",
        "introduction": "The pulsar \\object{PSR B\\,1706$-$44} (Johnston et al.\\ \\cite{joh92}) is superposed on an incomplete arc of radio emission (McAdam et al.\\ \\cite{mca93}).  McAdam et al.~interpreted this arc as a shell-type supernova remnant (SNR), named \\object{G\\,343.1$-$2.3}, and suggested that the SNR is physically associated with PSR B\\,1706$-$44. This suggestion was questioned by Frail et al.\\ (\\cite{fra94a}) and Nicastro et al.\\ (\\cite{nic96}; see, however, Dodson et al.\\ \\cite{dod01}).  Usually, a particular claimed pulsar/SNR association is considered reliable if the following five criteria are fulfilled (e.g. Kaspi \\cite{kas96}): \\begin{enumerate} \\item agreement of independent distance estimates for pulsar and SNR;   \\item agreement of independent age estimates for pulsar and SNR;   \\item consistence of the implied pulsar transverse velocity (i.e.~the velocity inferred by the displacement of the pulsar from the geometrical centre of the associated SNR) with the measured (proper motion and/or scintillation) velocity;   \\item existence of any sign of interaction between the pulsar and the SNR;   \\item ``correct\" (inferred or measured) orientation of the vector of pulsar transverse velocity (it is assumed that this vector should be pointed away from the geometrical centre of the associated SNR). \\end{enumerate}   Although the distance and age estimates for PSR B\\,1706$-$44 and G\\,343.1$-$2.3 are in reasonable agreement, the implied transverse velocity is at least an order of magnitude larger than the scintillation velocity calculated by Nicastro et al.\\ (\\cite{nic96}). This inconsistency along with the absence of any apparent interaction between the pulsar and the SNR constitute the two main arguments against the physical association between these two objects (Frail et al.  \\cite{fra94a}, Nicastro et al.\\ \\cite{nic96}). The fifth criterion is not applied to the system, since the direction of the pulsar proper motion is still unknown (cf. Giacani et al.\\ \\cite{gia01} with Frail et al.\\ \\cite{fra94a}; see also Sect.  3.5). Additional (indirect) arguments against the association are based on Gaensler \\& Johnston's (\\cite{gae95}) statistical study, which suggests that young pulsars cannot overrun their parent SNR shells (Nicastro et al.\\ \\cite{nic96}; see however Arzoumanian et al.\\ \\cite{arz02}) and on the large extent of the ``halo\" around the pulsar (Frail et al.\\ \\cite{fra94a}).   In this paper we show how the existing observational data on PSR B\\,1706$-$44 and G\\,343.1$-$2.3 can be interpreted in favour of their physical association (Sect.~2) and discuss the criteria for evaluating the reliability of pulsar/SNR associations as applied to this system (Sect.~3).  The main suggestion of the paper is that the association between PSR B\\,1706$-$44 and the SNR G\\,343.1$-$2.3 could be real if both objects are the remnants of a SN which exploded within a mushroom-like cavity created by the SN progenitor wind breaking out of the parent molecular cloud (Sect.~2.2).  This suggestion implies that in addition to the known bright ``half\" of the SNR G\\,343.1$-$2.3 there should exist a more extended and weaker component, so that the actual shape of G\\,343.1$-$2.3 is similar to that of the well-known SNR \\object{VRO\\,42.05.01}. It is remarkable that the 2.4 GHz Parkes Survey of Duncan et al.\\ (\\cite{dun95}) shows the existence of such an extended component. ",
        "conclusions": "We have analyzed the available observational data on the pulsar PSR B\\,1706$-$44 and the SNR G\\,343.1$-$2.3 and suggested that these objects could be the remnants of a SN which exploded within a mushroom-like cavity created by the SN progenitor wind breaking out of the parent molecular cloud.  This accounts for the disparity between the measured and implied velocities of the pulsar. Our suggestion implies that in addition to the known bright ``half\" of the SNR G\\,343.1$-$2.3 there should exist a more extended and weaker component, so that the actual shape of G\\,343.1$-$2.3 is similar to that of the well-known SNR VRO\\,42.05.01. We have found such a component in archival radio data. Further observations, such as those discussed in Sect.\\ 2.2, would be useful to confirm or reject the association between this component and the SNR G\\,343.1-2.3."
    },
    "0208/astro-ph0208043_arXiv.txt": {
        "abstract": "Massive  structures, such  as  galaxies, act  as strong  gravitational lenses on background sources. When  the background source is a quasar, several lensed images are  seen, as magnified or de-magnified versions of the same object.  The detailed study of the image configuration and the  measurement  of  ``time-delays''  between  the  images  yield estimates of the  Hubble parameter $H_0$.  We describe  in  a simple way  the phenomenon of strong lensing and  review recent progress made in the field, including microlensing by stars in the main lensing galaxy. ",
        "introduction": "\\input QSOlens_imform  \\input QSOlens_macex  \\input QSOlens_micro  \\input QSOlens_cos  \\input QSOlens_degen ",
        "conclusions": ""
    },
    "0208/astro-ph0208018_arXiv.txt": {
        "abstract": "We compare the distribution of molecular gas and star formation activity in the bar region of six spirals (NGC 2903, NGC 3627, NGC 4321, NGC 5457, NGC 6946, \\& IC 342) from the BIMA Survey of Nearby Galaxies (SONG). The molecular gas, traced using the CO (J=1--0) emission line, is brightest along the leading edge of the stellar bar in the bar dust lanes.  The star formation activity, traced using the \\Ha emission line, is offset towards the leading side of the CO emission.  A cross-correlation analysis shows that a) the HII regions are offset 0--800 pc on the leading side of the CO emission, b) the largest offsets are found in the strongest bars, and c) there is a wide range in offsets in a single bar with no systematic pattern as a function of the galacto-centric radius.  The CO-\\Ha offset constrains how stars may form depending on the gas flow.  We examine possible star formation scenarios in context of the two main classes of bar gas flow simulations, the N-body/sticky particle and hydrodynamic models. Though both model gas flows are generally consistent with the observed offsets, we suggest the inclusion of a two- or multi-phase medium to improve the agreement between models and observation. ",
        "introduction": "\\label{intro}  Barred spiral galaxies are ideal laboratories for the study of star formation because they host a variety of environments with distinctive star formation activity and gas dynamics. These environments include the circumnuclear, inner and outer rings (see discussion of rings in \\citealt{buta96,regan02}), the bar ends, and the bar region itself, located in between the bar ends and the nucleus. From region to region, the star formation activity can vary dramatically: bars have star formation rates of $\\sim$0.1--0.4 \\Msun \\yr\\ (e.g., \\citealt{martin97}), whereas in circumnuclear rings, in an area 10-30 times smaller, star formation rates may be as high as 1 \\Msun \\yr\\ (e.g., \\citealt{buta00a}).  Since star formation occurs in molecular clouds, comparative studies of the distribution of molecular gas and star formation activity in different environments can shed light on how star formation may be induced or inhibited.  Along the spiral arms, for example, observations find that the molecular gas and HII regions are usually not co-spatial; HII regions are preferentially offset towards the leading side of the molecular gas or dust lanes \\citep{vogel88,rand93,knapen96,loinard96}.  This offset is interpreted as evidence of star formation induced by a spiral density wave \\citep{vogel88,rand93}.  However such offsets are not universally present.  In M100, \\citet{sempere97} find that the offset is absent, or even inverted, along the spiral arms.  Still, in other cases, the offset is more pronounced between the HII regions and the dust lanes as the molecular gas and dust lanes diverge (e.g., \\citealt{lord91,rand99}). \\citet{rand99} attribute the divergence to heating of the gas by young stars or cosmic rays, or to a two-phased molecular gas medium, but admit that neither of these explanations works satisfactorily.  In barred spirals, most previous studies comparing the molecular gas distribution and star formation activity have focused on the highly active circumnuclear region (e.g., \\citealt{eckart91,roy93,kenney93,sakamoto95,benedict96}).  In this paper, we focus on the unique region between the bar ends and the nucleus. This region is unlike any other in the galactic disk because it is dominated by highly elliptical stellar orbits (see reviews by \\citealt{sellwood93,a92a,a92b}). Throughout this paper, we refer to it simply as the {\\sl bar}.  Only one previous study has studied the location of HII regions relative to the stellar bar \\citep{martin97}; they found that the ``\\Ha bar'' was usually offset towards the leading side of the stellar bar with misalignments as large as 15$^{\\circ}$.  Studies of molecular gas or dust in bars have found that the gas and dust are also on the leading side of the stellar bar (e.g., \\citealt{ondrechen85,handa90,regan95,downes96,sheth00}).  But the relationship between the molecular gas (or dust) and star formation has only been studied in a few cases.  In M101, for example, \\citet{kenney91} note that the molecular gas and \\Ha emission are at the same position angle.  In contrast, \\citet{sheth00} found that the HII regions in the bar of NGC 5383 are offset towards the leading side of the bar dust lanes.  It is unclear whether such offsets are common in bars.  Since bars have distinctive (and well-studied) gas kinematics, a study of the relative distribution of the gas and stars can further elucidate the complex phenomenon of star formation.  With this goal, we have studied six barred spirals from the recently completed BIMA Survey of Nearby Galaxies (SONG) \\citep{regan01, helfer02b}.  The sample selection is discussed in \\S \\ref{sample}, and the observations and data reduction in \\S \\ref{obs}.  Using the CO (J=1-0) emission line to trace the molecular gas, and the ionized hydrogen (\\Ha) line to trace recent star formation activity, we compare the relative distribution of the two in \\S \\ref{qual}.  In all six galaxies we find that the majority of the \\Ha emission is offset towards the leading side of the molecular gas.  We quantify the offset using one and two dimensional cross-correlation analysis in \\S \\ref{quant}.  The CO-\\Ha offset constrains how stars may form depending on the gas flow into the dust lane.  We discuss the results of the cross-correlation analysis in the framework of the two main classes of bar gas flow models (the N-body sticky particle and hydrodynamic models) in \\S \\ref{disc}, and summarize our results in \\S \\ref{conc}.  \\subsection{Sample Selection} \\label{sample}  In BIMA SONG (see details of the survey in \\citealt{regan01,helfer02b}), we detect CO emission in 27 of the 29 barred spirals (e.g., Figures A1 and A2 in \\citealt{sheth01}). Typically, the emission is detected in the circumnuclear region, where it is usually the brightest. In some bars, CO emission is also detected at the bar ends, along the bar, and even in inner rings \\citep{regan02}.  Since we are mainly interested in studying star formation in the bar region we limit ourselves to those BIMA SONG galaxies in which {\\em both} CO and \\Ha emission are clearly detected over a significant portion of the bar. Six galaxies satisfy this criterion: NGC 2903, NGC 3627, NGC 4321, NGC 5457, NGC 6946 and IC 342. Global properties of these six are listed in Table \\ref{tab1}.  Though small, our sample spans a range of Hubble types with 1 Sb, 2 Sbc and 3 Scd galaxies. This range may be important because most of the observed differences in star formation activity in bars occur between early and late Hubble type galaxies (e.g., \\citealt{elmelm85,ohta86}). For instance, early Hubble type bars have low star formation activity along the bar and high star formation activity at the bar ends, whereas late Hubble type galaxies have higher star formation activity in the bar, but have a gap in star formation at the bar ends \\citep{phillips96}.  All the galaxies in our sample are classified as SAB in the RC3 catalog. However this does not mean that they are all weak bars because the Hubble classification of SAB types is not rigorous; in fact, a recent analysis of infrared data has shown that the true fraction of strong bars may be as high as 56\\%, higher than the typically quoted 33\\% in the RC3 \\citep{eskridge00}.  The strength of a bar may be correlated with its Hubble type.  \\citet{elmelm85} concluded that early Hubble type galaxies have stronger bars because these bars are longer, relative to their disks, and have flat profiles; these galaxies also have strong spiral arm patterns. However, bar strength is a difficult parameter to quantify; other structural properties such as the bar ellipticity and bulge size are also important (see discussion in \\citealt{buta00b}).  A good indicator of the bar strength is the shape of the bar dust lanes because it reflects the gas response to the stellar bar. \\citet{a92b} showed that strong bars have relatively straight dust lanes whereas weaker bars have curved dust lanes. Extending her analysis to our sample, we infer that NGC 2903 and NGC 3627 are strongly barred because of their straight dust lanes. NGC 5457 and NGC 4321 with slightly curved dust lanes are weaker bars. IC 342 has an even more curved dust lanes indicating an even weaker bar. In NGC 6946, the northern dust lane appears to be straight but the southern dust lane is curved so it is difficult to classify this bar solely on the shape of the dust lanes.  \\citet{regan95} suggest that if there is bar in NGC 6946, it is rather weak. Using bar dust lanes as a measure of bar strength, we find that the three latest Hubble types in our sample are classified as relatively weak bars.  In summary, our sample contains a broad range of bar strengths, from strong bars in NGC 2903 and NGC 3627, to intermediate strength bars in NGC 4321 and NGC 5457, to relatively weak bars in NGC 6946 and IC 342. ",
        "conclusions": "\\label{conc}  We have investigated the distribution of molecular gas and star forming regions in the bars of six spirals from the BIMA Survey of Nearby Galaxies. Our main conclusions are as follows: \\\\  \\noindent 1. The CO emission is brightest along the leading edges of the bar. Weak spurs of CO emission are seen on the trailing side of the dust lanes.  These spurs may play a role in star formation upstream of the dust lane. At the bar ends, strong CO and \\Ha emission are seen on the trailing and leading sides; these may be the beginnings of inner rings.  \\noindent 2. The \\Ha emission is distributed in compact and diffuse structures. There are a few instances where the \\Ha is coincident with, or on the trailing side of the CO emission.  But the main result is that in all six cases, the majority of the \\Ha emission is offset towards the leading side of the CO.  \\noindent 3.  We quantify the offsets using a cross-correlation analysis and find a range of 0--800 pc, with larger offsets in stronger bars.  However, in a given bar there is a range of offsets and there is no systematic pattern as a function of the galacto-centric radius.  \\noindent 4. In the two dimensional cross-correlation analysis, there is a tendency for the \\Ha emission to be offset radially outwards from the CO emission.  However, this trend is less significant than the azimuthal offsets because correlations in two-dimensions can be between physically unrelated HII and CO regions.  \\noindent 5. The observed CO-\\Ha distributions may be explained by either the N-body/sticky particle models or the hydrodynamic models with different, plausible, star formation scenarios. In the context of the N-body simulations, stars may form via cloud-cloud agglomeration in the dust lanes. In the context of the hydrodynamic models, the stars could form in dust spurs on the trailing side of the dust lane. We suggest that addition of a two-phased or multi-phased molecular medium can improve the agreement between these and previous observations, and gas flow models in bars.  \\centerline{\\bf Acknowledgements}  This work would not have been possible without the rest of the SONG team members (T. Wong, T. Helfer, L. Blitz and D. Bock) and the dedicated observatory staff at Hat Creek and at the Laboratory for Millimeter-wave Astronomy at the University of Maryland.  We thank S. Aalto, S. H\\\"uttemeister, J. Kenney, E. Ostriker, N. Scoville, E. Schinnerer, J. Stone, and T. Treu for invaluable and insightful discussions about gas kinematics and star formation.  We are grateful to N. Reddy and E. Schinnerer for their careful reading and helpful comments which significantly improved this paper.  Research with the BIMA array is supported by NSF grant AST-9981289. Support for the Laboratory for Millimeter-wave Astronomy is also provided by the state of Maryland.  This research is also partially funded by NSF grant AST-9981546.  \\clearpage"
    },
    "0208/astro-ph0208532_arXiv.txt": {
        "abstract": "\\noindent We present a generalization of the concept of magnification bias for gravitationally-lensed quasars, in which the quasars are selected by flux in more than one wavelength band.  To illustrate the principle, we consider the case of two-band selection, in which the fluxes in the two bands are uncorrelated, perfectly correlated, or correlated with scatter. For uncorrelated fluxes, we show that the previously-held result---that the bias is the product of the single-band biases---is generally false.  We demonstrate some important properties of the multi-band magnification bias using model luminosity functions inspired by observed correlations among X-ray, optical, infrared and radio fluxes of quasars.  In particular, the bias need not be an increasing function of each flux, and the bias can be extremely large for non-linear correlations.  The latter fact may account for the high lensing rates found in some X-ray/optical and infrared/radio selected samples. ",
        "introduction": "\\label{sec:intro}  If a massive galaxy lies along the line of sight to a background quasar, the galaxy may act as a gravitational lens, magnifying and forming multiple images of the quasar.  Beginning with the pioneering work of \\citet{tog84}, many authors have computed the number of lenses that should appear in well-defined samples of quasars, with particular attention given to the dependence of this statistic on the vacuum energy density \\citep{turner90,kochanek96,helbig99,sarbu01,li02}.  These calculations must take into account not only the probability that a massive galaxy will be aligned closely enough with a background quasar (the lensing cross-section), but also the enhancement of the quasar flux due to lensing (the magnification bias).  This is because quasar samples are usually defined by observed flux in some wavelength band, and gravitational lensing boosts the observed flux, thereby sampling a fainter portion of the quasar luminosity function.  For example, if intrinsically faint quasars are sufficiently more numerous than bright quasars, then a quasar with a given observed flux is more likely to be lensed than the cross-section alone would imply.  More recently attention has turned toward the information about galaxy mass profiles that can be gleaned from lens statistics. These statistics include lensing rates \\citep[see, e.g.,][]{keeton01b,wyithe01,li02}, the ratio of four-image to two-image lenses \\citep[see, e.g.,][]{rusin01b,finch02}, the image separation distribution \\citep[see, e.g.,][]{kochanek01} and the brightness distribution of central images \\citep[see, e.g.,][]{rusin01,keeton01,keeton02a,evens02,oguri02}. All of these applications of lens statistics require a good understanding of magnification bias.  \\citet{blr91} noted that quasar samples selected by both radio and optical flux measurements are subject to what they called a ``double magnification bias.''  If the radio and optical fluxes from a given quasar are nearly independent, then quasars bright in both bands are especially likely to be lensed\\footnote{ Note that the important property of the two bands is independence, not a large separation in wavelength (as has since been stated in the literature; \\cite{bade97}), although of course these two properties are related.}. By assuming that gravitational lensing produces only one possible value of magnification, and using power-law luminosity functions for the optical and radio bands, \\citet{blr91} showed that the resulting two-band magnification bias is the product of the bias factors computed separately for each band.  It is timely to revisit the issue of multi-band magnification bias with a more general approach.  With the advent of large-area sky surveys at many wavelengths, it has become possible to define samples of thousands of quasars by their observed fluxes in X-ray, optical, infrared, and radio bands.  Quasars appear in large numbers in, for example, the RASS \\citep[ROSAT All-Sky Survey:][]{truemper82,voges99} and eventually ChaMP \\citep[Chandra Multi-wavelength Project:][]{wilkes01,silverman02} at X-ray wavelengths; NVSS \\citep[NRAO-VLA Sky Survey:][]{condon98} and FIRST \\citep[Faint Images of the Radio Sky at Twenty centimeters:][]{bwh95,white97} at radio wavelengths; 2MASS \\citep[Two Micron All Sky Survey:][]{kleinmann94} at near-infrared wavelengths; and SDSS \\citep[Sloan Digital Sky Survey:][]{york00,schneider02} at optical wavelengths. Cross-correlation of these catalogs \\citep[see, e.g.,][]{mcmahon01,ivezic02} will become an increasingly important source of information about quasars in general, and gravitational lens statistics in particular.  A few lenses have already been discovered using multi-band selection criteria, at lensing rates that are larger than the 0.2--1\\% typical of single-band lens surveys. \\citet{bade97} discovered the gravitational lens RX~J0911.4+0551 by matching RASS sources with optical sources from Schmidt plates. Of the $\\sim40$ radio-quiet X-ray--luminous high-redshift quasars known, two are lensed \\citep{wbb99}. A search for very red quasars through the matching of FIRST and 2MASS has identified two gravitational lenses out of thirteen sources \\citep{gregg02,lacy02}.  None of these projects were designed explicitly to discover gravitational lenses, although this is a realistic possibility for the future.  In this paper we investigate the magnification bias for quasar samples defined by measurements in multiple wavelength bands.  After presenting the basic formalism for $N$ bands (\\S\\ref{sec:mbias}), we specialize to the case of two bands and consider some illustrative examples.  We consider the cases in which the two fluxes are uncorrelated (\\S\\ref{subsec:independent}), perfectly correlated (\\S\\ref{subsec:perfectly-correlated}), and correlated with non-zero scatter (\\S\\ref{subsec:correlated}).  We then use a realistic model of the optical luminosity function for quasars to demonstrate a few interesting properties of the multi-band magnification bias (\\S\\ref{sec:examples}); in particular, the bias does not necessarily increase with flux in each band, and there is a profound difference between the case of a linear correlation and a non-linear correlation with flux in another band. Finally, in \\S\\ref{sec:discussion} we summarize our results, and discuss possible applications of this formalism to real quasar samples. ",
        "conclusions": "\\label{sec:discussion}  The multi-band magnification bias is an {\\em a posteriori} statistic. It is used to estimate the probability that the apparent luminosities of a given quasar, as measured in several bands, are due to gravitational magnification, rather than being intrinsic to the quasar.  When a sample of quasars is selected through the matching of sources in two different catalogs, both fluxes must be used to perform this calculation.  One must also have some knowledge of the intrinsic correlation (if any) of the fluxes, and the distribution of magnifications produced by lensing.  One might expect that the multi-band bias is maximized when the bands are uncorrelated (an example of which is shown for the radio-optical correlation of SDSS early data release quasars in Fig.~\\ref{fig3}), since in that case there is no redundant information in the flux measurements.  Upon further reflection, or using the mathematics developed in this paper, one realizes that this is not true---the relevant information is how discrepant the observed fluxes are from the correlation, and whether the discrepancy can be made smaller if the observed fluxes are all reduced by a constant factor.  Many of the illustrative examples presented in this paper approximate certain correlations that have been observed for real quasars.  In particular, the multi-band magnification bias may result in very high lens fractions for certain quasar samples.  First, we consider the case of a quasar sample selected by optical colors.  The top left panel of Fig.~\\ref{fig3} is a logarithmic plot of SDSS $i$-band {\\em vs.} $r$-band fluxes for the SDSS early data release quasars (Schneider et al.~2002).  The data show a linear correlation with scatter, the magnification bias for which is illustrated in the top panels of Fig.~\\ref{fig:bias-linear-correlation}. The magnification bias must be computed using both optical measurements, unless the sample is 100\\% complete in one filter (i.e., unless after selecting quasars in $i$, the $r$-band magnitude was measured in every single case).  As an example consider the sample of SDSS $z>5.8$ quasars (Fan et al.~2001). Since the $z$-band selection is at $\\sim1100\\AA$ in the rest-frame, quasars with fixed absolute B magnitude ($\\sim4400\\AA$) are more likely to be selected if they are bluer than average. Thus a sample of quasars selected in this manner will be bluer than average and lie blueward of the correlation on a plot of the intrinsic correlation between $M_{\\rm V}$ and $M_{\\rm B}$. The magnification bias for these sources may be significantly smaller than that computed using only extrapolations of the $B$-band luminosity function \\citep{wl02,chs02}.  Next, we consider examples of non-linear luminosity correlations. \\citet{brinkmann00} measured the correlation between ROSAT X-ray and FIRST radio fluxes for matched quasar samples. They find that while radio-quiet quasars show a linear relationship between X-ray and radio luminosity, radio-loud quasars have an X-ray flux $L_x$ that varies with the radio luminosity $L_{\\rm r}$ as $L_x\\propto L_{\\rm r}^{0.48\\pm0.05}$ with an intrinsic scatter of $\\sim0.2$ dex. Furthermore, \\citet{brinkmann00} showed that the X-ray luminosity correlates non-linearly with optical luminosity $L_{\\rm o}$, following $L_x\\propto L_{\\rm o}^{1.42\\pm0.09}$. The second of these correlations (in flux) is plotted in Fig.~\\ref{fig3} for quasars in the SDSS Early Data Release \\citep{schneider02}, but can be seen more clearly in Fig.~14 of \\citet{brinkmann00}. The multi-band magnification bias corresponding to the second correlation\\footnote{Note that while we have presented results for a non-linear correlation with index $\\gamma>1$, the result for $\\gamma<1$ is simply obtained through reversal of the axes.} is illustrated in the top panels of Fig.~\\ref{fig:bias-non-linear-correlation}. The magnification bias can be extremely large for sources that are luminous at both optical and X-ray wavelengths. This may be the explanation for the apparently high probability of lensing in bright X-ray selected quasar catalogs \\citep{bade97}. The location of the gravitational lens RX~J0911.4+0551, which was selected from cross-correlation of optical and X-ray catalogs, is shown on this plot by the large dot\\footnote{We used the integrated $R$ from Bade et al.~(1997) and color transformations from Fukugita, Shimasaku \\& Ichikawa~(1995).}. The fluxes place the quasar below the correlation, in the region where we expect the magnification bias to be large (see the upper panels of Fig.~\\ref{fig:bias-non-linear-correlation}). The lens HE 1104--1805 is also X-ray loud (Wisotzki et al.~1993; Reimers et al.~1995). While this lensed quasar was not discovered through cross-correlation between catalogs, it is interesting to note its location on this plot, shown by the open square in Fig.~\\ref{fig3}. The quasar is found to be very bright in both bands, and is again in the region of high magnification bias. It is suggestive that the two X-ray loud gravitational lenses both appear to lie in the region of high magnification bias, as expected.  Finally, the multi-band magnification bias may also provide an explanation for the large gravitational lens fraction (2 out of 13) found through the matching of FIRST and 2MASS sources \\citep{gregg02,lacy02}. Fig.~\\ref{fig3} shows the correlation for near infrared luminosities verses radio luminosities compiled from table~1 of \\citet{bh01}. The radio/near-IR correlation appears to be steeper than linear. If true we might expect very large biases for luminous near-IR sources. The top panel of Fig.~\\ref{fig:bias-non-linear-correlation} demonstrates magnification bias for a non-linear correlation, and shows that the bias of around 100 necessary to achieve a lens fraction of 2/13 is possible.    This paper has discussed the multi-band magnification bias for gravitational lensing with arbitrary luminosity functions in several bands.  Previous discussion of the multi-band magnification bias \\citep{blr91} focused on the case where the fluxes in the two bands are independent.  If a single value for the lens magnification is considered, they showed that this assumption leads to a multiple magnification bias that is equal to the product of the single-band biases.  However, we have shown that this equality breaks down in the more realistic case when there is a distribution of possible magnifications.  We also discussed the multi-band magnification bias when the fluxes in the various bands are correlated.  In the case of a perfect (i.e.\\ zero scatter) linear correlation, the information from the second band does not change the magnification bias. However, if the correlation is non-linear, then sources with fluxes that obey the correlation cannot be lensed. On the other hand, sources with fluxes that do not obey the correlation must be lensed.  Of course, real correlations have intrinsic scatter. We have calculated the multi-band magnification bias for bi-variate luminosity functions with finite scatter about both linear and non-linear correlations.  For a linear correlation (as expected for a quasar sample selected by optical colors) we find that the magnification bias is an increasing function of either flux.  Calculations of lens statistics from incomplete color-selected quasar samples should therefore account for the multi-band magnification bias.  Non-linear correlations (and anti-correlations) with finite scatter were also explored. If the fluxes in two bands are correlated through a relation that is steeper than linear, then sources that lie below the correlation can be subject to a very large bias. The observed correlation between X-ray and optical flux (and possibly between infrared and radio flux) for quasars is steeper than linear. Suggestively, the two known X-ray loud gravitationally lensed quasars lie below the X-ray/optical correlation in the region of large magnification bias. Thus the multiple magnification bias may provide an explanation for the large lensing rates found in X-ray/optical and infrared/radio selected samples."
    },
    "0208/astro-ph0208197_arXiv.txt": {
        "abstract": "{ We present circular polarization measurements of circumstellar H$_{2}$O masers. The magnetic fields in circumstellar envelopes are generally examined by polarization observations of SiO and OH masers. SiO masers probe the high temperature and density regime close to the central star. OH masers are found at much lower densities and temperatures, generally much further out in the circumstellar envelope.  The circular polarization detected in the (6$_{16}$--5$_{23}$) rotational transition of the H$_{2}$O maser can be attributed to Zeeman splitting in the intermediate temperature and density regime.  The magnetic fields are derived using a general, LTE Zeeman analysis as well as a full radiative transfer method (non-LTE), which includes a treatment of all hyperfine components simultaneously as well as the effects of saturation and unequal populations of the magnetic substates. The differences and relevances of these interpretations are discussed extensively. We also address a non-Zeeman interpretation as the cause for the circular polarization, but this is found to be unlikely. We favor the non-LTE analysis. The H$_2$O masers are shown to be unsaturated, on the basis of their line widths and the lack of linear polarization. The field strengths are compared with previous detections of the magnetic field on the SiO and OH masers. Assuming a $r^{-2}$ dependence of the magnetic field on the distance to the star, similar to a solar-type magnetic field, our results seem to indicate that we are probing the highest density maser clumps at the inner edge of the H$_2$O maser region. This allows us to estimate the density of the clumps, and the extent of the H$_2$O maser region. We show that the magnetic pressure dominates the thermal pressure by a factor of $20$ or more. We also give an order of magnitude estimate of the magnetic field on the surface of the stars.  In particular we discuss the differences between Supergiants and Mira variable stars. ",
        "introduction": "\\label{intro} A large majority of stars go through a period of high mass loss at the end of their evolution while climbing the asymptotic giant branch (AGB). This mass loss, of the order of $10^{-7}$ to $10^{-4} M_\\odot /{\\rm yr}$, is the main source for replenishing interstellar space with processed materials. Thus the mass loss mechanism is an important subject of study. In late type stars, this high mass loss produces a circumstellar envelope (CSE) in which several maser species can be found. These masers, especially SiO, H$_{2}$O and OH, are excellent tracers of the kinematics of the CSEs. The role of magnetic fields in the mass loss and formations of CSEs is still unclear. Polarization observations of circumstellar masers can reveal the strength and structure of the magnetic field throughout the CSE.  Observations of SiO maser polarization have shown highly ordered magnetic fields close to the central star, at radii of 5-10 AU where the SiO maser emission occurs (e.g. Kemball \\& Diamond, 1997). The measured circular polarization indicates magnetic field strengths of $\\approx 5-10$ G, when assuming a standard Zeeman interpretation. However, a non-Zeeman interpretation has been proposed by Wiebe \\& Watson (1998), which only requires fields of $\\approx$~30 mG.  At much lower densities and temperatures and generally much further from the star, OH maser observations measure fields of $\\approx 1$ mG (e.g. Szymczak \\& Cohen, 1997; Masheder et al., 1999) and there is little dispute of the Zeeman origin of the polarization. Until recently, no information on the magnetic fields at distances of a few hundred AU from the star was available.  This is the region where the H$_{2}$O maser emission occurs. Because water is a non-paramagnetic molecule, determination of the magnetic field is significantly more difficult. The fields expected in the H$_{2}$O maser region are stronger than the fields measured for the OH masers; since they probably occur in gas that is a factor of $10-1000$ more dense and closer to the central star than the OH masers. So we expect fields of a few tens to a few hundred mG.  For these fields the Zeeman splitting of H$_{2}$O is extremely small, only $\\approx~10^{-3}$ times the typical half-power width of the H$_{2}$O maser line ($\\Delta\\nu_{\\rm L} \\approx 30$ kHz). However, Vlemmings et al. 2001 (hereafter V01) have shown that in the presence of such magnetic fields the Zeeman splitting can be detected with high spectral resolution polarization observations. They presented the first results of circular polarization measurements of circumstellar H$_2$O masers on the masers around the supergiant star S~Per. A 'basic', LTE, Zeeman analysis was used to infer the magnetic field strength along the line of sight. This analysis did not include the interaction between the various hyperfine components of the 22~GHz H$_2$O transition. The observations confirmed the expected field strength, finding a field of $\\approx 250$ mG. The method used in V01 was similar to the method used by Fiebig \\& G\\\"usten (1989, hereafter FG) to analyze the circular polarization of strong interstellar H$_2$O maser features.  Here we present observations for a sample of 4 stars. We discuss the data reduction path, and in addition to the FG method, we fit our observation to the theoretical, non-LTE, models of Nedoluha \\& Watson (1992), hereafter NW. These include hyperfine interaction as well as saturation effect. The non-Zeeman interpretations presented in Nedoluha \\& Watson (1990) and Wiebe \\& Watson (1998) are also discussed. Finally we present measurements of the linear polarization. ",
        "conclusions": "The analysis of the circular polarization of H$_2$O masers depends strongly on the models used. The interpretation of our observations has shown that the non-LTE, full radiative transfer models presented by NW improve the accuracy of the fits. Some aspects of the observations however, remain unclear, even though the non-LTE approach can explain most of the narrowing of the circular polarization spectra. Multi-dimensional models seem promising to explain the narrowing observed. We have found magnetic field strengths of $\\approx 150-500$ mG for the supergiants and $\\approx 1.5$ G for the Mira variable U~Her. They occur in the most dense H$_2$O maser spots at the inner boundaries of the maser region and indicate strong magnetic fields at the surface of the star. The fields dominate the thermal pressure of the gas and are strong enough the drive the stellar outflows and shape the stellar winds.  {\\it Acknowledgments:} This project is supported by NWO grant 614-21-007."
    },
    "0208/astro-ph0208474_arXiv.txt": {
        "abstract": "In this first of two companion papers on the Orion Nebula Cluster (ONC), we present our analysis of a 63 Ksec {\\em Chandra} HRC-I observation that yielded 742 X-ray detections within the $30^{\\prime}\\times 30^{\\prime}$ field of view. To facilitate our interpretation of the X-ray image, here we collect a multi-wavelength catalog of nearly 2900 known objects in the region by combining 17 different catalogs from the recent literature.  We define two reference groups: an {\\em infrared sample}, containing all objects detected in the $K$ band, and an {\\em optical sample} comprising low extinction, well characterized ONC members. We show for both samples that field object contamination is generally low.  Our X-ray sources are primarily low mass ONC members. The detection rate for optical sample stars increases monotonically with stellar mass from zero at the brown dwarf limit to $\\sim 100\\%$ for the most massive stars but shows a pronounced dip between 2 and 10 solar masses.  We determine $L_X$ and $L_X/L_{bol}$ for all stars in our {\\em optical sample} and utilize this information in our companion paper to study correlations between X-ray activity and other stellar parameters. ",
        "introduction": "The Orion region is one of the most frequently observed areas of the sky. It comprises several molecular clouds and stellar associations, among which the Orion Nebula Cluster (ONC - also referred to as the Trapezium region), is of particular interest.  With an age of about 1 million years, more than two thousand young stellar objects with masses between 0.08 and 50$M_\\odot$, an estimated central density of $\\sim 2 \\times 10^4$ stars per cubic parsec, and at a distance of $\\sim 470$pc, it is the largest and densest concentration of young pre-main sequence (PMS) stars in our region of the Galaxy. Only part of the ONC members -- those lying on  the near side of the Orion Molecular Cloud from which the cluster is forming -- are optically visible; more than half are so deeply embedded in the cloud that they are observable only at infrared (IR) or X-ray wavelengths where the cloud becomes more transparent. For more complete descriptions of the ONC structure, dynamics, and stellar content, see \\citet{hil97,hil98b,hil00,car00,luh00,ode01}, and references therein. Among others, Hillenbrand and collaborators \\citep{hil97,hil98a,hil00} have extensively studied both the optically visible and the embedded ONC stellar content. Several studies \\citep{sta99,her00,her01,reb01} have measured rotational periods  via photometric modulation of stellar spots. Spectacular HST images \\citep[see e.g.,][]{bal00} have allowed the direct observation of many circumstellar disks and {\\em proplyds} (photoionization structures due to the evaporation of disks) that seem to surround a large fraction of the PMS stars in the vicinity of the central bright star $\\theta^1$ Orionis C.  The ONC was first detected in the X-rays by the {\\em Uhuru} satellite \\citep{gia72}. Following imaging observations performed with the {\\em Einstein} and ROSAT observatories \\citep{ku79,gag95a} indicated that: {\\em a}) ONC members are powerful X-ray emitters, with typical $L_X/L_{bol}$ close to the saturation value, $10^{-3}$, observed for fast rotators on the main sequence; {\\em b}) no relation seems to hold between X-ray activity and stellar rotation; {\\em c}) stars with circumstellar accretion disks may have lower levels of X-ray activity. These works were however hindered by the lack, at that time, of complete optical information on the ONC population and by the low sensitivity and spatial resolution of the X-ray instrumentation, especially important to resolve the dense central region and to unambiguously identify X-ray sources with optical counterparts. Both of these limitations have been recently lifted: {\\em Chandra} observations \\citep{sch01,gar00} have revealed about 1000 X-ray point sources, for the most part associated with low mass stars down to the substellar mass limit.  In this work we describe the analysis of a {\\em Chandra} High Resolution Camera \\citep[HRC, ][]{mur00} observation of the ONC and correlate our data with relevant data from the literature. In a companion work (\\citealt{fla02}; hereafter Paper~II), we investigate statistical correlations between X-ray activity and other stellar characteristics (such as rotational period, mass, age and disk accretion indicators) and search for insights into the physical mechanisms that drive activity in PMS stars. An investigation of the X-ray variability characteristics of ONC members based on these data is in preparation.  The structure of this paper is as follows: we first introduce our {\\em Chandra} observation and discuss its reduction and X-ray source detection (\\S \\ref{sect:obs}). We then describe the preparation of an extensive catalog of known X-ray/optical/IR/radio objects in the region of our HRC observation, collecting from the literature useful observable properties and defining two reference object samples (\\S \\ref{sect:cat}). In the same section we also report the results of our observation in the BN/KL region. In \\S \\ref{sect:optIRprop} we concentrate on the optical/IR properties of our X-ray sources, and in \\S \\ref{sect:Xprop} and a supporting Appendix, we determine the X-ray activity indicators $L_X$ and $L_X/L_{bol}$, along with their uncertainties. Our findings are summarized in \\S \\ref{sect:disc}, setting the stage for Paper~II. ",
        "conclusions": "}  Our {\\sc pwdetect} wavelet algorithm finds 742 X-ray sources in the $30^{\\prime}\\times 30^{\\prime}$ FOV of our 63 ksec {\\em Chandra} HRC-I observation of the ONC.  As a basis for our study of ONC X-ray activity (see Paper~II), we have combined 17 different catalogs from the recent literature to assemble a catalog of all available information for nearly 2900 objects, the large majority of which are ONC members.  We then calculated mass and age for a significant subset using the evolutionary tracks of \\citet{sie00} and defined two reference stellar samples: the {\\em optical sample}, comprising $\\sim 700$ well characterized members with low extinction ($A_V \\le 3.0$), and the {\\em IR sample}, largely including the {\\em optical sample} and comprising $\\sim 2500$ stars observed in the $K$ band.  We have shown that field object contamination is quite limited in both of these samples.  In order to characterize the population of X-ray detected objects, we have presented distributions of $K$ magnitudes and masses and have noted, in agreement with the results of \\citet{gag95a} and \\citet{gar00}, a clear dependence of the percentage of detections on stellar mass, from $\\sim 0$\\% at the brown dwarf limit, to $\\sim 100$\\% for $\\sim 2.0 M_\\odot$ stars.  For 2 -- 10 $M_\\odot$ stars, detection fraction drops sharply, followed by a marked  increase (to 100\\%) for the six highest mass ONC stars.  We concluded here by estimating activity indicators ($L_X$ and $L_X/L_{bol}$) for stars in our {\\em optical sample}; Paper~II studies correlations between this X-ray activity and other stellar parameters. Our X-ray luminosities were computed from {\\em basal} count rates, assuming single temperature spectra for all sources and proportionality between X-ray absorption and optical extinction, with these assumptions verified by our analysis of medium resolution archival ACIS X-ray spectra for a subset of our sources."
    },
    "0208/astro-ph0208312_arXiv.txt": {
        "abstract": "We consider the evolution of magnetic fields under the influence of Hall drift and Ohmic decay.  The governing equation is solved numerically, in a spherical shell with $r_i/r_o=0.75$.  Starting with simple free decay modes as initial conditions, we then consider the subsequent evolution.  The Hall effect induces so-called helicoidal oscillations, in which energy is redistributed among the different modes.  We find that the amplitude of these oscillations can be quite substantial, with some of the higher harmonics becoming comparable to the original field.  Nevertheless, this transfer of energy to the higher harmonics is not sufficient to significantly accelerate the decay of the original field, at least not at the $R_B=O(100)$ parameter values accessible to us, where this Hall parameter $R_B$ measures the ratio of the Ohmic timescale to the Hall timescale.  We do find clear evidence though of increasingly fine structures developing for increasingly large $R_B$, suggesting that perhaps this Hall-induced cascade to ever shorter lengthscales is eventually sufficiently vigorous to enhance the decay of the original field. Finally, the implications for the evolution of neutron star magnetic fields are discussed. ",
        "introduction": "Neutron stars have the strongest magnetic fields found in the universe, with fields perhaps as large as $10^{15}$ G for so-called magnetars (e.g.\\ Murakami 1999), around $10^{12}$ G for young ($\\sim10^7$ year) radio and X-ray pulsars, and a still appreciable $10^8 - 10^{10}$ G for much older ($\\sim10^{10}$ year) millisecond pulsars (e.g.\\ Chanmugam 1992; Bhattacharya 1995; Lyne 2000).  This correlation between field strength and age suggests that these very different strengths are due to the field decaying in time, rather than to any intrinsic differences between different neutron stars. One would therefore like to identify the processes causing the field to decay.  The additional observation that most weakly magnetic neutron stars have binary companions, whereas very few strongly magnetic ones do (e.g.\\ Bhattacharya 1995), suggests that accretion of mass from the companion is somehow causing the field to decay (by mechanisms that need not concern us here, but see for example Blondin \\& Freese 1986; Romani 1990; Urpin \\& Geppert 1995).  The observational evidence is unfortunately inconclusive, with Taam \\& van den Heuvel (1986) claiming a correlation between field strength and accreted mass, but Wijers (1997) disputing this.  One would therefore like to consider the possibility of other mechanisms besides accretion.  One such alternative is Hall drift, first proposed by Jones (1988), in which the magnetic field influences itself through a quadratic nonlinearity.  If it is relevant at all, Hall drift will therefore be most important for the very strongest fields -- which as we saw tend to occur in isolated neutron stars, where accretion is not acting at all. Hall drift is thus likely to be the dominant mechanism influencing the magnetic fields of these stars.  Of course, it could potentially be important in binaries as well, at least in the early stages while their fields are still relatively strong.  Again as a result of this quadratic nonlinearity, the timescale on which Hall drift might be expected to act is almost necessarily inversely proportional to $|{\\bf B}|$.  Jones suggests that it is given by \\begin{equation} t_{Hall}\\sim {10^8\\over B_{12}}\\;{\\rm years}, \\end{equation} where $B_{12}$ is the field strength in units of $10^{12}$ G.  See also Goldreich \\& Reisenegger (1992), who obtain a similar estimate. For these $O(10^{12})$ G radio pulsars, one therefore expects a timescale comparable to their age. And indeed, Lyne, Manchester \\& Taylor (1985) and Narayan \\& Ostriker (1990) have suggested that the fields of these young pulsars do decay on a $10^7$ year timescale (although this too is in dispute, see for example Hartman et al.\\ 1997; Regimbau \\& Pacheco 2001).  Apart from the strength of the observational evidence though, the mere fact that Hall drift could affect the fields of neutron stars on timescales so short compared to their evolutionary timescales makes it worthy of study.  There is one slight difficulty though in attributing this possible $10^7$ year decay rate to Hall drift, namely that the Hall effect conserves magnetic energy, and therefore by itself cannot cause any field decay at all. The suggestion therefore is that the Hall term, being nonlinear, will redistribute energy among the different modes, and in particular will initiate a cascade to ever shorter lengthscales, where ordinary Ohmic decay (which only acts on $O(10^{10})$ year timescales at the longer lengthscales) can destroy the field.  Since this mechanism was first proposed by Goldreich \\& Reisenegger, detailed calculations have been done by a number of authors, including Naito \\& Kojima (1994), Muslimov (1994), Muslimov, Van Horn \\& Wood (1995), Shalybkov \\& Urpin (1997) and Urpin \\& Shalybkov (1999).  Of these, only the last two were in the astrophysically relevant limit of large Hall parameter $R_B$ though, where $R_B$ measures the ratio of the Ohmic timescale to the Hall timescale, and is defined more precisely below.  However, it is not certain whether their results were fully resolved, as they had only 20 radial by 40 latitudinal finite difference points.  In contrast, we have 25 radial by 100 latitudinal spectral expansion functions, and obtain fully resolved solutions for $R_B$ up to $O(100)$ (comparable to what Shalybkov \\& Urpin achieved, and indeed in broad agreement with their results, suggesting that perhaps their resolution was good enough after all to resolve the most important features anyway). Even at these large values, however, we find that while the Hall effect does indeed induce a significant redistribution of energy among the different modes, it does not appear to be enough to cause the lowest modes to decay substantially faster than they would have otherwise.  Before applying this conclusion to real neutron stars though, it is important to qualify it by noting that our calculations (as well as the others cited above) are restricted to $\\bf B$ being axisymmetric, and the various material properties such as the density being independent of depth.  Neither of these assumptions holds in real neutron stars, and relaxing either could significantly alter the results.  For example, Vainshtein, Chitre \\& Olinto (2000) show that including variations in density introduces new effects even for field configurations where no ordinary Hall drift would be present at all. Similarly, Rheinhardt \\& Geppert (2002) also consider field configurations where no ordinary Hall cascade is present, but claim that instabilities, including non-axisymmetric ones, can nevertheless arise.  We will discuss both of these papers more fully below, as well as how these two restrictions might be relaxed in future work. ",
        "conclusions": "The results presented here suggest that Hall drift could indeed have a significant influence on the evolution of a neutron star's magnetic field. Particularly if the internal toroidal field is as strong or stronger than the poloidal field, Hall drift can excite some of the higher harmonics to amplitudes comparable to the original mode. However, as substantial as some of these higher harmonics are, this still does not appear to be enough to cause the original mode to decay significantly faster than it otherwise would have.  This conclusion must be qualified though by our inability to increase $R_B$ indefinitely.  Indeed, the very feature that caused the code to fail beyond certain limits, namely the fact that the spectra got flatter and flatter, also indicates that this transfer of energy to the higher harmonics gets more and more efficient as $R_B$ is increased. It is conceivable, therefore, that the solutions for, say, $R_B=1000$ would show a very rapid decay of the original mode.  Also, the cascade may well be very different in 3D than in 2D, just like ordinary fluid turbulence is very different.  Extending our model here from 2D to 3D is possible in principle, but will obviously require considerable computational resources.  Finally, even if it should turn out that Hall drift alone, in either 2D or 3D, simply does not generate a sufficiently strong cascade at any value of $R_B$, the combination of Hall drift and stratification may still do so.  We've already noted in the introduction that the electron number density $n$ in equation (2) is in fact not constant, but rather varies by several orders of magnitude over the depth of the crust.  Vainshtein et al.\\ (2000) show that if one includes this effect, one can obtain a very rapid decay of a toroidal field at least.  In their highly idealized analytical model it was not possible to include poloidal fields though (we recall from section 2.4 that one can indeed consistently consider only toroidal fields).  In contrast, our numerical model here already includes poloidal fields, and including radial variations in $n$ is possible too.  Calculations are therefore currently under way to see if this Vainshtein et al.\\ result applies to poloidal fields as well."
    },
    "0208/gr-qc0208025_arXiv.txt": {
        "abstract": " ",
        "introduction": "New observational evidence for black holes provides new motivations for the investigation of the general relativistic dynamics of particles and electromagnetic fields in the vicinity of the black holes. We shall start with a brief description of the situation. The results of astronomical observations over the last decade continue to point insistently to the existence of stellar-mass and supermassive black holes in some X-ray binary systems and in galactic centres (see  Horowitz $\\&$  Teukolsky 1999; Rees 1998 for reviews). The typical examples of the stellar-mass black holes in X-ray binaries are Cyg X-1 discovered back in 1971, the X-ray source LMC X-3 in the Large Magellanic Cloud, as well as the source in A0620-00  discovered in 1975 and a number of recently discovered sources, such as V404 Cyg, GS 2000+25, GRO J0422+32 (for full list see Charles 1999). New observational data, such as the detection of broad iron fluorescence lines and maser emission lines of water in the spectra provide the strongest suggestion for the presence of supermassive black holes in the centres of the active galaxies MCG 6-30-15, NGC 4258 and NGC 1068 Menou, Quataert  $\\&$ Narayan 1999; Rees 1998; Miyoshi et.al 1995; Watson $\\&$ Wallin 1994).   The supermassive black holes are also strongly believed to be in the centres of some low-level active, or non-active, galaxies. For instance, recent progress in the studies of the distributions and velocities of stars near the centres of the giant elliptical galaxy M87, Andromeda M31 and our own Galaxy have revealed the strong evidence for the existence of the supermassive black holes in these centres (Merritt $\\&$ Oh 1997; Richstone et al. 1998; Rees 1982, 1998).   On the other hand, a convincing explanation for a huge amount of energy output from the active cores of the galaxies associated with the supermassive black holes requires the searches for mechanisms responsible for the high-level energy release. One of these mechanisms is the extraction of the rotational energy from a rotating black hole surrounded by stationary magnetic fields. The magnetic fields threading the event horizon tap the rotational energy of the black hole due to the interaction between charged particles and the induced electric field (Blandford $\\&$ Znajek 1977; Thorne, Price $\\&$ Macdonald 1986). The interest in this model still continues to point out new gravito-electromagnetic phenomena in the strong and weak field domains around a rotating black hole (Bi\\u c\\'ak $\\&$ Ledvinka 2000; Chamblin, Emparan $\\&$ Gibbons 1998; Mashhoon 2001; see also Punsly 2001). Another mechanism responsible for high-level energy release is gas accretion by a black hole (see Shapiro $\\&$ Teukolsky 1983), where energy is released mostly at the expense of the binding energy of the particles and the strong gravitational field of the black hole. It is well known that the binding energy in the innermost stable orbits of the particles determines the potential efficiency of an accretion disc. It is about 5.7 \\% of the rest energy in the Schwarzschild field, but in the case of a maximally rotating black hole it approaches 42 \\% of the rest energy. The observational data from the core of some galaxies, such as the elliptical galaxy M87 reveal that the inner part of a gas disc around a putative supermassive black hole emits non-thermal radiation and radio waves. The reason for this is believed to be synchrotron emission from ultrarelativistic electrons moving in a strong magnetic field in the inner part of the accretion flow onto the supermassive black hole (Fabian \\& Rees 1995; Narayan \\& Yi 1995; Rees 1998). This gives us a new impetus to return back once again to the investigation of the motion of charged particles in the model of a rotating black hole in a uniform magnetic field, though much insight into this problem was given in 1980s (Prasanna $\\&$ Vishveshwara 1978; Prasanna 1980; Wagh, Dhurandeur \\& Dadhich 1985; Aliev $\\&$ Gal'tsov 1989; see also Frolov $\\&$ Novikov 1998 and references therein). In particular, the works of Prasanna \\& Vishveshwara (1978) and Prasanna (1980) have given a comprehensive numerical analysis of the charged particle motion in a magnetic field superposed on the Kerr metric by studying the structure of the effective potential for radial motion and integrating the complete set of equations of the motion for appropriate initial conditions.   In this paper we shall study a particular class of orbits, namely the marginally stable circular orbits of charged particles in the equatorial plane of a Kerr black hole immersed in a uniform magnetic field. The numerical analysis performed by Prasanna \\& Vishveshwara (1978) has revealed that the presence of a uniform magnetic field on the Kerr background increases the range of stable circular orbits. This result is confirmed in our model of the marginally stable circular motion, however we also obtain significant new results: First of all, we derive the basic equation determining the region of the marginal stability of the circular orbits, that comprises only two parameters, namely the rotation parameter of black hole and the strength of the magnetic field. We also find the closed analytical expressions for the associated angular momentum and energy of a charged particle moving in a marginally stable circular orbit. Further, for a sufficiently large values of the magnetic field, as well as for a maximum value of the black hole rotation parameter we find the limiting analytical solutions for the radii of stability of both direct and retrograde innermost circular orbits. This allows us not only to emphasize that the presence of a magnetic field enlarges the region of stability towards the event horizon, but also to find the limiting values for this enlargement both for direct and retrograde motions along with their associated energies and angular momenta. Our analytical and numerical calculations show that the combined effects of a sufficiently strong magnetic field and the rotation of a black hole give rise to the possibility of relativistic motion of the charged particles in the innermost stable direct and retrograde orbits. The existence of relativistic motion in the innermost stable direct orbits is especially important new feature worked out in our analysis. These orbits lie very close to the event horizon of a rotating black hole and they may provide a mechanism for synchrotron emission from relativistic charged particles with the signatures of the strong-gravity domain.   The paper is organized as follows. We shall first review the solution of the Maxwell equations describing a uniform magnetic field in the Kerr metric with a small electric charge (see Section 2). In Section 3 we consider the separation of variables in the Hamilton-Jacobi equation and derive the effective potential for the radial motion of charged particles around a Kerr black hole in a uniform magnetic field. These results are used in Section 4 to obtain the basic equations governing the region of the marginal stability of the circular orbits and their associated energies and angular momenta. To make the analysis more transparent in the general case, we shall first present the results of analytical and numerical calculations for the marginal stability of the circular orbits in the pure Kerr  metric, as well as in a uniform magnetic field in the Schwarzchild metric (see Sections 4.1 and 4.2). Section 4.3 is devoted to a comprehensive analysis of the effect of a uniform magnetic field on the radii and the assigned energy and angular momentum of the marginally stable circular orbits for various values of the magnetic field and the rotation parameter of the black hole.       \\begin{center} ",
        "conclusions": "\\vspace{4mm} \\noindent The main purpose of this paper was to study the combined effects of the rotation of a Kerr black hole and an external magnetic field on the marginally stable circular motion of charged particles. We have presented general equations governing the energy, the angular momentum and the region of the marginal stability of circular orbits around a rotating black hole in a uniform magnetic field. The analytical results obtained for large enough values of the magnetic field strength and for a maximum value of the black hole angular momentum, as well as the numerical analysis performed in general case have shown that in all cases of the circular orbits the magnetic field essentially enlarges the region of their marginal stability towards the event horizon. As for the effect of the rotation of a black hole it has been found to be different for direct and retrograde motions: For retrograde motion the rotation opposes to the magnetic field in its stabilizing effect, though the latter always remains be dominant and for extreme values of the magnetic field and rotation parameter there exists a limiting value for the enlargement of the region of stability towards the event horizon. In the case of direct motion the presence of a rotation produces an additional shift of the stable orbits to the event horizon. However, for large enough values of the magnetic field, when the innermost stable motion occurs at a radius that lies very close to the event horizon, the magnetic field lines are expelled from the black hole as the hole's rotation parameter increases. Accordingly, the direct motion occurs in two different orbits; one of them is also expelled from the black hole, while the other one being affected less and less by the magnetic field approaches the event horizon as the rotation becomes maximum.   We have shown that the presence of a strong magnetic field around a rotating black holes provides the possibility of retrograde motion in the innermost stable relativistic orbits and the rotation of the black hole further enhances this effect. It is very important that, unlike the Schwarzschild case, the combined effects of a sufficiently large magnetic field and the rotation of a black hole also result in relativistic motions in the innermost stable direct orbits that lie very close to the event horizon. Thus, nearby the event horizon of a rotating black hole there may exist a source of synchrotron emission from relativistic charged particles moving in stable circular orbits. This, of course, may play an important role both in the searches for black holes and in making feasible the probes of the metric in the strong-gravity domain.\\\\[2mm] \\noindent {\\bf"
    },
    "0208/astro-ph0208581_arXiv.txt": {
        "abstract": "We present V, I photometry of the Sagittarius Dwarf Spheroidal galaxy (Sgr) for a region of $\\sim 1^{\\circ} \\times 1^{\\circ}$, centered on the globular cluster M~54. This catalog is the largest database of stars ($\\sim$500,000) ever obtained for this galaxy. The wide area covered allows us to measure for the first time the position of the RGB-bump, a feature that has been identified in most Galactic globular clusters and only recently in a few galaxies of the Local Group. The presence of a single-peaked bump in the RGB differential Luminosity Function confirms that there is a dominant population in Sgr (Pop~A).  The photometric properties of the Pop~A RGB and the position of the RGB bump have been used to constrain the range of possible ages and metallicities of this population. The most likely solution lies in the range $-0.6< [M/H]\\le -0.4$ and $4~Gyr\\le age\\le 8~Gyr$. ",
        "introduction": "The Sagittarius dwarf spheroidal (Sgr dSph) galaxy \\citep{igi,iba97} is an {\\em in vivo} example of an accretion/disruption event of a Galactic satellite \\citep{new02}: a process that may have been one of the main mechanisms for the assembly of the Galactic halo \\citep{sz78,cote}. Hence, the study of the stellar content, star formation history and chemical evolution \\cite[see][for references]{ls00,b99b,sm02} of this disrupting system is a fundamental step in order to reconstruct the evolutionary history of the Milky Way.   As part of a long-term project devoted to understand the origin and the evolution of the {\\em building blocks} of the Galactic Halo, we present here the first result of a new wide-field V,I photometry of the Sgr in a region ($\\sim$1$^{\\circ}\\times $1$^{\\circ}$) around the globular cluster M~54. Almost 500,000 stars have been measured in the area, allowing us to unambiguously identify, for the first time, the RGB-bump in the Sgr. This feature was identified in a relevant number of Galactic globular clusters \\citep{fp90,f99,z99} and only in a few galaxies in the Local Group: Sculptor \\citep{maj99}, Sextans \\citep{b01} and Ursa Minor \\citep{b02}.  The RGB-bump magnitude is mainly driven by the metal content and, to a lesser extent, by the age of the population. In this {\\em Letter} we use this feature to constrain the metallicty and age of the main stellar population in the Sgr galaxy (Pop~A). Such constraints are particularly useful in the case of the Sgr dSph, where the strong foreground contamination makes a clear-cut interpretation of the Color Magnitude Diagram (CMD) more difficult \\citep[see][hereafter B99 and LS00, respectively]{b99b,ls00} and where some inconsistency between the photometric and spectroscopic metallicity estimates has emerged \\citep{co01,bo00,cse01,sm02}. ",
        "conclusions": "The growing wealth of independent observational material about the Sgr dSph is beginning to provide a self-consistent view of the properties of this galaxy. For instance, the presence of a metal poor population ($-2.0\\la [Fe/H] \\la -1.4$), first suggested by B99, has been confirmed by further photometric studies \\cite[LS00,][]{co01}, by the analysis of the pulsational properties of the RR Lyrae variables \\citep{cse01} and also by the spectroscopic survey by \\citet{sm02}. Moreover, the existence of a significant metallicity (and/or age) spatial gradient, with the stars in the outer regions being more metal deficient than those near the center of the galaxy, has been put into evidence by different authors \\cite[B99, LS00][]{ala01}.  Conversely, some reason of concern was provided by the estimate of the mean metallicity of the stellar population that dominates the inner region of the galaxy. While photometric estimates were tipically in the range $-1.0\\la [Fe/H]\\la -0.5$, the first high-resolution spectra of two member stars by \\citet{bo00} suggested a much higher metal content ($[Fe/H]\\simeq -0.2$). The larger sample provided by the spectroscopic analysis of \\citet{sm02} suggests that, while stars having $[Fe/H]\\simeq -0.2$ are indeed present in Sgr, they do not represent the dominant population. The majority of stars in the \\citet{sm02} sample has $[\\alpha/Fe]\\simeq 0.0$ (and thus [Fe/H]=[M/H]) and $-0.6\\le [Fe/H] \\le -0.3$. Furthermore, previous photometric estimates based on the comparison of the RGB with template globular clusters, neglected the effects of age and the difference in the $\\alpha$-elements enhancement, as correctly pointed out by \\citet{co01}.  In this {\\em Letter} we have revisited the problem of the metallicity and age of the Sgr main population, taking into account all of the above considerations and adding to the usual photometric metallicity indices (i.e., the position and the shape of the RGB) an additional constraint provided by the position of the RGB-bump. The main results can be summarized as follows:  \\begin{enumerate}  \\item The mere detection of a single-peaked bump in the differential LF of the RGB confirms the existence of a dominant population (Pop~A) in the Sgr dSph stellar mix.  \\item The Pop~A RGB is well fitted by the RGB ridge-line of the globular cluster 47~Tuc, that has $[M/H]\\simeq -0.6$. However, a significant difference in the RGB-bump luminosity between Sgr and 47~Tuc has been measured. This fact suggest that Pop A is several Gyr younger than 47~Tuc, in good agreement with previous results based on the comparison of the Main Sequence Turn Off's (B99, LS00). Hence, $[M/H]\\simeq -0.6$ has to be considered as a {\\em lower limit} to the mean metallicity of Pop~A.  \\item A full self-consistency among the spectroscopic and photometric constraints (including the RGB-bump) is achieved if a mean metallicity of $-0.6< [M/H]\\le -0.4$ and a mean age of $7\\ge age\\ge 4$ Gyr are assumed for Pop~A, in good agreement with the results by LS00. Ages younger than $\\sim 4$ Gyr are {\\em excluded} by the observed morphology of the TO (see Figure~11 by B99 and Figure~16 by LS00). The detailed discussion of this region of the CMD will be the subject of a forthcoming paper in preparation.  \\end{enumerate}"
    },
    "0208/astro-ph0208062_arXiv.txt": {
        "abstract": "We report spectroscopic observations in the wavelength region 0.8$\\mu$m $-$ 2.4$\\mu$m aimed at detecting near-infrared coronal lines in a sample of 5 narrow-line and 1 broad-line Seyfert 1 galaxies. Our measurements show that [\\ion{Si}{6}] 1.963$\\mu$m, [\\ion{S}{9}] 1.252$\\mu$m and [\\ion{S}{8}] 0.991$\\mu$m are present in most of the objects and are useful tracers of nuclear activity. Line ratios between coronal and low-ionization forbidden lines are larger in narrow-line Seyfert 1 galaxies. A positive correlation between FHWM and ionization potential of the forbidden lines is observed. Some coronal lines have widths similar to that of lines emitted in the broad line region (BLR), indicating that part of their flux originates in gas close to the outer portions of the BLR. Most coronal lines are blueshifted relative to the systemic velocity of the galaxy and this shift increases with the increase in line width. Assymetries towards the blue are observed in the profiles of high-ionization Fe lines, suggesting that the emitting gas is related to winds or outflows, most probably originating in material that is being evaporated from the torus. This scenario is supported by models that combine the effects of shock ionization and photoionization by a central continuum source in the gas clouds. The agreement between the coronal line emission predicted by the models and the observations is satisfactory; the models reproduced the whole range of coronal line intensities observed. We also report the detection of [\\ion{Fe}{13}] 1.074,1.079$\\mu$m in three of our objects and the first detection of [\\ion{P}{2}] 1.188$\\mu$m and [\\ion{Ni}{2}] 1.191$\\mu$m in a Seyfert 1 galaxy, ARK\\,564. Using the ratio [\\ion{P}{2}]/[\\ion{Fe}{2}] we deduced that most Fe present in the outer NLR of ARK\\,564 is locked up in grains, and the influence of shocks is negligible. ",
        "introduction": "Coronal lines are collisionally excited forbidden transitions within low-lying levels of highly ionized species ($\\chi \\geq$ 100 eV). They can be formed either in gas photoionized by a hard UV continuum \\citep{gr78, kf89, fkf97} or in a very hot, collisionally ionized plasma \\citep{vac89}. It is also possible that the excitation mechanism is a mixture of collisional ionization at the shock front and photoionization in the emitting clouds \\citep{cv92, vc94, cpv98}. In active galactic nuclei (AGN), the presence of coronal lines, mainly in the optical region (i.e. [\\ion{Fe}{7}] and [\\ion{Fe}{10}]), have long been known to be common features in the spectra of these sources, although the physical conditions of the gas from which they originate and the location of the emitting region are still poorly determined.  Observationally, coronal lines are, on average, broader and blueshifted, relative to the centroid position, of lines of lower ionization stages such as [\\ion{O}{3}]\\lb5007 \\citep{dro84, dro86, vei91, eaw97}. This has led to the speculation that they are formed in a separate region, known as the coronal line region (CLR), located at an intermediate distance between the classical narrow line region (NLR) and the broad-line region (BLR). Variability studies on [\\ion{Fe}{7}] \\lb6087 and [\\ion{Fe}{10}] \\lb6374 carried out by \\citet{veill88} also suggest that the most likely place for the CLR is between the BLR and NLR. \\citet{eaw97} found that the coronal line emission occurs predominantly in objects with a soft X-ray excess and suggested a relationship between these lines and the X-ray absorption edges (also known as warm absorbers) seen in 50\\% of AGN. In fact, \\citet{por99} demonstrated through photoionization modeling that the optical coronal lines could be formed in the warm absorber and that they may strongly constrain the physical parameters of that medium. However, the number of coronal lines effectively detected in the optical region is small ([\\ion{Fe}{7}] \\lb5721, \\lb6087; [\\ion{Fe}{10}] \\lb6374; [\\ion{Fe}{11}] \\lb7892 and [\\ion{Fe}{14}] \\lb5303). It is then necessary to expand the range of important diagnostics of the physical conditions that prevail in the coronal gas to other wavelength intervals. In this respect, the near-infrared (NIR) is promising because it offers a wealth of highly ionized species different from Fe.  Uptil now, coronal line emission between 1.5$\\mu$m and 4$\\mu$m has been detected in only a small number of AGN. \\citet{om90} reported, for the first time, the observation of [\\ion{Si}{6}] 1.962$\\mu$m and [\\ion{Si}{7}] 2.483$\\mu$m in the archetypical Seyfert 2 galaxy NGC\\,1068. \\citet{oli94} detected, in addition to those two lines, [\\ion{S}{9}] 1.252$\\mu$m, [\\ion{Ca}{8}] 2.321$\\mu$m and [\\ion{Si}{9}] 3.9346$\\mu$m in Circinus, another nearby Seyfert 2 nucleus. \\citet{mar94} detected [\\ion{Si}{6}] 1.962$\\mu$m in eight AGN (including the former two objects) in a sample composed of Seyferts, Starburst and Ultraluminous Infrared Galaxies (ULIRGs). They concluded that [\\ion{Si}{6}] emission was a common characteristics of Seyfert nuclei, consistent with modeling the line formation by photoionization of the active nucleus. That result had reinforced the use of [\\ion{Si}{6}] 1.962$\\mu$m as a diagnostic of AGN activity and had allowed the detection of hidden AGN, mostly in ULIRGs samples \\citep{vsk99, msm00}. \\citet{tho95} studied NGC\\,4151 in detail and reported the detection of several coronal lines in the 0.87$\\mu$m $-$ 2.5$\\mu$m interval, most notably [\\ion{S}{8}] 0.911$\\mu$m, [\\ion{Fe}{13}] 1.074,1.079$\\mu$m, [\\ion{S}{9}] 1.252$\\mu$m, [\\ion{Si}{6}] 1.963$\\mu$m and [\\ion{Si}{7}] 2.481$\\mu$m. \\citet{grr95} studied a sample of six Seyfert 1 galaxies (one of them NGC\\,4151) in the 2$\\mu$m region and detected [\\ion{Si}{6}] 1.963$\\mu$m in all the galaxies and [\\ion{Ca}{8}] 2.321$\\mu$m in two of them.  In this paper we present the results of a search for coronal emission lines in the wavelength range 0.8$\\mu$m $-$ 2.5$\\mu$m by means of long-slit spectroscopy at moderate resolution ($R \\sim$ 750). The sample chosen for this study is composed of six Seyfert 1 galaxies, five of them classified as narrow-line Seyfert 1 (NLS1). These data are the first measurements of coronal lines in the NIR made on this sub-class of objects. Here, we concentrate on the analysis of the line ratios derived from our measurements and the physical conditions for the CLR that they imply. We also discuss the kinematics of the CLR based on the analysis of the line profiles. In addition, we report the detection of other forbidden lines and molecular lines emitted by the NLR in the NIR, which are also useful for constraining the various excitation models proposed so far.  Our observations are described in $\\S$~\\ref{obs}, the main results in $\\S$~\\ref{res}, the study of the kinematics of the CLR in $\\S$~\\ref{kin} and the physical conditions for the CLR derived from the observations in $\\S$~\\ref{physical}. Some comments about the low ionization and coronal lines observed appear in$\\S$~\\ref{low}. We present the main conclusions of this work in $\\S$~\\ref{final}. ",
        "conclusions": "\\label{final}  NIR coronal lines in the 0.8$-$2.5 $\\mu$m interval are studied, for the first time, in a sample of Seyfert 1 galaxies. We have found that [\\ion{S}{8}], [\\ion{S}{9}], [\\ion{Fe}{13}], [\\ion{Si}{6}] and [\\ion{Si}{10}] are present in most of the objects. These lines are significantly broader than low-ionization lines such as [\\ion{S}{3}], [\\ion{Ca}{1}] and [\\ion{Fe}{2}]. In addition, blueshifts of up to 550 km s$^{-1}$, relative to the systemic velocity of the galaxies, are measured for [\\ion{Fe}{13}]. The amount of blueshift is strongly correlated with FWHM and varies among the species with similar IP. These results hold when NIR data are combined with existing optical spectroscopy. Moreover, the FWHM of the broadest coronal lines is larger than that of \\ion{O}{1} \\lb1.128$\\mu$m, a pure BLR feature, but lower than broad Pa$\\beta$.  The above findings give strong observational support to the picture in which coronal lines are emitted in the intermediate region between the NLR and BLR. The blueshifts and assymetries of the highest ionization lines suggest that they are formed from outflow gas, most probably associated with material evaporated from the torus. The combined effect of shocks between that material and the ambient gas and the intense radiation field from the central source would produce the observed coronal line emission. Models that take into account these two mechanisms successfully reproduce the observed values of line ratios between coronal optical and NIR lines.  Outward from the CLR, where most low- to intermediate-ionization lines are being emitted, thermal processes associated with starburst activity may give rise to the observed molecular H$_{2}$ emission, as is indicated by the value of the H$_{2}$ ratio 2.247$\\mu$m/2.121$\\mu$m. Our data also report the first detection of [\\ion{P}{2}] 1.188$\\mu$m and [\\ion{Ni}{2}] 1.191$\\mu$m in a Seyfert 1 object. The ratio between [\\ion{Fe}{2}] 1.257$\\mu$m and [\\ion{P}{2}] sets important constraints on the dominant excitation mechanism for the low-ionization gas. The low value found for [\\ion{Fe}{2}]/[\\ion{P}{2}] (0.82) indicates that most iron is locked up in grains, being negligible the excitation via shocks."
    },
    "0208/astro-ph0208548_arXiv.txt": {
        "abstract": "The results of deep long-slit optical spectroscopy for a sample of eight 6C radio galaxies at redshift $z\\sim 1$ are presented. Emission line ratios are derived for many emission lines with rest--frame wavelengths of $1500 - 4500$\\AA\\, and the kinematic properties of the emission line gas are derived from an analysis of the two dimensional structure of the [O\\textsc{ii}]3727\\AA\\, emission line at $\\approx 5\\rm{\\AA}$ spectral resolution.  In general, the 6C spectra display many characteristics similar to those of more powerful 3CR sources at the same redshifts. The emission line region gas kinematics are more extreme for the smaller radio sources in the sample, which often display distorted velocity profiles. The ionization state of the emission line region also varies with radio size: the spectra of large radio sources ($> 120$kpc) are consistent with photoionization by an obscured AGN, whilst smaller ($< 120$kpc) sources typically exist in a lower ionization state and have spectra which are better explained by additional ionization due to shocks associated with the expanding radio source. The kinematic and ionization properties of the 6C radio galaxies are clearly linked. As for the 3CR sources, smaller radio sources also typically possess more extensive emission line regions, with enhanced emission line luminosities. A high velocity emission line gas component is observed in 6C1019+39, similar to that seen in 3C265.  It is clear that the best interpretation of the spectra of radio sources requires a combination of ionization mechanisms. A simple model is developed, combining AGN photoionization with photoionization from the luminous shock associated with the expanding radio source. The relative contributions of ionizing photons from shocks and the central AGN to an emission line gas cloud varies with radio source size and the position of the cloud. This model provides a good explanation for both the ionization properties of the emission line regions and the radio size evolution of the emission line region extents and luminosities. ",
        "introduction": "\\footnotetext{E-mail: kji@mrao.cam.ac.uk}  Extended emission line regions are often observed around powerful radio galaxies. High redshift ($z \\gta 0.5$) radio galaxies typically have more extensive emission line regions than lower power radio sources at low redshift. These emission line regions can be up to 100\\,kpc in size, and are often elongated and aligned along the radio jet axis (McCarthy 1993;  McCarthy {\\it et al} 1995).  The kinematics of the emission line gas of higher redshift radio galaxies are generally more extreme than those seen in low redshift radio galaxies; more distant sources have larger emission line FWHM, and larger velocity shears. The luminosities of the emission lines are large, with rest frame equivalent line widths often exceeding 100\\AA\\, (e.g. Baum \\& McCarthy 2000).  Over the past few years it has become increasingly clear that in addition to photoionization by an obscured AGN, in some cases the emission line gas may also be ionized by radiative shocks associated with the radio source.  Evidence for this includes the fact that whilst the spectra of low redshift sources are well explained by AGN photoionization, the emission line regions of more distant galaxies do not always display the characteristic line ratios of photoionized regions, and require alternative heating mechanisms.  Furthermore, observations of individual sources reveal features suggesting significant interactions of the radio jet with the interstellar medium (ISM) of the host galaxy.  Shocks associated with the passage of the radio source clearly influence the kinematics and morphology of the emission line gas, and can also affect the ionization state within the gas. Another potential source of ionizing photons is UV emission from a young stellar population.   Dey {\\it et al} (1997) find that the extended UV continuum emission from 4C 41.17 (z = 3.8), is unpolarised, suggesting that scattered light from the AGN is not a dominant source of UV photons.  This galaxy appears to have undergone a major epoch of star formation at $z \\sim 4$.    At lower redshifts, the stellar populations of radio galaxies are generally old.  Most 3CR radio galaxies at $z \\sim 1$ exhibit strong continuum polarisation ($\\sim 10\\%$, e.g. di Serego Alighieri, Cimatti \\& Fosbury 1994); scattered light from an obscured AGN provides a large proportion of the extended UV continuum.  Although small amounts of radio source jet induced star formation are possible, this is clearly not the dominant source of aligned continuum emission, nor can it be a major source of ionizing photons for objects at $z \\sim 1$.  Spectroscopic observations of a sample of 14 3CR radio galaxies at $z \\sim 1$ by Best {\\it et al.} (2000a, 2000b) showed that the ionization state of such sources is strongly correlated with radio size, smaller radio sources generally existing in a lower ionization state. In addition, radio sources with linear sizes $\\lta 150$kpc typically have greater emission line fluxes and broader line widths than their larger counterparts. The emission line ratios of smaller sources are in good agreement with the predictions of shock ionization models, and their observed kinematics are more disturbed. The irregularities in the velocity structures of the small sources are likely to be due to shock acceleration of the emission line gas clouds. Compression of the emission line gas clouds by the shock front and the  ionizing photons associated with it combine to lower the ionization state of the gas in smaller sources.  The properties of the larger, older, radio sources, for which the shocks associated with the expansion of the radio cocoon are long in the past, are well explained by photoionization models.  A spectroscopic study of four intermediate redshift 3CR galaxies by Sol\\'{o}rzano-I\\~{n}arrea {\\it et al.} (2001) found evidence for shock acceleration of the emission line gas in the extended emission line regions (EELRs) of all four galaxies. The disturbed kinematics of these sources are reflected in the emission line spectra by line splitting and/or underlying broad components.  Two of these sources have radio sizes comparable to the physical extent of the emission line region, and both show evidence for an underlying broad component in their emission lines.  Of the two sources with radio sizes significantly larger than their emission line regions, only one shows evidence for an underlying broad component, although both sources exhibit line splitting.  Both of these features can be related to the effects of shocks on the emission line gas. %  It is clear that the age of a radio source and the dominant ionization mechanism play a determining role in the observed properties of the emission line regions of powerful high redshift radio galaxies.  It is important to understand exactly how radio luminosity correlates with the EELR properties; the precise effects of radio jet power on shock acceleration and the ionization mechanism are still uncertain.  To this end, we have carried out deep spectroscopic observations of a complete sample of 8 6C radio galaxies at $z \\sim 1$.  In this paper we investigate the effects of radio power on the ionization and kinematic properties of emission line regions by contrasting the properties of our sample of 6C galaxies with those of the more powerful 3CR galaxies at similar redshifts.   A comparison with a sample of low redshift 3CR sources matched in radio power to the 6C subsample will also enable us to  break the degeneracy between redshift and radio power present in any flux limited sample; this analysis is deferred to a second paper. Overall, this research is aimed at adding to our understanding of: (i) the physics of the ISM in relatively extreme conditions, (ii) the impact of AGN activity on its nearby environment and (iii) the origin of the cool gas which forms the extended emission line regions.  \\begin{table*} \\caption{Details of the ISIS observations. Where more than one slit position angle is listed on a single line, the radio galaxy emission line data were combined in 2--dimensions, due to the similarity of the 2--d emission line spectra.} \\begin{center} \\begin{tabular} {lccccccc} \\hline {Source}&{Redshift}&{Observation}&{Slit width}&{Isis Arm}&{Exposure}&{Wavelength}&{Slit PA}\\\\ {}&{}&{Date}&{[arcsec]}&{}&{time [s]}&{Range [\\AA]}&{[deg.]}\\vspace*{0.05cm}\\\\\\hline 6C0943+39 & 1.035 & 21/03/99, 22/03/99           & 1.5     & blue & 15360$^1$  & 3100--5400 & 105$^2$\\\\ &       & 21/03/99, 22/03/99, 01/03/00 & 1.5-2.5 & red  & 19950  & 6900--9000 & 105, 110$^{2}$\\vspace*{0.15cm}\\\\ 6C1011+36 & 1.042 & 21/03/99           \t\t & 1.5     & blue & 3000   & 3100--6000$^3$ & 47$^2$\\\\ &       & 21/03/99           \t\t & 1.5     & red  & 3000   & 6900--8250 & 47$^2$\\\\ &   \t  & 21/03/99, 01/03/00 \t\t & 1.5-2.5 & blue & 9320   & 3100--6000 & 345$^4$\\\\ &   \t  & 21/03/99$^5$   \t         & 1.5-2.5 & red  & 9000   & 6900--9000 & 345$^4$\\vspace*{0.15cm}\\\\ 6C1017+37 & 1.053 & 22/03/99, 29/02/00 \t\t & 1.5-2.5 & blue & 15440  & 3100--5400 & 48$^2$\\\\ &    \t  & 22/03/99, 29/02/00 \t\t & 1.5-2.5 & red  & 13180  & 6900--9000 & 48$^2$\\vspace*{0.15cm}\\\\ 6C1019+39 & 0.922 & 28/02/00           \t\t & 1.5-2.0 & blue & 9360$^1$   & 3100--5400 & 61$^2$\\\\ &       & 28/02/00, 01/03/00 \t\t & 1.5-2.5 & red  & 14800  & 6400--7900$^6$ & 61$^2$\\vspace*{0.15cm}\\\\ 6C1129+37 & 1.060 & 21/03/99, 22/03/99 \t\t & 1.5     & blue & 12590  & 3100--5400 & 105$^2$\\\\ &       & 21/03/99, 22/03/99 \t\t & 1.5     & red  & 12000  & 6950--8450$^7$ & 105$^2$\\vspace*{0.15cm}\\\\ 6C1217+36 & 1.088 & 29/02/00           \t\t & 2.0     & blue & 10540  & 3500--5400 & 61$^2$\\\\ &       & 29/02/00           \t\t & 2.0     & red  & 7300$^8$   & 7100--8600 & 61$^2$\\vspace*{0.15cm}\\\\ 6C1256+36 & 1.128 & 21/03/99, 22/03/99 \t\t & 1.5-2.0 & blue & 9390   & 3100--5400 & 40$^{2}$, 22$^9$, 0$^9$\\\\ &       & 21/03/99, 22/03/99 \t\t & 1.5-2.0 & red  & 9000   & 7000--9000 & 40$^{2}$, 22$^9$, 0$^9$\\\\ &       & 21/03/99, 22/03/99 \t\t & 1.5-2.0 & blue & 6240   & 3100--5400 & 115$^9$, 79$^9$\\\\ &       & 21/03/99, 22/03/99 \t\t & 1.5-2.0 & red  & 6000   & 7250--9000 & 115$^9$, 79$^9$\\vspace*{0.15cm}\\\\ 6C1257+36 & 1.004 & 28/02/00           \t\t & 2.0-2.5 & blue & 11720  & 3100--5400 & 317$^2$\\\\ &       & 28/02/00           \t\t & 2.0-2.5 & red  & 11220  & 6750--8250$^7$ & 317$^2$\\vspace*{0.05cm}\\\\\\hline \\multicolumn{8}{l}{Notes:} \\\\ \\multicolumn{8}{l}{[1]: The seeing conditions on 01/03/00 were extremely poor. Some blue arm observations on this night were necessarily excluded due}\\\\ \\multicolumn{8}{l}{ to their very low signal--to--noise.}\\\\ \\multicolumn{8}{l}{[2]: Slit aligned along the radio axis.} \\\\ \\multicolumn{8}{l}{[3]: Blue arm data beyond the dichroic is included because of the high signal--to--noise ratio of the Mg\\textsc{ii} 2800 line}\\\\ \\multicolumn{8}{l}{[4]: Slit aligned with the optical emission.} \\\\ \\multicolumn{8}{l}{[5]: The red arm observations of 6C1011+36 on 01/03/00 were excluded due to the very poor seeing conditions.}\\\\ \\multicolumn{8}{l}{[6]: The [Ne\\textsc{v}]3426 line for 6C1019+39 is very weak, and only one red arm central wavelength was used in order to maximise the}\\\\ \\multicolumn{8}{l}{ signal--to--noise ratio of this emission line. }\\\\ \\multicolumn{8}{l}{[7]: Only one red arm central wavelength was used for this object, due to the loss of the latter half of the night of 01/03/00.}\\\\ \\multicolumn{8}{l}{[8]: The red arm exposure time for 6C1217+36 is considerably lower than that for the blue arm, due to technical problems during}\\\\ \\multicolumn{8}{l}{ the observations.} \\\\ \\multicolumn{8}{l}{[9]: Slit angled in other directions to include other objects in the field.} \\\\ \\end{tabular} \\end{center} \\end{table*}  \\begin{table*} \\scriptsize{} \\caption{Spectroscopic properties of the 6C radio galaxies. The [O\\textsc{ii}] 3727\\AA\\, integrated flux (units of $10^{-19}\\rm{W\\,m}^{-2}$) corresponds to the line flux along the entire length of the slit calculated by integrating the \\oo 3727\\AA\\, intensities shown in Figs.~\\ref{Fig: 1}--\\ref{Fig: 10}(d). All the flux ratios and flux densities quoted in this table are corrected for galactic extinction using the {\\it E(B-V)} for the Milky Way from the NASA Extragalactic Database (NED) and the parameterized galactic extinction law of Howarth (1983). All flux ratios are measured relative to \\oo 3727\\AA\\, (value of 100) within the extracted one-dimensional spectrum.  The error on the \\oo 3727\\AA\\, line flux is dominated by calibration errors, estimated to be $\\lta 10$\\%. Errors on the other lines and this calibration error are added in quadrature with the errors arising from photon statistics. There may be a small $\\lta 5$-10\\% systematic offset between the red and blue arm measurements due to different spatial extraction regions - the extracted regions are of the same spatial extent, but may not be centred on the same position. Galaxies may not always be completely centred in the slit at the red and blue extremes due to differential refraction, as the slit position angle was not necessarily aligned with the parallactic angle. This should be minimal however, as observations were generally taken at as low an airmass as possible, with a maximum value typically $< 1.3$.} \\footnotesize \\begin{center} \\begin{tabular} {lccccccccc} \\multicolumn{2}{c}{Source}   &{6C0943}&{6C1011}&{6C1017} &{6C1019} & {6C1129} & {6C1217} & {6C1256} & {6C1257}\\\\ \\multicolumn{2}{c}{Redshift} & {1.035} & {1.042} & {1.053} & {0.922} & {1.060} & {1.088} & {1.128} & {1.004}\\\\ \\multicolumn{2}{c}{Radio Size (kpc)} & {92} & {444} & {65} & {67}    & {141}   & {38}    & {155}   & {336}\\\\ \\multicolumn{2}{c}{Milky Way E(B-V)}& {0.018} & {0.013}& {0.008}& {0.013}& {0.030}& {0.017}&{0.014}& {0.014}\\\\ \\multicolumn{2}{c}{Integrated [O\\textsc{ii}] 3727\\AA\\, flux} & {5.01} &{1.24} & {6.86} & {2.43} & {5.14} & {0.51} & {2.09} & {2.24}\\vspace*{0.15cm}\\\\ {CIV 1549}  & {Flux Ratio}            & {38.0} & {543}  & {123}  & {*}    & {29.3} & {*}    & {45.0} & {*}\\\\ & {Error}                 & {9.8}  & {111}  & {24.9} & {*}    & {7.6}  & {*}    & {22.6} & {*}\\\\ & {Equiv. width}          & {103}  & {90}   & {151}  & {*}    & {-}    & {*}    & {-}    & {*}\\vspace*{0.05cm}\\\\ {He\\textsc{ii} 1640} & {Flux Ratio}   & {36.8} & {197}  & {59.0} & {*}    & {16.7} & {*}    & {34.5} & {*}\\\\ & {Error}                 & {8.9}  & {42.8} & {12.4} & {*}    & {4.9}  & {*}    & {17.3} & {*}\\\\ & {Equiv. width}          & {97}   & {34}   & {66}   & {*}    & {39}   & {*}    & {-}    & {*}\\vspace*{0.05cm}\\\\ {C\\textsc{iii}] 1909} & {Flux Ratio}  & {29.6} & {161}  & {57.9} & {58.8} & {22.4} & {42.4} & {18.7} & {69.6}\\\\ & {Error}                 & {6.1}  & {32.7} & {11.9} & {22.7} & {4.8}  & {18.2} & {9.4}  & {22.2}\\\\ & {Equiv. width} \t      & {49}   & {43}   & {58}   & {43}   & {58}   & {22}   & {-}    & {25}\\vspace*{0.05cm}\\\\ {C\\textsc{ii}] 2326} & {Flux Ratio}   & {15.7} & {33.9} & {20.4} & {22.2} & {1.9}  & {35.1} & {9.1}  & {12.1}\\\\ & {Error}        \t      & {3.4}  & {7.3}  & {4.1}  & {4.6}  & {1.5}  & {25.6} & {4.6}  & {5.0}\\\\ & {Equiv. width} \t      & {23}   & {9}    & {24}   & {65}   & {-}    & {18}   & {-}    & {14}\\vspace*{0.05cm}\\\\ {[NeIV] 2425} & {Flux Ratio} \t      & {10.9} & {87.9} & {23.6} & {4.5}  & {16.0} & {7.8}  & {4.3}  & {23.2}\\\\ & {Error}        \t      & {2.4}  & {17.4} & {4.7}  & {2.9}  & {4.6}  & {7.4}  & {2.2}  & {7.1}\\\\ & {Equiv. width} \t      & {17}   & {26}   & {28}   & {14}   & {58}   & {-}    & {-}    & {27}\\vspace*{0.05cm}\\\\ {Mg\\textsc{ii} 2798} & {Flux Ratio}   & {*}    & {185}  & {*}    & {-}    & {*}    & {*}    & {*}    & {*}\\\\ & {Error}        \t      & {*}    & {50.5} & {*}    & {-}    & {*}    & {*}    & {*}    & {*}\\\\ & {Equiv. width}          & {*}    & {76}   & {*}    & {-}    & {*}    & {*}    & {*}    & {*}\\vspace*{0.05cm}\\\\ {[NeV] 3426} & {Flux Ratio}           & {16.1} & {83.1} & {25.4} & {2.1}  & {11.6} & {5.8}  & {14.8} & {28.6}\\\\ & {Error}                 & {4.7}  & {17.6} & {5.1}  & {1.9}  & {3.0}  & {4.6}  & {5.8}  & {6.4}\\\\ & {Equiv. width}          & {36}   & {38}   & {45}   & {-}    & {52}   & {3}    & {26}   & {33\\vspace*{0.05cm}}\\\\ {[O\\textsc{ii}] 3727} & {Flux Ratio}  & {100}  & {100}  & {100}  & {100}  & {100}  & {100}  & {100}  & {100}\\\\ & {Equiv. width}          & {216}  & {41}   & {195}  & {58}   & {226}  & {59}   & {147}  & {83}\\vspace*{0.05cm}\\\\ {[Ne\\textsc{iii}] 3869} & {Flux Ratio}& {35.5} & {98.4} & {32.5} & {17.3} & {30.0} & {5.8}  & {17.7} & {22.8}\\\\ & {Error}                 & {8.3}  & {20.7} & {6.6}  & {4.7}  & {6.0}  & {5.4}  & {4.9}  & {4.9}\\\\ & {Equiv. width}          & {94}   & {39}   & {69}   & {11}   & {56}   & {4}    & {34}   & {25}\\vspace*{0.05cm}\\\\ {H$\\zeta$ 3889}   & {Flux Ratio}      & {-}    & {21.0} & {7.3}  & {-}    & {-}    & {-}    & {12.4} & {8.5}\\\\ & {Error}        \t      & {-}    & {6.7}  & {1.5}  & {-}    & {-}    & {-}    & {3.7}  & {2.8}\\\\ & {Equiv. width}\t      & {-}    & {8}    & {16}   & {-}    & {-}    & {-}    & {24}   & {10}\\vspace*{0.05cm}\\\\ {H$\\epsilon$ + [Ne\\textsc{iii}] 3967}&{Flux Ratio}&{5.9}&{33.9}&{12.5}&{-}& {9.7}  & {13.7} & {-}    & {13.4}\\\\ & {Error}                 & {2.7}  & {9.5}  & {2.5}  & {-}    & {2.5}  & {6.4}  & {-}    & {13.4}\\\\ & {Equiv. width}          & {9}    & {13}   & {29}   & {-}    & {17}   & {6}    & {-}    & {12}\\vspace*{0.05cm}\\\\ {H$\\delta$ 4102} & {Flux Ratio}       & {-}    & {10.5} & {19.0} & {-}    & {-}    & {-}    & {-}    & {-}\\\\ & {Error}                 & {-}    & {9.1}  & {4.0}  & {-}    & {-}    & {-}    & {-}    & {-}\\\\ & {Equiv. width}          & {-}    & {-}    & {53}   & {-}    & {-}    & {-}    & {-}    & {-}\\vspace*{0.05cm}\\\\ {H$\\gamma$ 4340} & {Flux Ratio}       & {-}    & {49.2} & {16.3} & {*}    & {*}    & {*}    & {*}    & {*}\\\\ & {Error}                 & {-}    & {14.3} & {3.2}  & {*}    & {*}    & {*}    & {*}    & {*}\\\\ & {Equiv. width}          & {-}    & {11}   & {47}   & {*}    & {*}    & {*}    & {*}    & {*}\\vspace*{0.15cm}\\\\ {Mean Flux Density} & {2100-2300\\AA} & {1.45} & {2.41} & {2.60} & {0.93} & {0.72} & {1.92} & {0.01} & {1.73}\\\\ {Error} &                            & {0.21} & {0.30} & {0.32} & {0.32} & {0.13} & {0.47} & {0.04} & {0.49}\\vspace*{0.05cm}\\\\ {Mean Flux Density} & {2450-2700\\AA} & {1.17} & {-}    & {-}    & {0.96} & {-}    & {-}    & {-}    & {1.01}\\\\ {Error} &                            & {0.23} & {-}    & {-}    & {0.32} & {-}    & {-}    & {-}    & {0.37}\\vspace*{0.05cm}\\\\ {Mean Flux Density} & {3500-3700\\AA} & {1.01} & {1.22} & {1.63} & {2.63} & {0.74} & {1.01} & {0.52} & {1.67}\\\\ {Error} &                            & {0.16} & {0.20} & {0.23} & {0.33} & {0.15} & {0.27} & {0.12} & {0.25}\\vspace*{0.05cm}\\\\ {Mean Flux Density} & {4050-4250\\AA} & {1.09} & {2.11} & {0.68} & {-}    & {-}& {-}    & {0.87} & {1.91$^\\dag$}\\\\ {Error} &                            & {0.31} & {0.57} & {0.46} & {-}    & {-}& {-}    & {0.39} & {0.53$^\\dag$}\\\\ \\normalsize \\end{tabular} \\end{center} \\end{table*}  \\addtocounter{table}{-1} \\begin{table*} \\scriptsize{} \\caption{\\textbf{-- Continued.} Equivalent widths are measured in the rest frame of the galaxy. A single dash is used to represent emission lines of such low flux as to be unobservable, and asterisks are used to represent emission lines outside the wavelength range covered by our spectra. Mean continuum flux densities for line-free wavelength ranges (or as much of the range as possible provided at least 100\\AA\\, are covered in the spectrum) are measured from the extracted one-dimensional spectrum; the (weak) H$\\delta$ emission line was subtracted from the spectrum before calculating the mean continuum level for the wavelength range 4050--4250\\AA. Values are in units of $10^{-21}\\rm{W\\,m}^{-2}\\AA^{-1}$; the uncertainty given is the error on the mean value in that wavelength region. \\dag\\,\\, indicates that a shorter wavelength range has been used in measuring the flux density. } \\begin{center} \\begin{tabular} {lcc} \\end{tabular} \\end{center} \\end{table*}  The layout of the paper is as follows.  Section 2 gives details of the sample selection, the observations, and the data reduction techniques. In Section 3, the results of the spectroscopic observations are presented, including a two-dimensional analysis of the \\oo 3727\\AA\\, emission line. Section 4 describes an analysis of the ionization and kinematical properties of the emission line gas and compares these with the results of the more powerful 3CR galaxies at the same redshift.  Our results are discussed in section 5, and a summary of our conclusions is given in section 6.  Throughout the paper, values for the cosmological parameters of $\\Omega_0=0.3$, $\\Omega_{\\Lambda}=0.7$ and $H_{0} = 65 \\rm{km\\,s}^{-1}$Mpc$^{-1}$ are assumed. ",
        "conclusions": "Very deep spectroscopic observations have been made of an unbiased subsample of 8 6C galaxies at $z \\sim 1$. Many emission lines have been observed over the rest-frame wavelength range 1500-4500\\AA, and a study of the two-dimensional kinematics of the emission line gas has been carried out.  Our conclusions can be summarised as follows:  \\begin{enumerate} \\item The observed spectra of the 6C galaxies are quite varied, both in the strength of the emission lines observed, and the line ratios observed. \\item The composite spectra of the 6C galaxies are similar to those of Best {\\it et al} (2000a) for the 3CR sources at the same redshift. \\item 6C sources with $D_{\\rm{rad}} > 120\\,\\rm{kpc}$ host less powerful AGN than 3CR sources of a similar size at the same redshift. Their spectra are well explained by photoionization, typically with a lower ionization parameter than their more powerful 3CR counterparts.  The total emission line luminosities of large 6C sources are also smaller than that of large 3CR sources, and their emission line regions are observable out to smaller physical scales. \\item For small radio sources, $D_{\\rm{rad}} < 120$\\,kpc, a combination of AGN photoionization and shock ionization provides the best explanation of their spectra.  Their emission line regions typically have a similar size to the extent of the radio source. \\item The velocity profiles of shock ionized EELRs are distorted, whereas photoionized EELRs display smooth velocity profiles. \\item The ionization properties of the subsample can be explained by a simple model incorporating photoionization by the AGN and a luminous shock associated with the expanding radio source. \\item The kinematics of the EELRs of 6C radio sources are similar to those of more powerful 3CR sources. \\item A high velocity component is observed in the EELR of 6C1019+39 at $\\sim 700 \\rm{km\\,s}^{-1}$, close to the host galaxy. \\end{enumerate}  In summary, the properties of 6C radio galaxies at $z \\sim 1$ are similar to 3CR sources at the same redshift, despite the decrease in radio power between the samples.   A strong anticorrelation of emission line luminosity with source size is found.  In addition, we also find tentative evidence that the range of velocities observed in the emission line gas is most likely dependent on the properties of the radio source rather than the underlying gravitational potential of the host galaxy.   In a companion paper (Paper 2), this dataset is compared with lower redshift sources of the same radio power to break the radio power-redshift degeneracy, and to investigate the intrinsic dependencies of radio galaxy properties on redshift, radio power and source age."
    },
    "0208/astro-ph0208254_arXiv.txt": {
        "abstract": "\\noindent We present observations of the cluster of galaxies associated with the X-ray source RX~J0820.9+0752 and its dramatic central cluster galaxy in the optical and X-ray wavebands. Unlike other cooling flow central cluster galaxies studied in detail, this system does not contain a powerful radio source at its core, and so provides us with an important example where we expect to see only the processes directly due to the cooling flow itself. A 9.4\\thinspace ks Chandra observation shows that the hot intracluster gas is cooling within a radius of 20\\kpc\\ at a rate of a few tens of solar masses per year. The temperature profile is typical of a cooling flow cluster and drops to below 1.8\\keV\\ in the core. Optical images taken with the AAT and HST show that the central galaxy is embedded in a luminous (L$_{H\\alpha}\\sim 5\\times 10^{42}$\\ergps), extended line-emitting nebula that coincides spatially with a bright excess of X-ray emission, and separate, off-nucleus clumps of blue continuum that form part of a patchy structure arcing away from the main galaxy. The X-ray/H$\\alpha$ feature is reminiscent of the 40\\kpc\\ long filament observed in A 1795 which is suggested to be a cooling wake, produced by the motion of the central cluster galaxy through the intracluster medium. We present optical spectra of the central cluster galaxy and its surroundings, and find that the continuum blobs show stronger line emission, differing kinematic properties and more extreme ionization ratios than the surrounding nebula. Accounting for the strong intrinsic reddening and its significant variation over the extent of the line emitting region, we have fit the continuum spectra of the blobs and the nucleus using empirical stellar spectra from a library. We found that continuum emission from early main sequence stars can account for the blue excess light in the blobs. Kinematical properties associate the gas in the system with a nearby secondary galaxy, suggesting some kind of tidal interaction between the two. We suggest that the secondary galaxy has moved through the cooling wake produced by the central cluster galaxy, dragging some of the gas out of the wake and triggering the starbursts found in the blobs. ",
        "introduction": "\\label{introduction}  The hot intracluster medium in the cores of many clusters of galaxies emits so much energy in the X-ray band that the cooling time is substantially shorter than the cluster age. The resulting decrease in gas pressure leads to a highly subsonic inflow of material towards the cluster centre -- a cooling flow (Fabian 1994). Recent X-ray spectra from {\\sl XMM-Newton} have confirmed the short central cooling times and low central cluster temperatures required by a cooling flow, but show a surprising deficit of spectral lines from gas cooling below $1-2$\\keV{} (Peterson \\etal 2001; Kaastra \\etal 2001). The rates of cooling (or `mass deposition rate') thus appear to be reduced from those deduced from earlier X-ray missions, and in the `classical' (\\ie pre-{\\sl XMM-Newton} and {\\sl Chandra}) picture of cooling flows. It is possible that the age of the cooling flow can be substantially reduced to around a few Gyr, for example if the intracluster medium is stirred up by infall of subclusters (Allen, Ettori \\& Fabian 2001a). The current data are only consistent with the previously-deduced high mass deposition rates if excess intrinsic absorption is introduced to selectively remove the emission below 1\\keV\\ (David \\etal 2001; Schmidt, Allen \\& Fabian 2001; Ettori \\etal 2001).  Alternatively, a possible sink for the `missing' soft X-ray luminosity expected from steady cooling flows could be the luminous (and excess) line emission seen in the ultraviolet/optical waveband (Heckman \\etal 1989; Crawford \\etal 1999), and dust emission in the sub-mm/far-infrared (Edge \\etal 1999; Allen \\etal 2001c) seen associated with many central cluster galaxies in cooling flows. In a cold mixing model, the gas below $1-2$\\keV\\ drops from the flow to be rapidly cooled by being churned with embedded cold gas clouds, and then the `missing' energy is re-emitted at longer wavelengths (Fabian \\etal 2002).  Another explanation for the observed properties has recently been put foward by Voigt~\\etal~(2002). They suggest that cooling can be balanced by heat conduction along the temperature gradient in the outer parts of a cluster with only a small cooling flow operating at the centre ($r\\approxlt 20$\\kpc), thereby reducing the mass deposition rates to values consistent with observations.  Most cooling flow models imply some correspondence in both the spatial and luminous properties of the cooling gas seen in X-rays and the emission seen in the lower-energy wavebands -- perhaps such as the X-ray/\\ha\\ filament extending from the central cluster galaxy in the A1795 cluster (Fabian \\etal 2001a; Crawford \\etal 2002 in preparation), or the associated X-ray/emission-line regions seen in objects such as A2199 (Johnstone \\etal 2002), Virgo (Young, Wilson \\& Mundell 2002), or A2390 (Allen \\etal 2001a). These line-luminous central cluster galaxies, however, are also sites for extended regions of massive star formation that are presumed to be a final sink for material cooling from the intracluster medium (e.g. Crawford \\& Fabian 1993). Ionization from these stars may also be a large contribution to powering the emission-line nebulae.  A further heat source which could balance radiative cooling in a cluster is a central radio source. Most central cluster galaxies have a radio source with jets supplying energy into the surrounding gas, often blowing bubbles of relativistic plasma (e.g. McNamara \\etal 2000; Fabian \\etal 2000). At issue here is whether these bubbles lead to local heating of the intracluster medium (Churazov~\\etal~2002; Churazov \\etal 2001; Br\\\"uggen \\& Kaiser 2001; Reynolds, Heinz \\& Begelman 2002) or buoyantly rise and transport the energy to large radii (Fabian \\etal 2002). It is also possible that cooling is balanced by a combination of heating by the radio source and conduction (Ruszkowski \\& Begelman 2002). In this paper we present a central cluster galaxy which has a particularly weak central radio source. (Of course such sources may go through cycles of activity and our object may just be in a low state.) It also appears to have a strong correspondence between its cooling X-ray emission and its properties at optical wavelengths. The galaxy is at the centre of the cluster associated with the {\\sl ROSAT} X-ray source RX~J0820.9+0752\\footnote{For convenience, we shall use `RX~J0821' throughout the rest of this paper to refer to the central galaxy and its associated line-emitting nebula.}, the coordinates of the nucleus of the CCG being $\\alpha=08$:21:02.28, $\\delta=+07$:51:46.8 (J2000). The cluster was discovered as part of the extended Brightest Cluster Sample (Ebeling \\etal 2000), and optical spectroscopy on the central galaxy yielded a redshift of 0.110 (Crawford \\etal 1995). It has a {\\sl ROSAT} $0.1-2.4$\\keV\\ (rest frame) luminosity of L$_X=2.09\\times10^{44}$\\ergps\\footnote{We assume a cosmology with a Hubble constant of $H_0=50$\\kmpspMpc\\ and a cosmological deceleration parameter of $q_0=0.5$ throughout this paper.}, and an estimated temperature of 4.4\\keV. The spectrum of the central cluster galaxy showed it to have strong \\ha+[NII] emission lines (a slit \\ha\\ luminosity of around $3.3\\pm0.2\\times10^{41}$\\ergps) but with a lack of \\hb\\ that implied large amounts of intrinsic obscuration to be present in this system (Crawford \\etal 1995).  RX~J0820.9+0752 is a strong CO emitter (Edge 2001) with an implied molecular gas mass of 3.9$\\pm0.4\\times 10^{10}$~M$_\\odot$ and offset in velocity from the main galaxy by about $+260$\\kmps. The CO emission is centred on the central galaxy but extended to the W by about 5\\arc{} (Edge \\& Frayer 2002 in prep.). RX~J0820.9+0752 is also detected by {\\it IRAS} at 60$\\mu$m (Edge 2001) implying a dust mass of 2.2$\\times 10^7$~M$_\\odot$ for a dust temperature of 40~K. A near-infrared spectrum revealed Pa$\\alpha$ extended over 2.4 \\arc{} to the north of the central galaxy, but there was no significant detection of any \\mbox{1-0~S~series} H$_2$ line emission, despite the strong CO detection (Edge \\etal 2002). ",
        "conclusions": "The {\\sl Chandra} data show that the central galaxy of RX~J0820.9+0752 is embedded in cluster gas with a central temperature of 1.8\\keV and moderate X-ray luminosity. Assuming that no heat sources are present, the observed surface brightness profile implies that the hot intracluster gas is cooling within a radius of 20\\kpc\\ at rates of a few tens of solar masses a year. The X-ray emission is clearly extended by around 22\\kpc\\ towards the NW of the central galaxy. The scale and orientation of this extended emission is well matched to the region within the break of the $\\dot{M}$ profile, the luminous \\ha\\ nebula, and the strong CO emission. The line-emitting nebula cannot be seen in the HST image due to the latter's relatively blue passband. The image shows, if anything, a sharp decline in the diffuse continuum in this system in the same region at radii beyond around 8\\kpc. Thus it appears the strong extended X-ray/\\ha\\ feature is heavily obscured by dust at shorter wavelengths which agrees with our estimates of {\\sl E(B-V)} obtained from the Balmer decrement in this region. The HST image does, however, reveal large clumps of emission (blue continuum and/or blue emission lines) that cross in an arc from this region to the SE of the central galaxy; a secondary galaxy lies $\\sim$11\\kpc\\ in this direction, opposite to the X-ray/\\ha\\ extension.  The radio power of RX~J0821 is the third lowest of all line-emitting CCGs in the BCS (only A262 and RX~J1733.0+4345/IC1262 have lower powers) and may not be associated with the core of the CCG (as is found in the vast majority of BCS line-emitting CCGs) but instead be coincident with the main star-forming regions. The discovery of radio emission from star formation rather than nuclear activity is intriguing and needs more detailed radio imaging to verify.  We have taken spectra from two slit position angles across the main galaxy, including the extended nebula to the NW and across the two arcing lines of clumps seen in the HST image to the ENE. We detect \\ha\\ in emission out to 24\\kpc\\ to the NW of the galaxy. The gas in this extended nebula is redshifted by a roughly constant amount of $\\sim100$\\kmps\\ relative to the central galaxy, and then undergoes a strong gradient of about 150\\kmps in 8\\kpc\\ as it crosses this galaxy and out to the SE. The nucleus of the main galaxy shows little line emission, no intrinsic reddening or evidence for an excess blue continuum component. The radial velocities of the stellar and nebular components in the nucleus of the CCG are in very good agreement, but they are offset significantly by $\\sim 200$\\kmps to the NW, and this separation appears to be gradually decreasing to almost zero towards the E.  There is a marked overall decline in line-width from about 300\\kmps\\ to just below 200\\kmps\\ over the extent of the nebula from the NW towards the nucleus, followed by a sharp increase back to 300\\kmps\\ about 3\\kpc\\ from the nucleus. The line-width then decreases again over the inner region of the galaxy.  We isolate two main regions of excess line and continuum emission to the NW of the central cluster galaxy. The \\ha SB feature lies at the end of the inner arc of clumped emission seen in the HST image, where it tails off into the core of the \\ha\\ nebula. Consistent with its position also at the edge of the sharp `bite' of extinction, the strong line emission shows a Balmer decrement indicating a large degree of intrinsic reddening, with {\\sl E(B-V)}$\\sim0.7$. The excess blue continuum is well fit by a combination of O and A-type stars.  A more isolated blue blob (BB1) lies further to the NW, still within the core of the \\ha\\ nebula. Its spectrum again shows a strong excess blue continuum (again well fit by O and A main sequence stars), and a strong Balmer series in line emission. Again, the Balmer decrement indicates a degree of intrinsic reddening at {\\sl E(B-V)}$\\sim0.3$. Generally reddening is found to be very high within the region occupied by the \\ha\\ nebula.  We also obtain the spectra of two continuum blobs, one from each of the clumped lines of emission to the ENE of the main galaxy. BB3 is a conspicuous bright blob forming part of the inner arc, that is bright in \\ha, lying within the peak of emission. A high level of intrinsic reddening is indicated by the strong Balmer decrement ({\\sl E(B-V)}$\\sim0.73$) and evidence for a small amount of excess blue continuum, fit by a range of main sequence stars. BB2 forms part of the outer crossing arc to the ENE, and shows little line emission and an insignificant amount of intrinsic reddening. Its blue continuum, however, -- otherwise quite similar to that of BB3 -- is remarkable in showing a clear Balmer line sequence in absorption that is fit by a complex mixture of stars of all types.  The emission line gas at the locations of BB1, BB3 and \\ha SB has a lower [NII]/\\ha, smaller linewidths and stronger \\ha\\ emission than found in the surrounding regions. BB2 deviates in this and other respects from the other blobs."
    },
    "0208/hep-ph0208069_arXiv.txt": {
        "abstract": "{We derive the allowed ranges of the spin--independent interaction cross section $\\sigsip$ for the elastic scattering of neutralinos on proton for wide ranges of parameters of the general Minimal Supersymmetric Standard Model.  We investigate the effects of the lower limits on Higgs and superpartner masses from colliders, as well as the impact of constraints from $\\bsgamma$ and the new measurement of $\\gmtwo$ on the upper and lower limits on $\\sigsip$. We further explore the impact of the neutralino relic density, including coannihilation, and of theoretical assumptions about the largest allowed values of the supersymmetric parameters. For $\\mu>0$, requiring the latter to lie below $1\\tev$ leads to $\\sigsip\\gsim 10^{-11}\\pb$ at $\\mchi\\sim100\\gev$ and $\\sigsip\\gsim 10^{-8}\\pb$ at $\\mchi\\sim1\\tev$. When the supersymmetric parameters are allowed above $1\\tev$, for $440\\gev \\lsim \\mchi\\lsim 1020 \\gev$ we derive a {\\em parameter--independent lower limit} of $\\sigsip\\gsim 2\\times 10^{-12}\\pb$. (No similar lower limits can be set for $\\mu<0$ nor for $1020\\gev\\lsim\\mchi\\lsim2.6\\tev$.) Requiring $\\abundchi<0.3$ implies a {\\em parameter--independent upper limit} $\\mchi\\lsim2.6\\tev$. The new $\\epem$--based measurement of $(g-2)_{\\mu}$ restricts $\\mchi\\lsim 350\\gev$ at $1\\,\\sigma$~CL and $\\mchi\\lsim515\\gev$ at $2\\,\\sigma$~CL, and implies $\\mu>0$.  The largest allowed values of $\\sigsip$ have already become accessible to recent experimental searches.  } ",
        "introduction": "The hypothesis of the lightest neutralino $\\chi$, as the lightest supersymmetric particle (LSP), providing the dominant contribution to cold dark matter (CDM) in the Universe, has inspired much activity in the overlap of today's particle physics and cosmology. It is well--known that the relic density of the neutralinos is often comparable with the critical density~\\cite{ehnos84,jkg96}.  The expectation that the Galactic dark matter (DM) halo is mostly made of weakly--interacting massive particles (WIMPs) has further led to much experimental activity. In particular, the experiments looking for CDM WIMPs elastically scattering off underground targets have recently set limits on spin--independent (SI), or scalar, cross section of the order of $10^{-6}\\pb$~\\cite{cdms00,edelweiss-june02,ukdmc-zeplini-sep02}.  They have also nearly ruled out the region of ($\\mchi,\\sigsip$) that has been claimed by the DAMA experiment to be consistent with an annual modulation effect~\\cite{dama00}.  Initial and early studies~\\cite{gw85,griest89,ddstudies:past,dn93scatt:ref} were followed by more recent work~\\cite{ddstudies:recent}, where it was concluded that current experimental sensitivity is generally comparable with the ranges expected from the neutralino WIMP in the Minimal Supersymmetric Standard Model (MSSM). It is, however, still at least on order of magnitude, or so, above the ranges predicted by recent analyses of the Constrained MSSM (CMSSM)~\\cite{efo00,cmssm:recent, rrn1,lr-latalk,cn0208}.  For comparison, cross sections for spin--dependent (SD) interactions are in the case of the neutralino generally some two or three orders of magnitude larger than the SI ones. On the other hand, at present detectors are still not sensitive enough to explore the parameter space of the MSSM, despite recent progress~\\cite{ddspin:recentexpt}.  In light of the ongoing and planned experimental activities, it is timely to conduct a thorough and careful re--analysis of the predicted cross sections for SI scattering of neutralino WIMPs. Such a study is rather challenging because resulting ranges often strongly depend on a given SUSY model and on related theoretical assumptions. They are further affected by experimental limits on SUSY, both from colliders and from indirect searches, as well as by cosmological input, where the relic abundance of the CDM has been measured with better accuracy both directly and in CMBR studies~\\cite{cmbstudiesrecent}.  Over the last few years and months there have been also new results for LEP lower bounds on the masses of the lightest Higgs and electroweakly--interacting superpartners, Tevatron lower limits on strongly--interacting superpartners, as well as limits on allowed SUSY contributions to $\\bsgamma$, and especially to the anomalous magnetic moment of the muon $\\gmtwo$~\\cite{gm202}. The new result for $\\gmtwo$ indicates a sizable deviation from the Standard Model prediction, whose value is still a subject of much discussion.  As we will show, when interpreted in terms of SUSY, the required extra contribution to $\\gmtwo$ plays a unique role in implying a stringent {\\em upper} bound $\\mchi\\lsim 350\\gev$ ($1\\,\\sigma$~CL) and $\\mchi\\lsim 512\\gev$ ($2\\,\\sigma$), but it does {\\em not} affect much the allowed ranges of the SI scattering cross section $\\sigsip$.  In this paper we carefully study the impact of the above constraints. In an attempt to minimize theoretical bias, we work here in the context of the general MSSM, which will be defined below, with an additional assumption of $R$--parity conservation. We focus here on the SI cross section case. Other recent studies of the general MSSM include~\\cite{efo01:gmssm,mpmg00,bg02:mssm}.  The results presented here show that the level of experimental sensitivity that has recently been reached~\\cite{cdms00,edelweiss-june02,ukdmc-zeplini-sep02} has now indeed allowed one to start exploring cosmologically favored ranges of the neutralino WIMP mass and SI cross section. However, we point out a number of caveats and relations and further discuss the origin of, and robustness of, the upper and lower limits on $\\sigsip$. In particular, for a big range of the heavy neutralino mass $440\\gev\\lsim\\mchi\\lsim 1020\\gev \\gev$ we are able to derive {\\em parameter--independent} lower bounds on $\\sigsip$. Collider lower limits on Higgs and superpartner masses plus requiring $\\abundchi<0.3$ alone leads to a {\\em parameter--independent} upper bound $\\mchi\\lsim2.6\\tev$. ",
        "conclusions": "We have delineated the ranges of the SI cross section $\\sigsip$ in the general MSSM, which are consistent with  current experimental bounds and for which one finds the expected amount of dark matter. We have further discussed the dependence of our results on the experimental constraints and on the underlying theoretical assumptions. While the ranges which we have obtaine extend over more than six orders of magnitude, we find it encouraging that the experimental sensitivity that has recently been reached, now allows one to explore our theoretical predictions for the MSSM. As we have argued above, smaller values of the WIMP mass and also larger values of $\\sigsip$ may be considered as more natural, which will hopefully be confirmed by a measuring a positive WIMP detection signal in the near future.   \\bigskip"
    },
    "0208/astro-ph0208289_arXiv.txt": {
        "abstract": "We present results from a {\\it Chandra} observation of the NGC\\,346 cluster. This cluster contains numerous massive stars and is responsible for the ionization of N66, the most luminous \\hii\\ region and the largest star formation region in the SMC. In this first paper, we will focus on the characteristics of the main objects of the field. The NGC\\,346 cluster itself shows only relatively faint X-ray emission (with $L_X^{unabs}\\sim 1.5\\times 10^{34}$~erg~s$^{-1}$), tightly correlated with the core of the cluster. In the field also lies \\hd, a LBV star in a binary (or possibly a triple system) that is detected for the first time at X-ray energies. The star is X-ray bright, with an unabsorbed luminosity of $L_X^{unabs}\\sim 1.7\\times 10^{34}$~erg s$^{-1}$, but needs to be monitored further to investigate its X-ray variability over a complete 19\\,d orbital cycle. The high X-ray luminosity may be associated either with colliding winds in the binary system or with the 1994 eruption. \\hd\\ is surrounded by a region of diffuse X-ray emission, which is a supernova remnant. While it may be only a chance alignment with \\hd, such a spatial coincidence may indicate that the remnant is indeed related to this peculiar massive star. ",
        "introduction": "The Small Magellanic Cloud (SMC) is an irregular galaxy at a distance of 59~kpc \\citep{ma86} that forms a pair with the Large Magellanic Cloud (LMC). Both are satellites of our own Galaxy. The interstellar extinction towards the Magellanic Clouds is low, allowing studies of the X-ray sources to be undertaken with clarity. \\\\  Previous X-ray observations of the SMC have been made with the {\\it Einstein Observatory}, {\\it ASCA} and {\\it ROSAT}. These observations included surveys of the point source population \\citep{ka99,sc99,hfp,sa00,yo00} and in particular studies of the properties of the X-ray binary population \\citep{ha00}. The recent launch of {\\it Chandra}, however, provides an opportunity to study the SMC with a far greater sensitivity and spatial resolution than ever before. \\\\  We have obtained a Chandra observation of the giant \\hii\\ region N66 \\citep{he56}, the largest star formation region in the SMC, in order to study the young cluster NGC\\,346 and its interaction with the surrounding interstellar medium. This cluster contains numerous massive stars \\citep{ma89}, of spectral types as early as O2 \\citep{wal02}. The large number of massive stars is not the only feature of interest in this field. On the outskirts of NGC\\,346 lies the remarkable star \\hd, which underwent a Luminous Blue Variable (LBV)-type eruption in 1994. This massive binary (or triple?) system has been monitored and analysed for its varying spectral and photometric properties in visible and UV wavelengths since the early 80's (Koenigsberger et al. 2000; Sterken \\& Breysacher 1997, and references therein). Previous X-ray observations of NGC\\,346 \\citep{ikt,hfp} detected a bright extended source around \\hd\\ which has been attributed to a supernova remnant (SNR); however, the crude instrumental resolution prohibited unambiguous detections of a point source associated with \\hd.  In addition to \\hd, the young massive stars of NGC\\,346 and their interacting winds are likely to produce X-ray emission, but have not been detected previously. The sensitive, high-resolution {\\it Chandra} observation of the NGC\\,346 field thus provides an excellent opportunity to study a variety of phenomena involving emission at X-ray energies. \\\\  In this paper, we will describe in \\S~2 the observations used in this study, then discuss the available data on NGC\\,346, \\hd, and the extended emission in \\S~3, 4, and 5, respectively. Finally, conclusions are given in \\S~6. The second part of this work (Naz\\'e et al., paper II) will describe the properties of the other sources detected in the field.\\\\ ",
        "conclusions": "In this series of articles, we report the analysis of the {\\it Chandra} data of N66, the largest star formation region of the SMC. In this first paper, we have focused on the most important objects of the field: the NGC\\,346 cluster and \\hd. \\\\  The cluster itself is relatively faint, with a total luminosity of $L_X^{unabs}\\sim 1.5\\times 10^{34}$~erg~s$^{-1}$ in the 0.3-10.0~keV energy range. Most of this emission seems correlated with the location of the brightest stars of the core of the cluster, but the level of X-ray emission probably cannot be explained solely by the emission from individual stars. \\\\  In this field lies another object of interest: \\hd, a remarkable star that underwent a LBV-type eruption in 1994. {\\it Chandra} is in fact the first X-ray telescope to detect \\hd\\ individually. In X-rays, the star appears very bright, comparable only to the brightest WR stars in the Galaxy, but still fainter than $\\eta$ Carinae.  The comparison of our results with future X-ray observations will enable us to better understand \\hd: for example, phase-locked variations will be analysed in the perspective of the colliding wind behaviour of this binary, while other variations may be related to the recent LBV eruption.  Follow-up observations are thus needed to complete the study of this system. \\\\  A bright, extended X-ray emission is seen to surround \\hd. It is most probably due to a SNR whose progenitor is unknown. The spatial coincidence of this extended X-ray emission with the peculiar massive star suggest an association between these two objects. We also note the close resemblance of this X-ray emission to the Carina Nebula in which the LBV $\\eta$ Carinae lies. \\\\"
    },
    "0208/astro-ph0208130_arXiv.txt": {
        "abstract": "{\\small We report preliminary results of a RossiXTE campaign on the 2002 outburst of the black-hole candidate GX~339--4. We show power density spectra of five observations during the early phase of the outburst. The first four power spectra show a smooth transition between a Low State and a Very  High State. The fifth power spectrum resembles a High State, but a strong 6 Hz QPO appears suddenly within 16 seconds. } ",
        "introduction": "After almost three years of quiescence, the ``persistent\" BHC GX~339--4 became active again on 2002 March 26 \\cite{atel}. This source is important , as in the past it has shown in the past all of the ``canonical'' states of BHCs \\cite{mendez}.   \\begin{figure}[htb] \\centering \\psfig{file=belloni2_1.eps,width=9cm} \\caption{Top panel: full ASM light curve for GX~339--4. Bottom panel: zoom of the new activity period considered here. The times of our pointings are marked by dotted lines.} \\label{fig:ex} \\end{figure}  The full RossiXTE/ASM light curve from the start of the mission is shown in Fig. 1. We started our observations with RossiXTE on April 3rd and obtained roughly one pointing per week since then. The ASM rate (1.5-1 keV) increased to 0.8 Crab in less than two months, then decreased to $\\sim$0.5 Crab in three weeks, to start rising again to 1 Crab, and at the time of writing it is still brightening (see Fig. 1).  We present here preliminary results of the timing analysis of a subset of PCA observations: four during the early phase of the outburst, when the source count rate was increasing monotonically with time, and one a few days into the decay. The times of the five observations considered here are marked with dotted lines in the bottom panel of Fig. 1.  \\begin{figure}[htb] \\centering \\psfig{file=belloni2_2.ps,width=8cm} \\caption{Power Density Spectra of the first four observations marked in Fig.1. The spectra have been shifted by factors of ten in squared rms for clarity. } \\label{fig:ex} \\end{figure} ",
        "conclusions": ""
    },
    "0208/astro-ph0208460_arXiv.txt": {
        "abstract": "We have searched for nuclear radio emission from a statistically complete sample of 40 Sc galaxies within 30~Mpc that are optically classified as star-forming objects, in order to determine whether weak active galactic nuclei might be present.  Only three nuclear radio sources were detected, in NGC~864, NGC~4123, and NGC~4535.  These galaxies have peak 6-cm radio powers of $\\sim 10^{20}$~W~Hz$^{-1}$ at arcsecond resolution, while upper limits of the non-detected galaxies typically range from $10^{18.4}$~W~Hz$^{-1}$ to $10^{20}$~W~Hz$^{-1}$.  The three nuclear radio sources all are resolved and appear to have diffuse morphologies, with linear sizes of $\\sim 300$~pc.  This strongly indicates that circumnuclear star formation has been detected in these three \\hii\\ galaxies. Comparison with previous 20-cm VLA results for the detected galaxies shows that the extended nuclear radio emission has a flat spectrum in two objects, and almost certainly is generated by thermal emission from gas ionized by young stars in the centers of those galaxies. The 6-cm radio powers are comparable to predictions for thermal emission that are based on the nuclear H$\\alpha$ luminosities, and imply nuclear star formation rates of $0.08-0.8\\,M_\\odot$~yr$^{-1}$, while the low-resolution NRAO VLA Sky Survey implies galaxy-wide star formation rates of $0.3-1.0\\,M_\\odot$~yr$^{-1}$ in stars above $5M_\\odot$.  In a few of the undetected galaxies, the upper limits to the radio power are lower than predicted from the H$\\alpha$ luminosity, possibly due to over-resolution of central star-forming regions.  Although the presence of active nuclei powered by massive black holes cannot be definitively ruled out, the present results suggest that they are likely to be rare in these late-type galaxies with \\hii\\ spectra. ",
        "introduction": "\\label{sec:intro}  Most nearby normal and active galaxies with bulges appear to have supermassive black holes at their centers (see Kormendy \\& Gebhardt 2001 for a review), and the masses of these black holes ($\\sim 10^6 M_\\odot$--$10^9 M_\\odot$) seem well correlated with the galaxy bulge masses as inferred from the bulge luminosities \\citep{mag98}.  In fact, the relation of black hole mass to bulge stellar velocity dispersion is a correlation with far less scatter than that derived from the bulge luminosity \\citep{fer00,geb00a,tre02}.  The black hole masses from this relation are consistent with those derived from the widths of broad emission lines and their distances from the black hole as derived by reverberation mapping \\citep{geb00b,fer01}.  The relation of black hole mass to bulge mass appears confirmed in at least one bulgeless galaxy, the nearby spiral M33, where an upper limit of 1500--3000$M_\\odot$ has been found for the black hole mass \\citep{geb01,mer01b}.  Ho, Filippenko, \\& Sargent (1995, 1997a, 1997b) used the Palomar 200 inch telescope to perform a spectroscopic survey of 486 bright, nearby galaxies selected from the Revised Shapley-Ames Catalog of Bright Galaxies \\citep{san81}, in order to conduct a census of the population of active galactic nuclei (AGNs) in the nearby universe. Fifty-two Seyfert galaxies were found in that survey, and 82\\% of the objects in a statistical sample of 45 of these Seyferts were detected using the Very Large Array (VLA) at 6~cm, with observations of 15--18~min in length \\citep{ho01,ulv01}.  The detected radio sources in these low-luminosity Seyferts typically are associated with the AGNs, implying the presence of massive black holes powering the active nuclei.  In addition to the 52 Seyferts identified by \\citet{ho97a}, 206 galaxies from the Palomar sample were reported to contain \\hii\\ nuclei: galaxy nuclei with optical line ratios consistent with ionization by young, massive stars.  As a general (though not universal) rule, the AGNs (Seyferts and LINERs) are found in early-type galaxies (earlier than Sbc), while the \\hii\\ nuclei are found in galaxies of type Sb or later.  Although the late-type galaxies in the Palomar sample that are classified as \\hii\\ galaxies appear to be dominated by star formation, it also is possible that they contain weak AGNs whose optical signature has simply been overpowered by the more dominant signal from nuclear \\hii\\ regions. The presence of an AGN typically requires two constituents: a massive central black hole and a supply of gas to ``feed'' that black hole.  Since galaxies with \\hii\\ spectra generally harbor copious supplies of gas, the question of whether \\hii-dominated objects contain observable AGNs then may be equivalent to the question of whether these galaxies contain massive central black holes.  Good discriminators for the presence of an AGN include the presence of a compact source of hard X-rays or radio  \\begin{figure*}[t] \\centerline{\\psfig{file=table1.ps,width=18.5cm,angle=0}} \\end{figure*}  \\noindent emission identified with the galaxy nucleus.  Here, we report on a study of a subsample of \\hii\\ nuclei in Sc galaxies from the Palomar bright galaxy sample, in which we test for the presence of weak AGNs and central black holes by searching for radio emission from the galaxy nuclei.  \\vspace{0.6cm} ",
        "conclusions": "We have searched for nuclear radio sources in a statistical sample of 40 Sc galaxies hosting \\hii\\ nuclei, using imaging of new and archival VLA data at 6~cm.  Only three such sources were detected, in galaxies among those with the largest H$\\alpha$ luminosities in the sample.  The radio  powers and morphologies are consistent with the association of the detected radio sources with nuclear starbursts similar to that in NGC~253, rather than being caused by active galaxies powered by massive black holes. The ratio of radio power to H$\\alpha$ luminosity is generally consistent with thermal radio emission from nuclear starbursts, and the detected central sources could produce such thermal emission for star formation rates of only $0.08-0.8\\,M_\\odot$~yr$^{-1}$. However, nonthermal emission from supernova remnants also could contribute to the detected nuclear sources, requiring supernova rates of well under  0.01~yr$^{-1}$ and star formation rates of $\\sim 0.1M_\\odot$~yr$^{-1}$.  To the extent that radio cores are a common feature in low-luminosity AGNs, the nondetection of radio cores in \\hii\\ nuclei strongly suggests that they intrinsically lack AGNs, and, by inference, massive black holes.  This is not to say that late-type galaxies never host AGNs.  Indeed, $\\sim$15\\% of the Seyfert nuclei and $\\sim$10\\% of all AGNs in the Palomar survey are hosted in galaxies with Hubble types Sc or later.  In terms of AGN demographics, the present results imply that the existing statistics of AGNs in late-type galaxies based on optical searches (Ho et al. 1997b) are likely to be robust. There is not a large population of AGNs hidden among late-type galaxies that have \\hii\\ spectra."
    },
    "0208/astro-ph0208183_arXiv.txt": {
        "abstract": "{ We present a spectroscopic study of 41 hard X-ray sources detected serendipitously with high significance ($>$5$\\sigma$ in the 2-10 keV band) in seven {\\it EPIC} performance/verification phase observations. The large collecting area of {\\it EPIC} allows us to explore the spectral properties of these faint hard X-ray sources with 2$<$$F_{2-10}$$<$80 $\\times$ 10$^{-14}$ \\cgs~ even though the length of the exposures are modest ($\\sim$20 ks). Optical identifications are available for 21 sources of our sample. Using a simple power law plus Galactic absorption model we find an average value of the photon index $\\Gamma$$\\sim$1.6-1.7, broadly consistent with recent measurements made at similar fluxes with {\\it ASCA} and with {\\it Chandra} stacked spectral analyses. We find that 31 out of 41 sources are well fitted by this simple model and only eight sources require absorption in excess of the Galactic value. Interestingly enough, one third of these absorbed sources are broad line objects, though with moderate column densities. Two sources in the sample are X-ray bright optically quiet galaxies and show flat X-ray spectra. Comparing our observational results with those expected from standard synthesis models of the cosmic X-ray background (CXB) we find a fraction of unabsorbed to absorbed sources larger than predicted by theoretical models at our completeness limit of $F_{2-10}$$\\sim$5 $\\times$ 10$^{-14}$ \\cgs. The results presented here illustrate well how wide-angle surveys performed with {\\it EPIC} on board {\\it XMM-Newton} allow population studies of interesting and unusual sources  to be made as well as enabling constraints to be placed on some input parameters for synthesis models of the CXB. ",
        "introduction": "It is now widely accepted that the bulk of the Cosmic X-ray background (CXB) in the range 0.1-10 keV is the result of the integrated emission of unresolved sources over cosmic time (Giacconi et al. 2001). Recent extremely deep surveys by  {\\it Chandra} (Tozzi et al. 2001; Brandt et al. 2001a) and {\\it XMM-Newton} (Hasinger et al. 2001), have resolved up to $\\sim$75-90\\% of the CXB into discrete sources, down to a limiting 2-10 keV flux of $\\sim$10$^{-16}$ erg cm$^{-2}$ s$^{-1}$. On-going follow-up optical identification programs suggest that most of these sources are AGNs, while a sizeable fraction of the rest are optically faint ($I$$>$24) objects ($\\sim$30\\%), likely to be highly obscured AGNs (Alexander et al. 2001) or luminous bulge-dominated galaxies (Barger et al. 2001, Cowie et al. 2001). Large area surveys (e.g. from {\\it ASCA} and {\\it BeppoSAX}; Akiyama et al. 2000; Della Ceca et al. 1999; Giommi, Perri \\& Fiore 2000), limited to higher fluxes but providing a larger number of sources, confirm the above results. At a flux limit of $\\sim$5 $\\times$ 10$^{-14}$ erg cm$^{-2}$ s$^{-1}$ in the range 5-10 keV, the bulk of the sources are broad line AGNs, while most of the others are classified as Seyfert 1.8-2 galaxies and optically 'red' quasars (La Franca et al. 2000). Surprisingly, {\\it Chandra} serendipitously discovered that also some normal galaxies, without evidence for any activity (i.e. AGN and/or starburst) in the optical band, are X-ray loud (Fiore et al. 2000).  To date very few X-ray spectra of such faint sources are reported in the literature (but see pioneering results from Vignali et al. 2000; Sakano et al. 1998 and Crawford et al. 2001). Such spectral information is crucial in order to accurately determine source-by-source the amount of absorption, the spectral shape and the presence of reprocessed features (see e.g. Yaqoob 2000). These can also be used as input parameters for the synthesis models of the CXB in order to evaluate the contribution of the various AGN types and constrain their evolution (Comastri et al. 2001; Gilli, Salvati \\& Hasinger 2001). Important questions are still unanswered, such as the relationship between optical identification and X-ray characteristics as well as the role of absorption in the different types of active galaxies (Salvati \\& Maiolino 2000; Pappa et al. 2001a). There is growing evidence that not only narrow line but also  broad line AGNs suffer from intrinsic absorption (Risaliti et al. 2000), which is present both in radio-quiet  (Gallagher et al. 2000; Brandt al. 2001b; Maloney \\& Reynolds 2001; Collinge \\& Brandt 2000) and in radio-loud objects (Cappi et al. 1997; Yuan et al. 2000; Sambruna, Eracleous \\& Mushotzky 1999) at all redshifts. These obscured sources show hard spectral indices which match well with the CXB slope in the hard X-ray band ($\\sim$1.4, e.g. Marshall et al. 1980). Moreover doubts are emerging about the existence of the long sought after high luminosity type 2 AGNs (Halpern,Turner \\& George 1999), the so-called QSO 2s, which should play a significant (and perhaps a major) role in the production of the hard CXB (Fabian \\& Iwasawa 1999): from recent deep observations only two cases have been reported (Norman et al. 2001 and Stern et al. 2001). It would appear to be difficult to discover a large number of distant Compton-thick QSOs with present X-ray observatories (Fabian, Wilman \\& Crawford 2001).  Based on the above arguments, we have started an extensive program with {\\it XMM-Newton} aimed at studying the brightest (in the hard X-ray band) sources serendipitously detected in a number of fields with moderate exposure ($\\sim$20 ks). This work has been designed to complement at an intermediate flux level ($\\sim$10$^{-14}$ \\cgs) the X-ray population studies made by very deep pencil-beam observations. On account of its good positional accuracy ($\\sim$6 arcsec error radius) and its unprecedented sensitivity in the 2-10 keV band, the {\\it XMM-Newton} satellite is currently the most appropriate telescope with which to pursue such a study. \\begin{table*} \\caption{Journal of the {\\it XMM-Newton} observations.} \\label{tab1} \\begin{center} \\begin{tabular}{lcccccccc} \\hline \\multicolumn{1}{c} {Field} & \\multicolumn{1}{c} {R.A.} & \\multicolumn{1}{c} {Dec.}& \\multicolumn{1}{c} {Orbit}& \\multicolumn{1}{c} {Date} & \\multicolumn{3}{c} {Exposure (s)} & \\multicolumn{1}{c} {Filter} \\\\ & & & & &PN &M1 &M2&PN M1 M2\\\\ \\hline\\hline\\\\ PKS0312$-$770  &03 11 55.0 &$-$76 51 52&057 &2000-03-31 &26000 &25000&24000&Tc Tc Tc \\\\ MS1229.2$+$6430&12 31 32.0 &$+$64 14 21 &082 &2000-05-20 &22900&18600&22900 &Th Th Th\\\\ IRAS13349$+$2438&13 37 19.0 &$+$24 23 03&097&2000-06-20&$-$ &41300& 38600 &$-$ Me Th\\\\ Abell 2690&00 00 30.0 &$-$25 07 30&088 &2000-06-01 &21000 &16600 &15300 & Me Me Me\\\\ MS 0737.9$+$744&07 44 04.5 &$+$74 33 49&063 & 2000-04-12& 15000 &17800 &26100& Th Th Th\\\\ Markarian 205& 12 21 44.0 &$+$75 18 37&075&2000-05-07&17000 &$-$ &14800& Me $-$ Me\\\\ Abell 1835&14 01 02.0 &$+$02 52 41 &101 &2000-06-27&22900 &23700 &26400 &Th Th Th\\\\ \\hline \\end{tabular} \\end{center} Optical blocking filters used during observations: Th=thin, Me=medium and Tc=thick. \\end{table*} ",
        "conclusions": "We have reported spectral results for a sample of 41 hard X-ray sources detected serendipitously in seven {\\it EPIC} fields and selected in the 2-10 keV band. A detailed spectral analysis has been performed in order to measure source-by-source the 0.3-10 keV continuum shape, the amount of cold (and, possibly, ionized) absorbing matter and the strength of other spectral features. Complementary to deep pencil beam surveys, our shallower survey allows us to investigate in some detail the spectral properties of faint serendipitous sources. This is a field of study almost unexplored with previous X-ray satellites. We have found an average photon index  $\\langle\\Gamma\\rangle$=1.67$\\pm$0.04 using a simple power law fit with Galactic absorption for the whole sample. Considering only sources with $F_{2-10}\\geq$5 $\\times$ 10$^{-14}$ \\cgs~ (i.e. our completeness limit) we obtain a $\\langle\\Gamma\\rangle$=1.69$\\pm$0.04, which is consistent with average values reported in recent {\\it Chandra} and {\\it ASCA} works at similar fluxes.  We have also shown how surveys of the kind described here can constrain some of the assumptions used in CXB population synthesis models (either spectral shape, XLF and/or column density distribution). In particular, we found a mismatch between our observational results and those predicted by the CXB theoretical models relative to the fractions of absorbed versus unabsorbed sources above 5 $\\times$ 10$^{-14}$ \\cgs~ in the range 2-10 keV. Extremely deep pencil beam exposures do not stress this trend, very likely because of their bias towards fainter fluxes in the source selection. We have also been able to collect information about unusual objects such as broad line X-ray obscured AGNs and optically dull X-ray bright galaxies. We are currently analysing further {\\it XMM-Newton} observations with a goal of obtaining at least 100 source spectra  which will allow us to put our results on more sound statistical grounds."
    },
    "0208/astro-ph0208526_arXiv.txt": {
        "abstract": "Detection of Ultra High Energy Neutrinos (UHEN), with energy above 0.1 EeV~($10^{18}$ eV) is one of the most exciting challenges of high energy astrophysics and particle physics. In this article we show that the Auger Observatory, built to study ultra high energy cosmic rays, is one of the most sensitive neutrino telescopes that will be available during the next decade. Furthermore, we point out that the Waxman-Bahcall upper bound for high energy neutrino flux below 1 EeV turns into a lower bound above a few EeV. In this framework and given the experimental evidences for $\\nu_\\mu\\longrightarrow\\nu_\\tau$ with large mixing, we conclude that observation of Tau UHEN in the southern Auger observatory should most certainly occur within the next five years. \\vspace{1pc} ",
        "introduction": "The nature and origin of the observed Ultra High Energy Cosmic Rays (UHECR) have been for the past decades the subject of numerous debates and models (for a review see e.g. \\cite{SiglRev,BoratavRev,WatsonRev}). None of those models seemed to be able to fulfill simultaneously the source power requirement, its invisibility and the transport energy losses without requiring either new physics or large magnetic field together with a local over-abundance of transient (power sufficient?) sources. Moreover recent experimental results\\cite{HiresICRC,AgasaICRC} shown at the 27th ICRC conference\\cite{ICRC2001}, seemed somewhat contradictory in particular concerning the flux of events above the Greisen Zatsepin and Kuzmin cut-off\\cite{GZK}. Given the lack of statistics the interpretation of the same data has even led some authors to completely contradictory conclusions (see e.g. \\cite{WB0206217,BGG0204357}). What are the UHECR sources and their distribution ? What is their nature and energy spectra ? What is the flux above $10^{20}eV$ ? Those are the few fundamental questions that future experiments need to answer. \\par In the energy range [$10^{19}-10^{20}$] eV only stable hadrons (e.g. protons) or nuclei, among the known particles, can travel on distance much larger than a few tens of Mpc. However the maximum distance is still limited to about 50 Mpc (our neighborhood on cosmological scale) above $10^{20}$ eV because of photo-production or photo-dissociation processes against the Cosmic Microwave Background of radiation (CMB). If astrophysical objects such as AGN or GRB are to play a fundamental role in UHECR production then particle spectra from distant sources will be affected by these processes. Resulting fluxes should extinct exponentially above $10^{20}$ eV exhibiting the GZK cutoff. Taking into account additional distortions induced for example by extra galactic magnetic fields we conclude that hadron spectra in this energy range will mix the source characteristics together with the transport distortions. On the other hand neutrinos can travel on cosmological distances essentially unaffected and all astrophysical UHECR hadron sources as well as more exotic ones such as topological defect collapses or heavy relic decays are bound to produce Ultra High Energy Neutrino (UHEN). Their detection would be a very valuable clue to the characteristics of the UHECR sources. ",
        "conclusions": "The Auger observatory is a very sensitive neutrino telescope reaching for a null result a limit in $\\nu_\\tau$ flux as low as $5\\times10^{-9}$~GeV\\,cm$^{-2}$s$^{-1}$sr$^{-1}$ at 90\\% CL. Moreover, given the observation of UHECR in the range [10,100] EeV, given the lower limit on neutrino flux which can be derived from the WB bound and given the strong experimental evidence for \\mbox{$\\nu_\\mu\\longleftrightarrow\\nu_\\tau$} with large mixing the expectations for a positive result in the next five years are very high. Beside its excellent performances for the study of UHECR Auger may very well be the first experiment to observe  $\\nu_\\tau$ appearance from a $\\nu_\\mu$/$\\nu_e$ source."
    },
    "0208/astro-ph0208093_arXiv.txt": {
        "abstract": "\\baselineskip 11pt Reconstructed from lensing tomography, the evolution of the dark matter density field in the well-understood linear regime can provide model-independent constraints on the growth function of structure and the evolution of the dark energy density. We examine this potential in the context that high-redshift cosmology has in the future been fixed by CMB measurements.  We construct  sharp tests for the existence of multiple dark matter components or a dark energy component that is not a cosmological constant. These functional constraints can be transformed into physically motivated model parameters. From the growth function, the fraction of the dark matter in a smooth component,  such as a light neutrino, may be constrained  to a statistical precision of $\\sigma(f) \\approx 0.0006 \\sifsky$ by a survey covering a fraction of sky $\\fsky$ with redshift resolution $\\Delta z=0.1$.  For the dark energy, a parameterization in terms of the present energy density $\\Omega_{\\de}$, equation of state $w$ and its redshift derivative $w'$, the constraints correspond to $\\sigma(w)=0.016 \\sifsky$ and a mildly degenerate combination of the other two parameters.  For a fixed $\\Omega_{\\de}$, $\\sigma(w') = 0.046 \\sifsky$;  for $\\Omega_{\\de}$ marginalized $\\sigma(w')=0.069 \\sifsky$. ",
        "introduction": "The weak gravitational lensing of faint galaxies \\cite{weak} provides the most direct probe of mass distribution in the universe (e.g.~\\cite{BarSch01}).    Moreover the evolution of clustering in the mass distribution is arguably the best theoretically-grounded probe of the dark energy and dark matter \\cite{deprobes}. Although observations of weak lensing on large scales \\cite{weaklss} are still in the discovery phase \\cite{weakdet}, future wide-field surveys have the potential to rival the statistical precision and cosmological utility of luminosity distance measures from supernova surveys \\cite{Hut02,Hu01c}.  Even in the context of a precisely-determined homogeneous cosmology, lensing measurements are unique in that they probe the clustering properties of the dark matter and energy.  These are fixed by the homogeneous cosmology only under particular assumptions of the particle constituents (e.g. cold dark matter and scalar field dark energy) \\cite{CalDavSte98,Hu98}.  Much of the critical cosmological information lies in the temporal or radial direction. A potential obstacle for weak lensing  is that the observables are inherently two-dimensional.  All of the matter along the line-of-sight to a distant source contributes to the lensing. For a family of  cosmological models that is described by a handful of parameters, this is not a serious drawback.  Lack of radial information is largely compensated by a large angular dynamic range and external cosmological information.  Given the lack of compelling models for the dark energy and controversies surrounding the phenomenology of the dark matter on small scales, it is interesting to consider a more model-independent approach.   Indeed recent studies of alternate parameterizations of the dark energy have revealed potential ambiguities in the interpretation of luminosity distance measurements \\cite{WanGar01,MaoBruMcMSte02,WelAlb02,Teg01,HutSta02}. To address these issues with weak lensing,  recovery of the temporal dimension becomes critical.  With future surveys that possess source photometric redshift information, recovery of the lost information is possible in principle through tomography.   Photometric redshift techniques are already being applied and tested on current lensing data \\cite{Witetal01}. The full two-point statistical information can be regained by cross-correlating the lensing observables on all source redshift planes \\cite{Hu99}.   This method utilizes both the angular clustering and the temporal evolution of the density field but obscures the nature and  hence the model-dependence of the information. Additionally, the joint observables are survey dependent and computationally cumbersome to analyze.  In this paper, we instead isolate the temporal information by applying recently developed techniques to reconstruct the radial density field itself \\cite{Tay01,HuKee02}.   We will further focus solely on the linear regime where predictions are well-understood.  Even utilizing only this theoretically-clean subset of information in the data, future surveys can potentially provide interesting model-independent constraints on the properties of the dark energy and matter.  The outline of the paper is as follows.  In \\S \\ref{sec:tomography}, we discuss the method for reconstruction and statistical forecasts.  In \\S \\ref{sec:growth}, we study constraints on the growth function for a fixed homogeneous cosmology and in \\S \\ref{sec:density} the dark energy density evolution assuming pure cold dark matter. We discuss these results in \\S \\ref{sec:discussion}. ",
        "conclusions": "\\label{sec:discussion}  We have shown that lensing surveys covering more than a few percent of the sky with good photometric redshift information can probe the time evolution of the linear growth function and distance-redshift relation, both of which are sensitive to properties of the dark energy and dark matter. Specifically, we have tested a model-independent parameterization of the linear growth rate and/or dark energy density discretized into bins in redshift.    Deviations in the growth rate would indicate a component of the dark matter that is not effectively cold or dark energy that is not smooth on the lensing scale.  Deviations in the constancy of the dark energy density would rule out a cosmological constant model.  In this exploratory study, we have made several simplifying assumptions that would need to be addressed in a concrete implementation. Perhaps the primary one is that future CMB measurements will completely fix the high redshift cosmology.   The most uncertain piece involves the amplitude of the initial fluctuations on the scales relevant to the lensing pixels, $k \\sim 0.05$ Mpc$^{-1}$ for degree scales. Fortunately, the pivot point of CMB anisotropy experiments with several arcminute scale resolution is sufficiently close to the lensing scale that the slope and shape of the initial power spectrum do not cause much ambiguity \\cite{Hu01c}.  To fully utilize the lensing information, the initial amplitude must be fixed to an accuracy better than the amplitude of the growth function, which we have found to be  $\\sim 0.002 \\sifsky$, i.e. percent level accuracy for surveys of several thousand square degrees. For CMB anisotropies, this precision requires that the optical depth during reionization must be determined to $\\sigma(\\tau) \\sim 0.01$ to resolve the amplitude degeneracy. If determined from CMB observations alone, this will require polarization measurements with a precision comparable to the  Planck satellite \\cite{planck}, which can in principle achieve $\\sigma(\\ln \\delta_\\zeta)=0.0044$ at $k=0.05$ Mpc$^{-1}$ \\cite{Hu01c}.  Direct measurements of the reionization epoch can also resolve the ambiguity \\cite{Hu01c}. Even in the absence of this information, the {\\it evolution} of the growth and luminosity-distance relation are still constrained.   These issues are best addressed through joint parameter estimation.  On the lensing side, the most important assumption is that the noise in the convergence measurements is well-calibrated and not significantly larger than the projections based on intrinsic ellipticities.  Even aside from the demanding requirements for control of systematic errors, there may be intrinsic correlations in the ellipticities \\cite{intrinsic} that need to be modeled or avoided by increasing the pixel scale and redshift bin widths.   The recovered information is largely insensitive to the redshift bin width since the high signal-to-noise modes are all low frequency. Errors scale roughly as $A_{\\rm pix}^{1/2}$ due to the loss of independent modes in a fixed survey area.  We have also neglected sample covariance between the pixels but note that we have correspondingly neglected the information contained in such correlations.  Indeed, we have completely neglected the information contained in the non-linear regime which in fact contains the majority of the information from lensing tomography \\cite{HuKee02}.   Clearly, future studies will be required to see how best to mine the model-independent information contained in lensing tomography.   {\\it Acknowledgments:} I thank D. Huterer and C.R. Keeton for useful conversations and E. Linder for pointing out a typo in the draft.  This version includes errata from a coding bug, affecting constraints involving the present dark energy density, pointed out by K. Abazajian and S. Dodelson. WH is supported by NASA NAG5-10840 and the DOE OJI program.   \\vfill"
    },
    "0208/astro-ph0208216_arXiv.txt": {
        "abstract": "We use Owens Valley Radio Observatory (OVRO) cosmic microwave background (CMB) anisotropy data to constrain cosmological parameters. We account for the OVRO beamwidth and calibration uncertainties, as well as the uncertainty induced by the removal of non-CMB foreground contamination. We consider open and spatially-flat-$\\Lambda$ cold dark matter cosmogonies, with nonrelativistic-mass density parameter $\\Omega_0$ in the range 0.1--1, baryonic-mass density parameter $\\Omega_B$ in the range (0.005--0.029)$h^{-2}$, and age of the universe $t_0$ in the range (10--20) Gyr. Marginalizing over all parameters but $\\Omega_0$, the OVRO data favors an open (spatially-flat-$\\Lambda$) model with $\\Omega_0\\simeq$ 0.33 (0.1). At the 2 $\\sigma$ confidence level model normalizations deduced from the OVRO data are mostly consistent with those deduced from the DMR, UCSB South Pole 1994, Python I-III, ARGO, MAX 4 and 5, White Dish, and SuZIE data sets. ",
        "introduction": "Cosmic microwave background (CMB) anisotropy measurements have begun to provide interesting constraints on cosmological parameters.\\footnote{ See, e.g., Miller et al. (2002a), Coble et al. (2001), Scott et al. (2002), and Mason et al. (2002) for recent measurements, and, e.g., Podariu et al. (2001), Wang, Tegmark, \\& Zaldarriaga (2002), Durrer, Novosyadlyj, \\& Apunevych (2001), and Miller et al. (2002b) for recent discussions of constraints on cosmological parameters.} Ganga et al. (1997a, hereafter GRGS) developed a technique to account for uncertainties, such as those in the beamwidth and the calibration, in likelihood analyses of CMB anisotropy data. This technique has been used with theoretically-predicted CMB anisotropy spectra in analyses of the Gundersen et al. (1995) UCSB South Pole 1994 data, the Church et al. (1997) SuZIE data, the Lim et al. (1996) MAX 4+5 data, the Tucker et al. (1993) White Dish data, the de Bernardis et al. (1994) ARGO data, and the Platt et al. (1997) Python I-III data (GRGS; Ganga et al. 1997b, 1998; Ratra et al. 1998, 1999a, hereafter R99a; Rocha et al. 1999, hereafter R99). A combined analysis of all these data sets, excluding the Python data, is presented in Ratra et al. (1999b, hereafter R99b).  In this paper we present a similar analysis of CMB anisotropy data from the OVRO observations (Leitch et al. 2000, hereafter L00). The OVRO detectors and telescopes are described in Leitch (1998) and L00; here we review information about the experiment that is needed for our analysis.  OVRO data were taken in two frequency bands, one centered at 14.5 GHz (Ku band), the other at 31.7 GHz (Ka band). Thirty-six fields, along an approximate circle at declination $\\delta \\simeq 88^\\circ$ centered on the North Celestial Pole (NCP) were observed. In our computations we use the coordinates for the 36 fields given in Table 2 of L00. The OVRO measurements were made by switching the beam in a two-point pattern along the circle, resulting in a three-beam response to the sky signal. The beamthrow is $22'{\\!}.16$. The zero-lag window function parameters for the OVRO experiment are given in Table 1. This and other window functions are shown in Fig. 18 of L00.  L00 use multiepoch VLA observations to detect and remove non-CMB discrete source contamination from the OVRO data. We have also analyzed the OVRO data ignoring 3 of the 36 fields that were affected by the strongest variable discrete source; cosmological constraints derived from this restricted OVRO CMB anisotropy data set are very consistent with those derived from the full OVRO CMB anisotropy data set, so we do not discuss this restricted OVRO data set analysis further.  Since OVRO data were taken at two frequencies, it is possible to fit the data to both a non-CMB foreground component (parametrized by the frequency dependent temperature anisotropy $\\Delta T_{\\rm fore} \\propto \\nu^\\beta$) and a CMB anisotropy component with spectral index $\\beta = 0$.\\footnote{ See L00 and Mukherjee et al. (2002) for discussions of foreground contaminants in the OVRO microwave data.} We use the method in $\\S$ 11 of L00 to extract the CMB anisotropy component in the OVRO data, marginalizing over a foreground spectral index in the range $-3 < \\beta < 2$ in our likelihood analysis.\\footnote{ Although the data themselves are unable to rule out more negative values of beta (L00, Fig. 14), Leitch et al. (1997) use low frequency maps of the NCP region to rule out such values.}${^,}$\\footnote{ Following Mukherjee et al. (2002) we have also analyzed the 31.7 GHz OVRO CMB anisotropy data while marginalizing over possible 100 $\\mu$m and 12 $\\mu$m foreground contaminant template (Schlegel, Finkbeiner, \\& Davis 1998) correlated components. The cosmological constraints from these analyzes are quite consistent with results presented here. This is because although the foreground signal inferred in our analysis is not entirely fit by the dust data, they are significantly correlated, and the 31.7 GHz data, modelled either way, is almost entirely CMB anisotropy. The OVRO data at its two frequencies are shown in Fig. 13 of L00, the deduced CMB anisotropy and foreground signals are shown in Fig. 16 of L00, and the dust-correlated emission is shown in Fig. 1 of Mukherjee et al. (2002).}  CMB anisotropy constraints are derived from the foreground-corrected 31.7 GHz data.\\footnote{ At $\\beta = -2.2$ for the foreground contaminant, $96\\%$ of the 31.7 GHz data is CMB anisotropy.} The 31.7 GHz beam profile is well approximated by a circular Gaussian of FWHM $7'{\\!}.37 \\pm 0'{\\!}.26$ (one standard deviation uncertainty). We use the method of GRGS to account for the OVRO beam uncertainty.  As discussed in L00, the noise in the 31.7 GHz data indicates the presence of a component that is correlated between neighboring fields (this component is small compared to the uncorrelated noise in a single scan of data). As a result the 31.7 GHz OVRO data show only one-half of the anticorrelation for nearest neighbor fields expected for a triple beam chopped experiment.\\footnote{ Models that neglect these correlations are grossly discrepant with the data, while when these correlations are accounted for the model fits are reasonable and consistent with the data. This can be seen from Fig. 19 of L00 and we find the same.} This one-offdiagonal correlated noise is included and its amplitude marginalized over in our analysis.  A constant offset is removed from the OVRO data; we marginalize over the amplitude of the offset to account for this in our likelihood analysis. The 1 $\\sigma$ absolute calibration uncertainty of the OVRO data is $4.3\\%$, and the method developed by GRGS is used to account for it.  In $\\S$ 2 we summarize the computational techniques used in our analysis. See GRGS and R99a for detailed discussions. Results are presented and discussed in $\\S$ 3. We conclude in $\\S$4. ",
        "conclusions": "The OVRO data results derived here are mostly consistent with those derived from the DMR, SP94, Python I-III, ARGO, MAX 4+5, White Dish and SuZIE data. The OVRO data significantly constrains $Q_{\\rm rms-PS}$ (for the flat bandpower spectrum $Q_{\\rm rms-PS}\\ = \\ 38{+6 \\atop -5}\\ \\mu$K at 1 $\\sigma$) and weakly favors low-density, low $\\Omega_B h^2$, young models.  \\bigskip  We acknowledge valuable assistance from R. Stompor and helpful discussions with K. Ganga and E. Leitch. PM, BR, and TS acknowledge support from NSF CAREER grant AST-9875031. NS acknowledges support from the Alexander von Humboldt Foundation and Japanese Grant-in-Aid for Science Research Fund No. 14540290.  \\clearpage  \\begin{table} \\begin{center} \\caption{Numerical Values for the Zero-Lag Window Function Parameters\\tablenotemark{a}} \\vspace{0.3truecm} \\tablenotetext{{\\rm a}}{The value of $l$ where $W_l$ is largest, $l_{\\rm m}$, the two values of $l$ where $W_{l_{e^{-0.5}}} = e^{-0.5} W_{l_{\\rm m}}$, $l_{e^{-0.5}}$, the effective multipole, $l_{\\rm e} = I(lW_l)/I(W_l)$, and $I(W_l) = \\sum^\\infty_{l=2}(l+0.5)W_l/\\{l(l+1)\\}$.} \\begin{tabular}{ccccc} \\tableline\\tableline $l_{e^{-0.5}}$ & $l_{\\rm e}$ & $l_{\\rm m}$ & $l_{e^{-0.5}}$ & $\\sqrt{I(W_l)}$  \\\\ \\tableline 360 &  596 &  537 &  753 & 1.41  \\\\ \\tableline \\end{tabular} \\end{center} \\end{table}   \\begin{table} \\begin{center} \\caption{Numerical Values for $Q_{\\rm rms-PS}$ and $\\delta T_l$ from Likelihood Analyses Assuming a Flat Bandpower Spectrum} \\vspace{0.3truecm} \\tablenotetext{{\\rm a}}{The first of the three entries is where the posterior probability density distribution function peaks and the vertical pair of numbers are the $\\pm 1$ $\\sigma$ (68.3\\% highest posterior density) values.} \\tablenotetext{{\\rm b}}{Average absolute error on $Q_{\\rm rms-PS}$ in $\\mu$K.} \\tablenotetext{{\\rm c}}{Average fractional error, as a fraction of the central value.} \\tablenotetext{{\\rm d}}{Likelihood ratio.} \\begin{tabular}{ccccc} \\tableline\\tableline $Q_{\\rm rms-PS}$\\tablenotemark{a} & Ave. Abs. Err.\\tablenotemark{b} & Ave. Frac. Err.\\tablenotemark{c} & $\\delta T_l$\\tablenotemark{a} & LR\\tablenotemark{d} \\\\ ($\\mu$K) & ($\\mu$K) & {\\ } & ($\\mu$K) & {\\ } \\\\ \\tableline \\medskip 38 ${44 \\atop 33}$ & 5.5 & 14\\% & 59 ${69 \\atop 51}$ & $9 \\times 10^{66}$ \\\\ \\tableline \\end{tabular} \\end{center} \\end{table}  \\clearpage"
    },
    "0208/astro-ph0208020_arXiv.txt": {
        "abstract": "We report the abundances of neutron-capture elements in eight carbon-rich, metal-poor ($-2.7\\leq$[Fe/H]$\\leq -1.9$) stars observed with the Subaru Telescope High Dispersion Spectrograph. The derived abundance patterns indicate that the neutron-capture elements in these objects primarily originated from {\\it s}-process nucleosynthesis, although the [Ba/Eu] abundance ratios in some objects are lower than that of the solar-system {\\it s}-process component. The present analysis has yielded the Pb abundances for seven objects, as well as an upper limit for one object, from use of the \\ion{Pb}{1} 4057~{\\AA} and 3683~{\\AA} lines.  The values of [Pb/Ba] in these objects cover a wide range, between $-0.3$ and +1.2. Theoretical studies of {\\it s}-process nucleosynthesis at low metallicity are required to explain this large dispersion of the [Pb/Ba] values. Variations in radial velocity have been found for two of the eight objects, suggesting that, at least in these instances, the observed excess of {\\it s}-process elements is due to the transfer of material across a binary system including an AGB star.  Comparisons with predictions of AGB nucleosynthesis models are discussed. ",
        "introduction": "\\label{sec:intro}  Lead (Pb) isotopes form, along with those of bismuth, the group of the heaviest stable nuclei.  Understanding of the physical processes responsible for the synthesis of these nuclei, as well as their enrichment history in the Galaxy, is important in the study of stellar nucleosynthesis and Galactic chemical evolution, as well as for precision cosmochronology, since these nuclei are the expected decay products of Th and U \\citep{goriely01, schatz02}. The majority of Pb nuclei in the solar system are believed to have originated in the {\\it s}-process.  However, the so-called main-component of the {\\it s}-process in classical models cannot explain the solar system abundances of Pb, hence another component (the so-called strong component) was introduced \\citep[e.g., ][]{kappeler89}. Recent theoretical work on nucleosynthesis in asymptotic giant branch (AGB) stars, based on the idea of an {\\it s}-process that occurs in the radiative zone during the inter-pulse phase, have predicted large production of Pb and Bi ($Z=82$ and 83, respectively) compared with lighter {\\it s}-process elements such as Sr ($Z=38$) and Ba ($Z=56$) in low metallicity AGB stars \\citep{gallino98, goriely00, busso01}. These works suggested that the Pb production in low-metallicity AGB stars corresponds to the strong component assumed in the classical models. Studies of Pb enrichment in the Galaxy have also started using the yields predicted by these AGB models \\citep{travaglio01}.  Recent abundance studies of ${\\it s}$-process-element-enhanced, metal-poor stars, based on high-resolution spectroscopy, have measured the abundances of Pb and other neutron-capture elements in a number of stars, and made it possible to examine the models of AGB nucleosynthesis in greater detail.  For instance, \\citet{aoki00} determined the abundances of 16 neutron-capture elements, including Pb, for the ${\\it s}$-process-element-rich, very metal-poor ([Fe/H]=$-$2.7) star LP~625--44\\footnote{[A/B] = $\\log(N_{\\rm A}/N_{\\rm B})- \\log(N_{\\rm A}/N_{\\rm B})_{\\odot}$, and $\\log \\epsilon_{\\rm A} = \\log(N_{\\rm A}/N_{\\rm H})+12$ for elements A and B.}. The Pb abundance of this object is, however, much {\\it lower} than the prediction by the ``standard'' model of \\citet{gallino98}, when the abundance pattern is normalized by the abundance of lighter {\\it s}-process nuclei (e.g., Ba). A similar result was derived for another metal-poor star, LP~706--7, by \\citet{aoki01}. Their abundances could, however, be reconciled with the models of Gallino et al. if the extent of the $^{13}$C pocket is reduced by a factor of $\\simeq$24 \\citep{ryan01}.  On the other hand, \\citet{vaneck01} studied another three {\\it s}-process-element-enhanced, metal-poor objects, and found very {\\it large} Pb enhancements in these objects, compared to those of the lighter elements.  They concluded that the results can be explained well by the model of AGB nucleosynthesis presented by \\citet{goriely00}.  Thus, the abundance ratios of the Pb, and the neutron-capture elements at the second peak of the {\\it s}-process ($Z\\sim 56$), produced by metal-deficient AGB stars seem to show a large dispersion, or might even be classified into two distinct groups. In order to understand the nature of the {\\it s}-process nucleosynthesis at low metallicity, we have embarked on an extensive set of studies of the neutron-capture elements for metal-poor stars with excesses of {\\it s}-process elements. In this Paper we report the abundances of neutron-capture elements, including Pb, for eight metal-poor stars. ",
        "conclusions": "\\label{sec:disc}  The resulting abundances for two of the stars in our sample are shown in Figure~\\ref{fig:abund}, along with the scaled abundance patterns of solar system material, the main {\\it s}-process component, and the (inferred) {\\it r}-process pattern \\citep{arlandini99}. These comparisons are useful to distinguish the origins of the neutron-capture elements detected in our objects.  The abundance patterns of elements with $56\\leq Z\\leq 63$ in our stars agree with the {\\it s}-process pattern much better than with the {\\it r}-process pattern. This result implies that, as expected, the neutron-capture elements in our stars principally originate in the {\\it s}-process. The values of [Ba/Eu] for four of the stars in our sample (CS~29526--110, CS~22898--027, CS~31062--012, and CS~31062--050) are significantly lower ([Ba/Eu]=0.36$\\sim$0.47) than seen in the main {\\it s}-process component \\citep[{[Ba/Eu]=+1.15};][]{arlandini99}.  We suggest, however, that the derived abundance ratios of [Ba/Eu] have been produced by an {\\it s}-process which produces {\\it different} abundance ratios from that of the main {\\it s}-process component. Indeed, \\citet{goriely00} predicted lower values of [Ba/Eu] ($\\sim 0.4$) for yields of metal-deficient AGB stars, though the [Ba/Eu] value of our objects is not correlated with metallicity. To explain these low values of [Ba/Eu] by the mixture of the abundance ratios of the {\\it r}- and {\\it s}-process components in the solar system, we must assume that about 80-90\\% of Eu nuclei were produced by the {\\it r}-process. If this were true, it follows that these four stars show very large excesses of their {\\it r}-process elements ([Eu/Fe]=1.5$\\sim$1.8), similar to CS~31082--001, an extreme {\\it r}-process-element-enhanced star (Cayrel et al. 2001; Hill et al. 2002).  However, recent high-resolution studies of metal-poor giants indicate that stars with such large excesses of {\\it r}-process elements are quite rare.  An ongoing survey by Christlieb et al. (2002, in preparation) confirms the original suggestion by Beers (private communication) that roughly 3\\% of giants with [Fe/H] $< -2.5$ exhibit [{\\it r}/Fe]$ \\ge +1.0$. Hence, it would be difficult to appeal to this explanation to account for the excesses of Eu in the four stars in our sample. We treat our stars as having the abundance patterns produced by the {\\it s}-process for neutron-capture elements in this paper, because the elements discussed below (Ba and Pb) should principally originate from the {\\it s}-process, even though {\\it r}-process contamination may have contributed to a portion of the observed Eu excesses.  Figure~\\ref{fig:bapb} shows the ratio [Pb/Ba], representing the ratio of abundances between elements located at the second and third {\\it s}-process peaks, as a function of metallicity ([Fe/H]).  There exists a large scatter in [Pb/Ba] for these stars, larger than can be accounted for by the errors in the abundance analysis.  It is not apparent that our stars can be readily classified into separate groups on the basis of the present data. There may be evidence for an decreasing trend of [Pb/Ba] with decreasing [Fe/H] (the correlation estimated from objects in Figure 3, excluding CS~22942--019, is represented as [Pb/Ba]$=2.62+0.88$[Fe/H] with the standard error of the slope of 0.47, i.e., the possible slope is significant at the 1.9 sigma level). Additional data will presumably help to clarify these questions. Additional studies of {\\it s}-process nucleosynthesis at low metallicity should be carried out to explore the possible reasons for this large dispersion, and the possible dependence on metallicity, of the Pb/Ba ratios.  Comparison of these results with the predictions of AGB nucleosynthesis models of \\citet{gallino98} and \\citet{busso99} provides a constraint on their $^{13}$C pocket models\\footnote{ Uncertainties in the amount of $^{13}$C in the pocket reflects the unknown amount of mixing of protons from the H-rich envelope down into the C-rich zone.  Recent model calculations by Cristallo et al. (2001; see also Goriely \\& Mowlavi 2000) treat the proton mixing as an input parameter, rather than the amount of $^{13}$C in the ``pocket''.  The efficiency of the $^{13}$C as a neutron source for the {\\it s}-process is also affected by the profile of neutron poisons like $^{14}$N. }. Figure 1 of \\citet{ryan01} shows the metallicity dependence of [Pb/Fe] and [Ba/Fe] predicted by the model for 1.5~$M_{\\odot}$ AGB stars, compared with the abundance ratios observed in LP~625--44.  The value of [Pb/Ba] is sensitive to the adopted $^{13}$C profile, represented by the normalization factor of the $^{13}$C pocket in the standard model of \\citet{busso99}.  The [Pb/Ba] values found in our stars with $-2.7 \\leq$[Fe/H]$\\leq -1.9$ distribute from $-0.3$ to +1.2 (Figure~\\ref{fig:bapb}). This range of the abundance ratios is not explained by the standard model, which predicts [Pb/Ba]$\\sim +2$, but can be explained by models where the amount of $^{13}$C produced as the neutron source for the $s$-process is lowered by factors of 6-24 relative to the standard model.  We would also like to mention the model of the {\\it s}-process in AGB stars recently proposed by \\citet{iwamoto02}. Their model for AGB stars with [Fe/H] = $-2.7$ and $M=2M_{\\odot}$ shows a possible production of $^{13}$C, a candidate for the neutron source of the {\\it s}-process, as the result of proton mixing into the hot He shell at the second thermal pulse \\citep[see also ][]{fujimoto00}. Since the value of neutron exposure predicted by their model is rather high ($\\tau \\sim 1~$mb$^{-1}$), one may expect high Pb abundances relative to those of lighter neutron-capture elements. However, the small number of the exposures (only one in their model) is preferable to account for the low Pb abundances that we find in some stars in our sample \\citep{aoki01}. This mechanism for producing the required free neutrons is expected to occur only in very metal-poor ([Fe/H]$\\lesssim -2.5$) AGB stars, and may explain the abundances of some objects with lower metallicity and high [Pb/Ba] values.  The comparison with this model is, however, still quite speculative, and further theoretical study for the possibility of flash-driven proton mixing and its contribution to the {\\it s}-process is required.  Finally, we point out that clear temporal variations of radial velocity, which directly implies the binarity of the object, have thus far been detected for only three of the carbon-enhanced, metal-poor stars we have been studying in recent years:  CS~22942--019 \\citep{preston01}, CS~29526--110 (section \\ref{sec:obs}), and LP~625--44 \\citep{aoki00}. \\citet{preston01} reported that {\\it none} of the three carbon-rich, metal-poor subgiants in their sample, including the stars CS~22898--027 and CS~22880--074 studied here, exhibited radial velocity variations exceeding 0.5~kms$^{-1}$ over an 8 year period.  Furthermore, no radial velocity variation has been found for the subgiant (or dwarf) LP~706--7 \\citep{norris97}. These results suggest that the enhancement of neutron-capture elements in these subgiants may not be explained as a result of the transfer of material rich in {\\it s}-process elements across a binary system including an AGB star. For comparisons with the predictions of AGB models, confirmation of binarity based on the radial velocity monitoring is strongly desired.  It is interesting to note the large dispersion of [Pb/Ba] values of the three objects (CS~22942--019, CS~29526--110, and LP~625--44) for which the variation of radial velocity have been observed.  This suggests that the Pb/Ba ratios produced by {\\it s}-process nucleosynthesis in metal-deficient AGB stars show a large scatter, although the sample is still too small to be confident in this regard.  The present work clearly shows a variety in the abundance ratios of neutron-capture elements for carbon-rich, metal-poor stars, and underscores the importance of further studies for individual objects, including long-term radial velocity monitoring, for development of a better understanding of heavy element production by the {\\it s}-process operating at low metallicity."
    },
    "0208/astro-ph0208350_arXiv.txt": {
        "abstract": "We quantify the angular clustering of radio galaxies in the NVSS and FIRST surveys using the two-point correlation function and the moments of counts-in-cells -- both important points of comparison with theory. These investigations consistently demonstrate that the slope of the correlation function for radio galaxies agrees with that for optically-selected galaxies, $\\gamma \\approx 1.8$.  We describe how to disentangle the imprint of galaxy clustering from the two observational problems: resolution of radio galaxies into multiple components and gradients in source surface density induced by difficulties in processing ``snapshot'' radio observations (significant in both surveys below $\\fl \\sim 15$ mJy).  This study disagrees in some respects with previous analyses of the angular clustering of radio galaxies. ",
        "introduction": "\\renewcommand{\\thefootnote}{\\fnsymbol{footnote}} \\setcounter{footnote}{1} \\footnotetext{E-mail: cab@astro.ox.ac.uk}  Describing the large-scale structure of the Universe is of fundamental importance for testing theories of galaxy and structure formation and for measuring the cosmological parameters.  The largest structures require delineation by the deepest, widest surveys, currently represented by surveys for radio AGN.  These contain objects to redshifts of at least $z \\sim 4$: the radio emission marking these objects is not affected by dust obscuration, large-scale calibration effects should be minimal, and the number of objects in the current generation of radio surveys such as WENSS, FIRST and NVSS reach $\\sim 10^6$ over substantial fractions of the sky.  We see the distribution of galaxies projected on the sky, but this is still useful to quantify: it is easy to assemble a large sample of objects and the angular clustering can be de-projected (in a global statistical manner) to measure the spatial clustering, conclusions being reached in the absence of complete redshift information.  There are many sophisticated methods for quantifying the angular distribution of galaxies.  These include spherical harmonic analysis (e.g. Baleisis et al. 1998), percolation analysis (Bhavsar \\& Barrow 1983) and minimal spanning trees (Krzewina \\& Saslaw 1996).  Chiang \\& Coles (2000) emphasize the importance of maintaining the phase information of the clustering for describing morphology.  In contrast, here we use two of the crudest statistics for describing angular structure: the two-point angular correlation function and the moments of counts-in-cells.  It is well-known that these methods lose much of the clustering information: two very different distributions can have the same two-point correlation function.  However, these statistics are simple to interpret and hence reveal the fundamental observational problems and survey limitations.  They provide simple points of contact with prediction, have well-understood statistical errors and together provide a consistency check.  They must be understood and must give consistent results before application of more powerful techniques can be considered.  Correlation function analyses (Peebles 1980), widely used since the early days of clustering investigations, have been extensively applied in the optical regime, for example to the APM survey (Maddox et al. 1996).  Here, the correlation function typically shows a power-law behaviour $w(\\theta) \\propto \\theta^{1-\\gamma}$ with $\\gamma \\approx 1.8$ on small scales, with a steepening break to larger scales.  The key difference between angular correlation function analyses in the optical and radio regimes is in the latter, the wide redshift range of radio sources washes out much of the clustering amplitude through the superposition of unrelated redshift slices.  Hence an angular clustering signal has only been measurable in the most recent radio surveys, initially with marginal detections in the Green Bank 87GB survey (Kooiman et al. 1995) and the Parkes-MIT-NRAO (PMN) survey (Loan et al. 1997).  The latest generation of deep radio surveys -- FIRST (Becker et al. 1995), WENSS (Rengelink et al. 1998) and NVSS (Condon et al. 1998) -- reveal the imprint of structure more clearly.  The correlation function has been measured for WENSS by Rengelink et al. (1998), and for FIRST by Cress et al. (1996) and, in a pioneering and innovative series of papers, by Magliocchetti et al. (1998).  These studies concluded that the slope of the correlation function for radio galaxies was steep ($\\gamma > 2$).  Cress \\& Kamionkowski (1998) and Magliocchetti et al. (1999) modelled the 3D clustering from these analyses, including the behaviour of bias with epoch.  Magliocchetti et al. (1998) also carried out a counts-in-cells analysis of the FIRST survey, detecting significant skewness.  These results motivated our present investigation.  The NVSS had never been investigated for large-scale structure effects, and we wished to determine if the more extensive sky coverage and source list of $\\sim 2 \\times 10^6$ objects led to conclusions compatible with FIRST, with higher signal-to-noise ratio affording further insight.  We wished to understand how robust the conclusions from FIRST were, given the issue of over-resolution and the consequent need to ``combine'' catalogue sources from multiple-component radio galaxies.  We wished to examine the compatibility of results from counts-in-cells and correlation-function analyses.  With these aims in mind, NVSS and FIRST, at the same frequency but at resolutions differing by a factor of 9, suggest an ideal comparative study.  Our initial results (measurement of the NVSS angular correlation function) were presented in Blake \\& Wall (2002).  To proceed we first describe the two surveys, NVSS and FIRST.  Section \\ref{secquan} summarizes the clustering statistics we use and Section \\ref{secobseff} discusses the observational issues bound to impact upon large-scale structure analyses.  Sections \\ref{seccorr} and \\ref{seccell} derive angular correlation functions and counts-in-cells for each of NVSS and FIRST, and Section \\ref{secconc} compiles the conclusions. ",
        "conclusions": "\\label{secconc}  We have quantified the angular clustering in the NVSS and FIRST radio surveys using two independent methods: the two-point angular correlation function and the variance on counts-in-cells.  Our results may be summarized as follows:  \\begin{enumerate} \\item The results of angular correlation function and counts-in-cells analyses of the surveys are entirely consistent. \\item The larger area and greater number of sources in the NVSS yield a much clearer description of the clustering imprint.  The correlation function has two contributions: that due to multiple components of the same galaxy, dominant at $\\theta < 0.1^\\circ$, and that due to clustering between galaxies, which dominates at larger angles.  A clear break in $w(\\theta)$ is evident between these scales.  Both of these contributions are needed to explain the observed variance on counts-in-cells. \\item The clustering part of the correlation function has a slope consistent with that measured in the optical regime, $w(\\theta) \\propto \\theta^{-0.8}$; this is confirmed by our counts-in-cells measurements. \\item Both the NVSS and FIRST surveys suffer from systematic fluctuations in source surface density at flux-density thresholds at which they purport to be complete. \\end{enumerate}  Our work disagrees with some previous conclusions drawn from the FIRST survey:  \\begin{enumerate} \\item We find a galaxy correlation slope consistent with that measured in the optical, $\\gamma \\approx 1.8$, in contrast to somewhat steeper slopes reported in previous analyses.  These steeper slopes may have been produced by residual multiple component radio sources. \\item The skewness reported by Magliocchetti et al. (1998) may also be due to these residual double sources. \\end{enumerate}  This investigation has improved our understanding of the methodology of angular clustering analyses for large-scale radio surveys, of relevant observational effects present in such surveys, and of the derived structural parameters.  There is now the basis to use these surveys to derive three-dimensional information on the very largest structural scales, adopting more powerful statistical methods in conjunction with the redshift databases to be provided by surveys such as 2dF and SDSS."
    },
    "0208/astro-ph0208399_arXiv.txt": {
        "abstract": "We investigate the possibility of determining whether microlensing objects towards the Large Magellanic Cloud (LMC) are in a Galactic thick disc, or are in a Galactic halo, by using parallax measurements with an Earth-radius scale baseline. Our method makes use of EAGLE (Extremely Amplified Gravitational LEnsing) events which are microlensing events with an invisible faint source.  We show that the rate of EAGLE events is as high as that of normal microlensing events, even if they are caused by dark stars in the Galactic thick disc.  We explore the possibility of measuring the parallax effect in EAGLE events towards the LMC by using the {\\it Hubble Space Telescope} (HST) or the {\\it Very Large Telescope} (VLT). We find that EAGLE events enlarge the opportunity of parallax measurements by $4 \\sim 10$ times relative to that in normal microlensing events. We show that the parallax effect can be measured in $\\sim75\\%$ (from the HST) and $\\sim 60\\%$ (from the VLT) of all EAGLE events if most lenses are stars in the Galactic thick or thin disc, while $\\sim 20\\%$ (from the HST) and $\\sim 10\\%$ (from the VLT) can be measured if most lenses are halo MACHOs. In combination with the finite source size effect observations, we can strongly constrain the location of lenses. ",
        "introduction": "Several groups have carried out gravitational microlensing observations towards the Large Magellanic Cloud (LMC) in order to search for MAssive Compact Halo Objects (MACHOs) in the Galactic halo.  Until now, 13-17 candidates have been found towards the LMC and the microlensing optical depth $\\tau$ from the events is $1.2^{+0.4}_{-0.3}\\times10^{-7}$ (\\citealt{alc00b}). The estimated typical lens mass depends on the adopted Galactic kinematic model ranging over $0.01-1\\,M_{\\odot}$ (\\citealt{alc00b}; \\citealt{hon98}).   We have learned that there are lens objects along the line of sight towards the LMC. However, the issues of where lens objects are and what they are, are still unclear. This is because a degeneracy occurs in ordinary microlensing events for which the amplification is described by (\\citealt{pac86}) \\begin{equation} \\label{eq:amp-u} A(u)= \\frac{u^2+2}{u\\sqrt{u^2+4}}\\sim\\frac{1}{u} \\mbox{ for $u\\ll1$,} \\end{equation} where $u$ is the projected separation of the source and lens in units of the the Einstein radius $R_{\\rm E}$, which is given by  \\begin{equation} R_{\\rm E}(M,x) = \\sqrt{\\frac{4GM}{c^2}D_{\\rm s}x(1-x)}. \\label{eq:re} \\end{equation} Here $M$ is the lens mass, $x=D_{\\rm d}/D_{\\rm s}$ is the normalized lens distance and $D_{\\rm d}$ and $D_{\\rm s}$ are the observer-lens and the observer-source star distances. $D_{\\rm s}$ is hereafter assumed to be $50$ kpc. The time variation of the parameter $u=u(t)$ is \\begin{equation} \\label{eq:u} u(t)=\\sqrt{\\beta^{2} + \\left( \\frac{t-t_{0}}{t_{\\rm E}} \\right)^2}, \\end{equation} where $\\beta$, $t_{0}$, $t_{\\rm E}= R_{\\rm E}(M,x)/v_{\\rm t}$ and $v_{\\rm t}$ are the minimum impact parameter in units of $R_{\\rm E}$, the time of maximum magnification, the event time-scale and the transverse velocity of the lens relative to the line of sight towards the source star, respectively. From a light curve, one can determine the value of $\\beta$, $t_{0}$ and $t_{\\rm E}$, where $M$, $x$ and $v_{\\rm t}$ are degenerate in $t_{\\rm E}$. This three-fold degeneracy is the essential difficulty in determining the nature of the lens objects.  There are only marginally possible candidates, viz. old white dwarfs (\\citealt{alc97,alc00b}; \\citealt{han98}), old brown dwarfs (\\citealt{hon98}) and primordial black holes (\\citealt{iok00}).   Possibilities for non-halo lensing objects have also been discussed, for example, dark objects in a dark heavy component of the LMC itself (\\citealt{aub99}; \\citealt{gyu00}; \\citealt{alc01}). There is also the possibility that the lenses are dark objects in the Galactic thick disc.  While the microlensing optical depth $\\tau$ by known populations of stars in the Galactic thin disc and thick disc is of order $10^{-9}$ (\\citealt{alc00b}), a maximal heavy thick disc, which may be surrounded by an extended dark halo composed of particles, is also a possible Galactic component as the reservoir of lenses (\\citealt{gat95}; \\citealt{gat98}). Such a thick disc can have $\\tau\\simeq7\\times10^{-8}$ (\\citealt{gou94a}; \\citealt{gmb94}), so the summed optical depth including the contribution from the Galactic thin disc ($\\sim2\\times10^{-8}$) and the LMC itself ($\\sim1\\times10^{-8}$, \\citealt{sah94}) can be close to the observed value.    The three-fold degeneracy can be resolved in some kinds of exotic microlensing events, e.g., the binary event (\\citealt{har95}; \\citealt{alb99}; \\citealt{alc99a}; \\citealt{hon99}; \\citealt{afo00}; \\citealt{an01}) and the finite source transit event (\\citealt{gou92,gou94a}; \\citealt{nem94}; \\citealt{wit94}; \\citealt{pen97}).  \\citet{sum00} pointed out that an extensive transit events search would make it possible to discriminate between the lenses in the Galactic halo and in the LMC.   The third example is an event with parallax effect, which is essentially detected through the difference of light curves due to the spatial shift of an observer or observers. In such events we could determine the ``reduced transverse velocity'' $\\tilde{v}=v_{\\rm t}/(1-x)$ of the lens. Examples of the parallax effects are the following. The annual parallax effect due to the Earth's motion around the Sun during an event leads an asymmetry in the light curve (e.g. \\citealt{gou92}; \\citealt{gmb94}; \\citealt{miy95}; \\citealt{alc95};  \\citealt{mao99}; \\citealt{ben01}; \\citealt{bon01}; \\citealt{sos01};  \\citealt{smi02}; \\citealt{mao02}). However, such events are rare. This parallax effect generally requires $t_{\\rm E}>\\sim100$ days, while $t_{\\rm E}\\sim40$ days for typical events. Another effect is parallax by the positional difference of two well-separated observation sites. In this case, we could measure the relative difference of the peak amplifications and the time at the peak amplifications between both. By observing an event from both a solar-orbit satellite and the Earth, we could utilize the parallax effect in almost every event, while it is difficult from two distant observatories on the Earth (\\citealt{ref66}; \\citealt{gou94b,gou95b}; \\citealt{hol96}). Furthermore, the parallax could be measured by the positional shift of an observer due to the diurnal motion of the Earth, which was first advanced by \\cite{an02}, and due to the orbital motion of an Earth-orbit space telescope such as the {\\it Hubble Space Telescope} (HST) (\\citealt{hon99}) in binary events. In ordinary microlensing events this effect is quite small, but it is more efficient around the peak of high magnification events.    For Galactic bulge events, \\citet{gou97} discussed the possibility of detecting the finite source size and parallax effects by using two distinct ground-based telescopes in order to break the degeneracy in EME's (Extreme Microlensing Events, $A>200$). With an approximate estimate for EME observations towards the LMC, he concluded that it is not feasible. However, with a more careful estimate, \\cite{nak98} showed that it is feasible to detect EAGLE (Extremely Amplified Gravitational LEnsing) towards the LMC. EAGLE is similar to the so-called \"Pixel lensing\" events (\\citealt{gou96}) but more simply defined as the events in which the source star is dimmer than observational limiting magnitude (ex. $V_{obs} = 21 \\sim 22$), and not concerned whether the source star is resolved or not.  Some EAGLEs would be EME's. EAGLE events could be efficiently detected with the ``image subtraction method'' or  ''Difference Image Analysis (DIA)'' (\\citealt{ala98}; \\citealt{ala00}; \\citealt{alc99b,alc00a}; \\citealt{woz00}; \\citealt{bon01}), which has been recently developed and can perform more accurate photometry than DoPHOT and Pixel lensing method. So, we refer the term EAGLE in this paper.   If lens objects are in the thin or thick disc (disc events), $R_{\\rm E}$ projected onto the observer plane from the source star, \\begin{equation} \\label{eq:redein} \\tilde{R_{\\rm E}}(M,x)\\equiv \\frac{R_{\\rm E}(M,x)}{1-x} \\propto\\sqrt{\\frac{x}{1-x}}, \\end{equation} is much smaller than that for halo events. In this case the light curve is more sensitive to the small displacement of the observer position. Therefore the fraction of parallax-measurable EAGLE events out of all EAGLE events for disc events will be much larger than that for halo events. This fraction is useful to discriminate statistically whether the lens objects are mainly in the thick disc or not.   Here, we estimate the rate of parallax-measurable EAGLE events towards the LMC. In \\S\\,\\ref{sec:eagle} we summarize the basic equations of EAGLE events. In \\S\\,\\ref{sec:model} EAGLE event rates are estimated. In \\S\\,\\ref{sec:parallax_hst} and \\ref{sec:parallax_ground} we describe the measurement of parallax effect from space and ground telescopes, respectively. In \\S\\,\\ref{sec:fraction} we calculate the fraction of parallax-measurable events. Discussion and conclusion are given in \\S\\,\\ref{sec:disc}. ",
        "conclusions": "\\label{sec:disc}  We have seen that the EAGLE event rate is as high as that for normal events even for disc events towards the LMC. Since the period in which sources are visible ($V <21$) is usually short (1 day $\\sim$ 40 days), the detection efficiency heavily depends on the observational frequency. The observational programs currently undertaken by most groups are not adequate. Hourly monitoring with a 1-m class dedicated telescope and the real-time detection of EAGLE events by the DIA are required to issue alerts with a high detection efficiency. However, for the events in which the source star is $V>25$ this period is less than 2 days, so it is difficult to detect. Thus we estimated $\\Gamma_{\\!\\rm E}/\\Gamma_{\\!\\rm N}$ and $\\Gamma_{\\!\\rm P}/\\Gamma_{\\!\\rm E}$ in the case that the source star is $V<25$. These are decreased but still sufficiently high. Of course, the larger alert telescopes make it easier and faster to issue the alerts.  Estimating the parallax-measurable event rate, we advocate follow-up observations with a space telescope such as the HST and an 8-m class ground-based telescope such as the VLT. We have found that EAGLE events enlarge the opportunity of parallax measurements by $5 \\sim 9$ (from the HST) and $6 \\sim 10$ (from the VLT) times for disc events, and by $\\sim 8$ (from the HST) and $4 \\sim 6$ (from the VLT) times for halo events relative to that in normal microlensing events. We have also found we can measure the parallax effect in $\\sim75\\%$ (from the HST) and $\\sim 60\\%$ (from the VLT) of EAGLE events if the lenses are in the thick or thin disc, and in $\\sim 20\\%$ (from the HST) and $\\sim 10\\%$ (from the VLT) if the lenses are in the halo. Since $\\tilde{v}$ for halo objects are $5 \\sim 8$ times larger than that for disc stars, we can determine whether the lens objects are in the halo or discs for each event. We can also statistically constrain the lens locations by using the parallax-measurable EAGLE event rate.  In follow-up observations from the HST, in this paper we assumed that the observations start just after the peak as the most conservative case. Of course, an observation around the peak is better than that just after the peak. However predicting $t_0$ is not so easy for very faint source events. Extensive follow-up observations from small ground-based telescopes around the world are needed to predict $t_0$ and inform to the HST. Furthermore a flexible operating program of the HST for the alerts are required. If the alert telescope is at a latitude $-30^\\circ$, the alert can be issued for half of the year (Southern Summer), while at a latitude $-44^\\circ$ (such as New Zealand) it can be done all year round. The total operation time of the HST would be several days per year.  From the VLT, we assumed $\\delta = 45^\\circ$. However for the events with faint source ($V>24$) the time until $t_0$ is short, and the beginning of the follow-up would tend to be delayed. A delay of $\\sim 1$ day reduces the possibility of measuring the parallax by $\\sim 30 \\%$ as shown in \\S\\,\\ref{sec:results}. This observation can be done for only half of the year (Southern Summer).       The true source flux $f_0$ is needed to measure the precise value of $\\tilde{v}$ in the light curve fitting. Then follow-up observations by a high resolution telescope such as the HST are needed to get an accurate $f_0$ after the event.  In short, a practical observation strategy would be to observe hourly with a 1-m class telescope and perform real-time analysis with DIA to issue alerts to world observatories and the HST or the VLT for follow-up observations. Then after the events  $f_0$ should be measured by the HST.      To demonstrate this specifically we estimate the number of expected parallax-measurable events for the two cases that these are mainly halo events or disc events.  In both cases, the thin disc events are included.  For the typical parameters $\\alpha_{\\rm d} = 2.35$, $\\Sigma_{\\rm thin}=\\Sigma_{\\rm thick}=50 M_{\\odot}$pc$^{-2}$, $V_{\\rm th} = 20$, a detection efficiency of 50\\% and source stars of $V<25$, one can expect to find $\\simeq$13 EAGLE events from 3-year observations of 11 square degrees of the LMC central region (as the MACHO collaboration does). In the case of follow-up from the VLT, the expected number would be half.  In these 13 EAGLE events, $\\sim 2$ events $(15 \\% )$ are due to the stars in the thin disc and a further $\\sim 11$ events are due to MACHOs or the dark stars in the thick disc. In considering these 11 events, reasonable parameters are $M = 0.1$ or $0.5 M_{\\odot}$ except $M = 0.1M_{\\odot}$ in the case of the $\\delta$-function for halo events, and $M_{\\rm l} = 0.01M_{\\odot}$ for disc events, to be consistent with $\\langle t_E \\rangle\\simeq$40 days (\\citealt{alc00b}).  In this case, we will be able to measure $\\tilde{v}$ in $\\sim 10$ (from the HST) or $\\sim 4$ (from the VLT) events for disc events, $\\sim 4$ (from the HST) or $\\sim 1$ (from the VLT) event for halo events, which include thin disc events. We can constrain lens locations  strongly based on the 3-year statistics of these observations.   In conclusion, one could statistically discriminate whether the typical lens locations are in a thick disc or not, using parallax measurements even with $\\sim R_\\oplus$ scale baseline. One could also distinguish whether the lenses are MACHOs or stars in the LMC itself through finite source size effects measurements in EAGLE events (\\citealt{sum00}). Therefore one could identify lens objects as halo MACHOs, dark stars in the Galactic thick disc, or stars in the LMC through these observations.   A real-time alert system with DIA, has been introduced by the MOA collaboration\\footnote{see {\\tt http://www.phys.canterbury.ac.nz/\\~{}physib/alert/alert.html}} from 2000 and by the OGLE collaboration\\footnote{see {\\tt http://www.astrouw.edu.pl/\\~{}ogle/ogle3/ews/ews.html}} from 2002."
    },
    "0208/astro-ph0208485_arXiv.txt": {
        "abstract": "We present Chandra X-ray images of Tycho's supernova remnant that delineate its outer shock as a thin, smooth rim along the straight northeastern edge and most of the circular western half.  The images also show that the Si and S ejecta are highly clumpy, and have reached near the forward shock at numerous locations.  Most of the X-ray spectra that we examine along the rim show evidence of line emission from Si and S ejecta, while the continuum is well-represented by either a thermal or nonthermal model.  If the continuum is assumed to be thermal, the electron temperatures at the rim are all similar at about 2 keV, while the ionization ages are very low, because of the overall weakness of the line emission.  These electron temperatures are substantially below those expected for equilibration of the electron and ion temperatures, assuming shock velocities inferred from radio and X-ray expansion measurements; the electron to mean temperature ratios are $\\lesssim0.1-0.2$, indicating that collisionless heating of the electrons at the shock is modest.  The nonthermal contribution to these spectra may be important, but cannot be strongly constrained by these data.  It could account for as much as half of the flux in the 4-6 keV energy range, based on an extrapolation of the hard X-ray spectrum above 10 keV. ",
        "introduction": "As the bright remnant of an historically observed supernova in our Galaxy, Tycho's supernova remnant (SNR 1572) has been extensively studied, but the X-ray spectrum associated with its forward shock has not been directly measured until now.  By necessity, most X-ray spectral studies have focused on the spatially integrated spectrum, which is dominated at energies below a few keV by the ejecta.  A faint outer shelf of emission was identified in X-ray images from the Einstein Observatory, however, and is attributed to material behind the forward shock (Seward, Gorenstein, \\& Tucker 1983)---an association that is further supported by the excellent correspondence of the X-ray boundary with the sharp outer boundary seen in radio images (Dickel et al. 1991).  Imaging studies with the ROSAT X-ray Observatory establish that the expansion of the remnant varies both with azimuthal angle and radius (Hughes 2000).  The faster expansion observed at the outer boundary corresponds to an average forward shock velocity of 4600 $\\pm$ 400 ($D$/2.3 kpc) km/s, where $D$ is the distance in kpc.  The expansion of the radio boundary is also observed to vary with azimuthal angle (Strom, Goss, \\& Shaver 1982, Reynoso et al. 1997).  Strong deceleration in the east is caused by the remnant's interaction with dense H gas (Reynoso et al. 1999).  Although the outer boundaries of the X-ray and radio emission show excellent correspondence, the average expansions measured at these wavelengths are inconsistent: the expansion rate $m$, defined such that the time evolution of the remnant radius is $r \\sim t^m$, is 0.471 $\\pm$ 0.028 in the radio (Reynoso et al. 1997), and 0.71 $\\pm$ 0.06 for the outer radii in X-rays (Hughes 2000).  This discrepancy is unresolved, but appears to be a common pattern in young remnants (such as Cas A, Koralesky et al. 1998, Vink et al. 1998, and Kepler's SNR, Dickel et al. 1988, Hughes 1999).  In the case of Tycho's SNR, the forward shock velocity can be measured in yet another way.  The optical emission is almost exclusively H Balmer emission from nonradiative shocks propagating into partially neutral gas (Chevalier \\& Raymond 1978, Chevalier, Kirshner, \\& Raymond 1980, Kirshner, Winkler, \\& Chevalier 1987).  The neutral H atoms are not heated by the shock, and can be collisionally excited before being ionized; these excited atoms produce narrow Balmer lines consistent with their low temperatures.  Slow H atoms can also undergo charge exchange reactions with fast protons that have already been heated behind the shock; these fast H atoms contribute a broad component to the line profile.  The optical emission from Tycho's SNR is generally faint, and is detected only in the eastern and northern regions.  The most prominent feature is on the eastern side of the remnant (knot {\\it g} in the compilation of Kamper \\& van den Bergh 1978).  Smith et al. (1991) and Ghavamian et al. (2001) infer shock velocities through knot {\\it g} of $\\sim$ 2000 km/s independent of the distance to the remnant.  From modelling the line emission, while accounting for the effects of both the electron-ion temperature equilibration and Ly $\\alpha$ scattering, Ghavamian et al. constrain the electron to proton temperature ratio to be $\\lesssim$0.10, implying electron temperatures $\\lesssim$0.8 keV in the knot.  X-ray spectral studies of Tycho's SNR generally infer a relatively hard spectral component, presumed to be associated with the forward shock.  The total spectrum from the Advanced Satellite for Cosmology and Astrophysics (ASCA) X-ray Observatory suggests a forward shock component with a temperature of roughly 4 keV that accounts for some 30\\% of the X-ray flux between 0.5 to 10 keV (Hwang, Hughes, \\& Petre 1998).  This temperature is well below the mean equilibrium temperature behind a 4600 km/s shock.  Hughes (2000) points out that this implies either a low efficiency for electron heating, so that the electron temperature is well below the mean, or nonlinear particle acceleration, which could result in significantly higher compressions and lower temperatures than for test-particle shocks of the same velocity (Decourchelle, Ellison \\& Ballet 2000).  Furthermore, X-ray emission has been observed from the remnant at energies up to 25 keV (Pravdo \\& Smith 1979, Fink et al. 1994, Petre et al. 1999), and this emission has also been attributed to material behind the forward shock.  This interpretation requires a heating mechanism that can rapidly heat the electrons to sufficiently high temperatures (e.g., Cargill \\& Papadopoulos 1988).  The hard X-ray emission has alternatively been suggested to come from a nonthermal population of highly energetic electrons that has been accelerated at the shock (Aharonian et al.  2001).  X-ray instruments have lacked the capability to isolate the spectrum of the forward shock thus far.  Recent XMM-Newton observations have now provided the first truly spatially resolved spectra of Tycho's SNR on angular scales of several arcseconds (Decourchelle et al. 2001). Even higher spatial resolution (less than 0.5$''$ FWHM) is provided by the Chandra X-ray Observatory.  In this paper, we present images and spectra of selected portions of the forward shock of Tycho's SNR using the Advanced CCD Imaging Spectrometer (ACIS) on the Chandra Observatory. ",
        "conclusions": "\\subsection{Electron Temperatures}  We combined the X-ray spectral results with the radio and X-ray expansion velocities to estimate the ratio of electron to ion temperatures behind the forward shock in Tycho's SNR.  The temperature attained by particles as they pass behind the shock depends on their mass according to the shock jump conditions as $kT \\sim \\frac{3}{16} m v_s^2$, where $m$ is the particle mass and $v_s$ the shock velocity. The different particle species will exchange energy through Coulomb collisions and eventually attain a single equilibrium temperature, but it has been proposed that this equilibration may be effected much faster by collective plasma interactions (Cargill \\& Papadopoulos 1988).  The efficiency of these collisionless heating processes appears to decrease in inverse proportion to the Mach number of the forward shock, however, so that they may be less important for the fast shocks associated with young SNRs (Laming et al. 1996, Ghavamian et al. 2001).  The electrons behind the slower shocks in the Cygnus Loop, for example, have virtually equilibrated with the ions, in contrast to those behind the faster shocks of younger remnants such as SN 1006.  We plot in Figure 8 the equilibrium temperatures for Tycho's SNR corresponding to the radio expansion parameters determined as a function of azimuthal angle (defined counterclockwise from north) by Reynoso et al. (1997).  We use their Table 2 and Figure 5, and a distance of 2.3 kpc, to compute the velocities, and take the mean molecular weight per particle to be $\\mu m_p$, with $\\mu=0.6$ for solar abundance gas and $m_p$ being the proton mass. The rather large fluctuations for angles greater than 180$^\\circ$ are attributed by Reynoso et al. (1997) to uncertainties in determining the remnant radius.  Also shown in Figure 8 are the equilibrium temperatures corresponding to the average radio expansion velocity, and the average X-ray expansion velocity given by Hughes (2000).  It is not yet clear which of the radio or X-ray expansion values should be used, though the radio measurement is made with higher angular resolution data. Also, if the ejecta presently located near the forward shock have high velocities, they might bias the X-ray expansion measurement, as suggested by Wang \\& Chevalier (2001).  This question will hopefully be better resolved with a new study of the X-ray proper motion using higher spatial resolution data.  The data points in the Figure give the temperatures determined along the rim from our fits, assuming that the X-ray emission is thermal (Table 4); the appropriate azimuthal angle is determined with reference to the center adopted by Reynoso et al. (1997).  The electron temperatures we measure are clearly much lower than the mean temperatures expected for shock velocities determined from the radio expansion, with electron to mean temperature ratios of 0.10$-$0.20. If we scale the mean temperature by the ratio of the average X-ray velocity to the average radio velocity, it increases by a factor of two, and the electron to mean temperature ratios decrease accordingly. Although electron-ion temperature equilibration is not attained, the electron temperatures are higher than would be expected from Coulomb heating alone.  Ionization ages as low as $10^8$ cm$^{-3}$ s allow for negligible electron heating mediated by Coulomb collisions.  The measured electron temperature is thus essentially the electron temperature attained immediately after passage through the shock front, and represents extra heating in addition to Coulomb heating.  A low degree of electron-ion equilibration is consistent with results from the optical spectra for the brightest knot in Tycho's SNR (knot $g$), where the electron to proton temperature ratio is $\\lesssim 0.1$ (Ghavamian et al. 2001).  The degree of electron-ion equilibration has also been determined to be low in SN 1006, with the electron-to-proton temperature ratio $\\lesssim$ 0.05 using UV spectra (Laming et al. 1996). In SN 1987A, the temperatures measured from Chandra X-ray spectra are also lower than would be expected for equilibrium based on the observed radio and X-ray expansion of the remnant (Michael et al. 2002).  The implied ratio of electron and mean particle temperatures is about 0.1, giving an electron to proton temperature ratio of about 0.07 (with $\\mu$ = 7 for the N-enriched circumstellar material surrounding SN 1987A).  The X-ray line profiles (observed with the high spectral resolution High Energy Transmission Gratings on Chandra) are also consistent with thermal broadening due to high ion temperatures behind the shock.  All these results are summarized in Figure 9. For comparison, the model of Cargill \\& Papadopoulos (1988) for rapid electron heating can accommodate electron-to-ion temperature ratios of about 0.20.  In contrast to these cases, the measured electron temperature behind the forward shock in the Small Magellanic Cloud remnant E0102-72 is lower than would be expected for the X-ray determined shock velocity, even with Coulomb heating alone.  In this case, the low temperature is interpreted as evidence for highly efficient, nonlinear acceleration of particles behind the shock front (Hughes, Rakowski, \\& Decourchelle 2000).  There is independent evidence that nonlinear acceleration of electrons may be occuring in Tycho's SNR, in that the observed curvature of the radio electron synchrotron spectrum is predicted by such models (Reynolds \\& Ellison 1992).  The brightness of the radio emission at the rim also requires the fresh acceleration of particles at the shock (Dickel et al. 1991), and this process may extend to sufficiently high energies to affect the X-ray emission.  The temperatures expected behind shocks where efficient particle acceleration takes place are lower, making the ratio of measured electron temperature to mean temperature ratios higher than otherwise.  For the CKE, the observed temperature is actually consistent with temperature equilibration behind a shock at the radio velocity.  The radio velocity of the rim may not be applicable there, however, as the knot is interior to the rim and is probably only projected there. This knot is also very near the position of the optical knot {\\it g}, for which Ghavamian et al. (2001) determine that the electron temperature should be no higher than 0.8 keV, at 0.1 times the proton temperature.  However, there is a 10$''$ nominal offset between the X-ray and optical (Kamper \\& van den Bergh 1978) position, whereas the nominal X-ray coordinate uncertainty should be well under 3$''$ (Chandra Proposer's Observatory Guide); the optical measurements are probably not applicable to the continuum knot either.  One of the radio velocity points applicable to rim segment 4 is also consistent with the measured X-ray temperature, but this point represents a sharp excursion well below the average.  The foregoing discussion is based on the assumption that the emission is thermal, but it has been seen that nonthermal and thermal models are about equally successful in individually describing the featureless rim spectra.  In section 3.2, we argued that the spectra indicate that the temperatures should not be much higher than determined from the thermal fits alone, even in the presence of nonthermal emission.  In the next section, we present evidence that an additional nonthermal component is indeed likely to be present.  \\subsection{Nonthermal and Thermal Emission Components}  Hard X-ray emission has been unmistakably detected from Tycho's SNR, at energies up to 25 keV with HEAO-1 (Pravdo et \\& Smith 1979), 20 keV with Ginga (Fink et al. 1994); and 30 keV with RXTE (Petre et al. 1999).  Several models have been proposed to explain the hard X-ray continuum in Tycho's SNR as synchrotron radiation from electrons accelerated at the shock (Heavens 1984, Ammosov et al. 1994).  Given that the temperatures (in thermal models) at the forward shock in Tycho's SNR are only about 2 keV, the X-ray emission at higher energies may well have a nonthermal origin, as has been suggested in the literature.  We estimate an upper limit to the nonthermal X-ray luminosity from Tycho's SNR in the Chandra energy band by attributing {\\it all} the flux between 4-6 keV to a nonthermal component.  This is clearly an overestimate because there is a hot thermal component associated with the Fe K line emission (Hwang et al. 1998).  We first obtained the total source counts between 4-6 keV by estimating and subtracting the background counts from the continuum image and multiplying by a rough geometrical correction factor of 1.15 for the portion of the remnant that was not imaged.  We then took the 0.5-10 keV source luminosity of the nonthermal model for the northwest rim and scaled it up by the ratio of 4-6 keV counts in the total image to those in the rim segment, to estimate the nonthermal luminosity for the entire remnant. This maximum luminosity is about 2 $\\times 10^{35}$ ergs/s, but could be reduced by a factor of a few if we use only the bright X-ray continuum regions in the rim and the knots in our estimate.  We have not accounted for the variation in radio spectral index for different parts of the remnant, but if we use the spectrum fitted to the southwest rim, which has a much flatter radio slope, the estimated luminosity is essentially unchanged.  One can also estimate this nonthermal luminosity for Tycho's SNR by extrapolating the flat component in the Ginga or RXTE X-ray spectra down to energies in the Chandra band.  Extrapolating the power-law model given by Fink et al. (1994) for the Ginga data, we estimate a 0.5-10 keV nonthermal luminosity of about 1.5$\\times 10^{35}$ erg/s. Aharonian et al. (2001) fitted the RXTE spectrum with a model that includes an exponential cutoff, although they inferred a steeper radio slope than is actually observed overall, and obtained a higher turnover energy of 1.6 keV ($\\nu = 3.8 \\times 10^{17}$ Hz).  The 0.5-10 keV luminosity implied by their model is $1 \\times 10^{35}$ ergs/s.  Both estimates are a substantial fraction of the maximum possible luminosity from our estimate above, indicating that perhaps half of the hard 4-6 keV X-ray emission in Tycho's SNR is nonthermal. The nonthermal luminosity estimated from these hard X-rays is slightly higher than that of SN 1006 in the same energy range (Dyer et al. 2001).  The nonthermal X-ray emission should also be accompanied by $\\gamma$-ray emission, since the energetic electrons that emit the X-ray synchrotron radiation will also upscatter cosmic background photons.  The $\\gamma$-ray upper limits for Tycho's SNR have become tighter in recent years (Aharonian et al. 2001), at levels that are a few to several times lower than the detections for SN 1006 (Tanimori et al. 1998).  This may be understandable from a consideration of the densities and magnetic fields in these remnants.  The extrapolation of the radio spectrum to the X-ray range gives maximum electron energies comparable to those determined for SN 1006 (Reynolds \\& Keohane 1999), if equal magnetic field strengths of 10 $\\mu$G are assumed.  A magnetic field of $10^{-4}$ to $10^{-2}$ G in Tycho's SNR from Reynolds \\& Ellison (1992) is higher than the 9 $\\mu$G in SN 1006 (Dyer et al. 2001).  A higher magnetic field qualitatively accounts not only for the more severe synchrotron energy losses of the electrons in Tycho's SNR, but also its higher radio surface brightness.  The maximum electron energies will be lower in proportion to $B^{-1/2}$; on the basis of our spectral fits, the maximum electron energies would appear to be between 1-15 TeV, depending on the exact value of the magnetic field.  The much lower density environment of SN 1006 (e.g., see Kirshner et al. 1987) would also allow particle acceleration to proceed to higher energies in the first place (Baring et al. 1999).  \\subsection{Mixing of Ejecta}  The presence of ejecta at the forward shock requires a significant amount of mixing either during or after the explosion.  Various high resolution hydrodynamical studies (Dwarkadas 2000, Wang \\& Chevalier 2001) show that the ejecta cannot penetrate beyond about half the interaction region through Rayleigh-Taylor instabilities alone, though early Richtmyer-Meshkov instabilities might enhance the effectiveness of the Rayleigh-Taylor instabilities (Kane et al. 1999).  Clumps in the ejecta (Hamilton 1985) or in the interstellar medium (Jun, Jones \\& Norman 1996) would, however, allow the ejecta to penetrate to large radii.  Ejecta might also penetrate the forward shock in the presence of enhanced turbulence generated as the reverse shock propagates into low density bubbles of Fe, formed as the heat release from the radioactive decay of clumps of Ni ejecta allows them to expand (Blondin et al. 2001).  Yet another possibility is that the ejecta can affect the forward shock when the interaction region for the shock is very thin, as could be the case with nonlinear particle acceleration behind the forward shock---a situation that might reasonably be expected to occur in young remnants (Blondin \\& Ellison 2001), including Tycho's SNR in particular (Reynolds \\& Ellison 1992).  The action of the Rayleigh-Taylor instabilities out to the forward shock is further supported by the radial orientation of the magnetic field (Dickel et al. 1991), which is observed all the way to the rim of the remnant.  One might expect that if the ejecta do penetrate the forward shock, the boundary of the outer shock would be distorted, whereas the western boundary of Tycho's SNR is smooth and circular.  This might still be possible, however, if the mixing occured early enough for the protruberances to have subsided.  In their simulations, Wang \\& Chevalier (2001) found that it was actually difficult to deform the forward shock unless the ejecta density contrast was very high.  The qualititative appearance of the Si ejecta in Tycho's SNR is quite different from that of the core-collapse remnant Cas A, which has also been beautifully imaged by Chandra (Hughes et al. 2000, Hwang et al. 2000).  While the Si is clumpy in Tycho, it is compact and knotty in Cas~A.  Despite the fact that Tycho is located at a smaller distance, the angular scale of its Si features is larger than those seen in Cas A.  Differences are also seen in the 4-6 keV continuum images.  In Tycho's SNR, the ejecta shell has virtually disappeared in this image, but in Cas A, the ejecta are still clearly visible.  The emission associated with the ejecta is thus seen to be relatively more important at these energies in Cas A than in Tycho's SNR.  Indeed, the hard X-ray emission in Cas A has been proposed to be primarily bremsstrahlung emission from electrons that have been accelerated in the ejecta (Laming 2001ab, Bleeker et al. 2001).  The appearance of the forward shock in these two remnants is also strikingly different.  Tycho's outer rim is distorted because of its interaction in the east, but the outline of the rim is otherwise generally smooth and continuous.  By contrast, the outer rim of Cas A is broken up on small scales and shows tightly curved fragments (Gotthelf et al. 2000).  If some of this emission is nonthermal, this may reflect differences seen in their radio emission, as Tycho has a sharp radio rim suggesting efficient first order Fermi acceleration, whereas Cas A has no distinct radio rim and is a better candidate for second order Fermi acceleration mediated by turbulence (e.g., Dickel et al. 1991).  SN 1006, another Type Ia remnant, shows the same smooth outer rim seen in Tycho's SNR, and the same highly clumpy ejecta (in this case O ejecta; K. Long, private communication), but this is still much too small a sample for these differences to be more than suggestive."
    },
    "0208/astro-ph0208166_arXiv.txt": {
        "abstract": "We present an analysis of the first high-resolution spectra measured from an accretion-driven millisecond X-ray pulsar in outburst.  We observed XTE J1751--305 with {\\it XMM-Newton} on 2002 April 7 for approximately 35~ks.  Using a simple absorbed blackbody plus power-law model, we measure an unabsorbed flux of $6.6 \\pm 0.1 \\times 10^{-10}~ {\\rm erg}~ {\\rm cm}^{-2}~ {\\rm s}^{-1}$~ (0.5--10.0 keV).  A hard power-law component ($\\Gamma = 1.44 \\pm 0.01$) contributes 83\\% of the unabsorbed flux in the 0.5--10.0 keV band, but a blackbody component ($kT = 1.05 \\pm 0.01$~ keV) is required.  We find no clear evidence for narrow or broad emission or absorption lines in the time-averaged spectra, and the sensitivity of this observation has allowed us to set constraining upper-limits on the strength of important features.  The lack of line features is at odds with spectra measured from some other X-ray binaries which share some similarities with XTE J1751--305.  We discuss the implications of these findings on the accretion flow geometry in XTE~J1751--305. ",
        "introduction": "Millisecond radio pulsars are thought to be created in neutron star low-mass X-ray binaries (LMXBs). In those LMXBs, accreting matter may spin-up the neutron star (see, e.g., Bhattacharya \\& van den Heuvel 1991).  Therefore, it is expected that millisecond pulsars should also be found in LMXBs during the accretion phase of the binary as X-ray pulsars.  Although evidence for rapidly spinning neutron stars in LMXBs was inferred from the burst oscillations which were seen during type-I X-ray bursts in several systems (see Strohmayer 2001 for a review), the detection of millisecond pulsations in persistent emission remained elusive for many years.  In 1998, the first such system was discovered (SAX J1808.4--3658, which has a spin frequency of 401 Hz; Wijnands \\& van der Klis 1998a).  This source was extensively studied due to its obvious importance for binary evolution scenarios and accretion flow geometries and dynamics (see, e.g., Wijnands \\& van der Klis 1998b and references therein).  In the spring of 2002, Markwardt \\& Swank (2002a) reported the discovery of the second accretion-driven millisecond pulsar XTE J1751--305 (435 Hz) in the {\\it Rossi X-ray Timing Explorer} ({\\it RXTE}) bulge scan observation program.  About one month after the discovery of this system, XTE J0929--314 was discovered (185 Hz; Remillard, Swank, \\& Strohmayer 2002.).  The neutron stars in both systems are in orbit around a companion star, with an orbital period of $\\sim$42 minutes (Markwardt \\& Swank 2002b; Galloway et al. 2002). These systems are very tight binaries and the inferred mass of their companion stars is very low ($\\sim 0.01~M_{\\odot}$).  After the discovery of XTE J1751--305, we submitted a target-of-opportunity request to {\\it XMM-Newton} for the purpose of studying this source with high-resolution spectroscopy.  The X-ray spectrum of the first system --- SAX J1808.4--3658 --- could only be studied during outburst using {\\it RXTE} and with the Wide Field Cameras aboard\\textit{BeppoSAX}, which have only moderate spectral resolution.  In this {\\it Letter}, we present an analysis of the time-averaged EPIC-pn and Reflection Grating Spectrometer (RGS) data. The spectra obtained represent the first CCD- and gratings-resolution measurements from a millisecond X-ray pulsar in outburst. ",
        "conclusions": "We have analyzed the first CCD- and grating-resolution X-ray spectra of a millisecond X-ray pulsar in outburst.  We find no convincing evidence for broad or narrow emission or absorption features in the the EPIC-pn and RGS spectra (see Figure 1 and Table 1).  With our simple blackbody plus power-law model for the EPIC-pn spectrum, we find that the power-law component comprises 83\\% of the unabsorbed flux in the 0.5--10.0~keV band.  The power-law component is hard: $\\Gamma = 1.44 \\pm 0.01$, but within a range typical for X-ray pulsars (e.g., $\\Gamma =$ 1.0--1.5; White, Swank, \\& Holt 1983).  The EPIC-pn spectrum requires a blackbody with $kT = 1.05 \\pm 0.01$~keV and an implied emitting radius of $R = f^{2} \\times 3.01^{+0.09}_{-0.06}~ (d/10 {\\rm kpc})$~km.  It is fair to say that our knowledge of the spectral hardening factor $f$ is rather poor (for a discussion, see Lewin, van Paradijs, \\& Taam 1993).  If $f$ is less than 1.3, it would indicate that only part of the neutron star surface (a ``hot spot'') is involved.  Markwardt et al. (2002) fit a simple absorbed power-law model to the {\\it RXTE} spectrum of XTE~J1751--305, and measure a softer power-law index ($1.7 < \\Gamma < 1.9$).  The effective lower sensitivity bound of the {\\it RXTE} Proportional Counter Array (PCA) is 3 keV; fitting the 3--10~keV EPIC-pn spectrum with only a power-law, we find $\\Gamma = 1.72 \\pm 0.01$ for $N_{H} = (9.8\\pm 0.1) \\times 10^{21}~ {\\rm atoms}~{\\rm cm}^{-2}$ and $\\Gamma = 1.90 \\pm 0.02$ allowing $N_{H}$ to float (in this case, $N_{H} = [2.5\\pm 0.1] \\times 10^{22}~ {\\rm atoms}~{\\rm cm}^{-2}$).  Thus, the discrepant indices are easily explained in terms of different instrumental ranges.  In modeling simultaneous {\\it Chandra}/LETGS and {\\it RXTE} spectra of XTE J0929--314, Juett, Galloway, \\& Chakrabarty (2002) measure similarly discrepant power-law indices.  They suggest an astrophysical origin for the discrepancy in XTE J0929--314 and XTE~J1751--305 in the form of a power-law with a break or a roll-over in the 1.4--4.4~keV range.  A model consisting of blackbody and broken power-law components ($E_{break} = 3.7 \\pm 0.1~{\\rm keV}x,~ \\Gamma_{E < 3.7} = 1.34 \\pm 0.03,~ \\Gamma_{E > 3.7} = 1.60 \\pm 0.04$) provides an improved fit to the spectrum of XTE J1751--305 ($\\chi^{2}/\\nu = 1.090, \\nu = 1882$).  Although these results suggest that an astrophysical origin for the apparent spectral evolution is possible, a number of concerns remain.  Given that: (1) the effective area (flux) calibration of the EPIC-pn has 5\\% uncertainties (Kirsch 2002), (2) the LETGS (plus ACIS-S array) flux calibration has 10\\% uncertainties above 1~keV (H. Marshall, priv. comm.), (3) that {\\it RXTE} PCA and High Energy X-Ray Timing Experiment (HEXTE) observations of the ``Crab'' nebula (taken to be a pure power-law for calibration) differ in power-law index by $\\delta(\\Gamma) \\simeq 0.1$ (see, e.g., Wilms et al. 1999), and (4) that Markwardt et al. (2002) did not fit a blackbody component to the {\\it RXTE} spectra of XTE~J1751--305, we do not consider an astrophysical origin for the discrepant indices to be required by the present data.  The upper limits on the strength of narrow (consistent with instrument resolution; FWHM $\\sim$ 0.1~keV) and broad (FHWM $=$ 0.7~keV) Fe K$\\alpha$ emission lines are 4~eV and 6~eV, respectively (95\\% confidence).  This stands in contrast the detection of a weak Fe K$\\alpha$ line in outburst spectra of the 401~Hz millisecond X-ray pulsar SAX J1808.4--3658 obtained with {\\it RXTE} (Heindl \\& Smith 1998; Gierlinski, Done, \\& Barret 2002).  It is possible that a strong, hot corona (or the magnetosphere, if the coranae in these LMXBs is small or disrupted) which produces to the power-law component may ionize the disk in XTE~J1751$-$305 to a degree which prevents Fe~K$\\alpha$ emission.  The absence of H-like and He-like resonance emission lines like those observed in 4U 1626--67 (Schulz et al. 2001) in the spectrum of XTE J1751--305 is also consistent with a highly ionized disk in this source.  Secondarily, the absence of such lines may be due to differences in the accretion flow geometry due to the neutron star magnetic field.  Whereas the magnetic field in XTE~J1751$-$305 is likely to be similar to that in SAX~J1808.6$-$3658 ($B = (2-6) \\times 10^{8}$~G; Wijnands \\& van der Klis 1998a), the magnetic field in 4U~1626$-$67 is likely much higher ($B = 3 \\times 10^{12}$~G, Orlandini et al. 1998).  The stronger magnetic field in 4U 1626--67 may disrupt the inner disk, allowing cool material in the outer disk to be irradiated by a weaker corona.  Emission lines have also been seen in neutron star systems which are viewed edge-on (so-called accretion disk coronae or ``ADC'' sources; see, e.g., Kallman et al. 2002); the absence of emission lines in XTE J1751--305 may also be partially due to a low system inclination (Markwardt et al. 2002 report no evidence of dips or eclipses).  Studies of some ultra-compact systems suggest a Ne-rich companion (for observational results, see: Schulz et al. 2001; Juett, Psaltis, \\& Chakrabarty 2001; for theoretical discussions see Yungelson, Nelemans, \\& van den Heuvel 2002; Bildsten 2002).  Fits to the Ne photoelectric absorption edge in the RGS spectra of XTE J1751--305 constrain the abundance of Ne to be at most 77\\% of the solar value along this line of sight (95\\% confidence upper-limit; see Section 3.3).  This may suggest that a Ne-rich companion is unlikely in the case of XTE J1751--305."
    },
    "0208/astro-ph0208172_arXiv.txt": {
        "abstract": "Massive stars evolve across the HR diagram, losing mass along the way and forming a variety of ring nebulae.  During the main sequence stage, the fast stellar wind sweeps up the ambient interstellar medium to form an interstellar bubble.  After a massive star evolves into a red giant or a luminous blue variable, it loses mass copiously to form a circumstellar nebula.  As it evolves further into a WR star, the fast WR wind sweeps up the previous mass loss and forms a circumstellar bubble.  Observations of ring nebulae around massive stars not only are fascinating, but also are useful in providing templates to diagnose the progenitors of supernovae from their circumstellar nebulae.  In this review, I will summarize the characteristics of ring nebulae around massive stars throughout the HR diagram, show recent advances in X-ray observations of bubble interiors, and compare supernovae's circumstellar nebulae with known types of ring nebulae around massive stars. ",
        "introduction": "Since Johnson \\& Hogg (1965) reported the first three ring nebulae around Wolf-Rayet (WR) stars, nebulae of various shapes, sizes, and ionization conditions have been observed around massive stars of different spectral types, such as luminous blue variables (LBVs), blue supergiants (BSGs), and red supergiants (RSGs). As these various spectral types in the upper part of the Hertzsprung-Russell (HR) diagram are strung together by the evolutionary tracks of massive stars, their surrounding nebulae must be evolutionarily related. This relationship was illustrated clearly by Garc\\'{\\i}a-Segura, Langer, \\& Mac Low (1996) and Garc\\'{\\i}a-Segura, Mac Low, \\& Langer (1996) in their hydrodynamic simulations of the formation of WR ring nebulae taking into account the evolution and mass loss history of the central stars.  In this review, I will cover three topics.  First, I will present a gallery of ring nebulae around massive stars throughout the HR diagram and compare them to Garc\\'{\\i}a-Segura et al.'s framework of hydrodynamic evolution of circumstellar gas around massive stars.  Second, I will report an X-ray view of the hot gas in the interiors of bubbles blown by massive stars, using ROSAT, ASCA, Chandra, and XMM-Newton observations. Finally, I will compare the circumstellar nebulae observed around young supernovae (SNe) to those around massive stars and use the nebular properties to diagnose the spectral types of SN progenitors. ",
        "conclusions": ""
    },
    "0208/astro-ph0208491_arXiv.txt": {
        "abstract": "We  report on the proper  motion  measurement of  the proposed optical counterpart of the X-ray/radio   pulsar  PSR 1929+10.  Using    images obtained with the   HST/STIS (average epoch   2001.73)  we computed  a yearly displacement of $+97 \\pm 1$ mas yr$^{-1}$ in RA and $+46 \\pm 1$ mas yr$^{-1}$ in Dec since the epoch (1994.52) of the original HST/FOC detection.  Both   the magnitude and  direction of  the optical proper motion components are   found to be  fully  consistent with  the  most recent VLBA radio measurements.   This result provides an  unambiguous confirmation of  the pulsar optical  identification.  In  addition, we have used the combined STIS/FOC datasets  to derive information on the pulsar spectrum,  which seems characterized by  a power  law component, apparently unrelated to the X-ray emission. ",
        "introduction": "PSR1929+10 is an old  ($ \\sim 3~  10^{6}$ yrs)  radio pulsar.   With a distance  of  $\\sim$   330  pc,  determined from   VLBA radio parallax measurements (Brisken et al.  2002), it is also one  of the closest to the solar  system.  After the  original  X-ray detection with Einstein (Helfand   1983),  pulsations at the    radio   period (227 ms)   were discovered by ROSAT  (Yancopulous et al. 1994)  and later confirmed in ASCA data (Wang and Halpern  1997).  The X-rays pulse profile exhibits a single, broad,  peak markedly  different  from the sharp  radio one. The X-ray spectrum can be described either by a blackbody ($T \\sim 3-5 ~ 10^{6}~K$) produced  from hot polar caps  (Yancopulous et al.  1994; Wang and Halpern 1997) or by a power law with $\\alpha \\approx 1.27 \\pm 0.4$ (Becker  and Tr\\\"umper 1997).  A trail  of diffuse X-ray emission originated from the  pulsar and extending $\\sim$   10 arcmin to  South East    was discovered with  the ROSAT/PSPC   (Wang  et al.  1993).  A candidate  optical counterpart to PSR1929+10  was  identified with the HST/FOC by Pavlov et  al.  (1996) based  on the positional coincidence ($\\sim 0\\farcs4$)   with  the radio  coordinates.  Interestingly,  the measured flux of the PSR1929+10 counterpart  ($U \\sim 25.7$) was found to deviate by 3 orders of magnitude from the values predicted from the X-ray spectra.  This behaviour is  markedly different from that of the middle aged pulsars PSR0656+14 (Pavlov et al.  1997), Geminga (Mignani et al.  1998) and PSR1055-52 (Mignani et  al. 1997), where the optical data are not  too  far from the   extrapolation of the  X ray spectra. Confirming the optical identification  of PSR1929+10 becomes a crucial step to settle  a consistent scenario for the  long term evolution  of the   optical  luminosity  of  pulsars  and  to  investigate  possible turnovers in the emission physics.  While young ($ \\sim 10^{3}-10^{4}$ yrs) objects are relatively bright, their  optical throughput seems to decay on a timescale of few thousands  years and progressively turn to a composite magnetospheric/thermal   regime  (Mignani   1998;  Caraveo 2000).  Although  evidence  for  such  a trend  can be  recognized  in middle-aged objects ($\\sim 10^{5}$ yrs), like PSR 0656+14 and Geminga, little is known on the optical emission at later  stages of the pulsar lifetime. \\\\  Taking advantage of   new HST observations, we  use  the pulsar proper motion to secure the PSR 1929+10 optical identification, thus adding an important piece of information on the optical behaviour of old pulsars.  Observations and data reduction are described in \\S2, while the results are discussed in \\S3. ",
        "conclusions": "Using images collected  with the STIS camera  aboard HST together with archived HST/FOC images taken 7.2 years apart we  have measured a very significant angular displacement  of the proposed  optical counterpart tof PSR 1929+10.  This yields a proper motion $\\mu_{\\alpha}cos(\\delta) =  +97  \\pm    1$  mas~yr$^{-1}$  and  $\\mu_{\\delta}  =    +46 \\pm  1$ mas~yr$^{-1}$. These  values  agree with the  ones  derived  from very recent VLBA radio measurements  (Brisken et al.  2002), thus providing an unambiguous  confirmation  of the pulsar  identification.  Securing the   identification of an   old pulsar  such  as   PSR 1929+10  is an important step to assess the  pulsars' optical behaviour as a function of their  age.  At   variance  with the phenomenology  of  middle aged objects, such as PSR 0656+14, Geminga and PSR 1055-52, the optical emission of PSR 1929+10 seems unrelated to the X-ray one, be it either of thermal pr non-thermal origin. Although the new STIS data seem to favour a power law rather than a blackbody, the paucity of flux values and the limited  spectral coverage available  do not  allow us to  put firmer constraints  on the spectrum.  More  data, expecially at longer wavelengths, are required  to better characterize the  pulsar spectral shape and  to   unveil the  possible presence of    different spectral components.  The detection of the optical timing signature will add an important piece of information to understand  the optical behaviour of this old pulsar."
    },
    "0208/astro-ph0208344_arXiv.txt": {
        "abstract": "{ Both nulling and subpulse drifting are poorly understood phenomena. We probe their mechanisms by investigating how they interact in PSR~B0809+74. We find that the subpulse drift is not aliased but directly reflects the actual motion of the subbeams. The carousel-rotation time must then be over 200 seconds, which is much longer than theoretically predicted. \\newline The drift pattern after nulls differs from the normal one, and using the absence of aliasing we determine the underlying changes in the subbeam-carousel geometry. We show that after nulls, the subbeam carousel is smaller, suggesting that we look deeper in the pulsar magnetosphere than we do normally. The many striking similarities with emission at higher frequencies, thought to be emitted lower too, confirm this. The emission-height change as well as the striking increase in carousel-rotation time can be explained by a post-null decrease in the polar gap height. This offers a glimpse of the circumstances needed to make the pulsar turn off so dramatically. ",
        "introduction": "In pulsars, the emission in individual pulses generally consists of one or more peaks (`subpulses'), that are much narrower than the average profile and the brightness, width, position and number of these subpulses often vary from pulse to pulse.  In contrast, the subpulses in PSR~B0809+74 have remarkably steady widths and heights and form a regular pattern (see Fig. \\ref{img:data.fit}a). They appear to drift through the pulse window at a rate of $-0.09 P_2/P_1$, where $P_2$ is the average longitudinal separation of two subpulses within one rotational period $P_1$, which is $1.29$ seconds. Figure \\ref{img:data.fit}a also shows how the pulsar occasionally stops emitting, during a so-called null.  \\begin{figure}[t] \\centering \\includegraphics[]{H3808F1.eps} \\caption{Observed and fitted pulse sequences. A window on the pulsar emission is shown for 150 pulses. One pulse period is $360^\\circ$. The centre of the Gaussian that fits the pulse profile best is at $0^\\circ$. {\\bf a)} The observed pulse sequence, with a null after pulse 30. {\\bf b)} The Gaussian curves that fitted the subpulses best. Nulls are shown in lightest gray, driftbands fitted to the subpulse pattern are medium gray.} \\label{img:data.fit} \\end{figure}   In this paper, we will interpret the drifting subpulse phenomenon in the rotating carousel model \\citep{rs75}. In this model, the pulsar emission originates in discrete locations (`subbeams') positioned on a circle around the magnetic pole. The circle rotates as a whole, similar to a carousel, and is grazed by our line of sight. In between successive pulses, the carousel rotation moves the subbeams through this sight line, causing the subpulses to drift.  Generally, the average profiles of different pulsars evolve with frequency in a similar manner: the profile is narrow at high frequencies and broadens towards lower frequencies, occasionally splitting into a two-peaked profile \\citep{kis+98}. This is usually interpreted in terms of `radius to frequency mapping', where the high frequencies are emitted low in the pulsar magnetosphere. Lower frequencies originate higher, and as the dipolar magnetic field diverges the emission region grows, causing the average profile to widen.  The profile evolution seen in PSR~B0809+74 is different. The movement of the trailing edge broadens the profile as expected, but the leading edge does the opposite. The profile as a whole decreases in width as we go to lower frequencies until about 400 MHz. Towards even lower frequencies the profile then broadens somewhat \\citep{dls+84, kis+98}. Our own recent observations of PSR~B0809+74, simultaneously at 382, 1380 and 4880 MHz, confirm these results \\citep{rrl+02}. Why the leading part of the expected profile at 400 MHz is absent is not clear. While \\citet{bkk+81} suggest cyclotron absorption, \\citet{dls+84} conclude that the phenomenon is caused by a non-dipolar field configuration. We will refer to this non-standard profile evolution as `absorption', but none of the arguments we present in this paper depends on the exact mechanism involved.  In a recent paper (\\nocite{lkr+02}van Leeuwen et~al. 2002, henceforth Paper I) we investigated the behaviour of the subpulse drift in general, with special attention to the effect of nulls. We found that after nulls the driftrate is less, the subpulses are wider but more closely spaced, and the average pulse profile moves towards earlier arrival. Occasionally this post-null drift pattern remains stable for more than 150 seconds.  For a more complete introduction to previous work on PSR~B0809+74, as well as for information on the observational parameters and the reduction methods used, we refer the reader to Paper I. In this paper we will investigate the processes that underly the post-null pattern changes. We will quantify some of the timescales associated with the rotating carousel model and map the post-null changes in the drift pattern onto the emission region.  One of the interesting timescales is the time it takes one subbeam to complete a rotation around the magnetic pole. This carousel-rotation time is predicted to be of the order of several seconds in the Ruderman \\& Sutherland model. Only recently a carousel-rotation time was first measured: \\citet{dr99} find a periodicity associated with a 41-second carousel-rotation time for PSR~B0943+10.  The second goal is to determine the changes in the emission region that underly the different drift pattern we see after nulls. Mapping this emission region could increase our insight into what physically happens around nulls.  Achieving either goal requires solving the so-called aliasing problem: as the subpulses are indistinguishable and as we observe their positions only once every pulse period, we cannot determine their actual speed. ",
        "conclusions": "We have shown that the drift of the subpulses directly reflects the actual motion of the subbeams, without any aliasing.  In other pulsars with drifting subpulses this may be different: for those we predict drift-direction reversals or longitude jumps in the post-null drift pattern.  We find that the carousel-rotation time for PSR~B0809+74 must be long, probably over 200 seconds. The expected lifetime of the subbeam characteristics is less, which explains why thus far no periodicity from the carousel rotation could be found in the pulse sequence.  The rotation time we find is larger than theoretically predicted, not only in absolute numbers but also after extrapolating the rotation time found in PSR~B0943+10. Both the magnitude and the scaling relations that link the carousel-rotation time to the magnetic field and period of the pulsar are therefore incorrect.  When the emission restarts after a null the drift pattern is different, and having determined the alias mode, we identify the underlying changes in the geometry of the subbeam carousel. A combination of a decrease in carousel size and `absorption' already explains many of the changes seen in the post-null drift pattern.  The resemblance between the drift pattern after nulls and that seen at higher frequencies, thought to originate at a lower height, is striking. Assuming that similar effects have identical causes leads us to conclude that after nulls we look deeper in the pulsar magnetosphere, too.  Both this decrease in viewing depth and the striking increase in the carousel rotation time can be quantitatively explained by a post-null decrease in gap height."
    },
    "0208/astro-ph0208202_arXiv.txt": {
        "abstract": "{ All the {\\em Extreme Ultraviolet Explorer} (EUVE) observations of AD~Leo, totalling  1.1~Ms of exposure time, have been employed to analyze the corona of this single M dwarf. The light curves show a well defined quiescent stage, and a distribution of amplitude of variability following a power law with a $\\sim -2.4$ index. The flaring behavior exhibits much similarity with other M active stars like FK~Aqr or YY~Gem, and flares behave differently from late type active giants and subgiants. The Emission Measure Distribution (EMD) of the summed spectrum, as well as that of quiescent and flaring stages, were obtained using a line-based method. The average EMD is dominated by material at log~T(K)$\\sim$6.9, with a second peak around log~T(K)$\\sim$6.3, and a large increase in the amount of material with log~T(K)$\\ga$7.1 during flares, material almost absent during quiescence. The results are interpreted as the combination of three families of loops with maximum temperatures at log~T(K)$\\sim$6.3, $\\sim$6.9 and somewhere beyond log~T(K)$\\ga$7.1. A value of the abundance of [Ne/Fe]=1.05$\\pm$0.08 was measured at log~T(K)$\\sim$5.9. No significative increment of Neon abundance was detected between quiescence and flaring states. ",
        "introduction": "AD Leo (GL 388) is a dM3 star located at a distance of 4.9~pc \\citep{hen94}. It is a well known frequent source of flares. Its high activity is probably due to its high rotation rate \\citep[P$_{\\rm phot}\\sim$2.7~d,][]{spi86}, and a large number of studies on this star has been carried out because of its nature as a very active single star. M stars are supposed to have a large convective layer, resulting in a high level of photospheric spot coverage and frequent flaring activity. With a mass of $M = 0.40 M_{\\odot}$, \\object{AD Leo} would still be close to the limit where the radiative core of the star is still present \\citep{fav00}. Studies in X-rays reflect that M stars are the stellar class with highest L$_X$/L$_{\\rm bol}$ values, binary systems or very young stars apart.  AD~Leo is one of the paradigms of flaring stars, and has been subject of frequent studies in the EUV and X-rays. X-rays low resolution spectra have been used by several authors \\citep[see][ and references therein]{fav00} to obtain fits to 2 or 3 temperatures that could explain such spectra.  With the advent of the {\\em Extreme Ultraviolet Explorer} (EUVE) it was possible to obtain the first high resolution spectra, and an analysis based on 85~ks of observation permitted \\citet{cul97} to get an Emission Measure Distribution (EMD) during different activity levels. However, the low statistics of the observation did not allow an accurate analysis of the EMD to be performed, and in the best case only 3 lines had signal-to-noise ratio (S/N) higher than 4, making only indicative the analysis on the variations between quiescent and flaring stages. Some studies relative to the EUVE light curves have been carried out by \\citet{haw95}, \\citet{gud01}, and \\citet{haw01}, but no deep analysis in has been conducted from high quality EUVE spectra to date.  More recently, the analysis of Chandra/Low-Energy Transmission Grating (LETG) spectrum \\citep{mag01} has been used to get values of the abundances, and an EMD peaking at log~T(K)$\\sim$6.8. The EUVE satellite (\\gl\\gl$\\sim$70--750~\\AA) has the advantage in respect of Chandra (\\gl\\gl$\\sim$1--175) and the {\\em X-ray Multi Mirror telescope} (XMM-Newton) (\\gl\\gl$\\sim$5--35) of a good coverage of lines of just one element. Iron lines observed with EUVE permit calculating the structure of the EMD in the range log~T(K)$\\sim$5.8--7.4 without any ambiguity produced by the use of different values of abundances of the elements observed.  In contrast, Chandra or XMM-Newton can observe a larger number of lines, also with higher spectral resolution, from different elements, covering a larger temperature region [up to log~T(K)$\\sim$8]. But in order to get a good temperature coverage, the spectral lines corresponding to different elements must be used. Hence if no proper calculation of the abundances is performed in the analysis of such spectra, this can result in a wrong EMD.  While the combination of the observations with EUVE and Chandra or XMM-Newton is desirable, the frequent variability of a star like AD~Leo makes such analysis difficult. Finally, the observations made with EUVE tend to span long periods of time (several days) in order to achieve good statistics in the spectra, permitting to get a good analysis on coronal variability, and even a separate analysis of flaring and quiescent stages, as it has been done, for example, for several RS~CVn stars \\citep{sanz01,paper1}.  \\begin{table} \\caption{\\scshape EUVE exposure times (s) of the AD~Leo observations (SW and MW spectra)}\\label{times} \\begin{center} \\begin{tabular}{lrr} \\hline \\hline {Start date}  & {SW}  & {MW} \\\\ \\hline {1 Mar 1993}  & {84\\,622}  & 85\\,271 \\\\ {3 Mar 1996}  & {73\\,322}  & 73\\,320 \\\\ {2 Apr 1999}  & {46\\,441}  & 44\\,384 \\\\ {5 Apr 1999}  & {141\\,885} & 141\\,888 \\\\ {9 Apr 1999}  & {133\\,001} & 131\\,198 \\\\ {17 Apr 1999} & {177\\,093} & 176\\,239 \\\\ {25 Apr 1999} & {158\\,084} & 145\\,181 \\\\ {6 May 1999}  & {190\\,759} & 147\\,785 \\\\ {9 Mar 2000}  & {99\\,952}  & 52\\,686 \\\\ \\hline \\end{tabular} \\end{center} \\end{table}  A total of 1~Ms of observations converts AD Leo in the active star most observed with EUVE, yielding a combined spectrum with high statistics, and it permits to study the variability properties observed during an elapsed time of 46~d.  In this paper we are presenting the most accurate analysis to date on the EMD of an M star without the ambiguity derived from the use of different elements. The long duration of the observations of AD~Leo allows us to perform an analysis of the flaring behavior from high quality spectra, and from the light curves as compared to the Sun. Given the spectral range response of the EUVE/DS ($\\lambda\\lambda$70--175~\\AA), such light curve will be dominated by emitting material with log~T(K)$\\sim$6.7--7.2.  A description of the technical information of the observations is given in Sect.~\\ref{sec:observations}. The analysis of data (light curves, spectra, and Emission Measure Distribution) is described in Sect.~\\ref{sec:analysis}. Results are discussed in Sect.~\\ref{sec:discussion} in the context of coronal activity in this and other late type stars, followed by a summary of the conclusions of the work (Sect.~\\ref{sec:conclusions}). ",
        "conclusions": ""
    },
    "0208/astro-ph0208087_arXiv.txt": {
        "abstract": "We show the apparent redshift-space clustering of galaxies in redshift range of 0.2--0.4 provides surprisingly useful constraints on dark energy component in the universe, because of the right balance between the density of objects and the survey depth. We apply Fisher matrix analysis to the the Luminous Red Galaxies (LRGs) in the Sloan Digital Sky Survey (SDSS), as a concrete example. Possible degeneracies in the evolution of the equation of state (EOS) and the other cosmological parameters are clarified. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0208/astro-ph0208108_arXiv.txt": {
        "abstract": "We review  how some open issues on Astro Particle physics can be studied by space born  experiments  in a complementary way to what is being done at underground and accelerators facilities ",
        "introduction": "This figure shows the conference photo taken at the 1939 Chicago Symposium on Cosmic Rays. Among the partecipants we find a quite exceptional group of physicists: Kohloster, Bethe, Shapiro, Compton, Teller, Eckart, Gouldsmit, Anderson, Oppenheimer,  Hess, Wilson, Rossi, Auger, Heisenberg, Weehler and \\begin{figure} \\begin{center} \\includegraphics[width=1.0\\textwidth]{Figure1.eps} \\end{center} \\label{chicago} \\end{figure}   \\footnote[1]{Invited Talk at the  ESO-CERN-ESA Symposium on Astronomy, Cosmology and Fundamental Physics, March 4-7  2002, Garching, Germany}  many other which are the among the fathers of the modern physics, based on Quantum Mechanics, Elementary Particles and Fundamental Forces. Why Cosmic Rays, discovered by Hess nearly 30 years before were, still in 1939, such an interesting topics for these distinguished scientists?  The answer lies in Table 1. In the years preceding 1937 both the first antiparticle (the positron) and the first unstable elementary particles $(\\mu^{\\pm})$ were discovered in Cosmic Rays. Many more particles were to be discovered during the following years analyzing the Cosmic Radiation, making Cosmic Rays symposia very exciting until at least 1953, when experiments at accelerators  started  to systematically discover new elementary particles while Cosmic Rays experiments suddenly stopped finding them.  \\begin{table} \\caption{Discovery of elementary particles} \\begin{center} \\renewcommand{\\arraystretch}{1.4} \\setlength\\tabcolsep{5pt} \\begin{tabular}{llllll} \\hline\\noalign{\\smallskip} Particle &  Year  & Discoverer (Nobel Prize) & Method \\\\ \\noalign{\\smallskip} \\hline \\noalign{\\smallskip} $e^{-}$ & 1897 & Thomson (1906)& Discharges in gases \\\\ $p$ & 1919 & Rutherford & Natural radioactivity \\\\ $n$ & 1932 & Chadwik (1935) & Natural radioactivity \\\\ $e^{+}$ & 1933 & Anderson (1936) & Cosmic Rays \\\\ $\\mu^{\\pm}$ & 1937 & Neddermeyer, Anderson  & Cosmic Rays \\\\ $\\pi^{\\pm}$ & 1947 & Powell (1950) , Occhialini& Cosmic Rays \\\\ $K^{\\pm}$ & 1949 & Powell (1950) & Cosmic Rays \\\\ $\\pi^{0}$ & 1949 & Bjorklund & Accelerator \\\\ $K^{0}$ & 1951 & Armenteros & Cosmic Rays  \\\\ $\\Lambda^{0}$ & 1951 & Armenteros & Cosmic Rays  \\\\ $\\Delta$ & 1932 & Anderson & Cosmic Rays  \\\\ $\\Xi^{-}$ & 1932 & Armenteros & Cosmic Rays  \\\\ $\\Sigma^{\\pm}$ & 1953 & Bonetti  & Cosmic Rays \\\\ $p^{-}$ & 1955 & Chamberlain, Segre' (1959) & Accelerators \\\\ anything else & 1955 $\\Longrightarrow$ today & various groups & Accelerators \\\\ $ m_\\nu \\neq 0$ & 2000 & KAMIOKANDE  & Cosmic rays \\\\ \\hline \\end{tabular} \\end{center} \\label{Tab1a} \\end{table}  If Cosmic Rays have been instrumental  to give birth to  particle physics during the first half of the past century, starting from the fifties, however,  accelerators have been the tools for the experimental  triumph of the Standard Model of Particle Physics, including   the discovery of the electro-weak bosons at CERN or of the  heavy sixth quark at Fermilab.  During the last ten years, however, the rate of discoveries  at accelerators seems significantly reduced, possibly because of the limited energy scale which can be tested at existing or future facilities. A growing number of physicists is then turning  again to CR with  new experimental techniques aiming to extend  by orders of magnitude the sensitivities reached by past experiments.  In particular a  number of  space born experiments have  been proposed to measure, with unrivalled  accuracy, the composition of primary high energy CR, searching  for new phenomena not accessible to present accelerators.  In this paper we  will review these experiments and their physics potential. \\begin{figure} \\begin{center} \\includegraphics[width=.6\\textwidth]{Figure2.eps} \\end{center} \\caption[]{High Energy Cosmic Rays composition \\cite{choutko}} \\label{hecr} \\end{figure} The paper is organized in two parts. In the first part I will discuss the characteristics of the CR flux, the beam nature gives us, reviewing the status of our knowledge of their energy spectrum and composition. In the second part I will discuss some of the space born experiments planned in the next future which will contribute to the quest for  answers to various unsolved questions in Astro Particle physics. ",
        "conclusions": "One hundred years after their discovery  Cosmic Rays have still an important potential for   new physics.  In order to exploit this potential new, more sensitive experiments are planned which can take advantage of the  unique conditions of space for precision measurement of the primary CR flux. During the current decade these experiments might well deliver exciting surprises in Astro Particle physics, on issues like antimatter, dark matter or other exotic states of matter."
    },
    "0208/astro-ph0208422_arXiv.txt": {
        "abstract": "Stellar streams in galaxy halos are the natural consequence of a history of merging and accretion. We present evidence for a blue tidal stream of {\\it young} stars in the nearest giant elliptical galaxy, \\cena\\ (Centaurus~A).  Using optical {\\it UBVR} color maps, unsharp masking, and adaptive histogram equalization, we detect a blue arc in the northwest portion of the galaxy that traces a partial ellipse with an apocenter of 8~kpc.  We also report the discovery of numerous young star clusters that are associated with the arc.  The brightest of these clusters is spectroscopically confirmed, has an age of $\\sim350$~Myr, and may be a proto-globular cluster. It is likely that this arc, which is distinct from the surrounding shell system and the young jet-related stars in the northeast, is a tidally disrupted stellar stream orbiting the galaxy.  Both the age derived from the integrated optical colors of the stream and its dynamical disruption timescale have values of 200--400~Myr. We propose that this stream of young stars was formed when a dwarf irregular galaxy, or similar sized gas fragment, underwent a tidally triggered burst of star formation as it fell into \\cena\\ and was disrupted $\\sim300$~Myr ago.  The stars and star clusters in this stream will eventually disperse and become part of the main body of \\cena, suggesting that the infall of gas-rich dwarfs plays a role in the building of stellar halos and globular cluster systems. ",
        "introduction": "The accretion and subsequent tidal disruption of low mass galaxies is suspected to be an important agent in the evolution of galaxy halos. Both the outer halo of the Milky Way (Searle \\& Zinn 1978) and the globular cluster systems of nearby ellipticals (C\\^{o}t\\'e, Marzke, \\& West 1998) may have been formed by the late infall of low mass fragments. Hierarchical models of galaxy formation suggest that structure first forms on small scales, and later combines to form larger galaxies (e.g. Klypin et al.\\ 1999). In our own Galaxy, the discovery of tidal streams associated with the Sagittarius dwarf galaxy (Ibata et al.\\ 2001a) as well as tidal tails around Galactic globular clusters (Odenkirchen et al.\\ 2000) give credence to the idea of ``spaghetti halos''---that galaxy halos may be comprised of the many remnants of tidally disrupted dwarf galaxies (Morrison et al.\\ 2000).  The study of streams in galaxy halos is valuable, both for understanding the stellar mass assembly of galaxies, and as probes of the galaxy potential (Johnston, Sackett, \\& Bullock 2001, hereafter JSB01). However, the identification of accretion events that still maintain their spatial coherence (ages of less than a few Gyr) is challenging.  In our own Galaxy, debris streams have been identified using star counts and kinematics (Majewski et al.\\ 1999; Helmi et al.\\ 1999; Newberg \\etal 2002). In external galaxies, these streams can only be detected with deep imaging at surface brightness levels fainter than 27~mag/arcsec$^2$.  Low surface brightness features in many galaxies, which may or may not be debris trails, were detected by Malin \\& Hadley (1997) using photographic amplification techniques. Until now, perhaps the best example of an observed tidal streamer is the one in NGC~5907 (Shang et al.\\ 1998).  There is also strong evidence that tidal streams exist in the halo of M31 (Ibata \\etal 2001b; Choi, Guhathakurta \\& Johnston 2002).  However, in all cases outside the Local Group, distance and low surface brightness preclude any direct study of the stellar populations and kinematics of the accreted stars.  \\begin{figure*}[t] \\epsscale{1.9}\\plotone{pffw.fig1.ps} \\caption{A {\\it BVR} color image of \\cena\\ created by applying our AHE algorithm to our Mosaic observations. This image was processed using the modified adaptive histogram equalization (AHE) method described in section~\\ref{sec:ahe}.  Colors assigned to pixels were determined from flux ratios in the original images. The AHE processing allows one to see that the halo of the galaxy fills most of the frame, extending well beyond the familiar dust lane region along the photometric minor axis.  Small gradients and patchiness in color are due to sky variations and systematic flat-fielding errors at the $\\sim1$--2\\% level.  In all figures, north is up and east is to the left. \\label{bvrimage}} \\end{figure*} ",
        "conclusions": "The detection of tidal debris streams in the halos of nearby galaxies is a burgeoning field that promises to shed new light on galaxy building. Using broadband optical color maps, we have identified one of the first trails of {\\it young} stellar debris---a young blue tidal stream in the halo of the nearest giant elliptical galaxy, \\cena. Associated with this arc are numerous blue star clusters, one of which is both spectroscopically confirmed and massive enough to be a young globular cluster.  The mean age of the unresolved blue light, age of the star cluster, and dynamical disruption timescale all have values of 200--400~Myr. We propose that this stream of young stars was formed when a dwarf irregular galaxy, or similar sized gas fragment, underwent a tidally triggered star formation episode as it fell into \\cena\\ and was disrupted $\\sim 300$~Myr ago. Non-detections to date of neutral or molecular gas in the stream is consistent with the lack of obvious OB associations or resolved \\ion{H}{2} regions, implying that the star formation is no longer ongoing.  The larger merger event that formed the central gas disk has a wide range of published ages, from 200~Myr (Quillen \\etal 1993) to 750~Myr (Sparke 1996).  While it is strongly possible that the formation of the tidal stream occurred in tandem with the larger merger, perhaps as an infalling satellite of the larger galaxy, we emphasize that the stream's morphology, stellar age, and gas content now makes it distinct from both the central disk and the jet-induced star formation in the NE halo.  The stars and star clusters from this tidal stream will eventually disperse into the main body of \\cena, suggesting that the late infall of gas-rich dwarf galaxies may play an important a role in the building of stellar halos. Future spectroscopic observations of the blue star clusters will provide valuable information on the stream's metallicity distribution and kinematic structure."
    },
    "0208/astro-ph0208479_arXiv.txt": {
        "abstract": "{We report the discovery of a new carbon rich white dwarf that was identified during a proper motion survey for cool white dwarfs based on photographic material used for the construction of the Guide Star Catalog II. Its large proper motion ($\\mu\\simeq 0.48$ arcsec/yr) and faint apparent magnitude ($V\\simeq 18.7$) suggest a nearby object of low luminosity. A low-resolution spectrum taken with the William Herschel Telescope clearly shows strong C$_2$ Deslandres-d'Azambuja and Swan bands, which identify the star as a DQ white dwarf. The strength of the Deslandres-d'Azambuja bands and the depression of the continuum in the Swan-band region are signs of enhanced carbon abundance for the given $T_{\\rm eff}$.  Comparison of our spectrophotometric data to published synthetic spectra suggests 6000 K $ < T_{\\rm eff} <$ 8000 K, although further analysis with specialized synthetic models appear necessary to derive both  $T_{\\rm eff}$ and chemical composition. Finally, the range of spatial velocity estimated for this object makes it a likely member of the halo or thick disk population. ",
        "introduction": "Star GSC2U J131147.2+292348 was identified during a proper motion survey for cool halo white dwarfs (WDs) based on photographic material used for the construction of the Second Guide Star Catalogue (GSC-II) (see, e.g., Lasker et al.\\ 1995, McLean et al.\\ 2000). The object is located near the North Galactic Pole (NGP) at $l\\simeq 61^\\circ$, $b \\simeq 85^\\circ$, is fast moving ($\\mu\\simeq 0.48$ arcsec~yr$^{-1}$), and faint ($V\\simeq 18.7$), as expected for a low luminosity object in the solar neighborhood. An accurate check on the SIMBAD database revealed that the star is not in the NLTT catalogue (Luyten 1979) but, quite surprisingly, is listed as a quasar candidate (object OMHR 58793) by Moreau \\& Reboul (1995), who measured an UV excess but did not detect any proper motion. ",
        "conclusions": "We have discovered a new carbon rich white dwarf (DQ), which shows very strong C$_2$ Deslandres-d'Azambuja and Swan bands. To the best of our knowledge, no other object is known today which such a strong simultaneous evidence of the two molecular band systems associated with C$_2$.  Comparisons to published synthetic spectra suggest  6000 $ < T_{\\rm eff} < $ 8000~K, while a black-body fit to the observed fluxes at $\\lambda >$ 7000~\\AA, and to the peaks below $\\sim$ 4100 {\\AA} supports the possibility that $T_{\\rm BB}\\sim$ 6000~K. Therefore, it is evident that the reliable determination of temperature and chemical composition of GSCU J131147.2+292348 must await more detailed atmosphere model calculations. Anyhow, it is likely that the carbon abundance in the atmosphere of this WD is significantly enhanced compared to other known DQ stars of similar temperature.   A photometric distance of 70-90 parsecs has been estimated, which implies a relatively large spatial velocity and makes this new DQ white dwarf a likely member of the halo or thick disk population. Of course, a direct determination of the distance will be the only way to derive model independent absolute magnitude and kinematics for this object."
    },
    "0208/astro-ph0208153_arXiv.txt": {
        "abstract": "A careful analysis of the $HEAO1~A2~2-10~keV$ full-sky map of the X-ray background (XRB) reveals clustering on the scale of several degrees.  After removing the contribution due to beam smearing, the intrinsic clustering of the background is found to be consistent with an auto-correlation function of the form $3.6 \\pm 0.9 \\times 10^{-4} \\theta^{-1}$ where $\\theta$ is measured in degrees. If current AGN models of the hard XRB are reasonable and the cosmological constant-cold dark matter ($\\Lambda CDM$) cosmology is correct, this clustering implies an X-ray bias factor of $b_X \\sim 2$. Combined with the absence of a correlation between the XRB and the cosmic microwave background (CMB), this clustering can be used to limit the presence of an integrated Sachs-Wolfe (ISW) effect and thereby to constrain the value of the cosmological constant, $\\Omega_\\Lambda \\le 0.60$ (95\\% C.L.).  This constraint is inconsistent with much of the $\\Omega_\\Lambda$ parameter space currently favored by other observations. Finally, we marginally detect the dipole moment of the diffuse XRB and find it to be consistent with the dipole due to our motion with respect to the mean rest frame of the XRB. The limit on the amplitude of any intrinsic dipole is $\\delta I_x / I \\le 5 \\times 10^{-3}$ at the 95 \\% C.L. When compared to the local bulk velocity, this limit implies a constraint on the matter density of the universe of ${\\Omega_m}^{0.6}/b_X(0) \\gs 0.24$. ",
        "introduction": "The X-ray background (XRB) was discovered before the cosmic microwave background (CMB), but only now is its origin being fully understood. The hard ($2-10 ~ keV$) XRB has been nearly completely resolved into individual sources;  most of these are active galactic nuclei (AGN), but there is a minor contribution from the hot, intergalactic medium in rich clusters of galaxies (Rosati et al. 2002; Cowie et al. 2002; \\& Mushotzky et al. 2000). In addition, the spectra of these faint X-ray sources are consistent with that of the ``diffuse'' XRB.  If current models of the luminosity functions and evolution of these sources are reasonably correct, then the XRB arises from sources in the redshift range $0 < z < 4$,  making them an important probe of density fluctuations intermediate between relatively nearby galaxy surveys ($z \\ls 0.5$) and the CMB ($z \\sim 1000$).  While there have been several attempts to measure large scale, correlated fluctuations in the hard XRB, these have only yielded upper limits or, at best, marginal detections (e.g. Barcons et al. 2000, Treyer et al. 1998 and references cited therein). On small scales, a recent correlation analysis of 159 sources in the Chandra Deep Field South survey detected significant correlations for separations out to $100~arcsec$ (Giacconi et al. 2001). (At the survey flux level, these sources comprise roughly two thirds of the hard XRB.) On much larger scales, a recent analysis by Scharf et al. (2000) claims a significant detection of large-scale harmonic structure in the XRB with spherical harmonic order $1 \\le \\ell \\le 10$ corresponding to structures on angular scales of $ \\theta \\gs 10^\\circ$ The auto-correlation results we describe here complement this analysis, indicating clustering on angular scales of $3^{\\circ}$ to $10^{\\circ}$, corresponding to harmonic order of $ \\ell \\ls 30$. However, all three detections have relatively low signal to noise and require independent confirmation.  The dipole moment of the XRB has received particular attention, primarily because of its relation to the dipole in the CMB, which is likely due to the Earth's motion with respect to the rest frame of the CMB.  If this is the case, one expects a similar dipole in the XRB with an amplitude that is 3.4 times larger because of the difference in spectral indices of the two backgrounds (Boldt 1987).  In the X-ray literature, this dipole is widely known as the Compton-Getting effect (Compton \\& Getting 1935). In addition, it is quite likely that the XRB has an intrinsic dipole due to the asymmetric distribution in the local matter density that is responsible for the Earth's peculiar motion in the first place.  Searches for both these dipoles have concentrated on the hard XRB, since at lower energies the X-ray sky is dominated by Galactic structure.  There have been several tentative detections of the X-ray dipole (e.g. Scharf et al. (2000)), but these have large uncertainties. A firm detection of an intrinsic dipole or even an upper limit on its presence would provide an important constraint on the inhomogeneity of the local distribution of matter via a less often used tracer of mass and a concomitant constraint on cosmological models (e.g., Lahav, Piran \\& Treyer 1997).  This paper is organized as follows.  In section \\S2 we describe the hard X-ray map used in the analysis, the determination of its effective beam size, and cuts made to remove the foreground contaminants. In section \\S3, we describe the remaining large scale structures in the map and the determination of their amplitudes.  The dipole is of particular interest, and is the topic of section \\S4. The correlation function of the residual map and its implications for intrinsic correlations are discussed in section \\S5.  In section \\S6, we compare our results to previous observations and discuss the cosmological implications of these results in \\S7. ",
        "conclusions": "By carefully reconstructing the HEAO beam and analysing its auto-correlation function, we have been able to confirm the presence of intrinsic clustering in the X-ray background. This gives independent verification of the multipole analysis of Scharf et al. (2000) and the level of clustering we see is comparable. The clustering we see is in excess of that predicted by standard cold dark matter models and indicates that some biasing is needed. The amount of biasing required depends on the cosmological model and on how the bias evolves over time; if the bias is constant, typical models indicate that $b_X \\simeq 2.$  The biases of galaxies, clusters of galaxies, radio sources, and quasars have yet to be adequately characterized and so whether or not the above X-ray bias is excessive is a question that, for the present, remains unanswered.  We have also confirmed, at the 2-3 $\\sigma$ level, the detection of the Compton-Getting dipole in the X-ray background due to the Earth's motion with respect to the rest frame of the CMB. However, we have been unable to confirm the presence of an intrinsic dipole in the XRB and have actually been able to exclude a significant part of the range reported by Scharf et al. (2000).  While our dipole limit is still too small to conflict with any of the favored CDM models, combining our dipole limit with observations of the local bulk flow enable us to constrain $\\Omega_m^{0.6}/b_X(0) > 0.24$.  For constant bias models, this suggests a relatively large matter density, as is also seen in for other velocity studies; however, the uncertainty in this limit is still considerable.  With the observed X-ray clustering, large $\\Lambda-CDM$ models predict a detectable correlation with the cosmic microwave background arising via the integrated Sachs-Wolfe effect. That we have not observed this effect suggests $\\Omega_\\Lambda \\ls 0.60$. This is beginning to conflict with models preferred by a combination of CMB, LSS and SNIA data (e.g., de Bernardis et al. 2000 \\& Bahcall et al. 1999).  This work gives strong motivation for further observations of the large scale structure of the hard X-ray background. Better measurements of the full sky XRB anisotropy are needed, as is more information about the redshift distribution of the X-ray sources. This will be essential for cross correlation with the new CMB data from the MAP satellite and to bridge the gap between the CMB scales and those probed by galaxy surveys such as 2-dF and SDSS."
    },
    "0208/astro-ph0208015_arXiv.txt": {
        "abstract": "Cosmic microwave background (CMB) observations provide in principle a high-precision test of models which are motivated by M theory. We set out the framework of a program to compute the tensor anisotropies in the CMB that are generated in braneworld models. In the simplest approximation, we show the braneworld imprint as a correction to the power spectra for standard temperature and polarization anisotropies. ",
        "introduction": "The early universe provides a testing ground for theories of gravity. The standard cosmological model, based on general relativity with an inflationary era, is very effective in accounting for a broad range of observed features of the universe. However, the lack of a consistent theoretical framework for inflation, together with the ongoing puzzles on the nature of dark matter and dark energy, indicate that cosmology may be probing the limits of validity of general relativity.  M theory is considered to be a promising potential path to quantum gravity. As such, it is an important candidate for cosmological testing. In the absence of a sufficiently general M-theoretic model of cosmology, we can use phenomenological models that share some of the key features of M theory, including branes. In brane cosmology, the observable universe is a 1+3-dimensional ``brane\" surface moving in a higher-dimensional ``bulk\" spacetime. Standard-model fields are confined to the brane, while gravity propagates in the bulk. The simplest, and yet sufficiently general, phenomenological braneworld models are those based on the Randall-Sundrum~II scenario~\\cite{randall}. These models have the additional advantage that they provide a framework for investigating aspects of holography and the AdS/CFT correspondence.  In the generalized RSII models, analyzed via an elegant geometrical approach in Ref.~\\cite{shiromizu}, the bulk is a 1+4-dimensional spacetime, with non-compact extra spatial dimension. What prevents gravity from `leaking' into the extra dimension at low energies is the negative bulk cosmological constant $\\Lambda_5=-6/\\ell^2$, where $\\ell$ is a curvature scale of the bulk. In the weak-field static limit, null results in tests for deviations from Newton's law impose the limit $\\ell \\lesssim 1$~mm. The negative $\\Lambda_5$ is offset by the positive brane tension $\\lambda$, which defines the energy scale dividing low from high energies. The limit $\\ell<1~$mm implies $\\lambda>(100~{\\rm GeV})^4$, and the effective cosmological constant on the brane is \\begin{equation}\\label{cc} \\Lambda=\\frac{1}{2} (\\Lambda_5+\\kappa^2\\lambda)\\,, \\end{equation} where $\\kappa^2= 8\\pi G=8\\pi/M_4^2$, and $M_4\\sim 10^{19}~$GeV is the effective Planck scale on the brane. A further intriguing feature of the braneworld scenario is that, because of the large extra dimensions, the fundamental energy scale of gravity can be dramatically lower than the effective Planck scale on the brane -- as low as $\\sim$~TeV in some scenarios. In generalized RSII models, the fundamental scale is higher, $M_5>10^5~$TeV, and is related to $M_4$ via $M_5^3=M_4^2/\\ell\\,.$  At energies well above the brane tension $\\lambda$, gravity becomes 5-dimensional and significant corrections to general relativity occur. There are also corrections that can operate at low energies, mediated by bulk graviton or Kaluza-Klein (KK) modes. Both types of correction play an important role in tensor perturbations.  The background cosmological dynamics of a Friedmann brane in Schwarzschild-Anti de Sitter (AdS) bulk are well understood~\\cite{binetruy}, including the high-energy modifications to inflation~\\cite{maartens3}. High-energy inflation on the brane generates a zero-mode (4D graviton mode) of tensor perturbations, and stretches it to super-Hubble scales. This zero-mode has the same qualitative features as in general relativity, remaining frozen at constant amplitude while beyond the Hubble horizon, but the overall amplitude is higher~\\cite{langlois5}. The massive KK modes (5D graviton modes) remain in the vacuum state during slow-roll inflation. The evolution of the super-Hubble zero mode is the same as in general relativity, so that high-energy braneworld effects in the early universe serve only to re-scale the amplitude. However, when the zero mode re-enters the Hubble horizon, massive KK modes can be excited. Qualitative arguments~\\cite{hawking,gorbunov1} indicate that this is a very small effect, but it remains to be properly quantified, so that the signature on the CMB may be calculated, and constraints may be imposed on the braneworld parameters.  We develop here a formalism to compute the tensor anisotropies in the CMB, which incorporates the early-universe high-energy braneworld effects, and we carefully delineate what is known on the brane from what is required from bulk equations. Once the 5D solutions are provided, our formalism, with its modified CMB code (based on CAMB~\\cite{lewis,lewis1}), is able to compute these anisotropies. We illustrate this by using a simple approximation to the 5D effects (cf. the analysis of the braneworld scalar Sachs-Wolfe effect in Ref.~\\cite{barrow}). ",
        "conclusions": "As expected, we find that the power spectra are insensitive to high-energy effects, i.e. effectively independent of the brane tension $\\lambda$ : the $\\zeta=0$ curve in Fig.~1 is indistinguishable from that of the general relativity model (both power spectra are identical at the resolution of the plot). For the computations, we have used the lowest value of the brane tension $\\lambda=(100~{\\rm  GeV})^4$, consistent with the tests of Newton's law.   There are three notable effects visible in Figs.~1 and 2, arising as the physical consequences of our approximate model of the KK stress: (i)~the power on large scales reduces with increasing KK parameter $\\zeta$; (ii)~features in the spectrum shift to smaller angular scales with increasing $\\zeta$; and (iii)~the power falls off more rapidly on small scales as $\\zeta$ increases. Neglecting scattering effects, the shear is the only source of linear tensor anisotropies (see e.g.\\ Ref.~\\cite{challinor4}). For $1\\ll l < 60$ the dominant modes to contribute to the temperature $C_l$s are those whose wavelengths subtend an angle $\\sim 1 / l$ when the shear first peaks (around the time of Hubble crossing). The small suppression in the $C_l$s on large scales with increasing $\\zeta$ arises from the reduction in the peak amplitude of the shear at Hubble entry [see Eq.~(\\ref{e:matterlong})], qualitatively interpreted as the loss of energy in the 4D graviton modes to 5D KK modes.  Increasing $\\zeta$ also has the effect of adding a small positive phase shift to the oscillations in the shear on sub-Hubble scales, as shown e.g.\\ by Eq.~(\\ref{e:mattershort}). The delay in the time at which the shear first peaks leads to a small increase in the maximum $l$ for which $l(l+1)C_l$ is approximately constant, as is apparent in Fig.~\\ref{plot2b}. The phase shift of the subsequent peaks in the shear has the effect of shifting the peaks in the tensor $C_l$s to the right. For $l > 60$ the main contribution to the tensor anisotropies at a given scale is localized near last scattering and comes from modes with wavenumber $k \\sim l / \\tau_0$, where $\\tau_0$ is the present conformal time. On these scales the gravity waves have already entered the Hubble radius at last scattering. Such modes are undergoing adiabatic damping by the expansion and this results in the sharp decrease in the anisotropies on small scales. Increasing the KK parameter $\\zeta$ effectively produces more adiabatic damping and hence a sharper fall off of power. The transition to a slower fall off in the $C_l$s at $l \\sim 200$ is due to the weaker dependence of the amplitude of the shear on wavenumber at last scattering for modes that have entered the Hubble radius during radiation domination~\\cite{starobinsky}. [The asymptotic expansion of Eq.~(\\ref{e:matterlong}) gives the shear amplitude $\\propto k^{-(2 + \\zeta)}$ at fixed $\\tau$, whereas for modes that were sub-Hubble at matter-radiation equality Eq.~(\\ref{e:mattershort}) gives the amplitude $\\propto k^{-(1 + \\zeta/2)}$.]  Similar comments apply to the tensor electric polarization $C^E_l$, shown in Fig.~\\ref{plot3}. As with the temperature anisotropies, we see the same shifting of features to the right and increase in damping on small scales. Since polarization is only generated at last scattering (except for the feature at very low $l$ that arises from scattering at reionization, with an assumed optical depth $\\tau_C=0.03$), the large-scale polarization is suppressed, since the shear (and hence the temperature quadrupole at last scattering) is small for super-Hubble modes. In matter domination the large-scale shear is $\\sigma_k = - k\\tau / (5 + 2\\zeta)$; the reduction in the magnitude of the shear with increasing KK parameter $\\zeta$ is clearly visible in the large-angle polarization. The braneworld modification to the tensor magnetic polarization $C^B_l$ has the same qualitative features as in the electric case.  In principle, observations can constrain the KK parameter $\\zeta$, which controls the generation of 5D modes within our simplified local approximation, Eq.~(\\ref{e:Pansatz1}). The other braneworld parameter $\\lambda$, the brane tension, is not constrained within our approximation. In practice, the tensor power spectra have not been measured, and the prospect of useful data is still some way off. What is more important is the theoretical task of improving on the simplified local approximation we have introduced. This approximation has allowed us to encode aspects of the qualitative features of braneworld tensor anisotropies, which we expect to survive in modified form within more realistic approximations. However, a proper understanding of braneworld effects must incorporate the nonlocal nature of the KK graviton modes, as reflected in the general form of Eq.~$\\eqref{e:soln}$. It is also necessary to investigate the scalar anisotropies, which have a dominant contribution to the measured power spectra. These may reveal new braneworld imprints that are more amenable to observational testing."
    },
    "0208/hep-ph0208070_arXiv.txt": {
        "abstract": "\\noindent We present the first study of parametric resonance in quantum field theory from a complete next-to-leading order calculation in a $1/N$-expansion of the 2PI effective action, which includes scattering and memory effects. We present a complete numerical solution for an $O(N)$-symmetric scalar theory and provide an approximate analytic description of the nonlinear dynamics in the entire amplification range. We find that the classical resonant amplification at early times is followed by a collective amplification regime with explosive particle production in a broad momentum range, which is not accessible in a leading-order calculation. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0208/astro-ph0208290_arXiv.txt": {
        "abstract": "We have studied the O and Ne absorption features in the X-ray spectrum of Cyg X-2 observed with the Chandra LETG.  The O absorption edge is represented by the sum of three absorption-edge components within the limit of the energy resolution and the photon counting statistics. Two of them are due to the atomic O; their energies correspond to two distinct spin states of photo-ionized O atoms.  The remaining edge component is considered to represent compound forms of oxide dust grains. Since Cyg X-2 is about 1.4 kpc above the galactic disk, the H column densities can be determined by radio (21~cm and CO emission line) and ${\\rm H}_\\alpha$ observations with relatively small uncertainties.  Thus the O abundance relative to H can be determined from the absorption edges.  We found that the dust scattering can affect the apparent depth of the edge of the compound forms. We determined the amplitude of the effect, which we consider is the largest possible correction factor.  The ratio of column densities of O in atomic to compound forms and the O total abundance were respectively determined to be in the range $1.7^{+3.0}_{-0.9}$ to $2.8^{+5.1}_{-1.5}$ (ratio), and $0.63\\pm 0.12$ solar to $0.74 \\pm 0.14$ solar (total), taking into account the uncertainties in the dust-scattering correction and in the ionized H column density. We also determined the Ne abundance from the absorption edge to be $0.75 \\pm 0.20$ solar.  These abundance values are smaller than the widely-used solar values but consistent with the latest estimates of solar abundance. ",
        "introduction": "The metal abundance in the interstellar medium (ISM) is an important parameter for the understanding of the chemical evolution of the Universe.  However, large uncertainties remain in our knowledge of this abundance.  The solar abundance is sometimes regarded as the average of our Galaxy.  However, UV absorption lines due to interstellar gas and optical absorption lines in the atmosphere of young B-stars have indicated that the metal abundance in the ISM of our Galaxy is on average about two thirds of the solar abundance \\citep{Savage_Sembach_1996, Snow_Witt_1996}.  However, both methods contain uncertainties because the former method is only sensitive to matter in atomic forms, while the abundance of B-star atmospheres may be different from the ISM.  Indeed, \\citet{Sofia_Meyer_2001} recently claimed that B-star abundances are lower than the ISM abundance and that the solar abundance is close to the ISM abundance. Moreover, the O solar abundance itself is not determined well.  Recent values by \\citet{Grevesse_Sauval_1998} and \\citet{Holweger_2001} are significantly lower than the values in the widely used table of \\citet{Anders-Grevesse_1989}.  High resolution X-ray spectroscopy of absorption features due to the interstellar matter is a powerful tool to measure the amount of interstellar metal.  It is sensitive to both the atomic (gas) and compound (molecular and dust grain) forms and can distinguish them by the chemical shifts.  There are absorption lines and edges of many elements in the X-ray energy range.  We can determine the column densities of these elements in different forms separately,  integrating towards the X-ray source.  However, the observation requires both high energy resolution and good statistics.  \\citet{Schatternburg_Canizares_1986} determined the O column density towards the Crab nebula with the transmission grating on the Einstein observatory.  They could not separate atomic and compound forms.  Also they were not able to determine the O abundance since the H column density towards the Crab nebula was not known.  \\cite{Paerels_etal_2001} analyzed the O and Ne absorption structures of the X-ray binary X0614+091 observed with the low energy transmission grating (LETG) on the Chandra observatory. They successfully separated the absorption features of O in atomic and compound forms.  They also determined the abundance ratio of Ne to O.  However, they could not determine the O abundance because the H column density toward the source is not known. \\cite{Schulz_etal_2002} determined the column densities of O, Ne, Mg, Si, and Fe towards Cyg X-1 with the Chandra high energy transmission grating (HETG) data.  They discussed the abundances of O, Ne, and Fe.  However, it was necessary to assume the abundances of all other elements to be solar, in order to estimate the H column density towards Cyg X-1 from the X-ray spectrum.  In this paper we analyze the spectrum of Cyg X-2 obtained with the Chandra LETG and HRC (High Resolution Camera)-I. Cyg X-2 is a Low-Mass X-ray Binary (LMXB) located at ($l$,$b$)= ($87.33^\\circ$,$-11.32^\\circ$) and its distance is determined to be 7.2~kpc from the optical counterpart and X-ray burst with mass ejection \\citep{Cowley_Crampton_Hutchings_1979, Smale_1998, Orosz_Kuulkers_1999}. \\cite{Smale_1998} claimed a distance of 11.6 kpc, however, this is because he assumed the Eddington luminosity for a hydrogen rich atmosphere.  If we assume a helium rich atmosphere, which is more realistic for type-I bursts \\citep{Lewin_etal_1995}, the distance is consistent with 7.2~kpc.  Adopting this distance, Cyg X-2 is located $\\sim$ 1.4~kpc above the Galactic disk.  Thus the H column density towards Cyg X-2 can be estimated from 21 cm (\\ion{H}{1}), CO (${\\rm H}_2$), and ${\\rm H}_{\\alpha}$ (\\ion{H}{2}) observations with relatively small errors.  Because of the featureless X-ray spectrum of the LMXB, the high brightness, and the well-determined H column density to the source, Cyg X-2 is one of the best X-ray sources for the study of the interstellar abundance with X-ray absorption.  In the next section, we will first discuss the H column density towards Cyg X-2. Then in the following sections we will show the observational results and discuss the O and Ne abundances and the uncertainties of the present results.  Throughout this paper, we quote single parameter errors at the 90~\\% confidence level unless otherwise specified. ",
        "conclusions": "We have analyzed the O and Ne absorption structures in the X-ray spectrum of Cyg X-2 observed with the Chandra LETG/HRC-I.  The O edge is represented by the sum of three edge components within the limit of present statistics and energy resolution.  Two edge components represent absorptions by atomic O; their edge energies are consistent with the theoretical values corresponding to the two distinct final O states of different spin numbers.  The edge energy of the remaining edge is lower than the other two, and is likely to represent O in compound forms.  From the depth of the edges, we estimated the O column densities with atomic and compound forms separately.  We tried a four-edge component model in the spectral fits.  We find the sum of the optical depths of all the edge components does not change significantly from three to four edge models.  Thus we conclude that the total O column density does not depend significantly on the number of edge components assumed in the spectral models. We estimated the atomic, compound, and total O abundances to be $0.47 \\pm 0.16$, $0.21 \\pm 0.09$, and $0.68 \\pm 0.13$ times the solar abundance of \\cite{Anders-Grevesse_1989} respectively.  The Ne edge is represented with a single edge model and the Ne abundance is estimated to be $0.75\\pm0.20$ solar.  In the errors of abundance quoted above, we included the statistical errors of the spectral fits, and the systematic errors of $\\NH$ estimated from the spatial fluctuation of 21cm and ${\\rm H}_\\alpha$ emissions and the upper limit of CO emission.  There is an additional systematic error in the abundance of O in compound forms, arising from the uncertainty in the spatial distribution of the WIM.  If we consider the extreme case in which its volume filling factor is 1, the O abundance reduces to $0.16 \\pm 0.07$ (compound form) and $0.63\\pm0.12$ (total).    It is very unlikely that the O and Ne absorption edges are associated with the circumstellar matter in the Cygnus X-2 binary system. Suppose that the absorption column of $\\NH \\sim 2\\times10^{21} {\\rm cm}^{-2}$ is located in a distance $r$ from the X-ray source. The ionization parameter, $\\xi = L_{\\rm X}/(n r^2) \\sim L_{\\rm X}/(\\NH r)$ must be smaller than $\\sim 10$ so that O is neutral \\citep{Kallman_McCray_1982}. With $L_{\\rm X} = 10^{38}$ erg sec$^{-1}$, we find $r\\gtrsim 10^{16}$ cm.  This is much larger than the size of the binary system \\citep[$\\sim 10^{12}$ cm,][]{Cowley_Crampton_Hutchings_1979} This is consistent with the facts that all emission and absorption line features found in LMXBs are from highly ionized atoms \\citep[e.g.,][]{Asai_etal_2000, Cottam_etal_2001} and that low-ionization emission lines from the Cyg X-2 system are from near the companion star \\citep{Cowley_Crampton_Hutchings_1979}.   About one third of O contributing to the absorption edges is in compound form.  Dust grains are considered to contribute to this component.  In such a situation, the optical depth for dust scattering of X-rays is not negligible compared to the absorption if we assume a typical dust radius of $\\sim 0.1~\\mu$m.  Scattered X-rays form an extended halo on the few arcminute scale \\citep{Hayakawa_1970, Mauche_Gorenstein_1986, Predehl_Schmitt_1995}. Since the scattering cross section is dependent on the X-ray energy, the energy spectrum of the unscattered X-rays is modified by scattering.  However, if multiple scattering is negligible and if the spatial distribution of dust is uniform over the spatial scale corresponding to the dust halo, the energy spectrum of total photons, i.e., the sum of the unscattered core and the scattered halo, is not modified by dust scattering.  This is not the case for the present energy spectrum obtained with the LETG, since we derived the energy spectrum only from the central $\\sim$ 1 arc second of the X-ray image.  Because the scattering cross sections show anomalous features around absorption edges \\citep{Mitsuda_etal_1990}, the dust scattering will affect the estimation of the absorption column density; the scattering cross section is {\\it smaller} at the energy just above the edge (at the wavelength shorter than the edge), while the absorption cross section is {\\it larger}.  This leads to an underestimation of the O absorption column density.  We estimated the amplitude of this effect in the following way.  The total scattering cross section of an X-ray photon of a wavelength $\\lambda$ with a dust grain of a radius $a$ is given by \\[ \\sigma_\\mathrm{dust,sc}=2\\pi a^2 \\left(\\frac{2\\pi a}{\\lambda}\\right)^2 \\left|\\frac{N r_0 \\lambda^2}{2\\pi} (f_1+{\\rm i}f_2)\\right|^2, \\] under the Rayleigh-Gans approximation \\citep{van_de_Hulst_1957}, where $N$ is the number of atoms per unit volume, $r_0$ is the classical electron radius, and $f_1$ and $f_2$ are the atomic scattering factors \\citep{Henke_Gullikson_Davis_1993}. The Rayleigh-Gans approximation is valid for $(2\\pi a/\\lambda) | N r_0 \\lambda^2/(2\\pi) (f_1+{\\rm i}f_2) | \\ll 1$.  Because the factor $|f_1 + {\\rm i}f_2|$ becomes small near the absorption edge, the dust scattering cross section is reduced at the edge.  Assuming typical chemical compositions and densities of dust grains ($\\mathrm{FeSiO_4}$, $\\mathrm{FeSiO_3}$, $\\mathrm{Mg_2SiO_4}$ and $\\mathrm{MgSiO_3}$), we estimate the reduction of the scattering cross section at the O edge to be in the range, \\[ \\Delta \\sigma_\\mathrm{dust,sc}=(0.53-0.93)\\times10^{-10} \\left(\\frac{a}{0.1~\\mathrm{\\mu m}}\\right)^{4}~\\mathrm{cm}^2. \\] Assuming the number density of dust to be proportional to $\\propto a^{-3.5}$ with a cut off at $a_\\mathrm{max}$ \\citep{Mathis_Rumpl_Nordsieck_1977}, we calculate the reduction of the scattering cross section per O atom, \\[ \\Delta \\sigma_\\mathrm{sc}=(0.74-1.44)\\times10^{-19} \\left(\\frac{a_\\mathrm{max}}{0.1~\\mathrm{\\mu m}}\\right) ~ \\mathrm{cm}^2. \\] Thus the reduction of scattering cross section can be a few tens of percent of the absorption cross section, $\\sigma_\\mathrm{abs}=4.98\\times10^{-19}~\\mathrm{cm^2}$.  The maximum grain radius contributing to X-ray scatterings has been estimated from the spatial size of dust scattering halos. \\citet{Mauche_Gorenstein_1986} and \\citet{Mitsuda_etal_1990} suggested $a_\\mathrm{max} = 0.05 - 0.1~\\mathrm{\\mu m}$, but recently \\cite{Witt_Smith_Dwek_2001} suggested the presence of grains with radii as large as $0.5\\mathrm{\\mu m}$.  Adopting $a_\\mathrm{max} = 0.1~\\mathrm{\\mu m}$, and assuming that O of compound forms are all in dust grains, the column density of O of compound forms increases from $5.2\\times10^{17}~\\mathrm{cm^{-2}}$ to $(6.1 - 7.3) \\times 10^{17}~\\mathrm{cm^{-2}}$.  At the O edge energy, the Rayleigh-Gans approximation is valid only for $a_\\mathrm{max}~\\lesssim 0.1~\\mathrm{\\mu m}$ and the scattering cross section saturates at $a_\\mathrm{max}\\gtrsim 0.1~\\mathrm{\\mu m}$ \\citep{Alcock_Hatchett_1978,Smith_Dwek_1998}.  Thus the correction factor does not increase for larger grains and the correction factor we applied above can be regarded as the maximum correction.  The O abundance values calculated for four different cases (with and without correction for dust scattering, and two different values of the volume filling factor of the WIM) are summarized in \\Tab\\ref{tab:summary}.  The total O abundance is between $0.63\\pm 0.12$ and $0.74 \\pm 0.14$ times solar. On the other hand the Ne abundance is $0.75 \\pm 0.20$.  Our results are more consistent with the recent solar abundances than the most widely used `old' values. For example , the O and Ne solar abundances by \\cite{Holweger_2001} are 0.65 and 0.81 times the values in \\cite{Anders-Grevesse_1989}, respectively.  Another important parameter we can estimate from the present X-ray observation is the gas to dust ratio of the O column densities, which is estimated to be $2.2^{+3.9}_{-1.1}$, $2.8^{+5.1}_{-1.5}$, $1.7^{+3.0}_{-0.9}$, and $2.2^{+4.0}_{-1.1}$ for cases 1 to 4 of \\Tab\\ref{tab:summary}, respectively.  This should be compared with the value estimated from the gas abundance from interstellar UV absorption lines and the assumption that the total abundance is solar.  Adopting the O gas abundance of \\cite{Cartledge_Meyer_Lauroesch_2001} and solar abundances by \\cite{Anders-Grevesse_1989} and \\cite{Holweger_2001}, we obtain 0.7 and 1.7, respectively. Our result is again consistent with the latest solar abundance value.  For O, the present X-ray result contains an additional uncertainty due to the correction for dust scattering.  We can avoid this if we include the spectrum from the scattering halo in the analysis.  This requires non-dispersive high resolution spectrometers such as microcalorimeters, which will be used on board future X-ray missions, such as Astro-E2."
    },
    "0208/astro-ph0208545_arXiv.txt": {
        "abstract": "{ We report the discovery of a  % new, bright ($V\\sim 17^{\\rm m}$) AM Her system as the optical counterpart of the soft ROSAT All-Sky-Survey source \\rxj\\ (= 1RXS J184659.4+553834). Optical photometric and spectroscopic follow-up observations reveal a single period of 128.7 min, consistent with a high degree of spin-orbit synchronization, and a low polar field strength ($B<20$~MG) of the primary accretion region. The system was observed in optical intermediate and high states that differ by about 1 mag. These  brightness variations were accompanied by a correlated change of the optical light curve, which we interpret as a switch between one- and two-pole accretion. This explanation is also supported by the X-ray light curves, which at two different epochs display emission from two equally bright accretion regions separated by $\\sim 160\\degs$. Both spots possess distinct spectral X-ray properties as seen from the X-ray hardness ratio, where the secondary accretion region appears significantly softer, thus probably indicating a higher field strength compared to the primary region. In all ROSAT pointings  a deep dip is present during the primary flux maxima, very likely caused by absorption in one of the accretion streams. ",
        "introduction": "AM Her type variables  are a subgroup of cataclysmic variables in which the magnetic field of the white dwarf controls the geometry of the material flow between the main-sequence donor and the white dwarf primary (see e.g. Warner 1995 for a detailed review). The inflow of matter along the magnetic field lines (of one or occasionally also two magnetic poles) is decelerated above the white dwarf surface producing a shock front. This region is thought to emit hard X-rays (usually modelled in terms of thermal bremsstrahlung of 10--20 keV) and polarized cyclotron radiation (hence these systems are also named polars) in the IR to UV range. In addition, a strong soft component has been frequently observed from polars that is thought to arise from the heated accretion pole (usually modelled in terms of a blackbody of 20--50 eV).  \\begin{figure}[t] \\fbox{\\includegraphics[clip,width=0.95\\columnwidth]{f_rx1846_fc_newinset.ps}}\\par \\caption[fc]{White-light CCD image of \\rxj\\ obtained with the AIP 70 cm telescope. North is top and East to the left. The size of the field is approximately 8\\amin$\\times$8\\amin. The inset at the lower right corner shows a 30\\asec$\\times$30\\asec\\ blow up taken from the digitized Palomar Observatory Sky Survey (POSS) 2 together with the 2$\\sigma$ X-ray error circle derived from the HRI pointing. The bright star in the upper left of the blow up is star A. The cataclysmic variable \\rxj\\ was certainly in an optically low state during the POSS exposure. Its optical position was determined to $\\alpha_{2000}$ = 18\\h46\\m58\\fss9 and $\\delta_{2000}$ = 55\\grad 38\\amin 29\\asec\\ ($\\pm$1\\asec). } \\label{fc} \\end{figure}  It is this soft X-ray component which has led to the discovery of a few dozen new polars by ROSAT observations over the last decade, most notably the ROSAT all-sky survey (Beuermann \\& Burwitz 1995). The source described here has been discovered as a result of a systematic survey for supersoft X-ray sources from the all-sky survey data (Greiner 1996 for details of this survey) which revealed a large number of CVs and single white dwarfs. Other confirmed polars identified from this sample include V844 Her = RX\\,J1802.1+1804 (Greiner, Remillard and Motch 1995, 1998), RS Cae = RX\\,J0453.4--4213 (Burwitz \\etal\\ 1996) and V1007 Her = RX\\,J1724.0+4114 (Greiner, Schwarz and Wenzel 1998).  In this paper we present photometric, spectroscopic and X-ray observations (summarized in Table~\\ref{log} and \\ref{xlog}) that led to the discovery of a new polar, \\rxj\\ (henceforth referred to as \\117). ",
        "conclusions": "\\117 shows all major hallmarks of a cataclysmic variable of the AM Herculis subclass: an emission line spectrum with abundant high ionisation species like \\ion{He}{II} $\\lambda 4686$, and the Bowen blend, optical and X-ray light curves strongly modulated on the spin period due to the self eclipse of the primary accretion region, and a very red cyclotron spectrum from that primary accretion region, indicating a polar field strength in the range of 15-20 MG. At most occasions the X-ray light curves reveal sharp X-ray dips, likely caused by absorption within a focused accretion funnel. With an orbital period of $P=128.7$~min \\117\\ is another magnetic CV which populates the lower edge of the 2--3 hr CV period gap.  \\subsection{Accretion mode changes} The rather large  body of observations available for \\117\\ reveals frequent switches between one-pole accretion and additional activity from a second,  equally bright, pole. Although accretion onto a second pole has been seen in  a few polars, e.g. in DP Leo (Cropper et al. 1990), VV Pup (Wickramasinghe et al. 1989) or UZ For (Schwope et al. 1990), in most cases the activity from the secondary pole is at least one order of magnitude lower compared to the primary one. Remarkable exceptions include the 'anomalous' or 'reversed' state of  AM Her (Heise et al. 1985), which appeared to be a singular event, and the low accretion rate polar HS1023+39 which permanently accretes onto two poles having similar luminosities (Reimers et al. 1999, Schwarz et al. 2001). So far the only polar, beside \\117, which regularly changes between the single and two-pole accretion mode is the period-gap system QS Tel (Schwope et al. 1995a, Rosen et al. 1996, 2001).  There are two alternative mechanisms that can trigger the accretion mode changes observed in \\117. Firstly, it can be due to a slight asynchronism of the spin of the white dwarf compared to the orbital motion, which in the case of an oblique dipole results in a constantly changing orientation of the magnetic field with respect to the infalling gas stream. Matter can then be channeled along different field lines, possibly feeding opposite magnetic poles. While this so-called 'pole switching' is believed to operate in the four known asynchronous AM Herculis systems (Campbell \\& Schwope 1999), detailed studies indicate that the mass exchange might be more complicated. For example in CD Ind, the only asynchronous polar  with known accretion geometry, one primary pole dominates in the light curves for a large fraction of the beat cycle (Ramsay et al. 2000). Thus,  the fact that \\117 does not switch between two magnetic poles but has one primary pole  might still be consistent with a possible asynchronism. A more crucial test of this scenario is the requirement that the accretion mode changes should occur strictly periodically over the beat cycle, a possibility which can not yet be rejected for \\117\\ on the basis of the available data.  The second mechanism would be variations of the total mass accretion rate from the secondary star, which are commonly seen as low states in disk-less magnetic CVs. These variations are fast, and reoccur aperiodically on timescales of months to years. The working hypothesis that explains most of the observed properties invokes the blocking of the $L_{1}$ point by starspots on the surface of the secondary star as proposed by Livio \\& Pringle (1994). As the total mass accretion rate increases, the ram  pressure ($\\rho v$) increases in the stream,  leading to a deeper penetration of the ballistic mass stream into the white dwarf's magnetosphere. Possibly the stream will then connect to field lines which can also feed the less favoured  pole. By now,  all observed two-pole accretion states in \\117\\ have been related to epochs of enhanced total brightness in the optical and X-ray, thus indicating an increase of the mass accretion rate. This correlation is so far the strongest argument in favour of the above picture. If the pressure balance relation holds, the effective magnetospheric radius at which matter is controlled by the magnetic field would scale with the mass accretion rate with $r_{\\rm mag } \\propto \\dot{M}^{-2/7}$. The brightening of the total X-ray flux by a factor of $\\sim 2$ in the two-pole state implies that the accretion rate must have roughly doubled at that epoch. The corresponding reduction of the magnetospheric radius would then be only of the order of 20\\%, probably not enough to reach the less favoured magnetic pole in the case of the standard field orientation seen in AM Herculis binaries with an azimuthal angle of $\\psi = 45\\degr$ (see Cropper 1988). Indeed, long-term X-ray monitoring of the eclipsing polar HU Aqr (Schwope et al. 2001) through different accretion states revealed only a moderate shift of the stagnation region of only 30\\degr\\  as the accretion rate varied by a factor of 40. Thus,  in addition to the accretion rate changes, a special field geometry is possibly required to facilitate the frequent accretion mode changes in \\117.   Both accretion regions emit approximately the same accretion luminosities of $L_{\\rm X}$ = 7$\\times$10$^{31}$ (D / 400 pc)$^2$ erg/s, thus receiving quite similar mass accretion rates. However, the contributions from the hot thermal plasma,  emitted as hard X-rays,  and reprocessed radiation from the white dwarf emitted as a blackbody in the soft X-ray band is remarkably different in both spots. While the energy balance of the primary region is in agreement with that of the standard shock model (Lamb \\& Masters 1979), a substantial soft X-ray excess is found for the secondary spot. This violation has been theoretically explained in terms of 'blobby' accretion of high $\\dot{m}$~$(> 30$ g~cm$^{-2}$~s$^{-1})$ material buried below the surface of the white dwarf (Kuijpers \\& Pringle 1982). Observationally, a strong correlation between the magnetic field strength and the softening of the X-ray spectra has been found (Beuermann \\& Schwope 1994) from a comparative study of ROSAT data. Probably two different processes are involved, one increasing the soft X-ray output and the other lowering the hard X-ray component, both changing the energy balance as observed: Firstly,  the specific mass accretion rate in the accretion region will be higher for material which travels further along converging field lines, and was consequently coupled at larger magnetospheric radii. Since this quantity is directly related to the surface field strength,  for high $B$ systems the fraction of high $\\dot{m}$ material  will be larger on average, shifting the energy balance towards the reprocessed blackbody component (Beuermann 1998). Secondly, for the low $\\dot{m}$ part of the material, bremsstrahlung radiation will be suppressed due to the dominance of cyclotron cooling in the case of an increased magnetic field in the post shock plasma (Woelk \\& Beuermann 1996).  Adopting this picture for \\117, the X-ray softness of the secondary spot would be the consequence of  a higher field strength of $\\sim 30$ MG at this pole. Correspondingly, the energy balance of the primary region would imply a lower field strength of $\\sim 14$ MG, in agreement with the value derived from the cyclotron spectroscopy. As in the six polars with field measurements available for both magnetic poles, the more active pole in \\117\\ would be the one with the lower field strength.  \\subsection{Accretion geometry}  A special geometry of the mass transfer  might foster the drastic variations of the accretion modes observed in \\117. For example, an orientation  where the dipolar axis is perpendicular to the infalling stream would equally favour both poles. However, this is difficult to verify given the as yet unknown position of the secondary star with respect to the accretion regions, and the wide range of possible inclinations ($i\\leq 70\\degs$). There is a strong indication that the primary accretion region is located at the lower hemisphere of the white dwarf $(\\beta_{1} > 90\\degs)$, from the duration of the faint phase ($\\gamma > 0.5$) measured at some occasions in the optical and X-ray. Estimates for the secondary accretion region are much less constrained due to the insufficient phase coverage the of June 1992 observation and the overlapping visibility of both regions around $\\phi \\sim 0.5$. Using the short duration of the supersoft emission ($\\Delta \\phi \\sim 0.3$) as a measure would place this spot on the lower hemisphere of the white dwarf, too.  Both accretion spots are separated in azimuth  by $160\\degs$, since the two bright phases are centered at $\\phi_{1} = 0.75$ and $\\phi_{2} = 0.2$. Due to the finite coupling radius the accretion regions are likely to be offset from the magnetic pole by $\\alpha \\sim 10\\degs$. Therefore, the measured angular separation is consistent with both spots accreting via one closed field line of a dipole field. A geometry that can accomplish the all above constraints would imply a magnetic field axis inclined into the orbital plane.  Another puzzling question concerns the accretion stream dip and its relation to the two accretion regions. As the azimuth of the stream dip and the primary spot are approximately the same, one would naively expect that the absorbing matter is transferred to the primary accretion region.  This contradicts the condition for a stream occultation $i\\ge \\beta + \\alpha$ given the primary spots colatitude $\\beta_{1} \\ge 90\\degs$. The above conflict could be circumvented  if one allows  for a possible vertical extent of the ballistic part of the stream and the stagnation region, an explanation proposed to explain a partial stream dip observed in the extreme UV light curves of  the high inclination polar UZ For (Warren et al. 1995). It is not clear how far this is viable for the case of \\117, where the absorbing material is denser and must extend much further above the orbital plane in order to produce a dip at the given lower inclination.  The competing scenario, absorption by material channeled to a second pole in the upper hemisphere, would lead to the occultation of X-rays from the primary region for a wide range of possible stream geometries. Such a view is also supported by the putative correlation of the density of the stream with the respective activity from the secondary accretion region: no stream dip is observed when the faint phase emission is zero (ROSAT survey), a mildly dense dip ($N_{\\rm H} \\sim 10^{21}$~cm$^{-2}$) for low faint phase emission (Sep.~93), and a much denser dip in the case of obvious two-pole accretion (Jul.~92). Further detailed studies have to show whether the footpoint of such a stream trajectory can be reconciled with the geometrical constraints for the secondary pole, namely the azimuthal separation to the primary spot (which would place it on the opposite side of the white dwarf) and a high colatitude far from the white dwarf rotational axis."
    },
    "0208/astro-ph0208403_arXiv.txt": {
        "abstract": "The DAMA annual modulation signature, interpreted as evidence for a spin-independent WIMP coupling, seems in conflict with null results from CDMS. However, in models of ``inelastic dark matter'',  the experiments are compatible. Inelastic dark matter can arise in supersymmetric theories as the real component of a sneutrino mixed with a singlet scalar. In contrast with ordinary sneutrino dark matter, such particles can satisfy all experimental constraints while giving the appropriate relic abundance. We discuss the modifications to the signal seen at DAMA, in particular noting the strong suppression of low energy events in both modulated and unmodulated components. We discuss future experiments, with emphasis on distinguishing inelastic dark matter from ordinary dark matter, and stressing the significance of experiments with heavy target nuclei, such as xenon and tungsten. \\vspace{1pc} ",
        "introduction": "Recently the DAMA collaboration has reported evidence from four years of study of an annual modulation signal consistent with a WIMP \\cite{Bernabei:2000qi}. The data satisfy the six requirements of a WIMP signal in that it is a modulation of the single hit rate, contained entirely in the low energy bins, with the appropriate shape, phase, period, and amplitude that are expected. Studies of possible systematic errors have turned up no candidates to explain the modulation \\cite{Bernabei:2000ew}.  At the same time, the CDMS collaboration has reported no evidence for a WIMP signal in their experiment \\cite{Abusaidi:2000wg,Abrams:2002nb}, claiming a conflict with DAMA as interpreted as a WIMP coupled with spin-independent (SI) interactions at $99.8\\%$, even given no assumptions about  the background \\cite{Abrams:2002nb}.  Given this conflict, and the absence of a clear source of systematic error, it is worthwhile to consider alternative forms of dark matter, with different interaction properties from the neutralino. The neutralino is a very appealing candidate, both because it arises in a motivated theory (supersymmetry) and because it naturally has the right relic abundance. In looking for a alternative theory, we would like to retain both of these features, while still explaining the discrepancy between CDMS and DAMA.  \\subsection{Inelastic Dark Matter} Let us consider ``inelastic dark matter'' (iDM) \\cite{Smith:2001hy}. We assume the presence of two particles $\\chi_1$ and $\\chi_2$, such that $\\delta = m_2-m_1 >0$. We further assume that $\\chi_1$ constitutes the dark matter in the galaxy, and that $\\chi_1$ can only scatter off of nuclei by making an inelastic transition to $\\chi_2$. That is, the allowed scattering is $\\chi_1 N\\rightarrow \\chi_2 N$. Let us emphasize that the inelasticity is {\\em not due to a nuclear transition, but instead in the final state of the dark matter particle, itself.}  It is quite simple to construct models of this type, but we will address this later. For now, let us focus on the kinematical changes which occur in a scattering experiment. The central difference from ordinary elastic scattering is the requirement on the velocity $\\beta$ for a scattering to occur: \\begin{equation} \\beta^2 > \\frac{2 \\delta (m_N+ m_{\\chi})}{m_N m_{\\chi}}, \\ee where $m_N$ is the mass of the target nucleus. The important point is that the right hand side is an {\\em increasing} (more severe) function for {\\em lighter} target nuclei. In the halo there is a distribution of velocities. Because the DAMA target is NaI ($A_I=127$) while the CDMS target is Ge ($A_{Ge}=73$), a larger range of velocities will be visible at DAMA than at CDMS (figure \\ref{fig:dist}). As a consequence, the sensitivity of DAMA relative to CDMS is enhanced when compared with the elastic case.  \\begin{figure} \\psfig{file=range.eps,width=0.4\\textwidth} \\caption{Tail of the distribution of particles in the halo as a function of velocity. Many particles are visible to DAMA but not to CDMS. \\vskip -0.2in} \\label{fig:dist} \\end{figure}  A second effect enhances the sensitivity of DAMA relative to CDMS. Because we are sensitive to the number of particles above some velocity $\\beta_{min}$, as we orbit the sun, the number of particles which can scatter can change considerably. As a consequence, the modulated amplitude can be considerably larger than $7\\%$ of the unmodulated amplitude, as is the limit in the usual elastic case. In figure \\ref{fig:modamp}, we see this enhancement.  \\begin{figure} \\psfig{file=modulation.eps,width=0.4\\textwidth} \\vskip-0.2in \\caption{DAMA signal for $m_\\chi=50\\gev$ and elastic (dashed) and $\\delta=100 \\kev$  (solid). \\vskip-0.2in} \\label{fig:modamp} \\end{figure}   An interesting result of this involves the interpretation of the DAMA data. It is often noted (e.g., \\cite{Sadoulet:1999rq}) that the DAMA best fit point has an unmodulated signal comparable to their measured background in certain low energy bins. Here, this is no longer the case in general. In fact, the suppression of the DAMA unmodulated signal is most pronounced in the low energy bins in question, as we will discuss in the next section.  The combination of all changes allows us to fit the ``model-independent'' data of \\cite{Bernabei:2000qi}, and compare with the limits of CDMS in our framework, requiring fewer than six of the observed nuclear scatterers to be WIMP events. Making certain reasonable assumptions that not too much of the DAMA signal lies in the high energy bins, we achieve the allowed parameter space of figure \\ref{fig:paramspace}. We show the allowed region for $m_\\chi=100 \\gev$.  Recently, the DAMA collaboration has performed a complete analysis of their data within the iDM framework \\cite{Bernabei:2002pp}, and found regions in good agreement with the estimated regions of \\cite{Smith:2001hy}. Interestingly, the best fit points lie well in the inelastic regime, where both the spectrum and strength of modulation can differ signficiantly from the elastic case. \\vskip0.2in  \\subsection{Experimental Signals of iDM} Although inelastic dark matter offers an attractive resolution to the conflict between CDMS and DAMA, it is essential that the scenario offer additional experimental signals to distinguish it from elastically scattering dark matter.  Of great importance are the upcoming experiments using targets at least as heavy as iodine. Xenon ($A_{Xe}=131$) is being used in a number of upcoming experiments. Still, while xenon and iodine are similar in mass, the sensitivity of most experiments compared with DAMA changes as a function of $\\delta$. This would be at most a factor of a few, and so a test should be available in the near future. Also important is the upcoming CRESST experiment, with a tungsten target ($A_W=183$) \\cite{CRESST}, which would be significantly more sensitive than any of the existing experiments. \\begin{figure} \\psfig{file=region100.eps,width=0.4\\textwidth} \\vskip-0.2in \\caption{Allowed parameter space (dark shaded) as a function of cross section per effective neutron (see ref \\cite{Smith:2001hy}) and $m_\\chi$.\\vskip-0.3in} \\label{fig:paramspace} \\end{figure}  The suppression of event rate is not the only significant difference. Because the kinematics are changed, the shape of the signal as a function of energy can be considerably different. In figure \\ref{fig:modspec} we see how the modulated signal can appear as a function of energy at DAMA. While for certain values of $\\delta$ the elastic case and inelastic case appear comparable, there can also be significant spectral differences. Currently the DAMA spectrum is unpublished, only fits to it are available. It is possible that presently existing data could help to suggest whether this scenario is correct.  At experiments which study the total rate (as opposed to modulation), the effect is even more pronounced. The signal spectrum, which for elastic WIMPs would rise dramatically at low energies, would now dip at low energies. The position of the peak would depend sensitively on $\\delta$. We show an example spectrum as would be seen at CDMS in figure \\ref{fig:cdm}. Such a signal would be a smoking gun of the iDM scenario. Measuring it would give great information into both the inelastic WIMP and the distribution of dark matter in the halo. \\vskip-0.2in ",
        "conclusions": "In summary, we have seen that inelastic dark matter can easily explain the discrepancy between CDMS and DAMA. Like the neutralino, inelastic dark matter can arise naturally in supersymmetric theories as the real component of a mixed sneutrino.  This well-motivated candidate is natural both in that it arises simply out of supergravity theories, and that its important features (the small splitting $\\delta$) are radiatively stable.  Such models make exciting predictions for the upcoming round of dark-matter experiments. For small to moderate $\\delta$, CDMS would be expected to see a signal after its move to the Sudan mine with low energy events suppressed. For moderate to large $\\delta$, DAMA would see a spectral deformation from the elastic case. In all events, upcoming xenon and tungsten experiments would see a signal with varying degrees of spectral distortion.  \\begin{figure} \\psfig{file=scenario.eps,width=0.4\\textwidth} \\vskip-0.25in \\caption{Possible future detection scenarios.} \\vskip-0.21in \\label{fig:scenario} \\end{figure}"
    },
    "0208/astro-ph0208129_arXiv.txt": {
        "abstract": "{\\small It is now ten years since the first microquasar GRS 1915+105 was discovered. More than six years of observations with RossiXTE have shown a level of variability never observed in any other X-ray source. Here I try to address some issues, based on X-ray observations only, that have relevance for theoretical modeling. First, I ignore these peculiarities and concentrate on the similarities with other X-ray transients. Then I focus on the peculiar variability and present a number of obervational facts that need to be addressed by theoretical models. } ",
        "introduction": "As practically all articles about GRS 1915+105 start, the source was discovered ten years ago, in 1992, with the Watch instrument on board GRANAT \\cite{castro}. Figure 1 shows the original discovery light curve from that work, where one can see that the source was very variable from the beginning. As a comparison, the full RossiXTE/ASM light curve up to 2002 July is shown in the bottom panel of Fig. 1. Already from these plots, it is evident that the variability of GRS 1915+105 is rather unique and that in this source we are observing something which is not observed in other sources.  \\begin{figure}[htb] \\centering \\psfig{file=belloni_1a.ps,width=10cm} \\psfig{file=belloni_1b.eps,width=10cm} \\caption{GRS 1915+105 long-term light curves. Top: GRANAT/Watch [1]. Bottom: RossiXTE/ASM up to 2002 July 19.} \\label{fig:ex} \\end{figure}  When examining the variability at shorter time scales, as observed with the RossiXTE/PCA, things become even more complex (see Fig. 2), showing a definite structure, which in this particular case repeats almost unchanged after about 25 minutes. The complexity of these light curves, first shown by \\cite{gmr97}, is analyzed and categorized by \\cite{b2000}, who classify the light curves in a dozen variability classes and identify three basic spectral states called A, B and C. In these ten years, many papers have been devoted to the analysis and interpretation of the X-ray observations of GRS~1915+105, and it is not possible to review them all here. What I want to do is to put this unique source in perspective, comparing it with other transient BHCs. First, I will emphasize the similarities with other systems, then I will concentrate on the peculiarities.  \\begin{figure}[htb] \\centering \\psfig{file=belloni_2.eps,width=10cm} \\caption{Example of a RossiXTE/PCA light curve of GRS 1915+105, from 1997 October 31.} \\label{fig:ex} \\end{figure} ",
        "conclusions": "GRS~1915+105 is a peculiar black-hole transient in at least two ways. First, it can hardly be considered a transient, since it is in ``outburst'' since ten years at large values of the accretion rate. Second, as shown before, the variability of its flux and spectral properties are unique and extremely complex. However, not all its properties are unique (as in the case of XTE J0421+56, see \\cite{bel99,par2000,boi02}), indicating that we are not dealing with a source completely different from the others. This can be turned to our advantage: by linking ``normal'' and peculiar characteristics, we can learn something general about accretion onto black holes."
    },
    "0208/astro-ph0208417_arXiv.txt": {
        "abstract": "High-resolution infrared spectra ($\\lambda$/$\\Delta$$\\lambda$= 50,000) have been obtained for twelve red-giant members of the Large Magellanic Cloud (LMC) with the Gemini South 8.3m telescope plus Phoenix spectrometer.  Two wavelength regions, at 15540\\AA\\ and 23400\\AA, were observed.  Quantitative chemical abundances of carbon (both $^{12}$C and $^{13}$C), nitrogen, and oxygen were derived from molecular lines of CO, CN, and OH, while sodium, scandium, titanium, and iron abundances were obtained from neutral atomic lines.  The twelve LMC red giants span a metallicity range from [Fe/H]= -1.1 to -0.3.  It is found that values for both [Na/Fe] and [Ti/Fe] in the LMC giants fall below their corresponding Galactic values (at these same [Fe/H] abundances) by about $\\sim$ 0.1 to 0.5 dex; this effect is similar to abundance patterns found in the few dwarf spheroidal galaxies with published abundances. The program red giants all show evidence of first dredge-up mixing of material exposed to the CN-cycle, i.e. low $^{12}$C/$^{13}$C ratios, and lower $^{12}$C- with higher $^{14}$N-abundances.  The carbon and nitrogen trends are similar to what is observed in samples of Galactic red giants, although the LMC red giants seem to show smaller $^{12}$C/$^{13}$C ratios for a given stellar mass.  This relatively small difference in the carbon isotope ratios between LMC and Galactic red giants could be due to increased extra mixing in stars of lower metallicity, as suggested previously in the literature.  Comparisons of the oxygen to iron ratios in the LMC and the Galaxy indicate that the trend of [O/Fe] versus [Fe/H] in the LMC falls about 0.2 dex below the Galactic trend.  Such an offset can be modeled as due to an overall lower rate of supernovae per unit mass in the LMC relative to the Galaxy, as well as a slightly lower ratio of supernovae of type II to supernovae of type Ia. ",
        "introduction": "The Large Magellanic Cloud (LMC), one of the nearest galaxies and a much smaller system than the Milky Way, is a prime target in which to probe chemical evolution in stellar populations.  Unlike the Milky Way, where large fractions of the volume are obscured by dust, the sightlines into most of the LMC are relatively clear, rendering entire populations of stars visible. The LMC's distance demands that detailed stellar abundance studies be based on spectra from 4-8 meter class telescopes and be restricted to rather luminous stars.  Pioneering abundance studies have been conducted on LMC stellar samples already. One basic measure of overall chemical evolution in a stellar system is an age-metallicity relation.  In the LMC, Olszewski et al. (1991) and Dopita (1996) provide extensive results, with the Olszewski et al. work based on clusters and Dopita's results derived from planetary nebulae.  More recently, Geisler et al. (1997) and Bica et al. (1998) have provided additional results from clusters.  In summary, the LMC age-metallicity relation appears to reflect a rapid enrichment phase more than 10 Gyr ago, followed by apparent chemical quiescence until about 2 Gyr ago, when another enrichment period (still ongoing) began.  Additional insights into chemical evolution can be provided by studying elements arising from different nucleosynthetic origins, e.g. Type II supernovae (SN II), Type Ia supernovae (SN Ia), or asymptotic giant branch (AGB) stars.  Studies of this type require high-resolution spectra and are still a challenging endeavor to undertake in the LMC. Published abundance distributions for the LMC include Russell \\& Dopita (1992), Barbuy, Pacheco, \\& Castro (1994), Hill, Andrievsky, \\& Spite (1995), Luck et al. (1998), Korn et al. (2000), Hill et al. (2000), Spite et al. (2001), and Korn et al. (2002).  The analysed stars include main sequence and supergiant B-stars,  supergiant K- to F-stars, and Cepheid variables.  For the younger stars ($\\le$ 10$^{8}$ yr), these papers find a typical iron abundance of [Fe/H]$\\sim$ -0.3 for LMC field stars. The oxygen-to-iron abundance ratio in the LMC may be measurably lower than this same ratio at the same [Fe/H] abundance in the Galaxy. The trend of [O/Fe] as a function of [Fe/H] can be a crucial relation in establishing star formation histories in a stellar population.  Oxygen is made preferentially in the most massive stars, while iron comes from both massive, core-collapse supernovae (SN II) and the (presumably) binary Type Ia supernovae. The run of the O/Fe ratio with Fe/H in a stellar system is, therefore, a measure of the history of SN II to SN Ia rates, and, hence, the star formation history of  the LMC. Carbon and nitrogen abundances may also be lower when compared to Galactic stars at the same [Fe/H].  As C, N, and O (with Ne) are the most abundant heavy elements, their abundances carry much weight in setting the overall metallicity, Z, in a galaxy.  Most of the previous abundance determinations of C, N and O in the LMC have rested on atomic lines, some of which are quite weak, or hampered possibly by non-LTE processes.  Because of the importance of the   O/Fe ratio to the disentangling of the LMC's star fomration history, this study set out to measure oxygen, as well as carbon and nitrogen abundances, in a sample of LMC red giants by observing molecular unblended lines of OH, CO, and CN from high-resolution infrared (IR) spectra.   Using the 8.3m Gemini South reflector, along with the Phoenix high-resolution IR spectrometer, abundances of the isotopes $^{12}$C, $^{13}$C, $^{14}$N, and $^{16}$O could be determined along with abundances of Fe, Na, Sc, and Ti. ",
        "conclusions": "Abundances of seven elements (C, N, O, Na, Sc, Ti, and Fe) have been measured in twelve red-giant members of the LMC from high-resolution IR spectra obtained with Gemini South plus Phoenix.  Using IR spectra it is possible to extract quantitative chemical abundances for a number of elements in somewhat lower-mass stars (M$\\sim$ 2-4M$_{\\odot}$ red giants) in the LMC than previous high-resolution optical spectroscopic studies of main-sequence B-stars or F to K supergiants (with M$\\sim$ 8-10M$_{\\odot}$).  The IR abundance analyses add complementary stellar mass targets to the earlier works.  In addition, the molecular lines of such species as CO, OH, or CN in the IR spectra allow for the determination of the abundances of specific CNO isotopic species, such as $^{12}$C, $^{13}$C, $^{14}$N, and $^{16}$O here, or $^{17}$O, $^{18}$O, and $^{15}$N in future studies.  The iron abundances sampled here range from [Fe/H]= -1.1 to -0.3.  Both [Na/Fe] and [Ti/Fe] are found to be consistently lower than their Galactic values by $\\sim$ -0.1 to -0.5 over the metallicity range sampled in the LMC.  These characteristic underabundances of Na and Ti seem to also occur in a number of dwarf spheroidal galaxies (Shetrone et al. 2002).  The LMC red giants in this sample all show evidence of the mixing of CN-cycle material to their surfaces via the first dredge-up, with $^{14}$N enhanced by +0.4 to +0.8 dex over its estimated initial values, and $^{12}$C decreased by -0.3 to -0.5 dex.  No evidence is found, in these predominantly first ascent giants or early AGB stars, of the extreme nitrogen enrichments ($\\sim$ +1.0 dex or more) that might result from second dredge-up or hot bottom burning.  The $^{12}$C/$^{13}$C ratios in the LMC red giants are found to decrease with decreasing giant star mass in a manner similar to that found for Galactic red giants (Gilroy 1989); however, the LMC trend appears to be shifted to lower $^{12}$C/$^{13}$C ratios for a given red-giant mass.  This shift may be due to the increased mixing associated with lower-metallicity giants as suggested by Charbonnel et al. (1998).  A comparison of [O/Fe] versus [Fe/H] between the LMC and the Milky Way finds that the LMC trend falls below by about 0.2 dex over the range of [Fe/H]= -1.1 to -0.3.  Good agreement in [O/Fe] as derived from a sample of F-supergiants (Hill et al. 1995), main-sequence B-stars (Korn et al. 2002), and the red giants analyzed here suggests that the difference between the LMC and the Milky Way is real.  Lower values of [O/Fe] in the LMC can be explained by both a lower supernovae rate (caused by a lower star formation rate) and a lower ratio of supernovae type II to supernovae type Ia.  We thank the staff of Gemini South for their excellent assistance with these observations.  Nick Suntzeff is to be thanked for insightful comments concerning the distance to the LMC. This work is supported in part by the National Science Foundation through AST99-87374 (V.V.S.).  \\clearpage"
    },
    "0208/astro-ph0208551_arXiv.txt": {
        "abstract": "Optical spectra of 14 emission-line galaxies representative of the 1999 NICMOS parallel grism \\ha survey of McCarthy et al. are presented.  Of the 14, 9 have emission lines confirming the redshifts found in the grism survey.  The higher resolution of our optical spectra improves the redshift accuracy by a factor of 5.  The [O II]/\\ha values of our sample are found to be more than two times lower than expected from Jansen et al.  This [O II]/\\ha ratio discrepancy is most likely explained by additional reddening in our \\han-selected sample [on average, as much as an extra E(B-V) = 0.6], as well as to a possible stronger dependence of the [O II]/\\ha ratio on galaxy luminosity than is found in local galaxies.  The result is that star formation rates (SFRs) calculated from [O II] \\lam3727 emission, uncorrected for extinction, are found to be on average 4 $\\pm$ 2 times lower than the SFRs calculated from \\ha emission.  Classification of emission-line galaxies as starburst or Seyfert galaxies based on comparison of the ratios [O II]/\\hb~and [Ne III] \\lam3869/\\hb~is discussed.  New Seyfert 1 diagnostics using the \\ha line luminosity, {\\it H}-band absolute magnitude, and \\ha equivalent widths are also presented.  One galaxy is classified as a Seyfert 1 based on its broad emission lines, implying a comoving number density for Seyfert 1s of $2.5^{+5.9}_{-2.1}~\\times$ 10$^{-5}$ Mpc$^{-3}$.  This commoving number density is a factor of 2.4$^{+5.5}_{-2.0}$ times higher than estimated by other surveys. ",
        "introduction": "Recent studies have found that the comoving integrated star formation rate (SFR) increases by an order of magnitude as z increases from 0 to 1 (Connolly et al. 1997; Hogg et al. 1998, hereafter H98; Yan et al. 1999).  Surveys at redshifts greater than z = 2 show that this trend flattens or decreases in that range, indicating that there may be a peak in the SFR within the redshift range $1 < z < 2$.  Thus, it is necessary to observe galaxies at these redshifts to understand star formation history not only in individual galaxies, but also in the universe as a whole.  SFRs for local galaxies have been inferred by measuring the \\ha \\lam6563 emission line (Kennicutt 1983).  \\ha is a good SFR indicator because it measures the flux of ionizing photons from young, massive O and B stars.  Assuming an initial mass function (IMF) and case B recombination, it is then possible to calculate the total SFR.  This method is observationally easiest up to z $\\sim$ 0.3, beyond which \\ha is redshifted out of the optical wavelength range.  Other prominent optical emission lines are \\hb~and [O III] \\lam5007, which are observable up to z = 0.7, as well as [O II] \\lam3727 and [Ne III] \\lam3869, which are still in the optical up to a redshift of z = 1.2.  It is possible to use [O II] \\lam3727 (a blend of [O II] \\lam\\lam~3726,3729) as a SFR indicator, although it is not as reliable (Kennicutt 1992).  Recent studies of local galaxies (Jansen, Franx, \\& Fabricant 2001; Carter et al. 2001; Charlot et al. 2002) have found that the observed L([O II])/L(\\han) ratio is correlated with the galaxy luminosity.  These studies have shown that the dependence is due to both reddening and the metallicity-dependent excitation of the interstellar medium.  The SFRs derived from the [O II] fluxes therefore have large uncertainty.  In local galaxies Jansen et al. (2001) found this uncertainty to be a factor of $\\sim$ 3, if no correction for metallicity and dust is possible.  Although this wide range in [O II]/\\ha values will lead to large uncertainty in SFRs based on [O II] line flux, it is still valuable to determine an average value for calibration of [O II] and SFR.  Using this average value will not give accurate SFRs for individual objects; but, for a large survey, accurate SFR densities may still be attainable, making it possible to assess the SFR in the previously inaccessible redshift range of $1 < z < 2$.  McCarthy et al. (1999, hereafter M99) used parallel grism observations with NICMOS onboard the Hubble Space Telescope (HST) to survey blank fields for \\han-emitting galaxies at $0.7 < z < 1.9$.  The survey, using the G141 slitless grism on NICMOS, covered approximately 64 arcmin$^{2}$ and found 33 emission line galaxies.  Redshifts were measured with the assumption that the single observed emission line was \\han.  \\ha is the most likely identification, since \\ha is the strongest optical/near-IR line (Kennicutt 1992).  M99 found an average SFR per galaxy of 30 M$_{\\odot}$ yr$^{-1}$ and an emission-line galaxy comoving number density of $3.3 \\times 10^{-4}~ h^{3}_{50}~ Mpc^{-3}$.  This comoving number density is about half that of present-day galaxies brighter than L* in the {\\it B}-band (Ellis et al. 1996) and similar to bright Lyman break galaxies at z $\\sim$ 3 (Steidel et al. 1996).  The calculated SFR density (Yan et al. 1999) is consistent with results from a similar \\ha study by Hopkins, Connolly, \\& Szalay (2000).  Two of the 33 M99 galaxies were suggested to be active galaxies (Seyferts) based on their high equivalent widths and compact morphologies.  To confirm the redshifts given by M99, and thus the SFRs and comoving number density estimates, we have obtained optical spectra of 14 of the galaxies in the M99 survey.  Along with the redshifts of the galaxies, the SFRs based on the [O II] emission are considered in an effort to determine if [O II], instead of \\han, can be used for future surveys.  The comoving number density of Seyfert 1 galaxies is also investigated.  Section 2 describes the observations, reduction process, and discusses how well our 14 galaxies represent the 33 galaxies in the survey done by M99, and $\\S$3 presents the emission line detections.  The implications of these emission lines (SFR, reddening, classification of the galaxies as active or starburst, and Seyfert 1 luminosity function), as well as two new diagnostic diagrams, are given in $\\S$4.  Throughout this paper, $H_{0} = 50~ km~ s^{-1}~ Mpc^{-1}$ and $q_{0} = 0.5$ are adopted, unless stated otherwise, to simplify comparison with previous studies. ",
        "conclusions": "\\subsection{Intrinsic [O II]/\\ha Ratio} A wide range in the [O II]/\\ha in local galaxies has been noted by several authors (see, e.g., Hammer et al. 1997).  This range is partly attributed to a dependence of [O II]/\\ha on galaxy luminosity, with higher ratios for less luminous galaxies (see, e.g., Jansen et al. 2001).  The observed [O II]/\\ha ratios for our galaxies are given in Table 6.  For 4 of the14 galaxies, our spectra did not include the wavelength where [O II] was expected to be located; thus no [O II]/\\ha is measured for these objects.  Three of these four spectra are those with no spectral features detected and thus no redshift determined.  The fourth is J1237+6219c, for which the redshift given by M99 was confirmed on the basis of three features at wavelengths shorter than [O II].  Upper limits are given in Table 6 for the two galaxies, J0613+4752a and J1134+0406a, for which the expected location of [O II] is observed but no feature detected.  For those galaxies in our sample for which [O II] was observed, the average [O II]/\\ha is 0.18 $\\pm$ 0.12, if J0931-0449a, a Seyfert 1, and upper limits are excluded.  However, the average of the logarithmic ratio for these objects is log([O II]/\\han) = -0.82 $\\pm$ 0.27, or [O II]/\\ha = 0.15$^{+0.13}_{-0.07}$.  Furthermore, if we had instead calculated the median value so that the significant upper limit of J1134+0406a would be included, then the median would drop to log([O II]/\\han) = -0.97, or [O II]/\\ha = 0.11.  Our observed [O II]/\\ha is significantly less than the median ratio observed by Kennicutt (1992), which was [O II]/\\ha = 0.45 $\\pm$ 0.26.  Figure 8 shows a histogram of the observed [O II]/(\\han+[N II]) ratios in the Kennicutt sample, as well as our sample for comparison.  A value for the [N II]/\\ha ratio must be assumed for lower resolution spectra, such as some of the Kennicutt sample spectra and all of the M99 sample.  For those spectra in the Kennicutt sample observed at higher resolution, the mean observed [N II]/\\ha = 0.5.  However, as stated by Kennicutt, this value may be slightly biased toward higher values, because galaxies with blended lines were not included, which led to excluding objects with lower [N II]/\\ha values.  Gallego et al. (1997) observed an average [N II]/\\ha = 0.4.  If this lower value is assumed, then the mean observed value of Kennicutt's sample decreases to [O II]/\\ha = 0.42 $\\pm$ 0.24.  Using this same lower [N II]/\\ha value decreases our average observed [O II]/\\ha value to 0.17 $\\pm$ 0.11.  This dependence of [O II]/\\ha on the assumed [N II]/\\ha ratio, which could result in a difference of at least 7\\%, must be kept in mind when considering the results of further calculations.  The average {\\it B}-band absolute magnitude, M$_{B}$, of the Kennicutt sample differs by more than half a magnitude compared to our sample; therefore, a difference in the [O II]/\\ha [and thus the [O II]/(\\han+[N II]) ratio] is expected.  The anti-correlation between galaxy luminosity and [O II]/\\ha in local galaxies (Jansen et al. 2001) predicts a difference in the observed [O II]/\\ha values for the two samples of 11\\%.  The relationship used is that given by Jansen et al. (2001) for [O II]/\\ha, uncorrected for interstellar reddening and using $H_{0} = 50~ km~ s^{-1}~ Mpc^{-1}$.  Table 6 lists M$_{B}$ for each galaxy in our sample, as well as the corresponding predicted log([O II]/\\han) value according to the Jansen et al. relationship.  The predicted [O II]/\\ha value for the Kennicutt sample is 0.53$^{+0.17}_{-0.13}$, based on an average M$_{B}$ = -20.80 $\\pm$ 1.4.  For our sample of 14 galaxies, the average M$_{B}$ is -21.33 $\\pm$ 1.25, which gives a predicted [O II]/\\ha = 0.49$^{+0.13}_{-0.11}$.  For starburst galaxies with confirmed redshifts, the average M$_{B}$ is -20.99 $\\pm$ 0.88 resulting in a predicted [O II]/\\ha = 0.52$^{+0.10}_{-0.08}$.  In Figure 9, it can be seen that the majority of our galaxies, seven out of nine, have significantly lower [O II]/\\ha than predicted by the Jansen et al. curve.  The remaining two objects have values consistent with the Jansen et al. prediction; however, one object only has an upper limit.  J0622-0118a is placed on the plot even though it lacks a second broad band measurement necessary to calculate an accurate M$_{B}$ (indicated by its large error bars).  M$_{B}$ was calculated using the {\\it H}-band measurment given by M99 and assuming {\\it R-H} = 2.8, the average for our sample (see $\\S$ 3.2).  However, J0622-0118a is expected to be consistent with the Jansen prediction, given its high [O II]/\\ha of 0.40, which places it on the plot at log([O II]/\\han) = -0.39.  The other galaxies in our sample lack [O II] measurements, for reasons discussed in the $\\S$ 3.2, and are thus not included in Figure 9.  Glazebrook et al. (1999) obtained [O II]/\\ha for 13 galaxies with z $\\sim$ 1.  If only galaxies with L(\\han) greater than 1$\\sigma$ are considered (eight galaxies total), their average [O II]/\\ha is 0.67 $\\pm$ 0.33, or log([O II]/\\han) = -0.17$^{+0.17}_{-0.29}$.  At higher redshifts, z $\\sim$ 3, Pettini et al. (2001) found an average [O II]/H$\\beta$ = 2.0 $\\pm$ 0.9 for five galaxies.  If \\han/H$\\beta$ = 6 is assumed, this gives an average [O II]/\\ha = 0.33 $\\pm$ 0.14, or log([O II]/\\han) = -0.48$^{+0.15}_{-0.24}$.  Including these two studies in Figure 9, it can be seen that both Glazebrook et al. and Pettini et al. agree with the relationship given by Jansen et al.  Assuming any Balmer decrement in the plausible range of 3 - 10 to convert the Pettini et al. data from [O II]/H$\\beta$ to [O II]/\\ha results in consistency with the Jansen et al. relationship.   Possible explanations for our lower [O II]/\\han, compared to those studies mentioned above, are additional reddening and/or an intrinsically lower [O II]/\\han.  As Jansen et al. (2001) demonstrated, metallicity and reddening are equally responsible for the observed range in [O II]/\\ha in local galaxies.  Both of these properties are correlated with the luminosity of the galaxy, leading to lower intrinsic [O II]/\\han, as well as more reddening in higher luminosity galaxies.  As a result, more luminous galaxies have significantly lower values of [O II]/\\ha than less luminous ones.  As Figure 9 demonstrates, the range in [O II]/\\ha seen in our sample cannot be fully explained by the dependence of [O II]/\\ha on luminosity found in the Jansen et al. sample of local galaxies. Arag\\'{o}n-Salamanca et al. (2002) found a stronger dependence of [O II]/\\ha on galaxy luminosity in the Universidad Complutense de Madrid (UCM) sample of local \\han-selected galaxies (see Fig. 9).  The mean observed [O II]/\\ha in their sample is half that observed by Jansen et al. Arag\\'{o}n-Salamanca et al. attribute this discrepancy between their sample and that of Jansen et al. to a difference in average extinction because of the selection technique used (\\ha versus {\\it B}-band selection). Our sample of \\ha selected galaxies is more likely to have an [O II]/\\ha dependence closer to that found by Arag\\'{o}n-Salamanca et al., and thus our lower observed [O II]/\\ha can be partially explained.  Using the [O II]/\\han-M$_{B}$ relationship found by Arag\\'{o}n-Salamanca et al., our predicted average [O II]/\\ha value for confirmed starburst galaxies is [O II]/\\ha = 0.23$^{+0.06}_{-0.05}$, which, within the errors, is consistent with our observed average [O II]/\\ha = 0.18 $\\pm$ 0.12. However, as mentioned, if the average of the logarithmic ratios or the median is considered, then our observed [O II]/\\ha is significantly below their prediction.  Therefore, the [O II]/\\han-M$_{B}$ relationship at higher redshifts cannot be assumed to be equivalent to that observed locally.  As suggested by Hammer et al. (1997), reddening and metallicity might play an even more important role in the observed [O II]/\\ha at high redshifts.  If a wider range of metallicity and/or reddening is present at higher redshifts than in local galaxies, then a wider range of observed [O II]/\\ha is realistic.  A more detailed discussion on the explanation for our lower observed [O II]/\\ha follows in $\\S$ 4.5.   \\subsection{Additional Reddening} Assuming an intrinsic [O II]/\\ha value of 0.52 from the Jansen et al. (2001) relationship, it is possible to estimate how much additional reddening might be present in our galaxies.  The averaged interstellar extinction curve used throughout this section is that given by Seaton (1979), but adopting A$_{V}$/E(B-V) = 3.09, instead of 3.20.  For our sample, the average confirmed starburst galaxy extinction required is E(B-V)$_{[O II]/\\han}$ = 0.59 $\\pm$ 0.34 mag.  The required extinction for each individual galaxy is given in Table 6.  If our galaxies have an intrinsically lower [O II]/\\ha than predicted by Jansen et al., then this calculation results in an overestimate of the actual extinction.   Another measure of the extinction in each galaxy is obtainable from the observed Balmer emission line ratios.  However, the Balmer lines detected have uncertain fluxes due to unknown stellar absorption lines underlying the emission.  Thus, the resulting E(B-V)$_{Balmer}$ is an upper limit. Using the \\ha emission given by M99 in combination with any H$\\gamma$, H$\\delta$, or H$\\epsilon$ emission detected in this study, the E(B-V)$_{Balmer}$ of each galaxy can be found (see Table 6).  The averaged interstellar extinction curve and intrinsic Balmer line ratios used are those given by Osterbrock (1989).  For those galaxies with multiple Balmer lines detected, an average E(B-V)$_{Balmer}$ is given.  For some galaxies, only upper limits on the higher Balmer line flux were available.  In these cases a lower limit to the E(B-V)$_{Balmer}$ is measured, meaning the upper limit on the extinction is not lower than the reported E(B-V)$_{Balmer}$ in Table 6.  The average extinction for our sample, using this method, is E(B-V)$_{Balmer}$ = 0.63 $\\pm$ 0.53 magnitudes.   \\subsection{Star Formation Rate Based on [O II]} \\ha is a reliable indicator of the current SFR (Kennicutt 1983), using the calibration \\begin{equation} SFR(M_{\\odot}~yr^{-1}) = 8.9 \\times 10^{-42}L(\\han)E(\\han), \\end{equation} \\noindent where E(\\han) is the extinction at \\ha \\lam6563.  This can be transformed into a calibration with [O II] using the average observed relationships [O II]/(\\han+[N II]) = 0.30 and accounting for [N II] \\lam\\lam6583, 6548, which is blended with \\ha in low resolution spectra, by using [N II]/\\ha = 0.5 (Kennicutt 1992).  These values give the observed value of [O II]/\\ha = 0.45, which then results in the following relationship \\begin{equation} SFR(M_{\\odot}~yr^{-1}) = 2.0 \\times 10^{-41}L([O II])E(\\han). \\end{equation} \\noindent The [O II]/\\ha ratio is the observed value for nearby galaxies (Kennicutt 1992), uncorrected for reddening.  Therefore, no further extinction correction at [O II] is needed, if it is assumed that the reddening in our sample of galaxies is the same as in nearby galaxies.  Table 6 lists the luminosity of the [O II] emission for each galaxy in our sample along with the resultant SFRs using equation (2) and no extinction [E(\\han) = 0 magnitudes = 1 times the flux].  Table 6 also contains the SFRs based on \\ha with the [N II] correction used above applied.  Therefore, the \\ha SFRs in our table are 33\\% lower than those given by M99 in their Table 3, which did not correct for [N II].  Since J0931-0449a is a Seyfert 1, the line emission is not due to high-mass stars.  Thus, the SFR for J0931-0449a is an upper limit, but it is included in the table for completeness.  Some of the spectra did not reach far enough into the blue to measure [O II] emission, and thus no SFRs were measured for these galaxies.  After correcting for [N II], M99 found an average SFR of 21 $M_{\\odot}~yr^{-1}$.  Assuming no extinction at \\han, we find that, for confirmed starburst galaxies (all of our confirmed galaxies except J0931-0449a, which is a Seyfert 1), the average SFR based on our measured [O II] is 3 $\\pm$ 2 times lower than that given by the \\ha measured in M99.  If an extinction of E(\\han) = 1 mag, which corresponds to E(B-V) = 0.43 mag, typical of nearby spiral galaxies (Kennicutt 1983), is assumed, then the [O II] and \\ha relationships change to: \\begin{equation} SFR(M_{\\odot}~yr^{-1}) = 2  \\times 10^{-41}L(\\han) \\end{equation} and \\begin{equation} SFR(M_{\\odot}~yr^{-1}) = 5 \\times 10^{-41}L([O II]). \\end{equation} \\noindent The resulting SFRs based on [O II] are listed in Table 6 in column (11).  Since this extinction is applied to both equations (1) and (2) resulting in equations (3) and (4), respectively, the ratio of SFR(\\han)/SFR([O II]) remains unchanged when this extinction is included.  Thus, the average SFR([O II]) is still 3 $\\pm$ 2 times lower than SFR(\\han).  The above relationships do not take into account the possible dependence of [O II]/\\ha on luminosity (Jansen et al. 2001).  Since the average luminosity of our sample is slightly less than that of the Kennicutt sample (see $\\S$4.1), the intrinsic [O II]/\\ha is expected to be different, and thus the relationships given in equations (2) and (4) are not quite accurate for our sample.  If an intrinsic [O II]/\\ha of 0.52, which is the prediction based on the Jansen et al. relationship for our averaged confirmed starburst galaxies, is used, then the new conversion from L([O II]) to SFR is \\begin{equation} SFR(M_{\\odot}~yr^{-1}) = 4 \\times 10^{-41}L([O II]). \\end{equation} \\noindent  The SFRs for each galaxy, using their predicted [O II]/\\ha to derive the SFR-L([O II]) relationship, are given in column (12) of Table 6.  Making this correction for the [O II]/\\ha dependence on luminosity, it is found that the averaged SFR is 4 $\\pm$ 2 times lower than that found by M99.  It can therefore be concluded that our sample of galaxies has an intrinsically lower [O II]/\\ha than that predicted by the Jansen et al. (2001) relationship and/or they are reddened more than the E(\\han) = 1 mag found in local galaxies by Kennicutt (1992).  Including the average extinction calculated in $\\S$4.2, which is based on the [O II]/\\ha ratio, results in SFR relationships that account for the maximum amount of additional reddening that may be present in our sample.  Adding this additional reddening to the E(\\han) of 1 mag already included by Kennicutt in the SFR relationships, gives a total average reddening of E(B-V) = 1.02 $\\pm$ 0.34 mag.  This larger reddening, which we consider more realistic, increases the intrinsic \\ha and [O II] fluxes as follows: \\begin{equation} L(\\han)_{int} = 3.52^{+3.72}_{-1.81}  \\times L(\\han)_{obs} \\end{equation} and \\begin{equation} L([O II])_{int} = 12.59^{+41.34}_{-9.65} \\times L([O II])_{obs}. \\end{equation} The transformations to SFR for are then \\begin{equation} SFR(M_{\\odot}~yr^{-1}) = 8 \\times 10^{-41}L(\\han) \\end{equation} and \\begin{equation} SFR(M_{\\odot}~yr^{-1}) = 50 \\times 10^{-41}L([O II]). \\end{equation} The SFR for each galaxy, using its calculated additional reddening, is given in column (13) of Table 6.  Although including this additional reddening has the effect of raising the SFRs based on [O II], so that SFR(\\han) equals SFR([O II]), it also has the effect of raising the SFRs based on \\ha.  As is seen in the next section, this would result in an SFR density that is higher than reported by any other study.    \\subsection {Star Formation Rate Density} Assuming our sample is representative of the whole \\ha survey, we arrive at an uncorrected, volume-averaged SFR 4 $\\pm$ 2 times lower than that found by Yan et al. (1999), which is based on the M99 \\ha survey.  Their result was a volume-averaged SFR at z=1.3 $\\pm$ 0.5 of 0.13 \\Msun~yr$^{-1}$~Mpc$^{-3}$.  Therefore, we find, based on observed [O II] emission with no additional correction for reddening, a volume-averaged SFR of 0.03$^{+0.03}_{-0.02}$ \\Msun~yr$^{-1}$~Mpc$^{-3}$ in the same redshift range.  Comparing our volume-average SFR to other studies in a similar redshift range, we find that our [O II]-based SFR density is again lower.  In addition to using [O II] emission as a SFR indicator, \\ha emission (Gallego et al. 1995; Tresse \\& Maddox 1998; Glazebrook et al. 1999; Yan et al. 1999; Hopkins, Connolly, \\& Szalay 2000) and UV continuum (Lilly et al. 1996; Connolly et al. 1997; Treyer et al. 1998) have also been used.  Although both of these methods agree that there is an order of magnitude increase in the SFR from z = 0 to z = 1, the UV estimates are significantly less than those based on \\ha emission.  Figure 10 shows our volume-averaged SFR density, along with results from the studies mentioned above.  It can be seen that our results are most consistent with SFR densities based on the UV continuum.  Our SFR density is at least 3 times lower than any results found using the \\ha emission method.  The systematic discrepancy between SFR based on the UV continuum and \\ha emission can most likely be explained by dust extinction (see, e.g., Steidel et al. 1999), which suppresses the UV continuum more than the \\ha emission.  This same reddening may be responsible for the difference in the SFR predictions based on \\ha and [O II].  Hogg et al. (1998, hereafter H98) measured [O II] in 375 faint z $\\la$ 1 galaxy spectra, making it possible for our results to be compared to a study based on the same SFR indicator.  They found an order of magnitude increase in the SFR density from present day out to z $\\sim$ 1, similar to results based on the UV continuum and \\han.  Their volume-averaged SFR agrees with results based on \\ha (see Fig. 10), possibly indicating that the [O II] to SFR conversion in equation (4), which assumes reddening typical of local spiral galaxies, is correct for z $\\la$ 1 galaxies.  Their redshift range is a bit lower than ours, but there is enough overlap for a reasonable comparison.  Despite the use of the same SFR indicator, our SFR density from [O II] is almost 3 times lower than H98's results in a similar redshift range.  Since \\ha was not measured in the H98 sample of galaxies, the intrinsic [O II]/\\ha is unknown.  However, it is reasonable to assume that the intrinsic value is near [O II]/\\ha $\\sim$ 0.45, since using the SFR-L([O II]) relationship given by Kennicutt (1992) results in SFR densities similar to those found in other studies using Kennicutt's relationship for SFR-L(\\han).  Another way to determine [O II]/\\ha is to use the Jansen et al. relationship.  For the H98 sample, which has an average M$_{B}$ = -20.5$^{+1.4}_{-1.9}$, the relationship predicts an intrinsic [O II]/\\han = $0.58^{+0.19} _{-0.18}$ (see Fig. 9).  M$_{B}$ was calculated based on the average {\\it R}-band magnitude of the H98 sample, {\\it R} = 21.8 $\\pm$ 0.8, the average redshift, z = 0.65 $\\pm$ 0.2, assuming an average {\\it R-H} = 2.8 (the average of our sample; see $\\S$ 3.2).  Changing the assumed {\\it R-H} color to 0.5 or 5, changes [O II]/\\ha by less than 5$\\%$.  If the dependence of [O II]/\\ha on luminosity is ignored, then it is concluded that our sample has a SFR density at least 2.9 times lower that that found by H98.  Even after accounting for the differences in luminosity, our sample has a SFR density 2.6 times lower than the H98 sample.  The remaining discrepancy between the samples is most likely due to a difference in the amount of reddening.  It is also possible that the dependence of [O II]/\\ha on luminosity is greater than that predicted by the Jansen et al. relationship at higher redshifts, which could also play a role in the remaining discrepancy.  Why such a difference would exist in such similar samples is discussed in the next section.  \\subsection{Luminosity Function Comparison} Comparing our study and H98 in more detail, we calculate the luminosity function (LF) predicted from our [O II] measurements.  This can be done by assuming our sample is representative of the whole M99 sample, and transforming the \\ha LF given in Yan et al. (1999), which is based on the M99 \\ha galaxies, into a [O II] LF.  The transformation is applied to the \\ha LF given in Figure 1 of Yan et al. by assuming a constant [O II]/\\ha = 0.18 $\\pm$ 0.12 for all galaxies, which is our average observed value.  The new LF then represents what the same 33 galaxy sample would yield if they were all observed in [O II].  Comparing our LF in Figure 11a to the LF given by H98 (only considering their $0.35<z<1.5$ galaxies), our LF is lower, especially at the higher luminosities.  The \\ha LF found by Hopkins et al. (2000), transformed using the same constant [O II]/\\ha value, is also included in Figure 11a.  The transformed Hopkins et al. luminosity function also appears to confirm that using our observed [O II]/\\ha ratio to transform from an \\ha to an [O II] luminosity function results in a luminosity function that is lower than that of H98 at the highest luminosities.  However, the Hopkins et al. luminosity function may be incomplete at the higher luminosities.  The discrepancy between the Hopkins et al. and Yan et al. luminosity functions is most likely explained by Yan et al.'s much wider areal coverage, which allowed them to sample the higher end of the luminosity function more completely.  Even accounting for incompleteness of the Hopkins et al. sample at the higher luminosities, our observed [O II]/\\ha still results in a transformed luminosity function that is significantly lower than the H98 luminosity function at the higher luminosities.  The discrepancy between the transformed \\ha LFs and the H98 [O II] LF may be due to the use of a single value for the \\ha to [O II] transformation for a sample that we know has a wide range in [O II]/\\ha values.  To account for the wide range in [O II]/\\ha observed in this study, the value of [O II]/\\ha used in the transformation must take into account the value's dependence on the \\ha luminosity.  This \\ha luminosity dependent [O II]/\\ha would then be applied to the \\ha LF, resulting in a more accurately transformed LF.  However, the seven galaxies in our sample for which [O II]/\\ha were measured, only cover a very narrow range in log(L(\\han)) values, 41.9 to 42.2.  This, combined with the error on each individual measurement, leads to a fit with very large uncertainty.  The resulting transformed \\ha LFs have such large error bars that no conclusion can be made about the consistency of the transformed \\ha and [O II] LFs.  An alternative to transforming the \\ha LFs is to do the transformation in the other direction, i.e., transforming the H98 [O II] LF into a \\ha LF, which can then be compared to the Yan et al. (1999) and Hopkins et al. (2000) results.  Our sample of seven galaxies covers a wider range in L([O II]), enabling a more reliable fit to be obtained.  It is assumed that a straight line is a good approximation of the dependence of [O II]/\\ha on L([O II]).  The seven data points appear linear to the eye, and more complicated functions could not be justified with so little data.  The linear Pearson's correlation coefficient for a straight line fit is 0.92.  Since our sample is very small, the resulting LF is still uncertain.  In addition, the differences in the SFR density based on [O II] between our sample and the H98 sample may indicate that [O II]/\\ha dependence on L([O II]) may be different for the two samples.  Thus, using a transformation based on our sample causes even greater uncertainty in the resulting transformed [O II] LF.  The [O II]/\\ha values used for the eight H98 points, from left to right in Figure 11b, are as follows: 0.10, 0.10, 0.10, 0.16, 0.36, 0.52, 0.70, 0.90.  For the first three points, [O II]/\\ha=0.10 was assumed, instead of the negative value given by the fit.  As Figure 11b demonstrates, using these more accurate values results in the H98 transformed [O II] LF being slightly lower than the LFs given by Yan et al. and Hopkins et al. at the highest luminosities, although they agree within the errors.  The H98 [O II] LF is higher than the \\ha LFs in Figure 11a and slightly lower in Figure 11b as a result of our observed [O II] values being lower than those in the H98 study.  As previously stated, the local luminosity-[O II]/\\ha relationship, is not enough to explain the difference in the LFs between our sample and that of H98.  If the dependence of [O II]/\\ha on luminosity is stronger in higher redshift galaxies, then the LF discrepancy may be expected.  Also, the differences at the bright end of the LF could be due to the small numbers.  With the steep slope in the \\ha LF, small systematic errors could lead to large differences in the derived \\ha LF.  Should neither of the above scenarios be responsible for the discrepancy in the LFs, the difference is most likely attributable to additional reddening in our sample.  This does not necessarily mean the two surveys observed galaxies from completely separate populations.  It is possible that the two studies are observing galaxies from a single population of galaxies, but that the subsamples are from different sides of a broad distribution.  Thus, the differing selection effects imposed on the galaxy population, which has an intrinsically wide range of [O II]/Ha values, causes the two samples to contain different cuts of the same population.  The M99 sample was selected on \\ha luminosity in a grism survey.  This method has a bias towards bright, compact sources, which causes our sample to have a higher average absolute {\\it B}-band luminosity than those in the H98 sample, which were selected by {\\it R}-band magnitude.  Our higher-luminosity objects lead to a lower intrinsic [O II]/\\ha, and thus galaxies with lower [O II] are more likely to be observed in our sample.  As stated, this difference in intrinsic [O II]/\\ha alone does not, most likely, explain the difference in LFs of the two samples, unless the dependence on luminosity is greater at higher redshifts than observed in local galaxies.  To explain the large discrepancy in LFs, it is also necessary to assume that our galaxies have more reddening (see $\\S$4.2).  Based on the {\\it R}-band magnitudes of our 14 galaxies, only four would have been selected by H98, suggesting that, on average, the M99 galaxies have more dust extinction than the H98 sample.  Thus, if a sample is chosen based on \\ha emission, lower [O II]/\\ha than those observed in an {\\it R}-band magnitude selected sample are more likely to be observed due to the increased reddening and lower intrinsic [O II]/\\ha ratios.  The selection methods for the two samples may, therefore, have limited the surveys to different subsections of the total population, causing neither one to sample the entire population completely.  Further support for this theory is provided by recent work by Arag\\'{o}n-Salamanca et al. (2002).  As mentioned in $\\S$ 4.1, they found that their \\han-selected sample of local galaxies had significantly more reddening than the local Jansen et al. (2001) sample, which was {\\it B}-band-selected.  This additional reddening resulting in an observed mean [O II]/\\ha that is half that observed in the Jansen et al. sample. Therefore, even at low redshift, there is such a wide range in the distribution of [O II]/\\ha and E(B-V) that different selection methods select subsamples with significantly different properties.  This same argument can be used to explain why the Pettini et al. (2001) study, which consists of z $\\sim$ 3 galaxies, has [O II]/\\ha values consistent with the Jansen et al. relationship, while we do not.  The Pettini et al. sample has an average {\\it R-H} = 1.2 $\\pm$ 0.2, if {\\it H-K} = 1.2 is assumed, which is a reasonable estimate for z $\\sim$ 3 galaxies (Chen et al. 2002).  This {\\it R-H} color is much lower than our average, found to be {\\it R-H}=2.6$\\pm$1.0.  The {\\it H-K} color must be assumed to be less than 0.8 for the average color of the Pettini et al. sample to be consistent with our average color.  Therefore, it is likely that our sample has more reddening then the Pettini et al. sample, thus leading to a lower observed [O II]/\\ha.  However, the  Glazebrook et al. sample, which is also consistent with the Jansen et al. relationship with a sample at z $\\sim$ 1, cannot be explained in this way.  The {\\it J-H} colors for their sample are similar to our sample, indicating that there is no significant difference in the degree of reddening.  The Glazebrook sample was selected based on {\\it I}-band magnitude, and if {\\it R-I} = 2.0 is assumed then only 3 of our galaxies out of the 14 would have been missed in their sample.  The average M$_{B}$ of the two samples only differ by 0.4 mag, and the average redshift of the Glazebrook sample is lower than that of our sample.  However, there is enough overlap in the redshift ranges covered in our sample and the Glazebrook sample, that their average observed [O II]/\\ha is not expected to be a factor of 3.7 discrepant.  Although continuum reddening does not seem to be responsible for this discrepancy, it is possible that the difference in the studies is due to reddening in the line emission, instead of the continuum.  Thus, when deriving the SFR-L([O II]) relationship, it is necessary not only to consider the dependence of [O II]/\\ha on luminosity, but also, at least when working at z $\\gtrsim$ 1, to assess the degree of reddening.  Because of the wide range in [O II]/\\ha in our study and the others mentioned above, it is possible that the E(B-V) extinction in the observed galaxy population can vary by more than a factor of 4 mag.   \\subsection{Classifying Emission-Line Galaxies} Standard diagnostic diagrams (Baldwin, Philips, \\& Terlevich 1981) are valuable for classifying emission-line galaxies based on their optical lines.  However, these diagrams are only useful for those galaxies at low enough redshifts for the necessary emission lines to be available at optical wavelengths.  The only diagnostic emission lines generally available at the rest wavelength ranges observed for our eight galaxies are [O II] and [Ne III] \\lam3869.  By comparing the luminosity of each of these lines to \\hb, it is still possible to classify emission galaxies as either a Seyfert 2 or a starburst galaxy (Rola, Terlevich, \\& Terlevich 1997).  This diagnostic does not work well for identification of LINERs.  Using an assumed Balmer decrement of 6, the \\ha luminosities published by M99 can be converted into \\hb~luminosities.  The contribution of [N II]\\lam\\lam~6583, 6548 to the \\ha + [N II] blend is assumed to be [N II]/\\ha $\\sim$~0.5 (Kennicutt 1992) for starbursts.  Figure 12 shows the position of our galaxies that have both [O II] and [Ne III] measurements compared to those galaxies observed by Rola et al. (1997).  All of our galaxies lie within the range for starburst galaxies, which is given by Rola et al. (1997) to be at log([Ne III] \\lam3869/\\hb) $\\la$ -0.2.  Increasing the Balmer decrement would push them closer to the area containing the Seyfert 2s, but still well within the region of starburst galaxies.  An implausibly high decrement, greater than \\han/\\hb~= 15, is needed in order to classify J0622-0018a, the most likely candidate in our sample, as a Seyfert 2.  Another adjustment that can be made is the contribution of [N II], which is closer to [N II]/\\ha $\\sim$~1.0 (McQuade et al. 1995) for Seyfert 2s.  This would also increase the ratio [Ne III] \\lam3869/\\hb, but by a smaller amount than from increasing the Balmer decrement.  Even if a higher contribution of [N II] is used, which can be as high as [N II]/\\ha $\\sim$~1.5 (see, e.g., Malkan $\\&$ Oke 1983), in combination with a reasonably high Balmer decrement, it is still not possible to classify any of our galaxies as a Seyfert 2 using this diagnostic.  We thus conclude that even though J0622-0018a and J0055+8518a have emission features that might lead one to suspect they are Seyfert 2s, they should still be classified as starburst galaxies.  \\subsection{New Diagnostic Diagram} In an effort to identify Seyfert 1s that were assumed to be starburst galaxies in the M99 \\ha survey, a new diagnostic diagram is used.  Comparing the rest equivalent width to the luminosity of the \\ha line suggests that Seyfert 1s may be separated from starburst galaxies using the criteria that they have $log(L_{\\han}) > 42.5$ and $EW(\\han) > 100 \\AA$ (see Fig. 13).  These limits were determined by eye, guided by the number of Seyfert 1s expected in the M99 sample (see $\\S$4.8), and should only be taken to be approximate values.  A separation can also be made comparing the absolute H (1.65\\mic) magnitude to the luminosity of \\han.  Figure 14 shows that Seyfert 1s may be separated from starburst galaxies by again having $log(L_{\\han}) > 42.5$ as well as $M_{H} < -25$.  Galaxies that fit both of the above criteria are most likely Seyfert 1s.  These criteria indicate that not only is J0931-0449a a Seyfert 1, but also that J0917+8142a and J0923+8149a are good Seyfert 1 candidates.  In addition, J0917+8142c is a good candidate due to, as M99 noted, its large \\ha equivalent width and compact morphology.  J0917+8142c meets the first criterion, but no {\\it H}-band magnitude was obtained by M99 to enable its placement on the $log(L_{\\han})$ versus $M_{H}$ plot. It is most likely that it would be place well within the Seyfert 1 area of the plot.  J0738+0507a, although it only satisfies one of the above criterions, is also a good candidate due to its compact morphology observed by M99.  \\subsection{Seyfert 1 Number Density} Of our 14 galaxies, we were able to confirm the redshifts of nine, and of these, one is a Seyfert 1 (J0931-0449a).  Assuming our galaxies are representative of the \\ha survey and that a Seyfert 1 could not have been missed in our other eight spectra, this gives a Seyfert 1 comoving number density of $3.7^{+8.4}_{-3.1} \\times 10^{-5}~h^{3}_{50}~Mpc^{-3}$; this comoving number density is 1/9 of the total galaxy comoving number density found by M99.  Errors were estimated from Poisson statistics (Gehrels 1986).  If this comoving number density is correct, there should be 2 or 3 other Seyfert 1s in the 33-galaxy \\ha survey.  Furthermore, assuming the redshifts of the four unconfirmed galaxies (excluding J0613+4752a which can not be excluded as a possible Seyfert 1; see $\\S$3.2) are correct, based on the \\ha emission line, and that emission lines would have been detected if they were Seyfert 1s, the comoving number density decreases to 1/13 of the total density found by M99, giving $2.5^{+5.9}_{-2.1} \\times 10^{-5}~h^{3}_{50}~Mpc^{-3}$.  In this case there should then be 1 or 2 other Seyfert 1s among the \\ha survey galaxies.  If instead it is assumed that J0613+4752a is a Seyfert 1 as well, then the commoving number density is 2/14 of the total density found by M99, giving $4.7^{+10.8}_{-3.9} \\times 10^{-5}~h^{3}_{50}~Mpc^{-3}$.  A method to identify these likely Seyfert 1 galaxies was discussed in $\\S$4.7, as were the most likely candidates.  The number of quasars/Seyfert 1s of a given luminosity at a given redshift is fairly uncertain at the redshifts of the fourteen galaxies in this study.  Grazian et al. (2000), using a luminosity dependent luminosity evolution (LDLE) model, suggest that the number density of Seyfert 1s within the absolute blue magnitude range of -21 to -25 and redshifts $1.0 < z < 1.4$ is 1.1$^{+0.58}_{-0.36} \\times 10^{-5}~Mpc^{-3}$.  We therefore had an expectation value of 0.39 Seyfert 1 galaxies in our sample of 13 galaxies (which excludes J0613+4752a for reasons stated earlier).  The probability of the observation of one Seyfert 1 galaxy in our sample being consistent with the expected value of 0.39 is 26\\%.  The consistency of our result with the prediction of the LDLE model cannot, therefore, be ruled out at a high significance level.  However, if we take our observation of one Seyfert 1 galaxy in our sample as representative of the whole M99 sample, then this gives a comoving number density a factor of 2.4$^{+9.6}_{-2.2}$ times higher, which, within the error, is still marginally consistent with the LDLE model prediction."
    },
    "0208/astro-ph0208284_arXiv.txt": {
        "abstract": "\\textit{Hubble Space Telescope} Wide Field Planetary Camera 2 images of Hickson Compact Group 79, Seyfert's Sextet, are presented.  Both point sources and extended sources detected on the three WF chips were photometered in four filters:  F336W, F439W, F555W, and F814W.  Unlike other HCGs that have been imaged with \\textit{HST}, there do not appear to be any candidate young star clusters among the detected point sources.  The majority of the point sources that may be star clusters associated with the Sextet have red colors consistent with stellar populations older than 1 Gyr.  A similar conclusion is drawn with regard to the extended sources.  The majority of these appear to be background galaxies, but a few candidate dwarf galaxies are identified as potentially associated with Seyfert's Sextet.  However, no blue, star forming objects similar to the tidal dwarf galaxy candidates identified in other HCGs are found among the extended objects identified in this study.  A redshift for one dwarf galaxy candidate was measured from a spectrum obtained with the Hobby-Eberly Telescope, and this object was found to have a redshift similar to NGC6027e, the discordant spiral formerly identified as a member of this compact group.  The \\textit{HST} observations presented here and previous radio observations of the neutral gas content of this group suggest that the interactions that have taken place in the Sextet only redistributed the stars from the member galaxies within the group.  We speculate that future interactions may be strong enough to strip the gas from NGC6027d and trigger star cluster formation. ",
        "introduction": "Compact groups of galaxies are found at the extreme end of the distribution of galaxy surface densities; by definition, compact groups consist of four or more isolated galaxies found within a small area on the sky.  Using the \\citet{hickson82} compact group selection criteria, the surface densities of galaxies in compact groups are similar to or larger than those found in the centers of massive galaxy clusters.   A radial velocity survey of the 100 Hickson compact groups \\citep[HCGs; ][]{hickson92} has verified that the majority of these HCGs are physical associations of 3 or more galaxies with a median velocity dispersion of 200 km sec$^{-1}$.  One expects that in such dense groups with small velocity dispersions that interactions and mergers among the members are inevitable.  As expected, the morphologies of the galaxies in compact groups often show evidence of tidal interaction.  Galaxy mergers are complicated phenomena, and it is difficult to disentangle the past histories of the merging galaxies, even in isolated, merging pairs.  In compact groups there is often evidence for multiple interactions \\citep[e.g.,][]{hickson82}, so unravelling the history of these systems is even more of a challenge.  Fortunately, star formation is a useful tool for dating some of the discrete events in the interaction history of a galaxy merger.  Relying on \\textit{Hubble Space Telescope (HST)} images, a growing number of studies have found evidence for compact star cluster formation in systems of interacting galaxies or merger remnants \\citep{holtzman92, whitmore93, rwo95, schweizer96, miller97, zepf99, johnson00, sq01}. Using stellar population synthesis models, the ages of the observed young star clusters can be estimated from their photometric colors. In cases where discrete populations of young clusters are found in different regions of an interaction or merger remnant, variations among the ages of the clusters can be used to date events in the merger history.  For example, \\citet{whitmore99} identified four distinct populations of young star clusters in the ``Antennae'' (NGC 4039/39), and used the ages of these populations to infer some of the evolutionary history of this system.  \\citet{sq01} have identified a number of compact clusters in HCG 92, Stephan's Quintet.  This system is more complicated than the Antennae, but the star cluster ages have been used to identify distinct epochs of cluster formation and to date some of the interaction events.  Star formation in interacting galaxies is not necessarily limited to compact cluster formation; there is also observational evidence for the formation of extended, dwarf galaxy-sized objects in the tidal debris of a number of systems \\citep[e.g.,][]{mirabel91, mirabel92, hibbard94, deeg98, duc98, duc00, MdO01}.  \\citet{sdh96, sdh98} have performed a statistical analysis of a large sample of HCGs and found evidence that tidal dwarf galaxy formation in compact groups may be common. Numerical simulations suggest that dwarf galaxy formation in tidal tails is possible \\citep{barnes92, elmegreen93}; however, it is unclear whether the observationally identified ``tidal dwarf galaxies'' (TDGs) will evolve into the bound entities seen in the simulations.  Recently, a few groups have presented dynamical analyses of several TDGs \\citep{duc00, MdO01, hibbard01} in order to assess the likelihood that these objects will remain bound, but the results remain ambiguous.  \\textit{HST} imaging of HCGs is useful for addressing both the interaction history of a compact group as well as the presence and nature of any putative TDGs.  In this paper, we present Wide Field Planetary Camera 2 (WFPC2) observations of Hickson Compact Group 79, Seyfert's Sextet, taken in the $U$, $B$, $V$, and $I$ (F336W, F439W, F555W, and F814W) filters.  In Sections 2.2--2.4, the compact clusters in the group are identified and their ages estimated from stellar population models.  In Sections 2.5--2.6, we present the extended objects in the field and discuss their nature.  Section 3 discusses a spectroscopic follow-up observation of one candidate dwarf galaxy associated with the Sextet.  Finally, in Section 4, we compare the population of compact clusters and dwarf galaxies in Seyfert's Sextet to those found in the recent \\textit{HST} studies of HCG 92 \\citep[Stephan's Quintet;][]{sq01} and HCG 31 \\citep{johnson00}. ",
        "conclusions": "The case study of Seyfert's Sextet presented here is part of a continuing effort to determine whether dwarf galaxies form during tidal interactions among giant galaxies.  The Sextet appears to be the most logical choice to search for tidal dwarf formation; it is the most compact of the Hickson compact groups, contains two prominent tidal tails, has a low velocity dispersion, and previous ground-based imaging revealed a number of faint, extended objects within the boundaries of the group.  However, the results of the \\textit{HST} imaging of Seyfert's Sextet show that, contrary to expectations, there is very little evidence for dwarf galaxy formation or any other strong star formation in this group.  A large number of both point sources and extended sources were catalogued and photometered from the three Wide Field images. We find that very few objects are detected in either of the two blue filters, F336W and F439W, and those that are detected in the two red filters, F555W and F814W, have red colors consistent with those of old stellar populations.  The majority of the point sources detected appear to be old ($> 1$ Gyr) and the majority of the extended sources detected appear to be background galaxies.  These photometric results contrast sharply with \\textit{HST} imaging studies of other HCGs, such as HCG92 (Stephan's Quintet) and HCG31.  In HCG92, \\citet{sq01} found a number of bright, blue star cluster candidates in the tidal debris regions of this group.  The images of this compact group also show bright, blue extended sources in the ``Northern Starburst Region'' and in the tidal tails of NGC7319 and NGC7318a/b.  HCG31 contains a significant number of bright, blue point sources \\citep{johnson00} similar to those seen in HCG92.  Star-forming regions are also observed in HCG31 that are ``too small to be called galaxies themselves, but are not clearly associated with either galaxy AC or galaxy E'' \\citep{johnson00}.  Thus, both HCG92 and HCG31 contain what appear to be young star clusters and tidal dwarf galaxy candidates, while Seyfert's Sextet does not appear to contain a significant population of either type of object.  The star cluster candidates identified in this study have photometric properties consistent with those for models of massive ($\\sim10^{6} M_{\\sun}$) clusters with ages $10^{8.5} - 10^{9.5}$ years.  The ages of these objects suggest that they are not entirely a primordial population, but may be the product of an interaction within the compact group at some time within the past few Gyr.  \\citet{williams91} argue that the optical tail associated with NGC 6027b and the \\ion{H}{1} gas that they associate with NGC 6027d may have resulted from an interaction between these two disk galaxies more than $5 \\times 10^{8}$ years ago. The ages we derive for many of the cluster candidates are consistent with this hypothesis.  While there do not appear to be any young star clusters or tidal dwarf galaxies associated with Seyfert's Sextet, we did identify several candidate dwarf galaxies in the group.  This sample includes a few faint, blue extended sources, and two galaxies with peculiar morphologies:  an irregularly shaped galaxy located quite near the disk of NGC6027c and an unusual, ``cometary'' galaxy located within the tidal tail associated with NGC6027c.  An additional candidate, galaxy 4.2, has already been ruled out as a member of the Sextet; the Hobby-Eberly Telescope spectrum of this object instead shows that it is associated with NGC6027e, the discordant redshift member of Seyfert's Sextet.  Whether or not the other candidate dwarf galaxies are associated with the Sextet, they appear morphologically very different from the clumpy, blue tidal dwarf galaxy candidates in HCG92 and HCG31.  The data suggest that there is some fundamental, physical difference between the Sextet and the two HCGs that are known to contain young star clusters and tidal dwarf galaxy candidates.   One obvious difference between these groups are the types of galaxies contained in each:  Seyfert's Sextet is primarily made up of early-type (S0/E) galaxies, Stephan's Quintet contains spirals, and HCG31 contains mostly irregular galaxies.  Based on these morphologies, one initial expectation is that the neutral gas content in Seyfert's Sextet is likely to be lower than that of either HCG31 or HCG92.  Radio observations show that the Sextet contains only $2 \\times 10^{9}\\:M_{\\sun}$ of neutral Hydrogen \\citep{vm01}, about an order of magnitude less than that of HCG31 and HCG92 \\citep{williams91, vm01}.  The most recent observations of the gas content of HCG92 \\citep{williams02} revise the gas mass of this group downward, however it remains at least five times larger than the gas in Seyfert's Sextet.  What appears to be the more significant difference among these three HCGs, however, isn't the gas mass, but the \\textit{distribution} of the \\ion{H}{1}.  \\citet{williams91} present VLA neutral Hydrogen observations of Seyfert's Sextet that indicate that the majority of the \\ion{H}{1} mass is retained by the disk of NGC6027d, although some gas is found in a tail extending to the east of this galaxy and also in the optical tidal tail associated with NGC6027b.  In HCG31, the VLA neutral Hydrogen maps \\citep{williams91} show that the gas is found both in the galaxies themselves and in a large envelope of gas that is plausibly attributed to tidal interactions between the galaxies.  The distribution of \\ion{H}{1} in HCG92 is found to lie entirely outside of the galaxies \\citep{vm01, williams02}, however.  The \\ion{H}{1} is concentrated in clouds and tidal tails that are not coincident with the disks of the member galaxies.  \\citet{vm01} proposed an evolutionary sequence based on their VLA observations of the \\ion{H}{1} content of HCGs.  In their model, ``phase 1'' HCGs are those where the vast majority of the neutral gas remains bound to the member galaxies.  ``Phase 2'' HCGs are more evolved in the sense that the galaxies retain some of the gas, while approximately half of the gas mass is found in tidal features.  The final, most evolved phase is broken into two subclasses, ``phase 3a'' and ``phase 3b''.  Phase 3a groups are those where the gas is almost completely stripped from the galaxies and is found entirely within tidal features, while phase 3b groups are a few rare cases where the entire group seems to be contained in a single \\ion{H}{1} cloud.  HCG31 is considered a prototype phase 2 group, and HCG92 is considered an extreme example of phase 3a.  Both \\citet{williams91} and \\citet{vm01} find Seyfert's Sextet to be anomalous; its gas distribution suggests that the system has not experienced significant dynamical evolution, while optical observations suggest the opposite.  We propose one possible scenario for the history of Seyfert's Sextet that takes into account the following significant factors:  (1)  The tidal tails are evidence for interactions among the accordant redshift members some time in the past, (2) the interactions that have occurred have not triggered star and/or star cluster formation similar to that seen in other merging galaxies and compact groups, (3) the relatively small amounts of neutral gas in Seyfert's Sextet remains bound in the one late type galaxy and does not appear to be distributed among the group environment, and (4) the low velocity dispersion among the member galaxies and the small distances between the member galaxies suggests that future interactions among the galaxies are likely.  This accumulated evidence suggests that a number of gas-poor (and one gas-rich) galaxies have interacted beginning perhaps as long as 1 Gyr or more in the past (dated by the colors of the red globular cluster candidates).  The interactions in the group have created the optical tidal tails and perhaps created the elliptical member of the Sextet, NGC6027a, as well.  The interactions in the past stripped stars from the progenitor galaxies, redistributing them within the group.  The evidence for a red, low surface brightness halo encompassing all of the member galaxies, which is seen in our images as well as deeper ground-based images, is further evidence for a redistribution of the galaxies' stars within the group.  The only ongoing star formation and most of the neutral gas is found within the disk of NGC6027d, the only late type member of the group, suggesting that any interaction that involved this galaxy must have been minor, although the galaxy disk does appear somewhat irregular and perhaps warped.  We speculate that further interactions are probably inevitable, and a major interaction between NGC6027d and the other members of the group may trigger the stripping of its neutral gas and star cluster formation throughout the group in the future.  Moreover, the low velocity dispersion suggests that none of the four large galaxies are likely to escape the group, and thus the group members may merge into a single galaxy, rather than remaining distinct.  Thus, we believe that we are seeing Seyfert's Sextet at the ``beginning of the end''; we presume that the future interactions will be the end of this group, transforming it into a single galaxy."
    },
    "0208/gr-qc0208046_arXiv.txt": {
        "abstract": "I perform an independent analysis of radio Doppler tracking data from the Pioneer 10 spacecraft for the time period 1987--1994.  All of the tracking data were taken from public archive sources, and the analysis tools were developed independently by myself.  I confirm that an apparent anomalous acceleration is acting on the Pioneer 10 spacecraft, which is not accounted for by present physical models of spacecraft navigation.  My best fit value for the acceleration, including corrections for systematic biases and uncertainties, is $(8.60\\pm 1.34)\\times 10^{-8}$ cm s$^{-2}$, directed towards the Sun. This value compares favorably to previous results.  I examine the robustness of my result to various perturbations of the analysis method, and find agreement to within $\\pm 5\\%$.  The anomalous acceleration is reasonably constant with time, with a characteristic variation time scale of $> 70$ yr.  Such a variation timescale is still too short to rule out on-board thermal radiation effects, based on this particular Pioneer 10 data set. ",
        "introduction": "}  Measurements of spacecraft motions in the solar system can be used as tests of gravitation and relativity.  Recently, Anderson et al. \\cite{anderson98,anderson02} have presented the discovery of an anomalous effect seen in radio tracking data from the Pioneer 10 spacecraft.  When interpreted as a Doppler shift, this anomalous effect corresponds to a constant acceleration, directed towards the Sun, of approximately $(8\\pm 1)\\times 10^{-8}$ cm s$^{-1}$.  Anderson et al. (hereafter \\anderson) found that the anomalous effect could not be explained by previously known physics or spacecraft properties.  The discovery of the anomaly has stimulated numerous efforts to explain it.  Some of the explanations involve ``new physics,'' such as modified gravity or dark matter, while other explanations invoke a change in the physical properties of the Pioneer spacecraft, such as a asymmetric radiation profile.  I have considered a third avenue of exploration, which is to test the analysis procedure for flaws.  In this paper, I present an independent analysis of the Pioneer 10 trajectory and search for an anomalous acceleration.  \\anderson\\ studied radio tracking data from four deep space missions: Pioneer 10, Pioneer 11, Ulysses, and Galileo.  All four of these missions showed suggestions of an anomalous acceleration of order $10^{-7}$ cm s$^{-2}$.  However, \\anderson\\ considered the determination of the anomalous acceleration of the Pioneer 10 spacecraft to be the most reliable.  Therefore I have focussed exclusively on the Pioneer 10 data for my analysis.  Once I had verified the presence of an anomalous acceleration, I tested the result for robustness in several different ways.  All of the procedures discussed in this paper were developed by myself \\cite{fdf} and written in the Interactive Data Language (IDL) \\cite{rsinc,cmlibrary}.  During the development I had only minimal contact with the \\anderson\\ group authors, and as I detail below, these contacts had a minimal impact.  Thus, I consider this work to be an independent test of the analysis by \\anderson.  I have analyzed a subset of the Pioneer 10 tracking data that is available from the public archives, which is, of course, the same data that \\anderson\\ used in their analysis.  The time coverage of my analysis (years 1987--1994) is most comparable to that of the original discovery presented in \\anderson.  By necessity, many of the analysis procedures I developed will be at least similar to those of \\anderson, but I attempt to extend the analysis by considering additional models, including spacecraft spin, maneuvers, and time-variation of the anomalous acceleration.  The contents of the paper are as follows.  Section \\ref{sec:sc-comm} briefly describes the Pioneer 10 spacecraft and the Deep Space Network systems.  Section \\ref{sec:dataprep} presents the methods I used to acquire the data and perform initial filtering, while Sec. \\ref{sec:anal} describes the analysis and modeling techniques that I employed.  The results of the tracking and uncertainty analyses are presented in Sections \\ref{sec:results} and \\ref{sec:uncertain}. This is followed by a short discussion and conclusion in Sections \\ref{sec:discussion} and \\ref{sec:conclusion}. ",
        "conclusions": "}  I have confirmed by independent analysis that the Pioneer 10 anomalous acceleration exists in the Doppler tracking data, and is likely not to be an artifact of the software processing by \\anderson.  Direct comparison to \\anderson's SIGMA acceleration value in their Interval I yields agreement at better than the 1\\% level.  The anomaly is robust to different choices of spacecraft spin model, and also produces a consistent value even when all maneuvers are removed.  This data does not constrain whether anomalous acceleration is geocentric or heliocentric.  By including a jerk term, I have showed that the acceleration is reasonably constant as a function of time over a 7.5 year time baseline, but not constant enough to rule out thermal radiation effects due to radioactive decay of Plutonium on board the spacecraft."
    },
    "0208/astro-ph0208001_arXiv.txt": {
        "abstract": "Binary supermassive black holes are produced by galactic mergers as the black holes from the two galaxies fall to the center of the merged system and form a bound pair. The two black holes will eventually coalesce in an enormous burst of gravitational radiation. Here we show that the orientation of a black hole's spin axis would change dramatically even in a minor merger, leading to a sudden flip in the direction of any associated jet. We identify the winged or X-type radio sources with galaxies in which this has occurred. The implied coalescence rate is similar to the overall galaxy merger rate, suggesting that the prospects are good for observing gravitational waves from coalescing supermassive black holes. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0208/astro-ph0208237_arXiv.txt": {
        "abstract": "We first discuss why the uncomfortable fine-tuning of the parameters of the $\\Lambda$-CDM cosmological model provides continuing, strong motivation to investigate Hubble's Constant. Then we review  evidence from the HST Key Project that there is a significant scale error between raw Cepheid and Tully-Fisher distances.  An analysis of mainly HST Distance Scale Key Project data shows a correlation between host galaxy metallicity and the rms scatter around the Cepheid P-L relation, which may support a recent suggestion that the P-L metallicity dependence is stronger than expected. If Cepheids do have a significant metallicity dependence then the Tully-Fisher scale error increases and the distances of the Virgo and Fornax clusters extend to more than 20Mpc, decreasing the value of H$_0$. Finally, if the Cepheids have a metallicity dependence then so do Type Ia Supernovae since the metallicity corrected Cepheid distances to eight galaxies with SNIa would then suggest that the SNIa peak luminosity is fainter in metal poor galaxies, with important implications for SNIa estimates of q$_0$ as well as H$_0$. ",
        "introduction": "One major motivation for studying Hubble's Constant is the complicated nature of the current standard model in cosmology, $\\Lambda$-CDM. In this model, to order of magnitude, $\\Omega_{baryon}\\approx \\Omega_{CDM}\\approx \\Omega_{\\Lambda}$ and this seems unnatural. The coincidence between the CDM and Baryon densities worried some authors (Peebles 1984, Shanks 1985) when CDM was first postulated. The coincidence between $\\Omega_{\\Lambda}$ and the others worried many more (eg Dolgov, 1983, Peebles and Ratra, 1988 and Wetterich, 1988). These fine-tuning problems of the standard model are compounded by the fact that the inflation model on which the standard model sits, was partly based on a fine-tuning argument, the flatness-problem; to begin by eliminating one fine tuning problem only to end up with several gives the appearance of circular reasoning!  Shanks (1985, 1991, 1999, 2001) noted that a simpler  model immediately became available if H$_0$ actually lay below 50 kms$^{-1}$ Mpc$^{-1}$. An inflationary model with $\\Omega_{baryon}$=1 is then better placed to escape the baryon nucleosynthesis constraint. Simultaneously, the low value of H$_0$ means that the X-ray gas in the Coma cluster increases towards the Coma virial mass and the lifetime of an Einstein-de Sitter Universe extends to become compatible with the ages of the oldest stars. Now this model does predict a first acoustic peak in the CMB anisotropy at around l$\\approx$330 which disagrees with the position of the first peak at l=220$\\pm$10 found  by the Boomerang experiment (Netterfield et al, 2002). However, the above fine-tuning problems of the $\\Lambda$-CDM model plus the historical tendency for later data to overturn early confirmations of  previous `concordance' models such as the isocurvature model in the early 1980's and the SCDM model in the early 1990's suggests that  it may be best, (a) to wait for MAP to confirm the Boomerang results before abandoning other models and (b) to continue to study Hubble's Constant.  \\begin{figure} \\plotfiddle{fig1.eps}{2.5in}{0}{40}{40}{-120}{-65} \\caption{ A comparison between HST Cepheid and TF distances which suggests that TF distances show a significant scale-error with the TF distance to galaxies at the distance of the Virgo cluster being underestimated by 22$\\pm$5.2\\%. The dashed line shows the best fit with $(m-M)_{TF}= 0.915\\pm0.036\\times(m-M)_{Ceph}+2.204$.} \\end{figure} ",
        "conclusions": "Our conclusions are as follows:-  \\begin{itemize}  \\item Key Project  HST Cepheid distances imply Tully-Fisher distances at Virgo/Fornax are underestimated by $\\approx22\\pm5$\\%, reducing H$_0$ from $\\approx$85 to $\\approx$65kms$^{-1}$Mpc$^{-1}$.  \\item TF distances may be Malmquist biased,  suggesting there may be a bigger TF scale error at larger distances.  \\item If the UV excess of F stars in open cluster NGC7790 is caused by low metallicity then  Cepheids have  a strong metallicity dependence,  $\\Delta M \\approx -0.66  \\Delta[Fe/H]$.  \\item  Current HST Cepheid distances may be significantly underestimated at Virgo/Fornax redshifts due to metallicity and magnitude incompleteness  bias, implying that values of H$_0<$50kms$^{-1}$Mpc$^{-1}$  may still not be ruled out.  \\item If Cepheids have a strong metallicity dependence then so have SNIa . Thus significant metallicity corrections may need to be applied to the Type Ia Hubble Diagram before reliable estimates of q$_0$ or H$_0$ can be made.   \\end{itemize}"
    },
    "0208/astro-ph0208147_arXiv.txt": {
        "abstract": "The distances and \\hi\\ contents of 161 spiral galaxies in the region of Virgo cluster are used to gain insight into the complicated structure of this galaxy system. Special attention has been paid to the investigation of the suggestion presented in an earlier work that some peripheral Virgo groups may contain strongly gas-deficient spirals.  The three-dimensional galaxy distribution has been inferred from quality distance estimates obtained by averaging distance moduli based upon the Tully-Fisher relationship taken from eight published datasets previously homogenized, resulting in a relation with a dispersion of 0.41~mag. Previous findings that the spiral distribution is substantially more elongated along the line-of-sight than in the plane of the sky are confirmed by the current data. In addition, an important east-west disparity in this effect has been detected. The overall width-to-depth ratio of the Virgo cluster region is about 1\\,:\\,4, with the most distant objects concentrated in the western half. The filamentary structure of the spiral population and its orientation are also reflected by the \\hi-deficient objects alone. The \\hi\\ deficiency pattern shows a central enhancement extending from \\sm16 to 22 Mpc in line-of-sight distance; most of this enhancement arises from galaxies that belong to the Virgo cluster proper. However, significant gas deficiencies are also detected outside the main body of the cluster in a probable group of galaxies at line-of-sight distances \\sm25--30 Mpc, lying in the region dominated by the southern edge of the M49 subcluster and clouds W$^\\prime$ and W, as well as in various foreground galaxies. In the Virgo region, the \\hi\\ content of the galaxies then is not a straightforward indicator of cluster membership. ",
        "introduction": "This article is the continuation of a series of papers on the \\hi\\ content of spirals from the 21 cm line of neutral hydrogen data devoted to examine the extent to which the cluster environment influences the evolution of the galaxies.  The starting point, developed in \\citet*[hereafter Paper~I]{SGH96}, was the establishment of reliable standards of \\hi\\ mass for the various morphological subgroups of luminous spirals from a complete \\hi-flux-limited sample of these galaxy types in low density environments. That work was followed by a second paper examining the possible connections between gas deficiency and the properties of both the underlying galaxies and their environment in the fields of eighteen nearby clusters by \\citet*[hereafter Paper~II]{Sol01}. The main motivation was to gain insight into the mechanisms responsible for the atomic gas depletion. While no clearly discriminating circumstances were found among those clusters which show significant \\hi\\ deficiency and those which do not, this work definitely confirmed previous findings \\citep[e.g.,][]{GH85,HG86,Mag88} that in \\hi-deficient clusters the proportion of gas-poor spirals increases monotonically towards the center. Moreover, \\citeauthor{Sol01} clearly demonstrated, as first suggested by \\citet*{Dre86}, that \\hi-deficient objects move on orbits more radial than those of their gas-rich counterparts. This result made a strong case for the ram-pressure stripping of the spirals by the hot X-ray emitting intracluster medium (ICM) as the most likely process responsible for the gas deficiencies observed in rich cluster environments.  The wealth of 21-cm data gathered for the Virgo region in \\citeauthor{Sol01} also made it possible to examine the distribution in two-dimensional space of the neutral gas deficiency in the Virgo central area. The sky distribution of \\hi\\ deficiency was found to be in overall agreement with the radial pattern characteristic of rich clusters, showing that the maximum depletion occurred at the cluster center. But quite unexpectedly, the same map of the HI deficiency pattern, a variation of which is produced here as Fig.~\\ref{virgohiam}, also revealed peripheral groups of galaxies with a dearth of atomic hydrogen but found in areas where the density of X-ray luminous gas is very low, raising into question the feasibility that the ram-pressure of the ICM was responsible for the observed \\hi\\ deficiency.  In the current paper, we conduct a further investigation into the nature of and conditions within the three-dimensional structure of the Virgo region, by incorporating into the analysis the \\hi\\ content of its spiral population. The proximity of the region under study facilitates the gathering of a large number of 21-cm single-dish observations, which we complement with a large number of Tully-Fisher \\citetext{\\citeyear{TF77}, hereafter TF} distance estimates also reported in the literature. Section~\\ref{data} presents a catalog of 161 galaxies with good \\hi\\ and TF distance measurements. After reviewing in Section~\\ref{hidef} the manner in which the \\hi\\ deficiency is calculated, our galaxy sample is used in the following two sections to discuss, first the radial pattern of \\hi\\ deficiency, and then its spatial distribution in the Virgo region. We conclude with a summary and some remarks in Section~\\ref{conclusions}. ",
        "conclusions": "\\label{conclusions}  In this paper, we have examined in more detail the suggestion presented in \\citeauthor{Sol01} that, in addition to the main galaxy concentration around M87, some of the well-known peripheral Virgo groups also contain strongly gas-deficient spirals. The overall distribution of \\hi\\ deficiency in the Virgo region has been compared with the three-dimensional galaxy distribution. The following conclusions have been reached:  (1) We confirm that the distribution of the spirals in the Virgo I cluster region is very elongated along the LOS; the galaxies associated with this region have LOS distances raging from less than 10 to more than 50 Mpc. The projected sky distribution of the Virgo spirals, however, looks (lumpy but) relatively compact, with a typical extent of only about 10 Mpc. The overall width-to-depth ratio is approximately 1\\,:\\,4, although with a strong east-west variability. The most distant objects concentrate in the western quadrant, while in the eastern half few spirals are seen at LOS distances larger than 25 Mpc. The Virgo filamentary structure appears to split into two branches around the W$^\\prime$ cloud region.  (2) The distribution of spiral galaxies with significant \\hi\\ deficiency is also characterized by great depth along the LOS. The highly gas-deficient spirals tend to concentrate along the upper branch of the spiral galaxy distribution, which is roughly aligned with the principal axis of the Virgo cluster.  (3) Within 4 Mpc of M87, the measured \\hi\\ deficiency is essentially a monotonically decreasing function of the distance from that galaxy, in agreement with the behavior observed in other \\hi-deficient clusters. Moreover, in the Virgo region, significant \\hi\\ deficiency enhancements are also identified at large distances from the Virgo core, well beyond the typical distance where the hot X-ray emitting ICM is concentrated. Tests of whether locally-high peripheral gas deficiencies are a rather common feature in cluster regions must await the equally-careful tracing of the \\hi\\ deficiency, incorporating quality 3D distance measures around other clusters.  (4) While the principal peak in the distribution of \\hi\\ deficiency arises from numerous gas-poor galaxies coincident with the core and with LOS distances ranging from \\sm16 to 22 Mpc, other important enhancements of the gas deficiency are associated with several nearby galaxies ($d\\lesssim 15$ Mpc) moving away from the cluster with large relative velocities, and with what appears to be a compact background group of galaxies between \\sm25--30 Mpc, most with roughly the same systemic velocities as the cluster mean, which matches the original definition of the W$^\\prime$ cloud. In addition, we have demonstrated that the localized enhancement in $\\df$ observed in the Virgo sky map around the M cloud position actually arises from several galaxies at very different distances aligned along the LOS and without any physical connection. In agreement with results presented by \\cite*{Dal01}, nothing in our analysis suggests that TF distance measurements are unreliable in objects with severe gas depletion.  Further progress in (1) and (2) needs a careful revision of TF distances ---at least until Cepheid distance measurements in Virgo galaxies become more commonplace. Even after the elimination of systematic differences among published Virgo catalogs, a few galaxies still exhibit strongly inconsistent distance measurements: 16 of the 161 members of the 21-cm sample have $1\\sigma$ uncertainties larger than 5 Mpc. Nevertheless, although the details may be questioned, the general picture reporting the elongated structure of the distribution of both the spirals and their \\hi\\ deficiency, as well as the clumped nature of the latter, should be correct. We can make a simple estimate of the \\emph{typical} elongation introduced by the uncertainty in the distances. The standard deviation of individual distances ($0.29$ mag) is likely responsible for an increase of \\sm40$\\%$ in the scale of the true LOS distance distribution by assuming an average distance of the spiral galaxies of 20 Mpc. Given that the absolute errors in distance increase with the values of this quantity, as it is obvious from inspection of Figure~\\ref{hivsd}, one can expect an artificial increment of the depth by a somewhat larger factor for the most distant galaxies which, in any event, would be clearly insufficient to account for the very strong LOS elongation of the galaxy distribution.  On the other hand, results (3) and (4) have profound implications on our understanding of the gas removal events and the influence of the environment on the life of the galaxies. While the characteristics exhibited by the \\hi\\ deficiency in cluster centers tend to support the interaction between the galaxies and the hot intracluster gas as the main cause of their gas depletion, our finding that a number of spirals with substantial \\hi\\ deficiencies lie at large radial distances from the Virgo cluster center is hard to reconcile with the proposition that this environmental process is also the cause. At this stage, it would be desirable to investigate whether these peripheral deficient objects have been produced \\emph{in situ} by alternative gas deficiency mechanisms, such as galaxy-galaxy interactions, predicted to operate in galaxy groups. In this sense, it would be of importance to perform multi-wavelength observations of the \\hi-deficient subclump detected in the Virgo background. Model calculations show that tidal stresses in disks generate extended tail structures in the stellar and neutral hydrogen distributions, the latter with surface densities well above the detection threshold of the most sensitive aperture-synthesis radio observations. In contrast, gas depletion arising from the ram-pressure sweeping of the interstellar medium should produce a dearth of atomic gas in the outer portions of the disks, as well as bow shocks and dense gaseous tails observable in X-rays."
    },
    "0208/astro-ph0208371_arXiv.txt": {
        "abstract": "We have examined {\\sl Chandra} observations of the recently discovered X-ray thread G0.13-0.11 in the Galactic center Radio Arc region. Part of the {\\sl Chandra} data was studied by Yusef-Zadeh, Law, \\& Wardle (2002), who reported the detection of 6.4-keV line emission in this region. We find, however, that this line emission is {\\sl not} associated with G0.13-0.11. The X-ray spectrum of G0.13-0.11 is well characterized by a simple power law with an energy slope of 1.8$^{+0.7}_{-0.4}$ (90\\% confidence uncertainties). Similarly, the X-ray spectrum of the point-like source embedded in G0.13-0.11 has a power law energy slope of 0.9$^{+0.9}_{-0.7}$. The 2 -- 10 keV band luminosities of these two components are $\\sim 3.2\\times 10^{33}{\\rm~ergs~s^{-1}}$ (G0.13-0.11) and $\\sim 7.5 \\times10^{32} {\\rm~ergs~s^{-1}}$ (point source) at the Galactic center distance of 8 kpc. The morphological, spectral, and luminosity properties strongly indicate that G0.13-0.11 represents the leading-edge of a pulsar wind nebula, produced by a pulsar (point source) moving in a strong magnetic field environment. The main body of this pulsar wind nebula is likely traced by a bow-shaped radio feature, which is apparently bordered by G0.13-0.11 and is possibly associated with the prominent nonthermal radio filaments of the Radio Arc. We speculate that young pulsars may be responsible for various unique nonthermal filamentary radio and X-ray features observed in the Galactic center region. ",
        "introduction": "The region around the dynamic center of our Galaxy is very active recently in massive star formation (e.g., Figer et al. 1999), which should have yielded various high-energy products such as supernova remnants (SNRs) and neutron stars (e.g., Morris \\& Serabyn 1996; Cordes \\& Lazio 1997; Wang, Gotthelf \\& Lang 2002). Young and fast-rotating neutron stars, in particular, may be observable as pulsars, although none is yet known within $1^\\circ$ radius of the Galactic center (GC). Detection of radio pulsars in the GC region is difficult, because of the severe radio wave scattering by intervening interstellar ionized gas (Cordes \\& Lazio 1997). On the other hand, two of the polarized radio nonthermal filaments (NTFs; in the Radio Arc and G359.96+0.09) show flat or slightly rising positive spectral indices --- a characteristic of Crab-like pulsar wind nebulae (PWNe; e.g., Anantharamaiah et al. 1991). {\\sl Chandra} observations have further revealed various diffuse X-ray filaments with unusually hard spectra, which are also signatures of PWNe. However, no specific link has so far been proposed between such radio/X-ray features and PWNe.  Here we report a strong candidate for a PWN that links an X-ray thread G0.13-0.11 and the prominent NTFs in the Radio Arc region (Galactic longitude $l \\approx 0^\\circ.2$; Fig.\\ 1; Yusef-Zadeh, Morris, \\& Chance 1984). This X-ray thread G0.13-0.11 was first apparent in the images of Yusef-Zadeh et al. (2002). The diffuse X-ray emission from the neighboring molecular cloud G0.13-0.13 (Oka et al. 2001) has been further investigated by Yusef-Zadeh, Law, \\& Wardle (2002). They considered the X-ray thread as part of a large-scale diffuse feature that emits the 6.4-keV fluorescent line, which results from the filling of K-shell vacancies of neutral or weakly-ionized irons (Koyama et al. 1996; Wang, Gotthelf, \\& Lang 2002; Wang 2002; Yusef-Zadeh, Law, \\& Wardle 2002). Because the prominent X-ray thread is on the side of the molecular cloud that is opposite to Sgr A$^*$, they concluded that the 6.4-keV line emission could not represent the reflection of a possible recent radiation burst from the central massive black hole. However, our examination of related {\\sl Chandra} observations shows that the 6.4-keV line is clearly not associated with the X-ray thread G0.13-0.11, although the molecular cloud is indeed a strong 6.4-keV line emitter (e.g., Fig.\\ 2; Wang 2002). We have conducted morphological and spectral analyses of both G0.13-0.11 and an embedded point-like source. This X-ray study, together with an investigation of the radio emission from the region, has led us to conclude that G0.13-0.11 most likely represents a PWN.  \\begin{figure*}[!hbt] \\unitlength1.0cm \\begin{picture}(16,8) \\put(-1.1,0){ \\begin{picture}(8,8) \\psfig{figure=f1a.eps,height=8cm,angle=90, clip=} \\end{picture} } \\put(7.8,0){ \\begin{picture}(8,8) \\psfig{figure=f1b.eps,height=8cm,angle=90, clip=} \\end{picture} } \\end{picture} \\caption{The left panel shows an overview of the region surrounding the X-ray thread G0.13-0.11: VLA 6-cm radio continuum image (\\S 5) and the {\\sl Chandra} ACIS-I intensity contours in the 2 -- 6 keV band. The square box illustrates the field of the close-up shown in the right panel. The 6-cm image is constructed from the combined CnB and DnC data and has a resolution of $\\sim$5\\arcsec. The X-ray image is adaptively smoothed, using the CIAO routine CSMOOTH with the smoothing kernel determined to achieve a signal-to-noise radio of 2.5-4. The contours are at 31, 33, 37, 45, 61, 93, 158, and 328 $\\times 10^{-3} {\\rm~counts~s^{-1}~arcmin^{-2}}$. } \\end{figure*}  \\begin{figure*}[!thb] \\centerline{\\psfig{figure=f2.eps,height=6in,angle=0,clip=}} \\caption{A close-up of the region around G0.13-0.11. The image represents the 2 -- 6 keV intensity distribution, which has been smoothed with a Gaussian of size $\\sim 1^{\\prime\\prime}$. The contours represent the smoothed distribution of 6.4-keV line emission at levels of 0.8, 0.9, 1.1, 1.4, 1.8, and 2.3 $\\times 10^{-3} {\\rm~counts~s^{-1}~arcmin^{-2}}$ (Wang 2002). Regions for the X-ray spectral analysis (\\S 3) are also outlined. } \\end{figure*}  \\begin{figure} [!bht] \\centerline{\\psfig{figure=f3.eps,height=6in,angle=90,clip=}} \\caption{ACIS-I effective exposure map in the same field as in Fig.\\ 2. The grey-scale represents the exposure, which ranges from 25 ks to 111 ks.} \\end{figure} ",
        "conclusions": "We have studied the X-ray thread G0.13-0.11 and the embedded point-like X-ray source \\xs\\ as well as the adjacent nonthermal radio emission in the Radio Arc region. The morphological and spectral properties of these features appear to be consistent with our PWN interpretation. We hope that this study will stimulate observational and theoretical studies to further explore the connection between pulsars and various nonthermal features observed in the unique GC environment. The ultimate confirmation of the PWN interpretation requires the detection of the pulsed signal from the putative pulsar, which could be achieved with future radio and/or X-ray observations with fast timing capabilities."
    },
    "0208/hep-ph0208191_arXiv.txt": {
        "abstract": "{ We point out that there is no cosmological gravitino problem in a certain class of gauge-mediated supersymmetry-breaking (GMSB) models. The constant term in the superpotential naturally causes small mixings between the standard-model and messenger fields, which give rise to late-time decays of the lightest messenger fields. This decay provides an exquisite amount of entropy, which dilutes the thermal relics of the gravitinos down to just the observed mass density of the dark matter. This remarkable phenomenon takes place naturally, irrespective of the gravitino mass and the reheating temperature of inflation, once the gravitinos and messenger fields are thermalized in the early Universe. In this class of GMSB models, there is no strict upper bound on the reheating temperature of inflation, which makes the standard thermal leptogenesis  the most attractive candidate for the origin of the observed baryon asymmetry in the present Universe. }   \\end{center} \\end{titlepage}  \\renewcommand{\\thefootnote}{\\arabic{footnote}} \\setcounter{footnote}{0} ",
        "introduction": "% The minimal supersymmetric standard model (MSSM) is the most promising candidate for physics beyond the Standard Model (SM), since it naturally solves the ``hierarchy problem'' and leads to a successful unification of the gauge coupling constants~\\cite{coupling-unification}. Because SUSY is not observed in the real world, it should be broken around the TeV scale. Once we allow generic soft SUSY-breaking terms, we must face hundreds of new parameters, which makes the rate of flavour-changing neutral-current (FCNC) interactions many orders of magnitude larger than the present experimental bounds. In order to obtain a successful low-energy effective theory, various mediation mechanisms of SUSY-breaking effects have been proposed.   The scenario most commonly considered in phenomenology is the minimal gravity-mediated SUSY-breaking (mSUGRA) models.  This scenario is very simple and aesthetically  attractive, since gravity does exist in nature. In the mSUGRA models, we also have a promising candidate for  dark matter. That is the lightest SUSY particle (LSP), which is usually the lightest neutralino. The severest difficulty in this scenario is the lack of a natural explanation for the suppression of the FCNC interactions. A specific form of the soft SUSY-breaking masses must be imposed by hand in order to suppress the FCNC interactions.  It has been argued that the suppression of the FCNC interactions can be naturally obtained in brane-world SUSY-breaking scenarios, such as anomaly-~\\cite{AMSB} and gaugino-~\\cite{gMSB} mediated SUSY-breaking models. In these scenarios, the fields relevant to SUSY breaking are assumed to reside on the hidden brane, which is geometrically separated from the visible brane where the SM fields are localized. Recently, however, a crucial observation has been made in Ref.~\\cite{Dine}, where  the authors have found that the separation of the visible and hidden branes in a higher-dimensional space-time is not sufficient for suppressing the FCNC interactions. Consequently, we need additional {\\it ad hoc} assumptions in those models.~\\footnote{There is an attempt to realize anomaly-mediation models in a four-dimensional framework~\\cite{s-conformal1}.}   At present, the most attractive scenario seems to be the gauge-mediated SUSY breaking (GMSB)~\\cite{GMSB}. In the GMSB models, the suppression of the FCNC interactions is realized in an automatic way, just because SUSY breaking occurs at a very low-energy scale. Furthermore, a whole spectrum of the superparticles in the MSSM sector is completely determined by only a few parameters, which allows us to discriminate the GMSB models from other candidates in the future collider experiments.  Unfortunately, from a cosmological perspective, there exists a big difficulty in the GMSB models. That is the so-called cosmological gravitino problem. In these models, we have no natural explanation for the dark matter in the present Universe. Thermal relics of the gravitino, the LSP in the GMSB models, overclose the Universe once they are thermalized in the early Universe. In order to avoid the overproduction of the gravitinos, there is a severe upper bound on the reheating temperature of inflation $T_{R}$, which is about $T_{R}\\lsim 10^6\\GEV$ for $m_{3/2}=10\\MEV$ for instance, and it even reaches $T_{R}\\lsim 10^3\\GEV$ in the case of the lighter gravitinos $m_{3/2}\\lsim 100\\KEV$~\\cite{gravitino-in-GMSB}. Furthermore, we have to fine-tune the reheating temperature just below this upper bound to explain the required mass density of the dark matter.  We also have to generate the observed baryon asymmetry at just the same reheating temperature. Therefore, for successful cosmology in the GMSB models, we need incredible fine-tunings of various parameters, which apparently belong to independent physics, such as SUSY breaking, inflation and baryo/leptogenesis.~\\footnote{If there exist extra matter multiplets of a SUSY-invariant mass of the order of the ``$\\mu$-term'', the observed baryon asymmetry and gravitino dark matter can be simultaneously explained in a way totally independent of the reheating temperature~\\cite{F-Y-extra}.} This is a big  drawback of the GMSB models with respect to the standard mSUGRA scenario.  In this letter, we point out that there are indeed  no such difficulties in a certain class of GMSB models. We consider direct gauge-mediation models where the SUSY-breaking effects are directly transmitted to the messenger sector without loop suppressions~\\cite{direct-GMSB}. In this class of  models, the specific form of the superpotential of the messenger sector is usually provided by the $R$-symmetry: since it is violated by the constant term in the superpotential $\\vev{W}$, which is anyhow required to cancel the cosmological constant, it is quite natural to expect that there are small mixings between the SM and messenger multiplets induced by the condensation $\\vev{W}$. As a result, the lightest messenger particle decays into the SM particle and gaugino through the mixings. As we will see, the resultant late-time decays of the lightest messengers provide an exquisite amount of entropy, which dilutes the thermal relics of the gravitinos down to just the observed mass density of the dark matter in the present Universe.  Surprisingly, this miracle turns out to be true almost irrespective of the mass of the gravitino and the reheating temperature, once the gravitinos and messenger particles are thermalized in the early Universe. Consequently, by assuming natural mixings between the SM and messenger fields, the severe upper bound on the reheating temperature is completely eluded. This fact makes it much easier to construct a realistic inflationary scenario in the GMSB models. This result also has an important implication on the origin of the observed baryon asymmetry. We will see that the standard thermal leptogenesis, through out-of-equilibrium decays of the  right-handed Majorana neutrinos~\\cite{Fuku-Yana}, is now the most attractive mechanism to generate the observed baryon asymmetry in the GMSB models. ",
        "conclusions": "% In this letter, we have pointed out that there is no fine-tuning problem to obtain the required mass density of the dark matter in a certain class of GMSB models. By virtue of the small mixing between the SM fields, which is induced by the $R$-symmetry-breaking effects, the lightest messenger particle has a finite lifetime and provides an exquisite amount of entropy, which dilutes the thermal relics of the gravitinos down to just the mass density required for  the dark matter. This phenomenon takes place naturally, regardless of the gravitino mass and the reheating temperature of inflation as long as the gravitinos and messenger fields are thermalized in the early Universe. There is no severe upper bound on the reheating temperature in this class of GMSB models, which makes the standard thermal leptogenesis very attractive as the origin of the observed baryon asymmetry.  The present scenario should have important implications also on other candidates for the origin of the present baryon asymmetry. It would be very interesting to reanalyse those models with the disappearance of the cosmological gravitino problem taken into account.  \\small"
    },
    "0208/astro-ph0208587_arXiv.txt": {
        "abstract": "We report in this {\\it paper} the ASCA discovery of the first (to our knowledge) radio-loud Active Galactic Nucleus (AGN) covered by a Compton-thick X-ray absorber, in the GigaHertz Peaked Spectrum radio source OQ+208. It represents one of the few available direct  measurements of dense matter in the nuclear environment of this class of sources, which may provide the confining medium to the radio-emitting region if GPS sources are indeed \"frustrated\" classical radio doubles. The perspective of future studies  with XEUS are discussed. ",
        "introduction": "GigaHertz Peaked Spectrum radio sources are a class of powerful ($L_{radio} \\sim 10^{45}$~erg~s$^{-1}$) radio sources, defined by a simple convex spectrum peaking near 1 GHz. They represent $\\simeq$10\\% of the 5~GHz selected sources, {\\it i.e.} a significant fraction of the powerful radio sources in the universe. They are characterized by compact radio cores, most likely not extending beyond the narrow-line regions (NLRs; $\\le$1 kpc). Some of them exhibit very faint extended radio emission on scales larger than the host galaxy (Baum et al. 1990; Stanghellini et al. 1990; Fanti et al. 2001), but rarely on Mpc scales (Schoenmakers et al. 1999). Their radio morphologies and host galaxies are generally consistent with classical 3CR doubles. Emission-lines optical spectra suggest interaction between the radio source and emission-line gas, as well as dust obscuration. Mid-IR measurements suggest the existence of a powerful hidden active nucleus (Heckman et al. 1994: see, however, de Vries et al. 1998 and Fanti et al. 2000 for a different point of view).  Two scenarios are currently proposed to explain the nature of GPS sources:  \\begin{enumerate}  \\item the {\\it young scenario}: GPS could be young versions of large scale radio galaxies at an earlier stage of their formation (Carvalho 1985: Mutel \\& Philips 1988; Fanti et al. 1995; Readhead et al. 1996: O'Dea \\& Baum 1997). In this scenario, objects which are $\\sim$100 pc in size are about 10$^4$~years old, which is consistent with proper motion measurements on about 10 of them (Fanti 2000)  \\item the {\\it frustrated scenario}: GPS may represent \"aborted\" classical doubles, which will never reach their full maturity because they are embedded in a dense and turbulent medium, able to confine and trap the radio-emitting region on the scale-length of the NLRs (van Brueghel et al. 1984; O'Dea et al. 1991)  \\end{enumerate}  X-ray observations can provide an important contribution to elucidate the nature of this class of objects. In facts:  \\begin{itemize}  \\item measurements of hot gas, through its optically thin emission peaking in the soft X-rays, may provide indication for the presence of hot confining medium (O'Dea et al. 1996) . Constraints on the presence of such a gas phase from other wavelengths are not conclusive (Kameno et al. 2000; Marr et al. 2001)  \\item measurements of heavy X-ray absorption (in the most extreme case Compton-thick, $N_H > \\sigma_T^{-1} \\simeq 10^{24}$~cm$^{-2}$) may indicate the presence of cold confining matter, to be compared with hydrodynamical models (De Young 1993; Carvalho 1994, 1998) to identify possible mechanisms responsible for the confinement of the jet  \\item the detection of large-scale X-ray jets may challenge one or more of our current assumptions on the nature of GPS sources (Siemiginowska et al. 2002)  \\end{itemize}  The properties of the cold and hot phases, that one can derive from the X-ray spectral fitting may therefore provide a test for the \"frustration\" scenario.  Unfortunately, GPS sources are X-ray weak: only 3 out of 9 GPS sources observed in the soft X-rays have been detected. Hints that this low rate may be due to absorption were put forward by Elvis et al. (1994) and Zhang \\& Marscher (1994). O'Dea et al. (1996) report the detection of a luminous ($L_X \\sim 2 \\times 10^{43}$~erg~s$^{-1}$) and highly absorbed ($N_H \\simeq 4 \\times 10^{22}$~cm$^{-2}$) hard X-ray source in 1314+125 ($z = 0.122$). Siemiginowska et al. (2002) recently discovered with {\\it Chandra} a prominent X-ray jet on scales $\\sim 300 h^{-1}_{50}$ in PKS~1127-145 ($z = 1.187$). ",
        "conclusions": ""
    },
    "0208/astro-ph0208064_arXiv.txt": {
        "abstract": "We present the final installment of an intensive 13-year study of variations of the optical continuum and broad \\Hbeta\\ emission line in the Seyfert 1 galaxy NGC~5548. The data base consists of 1530 optical continuum measurements and 1248 \\Hbeta\\ measurements. The \\Hbeta\\ variations follow the continuum variations closely, with a typical time delay of about 20 days. However, a year-by-year analysis shows that the magnitude of emission-line time delay is correlated with the mean continuum flux. We argue that the data are consistent with the simple model prediction between the size of the broad-line region and the ionizing luminosity, $r \\propto L_{\\rm ion}^{1/2}$. Moreover, the apparently linear nature of the correlation between the \\Hbeta\\ response time and the nonstellar optical continuum $F_{\\rm opt}$ arises as a consequence of the changing shape of the continuum as it varies, specifically $F_{\\rm opt} \\propto F_{\\rm UV}^{0.56}$. ",
        "introduction": "The nature of the broad-line region (BLR) is arguably one of the major remaining mysteries in active galactic nuclei (AGNs). Whereas the supermassive black hole/accretion-disk paradigm has become increasingly more secure, there is still no consensus about the origin of the broad emission lines that are prominent features of the UV/optical spectra of these sources. This is not to say that progress has not been made. Years of spectroscopic study have uncovered a rich phenomenology (see Sulentic, Marziani, \\& Dultzin-Hacyan 2000 for a recent review).  Of particular importance has been the recognition that the emission-line fluxes vary in response to continuum variations, with a small time delay (days to weeks for Seyfert galaxies) due to light travel-time effects within the BLR. Well before these time delays were first accurately measured, this led to the seminal paper on ``reverberation mapping'' (Blandford \\& McKee 1982), a tomographic method of determining the structure and kinematics of the BLR. While the full potential of reverberation mapping has yet to be realized, use of this technique has provided BLR sizes in approximately three dozen AGNs (see compilations by Wandel, Peterson, \\& Malkan 1999 and Kaspi et al.\\ 2000). Even more importantly, measurement of the emission-line time delays (or ``lags''), combined with measurements of the line width, have led to estimates of the masses of the central black holes.  In late 1988, we began a program of spectroscopic monitoring of optical variations in the Seyfert 1 galaxy NGC~5548. This activity was organized by an informal consortium called the ``International AGN Watch'' (Alloin et al.\\ 1994; Peterson 1999). Initially, the optical monitoring program was undertaken in support of an ultraviolet monitoring campaign carried out with the {\\it International Ultraviolet Explorer} (\\IUE\\,; Clavel et al.\\ 1991 [the first paper in the same series as this work, and hereafter referred to as Paper I]). The resulting optical data set (Peterson et al.\\ 1991, hereafter Paper II) proved to be so rich that it was decided to continue this program. Since that time, a number of additional contributions to this series (Peterson et al.\\ 1992 [Paper III]; Peterson et al.\\ 1994 [Paper VII]; Korista et al.\\ 1995 [Paper VIII]; and Peterson et al.\\ 1999 [Paper XV]) have described our continuing optical monitoring program. A complete summary of the work on NGC~5548 by this consortium appears in Table 1. Reviews of the progress of this program are also available (e.g., Peterson 1993, 2001; Netzer \\& Peterson 1997).  This paper represents the final installment of our 13-year program of optical monitoring of NGC~5548. As is customary for the papers in this series, we will focus primarily on the data and the salient results of cross-correlation analysis of the continuum and emission-line light curves. Further in-depth analysis will be left to subsequent papers. In \\S\\,2, we describe the observations, data reduction, and intercalibration procedures that we have used to construct a homogeneous data base of optical continuum and \\Hbeta\\ emission-line fluxes. In \\S\\,3, we describe the time-series analysis that we have undertaken to determine the time scale for response of \\Hbeta\\ to continuum variations. In \\S\\,4, we discuss some of the implications of our study. Our results are summarized in \\S\\,5. We also note that some of the data presented here have already appeared in other contexts, specifically as part of a short but intensive monitoring program carried out in 1998 June (Dietrich et al.\\ 2001) and as part of a long-term photometric monitoring campaign (Doroshenko et al.\\ 2001). ",
        "conclusions": "It was noted in Paper XV that statistically significant year-to-year changes in the \\Hbeta\\ lag have occurred, and that these changes are correlated with the mean continuum flux. The additional data described here confirm this result and allow us to investigate it further. In Figure 4,  we plot \\tcent\\ as a function of the mean starlight-corrected continuum flux for each year $\\bar{F}_{\\rm opt} = \\langle F_{\\lambda}({\\rm 5100\\,\\AA})\\rangle - F_{\\rm gal}$, where $F_{\\rm gal}$ is our best estimate of the host-galaxy starlight contribution through our standard aperture, $3.4 \\times 10^{-15}$\\,\\contunits\\ at 5100\\,\\AA\\ (Romanishin et al.\\ 1995); $\\bar{F}_{\\rm opt}$ is thus the mean nonstellar flux from the AGN component only. We add to this a measurement from an earlier monitoring campaign at Wise Observatory (Netzer et al.\\ 1990), adjusted as described in Paper XV. The best-fit power-law relationship between the continuum flux and time lag is $\\tcent \\propto \\bar{F}_{\\rm opt}^{0.95}$, nearly a linear relationship.  This result can be compared with a simple theoretical prediction. Photoionization equilibrium models for the BLR are characterized by (a) the shape of the ionizing continuum incident upon the BLR gas, and (b) an ionization parameter $U$, which is the ratio of ionizing photon density to particle density at the cloud inner face, \\begin{equation} U = \\frac{Q{\\rm (H)}}{4 \\pi r^2 n_{\\rm H} c}, \\end{equation} where $Q{\\rm (H)}$ is the rate at which ionizing photons are produced by the continuum source, $r$ is the separation between the BLR gas and the continuum source, and $n_{\\rm H}$ is the BLR particle density. Since the luminosity in ionizing photons $L_{\\rm ion}$ is proportional to $Q{\\rm (H)}$, we expect that $r^2 \\propto L_{\\rm ion}/U n_{\\rm H}$. Essentially, as $L_{\\rm ion}$ varies, we expect that the response in a particular emission line will be greatest at some particular value of the product $U n_{\\rm H}$, which will thus lead to $r \\propto L_{\\rm ion}^{1/2}$ for any given line; in other words, the response of a particular line to a continuum variation ought to be dominated by gas in which the change in emissivity is greatest. In Figure 3, we show a test of this naive prediction based on  fitting a $r \\propto L^{1/2}$ (i.e., $\\tcent \\propto \\bar{F}_{\\rm opt}^{1/2}$) power law to these data. The resulting fit is quite poor.  The simple model above makes the implicit assumption that the continuum does not change shape as it varies. This allows us to use $L_{\\rm opt}$ as a surrogate for $L_{\\rm ion}$, which cannot be measured directly. However, a flux at an ultraviolet wavelength closer to the Lyman edge at 912\\,\\AA\\ would obviously provide a better surrogate for $L_{\\rm ion}$ than $L_{\\rm opt}$, and such data do exist in archival form, mostly from our previous UV monitoring experiments (Papers I and VIII) on NGC 5548. We have therefore recovered the NEWSIPS-extracted \\IUE\\ spectra of NGC~5548 obtained since late 1989 from the {\\em IUE Final Archive} in order to effect a direct comparison of the UV and optical amplitude of continuum variability. We measured the UV continuum flux $F_{\\rm UV} \\equiv F_{\\lambda}(\\mbox{1350\\,\\AA)}$ by averaging the flux in the observed wavelength range 1370--1380\\,\\AA\\ for all \\IUE\\ SWP spectra obtained within one day of optical continuum measurements. All optical fluxes within one day of a UV continuum measurement were used to form a weighted average optical flux. This procedure yielded 83 pairs of UV/optical continuum measurements. We then fitted a power-law function $F_{\\rm opt} \\propto F_{\\rm UV}^{\\alpha}$ to these data, as shown in Figure 4, yielding a best-fit slope $\\alpha = 0.56$. Combining this with the above relationship between \\tcent\\ and the optical continuum yields $\\tcent \\propto F_{\\rm UV}^{0.53}$, which is consistent with the simple model. We note that there is no statistically significant time delay between the UV and optical continuum variations (Peterson et al.\\ 1998b, and previous papers in this series).  Thus, the apparently linear correlation between the variations in the optical continuum and those in the \\Hbeta\\ emission line is attributable to similar relationships between the UV and optical continuum variations on the one hand and the UV continuum and \\Hbeta\\ emission-line variations on the other. This probably accounts for the apparent absence of scatter (relative to, say, \\Lya\\ or \\civ\\,$\\lambda1549$) in comparisons of the optical continuum and \\Hbeta\\ fluxes of AGNs (e.g., Yee 1980; Peterson 1997).  It is interesting that our observational result agrees with the naive theory once we account for the change in the shape of the continuum between the UV and optical, but that naive theory supposes that the shape of the ionizing continuum remains unchanged. Either the shape of the continuum between the UV and extreme ultraviolet still remains constant, or the \\Hbeta\\ line is surprisingly insensitive to the shape of the continuum.  It should be pointed out that our result differs from the BLR radius--luminosity relationship found by Kaspi et al.\\ (2000), namely $r \\propto L^{0.7}$. The Kaspi et al.\\ result describes how the BLR radius varies from object-to-object as a function of the mean optical luminosity of the source. The relationship discussed here  describes how the BLR radius in an individual object changes as the luminosity of the central source varies with time."
    },
    "0208/astro-ph0208252_arXiv.txt": {
        "abstract": "I review the current state of our understanding of the galaxy formation and evolution process from the modeler's perspective. With the advent of the cold dark matter model and the support of fast computers and advanced simulation techniques, there has been considerable progress in explaining the growth of structure on the largest scales and in reproducing some of the basic properties of galaxies and their evolution with redshift. However, many properties of galaxies are still only poorly understood or appear to be in conflict with the prediction of the cold dark matter model. I discuss in what direction the next generation of galaxy formation models may go and why a large space-based optical-UV telescope could be critical for the calibration and testing of these advanced models. ",
        "introduction": "The past couple of years have witnessed a dramatic increase in the quantity and quality of observations on the formation and evolution of galaxies. Galaxies are routinely identified at redshifts exceeding three and high resolution imaging allows us to study their internal structure. These data are complemented by high resolution spectroscopy of QSO absorption systems that provide further clues on the evolution of baryons in the universe. In fact, this increase has been so rapid that observations have outgrown their theoretical framework. Traditional approaches, which rely heavily on the morphological classification of galaxies and which intend to disentangle the star formation history of galaxies, seem outdated if compared with the much richer structure seen in galaxies at different redshifts.  Motivated by the increasing body of evidence that most of the mass of the universe consist of invisible ``dark'' matter, and by the particle physicist's inference that this dark matter consists of exotic non-baryonic particles, a new and on the long run much more fruitful approach has been developed: rather than to model the formation and evolution of galaxies from properties of present day galaxies, it is attempted to prescribe a set of reasonable initial conditions. The evolution of galaxies is then modeled based on physical processes that are considered to be relevant such as gravity, hydrodynamics, radiative cooling and star formation. The outcome at different epochs is then confronted against observational data. One scenario that has been extensively tested in this way is the model of hierarchical clustering, currently the most successful paradigm of structure formation.  In this contribution I review the main successes but also some of the generic problems of models in which structure forms by hierarchical clustering. I briefly compare the current state of the field with that 10-15 years ago followed by some speculations in which direction the field may develop in the next decade and how a large optical-UV telescope in space may support such developments.   \\begin{figure*} \\epsfig{file=figs/fig1.eps,height=3.2cm} \\caption[]{Time sequence of structure formation in a hierarchical clustering universe, here for the so-called $\\Lambda$CDM model. The four snapshots correspond (from left to right) to redshifts of 9, 3.5, 1 and 0, respectively. The simulation box is 50 Mpc (comoving) on the side.} \\end{figure*} ",
        "conclusions": ""
    },
    "0208/astro-ph0208408_arXiv.txt": {
        "abstract": "The present work  is designed to explore the  evolution of helium-core white dwarf  (He WD)  stars for the  case of metallicities  much lower than the solar metallicity ($Z$= 0.001 and $Z$= 0.0002).  Evolution is followed in a self-consistent way with the predictions of detailed and new non-grey  model atmospheres, time-dependent  element diffusion and the history  of the white  dwarf progenitor.  Reliable  initial models for low mass  He WDs are obtained  by applying mass loss rates  to a 1 M$_\\odot$ stellar model in such  a way that the stellar radius remains close to the Roche lobe radius. The loss of angular momentun caused by gravitational  wave emission  and  magnetic stellar  wind braking  are considered.  Model  atmospheres, based on  a detailed treatment  of the microphysics  entering the  WD atmosphere  (such as  the  formalism of Hummer-Mihalas  to deal with  non-ideal effects)  as well  as hydrogen line  and pseudo-continuum  opacities,  enable us  to provide  accurate colours and magnitudes at both early and advanced evolutionary stages.  We  find that most  of our  evolutionary sequences  experience several episodes of  hydrogen thermonuclear flashes. In  particular, the lower the metallicity, the larger the minimum stellar mass for the occurrence of  flashes  induced  by  CNO  cycle  reactions.  The existence of a mass  threshold for the occurrence of diffusion-induced CNO flashes  leads to  a marked  dichotomy in the  age of  our models. Another  finding of this  study is  that our  He WD  models experience unstable  hydrogen burning via  PP nuclear  reactions at  late cooling stages as a  result of hydrogen chemically diffusing  inwards. Such PP flashes take place in models with very low metal content.  We also find  that  models experiencing CNO flashes exhibit  a pronounced turn-off in  most of their colours at M$_{\\rm V} \\approx$ 16.Finally, colour-magnitude  diagrams for  our models are  presented and compared with  recent observational  data of He  WD candidates  in the globular  clusters  NGC 6397  and  47  Tucanae. ",
        "introduction": "\\label{sec:intro}  Low-mass helium-core white dwarf stars  (He WDs) are thought to be the result of the evolution of certain close binary systems. Mass-transfer episodes  in binary  systems  are required  to  form low  mass He  WDs because an isolated  star would need a timescale  much longer than the present  age  of the  universe  to reach  a  WD  configuration with  a helium-rich interior. Low mass WDs have been detected in large surveys (Bragaglia et  al. 1990; Bergeron, Saffer \\&  Liebert 1992; Bragaglia, Renzini  \\&  Bergeron  1995;  Saffer,  Livio \\&  Yungelson  1998)  and represent an appreciable fraction of the total population of WD stars.  Since  He WDs  began to  be  found in  numerous binary  configurations (Marsh 1995; Marsh, Dhillon \\& Duck 1995; Lundgren et al. 1996; Moran, Marsh  \\& Bragaglia  1997; Orosz  et al.   1999; van  Kerkwijk  et al. 2000), they have  captured the attention of many  researchers who have devoted much effort to their study.  Recent works with the emphasis on the  evolutionary  properties  of  these stars  include  Benvenuto  \\& Althaus (1998), Hansen \\& Phinney (1998), Driebe et al. (1998), Sarna, Ergma \\& Antipova (2000),  Althaus, Serenelli \\& Benvenuto (2001a) and Serenelli et al.  (2001).  In particular, Althaus et al.  (2001a) have explored  their  evolution  in  a  self-consistent  way  with  nuclear burning, time  dependent element diffusion  and the history of  the WD progenitor.   Althaus  et al.   (2001a)  find  that element  diffusion induces thermonuclear  hydrogen shell flashes  in He WDs  with stellar masses greater  than $\\approx$  0.18 M$_\\odot$.  As  a result,  He WDs more massive  than 0.18 M$_\\odot$  are characterized by  thin hydrogen envelopes and a fast evolution,  while less massive He WDs evolve much more  slowly. As  shown by  Althaus  et al.   (2001a), this  behaviour solves the discrepancies between the spin-down ages of the millisecond pulsars B1855+09,  PSR J0034-0534 and  PSR J1012+5307 and  the cooling ages of their  He WD companions.  The evolution of  the Althaus et al. (2001a)   stellar   models   in   the   colour-colour   and   in   the colour-magnitude  diagrams  have been  analysed  by  Serenelli et  al. (2001) who find, on the  basis of detailed non-grey model atmospheres, that the  emergent spectrum  of low mass  He WDs becomes  bluer within time-scales of  astrophysical interest when  the effective temperature decreases  below  4000K.   Because   Serenelli  et  al.   (2001)  were interested in the late stages of He WD evolution, they did not attempt a detailed  modeling of the emergent  spectrum of these  stars at high effective temperature stages.  Interestingly, He WDs have also  been detected or inferred in open and globular clusters (Anderson 1997; Landsman et al. 1997; Edmonds et al. 1999).  More recently, Edmonds  et al.  (2001) have optically detected the  He  WD companion  to  a millisecond  pulsar  in  47 Tucanae.   In addition, Taylor et al. (2001)  have presented evidence for a sequence of  He WD  canditates  in the  globular  cluster NGC  6397.  A  proper interpretation  of the  observations of  He WDs  in  globular clusters requires evolutionary  calculations for HeWD progenitors with much lower metallicities than the solar one. In  this  regard,   the  present  work  is  designed   to  extend  the evolutionary calculations presented in Serenelli et al.  (2001) to the case  of low  metallicities.   In addition,  the present  calculations constitute  an improvement over  those presented  in Serenelli  et al. (2001).  Here  a more detailed treatment of  the microphysics entering the  WD  model  atmosphere  than   that  presented  in  that  work  is considered, thus enabling us to derive accurate colours and magnitudes for He  WDs at high effective  temperatures where the  effects of line broadening  opacities  are  not  negligible.   Finally,  a  much  more realistic treatment of mass loss phases than that attempted in Althaus et al. (2001a) and Serenelli et al. (2001) is considered in the present study. Details  about our atmosphere models, evolutionary code and mass  loss treatment are briefly  described in  Section 2.   Results are presented in Section  3 and  Section 4 is devoted to making some concluding remarks. ",
        "conclusions": "\\label{sec:conclusion}  In  this study  we have  explored the  evolution of  helium-core white dwarf (He  WD) stars with progenitors having  much lower metallicities than the  solar metallicity usually  assumed in the modeling  of these stars.    The  models   presented   here  are   appropriate  for   the interpretation of  recent and future  observations of low-mass  WDs in globular clusters.  Specifically, two low metallicity values have been considered: $Z$= 0.001  and $Z$= 0.0002.  In the  case of $Z$= 0.0002, we have followed the evolution of  He WD models with stellar masses of 0.199,  0.209, 0.219,  0.225,  0.243, 0.266,  0.280,  0.300 and  0.319 M$_\\odot$; and for the case  of $Z$= 0.001 we considered 0.172, 0.183, 0.197,  0.230, 0.244,  0.300,  0.336, 0.380,  0.390,  0.422 and  0.449 M$_\\odot$.  All of these models were evolved from the end of mass-loss phase down to very  advanced phases of evolution.  The binary nature of  our He  WD models has  been simulated by  abstracting mass from a  1 M$_\\odot$ model at  appropriate stages of  its evolution off the main  sequence.  Specifically, to obtain  physically sound initial low-mass He WD  models, mass transfer rates  were derived by  imposing that the stellar radius remains close to the radius of the Roche lobe.  A fully non-conservative approach was assumed and the loss of angular momentum caused by gravitational wave emission as well as magnetic stellar wind braking have been  taken into account.  The  evolution   of  our  He  WD   models  has  been   computed  in  a self-consistent  way with  the predictions  of  time-dependent element diffusion and nuclear burning. A non-gray treatment for the atmosphere that considered  the energy transfer  by radiation and  convection has been employed to derive the  outer boundary conditions of our evolving models. Models  atmosphere are  based on a  detailed treatment  of the microphysics entering the WD atmospheres, such as non-ideal effects in the equation of state (as  given by the occupation formalism of Hummer \\& Mihalas 1988) and the  inclusion of hydrogen line (from the Balmer, Lyman  and  Paschen  series)  and  pseudo-continuum  opacities.  Also, up-to-date  collision-induced absorption  data (Rohrmann  et  al. 2002) were  incorporated  in the  computations.  Such  a detailed description allowed  us to provide a grid  of accurate colour indices and magnitudes at both early and advanced evolutionary stages, strongly improving  previous efforts (Serenelli et al. 2001).  A  feature worthy  of comment  predicted  by our  calculations is  the existence  of thermonuclear  flash  episodes  for most  of  our He  WD sequences.  In part, this  is  a  result  of  including  in  our computations the  various chemical  diffusion processes.   We find that the lower the metallicity  $Z$, the larger the minumum stellar mass for the  occurrence of hydrogen  thermonuclear flashes induced  by the CNO cycle reactions.  Specifically, for  $Z$= 0.001 and $Z$= 0.0002 we find  a lower  mass  limit for  the  existence of  CNO  flashes of  $M \\approx$ 0.22  and 0.26 M$_\\odot$,  respectively (in the case  of $Z$= 0.02 such a limit becomes $\\approx$ 0.18 M$_\\odot$; see Althaus et al. 2001a).  In  addition, CNO  flashes become less  intense as  the metal content of the star is decreased and the mass range for the occurrence of such  instabilities depends strongly  on the inclusion  of chemical diffusion processes  in evolutionary calculations.  We  also find that the   existence  of   a   mass  threshold   for   the  occurrence   of diffusion-induced CNO flashes leads  to an age dichotomy (particularly for  $Z$=   0.001)  between  He   WD  models  with  and   without  CNO thermonuclear   flashes:  He   WDs   which  do   not  experience   CNO thermonuclear flashes  evolve very  slowly, so they  remain relatively bright even  at very large ages,  whilst those which  suffer from such flashes  are  characterized  by  a  fast cooling,  reaching  very  low effective temperature  stages within cooling  times less than  15 Gyr. Such an age  dichotomy is translated into distinctive  features in the isochrone plots which could  eventually be compared with observational expectations. It is worth mentioning that  during the short-lived, CNO diffusion-induced flashes, an appreciable amount of hydrogen is burnt, ultimately implying that the WD is left with a relatively thin hydrogen envelope which prevents stable hydrogen burning from being an important energy source at late cooling stages. As a result, the star has a much lower amount of available energy, which implies much shorter evolutionary time scales as compared with the situation when CNO flashes are absent.    Another finding of  this work is related to the fact  that some of our He  WD models  experience  several episodes  of thermal  instabilities related  to  unstable  hydrogen  burning via  the  proton-proton  (PP) nuclear  reactions.   These  PP  flashes,  which take  place  at  more advanced  stages of  evolution than  those  at which  the CNO  flashes occur, are  experienced by  all of our  sequences with $Z$=  0.0002.  In addition, our  0.172 and 0.183  M$_\\odot$ models with $Z$=  0.001 suffer from   PP flashes,  but at exceedingly  high ages.  We find that PP  thermal instabilities are triggered by  chemical diffusion that carries some  hydrogen downwards  deep enough for  the star  to ignite hydrogen there.  Except for the less massive models ($M \\lesssim$ 0.25 M$_\\odot$), PP flashes take place between 2 and 4 Gyr after the end of mass loss  episodes, and in general,  the more massive the  He WD, the earlier in the star life they occur.  The  evolution of  our He  WD  models has  also been  analysed in  the colour-magnitude  diagrams. We  find that  models which  have suffered from  CNO flashes  exhibit  a turn-off  in  most of  their colours  at M$_{\\rm V} \\approx$  16. This turn-off, which results  from the strong CIA opacity by molecular hydrogen at low temperatures, is reached well within  15  Gyr,  mostly by  He  WDs  with  $Z$= 0.001.  Finally,  the predictions of our models  for the colour-magnitude diagrams have been compared with  recent observational  data of He  WD candidates  in the globular clusters  NGC 6397  and 47 Tucanae  (Taylor et al.   2001 and Edmonds et al.  2001, respectively).  In this connection, we find that the  three  brightest HeWD  candidates in  NGC  6397  can indeed  be identified  as  HeWDs   characterized  by  stellar  mass  values  of 0.20-0.22~M$_\\odot$ (which is below  the mass value for the occurrence of CNO flashes) and ages ranging  from 0.5 to 1.5 Gyr. However, in the case of  the three  dimmest candidates, the agreement with observational data is not so evident as in the case of the brightest objects. Indeed, our  models appear to be  more massive than required by  observations. Specifically, our mass-loss  treatment gives  rise to  a minimum value of $\\sim$0.2~M$_\\odot$ an  HeWD may have  from progenitors with initially $Z$= 0.0002. However, it is worth noting that a $\\Delta(V-I)\\approx 0.1$ would allow a very good agreement between theoretical predictions and observations. In this sense, it is remarkable that the quoted uncertainty in the case of the HeWD candidate in 47 Tucanae is very close to this value (see Fig.10)\\footnote{It is worth mentioning that Townsley \\& Bildsten (2002) have recently suggested that the three dimmest HeWD candidates in NGC 6397 could be cataclismic variables involving a hot C-O WD and a low-mass main sequence companion.}. Finally the He WD in 47 Tucanae  is particularly  relevant because  the spin-down  age  of its millisecond pulsar  companion yields a WD  age estimate. Specifically, the pulsar age of 2 Gyr (see Edmonds et al. 2001) is in agreement with the prediction  of our $Z$= 0.001  He WD sequences.   In addition, the observational data  for the M$_{\\rm  V}$ and $U-V$ is  consistent with our 0.17~M$_\\odot$ sequence.  Complete tables containing the results of the present calculations are available  at  http://www.fcaglp.unlp.edu.ar/evolgroup/  or  upon request to the authors at their e-mail addresses.  \\begin{table*} \\centering \\begin{scriptsize} \\begin{minipage}{135mm} \\caption{Selected  stages for  0.199, 0.243  and 0.30~M$_\\odot$  He WD models for metallicity $Z$= 0.0002} \\begin{tabular}{@{}ccccccccccccc@{}} \\hline $M_*/{\\rm M_{\\odot}}$  & \\teff & $Log (g)$ &  $Age$ (Gyr) & U-B &B-V & V-R & V-K & R-I & J-H & H-K & BC & M$_V$ \\\\ \\hline 0.199 & 8920 & 4.3218 & 0.295 & -0.09 & 0.10 &  0.04 & 0.12 & 0.06 & 0.07 & -0.09 & -0.09 & 4.41 \\\\ '' & 10780 &  4.5281 & 0.494 & -0.24 & -0.05 & -0.03 & -0.20 & -0.02 &  0.02 & -0.10 & -0.40 & 4.42 \\\\ ''  & 14220 & 5.0085 & 0.683 & -0.52 & -0.13 & -0.07 &  -0.45 & -0.07 & -0.02 & -0.12 & -1.08 & 5.09 \\\\ '' & 16150 & 5.3500  & 0.791 & -0.64 & -0.15 & -0.08 & -0.54 & -0.09 & -0.04 & -0.13 & -1.41  & 5.72 \\\\ '' & 16880 & 5.5465 & 0.861 & -0.68 & -0.16 & -0.08 & -0.57 & -0.10 & -0.04 & -0.13 & -1.52 & 6.14 \\\\ '' & 17110 & 5.6901 & 0.924 & -0.69 & -0.15 & -0.08 & -0.58 & -0.10 & -0.04 & -0.13 & -1.56 & 6.47  \\\\'' & 16810 & 5.9008 & 1.054 & -0.68 & -0.14 & -0.08 & -0.56 & -0.10 & -0.04 & -0.13 & -1.51 & 7.03 \\\\ ''  & 16430 & 5.9817 & 1.124 & -0.66 & -0.13 & -0.08 & -0.55 & -0.09 & -0.04 & -0.13 & -1.45 & 7.27 \\\\ '' & 15960 & 6.0522 & 1.201 & -0.64 & -0.12 & -0.08 & -0.53 & -0.09 & -0.03 & -0.13 & -1.37 & 7.49 \\\\ '' & 15430  & 6.1141 & 1.283 & -0.61 & -0.11 & -0.07 & -0.50 & -0.08 & -0.03 & -0.12  & -1.28 & 7.71 \\\\ '' & 14870 & 6.1704 & 1.375  & -0.58 &  -0.09 & -0.07  & -0.47 & -0.08  & -0.02 &  -0.12 & -1.19 & 7.91 \\\\ '' & 14280 &  6.2219 & 1.478 & -0.54 & -0.08 & -0.07 & -0.45 & -0.07 & -0.02 & -0.12 &  -1.08 & 8.11 \\\\ '' & 13700 & 6.2699 & 1.594 & -0.51 & -0.06 & -0.06 &  -0.41 & -0.06 & -0.01 & -0.12 & -0.97 & 8.30 \\\\ '' & 13110 & 6.3151  & 1.725 & -0.47 & -0.04 & -0.06 & -0.38 & -0.06 & -0.01 & -0.12 & -0.86  & 8.49 \\\\ '' & 12530 & 6.3582 & 1.876 & -0.43 & -0.01 & -0.05 & -0.34  & -0.05 & 0.00 & -0.12 & -0.74 & 8.68 \\\\ '' & 11960 & 6.4003 & 2.058  & -0.39 & 0.02 & -0.04 & -0.29 & -0.04 & 0.01 & -0.11 & -0.63 & 8.88 \\\\ '' & 11430 & 6.4419 & 2.295 & -0.36 & 0.05 &  -0.03 & -0.24 & -0.02  & 0.02 & -0.11  & -0.53 & 9.07  \\\\ '' & 10930 & 6.4858 & 2.669 & -0.35 & 0.08 & -0.02 & -0.18 & -0.01 & 0.03 & -0.11 & -0.43 &  9.28 \\\\ '' & 10470 & 6.5314 & 3.356  & -0.35 & 0.11 & 0.00 &  -0.12 & 0.01  & 0.03  & -0.11 &  -0.34 & 9.49  \\\\ '' &  9990 & 6.5726 & 4.351 &  -0.37 & 0.15 & 0.03 & -0.04 & 0.04  & 0.05 & -0.10 & -0.26 &  9.72 \\\\ '' & 9510  & 6.6103 & 5.577  & -0.39 & 0.18  & 0.06 & 0.08 &  0.07 & 0.06  & -0.10 & -0.21  & 9.98 \\\\  '' & 9040 &  6.6448 & 6.985 &  -0.42 & 0.22 & 0.10  & 0.21 & 0.10  & 0.08 & -0.09  & -0.17 & 10.24 \\\\  '' & 8570 & 6.6772  & 8.658 & -0.45  & 0.26 & 0.13  & 0.34 & 0.13 & 0.10 &  -0.08 & -0.14 & 10.53 \\\\ '' & 8110  & 6.7069 & 10.622 & -0.47 & 0.29 & 0.17 & 0.49 & 0.17 & 0.13 & -0.07 & -0.13 & 10.82 \\\\'' &  7250 & 6.7575 & 15.738 & -0.48 & 0.36 & 0.23 & 0.81 & 0.23 & 0.18 &  -0.04 & -0.11 & 11.42 \\\\ \\\\ 0.243 & 15050 & 4.4210  & 0.131 & -0.59  & -0.18 & -0.08  & -0.51 & -0.09  & -0.03 & -0.12 & -1.21 & 3.29 \\\\ '' &  21020 & 5.0959 & 0.161 & -0.84 & -0.23 & -0.10 & -0.72 &  -0.13 & -0.07 & -0.14 & -2.10 & 4.42  \\\\ '' & 23930 & 5.4719 & 0.178 & -0.93 & -0.24 & -0.11 & -0.80 & -0.14 & -0.08 & -0.15 & -2.44 & 5.13 \\\\ '' & 25070  & 5.7041 & 0.190 & -0.95 & -0.24 & -0.12 & -0.82 & -0.15 & -0.08 & -0.15  & -2.56 & 5.63 \\\\ '' & 25360 & 5.8768 & 0.201  & -0.96 &  -0.24 & -0.12  & -0.83 & -0.15  & -0.09 &  -0.16 & -2.59 & 6.04 \\\\'' & 24650 & 6.1290 & 0.229 & -0.94 & -0.23 & -0.12 &  -0.81 & -0.14 & -0.08 & -0.15 & -2.51 & 6.72 \\\\ '' & 23920 & 6.2282  & 0.250 & -0.93 & -0.22 & -0.11 & -0.79 & -0.14 & -0.08 & -0.15 & -2.43  & 7.02 \\\\ '' & 23020 & 6.3141 & 0.277 & -0.91 & -0.21 & -0.11 & -0.77 & -0.14 & -0.07 & -0.15 & -2.33 & 7.30 \\\\ '' & 22000 & 6.3880 & 0.310 & -0.88 & -0.19 & -0.10 & -0.74 & -0.13 & -0.07 & -0.15 & -2.21 & 7.56  \\\\ '' & 20900 & 6.4526 & 0.350 & -0.85 & -0.18 & -0.10 & -0.71 & -0.12 & -0.06 & -0.14 & -2.08 & 7.81 \\\\ '' & 19780 & 6.5089 & 0.396 & -0.81 & -0.16 & -0.10 & -0.67 & -0.12 & -0.06 & -0.14 & -1.93 & 8.05 \\\\ ''  & 18650 & 6.5594 & 0.451 & -0.77 & -0.14 & -0.09 & -0.64 & -0.11 & -0.05 & -0.14 & -1.78 & 8.27 \\\\ '' & 17540 & 6.6058 & 0.517 & -0.72 & -0.12 & -0.09 & -0.59 & -0.10 & -0.04 & -0.13 & -1.61 & 8.49 \\\\ '' & 16460  & 6.6489 & 0.596 & -0.68 & -0.10 & -0.08 & -0.55 & -0.09 & -0.04 & -0.13  & -1.44 & 8.71 \\\\ '' & 15420 & 6.6900 & 0.695  & -0.63 &  -0.08 & -0.07  & -0.51 & -0.08  & -0.03 &  -0.13 & -1.27 & 8.92 \\\\ '' & 14450 &  6.7296 & 0.826 & -0.57 & -0.05 & -0.07 & -0.46 & -0.07 & -0.02 & -0.12 &  -1.10 & 9.13 \\\\ '' & 13540 & 6.7702 & 1.031 & -0.52 & -0.02 & -0.06 &  -0.41 & -0.06 & -0.01 & -0.12 & -0.93 & 9.34 \\\\ '' & 12700 & 6.8121 & 1.418 & -0.46 & 0.01 & -0.05 & -0.35 & -0.05 & 0.00 &  -0.12 & -0.77 & 9.57 \\\\ '' & 11870  & 6.8494 & 1.999 & -0.42 & 0.05 & -0.04 & -0.28 &  -0.03 & 0.01 & -0.11 & -0.61 & 9.79 \\\\ '' & 11060  & 6.8826 & 2.771 & -0.41  & 0.09 & -0.01 &  -0.20 & 0.00 & 0.03 & -0.11 & -0.45 & 10.02 \\\\  '' & 10290 & 6.9114 & 3.712 & -0.42 & 0.14 & 0.02 & -0.08 & 0.03 & 0.04 & -0.11 & -0.31 & 10.27 \\\\ '' & 9560 & 6.9384 & 4.897 & -0.44 & 0.18 &  0.07 & 0.09 & 0.07 & 0.06 & -0.10 & -0.23 & 10.58  \\\\'' & 8240 & 6.9865 & 8.308 &  -0.49 & 0.26 & 0.16  & 0.46 & 0.16  & 0.12 & -0.07  & -0.14 & 11.25 \\\\\\\\ 0.300  & 24990 & 6.9774 &  0.065 & -0.97 & -0.21  & -0.12 & -0.82 & -0.15 & -0.08 & -0.16 &  -2.53 & 8.57 \\\\ '' & 23390 & 6.9941 & 0.067 & -0.93 & -0.19 & -0.11 &  -0.78 & -0.14 & -0.07 & -0.15 & -2.36 & 8.73 \\\\ '' & 21800 & 7.0043  & 0.076 & -0.89 & -0.17 & -0.10 & -0.74 & -0.13 & -0.07 & -0.15 & -2.18  & 8.88 \\\\ '' & 20050 & 7.0178 & 0.105 & -0.83 & -0.15 & -0.10 & -0.68 & -0.12 & -0.06 & -0.14 & -1.96 & 9.06 \\\\ '' & 18730 & 7.0331 & 0.133 & -0.79 & -0.13 & -0.09 & -0.64 & -0.11 & -0.05 & -0.14 & -1.78 & 9.21  \\\\ '' & 17510 & 7.0495 & 0.162 & -0.74 & -0.10 & -0.09 & -0.60 & -0.10 & -0.04 & -0.13 & -1.60 & 9.36 \\\\ '' & 16160 & 7.0699 & 0.199 & -0.68 & -0.08 & -0.08 & -0.54 & -0.09 & -0.03 & -0.13 & -1.39 & 9.55 \\\\ ''  & 14930 & 7.0903 & 0.241 & -0.62 & -0.05 & -0.07 & -0.48 & -0.07 & -0.02 & -0.12 & -1.18 & 9.73 \\\\ '' & 13770 & 7.1113 & 0.293 & -0.55 & -0.01 & -0.06 & -0.42 & -0.06 & -0.01 & -0.12 & -0.97 & 9.93 \\\\ '' & 12870 & 7.1293 & 0.350 & -0.50 & 0.03 & -0.05 & -0.36 & -0.05 & 0.00 & -0.12 &  -0.80 & 10.10 \\\\ '' & 12050 & 7.1470 & 0.424 & -0.46 & 0.06 & -0.04 &  -0.29 & -0.03 & 0.01 & -0.12 & -0.64 & 10.27 \\\\ '' & 11110 & 7.1691 &  0.558 & -0.45 & 0.10 & -0.01 & -0.20 & 0.00 & 0.02 &  -0.11 & -0.45 & 10.48 \\\\ '' & 10270  & 7.1897 & 0.749 & -0.46 & 0.15 &  0.03 & -0.06 & 0.04 & 0.04 & -0.11  & -0.32 & 10.75 \\\\ '' & 9470 & 7.2094 & 1.024 & -0.48  & 0.19 & 0.08 & 0.12 & 0.08 & 0.07 & -0.09 & -0.23 & 11.06 \\\\ '' & 8870 & 7.2239 & 1.336 & -0.50 & 0.22 & 0.12 & 0.28  & 0.12 & 0.09  & -0.08 & -0.19 &  11.34 \\\\ A ''  & 8290 & 7.2368 & 1.766  & -0.51 & 0.26 & 0.16  & 0.46 & 0.16 &  0.12 & -0.07 & -0.15 & 11.63 \\\\ B '' & 7800 &  7.2634 & 2.776 & -0.51 & 0.30 & 0.19 & 0.61 &  0.20 & 0.15 & -0.06  & -0.13 & 11.94  \\\\ '' & 7150  & 7.2688 & 2.796 &  -0.48 & 0.37 & 0.24  & 0.85 & 0.25  & 0.19 & -0.03  & -0.12 & 12.32 \\\\  '' & 6570 & 7.2731  & 3.751 & -0.42  & 0.44 & 0.29  & 1.09 & 0.30 & 0.23  & -0.01 & -0.11 & 12.69  \\\\ '' & 6060 &  7.2852 & 4.394 & -0.33 & 0.53 & 0.35 & 1.33 & 0.35  & 0.27 & 0.01 & -0.12 & 13.08 \\\\'' & 5210  & 7.3257 & 5.748 &  -0.11 & 0.72 & 0.47 & 1.82 & 0.47 & 0.32 & 0.07 & -0.21 & 13.93 \\\\ '' & 4840 & 7.3486 & 6.969 &  0.00 & 0.82 &  0.53 & 2.07 & 0.53  & 0.35 & 0.09  & -0.31 & 14.40 \\\\ '' & 4470 & 7.3622 & 8.269 & 0.12 & 0.92 & 0.59 & 2.23 & 0.59 & 0.33 & 0.04 &  -0.42 & 14.89 \\\\ '' & 4110 & 7.3696  & 9.479 & 0.22 & 1.01 & 0.65 & 2.12 & 0.65 & 0.12  & -0.02 & -0.46 & 15.32 \\\\ '' & 3770 & 7.3743 & 10.729 & 0.29 & 1.07 & 0.69 & 1.67 & 0.68 & -0.14 & -0.14 & -0.41 & 15.65  \\\\ '' & 3460 & 7.3785  & 12.260 & 0.36 &  1.13 & 0.72 & 1.12 & 0.68  & -0.30 & -0.28 & -0.31  & 15.93 \\\\ '' &  3180 & 7.3835 & 15.016 & 0.44  & 1.19 & 0.72 & 0.59  & 0.60 & -0.37 &  -0.36 & -0.16 & 16.16 \\\\  '' & 2920 & 7.3858  & 16.606 & 0.51  & 1.24 & 0.69  & 0.05 & 0.43 & -0.38 & -0.47 & 0.01 & 16.37 \\\\ \\hline \\end{tabular} \\label{tab.z1}  {\\small Ages are counted from the end of mass transfer. Letters A and B denote the age  interval during which the 0.30  M$_\\odot$ model return several  times to  high  \\teff values  as  a result  from PP  hydrogen flashes.} \\end{minipage} \\end{scriptsize} \\end{table*}   \\begin{table*} \\centering \\begin{scriptsize} \\begin{minipage}{135mm} \\caption{Selected stages  for 0.172, 0.230, 0.336  and 0.449 M$_\\odot$ He WD models for metallicity $Z$= 0.001} \\begin{tabular}{@{}ccccccccccccc@{}} \\hline $M_*/{\\rm M_{\\odot}}$  & \\teff & $Log (g)$ &  $Age$ (Gyr) & U-B &B-V & V-R & V-K & R-I & J-H & H-K & BC & M$_V$ \\\\ \\hline  0.172 & 9390 & 4.6630 & 0.349 &  -0.13 & 0.07 & 0.02 & 0.02 & 0.03 & 0.05 & -0.09 & -0.16 & 5.27 \\\\ '' & 12420 & 5.2994 & 0.973 & -0.40 & -0.07 & -0.05 & -0.33 & -0.05 &  0.00 & -0.11 & -0.74 & 6.22 \\\\ '' & 12760 &  5.4971 & 1.173 &  -0.42 & -0.07 &  -0.05 & -0.35  & -0.05 & 0.00 & -0.11 & -0.81 & 6.67 \\\\ '' & 12700 & 5.6392 & 1.356 & -0.42 & -0.06 & -0.05 & -0.35 & -0.05 &  0.00 & -0.11 & -0.79 & 7.03 \\\\ '' & 12410 &  5.7509 & 1.544 &  -0.40 & -0.05 &  -0.05 & -0.33  & -0.05 & 0.00 & -0.11 & -0.73 & 7.35 \\\\ '' & 12010 & 5.8440 & 1.746 & -0.37 & -0.03 & -0.04 & -0.30 & -0.04 &  0.01 & -0.11 & -0.65 & 7.64 \\\\ '' & 11540 & 5.9253 & 1.971 & -0.33 & 0.00 & -0.04 & -0.26 & -0.03 & 0.01 & -0.11 & -0.56 & 7.93 \\\\ '' & 11040 & 5.9987 & 2.223 & -0.30 & 0.04 & -0.03 & -0.20 & -0.02 & 0.02  & -0.11 & -0.46 & 8.21 \\\\ '' & 10520 & 6.0664  & 2.511 &  -0.29 & 0.07  & -0.01 & -0.14  & 0.00 &  0.03 & -0.10 & -0.36 & 8.48 \\\\ '' & 10000 & 6.1303 & 2.849 & -0.30 & 0.12 & 0.02 &  -0.06 & 0.03 & 0.04  & -0.10 & -0.26  & 8.77 \\\\ ''  & 9500 & 6.1927 & 3.279 & -0.33 & 0.16 &  0.05 & 0.06 & 0.06 & 0.06 & -0.10 & -0.20 & 9.08 \\\\  '' & 9050 & 6.2590 & 3.974 & -0.36  & 0.19 & 0.09 & 0.18 & 0.09  & 0.07 & -0.09 & -0.16  & 9.42 \\\\ '' &  8660 & 6.3336 & 5.723 & -0.39 &  0.22 & 0.12 & 0.30 & 0.12 & 0.09  & -0.08 & -0.14 & 9.78 \\\\ ''  & 8230 & 6.3987 & 8.461  & -0.42 & 0.26 &  0.15 & 0.44 & 0.15 & 0.12 & -0.07 & -0.12 & 10.14 \\\\'' & 7330  & 6.5018 & 15.655 & -0.45  & 0.34 & 0.23 &  0.77 & 0.23 & 0.17 & -0.04 & -0.10 & 10.88 \\\\  '' & 6880 & 6.5439 & 20.326 & -0.42 & 0.40  & 0.26 & 0.95 &  0.27 & 0.20 &  -0.02 & -0.09 &  11.26 \\\\ \\\\ 0.230 &  30010 & 6.4188 &  0.203 & -1.06 &  -0.27 & -0.13  & -0.94 & -0.17 & -0.10 & -0.17 & -2.98 &  7.12 \\\\ '' & 28220 & 6.4678 & 0.203 & -1.02 &  -0.25 & -0.13 & -0.90  & -0.16 & -0.10 & -0.17  & -2.84 & 7.37 \\\\ '' & 26430 & 6.5134 &  0.203 & -0.99 & -0.23 & -0.12 & -0.86 & -0.15  & -0.09 &  -0.16 & -2.68  & 7.61 \\\\ ''  & 25000 &  6.5480 & 0.203 &  -0.96 & -0.22  & -0.12 &  -0.82 & -0.15  & -0.08 &  -0.15 & -2.54 & 7.80 \\\\ '' & 23350 &  6.5868 & 0.203 & -0.92 & -0.20 & -0.11 & -0.78  & -0.14  & -0.07 &  -0.15 & -2.36  & 8.02  \\\\ '' &  22030 & 6.6168 &  0.204 & -0.89 &  -0.19 & -0.10 &  -0.74 & -0.13  & -0.07 & -0.15 & -2.21 & 8.19 \\\\ '' &  20710 & 6.6461 & 0.205 & -0.85 & -0.17 & -0.10 & -0.70 & -0.12 & -0.06 & -0.14 & -2.05 & 8.37 \\\\ '' & 19200 & 6.6727 & 0.214  & -0.79 & -0.15 & -0.09 & -0.65  & -0.11 & -0.05 & -0.14 & -1.85 & 8.56 \\\\ '' &  17860 & 6.6782 & 0.251 & -0.74 & -0.13 & -0.09 & -0.61 & -0.10 & -0.04 & -0.13 & -1.66 & 8.70 \\\\ '' & 16620 & 6.6907 & 0.299  & -0.69 & -0.10 & -0.08 & -0.56  & -0.09 & -0.04 & -0.13 & -1.47 & 8.85 \\\\ '' &  15380 & 6.7116 & 0.345 & -0.62 & -0.08 & -0.07 & -0.50 & -0.08 & -0.03 & -0.13 & -1.26 & 9.04 \\\\ '' & 14310 & 6.7331 & 0.390  & -0.56 & -0.05 & -0.07 & -0.45  & -0.07 & -0.02 & -0.12 & -1.07 & 9.21 \\\\ '' &  13320 & 6.7549 & 0.442 & -0.50 & -0.02 & -0.06 & -0.39 & -0.06 & -0.01 & -0.12 & -0.89 & 9.40 \\\\ '' & 12400 & 6.7768  & 0.501 & -0.44  & 0.02 & -0.05  & -0.33 & -0.04  & 0.01 & -0.12 & -0.71 & 9.59 \\\\ '' & 11530 & 6.7990 & 0.570 & -0.40 & 0.06 & -0.03 & -0.25 & -0.02 & 0.02 &  -0.11 & -0.55 & 9.79 \\\\ '' & 10730 & 6.8210 & 0.648 & -0.40 & 0.11 & 0.00 & -0.15 & 0.01 & 0.03 & -0.11 & -0.38 & 9.99 \\\\  '' & 9910 & 6.8455 & 0.751 & -0.41  & 0.15 & 0.04 & 0.00 & 0.05 &  0.05 & -0.10 & -0.26 & 10.28 \\\\ ''  & 9220 & 6.8699 & 0.868 & -0.44 &  0.20 & 0.09 & 0.17 & 0.09 & 0.07  & -0.09 & -0.20 & 10.59 \\\\ '' &  8600 & 6.8939 & 1.015 & -0.47 & 0.24  & 0.13 & 0.35 & 0.14 & 0.10 & -0.08 & -0.16 &  10.91 \\\\ '' & 8070 & 6.9151 & 1.193 & -0.48 & 0.28 & 0.17 & 0.51 &  0.17 & 0.13 & -0.06 & -0.13 & 11.21 \\\\'' & 7000 & 6.9588 & 1.824 & -0.46 & 0.38 &  0.25 & 0.90 & 0.26  & 0.20 & -0.03  & -0.10 & 11.91  \\\\ '' & 6460 & 6.9796 &  2.341 & -0.40 & 0.45 & 0.30 & 1.13  & 0.30 & 0.24 & -0.01 & -0.10 & 12.31 \\\\ '' & 5970 & 6.9984 & 2.930 & -0.31 & 0.54 & 0.36 &  1.37 & 0.36  & 0.27 & 0.02  & -0.12 &  12.71 \\\\'' & 5150 & 7.0477 & 4.075 & -0.09 & 0.74 & 0.47 & 1.86 & 0.47  & 0.32 & 0.08 & -0.23  & 13.59 \\\\ '' &  4810 & 7.0820 & 4.780 &  0.02 & 0.83 & 0.54  & 2.11 & 0.53  & 0.35 & 0.10  & -0.32 & 14.07 \\\\ ''  & 4460 & 7.1098 & 5.792  & 0.14 & 0.93 &  0.60 & 2.29 & 0.60 & 0.35 &  0.06 & -0.43 & 14.58 \\\\ '' & 4110  & 7.1265 & 6.876 & 0.24 & 1.02 & 0.66 & 2.23 & 0.66 & 0.18 & 0.00 & -0.49 & 15.04 \\\\ '' & 3780 & 7.1369 & 7.948 & 0.32 & 1.09 & 0.70 & 1.85 & 0.69 & -0.08 & -0.11 & -0.46 & 15.39 \\\\ '' &  3450 & 7.1443 & 9.157 & 0.39 & 1.15 & 0.73 & 1.25  & 0.69 & -0.27 & -0.26  & -0.35 & 15.69 \\\\  '' & 3210 & 7.1486 & 10.232 & 0.46 & 1.20 & 0.74 & 0.74 & 0.64 & -0.36 & -0.35 & -0.22 & 15.89 \\\\ '' & 2980 &  7.1522 & 11.469 & 0.52 & 1.25 & 0.72 & 0.27 & 0.51 & -0.38 & -0.43 &  -0.07 & 16.07 \\\\ '' & 2760 & 7.1552 & 12.897 & 0.59 & 1.29 & 0.69 &  -0.16 & 0.33 & -0.36 & -0.54 & 0.08 & 16.26 \\\\ '' & 2530 & 7.1593 &  15.744 & 0.68 & 1.35 & 0.62 & -0.70 & 0.04 & -0.29 & -0.69 & 0.26 & 16.46 \\\\ '' & 2310 & 7.1624 & 18.216 & 0.78 & 1.40 & 0.55 & -1.27 & -0.34 & -0.20 & -0.87 & 0.45 & 16.69 \\\\ \\\\ 0.336 & 21420 & 7.1810 & 0.045  & -0.88 & -0.16 & -0.10 & -0.73 & -0.12 & -0.06 & -0.14 & -2.13 &  9.22 \\\\ '' & 19620 & 7.1907 & 0.049 & -0.82 &  -0.13 & -0.10 & -0.67  & -0.11 & -0.05 & -0.14  & -1.90 & 9.40 \\\\ '' & 18260 & 7.1978 &  0.065 & -0.78 & -0.11 & -0.09 & -0.62 & -0.10  & -0.05 &  -0.14 & -1.71  & 9.54 \\\\ ''  & 17010 &  7.2109 & 0.090 &  -0.72 & -0.09  & -0.08 &  -0.58 & -0.09  & -0.04 &  -0.13 & -1.52 & 9.69 \\\\ '' & 15840 &  7.2270 & 0.121 & -0.67 & -0.06 & -0.08 & -0.53  & -0.08  & -0.03 &  -0.13 & -1.33  & 9.85  \\\\ '' &  14640 & 7.2464 &  0.161 & -0.61 &  -0.03 & -0.07 &  -0.47 & -0.07  & -0.02 & -0.12 & -1.12 & 10.03 \\\\ '' &  13540 & 7.2658 & 0.210 & -0.55 & 0.01 & -0.06  & -0.41 &  -0.06 & -0.01  & -0.12 & -0.92  & 10.22 \\\\  '' & 12620 & 7.2830 & 0.268 & -0.50 & 0.05 & -0.05 & -0.34 & -0.04 & 0.01 & -0.12  & -0.75 & 10.40  \\\\ '' & 11810  & 7.2986 & 0.337  & -0.48 & 0.08 & -0.03 & -0.27 & -0.02 &  0.01 & -0.12 & -0.59 & 10.56 \\\\ '' & 10880 & 7.3163 & 0.445 & -0.48 & 0.12 & 0.00 & -0.16 & 0.01 & 0.03 & -0.11 & -0.41 & 10.78 \\\\ '' &  10020 & 7.3328 & 0.582 & -0.48 & 0.16 & 0.05 & 0.00 &  0.06 & 0.05 & -0.10 & -0.29 & 11.06  \\\\ '' & 9370 & 7.3454 & 0.719 & -0.50 & 0.20 &  0.09 & 0.16 & 0.09 & 0.07 & -0.09 & -0.23 & 11.33 \\\\ '' & 8720 &  7.3573 & 0.889 & -0.52 & 0.23 & 0.13 & 0.33 & 0.13 &  0.10 & -0.08 & -0.18 & 11.62 \\\\ ''  & 8120 & 7.3681 & 1.092 & -0.52 &  0.27 & 0.17 & 0.51 & 0.17 & 0.13  & -0.06 & -0.15 & 11.93 \\\\ '' &  7460 & 7.3795 & 1.370 & -0.51 & 0.33  & 0.22 & 0.73 & 0.22 & 0.17 & -0.04 & -0.12 &  12.29 \\\\ '' & 6950 & 7.3883 & 1.640 & -0.47 & 0.39 & 0.26 & 0.93 &  0.26 & 0.20 & -0.03 & -0.11 & 12.61 \\\\ '' &  6390 & 7.3987 & 2.023  & -0.40 & 0.47  & 0.31 & 1.17  & 0.31 & 0.24 & 0.00 & -0.11 & 13.00 \\\\  '' & 5870 & 7.4099 & 2.492 & -0.29 & 0.56 & 0.37 & 1.43 & 0.37 & 0.28 & 0.03 & -0.13 & 13.41 \\\\'' & 5000 & 7.4416 & 3.926 & -0.05 & 0.77 & 0.50 & 1.96 & 0.50  & 0.34 & 0.08 & -0.26  & 14.32 \\\\ '' &  4610 & 7.4569 & 5.361 &  0.07 & 0.88 & 0.57  & 2.18 & 0.57  & 0.35 & 0.06  & -0.38 & 14.83 \\\\ ''  & 4240 & 7.4669 & 6.985  & 0.18 & 0.97 &  0.63 & 2.16 & 0.63 & 0.20 & -0.01 & -0.44 &  15.29 \\\\ '' & 3890 & 7.4727 & 8.597 & 0.26 & 1.05 & 0.68 & 1.80 &  0.67 & -0.08 & -0.10 & -0.42 & 15.65 \\\\ '' &  3610 & 7.4760  & 9.997 & 0.32  & 1.10 &  0.70 & 1.28 &  0.68 & -0.26 & -0.25 & -0.34 & 15.90 \\\\  '' & 3300 & 7.4790 & 11.832 & 0.40 & 1.16 & 0.72 & 0.76 & 0.64 &  -0.36 & -0.34 & -0.21 & 16.17 \\\\ '' & 3020 & 7.4813 & 14.033 & 0.47 &  1.21 & 0.70 & 0.22 & 0.51 & -0.39 & -0.43 & -0.05 & 16.39 \\\\ \\hline \\end{tabular} \\label{tab.z2}  {\\small Ages are counted from the end of mass transfer.} \\end{minipage} \\end{scriptsize} \\end{table*}  \\begin{table*} \\centering \\begin{scriptsize} \\begin{minipage}{135mm} \\setcounter{table}{2} \\caption{Continued.} \\begin{tabular}{@{}ccccccccccccc@{}} \\hline $M_*/{\\rm M_{\\odot}}$  & \\teff & $Log (g)$ &  $Age$ (Gyr) & U-B &B-V & V-R & V-K & R-I & J-H & H-K & BC & M$_V$ \\\\ \\hline   \\\\ 0.449 & 36370 & 7.2604 & 0.002 & -1.16 & -0.28 & -0.14 & -1.00 & -0.18 & -0.12 & -0.17 &  -3.45 & 8.13 \\\\ '' & 34060 & 7.2999 & 0.004 &  -1.13 & -0.27 & -0.14  & -0.99 & -0.18 & -0.11  & -0.17 & -3.27 & 8.34 \\\\ '' & 32270 &  7.3577 & 0.011 & -1.11 & -0.26 & -0.14 & -0.97  & -0.17  & -0.11 &  -0.17 & -3.13  & 8.58  \\\\ '' &  30310 & 7.4037 &  0.023 & -1.09 &  -0.25 & -0.13 &  -0.94 & -0.17  & -0.10 & -0.17 & -2.98 & 8.82 \\\\ '' &  28220 & 7.4342 & 0.035 & -1.05 & -0.23 & -0.13 & -0.90 & -0.16 & -0.10 & -0.17 & -2.82 & 9.04 \\\\ '' & 26220 & 7.4605 & 0.049  & -1.01 & -0.21 & -0.12 & -0.86  & -0.15 & -0.09 & -0.16 & -2.64 & 9.24 \\\\ '' &  24310 & 7.4844 & 0.068 & -0.97 & -0.19 & -0.12 & -0.81 & -0.14 & -0.08 & -0.15 & -2.45 & 9.44 \\\\ '' & 22530 & 7.5065 & 0.093  & -0.93 & -0.16 & -0.11 & -0.76  & -0.13 & -0.07 & -0.15 & -2.25 & 9.63 \\\\ '' &  20860 & 7.5270 & 0.128 & -0.88 & -0.14 & -0.10 & -0.71 & -0.12 & -0.06 & -0.14 & -2.05 & 9.81 \\\\ '' & 19290 & 7.5462 & 0.174  & -0.83 & -0.11 & -0.10 & -0.66  & -0.11 & -0.05 & -0.14 & -1.84 & 9.99 \\\\ '' &  17830 & 7.5638 & 0.237 & -0.78 & -0.09 & -0.09  & -0.61 &  -0.10 & -0.04  & -0.14 & -1.63  & 10.17 \\\\  '' & 16470 &  7.5796 & 0.320 &  -0.72 & -0.06 &  -0.08 & -0.56  & -0.09 & -0.03 & -0.13 & -1.42 & 10.34 \\\\ '' & 15200 & 7.5938 & 0.429 & -0.66 & -0.02 & -0.07  & -0.50 & -0.07 & -0.02 & -0.13  & -1.21 & 10.51 \\\\ '' & 14030 & 7.6063 & 0.570 & -0.60 & 0.01 & -0.07 & -0.44 & -0.06 & -0.01 & -0.12 & -1.00 & 10.69 \\\\ '' & 12920 & 7.6173 & 0.749 & -0.55 & 0.06 & -0.05 & -0.36 & -0.04  & 0.00 & -0.12 & -0.80 & 10.87 \\\\ '' & 11890 &  7.6273 & 0.982 & -0.54  & 0.09 & -0.03 & -0.28  & -0.02 & 0.01 & -0.12 & -0.59 & 11.04 \\\\  '' & 10910 & 7.6365 & 1.289 & -0.53 & 0.13 & 0.01 & -0.15 & 0.02 &  0.03 & -0.11 & -0.41 & 11.26 \\\\ '' & 10030 & 7.6452 & 1.672 & -0.53 &  0.17 & 0.05 & 0.02 & 0.06 & 0.05 & -0.10 & -0.30 & 11.54 \\\\ '' & 9220 & 7.6532 & 2.148 & -0.54 & 0.21 & 0.10 &  0.21 & 0.11 & 0.08  & -0.09 & -0.23  & 11.85 \\\\ ''  & 8460 & 7.6606 & 2.728 & -0.55 & 0.25 &  0.15 & 0.41 & 0.15 & 0.11 & -0.07 & -0.18 & 12.19 \\\\ '' & 7770 &  7.6674 & 3.349 & -0.54 & 0.30 & 0.20 & 0.63 & 0.20 &  0.15 & -0.05 & -0.14 & 12.54 \\\\ ''  & 7130 & 7.6741 & 3.969 & -0.49 &  0.36 & 0.24 & 0.86 & 0.25 & 0.19  & -0.03 & -0.12 & 12.91 \\\\ '' &  6540 & 7.6806 & 4.611 & -0.42 & 0.44  & 0.30 & 1.10 & 0.30 & 0.23 & -0.01 & -0.10 &  13.28 \\\\ '' & 6020 & 7.6874 & 5.330 & -0.32 & 0.54 &  0.35 & 1.34 & 0.35 & 0.26 & 0.02  & -0.12 & 13.67 \\\\ '' &  5540 & 7.6959 & 6.189  & -0.21 & 0.63  & 0.41 & 1.59  & 0.41 & 0.29 & 0.05 & -0.14 & 14.07 \\\\  '' & 5110 & 7.7067 & 7.472 & -0.09 & 0.74 & 0.48 & 1.88 & 0.48 & 0.33 & 0.07 & -0.24 & 14.55 \\\\ '' & 4710 & 7.7160 & 9.455 & 0.04 & 0.85 &  0.55 & 2.10 & 0.55 & 0.34 & 0.05 & -0.34 & 15.04 \\\\ '' & 4330 &  7.7221 & 11.582 & 0.14 & 0.94 & 0.61 & 2.10 & 0.61 &  0.20 & -0.01 & -0.41 & 15.49 \\\\ ''  & 3970 & 7.7258 & 13.651 & 0.22 & 1.02 & 0.66 &  1.79 & 0.65 & -0.06 & -0.09 & -0.40 & 15.85 \\\\ '' &  3640 & 7.7282 & 15.838 & 0.30 & 1.08  & 0.69 & 1.19 & 0.66 & -0.28 & -0.26 & -0.31 & 16.16 \\\\ \\hline \\end{tabular} \\label{tab.z2}  {\\small Ages are counted from the end of mass transfer.}  \\end{minipage} \\end{scriptsize} \\end{table*}"
    },
    "0208/physics0208090_arXiv.txt": {
        "abstract": " ",
        "introduction": "Charge-coupled devices (CCDs) have emerged as the preferred detectors on all new X-ray astronomy mission in recent years. This is because they possess a high spatial resolution as well as a moderate energy resolution, simultaneously~\\cite{tanaka}. The dead layer above CCD must be thin enough to attain a high quantum efficiency at soft X-ray regions. As the result, devices cannot be protected against the high energy particles in space in the incident direction of X-rays.  Soon after the launch of the X-ray astronomy satellite, {\\it Chandra}, all of the front-illuminated CCD of the X-ray CCD camera (ACIS) have suffered some damage caused by the charge transfer inefficiency (CTI)~\\cite{acis_damage}.  The CTI is defined as an average fraction of charge packet lost at each transfer. Similar type of devices to ACIS CCDs have been tested by the high energy protons (40\\,MeV and 10\\,MeV) but not by the low energy protons before launch. The low energy protons having energy of $\\sim\\,$150\\,keV release major part of its energy at the transfer channel of the ACIS CCDs, which is located roughly 1\\,$\\mu$m below the electrodes. They cause the displacement damages in Si, leading to the formation of trapping sites for the charge packet. Since the flux of low energy protons at the orbit of {\\it Chandra} is much higher than that at the low earth orbit such as ASCA~\\cite{acis_proton} and low energy protons reflecting through the X-ray mirror assembly (HRMA) can reach the focal plane~\\cite{acis_reflect}, a significant degradation of the CTI has occurred.  The Monitor of All-sky X-ray Image (MAXI) has been selected as an early payload of the JEM (Japanese Experimental Module; {\\it KIBO}) Exposed Facility on the International Space Station (ISS)~\\cite{maxi}. MAXI has slit scanning cameras which consist of two kinds of X-ray detectors; one-dimensional position sensitive proportional counters with total area of $\\sim$\\,5000\\,cm$^2$ named GSC and the X-ray CCD camera with total area of $\\sim$\\,200\\,cm$^2$ named SSC.  SSC carries 32 CCDs which are three-side buttable with full-frame transfer and have 1024$\\times$1024 pixels of $24\\,\\mu {\\rm m}\\times24\\,\\mu {\\rm m}$ size with two phase gate structures. The CCD chips are fabricated by the Hamamatsu Photonics K. K. (HPK). In order to perform useful X-ray spectroscopy over the whole device, the CTI must be less than roughly $2\\times 10^{-5}$\\, per transfer where the shift of the peak energy is similar to that of the Fano-limited noise of 120\\,eV at 5.9\\,keV.  Previous studies of the radiation hardness for HPK CCDs were also focused on high energy protons above 1\\,MeV~\\cite{tomida} and no data are available for low energy protons. We thus performed the irradiation test focusing on the low energy protons. In this paper, we describe the device architecture, irradiation experiment, and the measurement of the CTI at $-100\\,^\\circ$C. ",
        "conclusions": "\\subsection{Proton Bragg curve}  We found that protons having energies of 292 and 391\\,keV seriously damaged HPK CCDs on the CTI performance. The degradation of the CTI obtained with protons having lower and higher energies is much less than those with 292 and 391\\,keV protons. This strongly suggests that the low radiation-tolerant region inside the HPK CCD is located in relatively a narrow region.  We calculated the Bragg curves of protons in Si. We employed {\\sl Geant4} with the {\\sl G4EMLOW0.3} data and considered the energy straggling due to the Al degrader of 5\\,$\\mu$m in thickness. Figure~\\ref{fig:bragg_curve} ({\\it upper}) shows the energy loss of protons as a function of depth of Si. The dotted line represents the minimum energy to displace Si atoms ($\\simeq 6$ eV $\\rm \\AA^{-1}$)~\\cite{sze}. The energy deposition due to 292 and 391\\,keV protons are concentrated at the depth of $2-4\\,\\mu$m inside Si. In this depth, the energy deposition of protons with other energies is less than those of 292 and 391\\,keV. Therefore, the radiation tolerance at depth of $2-4\\,\\mu$m is much lower than those in other region inside the HPK CCD.  Figure~\\ref{fig:bragg_curve} ({\\it lower}) shows the schematic view of the cross section of the HPK CCD employed. Since the HPK CCD is a buried-channel type, the charge packet is transferred in a narrow region along the depth of the CCD. This transfer channel well coincides with the Bragg peak region. We thus conclude that the transfer channel of the CCD possesses the lowest radiation tolerance for protons. This result is consistent with the ACIS result but the serious proton energy is slightly different from our value. Prigozhin {\\it et al.}~\\cite{acis_damage} estimated the minimum proton energy to reach the buried channel to be somewhat higher than 50$-$70\\,keV in order to penetrate the optical blocking filter, covering layer, and electrodes. Therefore, the thickness of the covering material is much thinner than our case, resulting that lower energy protons seriously affected the ACIS CCDs.  As described in section~\\ref{sec:wafer_result}, there is no difference in CTI values between CCDs fabricating from high resistivity wafer and those from low resistivity wafer. The acceptor doping concentration of our device is only an order of $10^{13}-10^{14}$\\,cm$^{-3}$ and the difference between epi-2 wafer and epi-3 wafer is roughly an order of magnitude~\\cite{ssc_em}.  Therefore, the probability that protons encounter Si atom is essentially the same between these devices.  Since the thickness of $n$-type layer is the same between them, their difference is the thickness of a depletion layer. It means that the location of the transfer channel is at the same depth between them. Our results are, therefore, expected if the radiation tolerance depends not on the depletion depth but on the transfer channel. This is consistent with the previous work~\\cite{holland}.  We are now developing CCDs from newly-obtained epitaxial wafer having much higher resistivity than that of epi-3. Since, however, the location of the transfer channel of new CCDs is the same depth as current devices, we are convinced that we can apply these results to new CCDs.   \\subsection{Modeling the CTI degradation}  As shown in Fig~\\ref{fig:delta_cti}, the degradations of $\\Delta$CTI are expressed as a linear function of the proton fluence. Since $\\Delta$CTI is expressed as a linear function of the electron trap density~\\cite{physics}, the formation of electron traps proportionally corresponds to proton fluence.  Values of $\\Delta$CTI are fitted to a linear function of proton fluence. The best fit parameters, a slope and an intercept ($\\Delta$CTI$_0$), are shown in Table~\\ref{table:cti}. Figure~\\ref{fig:slope} shows the slope obtained as a function of proton energy. Since the obtained values of slope correspond to an efficiency to create the electron trap, Fig~\\ref{fig:slope} shows that 292\\,keV protons most seriously affect the CTI degradation.  As shown in Fig~\\ref{fig:bragg_curve}, low energy protons deposit major part of their energy within a confined depth. The peak of the Bragg curve corresponds to the depth of 2.3\\,$\\mu$m in Si in the case of 292\\,keV protons. We thus assume there is a thin radiation-sensitive area within the CCD at depth of 2.3\\,$\\mu$m with thickness of 0.05\\,$\\mu$m. We should note that 0.05\\,$\\mu$m is the shortest unit we can simulate. Ignoring the $\\Delta$CTI degradation from other depths, we can calculate the energy deposition by protons that affect the CTI. Results are plotted in Fig~\\ref{fig:slope} with filled circles normalized by value at 292\\,keV.  For all proton energies, calculated values are much larger than those obtained.  As shown in Fig~\\ref{fig:bragg_curve}, if the thickness of the radiation-sensitive region increases, the energy deposit of 391 or 522 keV protons becomes relatively larger than that of 292 keV protons. It drove the calculated values for 391 and 522 keV to be increased much more than current values, resulting the deviation from data to be more significant.  In this calculation, we assumed that the probability to create an electron trap is linearly proportional to the proton energy loss. This assumption leads to a large discrepancies between the data and the calculations.  Therefore, there may be some nonlinear effects in their probabilities.  There are two types of process for proton energy loss: an ionization energy loss (IEL) and a nonionization energy loss (NIEL).  These two different forms of energy dissipation are translated into two major damage mechanisms for CCDs: an ionization damage and a bulk damage. The ionization damage leads to a flat-band shift which causes the operating voltage to be shifted. This damage is caused by all types of charged particles. On the other hand, energetic charged particles undergo Rutherford-scattering-type Coulombic interactions with the Si lattice structure. The energy deposited by the interacting ion is enough to pull a silicon atom out of its lattice position, forming an interstitial Si atom and a vacancy. The displaced atom, called the primary knock-on atom (PKA), may have sufficient energy to undergo collisions with lattice, producing more vacancies. NIEL is responsible for a part of the energy producing the initial vacancy-interstitial pairs and phonons.  Ziegler {\\it et al.}~\\cite{iel} and Burke~\\cite{niel} calculated the IEL and the NIEL, respectively. Based on their calculations, more than 98\\,\\% of incident proton energies ($E_p$ [keV]) release as the IEL for $E_p \\ge 100$\\,keV.  For a proton of relativistic energies, the NIEL is almost constant whereas with lower energies the NIEL has a 1/$E_p$ dependence.  This suggests that the probability to create displacements is not linearly proportional to the total energy loss but is proportional to $E_p^{-\\gamma}$.  We then fitted the function $E_p^{-\\gamma}$ to the results of NIEL calculated by Burke.  We found that $\\gamma$ can be approximated to be $\\simeq 0.76$ at the energy range of $100 \\,{\\rm keV} \\le E_p \\le 4 $\\,MeV.  In order to take into account the nonlinear effect in creating traps due to the NIEL, we need to employ not the incident proton energy but the energy at the depth of 2.3\\,$\\mu$m. We calculated the energy reduction of $E_p$ during the passage of 2.3\\,$\\mu$m in Si with {\\sl Geant4}.  We then calculated the fraction of the NIEL among the total energy loss with taking into account of the energy dependence of the NIEL for each reduced $E_p$.  We normalized the fraction of the NIEL for each proton energy by that of 292 keV and took them into account for the previous calculations. Results are shown by filled squares in Fig~\\ref{fig:slope}. Our calculations considering the NIEL represent the data obtained. However, values of slope measured suddenly decreases as $E_p$ increases whereas they cannot be reproduced by our calculations. In our model, we only consider the NIEL which represent the energy deposition as the initial vacancy-interstitial pairs and phonons. If the energy of PKA is large enough to undergo collisions with Si atoms, the number of vacancies increase. Therefore, to take into account the spectrum of PKA and collisions between PKA and Si atoms is important for future modeling.  Empirical relations between the slope of the $\\Delta$CTI versus the proton energy are described as:  \\begin{eqnarray}\\label{eqn:cti} {\\rm slope}(E_p\\,{\\rm [keV]}) & = & 1.2\\times 10^{-10}\\times E_p - 2.0 \\times 10^{-11} \\ \\ \\ {\\rm for} \\ E_p \\le 292\\, {\\rm keV} \\\\ {\\rm slope}(E_p\\,{\\rm [keV]}) & = & 1.2\\times 10^{-9}\\times\\exp(-E_p/6.6\\times 10^{-2}) + 3.0 \\times 10^{-13} \\ \\ \\ {\\rm for} \\ E_p \\ge 292\\, {\\rm keV} \\end{eqnarray}  \\noindent Solid lines in Fig~\\ref{fig:slope} represent above empirical relations. For a given proton spectrum in orbit, we can calculate the $\\Delta$CTI value by summing contributions from all proton energies.  \\subsection{Estimate the CTI for the MAXI mission}  We found that low energy protons with energies of $290-400$\\,keV seriously damaged the spectroscopic performance of the MAXI CCDs. The degradation of CTI as a function of mission life for the MAXI based on our experiments has been estimated.  There is a slit at the top of the SSC camera with a size of 5$\\times$0.3\\,mm$^2$ and the slat collimators just above the CCDs~\\cite{sakano}. The thickness of the slat collimator is $\\sim 100\\,\\mu$m, which is aligned by $\\sim 3\\,$mm pitch, resulting the field of view of each CCD to be $\\sim 1.5\\,^\\circ$ square. Within the field of view, no shield protects devices whereas the column density at other directions on the camera is $\\sim 2.5$\\,g cm$^{-2}$, suggesting the proton component passing through the camera to be negligibly small.  We thus calculate the proton flux coming through $1.5\\,^\\circ \\times 1.5\\,^\\circ$ area.  We employed the proton flux described in the literature~\\cite{ssp}, in which the attitude of the ISS is 500\\,km and solar activity is the maximum.  The proton flux at 500\\,km is the largest among attitudes expected for the ISS~\\cite{ssp} and we therefore use it for the worst case analysis.  The number of proton at the solar minimum is factor of $\\sim 2$ larger than that at the solar maximum.  We thus increase the proton flux with a factor of 1.5 as the average value. Figure~\\ref{fig:estimated_cti} shows the CTI estimated for the MAXI as a function of its mission life. The dotted line shows the acceptable limit for the MAXI mission. Since the mission life of the MAXI is two years, the degradation of the CTI is well below the acceptable limit even for the worst case analysis. We therefore confirm the high radiation torelance of MAXI CCDs.  \\acknowledgement  This work is partly supported by the Grant-in-Aid for Scientific Research by the Ministry of Education, Culture, Sports, Science and Technology of Japan (13874032, 13440062)."
    },
    "0208/astro-ph0208314_arXiv.txt": {
        "abstract": "Models of the solar transition region made from lines other than those of helium cannot account for the strength of the helium lines. However, the collisional excitation rates of the helium resonance lines are unusually sensitive to the energy of the exciting electrons. Non-thermal motions in the transition region could drive slowly-ionizing helium ions rapidly through the steep temperature gradient, exposing them to excitation by electrons characteristic of higher temperatures than those describing their ionization state. We present the results of calculations which use a more physical representation of the lifetimes of the ground states of He~{\\sc i} and He~{\\sc ii} than was adopted in earlier work on this process. New emission measure distributions are used to calculate the temperature variation with height. The results show that non-thermal motions can lead to enhancements of the He~{\\sc i} and He~{\\sc ii} resonance line intensities by factors that are comparable with those required. Excitation by non-Maxwellian electron distributions would \\emph{reduce} the effects of non-thermal transport. The effects of non-thermal motions are more consistent with the observed spatial distribution of helium emission than are those of excitation by non-Maxwellian electron distributions alone. In particular, they account better for the observed line intensity ratio $I$(537.0~\\AA)/$I$(584.3~\\AA), and its variation with location. ",
        "introduction": "\\footnotetext{E-mail: g.smith2@physics.ox.ac.uk} The resonance lines of He~{\\sc i} (at 584.3 \\AA) and He~{\\sc ii} (at 303.8 \\AA) show unusual behaviour when compared with other strong emission lines in solar {\\sc euv} spectra. Reviews of observations of the helium resonance lines and attempts to model their formation can be found in Hammer \\shortcite{rh97} and Macpherson \\& Jordan \\shortcite{mj99} (hereafter MJ99), but the details relevant to the present work are summarized here.  Early observations (e.g.\\ Tousey 1967) showed that the helium resonance lines are reduced significantly in intensity in coronal holes compared with the average quiet Sun, while other lines formed at similar temperatures show only very small reductions in intensity. In the quiet Sun, the helium resonance line intensities are too large to be reproduced by emission measure distributions that account for other transition region (TR) lines, a problem identified by Jordan (1975), who found disagreements by factors of 15 for He~{\\sc i} and 6 for He~{\\sc ii}. Corresponding factors of at least 10 and 13, respectively, were found by MJ99. Radiative transfer calculations have also been unable to reproduce the observed resonance line intensities without invoking either a `plateau' in the model atmosphere (see e.g. Vernazza, Avrett \\& Loeser 1981; Andretta \\& Jones 1997), which leads to other lines formed in the lower TR being over-estimated \\cite{grs00}, or some departure from equilibrium (e.g.\\ Fontenla, Avrett \\& Loeser 1993; Avrett 1999). These results have been taken to imply that some process preferentially enhances the helium resonance line intensities with respect to other lines formed in the TR in the quiet Sun, and that the enhancement is reduced in coronal holes. Recent results from the Solar and Heliospheric Observatory ({\\em SOHO}) show that the helium resonance line intensities are reduced by factors of 1.5--2.0 (Peter 1999; Jordan, Macpherson \\& Smith 2001) in coronal holes.  Many possible enhancement processes have been investigated, but a completely convincing explanation has not yet been found. The effects of photoionization by coronal radiation have been suggested as a natural explanation of the coronal hole/quiet Sun contrast \\cite{hz75}, and photoionization-recombination (PR) appears to be important in the formation of some lines in the helium spectrum (He~{\\sc ii} 1640.4 \\AA\\ -- Wahlstr\\o m \\& Carlsson 1994, He~{\\sc i} 10830 \\AA\\ -- Andretta \\& Jones 1997). Although PR probably contributes to the formation of the He~{\\sc i} 584.3-\\AA\\ line, evidence \\cite{rm75,aj,and99} suggests that the He~{\\sc ii} 303.8-\\AA\\ line and, to a lesser extent, the 584.3-\\AA\\ line, are formed principally by collisional excitation. If this is the case, then because the helium resonance lines have unusually large values of $W/kT_{\\textrm{e}}$, where $W$ is the excitation energy, their collisional contribution functions are sensitive to excitation by suprathermal electrons. Any process exposing helium ions to larger populations of suprathermal electrons than in equilibrium will tend to increase the collisional excitation rates of the helium lines, while lines with smaller $W/kT_{\\textrm{e}}$ will be relatively unaffected.  Two processes by which this might occur were suggested by Jordan (1975,1980): the transport of high energy electrons from the upper TR or corona to the lower TR, or the transport of He atoms and ions by turbulent motions to regions of higher electron temperature. Both processes would be expected to depend on the magnitude of the temperature gradient, which could explain the coronal hole/quiet Sun contrast, since Munro \\& Withbroe \\shortcite{mw72} found d$T/$d$h$ to be an order of magnitude smaller inside coronal holes.  Shoub \\shortcite{es83} investigated the former process in his study of the shape of the electron velocity distribution function (EVDF) in the transition region. His calculations suggested that EVDFs in the lower TR should have more heavily populated suprathermal tails than the local Maxwellian distribution. He found that this could lead to enhanced collisional excitation and ionization in helium, but did not calculate intensities to compare with observations. Anderson, Raymond \\& Ballegooijen (1996) did calculate intensities for the He~{\\sc ii} resonance line under excitation by several different non-Maxwellian EVDFs, and found enhancements over Maxwellian collisional excitation, but their calculations could not reproduce the observed intensities of all the lines they studied simultaneously.  Radiative transfer calculations of the helium resonance line intensities using EVDFs approximating those derived by Shoub (1982,1983) have been performed, and are reported in a companion paper (Smith, in preparation -- hereafter paper II). The calculations suggest that the He~{\\sc i} and He~{\\sc ii} resonance line intensities could be increased by non-Maxwellian collisional excitation, but that this process would produce signatures in the line ratios that contradict observations.  The second process is the subject of this paper. The temperature gradients in the solar TR are such that velocities of order 10 km s$^{-1}$ could carry material into regions of significantly higher electron temperature. Such mixing could be caused by non-thermal motions in the TR that can be inferred from observations of the excess widths of optically thin lines (e.g.\\ Berger, Bruner \\& Stevens 1970; Boland et al.\\ 1973; Doschek et al.\\ 1976; Chae, Sch\\\"uhle \\& Lemaire 1998. See also Section \\ref{sec2.3}). Helium ions would reach statistical equilibrium at the new local temperature much more slowly than the bulk of the material owing to their long excitation and ionization times. The helium resonance lines would be excited collisionally at higher temperatures and hence with greater rates than would be the case in statistical equilibrium at the temperatures determining the ground state populations, while other transition region lines are relatively unaffected. Jordan \\shortcite{cj80} investigated the effects of the process on the He~{\\sc ii} 303.8-\\AA\\ line, and a more extensive study has been made recently by Andretta et al.\\ \\shortcite{aea00}, who termed the process `velocity redistribution.'  This work represents the application of similar methods to more specific examples of temperature gradients, which are no longer assumed to be constant over the path of the moving clump of plasma. A new treatment is used for the mean lifetime of the helium ground state, and the analysis is extended to the cases of the He~{\\sc i} 584.3-\\AA\\ and 537.0-\\AA\\ lines in a first approximation.  In Section \\ref{sec2} earlier work on non-thermal transport of helium is reviewed, and the methods and atmospheric parameters used here to calculate the enhancements of the helium line intensities are introduced. In Section \\ref{sec3} the calculations of the velocity redistribution enhancement factors for the \\hbox{He\\,{\\sc ii}} 303.8-\\AA\\ line and the \\hbox{He\\,{\\sc i}} 584.3-\\AA\\ and 537.0-\\AA\\ lines are described in detail. The possible effects of non-Maxwellian electron distributions in the transition region on the calculations are assessed. In Section \\ref{sec4} the results of the calculations are discussed. The derived enhancement factors are compared with the results of MJ99, to determine whether the process can explain quantitatively the anomalously large helium line intensities. More qualitative comparisons are made with observations of the spatial variations of the helium line intensities in the quiet Sun with respect to each other and to other transition region lines. Comparisons are also made with the results of a separate investigation into the possibility that the helium resonance line intensities are enhanced by excitation by non-local suprathermal electrons (Paper II). ",
        "conclusions": "\\label{sec4} The intensity enhancement factors calculated for the helium resonance lines in the presence of turbulent transport are given in Table \\ref{tab5}. As discussed above, these should be regarded as upper limits. In the case of He~{\\sc ii} the over-estimates should be relatively small, and an intensity enhancement of approximately a factor of 5 is plausible, given the parameters assumed to describe the solar atmosphere. This is similar to the factors predicted by Jordan (1980) and Andretta et al.\\ (2000) using a less satisfactory expression for the excitation time. An enhancement factor of 5 cannot account entirely for the enhancement of at least a factor of 13 apparently required in the network according to the analysis of MJ99. However, using the current second order correction of $sim25$ for the {\\sc nis}~2 waveband of {\\sc cds} instead of the factor of 55 derived by Landi et al.\\ \\shortcite{lan97}, the required enhancement is only a factor of 6 (as found by Jordan 1975), which is within the range calculated here. Radiative transfer calculations (Smith 2000) using the VAL \\cite{val} models of the quiet solar atmosphere under-produce the 303.8-\\AA\\ line intensity by factors of 3--4, which the present work shows could be explained entirely by the enhancement mechanism suggested here.  The enhancement factors derived for the \\hbox{He\\,{\\sc i}} 584.3-\\AA\\ line are also upper limits, but show that enhancement factors of order 10 are plausible. Thus non-thermal transport of He~{\\sc i} could produce the enhancement of the resonance line intensity required to account for the results of MJ99.  \\begin{table} \\caption{Upper limits on intensity enhancement factors for the helium lines calculated for EMDs S and X. An assumed pressure of $P_{\\textrm{e}} = 2 \\times 10^{14}$ cm$^{-3}$ K is denoted by (a), $P_{\\textrm{e}} = 6 \\times 10^{14}$ cm$^{-3}$ K is denoted by (b).\\label{tab5}} \\begin{center} \\begin{tabular}{lccccc} \\hline Ion & Wavelength & \\multicolumn{4}{c}{Enhancement factor}\\\\ & (\\AA) & S(a) & X(a) & S(b) & X(b) \\\\\\hline He {\\sc ii} & 303.8 & 6.9 & 7.0 & 5.3 & 5.5 \\\\ He {\\sc i}  & 584.3 & 15.3 & 17.6 & 11.2 & 18.7 \\\\ He {\\sc i}  & 537.0 & 22.4 & 26.0 & 14.1 & 26.5 \\\\\\hline \\end{tabular} \\end{center} \\end{table}  Calculations of the 584.3-\\AA\\ line intensity using the VAL C (average quiet Sun) and VAL D (average network) models produce fairly good matches to observations \\cite{aj,grs00}, but produce too much emission in other low transition region lines, owing to the presence of the temperature plateau at $T_{\\textrm{e}} \\simeq 2.5 \\times 10^{4}$ K. However, models of the atmosphere in which the plateau is removed do not produce high enough intensities in the \\hbox{He\\,{\\sc i}} line to match the observations, but by smaller factors than required by MJ99 \\cite{aj,grs00}. These discrepancies could also be explained by the factors found here.  Similar arguments hold for the \\hbox{He\\,{\\sc i}} 537.0-\\AA\\ line, for which an enhancement of a factor of up to about 25 is predicted by the calculations presented here. It is interesting that greater enhancement factors are predicted for the 537.0-\\AA\\ line than for the 584.3-\\AA\\ line, as radiative transfer calculations using the VAL models \\cite{aj,grs00} show the predicted ratio $I$(537.0~\\AA)/$I$(584.3~\\AA) to be smaller than observed. MJ99 found a ratio of $0.116 \\pm 0.015$ in the network (using the Landi et al.\\ 1997 calibration of {\\sc cds}; $0.105 \\pm 0.014$ using the Brekke et al.\\ 2000 calibration), but calculations using the VAL models give values of about 0.08. Enhancements of the two lines in the ratio derived here would help resolve this disagreement. Radiative transfer calculations using a model atmosphere based on EMD S produce a value of 0.117 (Paper II); the enhancements predicted here would therefore worsen agreement with observations. Nevertheless, the enhancement factors calculated here give an intensity ratio much closer to that observed than do radiative transfer calculations of the effects of non-local suprathermal electrons which produce enhancement factors of the same order (Paper II).  In the formulation used here, the efficiency of the non-thermal transport process increases with decreasing pressure. For example, the predicted 303.8-\\AA\\ and 584.3-\\AA\\ line intensity enhancements are increased by about 30 per cent when the pressure is reduced by a factor of 3. This would appear to conflict with observations of coronal holes, where electron pressures are lower than in the quiet Sun by factors of 2 -- 3 \\cite{mw72,jea01}, but where helium line intensities (and required enhancements) are smaller. However, the energy balance adopted here in the upper TR (equation (\\ref{eq5})) takes no account of the fast solar wind and the term in $A(r)$ does not allow for super-radial expansion. For non-thermal transport calculations appropriate to coronal holes, energy balance calculations specifically for coronal holes are required; these could be combined with observed EMDs in a similar manner to that described here. Further observations of the helium lines in coronal holes on the disk would be useful (to avoid the problems of limb darkening/brightening seen in polar coronal holes). It would also be interesting to investigate non-thermal transport in active regions (for which Andretta et al.\\ 2000 predict smaller enhancements), where electron pressures are higher and turbulent velocities lower than in the quiet Sun.  Turbulent transport could help explain the spatial variations of the intensities of the helium lines with respect to each other and to other TR lines in the quiet Sun. The increases in the heights of peak line formation predicted here are not large enough to be resolved directly at the limb, but the effects of the increased \\emph{temperatures} of emission may perhaps be seen in observations of the network, the appearance of which changes between the low and high TR. The turbulent motions of the largely ionized gas would be influenced by the direction of the network field, so that material moving to greater heights would trace the expansion of the network (see e.g.\\ Gabriel 1976). The width and the contrast of the network observed in the \\hbox{He\\,{\\sc i}} and \\hbox{He\\,{\\sc ii}} lines seem to resemble that seen in high TR lines like those of Ne~{\\sc vi} (log~$T_{\\textrm{e}} = 5.6$) and Ne~{\\sc vii} (log~$T_{\\textrm{e}} = 5.7$), rather than that seen in lower TR lines of C~{\\sc ii} (log~$T_{\\textrm{e}} = 4.4$), C~{\\sc iii} (4.9), or even O~{\\sc iii} (5.05), O~{\\sc iv} (5.25) or O~{\\sc v} (5.4) \\cite{bb74,glw76,gea98,pea99}. Patsourakos et al.\\ (1999) found that Gabriel's (1976) model predicts a network width which increases with temperature, reaching the width observed in helium at log~$T_{\\textrm{e}} \\simeq 5.75$. The present calculations predict He~{\\sc i} emission to peak at log~$T_{\\textrm{e}} \\simeq 5.2$, and He~{\\sc ii} emission to peak at log~$T_{\\textrm{e}} \\simeq 5.3$, but with significant contributions in each case from temperatures up to log~$T_{\\textrm{e}} \\simeq 5.6$. There is little contribution to intensity above log~$T_{\\textrm{e}} \\simeq 5.7$, but the calculations assume different pressures and temperature gradients (and overall geometry) to those in the Gabriel (1976) model, making quantitative comparisons difficult. Qualitatively, however, turbulent transport in an expanding network could explain why the pattern of helium emission resembles that seen in lines formed in equilibrium at higher temperatures and heights. In a similar geometry, suprathermal electrons streaming down the field lines from the upper TR would be expected to be funnelled into the network in the low TR, which would concentrate any enhanced helium emission in the centre of the network, which would be less consistent with the observed width of the network.  Turbulent transport of helium in an expanding network is also qualitatively consistent with observed variations in the ratio of the intensities of the \\hbox{He\\,{\\sc i}} lines, $I$(537.0~\\AA)/$I$(584.3~\\AA). The excitation rate ratio for the 537.0-\\AA\\ line peaks at a slightly higher temperature and falls off more slowly at higher temperatures than for the 584.3-\\AA\\ line. This suggests that the network could appear wider in the 537.0-\\AA\\ line than in the 584.3-\\AA\\ line, which is broadly consistent with rastered images in the two lines (MJ99), although further observations of the two lines would be useful. The ratio $I$(537.0~\\AA)/$I$(584.3~\\AA) would be expected to increase towards the edges of the network, but the variation would be relatively small, since the ratio of the enhancement factors of the two lines changes by only 25 per cent between log~$T_{\\textrm{e}}$ = 4.8 and log~$T_{\\textrm{e}}$ = 5.7. This factor is consistent with values of the ratio observed by MJ99, who found only small variations of the ratio, but a consistent increase in the ratio in regions of smaller absolute intensity. Enhanced excitation by non-local suprathermal electrons in the low TR would tend to produce a larger variation in the opposite sense with absolute intensities (Paper II).  In a more refined approach to turbulent transport calculations, improvements could be made to the simple geometry adopted here. Flux tubes with various inclinations could be modelled, allowing a more quantitative examination of how the spatial variation of helium emission may be affected. MJ99 found steeper temperature gradients in cell interiors than in the network, but these gradients may be \\emph{across} magnetic structures. In the expanding network picture, the temperature gradient \\emph{along the field} is smaller, which would reduce the enhancement factor, as would the higher pressures found by MJ99 in cell interiors. The greater helium line enhancements required by MJ99's observations in cell interiors could possibly be caused by photon scattering from the boundaries; radiative transfer using two-component model atmospheres is needed to test this point.  As described in Section \\ref{sec3.3}, even significantly non-Maxwellian EVDFs are expected to have relatively little effect on turbulent transport of helium in the quiet Sun. As enhancements of the helium line intensities by non-thermal motions occur at temperatures much higher than the normal temperature of peak line formation, and enhancement by non-local suprathermal electrons occurs largely at lower temperatures (Paper II), the two processes could be operating simultaneously. If non-thermal motions do produce the enhancement factors calculated here, these would dominate over the effects of non-local electrons, provided departures from Maxwellian EVDFs are realistic (Paper II).  The present calculations do not include radiative transfer, which could have significant effects on \\hbox{He\\,{\\sc i}}, but which would be less important for \\hbox{He\\,{\\sc ii}}. Radiative transfer calculations including non-thermal motions (and possibly non-Maxwellian EVDFs as well) are needed to place our conclusions on a firmer footing."
    },
    "0208/astro-ph0208158_arXiv.txt": {
        "abstract": "Based on observations of the Seyfert nucleus in NGC~1068 with ASCA, RXTE and BeppoSAX, we report the discovery of a flare (increase in flux by a factor of $\\sim$ 1.6) in the 6.7~keV Fe~K line component between observations obtained four months apart, with no significant change in the other (6.21, 6.4, and 6.97~keV) Fe~K$\\alpha$ line components. During this time, the continuum flux {\\it decreased} by $\\sim$20\\%. The RXTE spectrum requires an Fe~K absorption edge near 8.6 keV (Fe XXIII $-$ XXV). The spectral data indicate that the 2$-$10 keV continuum emission is dominated ($\\sim$ 2/3 of the luminosity) by reflection from a previously unidentified region of warm, ionized gas located $\\lapprox$0.2 pc from the AGN. The remaining $\\sim$ 1/3 of the observed X-ray emission is reflected from optically thick, neutral gas. The coronal gas in the inner Narrow-Line Region (NLR) and/or the cold gas at the inner surface of the obscuring ``torus'' are possible cold reflectors. The inferred properties of the warm reflector are: size (diameter) $\\lapprox$0.2 pc, gas density $n \\gapprox$ 10$^{5.5}$ cm$^{-3}$, ionization parameter $\\xi \\approx$ 10$^{3.5}$ erg~cm~s$^{-1}$, and covering fraction 0.003 (L$_0$/10$^{43.5}$ erg~s$^{-1}$)$^{-1} < (\\Omega/4\\pi) <$ 0.024 (L$_0$/10$^{43.5}$ erg~s$^{-1}$)$^{-1},$ where L$_0$ is the intrinsic 2$-$10 keV X-ray luminosity of the AGN. We suggest that the warm reflector gas is the source of the (variable) 6.7 keV Fe line emission, and the 6.97 keV Fe line emission. The 6.7 keV line flare is assumed to be due to an increase in the emissivity of the warm reflector gas from a {\\it decrease} (by 20$-$30\\%) in L$_0$. The properties of the warm reflector are most consistent with an intrinsically X-ray weak AGN with L$_0 \\approx$ 10$^{43.0}$ erg~s$^{-1}$. The optical and UV emission that scatters from the warm reflector into our line of sight is required to suffer strong extinction, which can be reconciled if the line-of-sight skims the outer surface of the torus. Thermal bremsstrahlung radio emission from the warm reflector may be detectable in VLBA radio maps of the NGC~1068 nucleus. ",
        "introduction": "NGC~1068 is one of the original ``spiral nebulae'' that were noted by Seyfert (1943) to have strong, high-excitation emission lines in their nuclear optical spectra.  Its lack of ``broad'' (FWHM $\\gapprox$ 1000 km~s$^{-1}$) emission lines from a ``broad-line region'' (BLR), argued to be present in type~1 Seyfert nuclei, identify it as a type~2 Seyfert galaxy. However, when broad Balmer emission lines were discovered in its {\\it polarized} optical nuclear spectrum by Antonucci \\& Miller (1985), it became clear that this galaxy contained a ``hidden'' broad-line region, seen only in light reflected around an intervening obscuring medium. This result, combined with results for other Seyfert galaxies, gave rise to the so-called unified model of active galactic nuclei (AGNs), which asserts that the difference between type~1 and type~2 AGNs is due to the relative orientation of an optically thick torus which surrounds the central engine and BLR (cf. Antonucci 1993 and references therein).  X-ray observations of AGNs have, for the most part, supported this unified model.  For example, observations of X-ray bright type~2 Seyfert galaxies with the HEAO-1 satellite (Mushotzky 1982) showed that their X-ray spectra were well represented by a power-law with photon index $\\Gamma \\approx$ 1.8 (characteristic of type~1 Seyferts), with additional absorption with column densities N$_H \\gapprox$ 10$^{22}$ atoms cm$^{-2}$.  The absorption column in type~2 Seyfert galaxies, typically $\\sim$10$^{23}$$-$10$^{24}$ cm$^{-2}$, is assumed to be due to the blocking torus (e.g., Mulchaey, Mushotzky \\& Weaver 1992).  Higher quality data now show that Seyfert X-ray spectra are more complex than a simple absorbed power-law.  In many type~2 objects, a `soft excess' (soft X-ray emission in excess of that expected from a simple absorbed power-law) is common below $\\sim$2 keV (e.g.  Turner \\& Pounds 1989). Above $\\sim$10 keV, the power-law slope generally flattens (e.g., Nandra \\& Pounds 1994) due to Compton scattering of nuclear light reflected from the accretion disk (ergo, the ``reflection component''). Soft X-ray absorption lines (e.g., due to O, Ne, Mg, Si, Fe) from an ionized medium are also present (cf. Reynolds \\& Fabian 1995).  And finally, between 6 and 7 keV, strong Fe~K$\\alpha$ fluorescence lines are present, and are thought to originate from the accretion disk, or, in some cases, from the inner edge of the torus.   The central region of NGC~1068 contains a starburst (e.g., Wilson et al. 1992), which produces a large fraction of the X-ray emission below $\\sim$4 keV (e.g., Levenson et al. 2001).  At higher energies, the AGN dominates the spectrum. The X-ray spectrum of the AGN is particularly interesting, in that it has an Fe K$\\alpha$ line with a very large equivalent-width (EW), first discovered with GINGA by Koyama et al. (1989), who modelled the spectrum with a relatively flat ($\\Gamma \\sim$ 1.5) power-law with little or no absorption, plus a single Gaussian line centered at 6.55~keV with EW of 1.3~keV. Compare this with typical EWs for type~1 Seyferts, which are $\\lapprox$500 eV (e.g., Nandra et al. 1997).  Such a large EW was predicted by the X-ray scattering model of Krolik \\& Kallman (1987), although the measured EW in NGC~1068 is larger than predicted. It was suggested that the observed (reflected) X-ray continuum flux is much smaller than that illuminating the fluorescing material, resulting in a larger EW than for Seyfert~1 galaxies. Higher-resolution X-ray spectral data obtained with BBXRT showed three separate narrow Fe~K$\\alpha$ components with a net EW of 2.7 keV (Marshall et al. 1993). Marshall et al. suggested that the multiple Fe~K features arose from two components of the nuclear gas, a hot and a warm component. A revised model of the X-ray continuum for NGC~1068 was later put forward by Smith, Done \\& Pounds (1993), who found that although a reflection component was not required by the GINGA data, if it were included, it would change the inferred intrinsic power-law slope to $\\Gamma \\sim$ 1.9, more typical of Seyfert~1 X-ray spectra. More recent ASCA data have confirmed the three separate line components, this time with a total EW of 3.2 keV (Ueno et al. 1994). Using the same ASCA data, Iwasawa et al. (1997) proposed the existence of a fourth Fe~K component $-$ a weak red wing of the 6.4 keV fluorescence line, possibly due to Compton scattering in optically thick, cold matter (Matt et al. 1996).  Hard ($\\gapprox$10 keV) X-ray emission detected by BeppoSAX (Matt et a. 1997) is consistent with reflection from a mixture of neutral and ionized gas, with an intrinsic power-law slope of $\\Gamma \\sim$ 1.7.  In this paper, we further investigate the nature of the X-ray emission due to the AGN by analyzing the broad-band (E $\\approx$ 4$-$200 keV) X-ray spectrum using all of the available data from the ASCA, RXTE and BeppoSAX missions.  We focus here only on the X-ray spectrum above 4~keV to avoid the nuclear starburst. The hard ($\\gapprox$ 10 keV) X-ray spectral coverage provided by the RXTE and BeppoSAX Phoswich Detector System (PDS) data is vital for placing constraints on continuum models. After the continuum is well modelled with the best data available, we can then accurately model the Fe~K$\\alpha$ complex.  Such is the goal of the current paper. In section 2, we describe the data used for our analyses and describe details of the data reduction.  Section 3 contains the primary observational results from our analyses and section 4 contains a discussion of the implied physical constraints on the geometry of the nucleus of NGC~1068 and the origin of the X-ray emission. Throughout this paper, we assume a distance of 14.67~Mpc (H$_0 =$ 75 km~s$^{-1}$~Mpc$^{-1}$) to NGC~1068.  \\begin{figure}[hbn] \\epsscale{0.8} \\plotone{f1.eps} \\caption{ Timeline of relevant X-ray spectral observations of NGC~1068. ASCA observations are shown as filled triangles, RXTE observations are shown as filled rectangles, and BeppoSAX observations are shown as open rectangles.  Note that the RXTE observation is simultaneous with the second ASCA observation (ASCA2). \\label{figallobs}} \\end{figure} ",
        "conclusions": "}  We propose that the 6.7/6.97 keV line emitting region and the previously unidentified warm, ionized reflector in NGC~1068 are indeed the same region. Approximately 2/3 of the total observed 2$-$10 keV X-ray emission comes from the warm reflector, which is located $\\lapprox$0.2 pc from the AGN. The geometrical and physical properties of the warm reflector (Table~\\ref{tabwarmrefl}) are quite different from those of the optical/UV reflector (e.g. Miller et al. 1991).  The warm X-ray reflector should reflect optical and UV light; however, no such emission is seen in HST images at the location of the proposed nucleus (e.g., Capetti et al. 1995).  This suggests that the dust optical depth toward the nucleus is high enough to block the UV and optical light, but the equivalent X-ray column is not large enough to block the harder X-ray emission.  Equivalent columns up to $\\sim$10$^{22}$ cm$^{-2}$ are allowed when fitting the X-ray models to the ASCA/RXTE/BeppoSAX data, so such a scenario is feasible from the X-ray point of view.  We can achieve this scenario if the torus is viewed edge-on so that the line-of-sight toward the nucleus skims the edge of the torus (see Figure~\\ref{warmrefltorus}).    We estimate that $\\sim$ 1/3 of the total observed 2$-$10 keV X-ray emission is reflected by optically thick, neutral reflectors. See Figure~\\ref{neutralreflectors} for a cartoon illustrating the possible reflection regions. \\begin{figure}[hbn] \\epsscale{1.0} \\plotone{f10.eps} \\caption{ Cartoon of the two optically-thick `cold' reflection regions: the inner surface of the dense, obscuring torus, and the optical reflection region (the inner NLR), located $\\sim$30 pc from the AGN (Capetti et al. 1995). The shaded region of the torus illustrates that only a fraction ($\\lapprox$ 0.5) of the inner surface of the torus reflects X-rays toward the observer.  The coronal gas in the inner NLR has a moderately low ionization parameter ($\\xi \\sim$ 50) and may be optically thick to electron scattering, and therefore could also be a ``cold'' X-ray reflector. \\label{neutralreflectors} } \\end{figure} If the coronal gas in the NLR is Compton thick, it is possible that all of the cold reflection component is from the NLR. We estimate a covering factor of $\\approx$0.15 for the inner NLR, and $\\sim$0.1 for the inner surface of the torus.  Thus, either of these two regions could contribute significantly to the observed cold reflection component. Likewise, the 6.4 keV Fe emission line could originate from either of these regions.   VLBA observations of NGC~1068 (Gallimore et al. 1997) show that the size, radius and density of the radio-emitting clumps at the inner edge of the torus ($R \\approx$ 0.6~pc, height $h \\lapprox$ 0.1 pc) are very similar to those of our proposed warm reflector. If the warm reflector does indeed have a density $\\gapprox$10$^6$~cm$^{-3}$ and a spherical volume of diameter $\\sim$0.1~pc, and is at temperature of $\\sim$0.1-1 keV, the emission measure will be large enough to see thermal bremsstrahlung radio emission (cf. Gallimore et al. 1997). However, since the emission $\\propto$~n$^2$, values of $n$ below a few $\\times$ 10$^5$ (e.g., if D(NGC~1068) $<$ 14.67 Mpc) would force the bremsstrahlung luminosity below Gallimore et al.'s detection limit. It is worth noting that since the morphology of the radio clumps of S1 are not co-linear (as one would expect from an unwarped, edge-on disk/torus), it is quite possible that one of the clumps is in the Gallimore et al. radio image is the primary warm X-ray reflector.      How does our proposed warm reflector region fit in with the AGN paradigm? Krolik \\& Kriss (1995) suggest that the ``warm absorbers'' seen in type~1 AGN may be associated with the warm reflection region.  Typical ionization parameters associated with the warm absorbers are $\\xi \\sim$ 10$-$100 erg~cm~s$^{-1}$ (e.g., Reynolds 1997, Kaspi et al. 2001), which is much lower than the values we find ($\\xi \\sim$ 10$^3$). In principle, there could be clouds with different densities and mean ionization parameters in each of these regions, which could be responsible for the warm absorbers see in AGN X-ray spectra. One possibility is that the more highly ionized, inner regions of the warm reflector clouds reflect X-rays toward our line of sight, while the more weakly ionized, outer regions of the clouds cause the warm, ionized edges.  We suggest the existence of ionized, warm reflector gas clouds distributed within a radius of $\\sim$0.2 pc. The highly ionized gas in this region produces an Fe~K edge near 8.6 keV in the reflected nuclear spectrum, as well as both the observed 6.7~keV and 6.97~keV emission lines. Cold reflection, required to model the observed Compton reflection hump, originates from optically thick ``cold'' gas in the inner NLR and/or from the inner surface of the torus. Our data and proposed models are most consistent with an intrinsically X-ray weak AGN with L$_0 \\approx$ 10$^{43.0}$ erg~s$^{-1}$."
    },
    "0208/astro-ph0208191_arXiv.txt": {
        "abstract": "In order to study systematically the soft X-ray emission of Active Galactic Nuclei (AGNs) at medium to high redshifts, we have analyzed {\\it ROSAT} PSPC and HRI data of QSOs at $0.26\\le z \\le3.43$ selected from the second deepest {\\it ROSAT} PSPC survey carried out in 1991-1993 by McHardy \\etal (1998). Our sample of 22 type~1 QSOs is nearly complete above a flux limit of $1.4\\times10^{-14}{\\rm~erg~cm^{-2}~s^{-1}}$ in the $0.1-2{\\rm~keV}$ band. Of these, nine QSOs show long term ($\\sim 2{\\rm~yr}$) X-ray variability by a factor of $1.5-3.5$. Significant excess absorption above the Galactic column is seen in three QSOs. The soft X-ray photon index of the QSOs ranges from $1.4$ to $3.7$. Three QSOs have steep soft X-ray spectra ($\\Gamma_{X} > 3.0$), one of which is a narrow-line QSO -- a high luminosity version of narrow-line Seyfert~1 galaxies. The average photon index ($<\\Gamma_{X}>$) is $2.40\\pm0.09 $ (with a dispersion of $0.57$) in the $0.1-2{\\rm~keV}$ band. The average QSO spectra in four redshift bins flatten from an average photon index of $<\\Gamma_{X}> \\sim 2.53$ at $0.25 \\le z \\le 1$ to $<\\Gamma_{X}> \\sim 2$ at $2 \\le z \\le 3.4$. The flattening of the average photon index  can be understood in terms of the redshift effect of the mean intrinsic QSO spectrum consisting of two components -- a soft X-ray excess and a power-law component. We have also studied optical spectra of 12 of the 22 QSOs. ",
        "introduction": "QSOs or quasars represent the high luminosity end of a class of objects known as active galactic nuclei (AGNs). They are the most luminous continuously emitting objects in the universe and emit over the entire range of electromagnetic waves. It is widely believed that high energy emission (X-rays and $\\gamma$-rays) from an AGN originates in the innermost region of accretion disk around a super-massive black hole (SMBH) and much of the low energy emission e.g., infrared, optical, ultraviolet, is due to the reprocessing of high energy photons in a medium surrounding the accretion disk. While the atomic line emission in the optical, UV, and X-ray probe the circumnuclear medium surrounding the SMBH, the shape of the X-ray spectrum described by photon index ($\\Gamma_{X}$) is a critical parameter to constrain competing models for the mechanisms of X-ray continuum emission.  The increased sensitivity of the position sensitive proportional counter (PSPC) on-board {\\it ROSAT} compared to earlier missions allowed a significant improvement in the study of the soft X-ray emission of AGNs. The X-ray spectral shape of quasars has been studied extensively using earlier missions such as {\\it HEAO-I, Einstein, EXOSAT, and Ginga} (e.g., Mushotzky 1984; Wilkes \\& Elvis 1987; Canizares \\& White 1989; Comastri \\etal 1992; Lawson \\etal 1992; Williams \\etal 1992). Some of these missions were not sensitive below $2{\\rm~keV}$. These earlier studies suggested that the X-ray emission of quasars is well described by a power-law with photon index $\\Gamma_X \\sim 1.5$ for radio-loud quasars and $\\sim 2.0$ for radio quiet quasars. Large samples of AGNs have been studied by Walter \\& Fink (1993), Wang, Brinkmann \\& Bergeron (1996), and by Rush \\etal (1996) using the {\\it ROSAT} PSPC. However, most of the objects studied in the above papers are nearby and intrinsically bright AGNs. Also the AGN samples studied in the above papers are not complete and the derived results may be biased by selection effects. Laor \\etal (1994, 1997) studied X-ray spectra of a complete sample of 23 quasars from the Bright Quasar Survey (BQS) with $z\\le0.4$ and ${\\rm \\NH} < 1.9\\times10^{20}{\\rm~cm^{-2}}$. They found an average photon index of $2.63\\pm0.07$ for their complete sample. Reeves \\& Turner (2000) studied a larger sample of 62 quasars  with redshift in the range 0.06 -- 4.3 and ${\\rm M_{V}<-23}$. They found an average photon index, $\\Gamma_{X} = 1.66\\pm0.04$ for 35 radio-loud quasars and $\\Gamma_{X} = 1.89\\pm0.05$ for 27 radio-quiet quasars in their sample. Although the sample of Laor \\etal (1997) is complete, it is confined to the nearby universe. The  sample of Reeves \\& Turner (2000) includes some high redshift quasars but most of the quasars are nearby and the sample is not complete.  We present the results of a detailed spectral and timing analysis of a nearly complete sample of 22 type~1 QSOs obtained from the {\\it ROSAT} PSPC deep survey of McHardy \\etal (1998). The basic parameters of the QSOs are listed in Table~\\ref{basic_par}. We will refer to all the objects as QSO, independent of their luminosity, but note that some have low X-ray luminosity (see Table~\\ref{fit91_result}). The aims are to extend the study of soft X-ray properties of QSOs at higher redshifts and to investigate whether QSOs at high redshifts have excess soft X-ray emission similar to that seen in the narrow-line Seyfert~1 galaxies. The outline of the paper is as follows. In $\\S 2$ we describe the sample; in $\\S 3$ we describe the X-ray observations and analysis. Section 4 deals with optical spectroscopy. In $\\S 5$, we compare our results with other studies and discuss some of the implications. Finally, we conclude our study in $\\S 6$.  Throughout the paper, luminosities are calculated assuming isotropic emission, a Hubble constant of $H_{0}=75{\\rm~km~s^{-1}~Mpc^{-1}}$ and a deceleration parameter of $q_{0}=0$ unless otherwise specified.   {\\begin{figure*} \\centering \\includegraphics[angle=0,width=15cm]{mjm_hri_marked.ps} \\caption{MJM QSOs observed with the {\\it ROSAT} HRI. The image was binned by $16{\\rm~pixels}$ ($8{\\rm~arcsec}$) and then smoothed by convolving with a Gaussian of $\\sigma = 2{\\rm~pixels}$.} \\label{hri_qso} \\end{figure*}}   \\begin{table*} \\caption{Basic parameters of QSOs} \\label{basic_par} \\begin{flushleft} \\begin{tabular}{@{}ccccccccc} \\hline MJM$^1$ No. & {\\it ROSAT} name & \\multicolumn{2}{c}{HRI position$^2$} & \\multicolumn{2}{c}{Optical position$^3$} & z$^3$ & $m_{R}^3$ & Radio$^4$  \\\\ &           & $\\alpha (2000)$ & $\\delta (2000)$ & $\\alpha (2000)$ & $\\delta (2000)$      &               & \\\\ \\hline 2 &  RX J1334.7+3800 & 13 34 41.87 & 38 00 11.4 &  13 34 41.82 &    38 00 11.3 &   0.26 & 18.69   & (r)       \\\\ 3 &  RX J1333.7+3803 & 13 33 42.41 & 38 03 35.3 &  13 33 42.36 &    38 03 36.3 &   1.069 & 18.60  &  \\\\ 7 &  RX J1334.3+3757 &  13 34 17.52 & 37 57 22.1 &  13 34 17.52 &    37 57 22.4 &   1.14 &18.35    &  \\\\ 10 &  RX J1334.2+3759 & 13 34 10.57 & 37 59 56.4  & 13 34 10.62 &    37 59 56.3 &   0.38 &19.55   &   \\\\ 11 &  RX J1333.5+3746 & 13 33 32.32 & 37 46 42.4  & 13 33 32.01 &    37 46 41.1 &   0.826 &20.74  &   \\\\ 13 &  RX J1333.9+3759 & 13 33 58.49 & 37 59 39.3  & 13 33 58.55 &    37 59 38.2 &   1.61 &21.10   &   \\\\ 15 &  RX J1334.7+3759 & 13 34 42.81 & 37 59 15.0  & 13 34 42.77 &    37 59 15.0 &   1.14 &19.83   &   \\\\ 17 &  RX J1334.1+3754 & 13 34 00.96 & 37 54 04.7  & 13 34 01.03 &    37 54 04.9 &   1.64 &21.12   &   \\\\ 18 &  RX J1335.7+3751 & 13 35 44.46 & 37 51 43.3  & 13 35 44.66 &    37 51 40.8 &   1.621 &20.25  &   \\\\ 20 &  RX J1335.5+3757 & 13 35 30.26 & 37 57 50.5  & 13 35 30.30 &    37 57 50.0 &   1.39 &20.86   &   \\\\ 21 &  RX J1334.5+3748 & 13 34 31.30 & 37 48 31.2  & 13 34 31.33 &    37 48 31.4 &   1.359 &22.27 &    \\\\ 23 &  RX J1333.7+3757 & 13 33 44.34 & 37 57 53.6  & 13 33 44.27 &    37 57 52.6 &   0.97 & 20.37 &   \\\\ 24 &  RX J1335.6+3757 & 13 35:35.48 & 37:57:46.3  & 13 35 35.48 &    37 57 46.2 &   1.63 &20.21  &  \\\\ 29 &  RX J1335.7+3755 & 13 35 42.38 & 37 55 43.6  & 13 35 42.51 &    37 55 41.8 &   1.90 &19.30   &  \\\\ 30 &  RX J1334.9+3757 & 13 34 52.23 & 37 57 44.7  & 13 34 52.16 &    37 57 44.8 &   1.89 &20.47   & (r)\\\\ 31 &  RX J1333.9+3752 & 13 33 55.94 & 37 52 57.5  & 13 33 55.81 &    37 52 58.6 &  2.14 &20.32  &  \\\\ 37 &  RX J1334.4+3746 & 13 34 24.43 & 37 46 15.9  & 13 34 24.57 &    37 46 15.2 &   1.570 &20.11 &  \\\\ 48 &  RX J1335.5+3804 & 13 35 30.00 & 38 04 31.5  & 13 35 29.69 &    38 04 32.7 &  0.692 &19.17  &  \\\\ 56 &  RX J1334.7+3757 & 13 34 45.43 & 37 57 21.5  & 13 34 45.35 &    37 57 22.8 &   1.89 &20.36  &  \\\\ 57 &  RX J1333.6+3754 & 13 33 35.69 & 37 54 16.7  & 13 33 35.62 &    37 54 13.2 &   1.525 &21.68 &  \\\\ 61 &  RX J1333.6+3749 & --  & -- & 13 33 34.86 &    +37 49 16.92 &   3.43 &19.73  &  \\\\ 63 &  RX J1334.4+3806 & 13:34:22.44 &  38:06:21.5  & 13 34 22.24 &    38 06 20.1 &   2.593 &21.40 &  \\\\ \\hline \\end{tabular} \\newline $~^1$Identification number in the catalog of McHardy \\etal (1998). \\\\ $~^2$Derived from {\\it ROSAT} HRI observation of 1997. \\\\ $~^3$McHardy \\etal (1998) \\\\ $~^4$ (r) indicates that the source is detected in the preliminary 20-cm radio map. \\\\ \\end{flushleft} \\end{table*} ",
        "conclusions": "We have derived soft X-ray spectral shapes and light curves of a nearly complete sample of 22 QSOs. We also presented optical spectra of 12 QSOs from our sample. Our main results are as follows. \\\\ (i) About $33\\%$ of the QSOs show a long term ($\\sim 2{\\rm~yr}$) soft X-ray variability while only one QSO MJM~10 shows rapid variability. \\\\ (ii) Only three QSOs, MJM~7, 10, and 15 out of 22 QSOs ($\\sim7\\%$) show indications  for the presence of significant intrinsic absorption. The former two QSOs show excess absorption during both the observations of 1991 and 1993, while MJM~15 does not show excess absorption during the observation 1993.  \\\\ (iii) The soft X-ray photon index of the QSOs in our sample ranges from 1.4 to 3.7. The average photon index of the sample is $2.40\\pm0.09$ with a dispersion of $0.57$. \\\\ (iv) The average photon index of the QSOs is found to flatten at higher redshift. This can be understood in terms of the redshift effect of mean intrinsic QSO spectra consisting two components -- a soft excess component and a power-law. \\\\ (v) Only one QSO MJM~10 out of 22 has been found to be an NLQSO. \\\\ (vi) The broad component of the Mg~II line in the spectra of MJM~3, 7, and 20 are found to be blueshifted by $\\sim1180$, $\\sim790$, and $\\sim 1080{\\rm~km~s^{-1}}$, respectively."
    },
    "0208/astro-ph0208472_arXiv.txt": {
        "abstract": "{\\small  We sum up progress accomplished, since the last microquasar workshop, on the physics of the Accretion-Ejection Instability (AEI), and its ability to explain the properties of the low-frequency QPO of microquasars. These results concern the basic theory of the instability, its numerical simulation and the resulting modelisation of the QPO, as well as detailed observations of the QPO properties. They converge to reinforce the `magnetic flood' scenario, extrapolated from the AEI to explain the $\\sim$ 30 minutes cycles of \\G1915. We then discuss directions in which this scenario might be extended toward a more global view of the evolution of this source.} ",
        "introduction": "\\label{sec:Overview} This contribution is first a status report, in which we summarize recent results on the theory and numerical simulation of the Accretion-Ejection Instability (AEI), and on detailed observations of the properties of the low-frequency Quasi-Periodic Oscillation of microquasars, for which we believe that it provides a convincing explanation. This progress has been made in complementary directions (see the contributions of Peggy Varni\\`ere and Jer\\^ome Rodriguez, these proceedings):  \\begin{itemize} \\item {\\bf Theory:} We have \\cite{PV02B} computed what was only shown approximately in the original description of the AEI \\cite{TP99}, and justifies the `E' in its name: its unique ability to send upward, as Alfv\\'en waves propagating in the corona, the accretion energy  and angular momentum it extracts from the disk. We find that this mechanism is highly efficient, so that a large fraction of the accretion energy can be converted into Alfv\\'en waves. Future work should show how this energy can be deposited in the corona to energize a wind or jet. \\item {\\bf Numerical simulation:} % We have improved the numerical simulations presented by S.~Caunt \\cite{CT00} at the last workshop.  A simple model of heating and thickening of the disk  at the spiral shock formed by  the instability allows us to start producing\\footnote{in collaboration with M. Muno(MIT)} synthetic light curves, which can be compared with the observed ones. Preliminary results show in particular that the AEI can reproduce the high rms amplitude of the QPO.  \\item{\\bf Observation:} We have now published in final form the results presented at the last workshop, comparing  theory and observation of relativistic effects when the inner disk edge approaches the last stable orbit, in particular in \\J1655 and \\G1915 \\cite{JR02A}-\\cite{PV02A}. Although the observational  evidence is fragile  and can only be taken as an indication, an additional hint is now provided by observations of \\X1550 \\cite{JR02B}. \\item {\\bf Energy spectrum of the QPO:} It is commonly observed that  the QPO is best correlated with the properties of the disk, although  it affects more strongly the coronal emission, {\\em i.e.} the power-law tail. However, in the same observations of \\X1550, we find  that the modulated emission has a different energy spectrum than this power-law. This is consistent with the presence of a hot point  in the disk {\\em i.e.},  in our interpretation, the spiral shock.  \\end{itemize} Thus these results converge to confirm the expected properties of the AEI and its ability to explain the main characteristics of the QPO. These expectations were the basis of the `Magnetic Flood' scenario\\cite{Flood99}, which starts from the identification of the QPO with the AEI, and extrapolates to give a tentative explanation for the $\\sim$30 minutes cycles of \\G1915.  In this scenario the cycles are determined by the processing of the poloidal (vertical) magnetic flux advected with the gas in the disk.  We will show below how we have found a positive indication in favor of this scenario, in the detailed analysis of one of these cycles.  We will then turn to unpublished work, presented in the undergraduate thesis of Fitzgibbon (1999) with E. Morgan and R. Remillard at MIT: defining states analogous to the ones of Belloni {\\em et al.}\\cite{BELLONI00}, they show a striking regularity in the succession of these states: the source does not err between them, but repeatedly follows a well-defined sequence between them. We will briefly discuss how the  Magnetic Flood scenario could be extrapolated again to explain  this regularity. ",
        "conclusions": ""
    },
    "0208/physics0208023_arXiv.txt": {
        "abstract": "The principle of the Karlsruhe dynamo experiment is closely related to that of the Roberts dynamo working with a simple fluid flow which is, with respect to proper Cartesian co--ordinates $x$, $y$ and $z$, periodic in $x$ and $y$ and independent of $z$. A modified Roberts dynamo problem is considered with a flow more similar to that in the experimental device. Solutions are calculated numerically, and on this basis an estimate of the excitation condition of the experimental dynamo is given. The modified Roberts dynamo problem is also considered in the framework of the mean--field dynamo theory, in which the crucial induction effect of the fluid motion is an anisotropic $\\alpha$--effect. Numerical results are given for the dependence of the mean--field coefficients on the fluid flow rates. The excitation condition of the dynamo is also discussed within this framework. The behavior of the dynamo in the nonlinear regime, i.e.\\ with backreaction of the magnetic field on the fluid flow, depends on the effect of the Lorentz force on the flow rates. The quantities determining this effect are calculated numerically. The results for the mean--field coefficients and the quantities describing the backreaction provide corrections to earlier results, which were obtained under simplifying assumptions.  Key words: dynamo, dynamo experiment, mean--field dynamo theory, $\\alpha$--effect, Lorentz force ",
        "introduction": "\\label{intro}  In the Forschungszentrum \\mbox{Karlsruhe} U.\\ M\\\"uller and R.\\ Stieg\\-litz have set up an experimental device for the demonstration and investigation of a homogeneous dynamo as it is expected in the Earth's interior or in cosmic bodies \\cite{stieglitzetal96}. The experiment ran first time successfully in December 1999, and since then several series of measurements have been carried out \\cite{muelleretal00,stieglitzetal01,stieglitzetal02,muelleretal02}. It is the second realization of a homogeneous dynamo in the laboratory. Its first run followed only a few weeks after that of the Riga dynamo experiment, working with a somewhat different principle, which was pushed forward by A.\\ Gailitis, O.\\ Lielausis and co--workers \\cite{Gai00,Gai01}.  The basic idea of the Karlsruhe experiment was proposed in 1975 by F.~H.\\ Busse \\cite{busse75,busse92}. It is very similar to an idea discussed already in 1967 by A.\\ Gailitis \\cite{gailitis67}. The essential piece of the experimental device, the dynamo module, is a cylindrical container as shown in Fig.~\\ref{module}, with both radius and height somewhat less than 1m, through which liquid sodium is driven by external pumps. By means of a system of channels with conducting walls, constituting 52 ``spin generators\", helical motions are organized. The flow pattern resembles one of those considered in the theoretical work of G.~O.\\ Roberts in 1972 \\cite{robertsgo72}. This kind of Roberts flow, which proved to be capable of dynamo action, is sketched in Fig.~\\ref{robflow}. In a proper Cartesian co-ordinate system $(x, y, z)$ it is periodic in $x$ and $y$ with the same period length, which we call here $2 a$, but independent of $z$. The $x$ and $y$--components of the velocity can be described by a stream function proportional to $\\sin (\\pi x /a) \\sin (\\pi y /a)$, and the $z$--component is simply proportional to $\\sin (\\pi x /a) \\sin (\\pi y /a)$. When speaking of a ``cell\" of the flow we mean a unit like that given by $0 \\leq x, y \\leq a$. Clearly the velocity is continuous everywhere, and at least the $x$ and $y$--components do not vanish at the margins of the cells. The real flow in the spin generators deviates from the Roberts flow in the way indicated in Fig.~\\ref{spingenflow}. In each cell there are a central channel and a helical channel around it. In the simplest approximation the fluid moves rigidly in each of these channels, and it is at rest outside the channels. We relate the word ``spin generator flow\" in the following to this simple flow. In contrast to the Roberts flow the spin generator flow shows discontinuities and vanishes at the margins of the cells.  The theory of the dynamo effect in the Karlsruhe device has been widely elaborated. Both direct numerical solutions of the induction equation for the magnetic field \\cite{tilgner96,tilgner97,tilgner00,tilgneretal01, tilgneretal02,tilgner02,tilgner02b} as well as mean--field theory and solutions of the corresponding equations \\cite{raedleretal96,raedleretal97a,raedleretal98a,raedleretal02a,raedleretal02b,raedleretal02c} have been employed. We focus our attention here on this mean--field approach. In this context mean fields are understood as averages over areas in planes perpendicular to the axis of the dynamo module covering the cross--sections of several cells. The crucial induction effect of the fluid motion is then, with respect to the mean magnetic field, described as an anisotropic $\\alpha$--effect. The $\\alpha$--coefficient and related quantities have first been calculated for the Roberts flow \\cite{raedleretal97a,raedleretal96,raedleretal97b,raedleretal02a,raedleretal02b}. In the calculations with the spin generator flow carried out so far, apart from the case of small flow rates, a simplifying but not strictly justified assumption was used. The contribution of a given spin generator to the $\\alpha$--effect was considered independent of the neighboring spin generators and in that sense determined under the condition that all its surroundings are conducting fluid at rest \\cite{raedleretal97a,raedleretal97b,raedleretal02a,raedleretal02b}. An analogous assumption was used in calculations of the effect of the Lorentz force on the fluid flow rates in the channels of the spin generators \\cite{raedleretal02a,raedleretal02c}. It remained to be clarified which errors result from these assumptions.  The main purpose of this paper is therefore the calculation of the $\\alpha$--coefficient and a related coefficient as well as the quantities determining the effect of the Lorentz force on the fluid flow rates for an array of spin generators, taking into account the so far ignored mutual influences of the spin generators. In Section \\ref{dynprob} the modified Roberts dynamo problem with the spin generator flow is formulated. In Section \\ref{nummeth} the numerical method used for solving this problem and the related problems occurring in the following sections are discussed. Section \\ref{excond} presents in particular results concerning the excitation condition for the dynamo with spin generator flow. In Section \\ref{mfappr} various aspects of a mean--field theory of the dynamo experiment are explained and results for the mean electromotive force due to the spin generator flow are given. Section \\ref{lorentz} deals with the effect of the Lorentz force on the flow rates in the channels of the spin generators. Finally in Section \\ref{conclus} some consequences of our findings for the understanding of the experimental results are summarized.  Independent of the recent comprehensive accounts of the mean--field approach to the Karlsruhe dynamo experiment \\cite{raedleretal02a,raedleretal02b,raedleretal02c}, this paper may serve as an introduction to the basic idea of the experiment. However, we do not strive to repeat all important issues discussed in those papers, but we mainly want to deliver the two supplements mentioned above.  \\begin{figure}\\includegraphics[width=.45\\textwidth]{fig1.ps}\\caption[]{ The dynamo module (after \\cite{stieglitzetal96}). The signs + and -- indicate that the fluid moves in the positive or negative $z$--direction, respectively, in a given spin generator. $R = 0.85\\,$m, $H = 0.71\\,$m, $a = 0.21\\,$m. }\\label{module}\\end{figure}  \\begin{figure}\\includegraphics[width=.45\\textwidth]{fig2.ps}\\caption[]{ The Roberts flow pattern. The flow directions correspond to the situation in the dynamo module if the co--ordinate system coincides with that in Fig.~\\ref{module}. }\\label{robflow}\\end{figure}  \\begin{figure}\\includegraphics[width=.45\\textwidth]{fig3.ps}\\caption[]{ The spin generator flow pattern. As for the flow directions the remark given with Fig.~\\ref{robflow} applies. The fluid outside the cylindrical regions where flow directions are indicated is at rest. There are no walls between the cells. }\\label{spingenflow}\\end{figure} ",
        "conclusions": "\\label{conclus}  We have first dealt with a modified Roberts dynamo problem with a flow pattern resembling that in the Karls\\-ruhe dynamo module. Based on numerical solutions of this problem a self--excitation condition was found. Since in these calculations neither the finite radial extent of the dynamo module nor realistic boundary conditions at its plane boundaries were taken into account this self--excitation condition deviates markedly from that for the experimental device.  A mean--field approach to the modified Roberts dynamo problem is presented. Two slightly different treatments are considered, assuming as usual only weak variations of the mean magnetic field in space, or admitting arbitrary variations in the $z$--direction. The coefficient $\\alpha_\\perp$ describing the $\\alpha$--effect and a coefficient $\\beta$ connected with derivatives of the mean magnetic field are calculated for arbitrary fluid flow rates. The result for $\\alpha_\\perp$ corrects earlier results obtained in an approximation that ignores the mutual influences of the spin generators \\cite{raedleretal97b}. It leads to a much better agreement of the calculated self--excitation condition with the experimental results \\cite{raedleretal02a,raedleretal02b}. We note in passing that in the case of small flow rates our result, although calculated for rigid--body motions only, applies also for more general flow profiles \\cite{raedleretal02a,raedleretal02b}. The result for $\\beta$ suggests that the enlargement of the effective magnetic diffusivity by the fluid motion can be partially compensated by another effect of this motion. The same has been observed in investigations with the Roberts flow \\cite{plunianetal02b}. This could be one of the reasons why the results calculated under idealizing assumptions, in particular ignoring the effect of the mean--field diffusivity, deviate only little from the experimental results \\cite{raedleretal02b}.  In the framework of the mean--field approach we have also given an estimate of the excitation condition which considers the finite radial extent of the dynamo module. It shows that the real extent enhances the critical value of $C$, which is a dimensionless measure of $\\alpha_\\perp$, by a factor 2. In other words, if in the region of $V_{\\mathrm{C}}$ and $V_{\\mathrm{H}}$, in which experimental investigations have been carried out, $V_{\\mathrm{C}}$ is fixed, $V_{\\mathrm{H}}$ has to be larger by a factor between 2.5 and 3.5. If the excitation condition is corrected in this way it does not underestimate the requirements for self--excitation.  We have also calculated the effect of the Lorentz force on the fluid flow rates in the channels of a spin generator. Again our result corrects a former one obtained in the approximation already mentioned which ignores the mutual influences of the spin generators \\cite{raedleretal02a,raedleretal02b}. The braking effect of the Lorentz force proves to be stronger than predicted by the former calculations. This means in particular that estimates of the saturation field strengths given so far \\cite{raedleretal02a,raedleretal02c} have to be corrected by factors between 0.8 and 0.9; for more details see Note added in proof in \\cite{raedleretal02c}.   \\bigskip   {\\bf Acknowledgement} The results reported in this paper have been obtained during stays of K.-H.R. at NORDITA. He is grateful for its hospitality. An anonymous referee is acknowledged for making useful suggestions."
    },
    "0208/astro-ph0208228_arXiv.txt": {
        "abstract": "We report a calculation of the expected rate of inclined air showers induced by ultra high energy cosmic rays to be obtained by the Auger Southern Observatory assuming different mass compositions. We describe some features that can be used to distinguish photons at energies as high as 10$^{20}$ eV. The discrimination of photons at such energies will help to test some models of the origin of ultra-high-energy cosmic rays. ",
        "introduction": "Uncovering the origin, composition and energy spectrum of the highest energy cosmic rays is one of the biggest challenges in astroparticle physics.  The Pierre Auger project is the next step in the search for answers to intriguing questions about the origin of these particles \\cite{augerreport}.  The Auger Southern Observatory will consist of 1600 water Cerenkov detector stations (each 10 m$^2$ x 1.2 m deep) on a hexagonal grid of 1500 m spacing overlooked by four detectors capable of detecting the fluorescence light emitted by the nitrogen molecules excited by the shower. The array covers a ground area of approximately 3000 km$^2$ at a mean altitude of 875 g cm$^{-2}$ ($\\sim 1400$~m), near Malarg\\\"ue in Mendoza State, Argentina (lat = -35.2$^\\circ$, long = -69.2$^\\circ$).  The Auger Southern Observatory will be able to measure the energy of the incoming cosmic ray using fluorescence detectors (FD) and thus calibrate the energy inferred from the surface detectors.  The proposal to use the Auger observatories to search for very inclined showers induced by ultrahigh energy neutrinos \\cite{zas} has led to an investigation of the characteristics of high energy showers at large zenith angles, i.e. showers arriving at zenith angles larger than 60$^0$~\\cite{hplp}. These showers would not be very different from vertical showers except for the fact that they develop in the upper part of the atmosphere. As a result the electromagnetic part of the shower, produced mainly from $\\pi^0$ decay, is mostly absorbed well before the shower front reaches ground level. The muon front propagates through the atmosphere and it gets attenuated more slowly through pair production, bremsstrahlung, and hadronic interactions that reduce the muon energy and increase their probability to decay in flight. Therefore the muon energy spectrum at ground will have a {\\sl low energy} cutoff which increases as the zenith angle rises. As a result the average muon energy at ground level also increases. Although the bulk of the overall increase in the average muon energy with zenith is due to the rise of the cutoff, there is a smaller contribution due to the rise of the pion critical energy (the energy at which the pions are more likely to decay than to interact). These energetic muons travel making a small angle to the incoming cosmic ray direction but their trajectories are deflected by the Earth's magnetic field. As a result the muon density patterns at ground are different from typical densities measured in vertical showers, reflecting the structure of the shower core and the geomagnetic field which acts as a kind of ``natural magnetic spectrometer''.  This work outlines the sensitivity of the Auger Observatory to inclined air showers induced by ordinary cosmic rays, that is cosmic rays produced by protons, heavy nuclei, and photons. They constitute the background to neutrino detection but their observation also provides a significant increase in the aperture of the array and may improve mass composition studies~\\cite{PRL}. Below, we predict the expected rate for the Auger Surface Array for zenith angles above $60^{\\circ}$ assuming different primary mass compositions: proton, iron and photons.  Primary photons can interact with the geomagnetic field before reaching the top of the atmosphere \\cite{magnetic}. This will have a major impact on the expected rate of detection for those primaries, as we will show. We can take advantage of this effect to outline a way to identify photon primaries at energies as high as 10$^{20}$ eV.  The article is organized as follows: in section 2 we describe the procedure to calculate the expected rate for the Auger array above 60$^\\circ$ for primary protons, giving the expected values for the energy resolution, core error reconstruction, and multiplicity of stations as a function of the energy: a detailed explanation of the procedure can also be found in \\cite{hplp}; in section 3 we describe the effect of the geomagnetic interactions in inclined air showers initiated by photon primaries; in section 4 we present the expected rate for three primary mass compositions (protons, iron and photons). Finally in section 5 we end up with some conclusions. ",
        "conclusions": "\\label{conclu}  Inclined showers will be seen in the Auger Observatory as spectacular events with as many as 30 or 40 hit detectors. We have calculated the approximate rate of inclined showers with zenith angle exceeding $60^\\circ$ expected to be observed at the Pierre Auger Observatory. This rate increases the aperture of the observatory by almost a factor 2. Assuming a pure proton composition, there will be over 1000 well reconstructed events above $10^{19}$ eV, with a mean error energy $\\sim 25$~\\%. The rate is sensitive to composition. If photons were dominant at high energy, the rate would be an order of magnitude smaller than if they were protons or nuclei allowing for a clear discrimination of the two cases. Uncertainties in the physics at very high energies have implications for our results on the detailed quantitative predictions but these uncertainties have little impact on the previous conclusions.  There are other signatures of the presence of photons (see for instance ref.~\\cite{bertou}). Surely the combination of the inclined shower rate measurements together with vertical flux determinations and detailed analysis of the expected photon signatures will be a great step in the establishment of the overall photon rate at very high energies with the forthcoming data from the Auger experiment."
    },
    "0208/astro-ph0208520_arXiv.txt": {
        "abstract": "{ We assume that the gravitational instability of standard thin accretion disks leads to the Broad Line Regions (BLRs), the B band luminosity comes from standard thin disk and the motion of BLRs is virial. The central black hole masses, the accretion rates and the disk inclinations to the line of sight for 17 Seyfert 1 galaxies and 17 Palomar-Green (PG) quasars have been calculated. Our results are sensitive to $\\alpha$ parameter of the standard $\\alpha$ disk. With the same values of $\\alpha$ ($\\alpha=1$), calculated central black hole masses for 17 Seyfert 1 galaxies are consistent with that from Kaspi et al. (2000) while that for 17 PG quasars are larger than that from Kaspi et al. (2000) by almost 2 orders of magnitude. Inclinations of 17 Seyfert 1 galaxies are about 6 times larger than that of 17 PG quasars. These inclinations, with a mean value of $32^{o}$ for 17 Seyfert 1 galaxies that agrees well with the result obtained by fitting the iron $K\\alpha$ lines of Seyfert 1 galaxies observed with ASCA (Nandra et al. 1997) and the result obtained by Wu \\& Han (2001), provide further support for the orientation-dependent unification scheme of active galactic nuclei. There is a relation between the FWHM of H$\\beta$ and the inclination, namely the inclination is smaller in AGNs with smaller FWHM of H$\\beta$. The effect of inclinations in narrow line Seyfert 1 galaxies (NLS1s) should be considered when one studies the physics of NLS1s. The need of higher value of $\\alpha$ for PG quasars maybe shows that our model is not suitable for PG quasars if we think inclinations of PG quasars are in the inclination levels of Seyfert 1 galaxies. With our model, we also show that the size of BLRs relates not only to the luminosity, but also to the accretion rates. More knowledge of BLRs dynamics, accretion disks and optical luminosity are needed to improve the determinations of black hole masses, accretion rates and inclinations in AGNs. ",
        "introduction": "\\label{intro}  Broad emission lines are one of the dominant features of many Active Galactic Nuclei (AGNs) spectrum. Broad Line Regions (BLRs) play a particularly important role in our understanding of AGNs by virtue of its proximity to the central source. With reverberation mapping technique the sizes of the BLRs can be obtained through the study of correlated variations of the lines and continuum fluxes (Peterson 1993). The BLR sizes for 17 Seyfert 1 galaxies (Wandel et al. 1999) and for 17 PG quasars (Kaspi et al. 2000) have been recently obtained. Assuming the Keplerian motions and the random orbits of the BLRs, the central black hole masses also have been obtained. In order to underline the physics of AGNs, these masses have been used to check the relations with other parameters of AGNs, such as the radio power (Ho 2002), the X-ray excess variance (Lu \\& Yu 2001), the velocity dispersions of their host galaxies (Ferrarese et al. 2001; Gebhardt et al. 2000b).  Based on the unified model of AGNs, the wide variety of AGN phenomena we see is due to a combination of real differences in a small number of physical parameters (e.g. luminosity) coupled with apparent differences which are due to observer-dependent parameters (e.g. orientation). AGNs with wide emission lines from BLRs are oriented at a preferred angle from which BLRs is visible. No edge-on disks in AGNs with wide emission lines would be seen (Urry \\& Padovani 1995). Therefore, inclinations of AGNs with wide emission lines are generally expected to be small although further evidence is obviously needed. It is important to test it with inclinations of AGNs. Random mean angle is often assumed in calculated black hole masses with the reverberation mapping method. Some authors (McLure \\& Dunlop 2001; Wu \\& Han 2001) have shown it is necessary to consider the effects of the inclination.  Though it is powerful for size determination, currently reverberation mapping method does not provide the velocity field of the BLRs, and can not distinguish between radial and rotational motions. Because there is no consensus about the dynamics of the BLRs, there has been little progress in understanding the physics of this region. Recently some authors (Collin \\& Hure 2001) suggest the gravitationally unstable disc is the source which releases BLRs in the medium.  In this paper, we also assume the gravitational instability leads to the formation of BLRs. With the analysis formulae of standard thin accretion disks and the observational sizes of the BLRs, we calculate the central masses, accretion rates and inclinations for 34 AGNs. We want to give the clues to the origin of BLRs, the values of three parameters of accretion disks and the inclination effect in mass determination in AGNs. The paper is structured as follows. In Sect. 2, we describe our model and available data. In sect. 3, we give our calculational results. Our discussion and conclusions are presented in the last section. ",
        "conclusions": "\\subsection{The large accretion rates for Seyfert 1 galaxies.}  In Fig.7 and Fig.8, though the accretion rates in units of the Eddington accretion rates for PG quasars are consistent with the results of Laor (1990), we can see that accretion rates in units of the Eddington accretion rates of  the most of Seyfert 1 galaxies is larger than one. We confirmed the results of Collin \\& Hure (2001). They also find it radiates at super-Eddington rates if a standard accretion disc accounts for the observed optical luminosity. They can't account for the effect of the inclination. In this paper we consider the effect of the inclinations and uncertainties of accretion rates from Q, BLRs sizes and B absolute magnitude. Although we find there also exists high accretion rates for Seyfert 1 galaxies, we should notice the uncertainties about accretion rates (Fig. 8). The large accretion rates are maybe due to the uncertainty of dynamics of BLRs, median absolute B magnitudes. More knowledge of BLRs dynamics, accretion disks and optical luminosity are needed to improve the determinations of accretion rates in AGNs.  \\subsection{The disk inclination to the line of sight}  With our calculation, there is apparent difference in inclinations between Seyfert 1 galaxies and PG quasars (Fig.3, Table 1) when we adopt  $\\alpha=1$ for all objects. Inclinations for quasars are smaller than that for Seyfert galaxies. The mean and error of inclinations for 17 Seyfert 1 galaxies are $32.2\\pm5.5$ (deg) ($\\alpha=1$). The mean and error of inclinations for 17 PG quasars are $5.65\\pm1.08$ (deg) ($\\alpha=1$). Only 4 quasars have inclinations larger than $10^{o}$. The mean uncertainty of inclinations of quasars is large, about $4.7^{o}$ (Table 4). The mean value of inclinations for PG quasar can be about $10^{o}$ as upper limit considering the uncertainties of Q, BLR sizes and B magnitude. We noticed that the ratio between the black hole mass determined by reverberation mapping technique ($M_{rm}$) and our calculated black hole mass ($M_{cal}$) can be approximated by $3(sini)^{2}$ (Wu \\& Han, 2001). With the mean value of inclinations ($32^{o}$) derived by us, we obtained $M_{rm}/M_{cal}=1.19$ for Seyfert 1 galaxies. It means that the black hole mass estimated by the standard method of reverberation mapping can still represent the \"true\" black hole mass well if the inclination of the Seyfert galaxy is not substantially different from $32^{o}$. With the mean value of inclinations ($5.65^{o}$) derived by us, we obtained $M_{rm}/M_{cal}=34$ for PG quasars. According to our model and when using $\\alpha=1$, it is necessary to consider the effect of inclinations to calculate the central black hole masses for quasars.  Based on AGNs unification schemes, AGNs with wide emission lines from BLRs are oriented at a preferred angle from which BLRs is visible. No edge-on disks would be seen (Urry \\& Padovani 1995). Our calculated inclinations, with a mean value of $32^{o}$ for 17 Seyfert 1 galaxies that agrees well with the result obtained by fitting the iron $K\\alpha$ lines of Seyfert 1 galaxies observed with ASCA (Nandra et al. 1997) and the result obtained by Wu \\& Han (2001), provide further support for the orientation-dependent unification scheme of active galactic nuclei. Based on the unified model of AGNs, the wide variety of AGN phenomena we see is due to a combination of real differences in a small number of physical parameters (e.g. luminosity) coupled with apparent differences which are due to observer-dependent parameters (e.g. orientation). The unification scheme does not explain the difference between Seyfert 1 galaxies and quasars in an inclination effect (as they both should have the same inclinations of close to face on) but explain it as a luminosity difference. We here find that inclinations of PG quasars are smaller than that of Seyfert 1 galaxies if we adopt the same value of $\\alpha$ for all objects ( Fig.3, Table 1) . Do luminous quasars prefer to face on ? We adopt different values of parameter $\\alpha$ and try to bring quasars inclinations to the inclination levels of Seyfert 1 galaxies. We use larger $\\alpha$ to recalculated inclinations for PG quasars. The mean and error of inclinations for 17 PG quasars is $13.4^{+10.4}_{-5.8}$ (deg) ($\\alpha=10$) and $29.7^{+15.0}_{-12.0}$ (deg) ($\\alpha=100$), which are in the inclination levels of Seyfert 1 galaxies. The errors of Q, absolute B band magnitude, BLRs sizes are considered to calculate the uncertainties of inclinations. It is possible that the difference between PG quasars and Seyfert 1 galaxies is maybe due to the difference of the value of $\\alpha$. From above all we may find calculated inclinations of PG quasars are smaller than that of Seyfert 1 galaxies unless we adopt larger value of $\\alpha$ for PG quasars to calculate in our model. However we often assume that the value of $\\alpha$ of disc in AGNs is often between 0 and 1 (Shakura \\& Sunyaev 1973) . The need of higher value of $\\alpha$ (larger than one) for PG quasars maybe shows that our model is not suitable for PG quasars if we think inclinations of PG quasars are in the inclination levels of Seyfert 1 galaxies.    In Fig.3 we can find the NLS1s (here we simply define the NLS1s with FWHM of H$\\beta$ is less than 2000 $km s^{-1}$) have the smaller inclinations except for NGC4051. We should notice the larger scatter about the value of inclination for NGC4051 (Table 1). Collin \\& Hure (2001) suggested the sizes of the BLRs will increase with the accretion rates expressed in Eddington units and decrease with the black hole masses. The larger size of BLRs led to the smaller widths of H$\\beta$ in NLS1s. They omitted the effect of the inclinations. The virial BH mass is given by $M=\\frac{V_{FWHM}^{2}}{4(sini^{2}+A^{2})}RG^{-1}$. Because the difference for the sizes of BLRs is not big, NLS1s with small values of $V_{HWHM}$ will have smaller BH masses when we don't consider the effect of inclinations. The effect of inclinations in NLS1s should be considered when we study the physics of NLS1s. In Fig.6 we show the velocity of BLRs and we find the intrinsic widths of H$\\beta$ for NLS1s are not smaller than that of broad line AGNs, which is different from the results of Wu \\& Han (2001), which is from 11 Seyfert 1 galaxies.   \\subsection{The size of BLRs}  There is a natural idea that BLRs are made of the atmosphere or winds of giant stars (Edwards 1980). Another idea is that BLRs are from the accretion disk or the wind released at the top of the disk (Murray \\& Chiang 1997). With the reasonable values of $M_{cal}$ and inclinations, we show that the gravitational instability can indeed explain the size of BLRs. Assuming the gravitational instability of standard thin accretion disk leads to the Broad Line Regions(BLRs), the B band luminosity comes from standard thin disk and the motion of BLRs is virial, we can obtain $R_{BLR} \\propto L_{B}^{0.5}\\dot{M}^{-37/45}$ from Eq. 1 and Eq. 2. If $L_{5100} \\propto L_{B}$, there is a correlation $R_{BLR} \\propto L_{5100}^{0.5}\\dot{M}^{-37/45}$. The size of BLRs relates not only to the luminosity, but also to the accretion rates. Nicastro (2000) proposed that the BLRs are released by the accretion disk in the region where a vertically outflowing corona exists and he found the sizes of BLRs would relate to $\\dot{M}$. Collin \\& Hure (2001) showed the size of BLRs mainly related to $M$, not $\\dot{M}$. Based on the formulae $R_{BLR} \\propto L_{5100}^{0.5}\\dot{M}^{-37/45}$, the difference of the accretion rate for the Seyfert 1 galaxies is not much and there is a relation as $R_{BLR} \\propto L_{5100}^{0.5}$, the power index is almost 0.5, which can explain the correlation between the size of BLRs($R_{BLR}$) and the monochromatic luminosity at 5100\\AA ($L_{5100}$), $R_{BLR} \\propto L_{5100}^{0.5}$ (Wandel et al. 1999). The $\\dot{M}$ is different for Seyfert 1 galaxies and quasars. The $\\dot{M}$ of luminous quasars are small (see Fig.7, Fig.8). The index will be higher for the sample of Kaspi et al. (2000) which includes 17 Seyfert 1 galaxies and 17 PG quasars. It is necessary to check this with a larger sample.   \\subsection{Conclusion}  With the formulae of the standard thin disk and the assumption of gravitational instability leading to BLRs, we calculate the central black hole masses, accretion rates and inclinations for 34 AGNs. The main conclusions can be summarized as follows: \\begin{itemize} \\item{The gravitational instability can indeed explain the size of BLRs. The size of BLRs relates not only to the luminosity, but also to the accretion rates.} \\item{Our results are sensitive to $\\alpha$ parameter of the standard $\\alpha$ disk. $\\alpha$ is 1 in all the calculations. The mean value of inclinations for 17 Seyfert 1 galaxies is $32^{o}$ , which is favoring the orientation-dependent unification scheme of AGNs. Inclinations of 17 Seyfert 1 galaxies are about 6 times larger than that of 17 PG quasars unless the value of $\\alpha$ of PG quasars is larger than Seyfert 1 galaxies. The need of higher value of $\\alpha$ for PG quasars maybe shows that our model is not suitable for PG quasars if we think inclinations of PG quasars are in the inclination levels of Seyfert 1 galaxies. There is a correlation between inclinations and the FWHMs of H$\\beta$. Though the observed FWHMs of H$\\beta$ for NLS1s is smaller than that for broad line AGNs, the intrinsic of FWHMs of H$\\beta$ for NLS1s are not smaller than that of broad line AGNs. Small inclinations lead to the small FWHMs of H$\\beta$ for NLS1s.} \\item{The uncertainty of absolute B band magnitude (here we adopt 1 magnitude) from variability or the hosts contribution leads to larger scatter of our results than error of size of BLRs. More knowledge of BLRs dynamics, accretion disks and optical luminosity are needed to improve the determinations of black hole masses, accretion rates and inclinations in AGNs.}   \\end{itemize}"
    },
    "0208/astro-ph0208185_arXiv.txt": {
        "abstract": "{ Stellar spectroscopic classification has been successfully automated by a number of groups.  Automated classification and parameterization work best when applied to a homogeneous data set, and thus these techniques primarily have been developed for and applied to large surveys.  While most ongoing large spectroscopic surveys target extragalactic objects, many stellar spectra have been and will be obtained.  We briefly summarize past work on automated classification and parameterization, with emphasis on the work done in our group.  Accurate automated classification in the spectral type domain and parameterization in the temperature domain have been relatively easy.  Automated parameterization in the metallicity domain, formally outside the MK system, has also been effective.  Due to the subtle effects on the spectrum, automated classification in the luminosity domain has been somewhat more difficult, but still successful. In order to extend the use of automated techniques beyond a few surveys, we present our current efforts at building a web-based automated stellar spectroscopic classification and parameterization machine.  Our proposed machinery would provide users with MK classifications as well as the astrophysical parameters of effective temperature, surface gravity, mean abundance, abundance anomalies, and microturbulence.} ",
        "introduction": " ",
        "conclusions": ""
    },
    "0208/astro-ph0208300_arXiv.txt": {
        "abstract": "How do planets form from circumstellar disks of gas and dust? What physical processes are responsible for determining the final masses of forming stars and ultimately the initial mass function (IMF)? The Hubble Space Telescope has made major contributions in helping to address these fundamental questions.  In the next decade, the Space Infrared Telescope Facility (SIRTF) and the Next Generation Space Telescope (NGST) will build on this heritage in the near-- to far--infrared.  However several crucial questions will remain. We review recent progress made in star and planet formation with HST, summarize key science objectives for SIRTF and NGST, and suggest problems that would be uniquely suited to a large aperture UV/optical space--based telescope.  We focus on studies that take advantage of high spatial resolution and the unique wavelength range not accessible from the ground such as: 1) circumstellar disk structure and composition (resolved images of dust and UV spectroscopy of gas); and 2) extreme populations of young stars in the local group (UV imaging and spectroscopy of massive star--forming regions). An 8m UV/O space telescope operating from 1100--6000 $\\AA$ over fields of view 4--10' and with spectroscopic capabilities from R = 3,000--10,000 down to $<$ 1000 $\\AA$ would be a powerful tool for star and planet formation research. ",
        "introduction": "Star and planet formation will remain key themes in the NASA Origins Program through the next decade and beyond.  During this time, the community will enjoy access to: 1) continued HST operation with STIS, NICMOS, ACS, COS, and WFC--3 with wavelength coverage from 0.1150--2.5 $\\mu$m; 2) SIRTF with launch in early 2003 (lifetime $\\sim$ 5 years) covering the wavelength range from 3--160 $\\mu$m; and 3) NGST scheduled for launch in 2010 with projected 5 year mission covering the wavelength range 0.6--28 $\\mu$m.  Because important processes in star and planet formation occur over a wide range of temperatures, multi--wavelength observations are required in order to make significant progress (see overview by Hartmann this volume).  Both SIRTF and NGST will execute programs aimed at understanding the emergence of planetary systems from circumstellar disks around young stars, and the origins of stellar masses. Approved SIRTF guaranteed time programs will focus on circumstellar disk evolution and the nature and frequency of brown dwarf objects. In addition, three of the six adopted Legacy Science Programs (GLIMPSE, From Molecular Clores to Planet--Forming Disks, Formation and Evolution of Planetary Systems) will directly impact star and planet formation research (http://sirtf.caltech.edu/SSC/legacy/).  Two of the five themes in the Design Reference Mission for the NGST (The Birth and Formation of Stars and The Origins and Evolution of Planetary Systems) deal with similar topics (http://ngst.gsfc.nasa.gov/science/drm.html).  Observations in the UV/optical range are also important to obtain a complete picture of star and planet formation. Key observations include: 1) sampling the Wien peak of massive star energy distributions; 2) measuring important gas diagnostics for infall/outflow; and 3) studying resolved images of dust disks.  As in the infrared spectral regime, space offers unique capabilities for UV/optical astronomy such as the accessible range from 912--3000 $\\AA$, and diffraction--limited imaging over a large field of view at wavelengths $<$ 1 $\\mu$m.  Here, we concentrate on two areas where a large aperture UV/Optical space telescope could make substantial contributions.  We begin by identifying key issues, review the legacy of HST in each area, summarize the promise of SIRTF for studies of circumstellar disks and NGST for investigations concerning the origin of the IMF, and suggest parameters required in order for a UV/optical space telescope to realize its potential for star and planet formation research. ",
        "conclusions": ""
    },
    "0208/astro-ph0208136_arXiv.txt": {
        "abstract": "With the aim of studying the nonlinear stellar and gaseous response to the gravitational potential of a galaxy such as the Milky Way, we have modeled 3D galactic spiral arms as a superposition of inhomogeneous oblate spheroids and added their contribution to an axisymmetric model of the Galactic mass distribution. Three spiral loci are proposed here, based in different sets of observations. A comparison of our model with a tight-winding approximation shows that the self-gravitation of the whole spiral pattern is important in the middle and outer galactic regions. A preliminary self-consistency analysis taking ${\\Omega}_{p}$ = 15 and 20~km~s$^{-1}$~kpc$^{-1}$ for the angular speed of the spiral pattern, seems to favor the value ${\\Omega}_{p}$ = 20~km~s$^{-1}$~kpc$^{-1}$. As a first step to full 3D calculations the model is suitable for, we have explored the stellar orbital structure in the midplane of the Galaxy. We present the standard analysis in the pattern rotating frame, and complement this analysis with orbital information from the Galactic inertial frame. Prograde and retrograde orbits are defined unambiguously in the inertial frame, then labeled as such in the Poincar\\'e diagrams of the non-inertial frame. In this manner we found a sharp separatrix between the two classes of orbits. Chaos is restricted to the prograde orbits, and its onset occurs for the higher spiral perturbation considered plausible in our Galaxy. An unrealistically high spiral perturbation tends to destroy the separatrix and make chaos pervasive. This may be relevant in other spiral galaxies. ",
        "introduction": "Modeling of spiral galaxies with sophisticated computational techniques has become the usual way to study systems of this nature. One of the important structures, which is in fact the one that gives the name to this type of galaxies, is the spiral pattern. In the spiral density wave theory (Lin \\& Shu 1964), the spiral structure of galaxies was modeled as a periodic perturbation term to the axisymmetric potential in the disk plane. This is known as the {\\it tight-winding} or WKB approximation (e.g., Binney \\& Tremaine 1994) for small pitch angles. In the case of a two-armed spiral pattern it gives a potential in the galactic plane of the form:  \\be \\Phi_{s} (R,\\varphi) =  f(R)cos [ 2\\varphi+g(R) ]; \\label{eq_1} \\en  The function $f(R)$ is the amplitude of the perturbation, $g(R)$ provides the geometry of the spiral pattern, and $R$, $\\varphi$ are cylindrical coordinates in the non-inertial reference frame of the arms (rotating with a given angular velocity).  All studies of spiral galaxies we know of, even in cases of large pitch angle in the spiral pattern, have used a spiral potential of the form in Eq. (1), e.g. Contopoulos \\& Grosb{\\o}l (1986, 1988; hereafter C\\&G86 and C\\&G88); Patsis, Contopoulos, \\& Grosb{\\o}l (1991, hereafter PC\\&G), and in particular in our Galaxy the models of Amaral \\& L\\'epine (1997, hereafter A\\&L) and L\\'epine, Mishurov, \\& Dedikov (2001). Self-consistency of the proposed spiral pattern has been analyzed by C\\&G86, C\\&G88, PC\\&G, and A\\&L.  The dependence of the spiral potential on $z$ (perpendicular distance to the galactic plane) has been accounted for by Patsis \\& Grosb{\\o}l (1996) as a  $sech^2(z/z_s )$ factor of a function of the form in Eq. (1), with $z_s$ a scaleheight. Martos \\& Cox (1998), in numerical MHD simulations, considered an exponential z-factor of an approximate local spiral potential in the galactic plane.  In barred galaxies the approach is analogous to that given above for spiral galaxies: the usual approximation for the potential in the galactic plane due to the bar is a function of the form $\\Phi_{b} (R,\\varphi) = f(R)cos ( 2\\varphi)$. Instead of taking an ad hoc dependence on the z coordinate, an alternative way to consider the extension to a 3D-bar potential is to begin directly with a 3D mass distribution representing the bar. This method has been considered by Athanassoula et al. (1983) and Pfenniger (1984). From a comparison on the galactic plane between their 3D-bar potential and a potential of the form $\\Phi_{b} (R,\\varphi) = f(R)cos ( 2\\varphi)$, Athanassoula et al. (1983) found important differences in the corresponding force fields. However, the consequences of this result were not pursued.  In this paper, rather than using a simple ad hoc model for a 3D spiral perturbation, we consider a procedure whose essence is exactly the same as the modeling of a barred galaxy made by Athanassoula \\etal (1983): instead of using a spiral potential of the form given by Eq. (1), we propose a 3D mass distribution for the spiral arms and derive their gravitational potential and force fields from previously known results in potential theory. Grand design galaxies with a very prominent spiral structure in red light suggest to us that such structure should be considered an important galactic component and are worthy of a modeling effort beyond a simple perturbing term. This approach amounts to little more than admitting the possibility that there is no simple formula that fits the spiral perturbation at all $R$.  In our model we use Schmidt's (1956) analytical expression for the potential of an inhomogeneous oblate spheroid and model the spiral mass distribution as a series of such components settled along a spiral locus. The overlapping of spheroids allows a smooth distribution, resulting in a continuous function for the gravitational force. The basic parameters of the excess density distribution contributing to the spiral perturbation include a description of the spiral locus, the dimensions and density law of the spheroids, the central density in the spheroids as a function of galactocentric distance, the total mass of the spiral arms, and the angular velocity of the spiral pattern.  Our aim in this work is to make a preliminary study of stellar orbits in the Galactic plane $z$ = 0 in a potential resulting from the superposition of our 3D-spiral mass distribution and the axisymmetric Galactic mass distribution considered by Allen \\& Santill\\'an (1991, hereafter A\\&S). Also, we compare the potential and force fields produced by the 3D-spiral mass distribution with a tight-winding approximation in Eq. (1). The resulting differences may have important consequences on the stellar and gaseous dynamical behavior in a potential of this type. An expected difference in the force field is the effect of the self-gravitation of the mass of the spiral arms, which is not accounted for in a potential like that of Eq. (1).  Detailed orbital studies have been made in barred and spiral galaxies (e.g. Contopoulos 1983, Athanassoula et al. 1983, Pfenniger 1984, Teuben \\& Sanders 1985). In this work our analysis of stellar motion in the Galactic plane, under the proposed Galactic potential, follows the usual technique of Poincar\\'e diagrams.  However, we propose an alternative interpretation of Poincar\\'e diagrams which has not been previously considered. This interpretation is based on defining the orbital sense of motion (prograde or retrograde) in the Galactic inertial reference frame, joined to the usual definition in the non-inertial reference frame (e.g. Athanassoula et al.  1983) in which the spiral arms (and/or a bar) are at rest. This leads to Poincar\\'e diagrams (meaningful only in the non-inertial reference frame) revealing two sharply separated regions: one corresponding to prograde orbits and the other to retrograde orbits.  Our orbital analysis emphasizes the properties of the Galactic spiral arms for which some orbits may show stochastic behavior. These properties and the resulting stellar behavior should be applicable to similar types of galaxies.  The structure of this paper is the following: in Section 2 we present our Galactic model for the 3D spiral arms, with a discussion of the required parameters. In Section 3 we give the preliminary self-consistency tests that we have made of the proposed spiral arms, and establish a line of attack that must be followed to improve the model. In section 4 we make a comparison between the potential and force fields given by our model and those given by a tight-winding approximation. We show the importance of the self-gravitation of the spiral arms. In Section 5 we present an orbital analysis on the Galactic plane for differing spiral arms properties including the total mass in the spiral arms, the number of arms, and the angular velocity of the spiral pattern. In Section 5.1 we clarify the distinction between prograde and retrograde motion and the importance of the frame of reference to establish the essential difference between the two classes of orbits in Poincar\\'e diagrams through the zero angular momentum separatrix, a concept we introduce in this section. We show here that our definition provides a direct connection between sense of orbital motion and chaotic motion. In the same subsection, Poincar\\'e diagrams for a number of families labeled by their Jacobi integral $E_J$ are shown.  An estimation of the required strength of the spiral perturbation for which the nonlinear effects are important is given, and we discuss the range of parameters explored and those we deem plausible for our Galaxy. In Section 5.2 we investigate the onset of chaos using Lyapunov exponents, and the comparison of resonances for prograde and retrograde motion. In Section 6 we discuss our results and give some conclusions, including the possible response of the interstellar gas to the Galactic potential. ",
        "conclusions": "We present a 3D model for a spiral mass distribution, consisting of inhomogeneous oblate spheroids superposed along a given spiral locus. The model is applied in particular to our Galaxy, but can easily be applied to spiral galaxies in general. Furthermore, it allows to look with a deeper physical insight into details that are inaccessible to the classical treatment of the spiral perturbation, which models it as a simple periodic function.  Our model of oblate spheroids is physically simple and plausible, with continuous derivatives and density laws.  In our Galaxy, the model parameters, such as the number of spiral arms, its pitch angle, its radial extent, the pattern speed, the dimensions and mass density of the spheroids, and the total mass in the arms, were taken in a range of possibilities suggested by observations and theory.  In principle, the dimensions and mass density of the oblate spheroids will depend on the type of spiral arms which are modeled, gaseous or stellar. In this first work the adopted dimensions resemble those of gaseous spiral arms (Kennicutt \\& Hodge 1982), and a linear density law in the spheroids has been considered. We assembled Galactic models with two-armed spirals, such as the 15.5$^\\circ$ pitch-angle, stellar arms discussed by Drimmel (2000), and with $six$ spiral arms: adding the four 12$^\\circ$ pitch-angle, optical arms, delineated by luminous HII regions. From a range of possibilities, we considered three values of the pattern speed: ${\\Omega}_{p}$ = 15, 20, and 60 km s$^{-1}$ kpc$^{-1}$, and the ratio of the mass in the spiral arms to the disk's mass in the A\\&S axisymmetric Galactic model, $M_S/M_D$, in the range 0.0175 to 0.05 . In this range of masses the average force due to the spiral arms is between 5 and 10 $\\%$ of the background axisymmetric force.  In an effort to achieve a self-consistent model of the spiral perturbation in our Galaxy, we have used the well-known, approximate method of C\\&G86 to analyze the density response to this imposed perturbation. We have computed the density response in a Galactic potential with two spiral arms, taking the pattern speed as ${\\Omega}_{p}$ = 15 and 20 km s$^{-1}$ kpc$^{-1}$, and the mass ratio $M_S/M_D$ around the lower limit given above. Our nearly self-consistent models favor the pattern speed of 20 km s$^{-1}$ kpc$^{-1}$. However, this preliminary analysis must be improved at least accounting for (a) a hot stellar population around the central periodic orbits, (b) a four-armed, stellar, spiral pattern in the density response, in addition to the main two-armed component (A\\&L), and (c) a properly modeling of the dimensions of two-armed stellar spirals, as the K-band arms given by Drimmel (2000); this type of arms is azimuthally broad (Rix \\& Zaritsky 1995), thus an increase with galactocentric distance of the major semi-axis $a_0$ of spheroids would be appropriate. This analysis will be presented in a future work.  Modeling of the gravitational potential produced by a spiral perturbation has usually been based on the {\\it tight-winding approximation} (TWA, Eq. 1). We have compared the potential and force fields of a two-armed spiral perturbation given by our model with a TWA model. We found that the self-gravity of the spiral pattern (i.e. contributions to the potential from the entire pattern), which is not accounted for in the TWA (which acts more like a local approximation), cause the local spiral potential to adopt shapes that are not correctly fit by the simple perturbing term that has been traditionally invoked to represent the local spiral potential. This fact may have far-reaching consequences; for instance, in the gas response to the spiral perturbation. We have performed modest 1D MHD simulations (Franco, Martos \\& Pichardo 2001) with the code Zeus to show the differences in the gas response using the conventional model of a cosine for the potential and the model presented in this work. These simulations show that shocks do not leave the arm downstream as in previous calculations (Baker \\& Barker 1974, Martos \\& Cox 1998) for a plausible range of entry speeds. And, in correspondence with observational expectations, shocks seek the upstream edge of the arm, i.e. the concave side inside corotation marked in optical observations of galaxies for accumulations of dust in the inner part of the spiral arms. The inclusion of the magnetic field is essential to this effect.  In this manner, results based on the TWA should be revised: the gas response depends strongly on the position in the Galaxy. A potential ``well'' in the arm may disappear as such at a different segment of the arm.  In the analysis of Poincar\\'e diagrams we found it is quite fruitful to use an inertial frame to define the prograde or retrograde sense of orbital motion around the Galactic center along with the usual definition in the non-inertial system, where the Poincar\\'e diagrams are defined. In the inertial frame the sense of motion is preserved with time for almost every orbit in our experiments exception being orbits with nearly zero angular momentum. This property relies upon the parameters we consider plausible for our Galaxy. If we include information of the inertial system in the non-inertial one, Poincar\\'e diagrams reveal that prograde and retrograde orbits, as defined in the inertial frame, occupy sharply separated regions, through a separatrix corresponding, loosely, to nearly zero angular momentum orbits in this system.  The definition of sense of orbital motion in the inertial frame goes beyond a mere matter of semantics, for it has a simple physical meaning and it appears to be intimately connected to the onset of chaos. Based on an analysis of Poincar\\'e diagrams and the first Lyapunov exponent we find that, within plausible amplitudes and pitch angles of the spiral arms for a Galaxy such as the Milky Way (and independently of the number of arms chosen), if there is chaos, only prograde orbits can exhibit it, and for a sufficiently weak perturbation, as it seems to be the case in our Galaxy, the separatrix is a well-defined narrow curve. The onset and extension of chaotic subregions of the prograde region, depends on two main parameters that are the mass in the spiral arms or the relative force; and the angular velocity. We stress out the point that the standard definition in the rotating frame, which calls the $x' > 0$ of the diagram the retrograde side, and $x' < 0$ the prograde side (for a spiral pattern moving clockwise), would not have shed light onto the connection chaos-prograde motion, since the same orbit (ordered or chaotic) can occupy both sections of Poincar\\'e diagrams. The different behavior regarding the onset of chaos of prograde and retrograde orbits, as defined in the inertial system, could be attributed to the overlapping of resonances (Contopoulos 1967, Athanassoula 2001). Figures \\ref{resonancias1} and \\ref{resonancias2} show that the spacing of the main resonances is wider for retrograde orbits, and is even smaller if we take higher angular velocities for the spiral pattern. In cases with the lower spiral masses ($M_S/M_D \\la .03$), we do not find chaos for angular speeds of the spiral pattern lower than 20 km s$^{-1}$ kpc$^{-1}$. We have also computed some orbital families with $\\Omega_p = 60$ km s$^{-1}$ kpc$^{-1}$ since n-body models predict thoses velocities (that corresponds to the bar), and we find that, even for the lowest spiral mass we considered, chaos appears for some families (Fig. \\ref{6brazosOMP6}, $E_J = -2150 \\times 10^2$ km$^2$ s$^{-2}$), where almost all the prograde region is chaotic but chaos do not invade retrograde region. The inclusion of more than two spiral arms does not seem change dramatically the results (Figs. \\ref{MSD0.05} for two arms and \\ref{6brOM2J1630} for 6 arms). A minimum strength of the perturbation is required for the appearance of stochastic motion in the models with the lowest angular speeds (15 and 20 km s$^{-1}$ kpc$^{-1}$). We find that the amplitude of approximately 6$\\%$ (in average) of the axisymmetric radial force is required (that corresponds in our model to a $M_S/M_D \\la .05$ for a pitch angle of 15.5$^\\circ$). For cases of very strong spiral perturbations (relative forces higher than 15$\\%$) the separatrix is no longer a well-defined curve and chaos is pervasive. However we don't think this spiral forcing is proper for a Sb galaxy.  It is worth noticing that our results are valid in the plausible range of parameters (and even in unrealistic cases with maximum relative forces for the spiral arms up to 15$\\%$) in a galaxies similar to the Milky Way, with 2, 4 or 6 arms. However, our results will surely be altered by the influence of the Galactic bar. We are currently studying this effect."
    },
    "0208/astro-ph0208246_arXiv.txt": {
        "abstract": "here has been an unprecedented and continuing growth in the volume, quality, and complexity of astronomical data sets over the past few years, mainly through large digital sky surveys.  Virtual Observatory (VO) concept represents a scientific and technological framework needed to cope with this data flood. We review some of the applied statistics and computing challenges posed by the analysis of large and complex data sets expected in the VO-based research. The challenges are driven both by the size and the complexity of the data sets (billions of data vectors in parameter spaces of tens or hundreds of dimensions), by the heterogeneity of the data and measurement errors, the selection effects and censored data, and by the intrinsic clustering properties (functional form, topology) of the data distribution in the parameter space of observed attributes. Examples of scientific questions one may wish to address include: objective determination of the numbers of object classes present in the data, and the membership probabilities for each source; searches for unusual, rare, or even new types of objects and phenomena; discovery of physically interesting multivariate correlations which may be present in some of the clusters; etc. ",
        "introduction": "Observational astronomy is undergoing a paradigm shift.  This revolutionary change is driven by the enormous technological advances in telescopes and detectors (e.g., large digital arrays), the exponential increase in computing capabilities, and the fundamental changes in the observing strategies used to gather the data.  In the past, the usual mode of observational astronomy was that of a single astronomer or small group performing observations of a small number of objects (from single objects and up to some hundreds of objects). This is now changing: large digital sky surveys over a range of wavelengths, from radio to x-rays, from space and ground are becoming the dominant source of observational data.  Data-mining of the resulting digital sky archives is becoming a major venue of the observational astronomy.  The optimal use of the large ground-based telescopes and space observatories is now as a follow-up of sources selected from large sky surveys.  This trend is bound to continue, as the data volumes and data complexity increase.  The very nature of the observational astronomy is thus changing rapidly. See, e.g., Szalay \\& Gray (2001) for a review.  The existing surveys already contain many Terabytes of data, from which catalogs of many millions, or even billions of objects are extracted.  For each object, some tens or even hundred parameters are measured, most (but not all) with quantifiable errors. Forthcoming projects and sky surveys are expected to deliver data volumes measured in Petabytes.  For example, a major new area for exploration will be in the time domain, with a number of ongoing or forthcoming surveys aiming to map large portions of the sky in a repeated fashion, down to very faint flux levels. These synoptic surveys will be generating Petabytes of data, and they will open a whole new field of searches for variable astronomical objects.  This richness of information is hard to translate into a derived knowledge and physical understanding.  Questions abound: How do we explore datasets comprising hundreds of millions or billions of objects each with dozens of attributes? How do we objectively classify the detected sources to isolate subpopulations of astrophysical interest? How do we identify correlations and anomalies within the data sets? How do we use the data to constrain astrophysical interpretation, which often involve highly non-linear parametric functions derived from fields such as physical cosmology, stellar structure, or atomic physics? How do we match these complex data sets with equally complex numerical simulations, and how do we evaluate the performance of such models?  The key task is now to enable an efficient and complete scientific exploitation of these enormous data sets.  The problems we face are inherently statistical in nature.  Similar situations exist in many other fields of science and applied technology today.  This poses many technical and conceptual challenges, but it may lead to a whole new methodology of doing science in the information-rich era.  In order to cope with this data flood, the astronomical community started a grassroots initiative, the National (and ultimately Global) Virtual Observatory (NVO).  The NVO would federate numerous large digital sky archives, provide the information infrastructure and standards for ingestion of new data and surveys, and develop the computational and analysis tools with which to explore these vast data volumes.  Recognising the urgent need, the National Academy of Science Astronomy and Astrophysics Survey Committee, in its new decadal survey {\\em Astronomy and Astrophysics in the New Millennium} (McKee, Taylor, \\etal 2001) recommends, as a first priority, the establishment of a National Virtual Observatory (NVO).  The NVO would provide new opportunities for scientific discovery that were unimaginable just a few years ago.  Entirely new and unexpected scientific results of major significance will emerge from the combined use of the resulting datasets, science that would not be possible from such sets used singly.  In the words of a ``white paper'' \\footnote{ Available at http://www.arXiv.org/abs/astro-ph/0108115, and also published in Brunner, Djorgovski, \\& Szalay (2001), p.~353. } prepared by an interim steering group the NVO will serve as {\\em an engine of discovery for astronomy.}  Implementation of the NVO involves significant technical challenges on many fronts, and in particular the {\\em data analysis}. Whereas some of the NVO science would be done in the image (pixels) domain, and some in the interaction between the image and catalog domains, it is anticipated that much of the science (at least initially) will be done purely in the catalog domain of individual or federated sky surveys.  A typical data set may be a catalog of $\\sim 10^8 - 10^9$ sources with $\\sim 10^2$ measured attributes each, i.e., a set of $\\sim 10^9$ data vectors in a $\\sim 100$-dimensional parameter space.  Dealing with the analysis of such data sets is obviously an inherently multivariate statistical problem.  Complications abound: parameter correlations will exist; observational limits (selection effects) will generally have a complex geometry; for some of the sources some of the measured parameters may be only upper or lower limits; the measurement errors may vary widely; some of the parameters will be continuous, and some discrete, or even without a well-defined metric; etc.  In other words, analysis of the NVO data sets will present many challenging problems for multivariate statistics, and the resulting astronomical conclusions will be strongly affected by the correct application of statistical tools.  We review some important statistical challenges raised by the NVO. These include the classification and extraction of desired subpopulations, understanding the relationships between observed properties within these subpopulations, and linking the astronomical data to astrophysical models. This may require a generation of new methods in data mining, multivariate clustering and analysis, nonparametric and semiparametric estimation and model and hypothesis testing. ",
        "conclusions": "Given this computational and statistical complexity, blind applications of the commonly used (commercial or home-brewed) clustering algorithms could produce some seriously misleading or simply wrong results.  The clustering methodology must be robust enough to cope with these problems, and the outcome of the analysis must have a solid statistical foundation.  In our experience, design and application of clustering algorithms must involve close, working collaboration between astronomers and computer scientists and statisticians.  There are too many unspoken assumptions, historical background knowledge specific to the given discipline, and opaque jargon; constant communication and interchange of ideas are essential.  The entire issue of discovery and interpretation of multivariate correlations in these massive data sets has not really been addressed so far.  Such correlations may contain essential clues about the physics and the origins of various types of astronomical objects.  Effective and powerful data visualization, applied in the parameter space itself, is another essential part of the interactive clustering analysis. Good visualisation tools are also critical for the interpretation of results, especially in an iterative environment.  While clustering algorithms can assist in the partitioning of the data space, and can draw the attention to anomalous objects, ultimately a scientist guides the experiment and draws the conclusions. It is very hard for a human mind to really visualise clustering or correlations in more than a few dimensions, and yet both interesting clusters and multivariate correlations with statistical dimensionality $> 10$ or even higher are likely to exits, and possibly lead to some crucial new astrophysical insights.  Perhaps the right approach would be to have a good visualisation embedded as a part of an interactive and iterative clustering analysis.  Another key issue is interoperability and reusability of algorithms and models in a wide variety of problems posed by a rich data environment such as federated digital sky surveys in a VO.  Implementation of clustering analysis algorithms must be done with this in mind.  Finally, scientific verification and evaluation, testing, and follow-up on any of the newly discovered classes of objects, physical clusters discovered by these methods, and other astrophysical analysis of the results is essential in order to demonstrate the actual usefulness of these techniques for a VO or other applications.  Clustering analysis can be seen as a prelude to the more traditional type of astronomical studies, as a way of selecting of interesting objects of samples, and hopefully it can lead to advances in statistics and applied computer science as well."
    },
    "0208/astro-ph0208070_arXiv.txt": {
        "abstract": "We are developing photon-counting cameras employing cryogenic arrays of energy-resolving TES (Transition Edge Sensor) pixels. These are being tested in ground-based instruments, but will have their greatest impact when employed on space platforms, where they can cover the 10$\\mu$m-100nm range with high time- and moderate energy- resolution. Here we summarize briefly existing device performance, current directions in array camera development and anticipated capabilities. ",
        "introduction": "Cryogenic energy-resolving photon detectors show considerable promise for advanced instrumentation in several wavebands. In the near-IR through UV, these devices provide noise-free photon counting with $\\le \\mu$s time resolution at $\\delta E \\approx 0.15$eV energy resolution. Early work focused on superconducting tunnel junction devices (Perryman, Foden \\& Peacock 1993; Peacock, these proc.), but alternative cryogenic technologies show great promise. Our group has pursued the application of superconducting TES devices to this problem (Cabrera et al. 1998). Our detectors are pixel arrays of 40nm thick tungsten (W) films patterned on Si. When cooled below their $\\sim 100$mK transition temperature and voltage biased to produce negative electro-thermal feedback (Irwin 1995), an absorbed photon decreases the Joule-heating current. This pulse signal is read out with a DC SQUID array (Welty \\& Martinis 1993). Bi-layer thin films, a variety of substrates, and multiplex schemes to read out larger arrays are also under development and TES array technology is finding interesting application from the sub-mm (SCUBA-2) to the X-ray (Con-X) regimes.  Miller et al. (2000) summarize performance of present W TES devices. We routinely achieve $\\delta E\\sim 0.15$eV at $\\sim 3$eV, absolute GPS photon times to 300ns and single-pixel count rates of $\\sim 30$kHz. Energy discrimination rejects all cosmic rays and allows no `dark current' above the noise floor. The intrinsic bare W quantum efficiency is $\\sim 50$\\% and in astronomical systems we have achieved efficiencies on the sky as high as $\\sim$20\\%.  TES resolution is given by $\\Delta E_{FWHM} = 2.355 [ 4k_BT^2C(n/2)^{1/2}/\\alpha ]^{1/2}$, where $T$ and $C$ are the temperature and heat capacity of the W $e^-$ system, $n=5$ for electron-phonon conduction and $\\alpha = (d {\\rm ln}R/d{\\rm ln}T)_ {V=const}$. The pixel size sets $C$; the maximum (saturation) energy is $E_{max} \\approx CT_c/\\alpha$ and therefore the optimum detector design gives $\\Delta E_{FWHM,min}=2.355[4k_BT(n/2)^{1/2}E_{max} ]^{1/2}$.  Thus for our typical 20$\\mu$m pixel and $E_{max} = 10$eV we obtain $\\Delta E_{FWHM,min} \\approx 0.05$eV. Somewhat higher resolution can be obtained with lower $E_{max}$ by operating in the saturation regime, measuring pulse shape rather than peak height.  In the present devices on Si, 58\\% of the photon energy is lost from the W $e^-$ system to substrate phonons, giving an expected $\\Delta E =$0.088eV; for comparison the measured resolution is 0.12eV FHWM.  Interestingly, although the physics is different, these sums predict $\\Delta E_{FWHM,min}$ quite similar to theoretical limits and goal resolutions for existing STJ devices.  The eventual choice of detector is probably best decided on the basis of achieved performance and ease of manufacture.  At low $E$ the thermodynamic fluctuation noise floor allows detection to $\\sim 10 \\mu$m.  Of course, this full sensitivity range cannot be exploited from the ground, and we have employed a variety of filtering schemes to suppress unacceptably high thermal count rates beyond $2\\mu$m.  To illustrate the astronomical utility of these detectors, we have packaged TES arrays into simple demonstration systems, based on a He dilution refrigerator used for device development, and performed some basic observations. To filter the IR background, we employ a $\\sim 3$m length of high-OH optical fiber spooled at 4K in the dewar. The OH bands of this cold filter provided effective blocking at $\\sim 1.7\\mu$m, but allow transmission in the atmospheric windows. At the telescope, the atmosphere and optics limit us to $\\sim 3.5$eV, but in the lab we detected $E>7$eV photons. With such systems we demonstrated the first astronomical spectra taken with a cryogenic optical detector % and using small telescopes have made unique, scientifically valuable, measurements of the Crab pulsar and other astronomical objects (Romani et al. 1999, Romani et al. 2001), including photon counting spectro-photopolarimetry in the optical/IR.  On small telescopes these test set-ups have achieved good efficiency, although system efficiencies of $\\le 10$\\% are typical. As we are fiber-coupled we have to date only employed a small subset of the TES array (e.g. 2$\\times$3 out of $6\\times6$ for the McDonald 2.7m Crab observations). Also, for ease of fabrication, the present TES arrays only employ a single lithographic layer of wiring; these wiring leads are responsive to photons, which complicates the energy PSF and spectral calibration. Finally, fiber coupled schemes, while attractive for test purposes, do not match well to the eventual (space platform) imaging applications for which these TES devices should have their largest impact. ",
        "conclusions": ""
    },
    "0208/nucl-th0208035_arXiv.txt": {
        "abstract": "We calculate a $\\Lambda\\Lambda$ pairing gap in binary mixed matter of nucleons and $\\Lambda$ hyperons within the relativistic Hartree-Bogoliubov model.  Lambda hyperons to be paired up are immersed in background nucleons in a normal state. The gap is calculated with a one-boson-exchange interaction obtained from a relativistic Lagrangian.  It is found that at background density $\\rho_{N}=2.5\\rho_{0}$ the $\\Lambda\\Lambda$ pairing gap is very small, and that denser background makes it rapidly suppressed. This result suggests a mechanism, specific to mixed matter dealt with relativistic models, of its dependence on the nucleon density. An effect of weaker $\\Lambda\\Lambda$ attraction on the gap is also examined in connection with revised information of the $\\Lambda\\Lambda$ interaction. ",
        "introduction": "\\label{sec:intro}  Pairing correlation in hadronic matter has been attracting attention due to close relationship between properties of neutron stars and its interior superfluidity. Superfluidity inside neutron stars affects, for instance, heat capacity and neutrino emissivity. These quantities relate to the cooling processes of neutron stars.  In neutron stars, several types of baryon pairing appear. It is strongly believed that neutrons form the $^{1}S_{0}$ pair in the inner crust region~\\cite{takatsuka93:_super,wambach93:_quasip,chen93:_pairin}. At the corresponding density $10^{-3}\\rho_{0} \\lesssim \\rho_{B} \\lesssim 0.7\\rho_{0}$, where $\\rho_{0}$ is the saturation density of symmetric nuclear matter, the $^{1}S_{0}$ partial wave of the nucleon-nucleon (\\textit{NN}) interaction is attractive: In infinite matter an attraction, no matter how weak it is, brings about the BCS instability to the ground state. This type of pairing has been most extensively studied for decades using various models. Also important is the $^{3}P_{2}$ neutron pairing in the outer core region up to $\\rho_{B} \\sim 2\\rho_{0}$. The $^{3}P_{2}$ partial wave of the \\textit{NN} interaction is attractive enough there for neutrons to be in a superfluid state~\\cite{takatsuka93:_super,elgaroey96:_tripl}. On the other hand, the $^{1}S_{0}$ partial wave would become repulsive there so that the $^{1}S_{0}$ neutron pairs would disappear. Instead, the $^{1}S_{0}$ proton pairing is expected to be realized owing to its small fraction~\\cite{takatsuka93:_super,chen93:_pairin}.  In the inner core region, baryon density becomes much larger ($\\rho_{B} \\gtrsim 2\\rho_{0}$) and various hyperons may appear~\\cite{glendenning00:_compac_stars}. Some are expected to form pairs in the same way as the \\textit{NN} pairing owing to the attractive $^{1}S_{0}$ partial wave of the hyperon-hyperon (\\textit{YY}) interaction. Moreover, interspecies pairing such as $\\Lambda$-neutron pairing may be realized at the total baryon densities higher than $\\rho_{B} \\gtrsim 4\\rho_{0}$ where fractions of the two kinds of baryon are expected to be comparable. These kinds of pairings affect the properties of neutron stars through, say, suppression of the hyperon direct URCA processes. Whether $\\Lambda$ hyperons are in a super state or not plays a decisive role for the microscopic understanding of neutron stars: $\\Lambda$ hyperons in a normal state would lead to too rapid cooling of them and force one to modify the cooling scenarios. Conversely, one can extract information on baryonic force and inner structure of neutron stars from these phenomena.  Studying neutron stars thus is the driving forces for the study on baryon superfluidity.  Despite the situations, magnitude of the hyperon pairing gaps is still uncertain. More studies are needed exploiting available information from various sources such as the hypernuclear spectroscopy, direct observation of neutron stars, and so on.  Our aim of this study is twofold. One is to explore an effect of Dirac effective mass of $\\Lambda$ hyperons on the $\\Lambda\\Lambda$ pairing correlations in binary mixed matter composed of $\\Lambda$ hyperons and nucleons. In this respect, recognizing the significance of covariant representation led to the remarkable developments in nuclear/hadron physics for the last three decades.  As is well known nowadays, cancellation between large Lorentz scalar and vector fields provides a proper saturation mechanism of nuclear matter on the whole. Typical examples are the phenomenological relativistic mean field (RMF) model and the microscopic Dirac-Brueckner-Hartree-Fock (DBHF) approach. Especially in the latter, selfconsistent treatment of a nucleon spinor with a bare \\textit{NN} interaction brings the saturation points predicted by the nonrelativistic BHF approach towards the empirical one by a repulsive relativistic effect. This selfconsistency is the key ingredient of the relativistic models. Then, we would like to ask a following question: What does the selfconsistency bring to superfluidity in the composite hadronic matter?  This is an important issue on studying neutron star matter that has complex composition of baryons using relativistic models.  The other is to investigate an impact of the recent experimental finding on the $\\Lambda\\Lambda$ pairing. The KEK-PS experiment E373, especially the ``NAGARA'' event~\\cite{takahashi01:_obser_lambd_lambd_he_doubl_hyper} may explode the ``old'' information of $\\Lambda\\Lambda$ interaction which has ruled hypernuclear systems for three decades. The event unambiguously determined the binding energy of the two $\\Lambda$ hyperons $B_{\\Lambda\\Lambda}$ in $\\substack{\\phantom{\\Lambda}6 \\\\ \\Lambda\\Lambda}$He. Most importantly, it suggests that the $\\Lambda\\Lambda$ interaction is weaker than it was thought before. If this is confirmed, the new information ought to have a significant impact on the microscopic understanding of the properties of neutron stars.  Unlike the \\textit{NN} pairing, there are only a few studies on the hyperon pairing. It was first studied in a nonrelativistic framework by~\\citeauthor*{balberg98:_s_lambd}~\\cite{balberg98:_s_lambd}. Then their results were applied to the study on cooling of neutron stars~\\cite{schaab98:_implic_of_hyper_pairin_for}.  They obtained the $\\Lambda\\Lambda$ pairing gap in symmetric nuclear matter using an interaction based on the \\textit{G}-matrix in symmetric nuclear matter and an approximation of nonrelativistic effective mass obtained from single-particle energies with first order Hartree-Fock corrections, though their motive was application to the physics of neutron stars. Their conclusion was that the maximal pairing gap became larger as the background density increased; at the same time, the effective mass of $\\Lambda$ hyperons became smaller. Since smaller effective mass generally leads to smaller pairing gap, this conclusion is against general expectations. \\citeauthor*{takatsuka99:_super_lambd_hyper_admix_neutr_star_cores} subsequently studied the problem using two types of bare $\\Lambda\\Lambda$ interactions and of hyperon core models~\\cite{takatsuka99:_super_lambd_hyper_admix_neutr_star_cores}. Aiming at a better approximation of neutron star matter, they used the nonrelativistic effective mass which was obtained from the \\textit{G}-matrix calculation for composite matter of neutrons and $\\Lambda$ hyperons, and was dependent on a total baryon density and a $\\Lambda$ fraction. Their gaps were somewhat smaller than~\\citeauthor{balberg98:_s_lambd}'s due to smaller effective mass and appropriate choice of the interaction. They also showed that the result had considerable dependence on the interactions and the hyperon core models owing to related uncertainties. An important thing in common to these past studies is use of the $\\Lambda\\Lambda$ interactions that are too attractive considering the consequence of the NAGARA event, which was unavailable at that time.  Therefore, we study the $^{1}S_{0}$ $\\Lambda\\Lambda$ pairing in binary mixed matter of nucleons and $\\Lambda$ hyperons using relativistic interactions that reflects the new experimental information for the first time. The $\\Lambda$ hyperons are immersed in pure neutron matter or symmetric nuclear matter that is treated as a background. We use the relativistic Hartree-Bogoliubov (RHB) model in which density-dependence of the interaction is automatically taken into account via the Lorentz structure. The density-dependence that is an inherent mechanism in relativistic models may lead to novel behavior of the pairing gap: Since pairs are formed in medium, medium effects on a particle-particle (\\mbox{p-p}) channel interaction should be considered. In the RHB model, bare baryon masses are reduced by the scalar mean field. This decreased mass is the Dirac effective mass~\\cite{jaminon89:_effec,glendenning00:_compac_stars}. The mass decrease may change the pairing gap to some extent in comparison with that obtained with the bare masses.  Although the two preceding studies also introduced the medium effects, each had a purely nonrelativistic origin. It has nothing to do with the Lorentz structure and the Dirac effective mass. We thus intend to compare with the results of the first study by~\\citeauthor{balberg98:_s_lambd} neglecting, for the time being, complexity of $\\Lambda$-$\\Sigma^{0}$ mixing that probably occurs in asymmetric nuclear matter; this mixing will be discussed in Sec.~\\ref{sec:lambda-sigma-mixing}. Besides, other constituents predicted to exist in neutron stars and equilibration like chemical equilibrium of neutron star matter are ignored so that we narrow down arguments to the impact of the revision on the $\\Lambda\\Lambda$ pairing properties. Such a plain treatment should be taken as the very first step of our study on the hyperon pairing with the recently revised interactions in neutron star matter.  This paper is organized as follows: In Sec.~\\ref{sec:model}, we illustrate the Lagrangian of the system and the gap equation for the $^{1}S_{0}$ $\\Lambda\\Lambda$ pairing.  In Sec.~\\ref{sec:results-discussions}, we present results of the $\\Lambda\\Lambda$ pairing properties in the binary hadronic matter. Section~\\ref{sec:summ} contains a summary. ",
        "conclusions": "\\label{sec:results-discussions}  \\subsection{Effect of Dirac effective mass decrease} \\label{sec:effect-dirac-mass}  \\begin{figure}[tbp] \\centering \\includegraphics[bb=0 270 592 742,width=9cm,keepaspectratio,clip]{fig1.eps} \\caption{$\\Lambda\\Lambda$ pairing gap at the Fermi surface of $\\Lambda$ hyperons, for pure neutron background densities $\\rho_{N}=0$, $\\rho_{0}$, $2.5\\rho_{0}$, and $5\\rho_{0}$. The coupling ratio $\\alpha_{\\sigma^{\\ast}}=0.5$ is used.} \\label{fig:neutgap} \\end{figure}  Figure~\\ref{fig:neutgap} shows the resulting $^{1}S_{0}$ $\\Lambda\\Lambda$ pairing gap at the Fermi surface in pure neutron matter of densities $\\rho_{N}$ at $0$, $\\rho_{0}$, $2.5\\rho_{0}$, and $5\\rho_{0}$, with $\\alpha_{\\sigma^{\\ast}}=0.5$ chosen. This value of the coupling ratio can reproduce the bond energy, Eq.~\\eqref{eq:bond-energy}, of about $1$ MeV in the RMF model, which is suggested by the NAGARA event. Contrary to the results obtained by~\\citeauthor*{balberg98:_s_lambd}, the $\\Lambda\\Lambda$ pairing gap becomes suppressed as the neutron density increases. At $\\rho_{N}=2.5\\rho_{0}$, where $\\Lambda$ hyperons already appear in some models of neutron stars~\\cite{schaffner96:_hyper}, the maximal pairing gap is about $0.15$ MeV. Since there are probably no $\\Lambda$ hyperons at $\\rho_{N}=0$ and $\\rho_{0}$ in neutron star matter, the pairing gaps at these densities are quite hypothetical.  Figure~\\ref{fig:neuteffmass} shows the density dependence of baryon effective masses, $M_{N}^{\\ast}$ and $M_{\\Lambda}^{\\ast}$, as functions of the total baryon density $\\rho_{B}$. The background neutron densities are fixed here, so that variations in $\\rho_{B}$ correspond to those in the Fermi momentum of $\\Lambda$ hyperons. Since we ignore the chemical equilibrium here, curves of the effective masses have discontinuous jumps; each piece corresponds to the fixed neutron densities, $\\rho_{N}= 0$, $\\rho_{0}$, $2.5\\rho_{0}$, and $5.0\\rho_{0}$. Consideration of the chemical equilibrium should connect them with each other. Nevertheless, we obtain the values qualitatively similar to the ones shown in, for example, Fig.~4 of Ref.~\\cite{schaffner96:_hyper}. It is therefore concluded that in-medium property of the phenomenological $\\Lambda\\Lambda$ interaction used in this study is justifiable. The effective mass of neutrons decreases steeply as the total baryon density increases, while mildly does the effective mass of $\\Lambda$ hyperons due to the weaker coupling of $\\Lambda$ hyperons to the scalar bosons than the coupling of nucleons.  \\begin{figure}[tbp] \\centering \\includegraphics[bb=0 265 592 700,width=9cm,keepaspectratio,clip]{fig2.eps} \\caption{Effective masses of $\\Lambda$ hyperons and neutrons for pure neutron background densities $\\rho_{N}=0$, $\\rho_{0}$, $2.5\\rho_{0}$, and $5\\rho_{0}$. The coupling ratio $\\alpha_{\\sigma^{\\ast}}=0.5$ is used.} \\label{fig:neuteffmass} \\end{figure}  For the \\mbox{p-p} channel, we use the RMF interaction as stated above. The interaction contains the Dirac effective mass of $\\Lambda$ hyperons, Eq.~\\eqref{eq:lefmeq}, through which the medium effects are introduced; the coupling of $\\Lambda$ hyperons to $\\sigma$ bosons, to which nucleons also couple, brings about the dependence on the background density.  Figure~\\ref{fig:mexint} represents the $\\Lambda\\Lambda$ RMF interaction derived from the parameter set HS-m2. It is shown that increasing the background neutron density suppresses attractive contribution from low momenta. This is the main reason why the $\\Lambda\\Lambda$ pairing gap is smaller in denser background. \\begin{figure}[tbp] \\centering \\includegraphics[bb=0 270 592 742,width=9cm,keepaspectratio,clip]{fig3.eps} \\caption{$\\Lambda\\Lambda$ RMF interaction $\\bar{v}(M_{\\Lambda}^{\\ast};\\, k,k_\\mathrm{F})$ at the Fermi momentum of $\\Lambda$ hyperons $k_\\mathrm{F}=1.0$ fm$^{-1}$, for pure neutron background densities $\\rho_{N}=0$, $\\rho_{0}$, $2.5\\rho_{0}$, and $5\\rho_{0}$, corresponding to $M_{\\Lambda}^{\\ast}=1068$, $813$, $660$, and $605$ MeV, respectively. The coupling ratio $\\alpha_{\\sigma^{\\ast}}=0.5$ is used. The legend is the same as in Fig.~\\ref{fig:neutgap}. The inset shows a magnification of the region around the repulsive bumps.} \\label{fig:mexint} \\end{figure} This new mechanism of the suppression is inherent in relativistic models which respect the Lorentz structure as shown in Eq.~\\eqref{eq:lefmeq}. It is shown that the decrease of the effective baryon mass plays an indispensable role when it is used selfconsistently in the baryon spinor.  What is important is that relativistic models naturally lead to a density-dependent interaction through a selfconsistent baryon spinor where the bare mass in a free spinor is replaced with the Dirac effective mass. An apt example is the saturation of symmetric nuclear matter in the DBHF approach~\\cite{machleidt89:_meson}. Requirement of the selfconsistency for the nucleon spinor, that is, use of the Dirac effective mass in the nucleon spinor effectively gives repulsion to the binding energy of symmetric nuclear matter. Consequently, it pushes the saturation points predicted by nonrelativistic models toward the empirical one. It seems that our finding is similar to this repulsive effect. Furthermore, the mechanism is apparently not restricted to $\\Lambda\\Lambda$ pairs. It is probable that other kinds of \\textit{YY} pairs have the same trend.   \\subsection{Effect of NAGARA event} \\label{sec:effect-nagara}  \\begin{figure}[tbp] \\centering \\includegraphics[bb=132 493 452 675,width=8cm,keepaspectratio,clip]{fig4.eps} \\caption{Maximal $\\Lambda\\Lambda$ pairing gap as a function of the strength of $\\Lambda\\Lambda$ attraction and the background density in pure neutron matter.} \\label{fig:3dim} \\end{figure}  Next we explore the effect of the NAGARA event on the $\\Lambda\\Lambda$ pairing.  With relation to the revised information on the $\\Lambda\\Lambda$ interaction, we vary the ratio $\\alpha_{\\sigma^{\\ast}}=g_{\\sigma^{\\ast} \\Lambda} / g_{\\sigma N}$ between 0.4 and 0.6, referring to Fig.~1 of Ref.~\\cite{marcos98:_bindin_lambd}: Thereby we control the attractive component of the interaction. Figure~\\ref{fig:3dim} represents the maximal $\\Lambda\\Lambda$ pairing gap at the Fermi surface of $\\Lambda$ hyperons as a function of the strength of $\\Lambda\\Lambda$ attraction and the background density of pure neutron matter. From this figure as well as Fig.~\\ref{fig:neutgap}, one reads that the suppression of the gap occurs in denser background of neutrons. Moreover, it may even vanish in the end (though the result depends on the choice of RMF parameter sets and a cutoff mass as will be shown later).  This varying $\\alpha_{\\sigma^{\\ast}}$ reveals likely closing of the gap at smaller $\\alpha_{\\sigma^{\\ast}}$ (\\textit{i.e.} weaker $\\Lambda\\Lambda$ attraction) and its strong suppression in the denser neutron background. This result implies that the aforementioned mechanism acts in concert with the weakened attraction for closing the gap. Hence, the $\\Lambda\\Lambda$ pairing correlation in dense pure neutron matter becomes less likely than before. The result is almost the same with the background of symmetric nuclear matter.  In the light of the neutron star cooling, the absence of the $\\Lambda\\Lambda$ pairing might call for the pairing of other hyperonic species and a modification of its scenarios. More realistic approximation of the internal composition of neutron stars needs a condition of chemical equilibrium which plays a decisive role. Under the condition, other hyperons will emerge as the background density increases. \\citeauthor{takatsuka01:_possib_hyper_super_neutr_star_cores}  studied the $\\Sigma^{-}\\Sigma^{-}$ and $\\Xi^{-}\\Xi^{-}$ pairings and shown their possibility~\\cite{takatsuka01:_possib_hyper_super_neutr_star_cores}. Nevertheless, the possibility of the $\\Lambda\\Lambda$ pairing in neutron star matter with the concerted mechanism in this model stands unsettled due to complex composition of particles inside neutron stars.   \\subsection{Comparison with nonrelativistic study} \\label{sec:comp-with-nonr}  \\begin{figure}[tbp] \\centering \\includegraphics[bb=0 270 592 742,width=9cm,keepaspectratio,clip]{fig5.eps} \\caption{$\\Lambda\\Lambda$ pairing gap at the Fermi surface of $\\Lambda$ hyperons, for nucleon background densities $\\rho_{N}=0$, $\\rho_{0}$, $2.5\\rho_{0}$, and $5\\rho_{0}$. The coupling ratio $\\alpha_{\\sigma^{\\ast}}=0.5$ is used.} \\label{fig:nuclgap} \\end{figure}  Now we make a comparison between relativistic and nonrelativistic predictions. For the comparison with the nonrelativistic results of~\\citeauthor*{balberg98:_s_lambd}~\\cite{balberg98:_s_lambd}, we calculate the $\\Lambda\\Lambda$ pairing gap in symmetric nuclear matter. Figure~\\ref{fig:nuclgap} represents our result. Slightly smaller gap than that obtained from the calculation of pure neutron matter (Fig.~\\ref{fig:neutgap}) reflects the smaller Dirac effective mass of $\\Lambda$ hyperons in symmetric nuclear matter than that in pure neutron matter.  We would like to note two remarkable differences between their result and ours. One difference is the dependence of the gap on the background density. Strikingly, ours is opposite to theirs (\\textit{cf.} Fig.~4 of Ref.~\\cite{balberg98:_s_lambd}). This is brought about \\emph{directly} and \\emph{indirectly} by decrease of the Dirac effective mass. We intend by the word `directly' that we can grasp the decrease of the gap through an expression in the weak-coupling approximation, \\begin{equation} \\label{eq:weakcpl} \\Delta(k_\\mathrm{F}) \\propto \\exp\\left[{}-\\frac{1}{N(k_\\mathrm{F})\\, |\\bar{v}(k_\\mathrm{F},k_\\mathrm{F})|}\\right], \\end{equation} where $N(k_\\mathrm{F}) = E_{k_\\mathrm{F}}^{(\\Lambda)} k_\\mathrm{F} / 2 \\pi^{2} \\hbar^{2}$ is the density of states at the Fermi surface. Equation~\\eqref{eq:weakcpl} shows that the smaller the Dirac effective mass becomes, the smaller the density of states does, which makes the gap smaller. Note that we use the approximation for rough estimation here and the full integration of the gap equation~\\eqref{eq:gapeq} is done throughout the present study. Meantime, we intend by the word `indirectly' that the gap decreases as the density increases due to gradual weakening of the attraction in the \\mbox{p-p} interaction which is shown in Fig.~\\ref{fig:mexint}. The other difference is the region of the Fermi momentum of $\\Lambda$ hyperons where the gaps are open. While the regions in their result are similar in all densities presented, Fig.~\\ref{fig:nuclgap} shows that the regions in our result narrow as the background density increases.  Finally, the result at $\\rho_{N}=\\rho_{0}$ may have relevance to the $\\Lambda\\Lambda$ pairing correlation around the center of hypernuclei~\\cite{takatsuka01:_hyper_super_neutr_star_cores}: We obtain the maximal gap $\\Delta (k_\\mathrm{F}=0.9 \\mathrm{~fm}^{-1}) \\simeq 0.5$ MeV.  Prior to the present study, \\citeauthor{elgaroey96:_super} studied~\\cite{elgaroey96:_super} relativistic effects on the neutron and proton pairing in neutron star matter and made a comparison with a nonrelativistic result~\\cite{elgaroey96:_model_s_bonn}. Their result shows a large effect of ``minimal relativity''~\\cite{brown69:_nucleon_nucleon_poten_and_minim_relat} on the $^{3}P_{2}$ neutron pairing while a small one on the $^{1}S_{0}$ proton pairing. They explained that using DBHF single-particle energies and factors of the minimal relativity are the causes of much smaller neutron pairing gap. As for our model, the factor corresponding to the minimal relativity is already included in the \\mbox{p-p} interaction owing to the normalization of the Dirac spinor, $u^{\\dagger}u=1$.   \\subsection{Choice of form factor} \\label{sec:choice-form-factor}  Also noteworthy is a form factor: In this subsection, we investigate a dependence of the gap on the cutoff mass for each type of the form factor. We use the purely phenomenological form factor at each $\\Lambda$ hyperon-meson vertex to regulate the high-momentum components of the \\mbox{p-p} interaction as in Ref.~\\cite{matsuzaki01:_phenom}. We have chosen the Bonn-type form factor, Eq.~\\eqref{eq:bonnff}, with the cutoff mass $\\Lambda_\\mathrm{c}=7.26$ fm$^{-1}$ thus far in this paper.  In contrast to the \\textit{NN} pairing, there has yet been no proper guide to determine the cutoff mass in the form factors for the $\\Lambda\\Lambda$ pairing. We hence resort to borrow the value from our previous study of the \\textit{NN} pairing~\\cite{matsuzaki01:_phenom}.  However, the type of the form factor and the value of the cutoff mass significantly affect the magnitude of the pairing gap. We therefore calculate the dependence of the $\\Lambda\\Lambda$ pairing gap at the Fermi surface on the cutoff mass $\\Lambda_\\mathrm{c}$ in the form factors of Bonn-type, Eq.~\\eqref{eq:bonnff}. The cutoff mass is taken to be larger than $5$ fm$^{-1}$, which roughly corresponds to mass of the heaviest meson employed (namely $\\phi$), otherwise the interaction is unphysical. The result is shown in Fig.~\\ref{fig:cutoff-dep}, in which the Fermi momenta of $\\Lambda$ hyperons are fixed to $k_\\mathrm{F}=0.90$, $0.80$, and $0.75$ fm$^{-1}$ for background density of pure neutron matter $\\rho_{N}=\\rho_{0}$, $2.5\\rho_{0}$, and $5.0\\rho_{0}$, respectively.  As expected, varying the cutoff mass changes the gaps steeply since it changes the balance of the attraction and the repulsion of the interaction. The peaks around $\\Lambda_\\mathrm{c} \\sim 5$ fm$^{-1}$ are due to consecutive suppression of the attraction ($\\sigma^{\\ast}$ boson) and the repulsion ($\\phi$ meson) by the form factor. Nonetheless, the importance of this result lies in the fact that the gaps become smaller in denser background for \\emph{any} cutoff mass. Thus the arbitrariness does not alter our conclusions.  \\begin{figure}[tbp] \\centering \\includegraphics[bb=124 388 420 674,width=7.5cm,keepaspectratio,clip]{fig6.eps} \\caption{Cutoff mass dependence of the $\\Lambda\\Lambda$ pairing gap at the Fermi surface in pure neutron matter. The coupling ratio $\\alpha_{\\sigma^{\\ast}}=0.5$ is used.} \\label{fig:cutoff-dep} \\end{figure}  On the other hand, a form factor of monopole type, \\begin{equation} \\label{eq:monoff} f(\\mathbf{q}^{2})= \\frac{\\Lambda_\\mathrm{c}^{2}} {\\Lambda_\\mathrm{c}^{2}+\\mathbf{q}^{2}}\\, , \\end{equation} with moderate cutoff masses does not give a finite pairing gap in our model with the HS-m2 set; using other RMF parameter sets may give finite gaps and their gentle dependence on the cutoff mass is expected in the manner similar to the \\textit{NN} pairing~\\cite{matsuzaki01:_phenom}.  We would like to stress that we do not intend to provide the optimal parameter sets for the description of $\\Lambda\\Lambda$ pairing for the time being; or rather we intend to present its general trend of density dependence within the present model irrespective of a given set of parameters. Determining them precisely is inevitably deferred until the guide can be available.   \\subsection{Lambda-Sigma mixing} \\label{sec:lambda-sigma-mixing}  Before concluding the discussions, we present relevant issues for further study. We have employed pure neutron matter and symmetric nuclear matter as background in this study. Physics of neutron stars requires isospin asymmetricity of the background matter which should be considered in the next study. In connection with this, coherent and incoherent $\\Lambda$-$\\Sigma$ couplings should be mentioned. Reference~\\cite{akaishi00:_coher_lambd_sigma_coupl_shell_hyper} argued that they are important to understand \\textit{s}-shell hypernuclei and, in particular, resolve the longstanding problem of overbinding  in $\\substack{5 \\\\ \\Lambda}$He. Furthermore, the coherent $\\Lambda$-$\\Sigma$ coupling predicts the coherent $\\Lambda$-$\\Sigma^{0}$ mixing in dense neutron-rich infinite matter~\\cite{shinmura02:_coher_lambd_sigma}. As a consequence, the $\\Lambda$-$\\Sigma^{0}$ mixing shall come into play in asymmetric nuclear matter, which may change the critical density of hyperon emergence and eventually scenarios of the evolution of neutron stars. It is therefore important to introduce it into the models of dense hadronic matter. Concerning relativistic models, the introduction into both infinite and finite systems has been performed using the quantum hadrodynamics along with the concept of effective field theory~\\cite{mueller99:_effec_lambd_sigma,mueller00:_lambd_sigma}; these works also show the importance of the mixing. On the other hand, the QCD sum rules predict that relatively weak mixing would be realized for $\\Lambda$ and $\\Sigma^{0}$ of the positive energy state while strong for the negative~\\cite{yagisawa02:_in_sigma_lambd_qcd}. Hence, room for argument over this issue still remains and the effect of the mixing on the pairing is unknown so far. In all cases, we have ignored the mixing since it is beyond our scope of this study."
    },
    "0208/astro-ph0208501_arXiv.txt": {
        "abstract": "We present new \\sax LECS, MECS, and PDS observations of four flat--spectrum radio quasars (FSRQ) having effective spectral indices $\\alpha_{\\rm ro}$ and $\\alpha_{\\rm ox}$ typical of high-energy peaked BL Lacs. Our sources have X--ray--to--radio flux ratios on average $\\sim 70$ times larger than ``classical'' FSRQ and lie at the extreme end of the FSRQ X--ray--to--radio flux ratio distribution. The collected data cover the energy range $0.1 - 10$ keV (observer's frame), reaching $\\sim 100$ keV for one object. The \\sax band in one of our sources, RGB J1629+4008, is dominated by synchrotron emission peaking at $\\sim 2 \\times 10^{16}$ Hz, as also shown by its steep (energy index $\\alpha_{\\rm x} \\sim 1.5$) spectrum. This makes this object the {\\sl first} known FSRQ whose X--ray emission is not due to inverse Compton radiation. Two other sources display a flat \\sax spectrum ($\\alpha_{\\rm x} \\sim 0.7$), with weak indications of steepening at low X--ray energies. The combination of \\sax and ROSAT observations, (non-simultaneous) multifrequency data, and a synchrotron inverse Compton model suggest synchrotron peak frequencies $\\approx 10^{15}$ Hz, although a better coverage of their spectral energy distributions is needed to provide firmer values. If confirmed, these values would be typical of ``intermediate'' BL Lacs for which the synchrotron and inverse Compton components overlap in the \\sax band. Our sources, although firmly in the radio--loud regime, have powers more typical of high--energy peaked BL Lacs than of FSRQ, and indeed their radio powers put them near the low--luminosity end of the FSRQ luminosity function. We discuss this in terms of an anti-correlation between synchrotron peak frequency and total power, based on physical arguments, and also as possibly due to a selection effect. ",
        "introduction": "Blazars constitute one of the most extreme classes of active galactic nuclei (AGN), distinguished by their high luminosity, rapid variability, high ($> 3$\\%) optical polarization, radio core--dominance, and apparent superluminal speeds \\citep{kol94,urr95}. The broad--band emission in these objects, which extends from the radio to the gamma--ray band, appears to be dominated by non--thermal processes from the heart of the AGN, often undiluted by the thermal emission present in other AGN. Therefore, blazars represent the ideal class to study to further our understanding of non--thermal emission in AGN.  The blazar class includes BL Lacertae objects, characterized by an almost complete lack of emission lines, and a subclass of radio quasars (which by definition display broad emission lines) which have been variously called highly polarized quasars (HPQ), optically violently variable quasars (OVV), core-dominated quasars (CDQ). One of their observational properties which is easier to define is the flat--spectrum radio emission ($\\alpha_{\\rm r} \\la 0.5$) and so we will refer to them as flat--spectrum radio quasars (FSRQ).  Given the lack of prominent emission lines in BL Lacs, more than 95\\% of all known such objects have been discovered either in radio or X--ray surveys. Follow--up work on radio-- and X--ray--selected samples has shown that the two selection methods yield objects with somewhat different properties. The energy output of most radio selected BL Lacs peaks in the IR/optical band \\citep{gio94,pad95,pad96}; such objects are now referred to as LBL (low-energy peaked BL Lacs). By contrast, the energy output of most X--ray selected BL Lacs (referred to as HBL: high-energy peaked BL Lacs) peaks at UV/X--ray energies.  \\citet{pad95} and \\citet{sam96} have demonstrated that the difference in broad-band peaks for HBL and LBL is not simply phenomenological. Rather, it represents a fundamental difference between the two sub--classes. The location of the broadband peaks also suggests a different origin for the X--ray emission of the two classes. Namely, an extension of the synchrotron emission likely responsible for the lower energy continuum in HBL, which typically display steep (energy index $\\alpha_{\\rm x} \\sim 1.5$) X--ray spectra, and inverse Compton emission in LBL, which have harder ($\\alpha_{\\rm x} \\sim 1$) spectra \\citep{per96,urr96,pad96}. {\\it BeppoSAX} observations of BL Lacs are confirming this picture \\citep{wol98,padet01,bec02}.  In this respect, one could expect to find a similar range in peak frequencies in the FSRQ class -- for which, until recently, no evidence existed. Indeed, it was suggested by some authors \\citep{sam96}, based upon the similarities of the optical--X--ray broad-band spectral characteristics of LBL and FSRQ, that no FSRQ with synchrotron peak emission in the UV/X--ray band should exist.  Two studies have drastically changed this picture: 1. the multifrequency catalog of \\citet{pad97} identified more than 50 FSRQ ($\\sim 17$\\% of the FSRQ in their catalog) spilling into the region of parameter space once exclusively populated by HBL; 2. about 30\\% of FSRQ found in the deep X--ray radio blazar survey (DXRBS) \\citep{per98,lan01} were found to have X--ray--to--radio luminosity ratios, $L_{\\rm x}/L_{\\rm r}$, typical of HBL ($L_{\\rm x}/L_{\\rm r} \\ga 10^{-6}$ or $\\alpha_{\\rm rx} \\la 0.78$), but broad (FWHM $> 2,000 {\\rm ~km ~s^{-1}}$) and luminous ($L > 10^{43} {\\rm ~erg ~s^{-1}}$) emission lines typical of FSRQ. The discovery of a large population of ``X--ray strong'' FSRQ (labeled HFSRQ by \\citet{per98} to parallel the HBL moniker) represents a fundamental change in our perception of the broadband emission of FSRQ.  X--ray observations of these objects play a fundamental role in finding their place within the blazar class. For example, if the X--ray spectra were found to be relatively steep, one could infer a dominance of synchrotron emission, as observed in HBL. Flatter X--ray spectra, with corroborating evidence from the whole broad-band emission \\citep{pad96}, would instead suggest inverse Compton emission. In the latter case, the simple equations ${\\rm LFSRQ} \\equiv {\\rm LBL}$ (where by LFSRQ we mean the ``typical'' FSRQ with low-energy synchrotron peak) and ${\\rm HFSRQ} \\equiv {\\rm HBL}$ would not be valid, and some more complicated explanation for the existence of this class should be sought.  Our previous knowledge of the X--ray spectra of this new class of objects is scanty and mostly based on low signal--to--noise ratio (S/N) ROSAT data (Padovani et al. 1997; Padovani et al., in preparation) and is therefore also limited to the relatively narrow $0.1-2.4$ keV band. The \\sax satellite \\citep{boe97a}, with its broad-band X--ray ($0.1-300$ keV) spectral capabilities, is particularly well suited for a detailed analysis of the individual X--ray spectra of these sources.  In this paper we present \\sax observations of four HFSRQ candidates, selected as described below. In \\S~2 we present our sample, \\S~3 discusses the observations and the data analysis, while \\S~4 describes the results of our spectral fits to the \\sax data. \\S~5 discusses the ROSAT data, \\S~6 presents the spectral energy distributions and synchrotron-inverse Compton fits to the data, \\S~7 discusses our results, while \\S~8 summarizes our conclusions. Throughout this paper spectral indices are written $S_{\\nu} \\propto \\nu^{-\\alpha}$ and the values $H_0 = 50$ km s$^{-1}$ Mpc$^{-1}$ and $q_0 = 0$ have been adopted.  Readers wanting to skip the details of the data reduction and go directly to our results can read \\S~2, the summaries of our \\sax and ROSAT results in \\S\\S~\\ref{SAX_sum} and \\ref{ROSAT_sum}, and then go straight to \\S~6. ",
        "conclusions": "We have presented new \\sax observations of four flat--spectrum radio quasars selected to have effective spectral indices ($\\alpha_{\\rm ro}$, $\\alpha_{\\rm ox}$) typical of high-energy peaked BL Lacs. The main purpose of the paper was to see if these sources are indeed the broad-lined counterparts of BL Lacs with the synchrotron peak at UV/X--ray energies (HBL). Our objects are quite extreme in terms of their X--ray--to--radio flux ratios (a factor $\\sim 70$ higher than ``classical'' FSRQ), qualify as radio-loud sources both in terms of their radio-to-optical flux ratio and power, and are variable and radio-compact. Our main results can be summarized as follows:  \\begin{enumerate}  \\item We have discovered the first FSRQ (RGB J1629+4008) whose X--ray emission is dominated by synchrotron radiation, as clearly shown by our \\sax observations ($\\alpha_{\\rm x} \\sim 1.5$ in the $0.1-10$ keV band), in which the source was also rapidly variable. This object is therefore the first confirmed member of a newly established class of HBL--like FSRQ. This result is fully consistent with earlier ROSAT observations which detected a steep soft X--ray spectrum ($\\alpha_{\\rm x} \\sim 2$ in the $0.1-2.4$ keV band). The combination of the X--ray data with archival radio and optical data gives a spectral energy distribution typical of HBL sources, with a synchrotron $\\nu_{\\rm peak} \\sim 2 \\times 10^{16}$ Hz. The derived values follow very well the $\\alpha_{\\rm x} - \\nu_{\\rm peak}$ correlation seen for HBL sources (Fig. \\ref{fig8}).  \\item The other three sources show a relatively flat ($\\alpha_{\\rm x} \\sim 0.75$) \\sax spectrum. However, we have found for two of them (WGA J0546$-$6415 and RGB J1722+2436) some indication of steepening at low-energies. This is based on \\sax and ROSAT data and on their (sparsely sampled) SEDs. We interpret this as the tail of synchrotron emission, which is not strongly constrained but would peak around $\\approx 10^{15}$ Hz, as typical of ``intermediate'' BL Lacs for which the synchrotron and inverse Compton components meet within the \\sax band. The last source (S5 2116+81) turned out to be a more typical FSRQ, with the \\sax band fully dominated by inverse Compton emission and a synchrotron $\\nu_{\\rm peak}<10^{15}$ Hz.  \\item By fitting a synchrotron inverse Compton model which includes the contribution of an accretion disk, whose power we estimate from the broad-line region luminosity, to the spectral energy distributions, we have derived physical parameters (e.g., intrinsic power, magnetic field, etc.) for our sources. The thermal component was found to be non-negligible in three out of four objects, with ratios of thermal to synchrotron emission in the range $0.6 - 2$ at 5,000 \\AA.  \\item Although the original selection was based mostly on the effective broad-band (radio through X--ray) colors, which are in first approximation independent of luminosity, all four candidates are at relatively low powers, more typical of BL Lacs than of FSRQ and in any case close to the low-luminosity end of the FSRQ radio luminosity function \\citep{pad01}. We interpret this as due to two, possibly concurrent, causes: an inverse dependence of $\\gamma_{\\rm peak}$ (and therefore $\\nu_{\\rm peak} \\propto \\gamma_{\\rm peak}^2 \\delta B$), the Lorentz factor of the electrons emitting most of the radiation, on bolometric power, due to the more sever cooling at work in more powerful sources; and a selection effect, namely the fact that a high-power, HBL--like quasar will have such a strong optical/UV non-thermal component that its emission lines will be swamped and the object will be classified as a featureless BL Lac. \\end{enumerate}  A better understanding of these rare sources and the confirmation of the non-existence of high--$\\nu_{\\rm peak}$ --- high--power blazars will require the selection and study of a larger sample of objects."
    },
    "0208/astro-ph0208447_arXiv.txt": {
        "abstract": "We investigate the impact of an early population of massive stars on their surroundings.  Dissociation of molecular hydrogen by strong UV emission from such stars is expected to produce a global transition in the cooling mechanism of minihalos at a redshift of approximately 30, strongly inhibiting star formation until more massive halos can collapse.  Furthermore, chemical enrichment from Pop~III supernovae produces a second transition at $z\\sim15-20$, when the mean metallicity of the universe exceeds a critical threshold and Pop~III star formation gives way to Pop~II.  We show that observations of high redshift supernovae with the {\\it Next Generation Space Telescope} (NGST) have the potential to trace the cosmic star formation rate out to $z\\ga 20$, provided that Pop~III supernovae are at least as bright as, and ideally brighter than, type Ia supernovae.  We also propose a mechanism for the formation of a novel population of extremely low metallicity stars of intermediate mass at very high redshifts, which we term Pop~II.5.  In our model shock compression, heating, and subsequent cooling to high density reduces the fragment mass in primordial gas to $\\sim10\\,\\msun$, allowing low mass stars to form. We predict the number density of relic Pop~II.5 stars in the Milky Way halo today and find that, with certain assumptions, there should be $\\sim 10\\ {\\rm kpc^{-3}}$ in the solar neighborhood. ",
        "introduction": "An important challenge in modern cosmology is to understand how the cosmic ``dark ages'' ended \\citep[for a recent review, see][]{BarLoe01}.  There is growing theoretical evidence indicating that the first luminous objects to form in the universe were very massive stars with typical masses $M_{\\ast}\\ga 100\\,\\msun$ \\citep*{BroCopLar99,BroCopLar02,AbeBryNor00,AbeBryNor02,NakUme01}. These stars formed out of metal-free gas in dark matter (DM) halos of mass $\\sim 10^5 - 10^6 \\;\\msun$ at redshifts $z \\ga 20$ \\citep[e.g.,][]{Tegetal97,FulCou00}.  Typically, simulations predict that the halos hosting this process should contain either one or a few dense massive clumps of baryonic matter, with the most massive clump situated nearest to the center of the halo being the one to collapse first and to presumably form a very massive star (VMS).  On the other hand, conditions in the universe less than $10^{9}$ yr after the big bang must have allowed for the formation of low-mass stars, with typical masses of $\\sim 1 \\;\\msun$, as is implied by the ages of the oldest globular clusters in the Galactic halo with metallicities of $Z\\sim 10^{-2}\\zsun$ \\citep[e.g.,][]{AshZep98,BroCla02}.  In this paper, we investigate the question: {\\it How and when did the transition from the formation of very massive stars at high redshifts to that of more normal, low-mass stars at later times take place?}  Observations have recently provided hints to the character of star formation at $z\\ga 5$.  The abundance of C~IV in the low-column density Ly$\\alpha$ forest indicates that the intergalactic medium (IGM) was enriched with heavy elements to a level of $Z\\sim 10^{-3.5}\\zsun$ already at $z\\sim 5$ \\citep{Son01}. In addition, abundance patterns in Ly$\\alpha$ systems out to $z \\sim 4.6$ are suggestive of an early, prompt nucleosynthetic inventory of heavy elements from a generation of very massive stars \\citep*{QiaSarWas02}.  Elucidating the cosmic star formation history in the high redshift universe is crucial for predicting what observatories such as the {\\it Next Generation Space Telescope} (NGST) will observe at $z > 5$ less than a decade from now. In particular, it is important to ascertain the rates and properties of high redshift supernovae (SNe) which are likely to be the brightest beacons heralding the end of the ``dark ages'' \\citep{MirRee97}. Similarly, gamma-ray bursts (GRBs) are expected to trace the formation history of massive stars out to very high redshifts \\citep{Tot97,Wijetal98,BlaNat00}.  In fact, the top-heavy initial mass function (IMF) predicted for the first stars favors the massive stars which are the likely source of GRB progenitors.  Constraining how stars form at high $z$, therefore, is of great relevance for interpreting the results of the upcoming {\\it Swift} satellite \\citep{BroLoe02}.  Here, we examine processes related to star formation in the high redshift universe using a simple analytical model that synthesizes a number of recent theoretical and observational results into a coherent framework.  This idealized model is intended to emphasize the basic physical mechanisms, and we will complement it with detailed numerical simulations in future work.  The organization of the paper is as follows. In \\S 2 we describe our high-redshift star formation model.  A discussion of the various stellar populations that form at high $z$ is given in \\S 3, while \\S 4 presents observational consequences. Finally, \\S 5 contains our conclusions and the outlook for future work. ",
        "conclusions": "We have investigated the history of star formation in the high redshift universe, focusing on the role played by very massive Pop~III stars. We have argued that this history is shaped by the various feedback effects exerted by those stars, resulting in three distinct epochs of star formation.  The first impact Pop~III stars have on their surroundings is radiative, as their soft UV flux dissociates the molecular hydrogen in other minihalos \\citep{HaiReeLoe97,HaiAbeRee00}.  Because of their copious UV emission, and the small molecular fraction ($\\sim10^{-4}-10^{-6}$) in minihalos, this photodissociation is likely to happen quickly, and fairly completely.  A numerical investigation of this process was performed by \\citet{MacBryAbe01}, who simulated this suppression of star formation in low mass halos by incorporating a uniform soft UV background into a cosmological simulation.  While this assumption does not allow them to follow the detailed build-up of the UV radiation field as Pop~III star formation switches on, it does give a good indication of the overall effect.  It was found that the UV radiation can effectively suppress star formation due to molecular cooling in low mass halos, delaying their collapse from redshift $z\\sim 30$ to $z\\sim 20$.   It is encouraging that both analytic and numerical calculations agree quite well.  In Figure~\\ref{fig2}, we find that once a significant UV radiation field has been set up by $z\\sim30$, further star formation is strongly inhibited until more massive halos start to form, and the SFR does not recover to its $z\\sim 30$ value until $z\\la 20$.  This is all in good agreement with the simulation results, except that \\citet{MacBryAbe01} did find that more massive halos of $M\\ga 10^7\\;\\msun$ were able to retain some molecular hydrogen.  Hence, these halos were still able to cool via molecular lines, indicating that the SFR should not go all the way down to the atomic cooling curve, but it is likely to drop most of the way.  Future simulations will be able to follow this dissociation process in detail, taking into account the effect of self-shielding, with a time-varying radiation field generated by point sources at the sites of Pop~III star formation. In summary, the radiative feedback from Pop~III stars results in the transition, at $z\\sim 30$, between the first two epochs of star formation at high redshift.  The second impact Pop~III stars have on surrounding material is chemical, and we identify this feedback as the main effect governing the transition from forming predominantly very massive stars to forming normal, lower mass ones \\citep[e.g.,][]{Omu00,NisTas00,Broetal01,Schetal02}.  With the assumption that VMSs can only form in gas with a metallicity lower than a critical value, $Z_{\\rm crit}\\sim 10^{-3.5}\\zsun$, we estimate that most of the star-forming material in the IGM has reached this level in the redshift range $z\\sim 15 - 20$, which delineates the second transition in the star formation history, between epochs 2 and 3.  An important issue is whether such a well-defined transition redshift between massive and more normal star formation actually occurs.  This depends on how synchronized different parts of the universe are in crossing the critical metallicity threshold, which in turn depends on how well the supernova ejecta are mixed through the IGM at high redshifts.  It has been suggested \\citep{MadFerRee01} that enrichment becomes more synchronized and uniform at higher redshifts.  Dark matter halos are of much lower mass and have shallower potential wells, making it easier for metals to escape the halo.  Supernovae from VMSs are predicted to be very energetic, with explosion energies significantly larger than the binding energy of halos at $z\\ga 20$ \\citep{BarLoe01}.  Furthermore, the universe is much denser at high redshift and halos are consequently much closer together, again making it easier to pollute the IGM uniformly with metals.  It is also possible, however, that enrichment at high redshifts is similar to that at low redshifts, where it has been shown to be very inhomogeneous and incomplete \\citep[e.g.,][]{Gne98,CenOst99,Aguetal01a,Aguetal01b}.  If this were the case, the transition redshift at which the mean mass-weighted metallicity of the universe crosses the critical threshold would not be a very meaningful quantity.  Which of these two cases is true is an important and interesting question, and can only be addressed theoretically by more detailed cosmological simulations which can resolve star-forming halos at high redshifts.  The demise of Pop~III stars in SN explosions opens up the possibility for a new stellar population which we have termed Population~II.5. These stars are characterized by masses, $M_{\\ast}\\sim 10\\,\\msun$, intermediate to that of Pop~III and Pop~II stars, and by extremely low metallicities, below the critical value that would normally lead to the formation of VMSs.  This possibility relies on the gas in the vicinity of the SN being compressed in a radiative shock, leading to a significant increase in density, and thus possibly to a reduction of the fragment mass, $M_{\\rm BE}\\propto \\rho^{-1/2}$, to values much lower than the one realized in primordial gas.  During an intermediate epoch, therefore, low- and high-mass stars might form almost simultaneously. A similar idea had already been proposed in a prescient paper by \\citet{Cay86}, primarily to explain the so-called G-dwarf problem, i.e., the observed lack of metal-poor stars in our Galaxy \\citep[see also][]{KasRee83,NakUme01}.  It is to be expected that the mass range of Pop~II.5 stars extends down to $\\la 1\\;\\msun$, and these stars at the low-mass end of the Pop~II.5 IMF should still be around today.  Intriguingly and possibly providing a case in point, the recent discovery of the extremely metal poor star CS~29498-043 with [Fe/H]$=-3.7$ and a significant overabundance of Mg and Si, has been tentatively interpreted as hinting at a new class of stars \\citep{Aoketal02}. Ongoing surveys of extremely metal-poor halo stars, much improved in size and quality, should soon be able to test our prediction further.  Many of the uncertainties in our argument have to be addressed with detailed numerical simulations, and we plan to do so in future work. The framework presented in this paper does provide a coherent context for these numerical studies, and it highlights the important physical questions that define the challenge of elucidating the end of the cosmic ``dark ages''."
    },
    "0208/astro-ph0208392_arXiv.txt": {
        "abstract": "{We show how the temperature and the polarisation of the cosmic microwave background are affected by bulk rotation of clusters of galaxies owing to the kinetic Sunyaev-Zeldovich effect.  The main effects of rotation are (i) a shift of the position of the peak of the temperature fluctuation relative to the center of the cluster by a few percent of the core radius and (ii) a tilt of the direction of the plane of linear polarisation by several degrees. ",
        "introduction": "Several effects lead to anisotropies of the cosmic microwave background (CMB): primary effects, imprinted on the surface of last scattering, and secondary effects, arising after hydrogen-recombination or after reionisation has taken place \\citep[for references see][]{White2002}.  Among the secondary effects, the thermal Sunyaev-Zeldovich effect (th-SZE), which is due to inverse Compton scattering of the CMB photons off the hot intracluster medium (ICM) \\citep{SunyZeld69}, and the kinetic Sunyaev-Zeldovich effect (k-SZE), which arises from the peculiar motion of the cluster in the rest frame of the CMB \\citep{SunyZeld80}, are most important. The cosmological importance of the th-SZE is due to the fact that it is redshift independent. Therefore it can be used to detect clusters of galaxies at redshifts, where other observational methods fail \\citep{Korolev86,Kneissl2001}, and to determine the Hubble constant \\citep{Birkinshaw1999, Reese2002}. In principle the k-SZE can be used to determine both components of the peculiar velocity: the line of sight component via the relative change of intensity of the CMB and the velocity component in the celestial plane via the degree of linear polarisation of the CMB radiation. Measurements of the k-SZE are extremely difficult and will probably become feasible in the near future.  In this paper we address the k-SZE arising from bulk rotation of the ICM. In order to distinguish this effect from the k-SZE due to a bulk translatory motion of the cluster with respect to the CMB, we shall henceforth coin it the rotational kinetic Sunyaev-Zeldovich effect (rk-SZE). To our knowledge, the rk-SZE was previously only considered by \\cite{Cooray2002}. In their work the rk-SZE was discussed for a gas density profile following from hydrostatic equilibrium of the gas in the Navarro-Frenk-White dark matter density field within a halo \\citep{NFW96, Makino98}. Assuming isothermality this gas density profile is very well approximated by the commonly used isothermal $\\beta$-model \\citep{Caval76}, which is better applicable to analytical calculations. Here we focus on the contributions of the rk-SZE to the temperature fluctuations and the linear polarisation of the CMB for the isothermal $\\beta$-model finding a new method to measure the rotational properties of a cluster by performing multifrequency measurements of the position of the peak of the temperature fluctuation.  The paper is organized as follows. In Sect. \\ref{sec:calc} we state the model assumptions for the rotating cluster of galaxies and derive analytic formulae describing the rk-SZE. We also derive formulae describing combined thermal, kinetic and rotational kinetic SZE. In Sect. \\ref{sec:dis} we discuss our analytic results using recent observational data to estimate the effects for a set of 18 clusters in the redshift range from $z\\sim 0.14$ to $z\\sim 0.78$. Finally, we draw our conclusions in Sect. \\ref{sec:conc}.  \\newpage ",
        "conclusions": "\\label{sec:conc} We have derived analytic formulae describing the relative change of intensity and the degree of linear polarisation of the CMB radiation due to the kinetic SZE of a rotating cluster of galaxies (Eq. \\eqref{eq:ResultkinRotClusterInk} and \\eqref{eq:resultPOLrot}). We also have found analytic formulae for the superposition of all possible SZE contributions (Eq. \\eqref{eq:RELCHANGETOT} and \\eqref{eq:resultQUsup}). We have estimated the possible amplitude of the relative change of intensity due to the rk-SZE for a sample of 18 clusters of galaxies in a redshift range from $z\\sim 0.14$ to $z\\sim 0.78$ (see Table \\ref{tab:results}). Our results show, that the contribution of the rk-SZE to the peak temperature change for this sample can be expected to range from $\\Delta T_{\\rm rot} \\sim 3.5\\,\\mu$K for $\\beta_{\\rm c}\\sim 10^{-4}$ up to $\\Delta T_{\\rm rot} \\sim 146\\,\\mu$K for a recent merger ($\\beta_{\\rm c}\\sim 5\\cdot 10^{-4}$) of rich clusters. This agrees with the range predicted by \\cite{Cooray2002}.  We have also shown that due to the superposition of the thermal, the kinetic and the rotational kinetic SZE there is a frequency dependent displacement (Eq. \\eqref{eq:Versch}) of the peak value of the relative intensity change, which in principle can be used to examine the properties of the rotational velocity component of the ICM. Ground based interferometric telescopes should be able to detect this effect in the near future. Since the angular momentum distribution is not easily measurable with other observational techniques, this aspect of the rk-SZE might provide a new possibility of getting insights into the internal dynamics of clusters of galaxies.  In the future polarisation measurements of the CMB radiation may become feasible. In this work we have shown that the polarisation map following from the k-SZE can alter significantly due to the rk-SZE. Although the contribution of the rk-SZE to the degree of polarisation relative to the k-SZE is only of the order of a few $\\%$ (quantitatively $\\Delta T_{\\rm r} \\sim 10^{-4}\\mu$K), the directions of planes of polarisation can be tilted by an angle, which is of the order of a few degrees (Eq. \\eqref{eq:Tiltappr}). Since this tilt is frequency independent it can be easily separated from other effects, for example Faraday rotation. In the Wien region of the CMB spectrum the degree of polarisation due to the kinetic and the rotational kinetic SZE increases by a factor of $10-100$. Therefore measurements of the CMB polarisation should be performed in this frequency range."
    },
    "0208/astro-ph0208117_arXiv.txt": {
        "abstract": "We have used the publicly available data from the 2dF Galaxy Redshift Survey and the 2dF QSO Redshift Survey to test the hypothesis that there is a periodicity in the redshift distribution of quasi-stellar objects (QSOs) found projected close to foreground galaxies.  These data provide by far the largest and most homogeneous sample for such a study, yielding 1647 QSO--galaxy pairs.  There is no evidence for a periodicity at the predicted frequency in $\\log(1+z)$, or at any other frequency. ",
        "introduction": "\\footnotetext{E-mail: ppxeh@@nottingham.ac.uk}  Claims of periodicities or regularities in redshift distributions of various astronomical objects have been made for many years (e.g. Burbidge \\& Burbidge 1967; Broadhurst \\etal, 1990; Karlsson, 1990; Burbidge \\& Napier, 2001).  This effect, if real, has far-reaching implications for the interpretation of redshift as a cosmological phenomenon, and, indeed, for the nature of objects like quasi-stellar objects (QSOs) that appear to display the periodicities.  One particularly intriguing effect has been explored by Arp et al.\\ (1990) and Karlsson (1990) and extended to a larger sample by Burbidge \\& Napier (2001).  It involves the apparent strong periodicity in $\\log(1+\\zqso)$ for a sample of QSO redshifts, $\\zqso$, where the QSO appears projected close to a ``foreground'' galaxy at lower redshift. If confirmed, such an effect would be impossible to explain in conventional cosmological terms: it would either require that the QSOs be physically associated with the galaxies in an as-yet unexplained fashion, or that the QSO light passing the galaxy is somehow influenced to quantize its redshift.  The criticism usually levelled at this kind of study is that the samples of redshifts have tended to be rather small and selected in a heterogeneous manner, which makes it hard to assess their significance.  The more cynical critics also point out that the results tend to come from a relatively small group of astronomers who have a strong prejudice in favour of detecting such unconventional phenomena.  This small group of astronomers, not unreasonably, responds by pointing out that adherents to the conventional cosmological paradigm have at least as strong a prejudice towards denying such results.  In an attempt to circumvent these problems, Bill Napier contacted the authors of this paper.  The availability of the data from the 2dF Galaxy redshift Survey (2dFGRS) and the 2dF QSO Redshift Survey (2QZ) means that for the first time there exists a large homogeneous sample of data to carry out this kind of study.  Furthermore, Napier recognized the importance of the study being carried out independent from any of the researchers with vested interests one way or the other.  He therefore gave clear instructions as to what analysis should be performed and what periodic effect should be seen if the phenomenon is real, but chose to take no part in the subsequent analysis.  We have attempted to carry out this analysis without prejudice.  Indeed, we would have been happy with either outcome: if the periodicity were detected, then there would be some fascinating new astrophysics for us to explore; if it were not detected, then we would have the reassurance that our existing work on redshift surveys, etc, has not been based on false premises.  The remainder of this paper is laid out as follows. Section~\\ref{s:napier} presents Napier's prediction as to what signal we should expect to see if the data are analysed appropriately. Section~\\ref{s:data} describes the data set, and Section~\\ref{s:meth} presents the manner in which it has been analysed.  The results are described in Section~\\ref{s:res}. ",
        "conclusions": ""
    },
    "0208/astro-ph0208321_arXiv.txt": {
        "abstract": "{ We have analysed high resolution adaptive optics (AO) science demonstration data of the young, massive stellar cluster Arches near the Galactic Center, obtained with the Gemini North telescope in combination with the University of Hawai'i AO system Hokupa'a. The AO H and K' photometry is calibrated using HST/NICMOS observations in the equivalent filters F160W and F205W obtained by Figer et al. (\\cite{FKM}). The calibration procedure allows a detailed comparison of the ground-based adaptive optics observations against diffraction limited space-based photometry. The spatial resolution as well as the overall signal-to-noise ratio of the Gemini/Hokupa'a data is comparable to the HST/NICMOS data. The low Strehl ratio of only a few percent is the dominant limiting factor in the Gemini AO science demonstration data as opposed to space-based observations. After a thorough technical comparison, the Gemini and HST data are used in combination to study the spatial distribution of stellar masses in the Arches cluster. Arches is one of the densest young clusters known in the Milky Way, with a central density of $\\sim 3 \\cdot 10^5\\,M_\\odot\\,{\\rm pc^{-3}}$ and a total mass of about $10^4\\,M_\\odot$. A strong colour gradient is observed over the cluster field. The visual extinction increases by $\\Delta A_V \\sim 10$ mag over a distance of 15\\arcsec\\ from the cluster core. Extinction maps reveal a low-extinction cavity in the densest parts of Arches ($R \\leq 5\\arcsec$), indicating the depletion of dust due to stellar winds or photo-evaporation. We correct for the change in extinction over the field and show that the slope of the mass function is strongly influenced by the effects of differential extinction. We obtain present-day mass function slopes of $\\Gamma \\sim -0.8 \\pm 0.2$ in the mass range $6\\!<\\!M\\!<\\!65\\ M_\\odot$ from both data sets. The spatial analysis reveals a steepening of the mass function slope from close to zero in the cluster center to $\\Gamma \\sim -1.7 \\pm 0.7$ at $R > 10\\arcsec$, in accordance with a Salpeter slope ($\\Gamma = -1.35$). The bias in the mass function towards high-mass stars in the Arches center is a strong indication for mass segregation. The dynamical and relaxation timescales for Arches are estimated, and possible mass segregation effects are discussed with respect to cluster formation models. ",
        "introduction": "The Galactic Center (GC) is the most extreme star forming environment within the Milky Way. High stellar and gas densities, turbulent motion, tidal torques exerted by the steep gravitational potential, magnetic fields and an intense radiation field determine the physical environment of star formation in the GC region. Although disruptive forces exerted by the gravitational and radiation fields counteract the agglomeration of material, the high gas and dust densities cause star formation in the GC environment to be most efficient. In particular, the formation of high mass stars and massive clusters is more successful than in any other region of the Milky Way.  A detailed study of star formation processes and the stellar content of the GC region has until recently been limited to the brightest and most massive stars due to the large amount of extinction ($A_V \\sim 30$ mag) along the line of sight. Additional constraints are imposed due to the spatial resolution at the GC distance of $\\sim 8$ kpc ($DM = 14.47 \\pm 0.08$ mag, e.g., McNamara et al. \\cite{McNamara2000}), much farther than nearby star forming regions such as the Orion or $\\rho$ Ophiuchi star forming complexes, which have been studied in greater detail to date. Only with the advent of deep, high resolution near-infrared instruments, the analysis of stellar populations in young star clusters near the GC has become feasible.  During the past few years, it has become evident that three out of four young starburst clusters known in the Milky Way are located in the GC region - namely, the \\object{Arches} and Quintuplet clusters, as well as the Galactic Center Cluster itself. With a cluster age of only a few Myr for Arches and Quintuplet, the question arises how many clusters do actually form in the densest environment of the Milky Way. The 2MASS database yielded new insights into the estimated number of star clusters hidden in the dense stellar background. Dutra \\& Bica (\\cite{DB2000}, \\cite{DB2001}) report the detection of new cluster candidates of various ages located in the innermost 200 pc of the Galaxy found in 2MASS. Numerical simulations by Portegies Zwart et al. (\\cite{PZ2001}) suggest that clusters with properties similar to the massive Arches and Quintuplet may have formed in the past in the innermost 200 pc, but were then dispersed and are now indistinguishable from the dense stellar background. As dynamical evolution timescales are short due to the strong tidal field in the GC region (Kim et al. \\cite{Kim1999}), young star clusters are disrupted quickly after formation, contributing to the Galactic bulge population. Thus, only the youngest clusters remain intact for the study of star formation in this extraordinary environment.  The Arches cluster, at a projected distance of only 25 pc from the GC (assuming a heliocentric distance of 8 kpc to the GC), is one of the most massive young clusters known in the Milky Way. With an estimated mass of about $10^4\\,M_{\\sun}$ and a central density of $3 \\cdot 10^5\\,M_{\\sun}\\,{\\rm pc}^{-3}$, Arches is the densest young star cluster (YC) known (Figer et al. \\cite{FKM}). From physical properties of Wolf-Rayet stars, the age of the cluster is estimated to be between 2 and 4.5 Myr (Blum et al. \\cite{Blum2001}). The stellar content of Arches has been studied by Figer et al. (\\cite{FKM}) using HST/NICMOS data. They derived a shallow initial mass function in the range $6\\!<\\!M\\!<\\!120\\,M_{\\sun}$ with a slope of $\\Gamma = -0.7 \\pm 0.1$, but with significant flattening observed in the innermost part of the cluster ($\\Gamma = -0.1 \\pm 0.2$).  Most young star clusters and associations in the Milky Way display a mass function close to a Salpeter (1955) power law with a slope of $\\Gamma = -1.35$. Several such star forming regions have been studied by Massey et al. (\\cite{Massey1995a}), yielding slopes in the range $-0.7 < \\Gamma < -1.7$ with an average of $-1.1 \\pm 0.1$, which leads these authors to conclude that within the statistical limits no deviation from a Salpeter slope is observed.  A flat mass function as observed in Arches implies an overpopulation of the high-mass end as compared to ``normal'' clusters. The special physical conditions in the GC region have been suggested to enhance the formation of massive stars, thereby resulting in a flattened mass function (Morris \\cite{Morris1993}). The formation of high-mass stars in itself poses serious problems for the standard core collapse and subsequent accretion model, as radiation pressure from the growing star is capable of reversing the gas infall as soon as the mass is in excess of $10\\,M_{\\sun}$ (Yorke \\& Kr\\\"ugel \\cite{Yorke1977}). Assuming disk accretion instead of spherical infall, the limiting mass may be increased to $15\\,M_{\\sun}$ (Behrend \\& Maeder \\cite{Behrend2001}), still far below the mass observed in O-type stars. Various scenarios are suggested to circumvent this problem. Simulations with enhanced accretion rates and collision probabilities in dense cluster centers (Bonnell et al. \\cite{Bonnell1998a}), as well as growing accretion rates depending on the mass of the accreting protostar (Behrend \\& Maeder \\cite{Behrend2001}), allow stars of up to $100\\,M_{\\sun}$ to form in the densest regions of a rich star cluster. In case of the GC environment, a higher gas density may lead to a higher accretion rate and/or to a longer accretion process in the protostellar phase. As long as the gravitational potential is strongly influenced by the amount of gas associated with the cluster, gas infall causes a decrease in cluster radius and subsequent increase in the collision rate, reinforcing the formation of high-mass stars. Physical processes such as gravitational collapse or cloud collisions scale with the square root of the local density, $\\sqrt{\\rho}$ (Elmegreen \\cite{Elmegreen1999}, \\cite{Elmegreen2001}), causing an enhanced star formation rate (SFR) in high density environments. Elmegreen (\\cite{Elmegreen2001}) shows that the total mass as well as the maximum stellar mass in a cluster strongly depends on the SFR and local density. This is confirmed by observations of high-mass stars found predominantly in the largest star forming clouds (Larson \\cite{Larson1982}).  Both the growing accretion and the collision scenario predict the high-mass stars to form in the densest central region of a cluster, leading to primordial mass segregation, which may be evidenced in a flat mass function in the dense cluster center. As an additional physical constraint, both scenarios require the lower-mass stars to form first, and the highest-mass stars last in the cluster evolution process. As the strong UV-radiation field originating from hydrogen ignition in high-mass stars expells the remaining gas from the cluster center, the accretion process should be halted immediately after high-mass star formation.  The short dynamical timescales of compact clusters are, however, influencing the spatial distribution of stellar masses as well. On the one hand, high-mass stars are dragged into the cluster center due to the gravitational potential of the young cluster. On the other hand, low-mass stars may easily be flung out of the cluster due to interaction processes, especially given star densities as high as in the Arches cluster. The result of these processes would also be a flat mass function in the cluster center, steepening as one progresses outwards due to dynamical mass segregation. Dynamical segregation is predicted to occur within one relaxation time (Bonnell \\& Davies \\cite{Bonnell1998b}), which for compact clusters is only one to a few Myr, and should thus be well observable in Arches in the form of a spatially varying mass function.  In addition to the internal segregation process, the external GC tidal field exerts shear forces tearing apart the cluster entity. N-body simulations by Kim et al. (\\cite{Kim2000}) yield tidal disruption timescales as short as 10 to 20 Myr in the GC tidal field. We expect to find a mixture of all these effects in the Arches cluster.  We have analysed adaptive optics (AO) data obtained under excellent seeing conditions with the Gemini North 8m telescope in combination with the University of Hawai'i (UH) AO system Hokupa'a. We are investigating  the presence of radial variations in the mass distribution within the Arches cluster. We compare our ground-based results in detail to the HST/NICMOS data presented in Figer et al. (\\cite{FKM}, hereafter FKM), discussing possible achievements and limitations of ground-based, high-resolution adaptive optics  versus space-based deep NIR photometry.  In Section 2, we will introduce the data and describe the reduction and calibration processes. In this context, a thorough investigation of the quality of ground-based adaptive optics photometry as compared to space-based diffraction limited observations will be presented. The photometric results derived from colour-magnitude diagrams and extinction maps will be discussed in Section 3. A comparison of Gemini and HST luminosity functions will be given in Section 4. The mass functions will be derived in Section 5, and their spatial variation will be discussed with respect to cluster formation scenarios. We will estimate the relevant timescales for cluster evolution for the Arches cluster in Section 6, and discuss the implication on the dynamical evolution of Arches. We will summarise our results in Section 7. ",
        "conclusions": "We have analysed high-resolution Gemini/Hokupa'a adaptive optics and HST/NICMOS data of the Arches cluster near the Galactic Center with respect to spatial variations in the mass function and their implications for cluster formation. A detailed comparison of the Gemini data to HST/NICMOS observations allows us to investigate the instrumental characteristics of PSF fitting photometry with the Hokupa'a AO system.  \\subsection{Technical comparison of Gemini/Hokupa'a with HST/NICMOS}  The calibration of the Gemini/Hokupa'a data of the Arches cluster using HST/NICMOS data from Figer et al. (\\cite{FKM}) allows us to carry out a detailed technical comparison of the two datasets. Maps of photometric residuals show a strong dependence of the calibration error on the stellar density within the field. In particular, the vicinity of fainter objects to bright stars causes the Gemini magnitude to be underestimated in comparison with HST/NICMOS. This is understandable as the uncompensated seeing halos of bright stars enhance the background in a non-homogeneous way, thereby causing an overestimation of the background and a subsequent underestimation of the faint objects' magnitude. Conversely, the flux of very bright sources seems to be overestimated. The correlation of the photometric residual with the position of bright stars is more pronounced in $K^\\prime$, where crowding is the dominant source of photometric uncertainty, while the effect of angular anisoplanatism is less severe. In the $H$-band the anisoplanatism is more pronounced, and hence photometric uncertainties are a blend of uncertainties due to the distance to the guide star and due to the proximity to bright sources.  The incompleteness of the luminosity function measured with Hokupa'a increases faster at fainter magnitudes than the incompleteness observed in the NICMOS LF despite the comparable detection limit in both datasets. As expected, this effect is particularly pronounced in the dense cluster center, where crowding is most severe.  As the Strehl ratio determines the amount of light scattered into the seeing induced halo around each star, a good SR is crucial to achieve not only a diffraction limited spatial resolution, but to benefit from the adaptive optics correction in dense fields containing a wide range of magnitudes. In the Arches dataset, the SR of only 2.5 \\% in $H$ and 7 \\% in $K^\\prime$ as compared to 95 \\% in F160W and 90 \\% in F205W (NICMOS) is clearly the limiting factor for crowded field photometry. As Hokupa'a was initially designed and developed for the 3.6m CFH telescope, its performance at the 8m Gemini telescope is naturally constrained by the limited number of only 36 actuators. In the case of the Arches science demonstration data, additional constraints were given by the seeing, the high airmass due to the low latitude of the Galactic Center, and the guide star magnitude. Under better observing conditions and with a brighter guide star Strehl ratios of up to 30 \\% can be achieved with Hokupa'a at Gemini. Higher order AO systems are currently capable to produce SRs of up to 50\\%.  The Gemini ground-based AO data are comparable to the HST/NICMOS data in the resolution of bright sources ($K^\\prime < 18$ mag), and in  a non-crowded field. They do reach their limitations in the densest cluster area and in the case where faint stars are located close to a bright object. Higher Strehl ratios would of course reduce this unequality.  \\subsection{Photometric results}  A strong colour gradient is detected over the field of the Arches cluster, revealing an increase in visual extinction of approximately $9 < \\Delta A_V < 15$ mag when progressing outwards from the cluster center. The visual extinction is estimated from the Rieke \\& Lebofsky (\\cite{RL85}) extinction law to be $A_V \\sim 24$ mag in the cluster center, increasing to a maximum of $33 < A_V < 39$ mag in the vicinity, in accordance with Cotera et al. (\\cite{Cotera2000}), who found a maximum of $A_V = 37$ mag in the Arches field. Within the central 5\\arcsec\\ radius, however, no colour gradient is observed. This indicates that the cluster center has been stripped of the remaining dust and gas either by strong stellar winds from massive stars or by photo-evaporation, or both. Beyond 5\\arcsec, a linear increase in extinction is observed, suggesting an increasing amount of dust with distance from the cluster center. Photo-evaporation due to the strong UV radiation of the 8 WN7/8-stars and more than 100 O-stars found in the cluster center is most probably responsible for dust dissolution.  The $m160-m205$, $m205$ (equivalent to $H-K$, $K$) colour-magnitude diagrams derived from the HST and Gemini data sets both show a bent main sequence following this colour trend. The main sequence straightens out when correcting for this colour variation. A spatial analysis of the CMDs reveals the bulk of the bright stars on the Arches field to be located in the cluster center.  \\subsection{Mass functions}  Present-day mass functions have been derived from the CMDs after linear correction of the colour trend and corresponding change in extinction over the field, and selection of a reasonable main sequence colour cut. The integrated mass function derived from the Gemini photometry displays a slope of $\\Gamma = -0.8 \\pm 0.15$ for $6\\!<\\!M\\!<\\!65\\ M_\\odot$, less steep than the Salpeter slope of $\\Gamma = -1.35$. This value agrees with the slopes derived from the HST data in the same manner (Sect. \\ref{mfsec}), and with the values presented in FKM within the uncertainties. When the magnitudes are not corrected for differential extinction, the slope of the MF is significantly flatter, $\\Gamma \\sim -0.5 \\pm 0.2$. Particularly in young star forming regions, the effects of differential extinction are thus clearly non-negligible.  The analysis of the radial dependence of the mass function reveals a very flat IMF in the immediate cluster center with a slope close to zero. The IMF slope seems to increase outwards with $\\Gamma = -1.0 \\pm 0.3$ for $5 < R < 10\\arcsec$ and $\\Gamma = -1.7 \\pm 0.7$ beyond $R > 10\\arcsec$. We have created cumulative functions for the stars in each radial bin, and performed a KS test to derive the significance level of the variance in the mass distributions. The probability for the observed distributions to originate in the same mass function is below 1\\% when comparing each two of the three radial bins analysed.  \\subsection{Cluster dynamics}  The flat mass function in the Arches center is a strong indication for mass segregation. While the center seems to be dominated by high-mass stars, the cluster edges display the standard behaviour of a young stellar population. A similar radial dependence of the mass function is observed in the young, compact cluster NGC\\,3603 (Grebel et al. 2002, in prep.), located in a normal star forming environment in the Carina spiral arm. The fact that two out of three compact young clusters found in the Milky Way, which have been analysed in such detail to date, display a flat mass function slope in the core indicates that such a behaviour might be typical for starburst clusters and is not restricted to the extreme GC environment.  Mass segregation in a compact, massive cluster can either be caused by an enhanced production efficiency of massive stars, or by dynamical segregation during the cluster's evolution. We roughly estimate the present-day relaxation time of Arches from mass function considerations to be about a few Myr ($\\sim 20\\,t_{\\rm cross}$), and thus of the same order of magnitude as the cluster age of $\\sim 2$ Myr. Again, a similar timescale has been found for NGC\\,3603 as well (Grebel et al. 2002, in prep.). The comparison with N-body simulations suggests, however, that the dynamical evolution of massive clusters close to the Galactic Center whipes out the initial conditions within less than 1 Myr. We are therefore not able to distinguish between primordial and dynamical mass segregation. Both effects are intertwined at the current state of cluster evolution, such that a detailed dynamical analysis is crucial for a thorough understanding of the formation process. This analysis has to await deep, high resolution infrared spectroscopy to obtain radial velocities and proper motions for a significant fraction of stars belonging to the cluster population."
    },
    "0208/astro-ph0208098_arXiv.txt": {
        "abstract": "We discuss optical colors of 10,592 asteroids with known orbits selected from a sample of 58,000 moving objects observed by the Sloan Digital Sky Survey (SDSS). This is more than ten times larger sample that includes both orbital parameters and multi-band photometric measurements than previously available. We confirm that asteroid dynamical families, defined as clusters in orbital parameter space, also strongly segregate in color space. In particular, we demonstrate that the three major asteroid families (Eos, Koronis, and Themis), together with the Vesta family, represent four main asteroid color types. Their distinctive optical colors indicate that the variations in chemical composition within a family are much smaller than the compositional differences between families, and strongly support earlier suggestions that asteroids belonging to a particular family have a common origin. We estimate that over 90\\% of asteroids belong to families. ",
        "introduction": "Asteroid dynamical families are groups of asteroids in orbital element space (Gradie, Chapman \\& Williams 1979, Gradie, Chapman \\& Tedesco 1989, Valsecchi {\\em et al.} 1989). This clustering was first discovered by Hirayama (1918, for a review see Binzel 1993), who also proposed that families may be the remnants of parent bodies that broke into fragments. About half of all known asteroids are believed to belong to families; recent work (Zappal\\'{a} {\\em et al.} 1995, hereafter Z95), applying a hierarchical clustering method to a sample of 12,487 asteroids, finds over 30 families. The contrast between families and the background is especially strong in the space spanned by the so-called {\\it proper} orbital elements. These elements are nearly invariants of motion and are thus well suited for discovering objects with common dynamical history (Valsecchi {\\em et al.} 1989, Milani \\& Kne\\v{z}evi\\'{c} 1992, hereafter MK92). The current asteroid motion is described by {\\it osculating} orbital elements which vary with time due to perturbations caused by planets, and are less suitable for studying dynamical families.  Asteroid clustering is much weaker in the space spanned by directly observed osculating elements than in the space spanned by derived proper elements. Figure 1 compares the osculating (top panel, Bowell 2001) and proper (bottom panel, MK92) orbital inclination vs. orbital eccentricity distributions of 1,720 asteroids from the outer region of the main asteroid belt (proper semi-major axis larger than 2.84 AU). This region contains all three major asteroid families: Eos, Koronis and Themis, with approximate ($a, \\sin(i), e$) of (3.0, 0.18, 0.08), (2.9, 0.03, 0.05) and (3.15,0.02, 0.15), respectively. Here $a$ is proper semi-major axis, $\\sin(i)$ is the sine of the orbital inclination angle, and $e$ is eccentricity.  The proper elements are derived from the osculating elements by an approximate perturbation method (MK92), and it is possible that the overdensities evident in the bottom panel are at least partially created by that algorithm (Valsecchi {\\em et al.} 1989, Bendjoya 1993). A firm proof that families are real therefore requires their confirmation by a method that is {\\it not} based on dynamical considerations, for example, that dynamically selected groups have distinctive colors. While there is observational evidence that at least the most populous asteroid families have characteristic colors (Degewij, Gradie \\& Zellner 1978, Chapman 1989), even the most recent studies of the colors of asteroid families include fewer than 50 objects per family (Florczak {\\it et al.} 1998, Doressoundiram {\\it et al.} 1998, Florczak {\\it et al.} 1999). The large number (about 10,000) of color measurements for catalogued asteroids (Bowell 2001) recently made available by the Sloan Digital Sky Survey (SDSS, York {\\it et al.} 2000) allows a detailed investigation of this question. ",
        "conclusions": "A striking feature of Figures 3, 4 and 5 is the color homogeneity and distinctiveness displayed by asteroid families. Each of the three major Hirayama families, Eos, Koronis and Themis, and also the Vesta family at ($a, \\sin(i), e$) of (2.35, 0.12, 0.09), has a characteristic color. This strong color segregation provides firm support for the reality of asteroid dynamical families. The correlation between the asteroid colors and their heliocentric distance has been recognized since the earliest development of asteroid taxonomies (Chapman, Morrison \\& Zellner 1975, Gradie \\& Tedesco 1982, Zellner, Tholen \\& Tedesco 1985, Gradie, Chapman \\& Tedesco 1989). Our analysis indicates that this mean correlation (see e.g. Figure 23 in I01) is mostly a reflection of the distinctive colors of asteroid families and their heliocentric distribution.  When only orbital elements are considered, families often partially overlap each other (Z95), and additional independent information is needed to improve their definitions. With such a massive, accurate and public database as that discussed here (SDSSMOC), it will be possible to improve the classification of asteroid families by simultaneously using both the orbital elements and colors. For example, the SDSS colors show that the asteroids with ($a, \\sin(i)$) about (2.65, 0.20) are distinctively blue (Figure 3), proving that they do not belong to the family with ($a, \\sin(i)$) about (2.60, 0.23), but instead are a family in their own right. While this and several similar examples were already recognized as clusters in the orbital parameter space (Z95), this work provides a dramatic independent confirmation. Figures 3, 4 and 5 suggest that the asteroid population is dominated by families: even objects that do not belong to the most populous families, and thus are interpreted as background in dynamical studies, seem to show color clustering. Using the definitions of families based on dynamical analysis (Z95), and aided by SDSS colors, we estimate that at least 90\\% of asteroids are associated with families\\footnote{The preliminary analysis indicates that about 1--5\\% of objects do not belong to families. A more detailed discussion of the robustness of this result will be presented in a forthcoming publication. Similarly, it is not certain yet whether objects not associated with the families show any heliocentric color gradient.}.  Proper orbital elements (MK92) are not available for asteroids with large semi-major axis and orbital inclination. In order to examine the color distribution for objects with large semi-major axis, such as Trojan asteroids ($a\\sim 5.2$) and for objects with large inclination, such as asteroids from the Hungaria family ($a\\sim 1.9, \\sin(i)\\sim 0.38$), we use osculating orbital elements. Figure 6 shows the distribution of all the 10,592 known asteroids observed by the SDSS in the space spanned by osculating semi-major axis and the sine of the orbital inclination angle, with the points color-coded as in Figure 2. It is remarkable that various families can still be easily recognized due to SDSS color information. This figure vividly demonstrates that the asteroid population is dominated by objects that belong to numerous asteroid families.     \\vskip 0.4in \\leftline"
    },
    "0208/astro-ph0208051_arXiv.txt": {
        "abstract": "We present the first results of our 3D Monte Carlo maser radiative transfer code, used to model the 1612 MHz OH maser shell and the amplification of emission from the stellar radio-photosphere. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0208/astro-ph0208267_arXiv.txt": {
        "abstract": "We give a thorough investigation of sequences of uniformly rotating, homogeneous axisymmetric Newtonian equilibrium configurations that bifurcate from highly flattened Maclaurin spheroids. Each one of these sequences possesses a mass-shedding limit. Starting at this point, the sequences proceed towards the Maclaurin sequence and beyond. The first sequence leads to the well known Dyson rings, whereas the end points of the higher sequences are characterized by the formation of a two-body system, either a core-ring system (for the second, the fourth etc.~sequence) or a two-ring system (for the third, the fifth etc.~sequence). Although the general qualitative picture drawn by Eriguchi and Hachisu in the eighties has been confirmed, slight differences turned out in the interpretation of the origin of the first two-ring sequence and in the general appearance of fluid bodies belonging to higher sequences. ",
        "introduction": "If one moves along the Maclaurin sequence of uniformly rotating, axisymmetric, homogeneous fluid ellipsoids with fixed mass and fixed density, starting at the non-rotating configuration and proceeding towards increasing angular momentum, one first encounters the Jacobi branch point where the ellipsoids become secularly unstable with regard to the first non-axisymmetric perturbation (see e.g.~Chandrasekhar 1969). The corresponding Jacobi sequence branching off at this point leads to further bifurcations, in particular to Poincar\\'e's `pear-shaped' sequence (see Eriguchi \\& Hachisu 1982b). Farther along the Maclaurin path, one next comes to the bifurcation points of the non-axisymmetric `triangle', `square' and `ammonite' sequences (Chandrasekhar 1969, Eriguchi \\& Hachisu 1982a), before one arrives at the first axisymmetric sequence bifurcating at an eccentricity $\\varepsilon_1=0.98523$ (Chandrasekhar 1967, Bardeen 1971). As conjectured by Bardeen (1971) and confirmed by Eriguchi \\& Sugimoto (1981), the bodies of this sequence pinch together gradually at the centre and eventually form the anchor-ring configurations studied by Dyson (1892, 1893) and Wong (1974); see also Poincar\\'e (1885), Kowalewsky (1895), Lichtenstein (1933) and a somewhat related paper by Kley (1996).  Another not yet confirmed conjecture by Bardeen (1971) concerns a sequence of  axisymmetric `central-bulge' configurations likewise bifurcating at $\\varepsilon_1$. In this paper we indeed found this sequence, which might be considered to be a continuation of the Dyson-ring sequence beyond the Maclaurin ellipsoids and ends in a mass-shedding limit. However, the surface shapes of the corresponding fluid bodies do not generate a `central-bulge' region; their appearance is more `lens shaped'.  Apart from the study of the Dyson-ring sequence we shall give a detailed analysis of the next axisymmetric sequences bifurcating from the Maclaurin ellipsoids. The first core-ring sequence branches off at $\\varepsilon_2=0.99375$. Starting at a mass-shedding limit, this sequence proceeds towards the Maclaurin sequence and beyond, finally leading to the formation of a core-ring system.  In contrast to the results by Eriguchi \\& Hachisu (1982a) who stated that the first two-ring sequence branches off at $\\varepsilon_2$, we found that the bifurcation occurs at $\\varepsilon_3=0.99657$. Again, a mass-shedding limit marks one end point of this sequence, leading from here towards and beyond the Maclaurin sequence and ending in the formation of a two-ring system.  The same qualitative picture repeats as one moves to the higher axisymmetric bifurcation points at $\\varepsilon_4=0.99784, \\varepsilon_5=0.99851$ etc., of which there are infinitely many, accumulating at $\\varepsilon=1$ (Bardeen 1971). Starting at a mass-shedding limit, the sequences proceed towards the Maclaurin sequence and beyond, leading eventually to the formation of a core-ring system for $\\varepsilon_{2l}$ and a two-ring system for $\\varepsilon_{2l+1}$, see Figs 7-11. The body's surface is characterized by a particular number of grooves, notably $l$, for configurations close to the two-body systems that are the end-stages of the sequences bifurcating at $\\varepsilon_{2l}$ and $\\varepsilon_{2l+1}$. The outermost groove pinches together first and an outer ring without grooves separates. In some continuation process beyond this formation of a two-body system, the other grooves might also pinch off, leading eventually to a multi-body system, at most consisting of a core and $l$ rings or $(l+1)$ rings respectively\\footnote{This result is in contrast to the statement by Eriguchi \\& Hachisu (1982a), who expected the formation of $k$ rings for the $\\varepsilon_k$-sequence.}. However, since a continuation process of this kind is not unique, we consider the sequences in question precisely up to the formation of the two-body system.  These investigations have been carried out up to the $\\varepsilon_{10}$-sequence. The corresponding final configurations, when the two-body system is formed, can be seen in Figs 7-11.  Representative plots of physical quantities as well as of meridional cross-sections have been provided up to the $\\varepsilon_{5}$-sequence, see Figs 1-10. Note that the Maclaurin sequence becomes dynamically unstable with respect to axisymmetric perturbations for $\\varepsilon>0.99856$ (Bardeen 1971, Eriguchi \\& Hachisu 1985). Therefore, if we restrict our considerations to axial symmetry, the sequences bifurcating at $\\varepsilon_1,\\varepsilon_2,\\ldots,\\varepsilon_5$ are more relevant than those branching off at $\\varepsilon_6,\\varepsilon_7,\\ldots\\, (\\varepsilon_k\\geq\\varepsilon_6=0.99891$ for $k>5)$.  Table 1 lists physical quantities for the Maclaurin ellipsoids at the bifurcation points $\\varepsilon_k$ for $k=1,2,\\ldots,10$. Tables 2-4 contain numerical data with an accuracy of five digits for configurations of the $\\varepsilon_{1}$, $\\varepsilon_{2}$ and $\\varepsilon_{3}$-sequences. Additionally, Table 5 lists the ratios of the masses of the inner to the outer body at the two-body formation point for the $\\varepsilon_2\\ldots\\varepsilon_5$-sequences.  \\begin{table}{{\\bf Table 1:} Physical quantities for the Maclaurin ellipsoids at the bifurcation points $\\varepsilon_k$. For the definition of $\\omega^2$, $j^2$, $T$ and $W$ see Section 4.} \\begin{center} \\begin{tabular}[b]{ccccccl} \\hline\\hline $k$ & $\\varepsilon_k$ & Axis ratio & $\\omega^2$ & $j^2$    & $T/|W|$  \\\\ \\hline 1 & 0.98523 & 0.17126 & 0.087262 & 0.021741 & 0.35890 \\\\ 2 & 0.99375 & 0.11160 & 0.066105 & 0.029152 & 0.40345 \\\\ 3 & 0.99657 & 0.082750 & 0.052711 & 0.034638 & 0.42664 \\\\ 4 & 0.99784 & 0.065744 & 0.043714 & 0.039037 & 0.44084 \\\\ 5 & 0.99851 & 0.054534 & 0.037301 & 0.042740 & 0.45044 \\\\ 6 & 0.99891 & 0.046589 & 0.032513 & 0.045957 & 0.45736 \\\\ 7 & 0.99917 & 0.040664 & 0.028806 & 0.048814 & 0.46259 \\\\ 8 & 0.99935 & 0.036075 & 0.025854 & 0.051395 & 0.46667  \\\\ 9 & 0.99947 & 0.032417 & 0.023449 & 0.053755 & 0.46995 \\\\ 10 & 0.99957 & 0.029432 & 0.021451 & 0.055935 & 0.47265 \\end{tabular} \\end{center} \\end{table}  \\begin{figure*}[h] \\unitlength1cm \\hspace*{-2cm} \\epsfig{file=fig1.eps,scale=1} \\vspace*{-17cm} \\caption{For the first five axisymmetric sequences $(1), \\ldots,(5)\\,,$ the squared angular velocity $\\omega^2=\\Omega^2/(4\\pi G\\mu)$ is plotted against the radius ratio $r_1/\\rho_2$, where $r_1=\\zeta_1$ ($\\zeta_1:$ polar radius) for spheroidal figures and $r_1=-\\rho_1$ ($\\rho_1:$ inner equatorial radius) for toroidal shapes; $\\rho_2$ is always the radius of the outer equatorial rim. The dotted curve represents the Maclaurin sequence and the dashed one corresponds to the Dyson approximation.} \\end{figure*}  \\begin{figure*}[h] \\unitlength1cm \\hspace*{-2cm} \\epsfig{file=fig2.eps,scale=1} \\vspace*{-17cm} \\caption{For the first five axisymmetric sequences, the squared angular velocity $\\omega^2=\\Omega^2/(4\\pi G\\mu)$ is plotted against the dimensionless squared angular momentum $j^2$, given in formula (\\ref{j2}). Dotted and dashed curves again refer to the Maclaurin sequence and the Dyson approximation respectively. The $\\bullet$'s mark the bifurcation points on the Maclaurin sequence and $\\Box$ the transition configuration of spheroidal to toroidal bodies on the Dyson-ring sequence.} \\end{figure*} ",
        "conclusions": ""
    },
    "0208/astro-ph0208273_arXiv.txt": {
        "abstract": "{\\small We propose a scenario for a periodic filling and emptying of the accretion disc of GRS 1915+105, by computing the mass transfer rate from the donor and comparing it with the observed accretion rate (see the full paper \\cite{Vil01}). The binary parameters found by  \\cite{Gre01} predict evolutionary expansion of the donor along the giant branch with a conservative mass transfer rate (1 -- 2) $\\times$ 10$^{-8}$ \\msunyr. This reservoir can support the present observed accretion rate with a duty cycle 0.05 -- 0.1 (the active time as a fraction of the total life time). The viscosity time scale at the circularization radius (15 solar radii from the primary 14 \\msun black hole) is identified as the recurrent quiescent time during which a new disc is formed once consumed by the  BH. For small viscosity ($\\alpha$ = 0.001) it equals to 300 -- 400 years. The microquasar phase, with the duty cycle, will last around 10$^7$ years ending with a long-period black hole + white dwarf system. } ",
        "introduction": " ",
        "conclusions": ""
    },
    "0208/astro-ph0208045_arXiv.txt": {
        "abstract": "In popular cold dark matter cosmological scenarios, stars may have first appeared in significant numbers around a redshift of 10 or so, as the gas within protogalactic halos with virial temperatures $T_\\vir\\gta 10^{4.3}\\,$K (corresponding to masses comparable to those of present--day dwarf ellipticals) cooled rapidly due to atomic processes and fragmented. It is this `second generation' of subgalactic stellar systems, aided perhaps by an early population of accreting black holes in their nuclei, which may have generated the ultraviolet radiation and mechanical energy that ended the cosmic ``dark ages'' and reheated and reionized most of the hydrogen in the universe by a redshift of 6. The detailed history of the universe during and soon after these crucial formative stages depends on the power--spectrum of density fluctuations on small scales and on a complex network of poorly understood `feedback' mechanisms, and is one of the missing link in galaxy formation and evolution studies. The astrophysics of the epoch of first light is recorded in the thermal state, ionization degree, and chemical composition of the intergalactic medium, the main repository of baryons at high redshifts. ",
        "introduction": "At epochs corresponding to $z\\sim 1000$ the intergalactic medium (IGM) is expected to recombine and remain neutral until sources of radiation and heat develop that are capable of reionizing it. The detection of transmitted flux shortward of the \\lya wavelength in the spectra of sources at $z\\gta 5$ implies that the hydrogen component of this IGM was ionized at even higher redshifts. The increasing thickening of the \\Lya forest recently measured in the spectra of SDSS $z\\sim 6$ quasars (Becker \\etal 2001; Djorgovski \\etal 2001) may be the signature of the trailing edge of the cosmic reionization epoch. It is clear that substantial sources of ultraviolet photons and mechanical energy, like young star--forming galaxies, were already present back then. The reionization of intergalactic hydrogen at $z\\gta 6$ is unlikely to have been accomplished by quasi--stellar sources: the observed dearth of luminous optical and radio--selected QSOs at $z>3$ (Shaver \\etal 1996; Fan \\etal 2001), together with the detection of substantial Lyman--continuum flux in a composite spectrum of Lyman--break galaxies at $\\langle z\\rangle=3.4$ (Steidel et al. 2001), both may support the idea that massive stars in galactic and subgalactic systems -- rather than quasars -- reionized the hydrogen component of the IGM when the universe was less than 5\\% of its current age, and dominate the 1 ryd metagalactic flux at all redshifts greater than 3. An episode of pregalactic star formation may also provide a possible explanation for the widespread existence of heavy elements (like carbon, oxygen, and silicon) in the IGM. There is mounting evidence that the double reionization of helium may have occurred later, at a redshift of 3 or so (see Kriss et al. 2001, and references therein): this is likely due to the integrated radiation emitted above 4 ryd by QSOs.  Establishing what ended the dark ages and when is important for determining the impact of cosmological reionization and reheating on several key cosmological issues, from the role reionization plays in allowing protogalactic objects to cool and make stars, to determining the thermal state of baryons at high redshifts and the small--scale structure in the temperature fluctuations of the cosmic microwave background. Conversely, probing the reionization epoch may provide a means for constraining competing models for the formation of cosmic structures: for example, popular modifications of the CDM paradigm that attempt to improve over CDM by suppressing the primordial power--spectrum on small scales, like warm dark matter (WDM), are known to reduce the number of collapsed halos at high redshifts and make it more difficult to reionize the universe (Barkana \\etal 2001). In this talk I will summarize some recent theoretical developments in understanding the astrophysics of the epoch of first light and the impact that some of the earliest generations of stars, galaxies, and black holes in the universe may have had on the IGM. ",
        "conclusions": ""
    },
    "0208/astro-ph0208335_arXiv.txt": {
        "abstract": "Four evolutionary channels leading to the formation of wide binary millisecond pulsars are investigated. The majority of binary millisecond pulsars are found to descend from systems in which the most massive component undergoes a common-envelope phase prior to the supernova explosion leading to the birth of the neutron star.  The orbital period distribution of simulated samples of wide binary millisecond pulsars is compared with the observed distribution of Galactic binary millisecond pulsars for a variety of parameters describing the formation and evolution of binaries.  The distribution functions typically show a short-period peak below 10 days and a long-period peak around 100 days. The observed distribution is best reproduced by models with highly non-conservative mass transfer, common-envelope efficiencies equal to or larger than unity, a critical mass ratio for the delayed dynamical instability larger than 3, and no or moderate supernova kicks at the birth of the neutron star. Few systems are found with orbital periods longer than 200 days, irrespective of the accretion efficiency of neutron stars. This occurs as a result of the upper limit on the initial orbital periods beyond which the binary avoids the common-envelope phase prior to the supernova explosion of the primary. ",
        "introduction": "In close binaries containing a neutron star, the evolution of the system is driven by the nuclear evolution of the companion and the loss of angular momentum through processes as magnetic braking and gravitational radiation. When the combination of these processes causes the companion to fill its Roche lobe, mass and angular momentum are transferred to the neutron star which is then spun up to form a recycled pulsar, unless transient behaviour prevents effective accretion. At the end of the mass-transfer phase, the binary emerges as a binary millisecond pulsar (BMSP) consisting of a rapidly rotating neutron star in a nearly circular orbit around a low-mass white dwarf, which used to be the core of the Roche-lobe overflowing companion (for a review see Bhattacharya \\& van den Heuvel 1991, Phinney \\& Kulkarni 1994). If the mass transfer phase takes place during the companion's ascent of the red giant branch, the relation between the core mass and the radius of the star and the relation between the Roche-lobe radius and the orbital separation, lead to a correlation between the orbital period of the BMSP and the mass of the white dwarf. This correlation has been studied extensively by Joss, Rappaport \\& Lewis (1987), Savonije (1987), Rappaport et al. (1995), Ritter (1999), and Tauris \\& Savonije (1999).  The precise evolutionary path followed by a BMSP progenitor depends mainly on the orbital period and on the mass of the donor star at the onset of Roche-lobe overflow after the supernova explosion of the primary. Systems with orbital periods longer than $\\sim 20$ days and donors more massive than $\\sim 2.0\\,M_\\odot$ are likely to evolve through a common-envelope phase during which the envelope of the donor is expelled and the neutron star spirals in towards the donor star's core (Kalogera \\& Webbink 1998). If the orbital period is shorter, the common-envelope phase can be avoided for donor stars up to $\\sim 4.0\\,M_\\odot$ (Tauris, van den Heuvel \\& Savonije 2000; Kolb et al. 2000; Podsiadlowski, Rappaport \\& Pfahl 2002). Binaries with donor stars less massive than $\\sim 2.0\\,M_\\odot$ will become low-mass X-ray binaries, for which the evolution depends on whether the orbital period is longer or shorter than the bifurcation period separating diverging from converging systems (Pylyser \\& Savonije 1988, 1989). Only low-mass X-ray binaries with orbital periods longer than the bifurcation period will eventually evolve into wide BMSPs.  At present, 33 BMSPs with orbital periods longer than one day have been found in the Galactic disc. Their orbital periods, eccentricities, and white dwarf masses are listed in Table~\\ref{bmsp} (Ritter, private communication). Inspection of the observed orbital periods shows a paucity of systems with periods between 30 and 60 days, and an absence of systems with periods longer than 200 days. A theoretical interpretation of the possible period gap was given by Tauris (1996) and Taam, King \\& Ritter (2000), while an explanation for the upper limit was put forward by Ritter \\& King (2002).  \\begin{table} \\caption{Observed orbital periods, eccentricities, and white dwarf masses of Galactic BMSPs with orbital periods longer than 1 day. The data is derived from Taam et al. (2000) with updated information from Ritter (private communication).} \\label{bmsp} \\begin{tabular}{lr@{.}lr@{.}lr@{.}l} \\hline Name   & \\multicolumn{2}{c}{$P_{\\rm orb}$ (days)} &  \\multicolumn{2}{c}{$e$} & \\multicolumn{2}{c}{$M_{\\rm WD}/M_\\odot$} \\\\ \\hline J0613-0200  &    1&19851 &  0&000007 &  $>$0&13  \\\\ J1435-6100  &    1&35489 &  0&000010 &  $>$0&90  \\\\ J0034-0534  &    1&58928 & $<$0&00002  & $>$0&15  \\\\ J0218+4232  &    2&02885 & $<$0&00002  & $>$0&16  \\\\ J2317+1439  &    2&45933 & 0&0000005& \\multicolumn{2}{c}{} \\\\ J1911-1114  &    2&71656 & $<$0&000013 & $>$0&12 \\\\ J1157-5112  &    3&50739 &  0&000402 &  $>$1&18  \\\\ J1045-4509  &    4&08353 &  0&000024 &  $>$0&16  \\\\ J1745-0952  &    4&94345 &  0&000018 & $>$0&11   \\\\ J1732-5049  &    5&26300 &  0&000001 &  0&18       \\\\ J0437-4715  &    5&74104 &  0&000019 &  \\multicolumn{2}{c}{0.22-0.32} \\\\ J1603-7202  &    6&30863 & $<$0&00002  & $>$0&29  \\\\ J2129-5721  &    6&62549 & $<$0&000017 & $>$0&14 \\\\ J2145-0750  &    6&83890 & 0&000018 & $>$0&43 \\\\ J1022+1001  &    7&80513 &  0&000098 &  $>$0&73  \\\\ J0621+1002  &    8&31868 &  0&002458 &  $>$0&45  \\\\ J1518+4904  &    8&63400 &  0&249485 &  \\multicolumn{2}{c}{}  \\\\ J1918-0642  &   10&91318 &  0&000022 & $>$0&24   \\\\ J1804-2717  &   11&12871 &  0&000035 & $>$0&21   \\\\ B1855+09    &   12&32717 &  0&000027 &  \\multicolumn{2}{c}{0.24-0.29} \\\\ J1454-5846  &   12&42307 &  0&001898 &  $>$0&87  \\\\ J1810-2005  &   15&01220 &  0&000025 & $>$0&28   \\\\ J1709+23    &   22&7    & \\multicolumn{2}{c}{} & \\multicolumn{2}{c}{} \\\\ J1618-3919  &   22&8     & \\multicolumn{2}{c}{} & \\multicolumn{2}{c}{} \\\\ J2033+17    &   56&2     & $<$0&05     & $>$0&2  \\\\ J1713+0747  &   67&82513 &  0&000075 & $>$0&27   \\\\ J1455-3330  &   76&17458 &  0&000167 &  $>$0&27  \\\\ J2019+2425  &   76&51163 &  0&000111 &  0&33       \\\\ J2229+2643  &   93&01589 & 0&000256 & \\multicolumn{2}{c}{} \\\\ J1643-1224  &  147&01740 &  0&000506 & $>$0&13   \\\\ B1953+29    &  117&34910 &  0&000330 & \\multicolumn{2}{c}{}  \\\\ J1640+2224  &  175&46066 &  0&000797 & \\multicolumn{2}{c}{} \\\\ \\hline \\end{tabular} \\end{table}  With this paper, we initiate a systematic study based on population synthesis techniques to investigate the different evolutionary channels leading to the formation of wide BMSPs and to compare the theoretical with the observed orbital period distribution of wide BMSPs. We particularly investigate whether a set of standard evolutionary parameters can be found to reproduce the observed orbital period distribution without including observational selection effects or pulsar lifetime issues. Due to the large uncertainties involved in the determination of the masses of the white dwarfs in the observed BMSPs (see Table~\\ref{bmsp}) we presently do not consider the white dwarf mass distribution as good a tool as the orbital period distribution to confine stellar and binary evolution and formation parameters.  Since BMSPs located in globular clusters are thought to be formed through tidal encounters of neutron stars with primordial binaries (e.g. Rappaport, Putney \\& Verbunt 1989), we limit ourselves to BMSPs in the Galactic disc. We also leave aside systems with orbital periods shorter than 1 day since they are likely to follow evolutionary paths different from those of the BMSPs investigated here. In particular, the evolution of the short-period systems is dominated by angular momentum losses for which quantitative descriptions are still uncertain (Ergma \\& Sarna 1996; Kalogera, Kolb \\& King 1998; Ergma, Sarna \\& Antipova 1998; Rasio, Pfahl \\& Rappaport 2000; Dewi et al. 2002; and in particular Podsiadlowski et al. 2002 and references therein).  The plan of the paper is as follows. In Sect.~2, we briefly summarise the basic concepts of the binary evolution code used in this investigation.  In Sect.~3, the evolutionary channels leading to the formation of wide BMSPs are discussed. In Sect.~4, we explore the parameter space occupied by the BMSP progenitors during various phases of their evolution. The effects of the input parameters adopted in our binary evolution calculations on the orbital period distribution of wide BMSPs and the relative contributions of the different formation channels to the total population of wide BMSPs are investigated in Sect.~5. The final section is devoted to concluding remarks. ",
        "conclusions": "We have investigated the formation of wide BMSPs through four evolutionary channels using a rapid binary evolution code based on the analytical approximation formulae for the evolution of single stars derived by Hurley et al. (2000).  In three of the channels, dynamically unstable mass transfer from the primary prior to the supernova explosion results in a common-envelope phase during which the orbital separation is reduced and the helium core of the primary is exposed as a naked helium star. The primary continues its evolution until it becomes a neutron star in a type Ib/c supernova explosion. The further evolution depends on the mass of the secondary and on the post-supernova orbital period after circularisation:  \\begin{itemize} \\item If the mass of the secondary is smaller than $\\sim\\! 1.6\\, M_\\odot$, the binary undergoes a stable case~B mass-transfer phase when the secondary reaches the giant branch. The binary becomes a BMSP consisting of a rapidly rotating neutron star orbiting a helium white dwarf with an orbital period determined by the mass of the white dwarf. \\item If the mass of the secondary is larger than $\\sim\\! 1.6\\, M_\\odot$ and the orbital period shorter than $\\sim\\! 2.5$ days, the binary undergoes a thermal-timescale case~A mass-transfer phase, followed by a stable phase of case~A and/or case~B Roche-lobe overflow. The system again evolves into a BMSP containing a helium white dwarf whose mass determines the orbital period. \\item If the mass of the secondary is larger than $\\sim\\! 1.6\\, M_\\odot$ and the orbital period longer than $\\sim\\! 2.5$ days, the binary undergoes a thermal-timescale case~B mass-transfer phase when the secondary crosses the Hertzsprung gap. The system evolves into a binary containing a neutron star and a carbon/oxygen white dwarf without developing a tight correlation between the orbital period and the mass of the white dwarf. The accretion onto the neutron star in these binaries may not be efficient enough to spin them up to millisecond pulsars. \\end{itemize}  The fourth evolutionary channel corresponds to the direct supernova mechanism where no interaction between the binary components takes place until the primary gives birth to a neutron star in a type II supernova explosion on the AGB. If the resulting post-supernova orbit is wide enough to avoid any interaction until the secondary becomes an AGB star, stable case~C mass transfer leads to the formation of a BMSP containing a carbon/oxygen white dwarf whose mass is correlated with the orbital period.  Our main goal in this investigation was to compare the orbital period distribution of simulated samples of BMSPs resulting from different sets of input parameters for the binary evolution calculations with the orbital period distribution of wide BMSPs observed in the Galactic disc. The observed long orbital period peak is identified with the BMSPs containing a helium white dwarf whose progenitor underwent a mass-transfer phase that was stable at all times. The observed short orbital period peak on the other hand is identified with the BMSPs containing a helium white dwarf whose progenitor evolved through a thermal-timescale mass-transfer phase on the main sequence or in the Hertzsprung gap.  The simulated distribution functions all show a rapid decrease for orbital periods longer than 200 days, irrespective of the accretion efficiency of neutron stars. The lack of longer period systems is a consequence of the upper limit on the initial orbital periods beyond which the binary remains detached instead of going through a common-envelope phase prior to the supernova explosion of the primary. The agreement between the simulated and the observed orbital period distributions is best for models with highly non-conservative mass transfer, common-envelope efficiencies equal to or larger than unity, a critical mass ratio for the delayed dynamical instability larger than 3, and no or moderate supernova kicks at the birth of the neutron star.  The first results of the population synthesis study presented here are encouraging, but many uncertainties remain to be solved. The lack of BMSPs containing carbon/oxygen white dwarfs at orbital periods shorter than 20 days in our 'best-fit' solution, for instance, surely poses a problem that needs to be resolved. A more detailed treatment of the accretion process onto neutron stars and the potential effects of pulsar turn-on, evaporation, and spin-down are of particular importance to improve the population synthesis study of wide BMSPs. Preliminary calculations also reveal a dependency of the orbital period distribution on the mass-loss rates from stellar winds, but the results are still inconclusive.  Furthermore, our population synthesis study implicitly incorporates a model for the Galactic population of neutron star X-ray binaries. The long-standing apparent conflict between the observationally implied birth rate of BMSPs and their LMXB progenitors (e.g. Ruderman, Shaham \\& Tavani 1989; Podsiadlowski et al. 2002 and references therein) may point towards non-standard evolutionary effects in the X-ray binary phase which are not considered in our study. It is therefore desirable to consider the populations of LMXBs and BMSPs simultaneously.  We will address these matters in more detail in future investigations."
    },
    "0208/astro-ph0208429_arXiv.txt": {
        "abstract": "Jet-driven shocks are responsible for an important fraction of the emission of the narrow-line regions (NLRs) in many classes of AGN. However, this cannot explain all observations. It is clear that the remaining sources are photoionised by the active nucleus. The 2-d hydrodynamic models from the RSAA group support an evolutionary scenario whereby the shock-excited NLRs are initially jet-driven but later, ionizing photons from the central engine replace shocks as the main excitation mechanism and shock induced star formation may also become important. In their photoionized phase, dusty and radiation-pressure dominated evolution produces a self-regulated NLR spectrum. This model aso explains the coronal emission lines and fast (3000 km~s$^{-1}$) outflows seen in some Seyferts. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0208/astro-ph0208103_arXiv.txt": {
        "abstract": "Inflationary cosmology has become one of the cornerstones of modern cosmology. Inflation was the first theory within which it was possible to make predictions about the structure of the Universe on large scales, based on causal physics. The development of the inflationary Universe scenario has opened up a new and extremely promising avenue for connecting fundamental physics with experiment. This article summarizes the principles of inflationary cosmology, discusses progress in the field, focusing in particular on the mechanism by which initial quantum vacuum fluctuations develop into the seeds for the large-scale structure in the Universe, and highlights the important unsolved problems of the scenario. The case is made that new input from fundamental physics is needed in order to solve these problems, and that thus early Universe cosmology can become the testing ground for trans-Planckian physics. ",
        "introduction": "With the recent high-accuracy measurements of the spectrum of the cosmic microwave background (CMB) (see Fig. 1 \\cite{Scott}), cosmology has become a quantitative science. There is now a wealth of new data on the structure of the Universe as deduced from precision maps of the cosmic microwave background anisotropies, from cosmological redshift surveys, from redshift-magnitude diagrams of supernovae, and from many other sources. Standard Big Bang (SBB) cosmology provides the framework for describing the present data. The interpretation and explanation of the existing data, however, requires us to go beyond SBB cosmology and to consider scenarios like the Inflationary Universe in which space-time evolution in the very early Universe differs in crucial ways from what is predicted by the SBB theory. Since inflationary cosmology at later times smoothly connects with the SBB picture, we must begin this article with a short review of the framework of SBB cosmology.  Standard big bang cosmology rests on three theoretical pillars: the cosmological principle, Einstein's general theory of relativity, and the assumption that matter is a classical perfect fluid. The cosmological principle concerns the symmetry of space-time and states that on large distance scales space is homogeneous and isotropic. This implies that the metric of space-time can be written in Friedmann-Robertson-Walker (FRW) form. For simplicity, we consider the case of a spatially flat Universe: \\begin{equation} ds^2 \\, = \\, dt^2 - a(t)^2 \\left[dr^2 + r^2 (d \\vartheta^2 + \\sin^2 \\vartheta d\\varphi^2) \\right] \\, . \\end{equation} Here, $t, r, \\theta$ and $\\varphi$ are the space-time coordinates, and $ds$ gives the proper time between events in space-time. The coordinates $r, \\vartheta$ and $\\varphi$ are ``comoving'' spherical coordinates, and $t$ is the physical time coordinate. Space-time curves with constant comoving coordinates correspond to the trajectories of particles at rest. If the Universe is expanding, i.e. the scale factor $a(t)$ is increasing, then the physical distance $\\Delta x_p(t)$ between two points at rest with fixed comoving distance $\\Delta x_c$ grows as $\\Delta x_p = a(t) \\Delta x_c$.  \\begin{figure}[htb] \\begin{center} \\includegraphics[width=7.5cm,angle=+90]{ispectrum_01.ps} \\caption{Compilation of the spectrum of the CMB. In the region around the peak (the region probed with greatest precision by the COBE satellite) the error bars are smaller than the size of the data points.} \\end{center} \\end{figure}  The dynamics of an expanding Universe  is determined by the Einstein equations, which relate the expansion  rate to the matter content, specifically to the energy density $\\rho$ and pressure $p$.  For a Universe obeying the cosmological principle (and neglecting the possible presence of a cosmological constant) the Einstein equations reduce to the Friedmann-Robertson-Walker (FRW) equations \\begin{eqnarray} \\label{FRW} \\left( {\\dot a \\over a} \\right)^2 - {k\\over a^2} \\, &=& \\, {8 \\pi G\\over 3 } \\rho \\\\ \\dot \\rho \\, &=& \\, - 3 H (\\rho + p) \\, , \\end{eqnarray} where $H = {\\dot a}/a$ is the Hubble expansion rate, and an overdot denotes the derivative with respect to time $t$.  The third key assumption of standard cosmology is that matter is described by a classical ideal gas with some equation of state which is conveniently parametrized in the form $p = w \\rho$ which some constant $w$. For cold matter, pressure is negligible and hence $w = 0$. For radiation we have $w = {1/3}$. In standard cosmology, the Universe is a mixture of cold matter and radiation, the former dominating at late times, the latter dominating in the early Universe. In this case, the FRW equations can be solved exactly, with the result that the Universe had to be born in a ``Big Bang'' singularity.   SBB cosmology explains Hubble's redshift-distance relationship, and it explains the abundances of the light elements Helium, Deuterium and Lithium. These nucleosynthesis predictions of the SBB model depend on a single free parameter (which is the present ratio of the number densities of baryons to photons), and thus the fact that the abundances of more than one element can be matched by adjusting this ratio is remarkable. Most importantly, SBB cosmology predicts the existence and black-body spectral distribution of a microwave cosmic background radiation, the CMB. The precision measurement \\cite{Gush,COBE} of the spectrum of the CMB is a triumph for SBB cosmology (see Fig. 1).  \\begin{figure} % \\begin{center} \\leavevmode \\epsfysize=4.5cm \\epsfbox{dfig1.eps} \\caption{ A space-time diagram (physical distance $x_p$ versus time $t$) illustrating the homogeneity problem: the past light cone $\\ell_p (t)$ at the time $t_{rec}$ of last scattering  is much larger than the forward light cone $\\ell_f (t)$ at $t_{rec}$.} \\end{center} \\end{figure}  However, the triumph of the CMB also leads to one of several major problems for SBB cosmology: within the context of this theory, there is no explanation for the high degree of isotropy of the radiation. As is illustrated in Fig. 2, at the time the microwave radiation last scattered (which occurred when the temperature was about a factor $10^3$ of the present temperature), the maximal distance which light could have communicated information starting at the Big Bang (the forward light cone) is much smaller than the distance over which the microwave photons are observed to have the same temperature (the past light cone). This is the famous ``horizon problem'' of SBB cosmology. Within the context of SBB cosmology it is also a mystery why the Universe today is observed to be approximately spatially flat, since a spatially flat Universe is an unstable fixed point of the FRW equations in an expanding phase. This problem is called the ``flatness problem''. Finally, as illustrated in Fig. 3, within standard cosmology there is no causal mechanism which can explain the nonrandom distribution of the seed inhomogeneities which develop into the present-day large-scale structure. This constitutes the ``formation of structure problem''. Under the assumption that only gravity is responsible for the development of inhomogeneities on cosmological scales, the seeds for fluctuations must have been correlated at the time $t_{eq}$, the time when the energy densities in cold matter and radiation were equal (which occurred when the Universe was about $10^{-4}$ of its present size) when inhomogeneities can first start to grow by gravitational instability. But, at that time, the forward light cone was smaller than the separation between the seeds.  \\begin{figure} \\begin{center} \\leavevmode \\epsfysize=5.0cm \\epsfbox{dfig2.eps} \\caption{A sketch (conformal separation vs. time) of the formation of structure problem: the comoving separation $d_c$ between two clusters is larger than the forward light cone at time $t_{eq}$.} \\end{center} \\end{figure}  Standard Big Bang cosmology is also internally inconsistent as a theory of the very early Universe. We know that at very high energy densities which the theory predicts in the initial stages, the description of matter as a classical ideal gas is invalid. Since quantum field theory is a better description of matter at high energies, this naturally leads us to consider quantum field theory as the description of matter which must take over in the very early Universe, and this realization led to the discovery of the inflationary scenario. What follows is a brief overview of principles, progress and problems in inflationary cosmology. For more details, the reader is referred to \\cite{RHB}. ",
        "conclusions": ""
    },
    "0208/hep-ph0208093_arXiv.txt": {
        "abstract": "Relic neutrino $\\nu_{r}$ light masses clustering in Galactic and Local Hot Dark Halos act as a beam dump calorimeter. Ultra High Energy $\\nu$, above ZeV, born by AGNs,GRBs  at cosmic edges, overcoming the  Greisen, Zatsepin, Kuzmin (GZK) cut-off, may hit near Z resonance and WW-ZZ channels energies: their showering into nucleons  and $\\gamma$ Ultra High Cosmic Ray (UHECR) fit observed  data. Any tiny  neutrino mass splitting may reflect into  ({\\em a twin }) bump at highest GZK energy cut-off. The lighter the neutrino masses the higher the Z-Showering cut-off. The Z or WW,ZZ showering might explain a peculiar clustering in observed UHECR spectra at $10^{19}$, $2\\cdot 10^{19}$, $4 \\cdot 10^{19}$ eV found recently by AGASA.  Coincidence of clustered UHECR with highest $\\gamma$ BLac sources, originated either by neutral and charged particles (Q=0,+1,-1) is well tuned to Z-Showering Scenario. Additional prompt TeVs signals occur offering a natural solution of growing Infrared-TeV cut-off paradoxes related to distant TeV BLac sources, while electromagnetic cascades tail may explain correlation found with GeV-EGRET Sources. Such UHE $\\nu$ Astrophysics might trace near GZK  energy into Horizontal Tau Air-Showers originated by the UHE $\\nu_{\\tau}$ Earth-Skimming  in wide Crown Earth Crust around the observer. These Upward and Horizontal ${\\tau}$ Air-Showers \\textbf{UPTAUS}, \\textbf{HORTAUS}, test huge crown target volumes either from high mountains as well as observing from planes, balloons or satellites. The \\textbf{HORTAUS} from mountains measure crown masses at UHE $\\nu$ EeVs energies comparable to few $km^3$, while from satellites at orbit altitudes, at GZK energies $E_{\\nu}\\geq 10^{19}$, their corresponding Horizontal Crown Masses  may even exceed $150$ $km^3$. The expected event rate may produce at least a dozen of event a year within Z-WW Showering model from a satellite altitude. ",
        "introduction": "Light Neutrinos ($\\sim 0.1-3 eV$) clustering in Galactic, Local Hot Dark Halo, being an efficient calorimeter for ZeV $\\nu$ offer the possibility to overcome the Cosmic Black Body opacity ($\\gtrsim 4 \\cdot 10^{19}\\,eV$) (GZK)  at highest energy cosmic ray astrophysics. These rare events, being  nearly isotropic, are probably of cosmic origin. They are very possibly originated by blazars Jets AGN sources pointing their huge linear accelerators to us, as BLac; in standard scenario if the UHECR are originally of hadronic nature they must be  absorbed by the dragging friction of cosmic 2.75 K BBR or by  the inter-galactic radio backgrounds (the GZK cut-off). Indeed as it has been noted (K.Greisen,\\cite{Greisen 1966} Zat'sepin, Kuz'min \\cite{Zat'sepin et al.1966} 1966), proton and nucleons mean free path at E $> 5 \\cdot 10^{19} \\,EeV$ is less than 30 $Mpc$ and asymptotically nearly ten $Mpc$; also gamma rays at those energies have even shorter interaction length ($10 \\,Mpc$) due to severe opacity by electron pair production via microwave and radio background interactions (J.W.Elbert, P.Sommers \\cite{Elbert et all.1995}, 1995)(R.J.Protheroe and Biermann \\cite{Protheroe and Biermann 1997}, 1997. Nevertheless these powerful sources (AGN, Quasars, GRBs) suspected to be the unique source able to eject such UHECRs, are rare  at nearby distances ($\\lesssim 10 \\div 20 \\, Mpc$, as for nearby $M87$ in Virgo cluster); moreover there are not nearby $AGN$ in the observed UHECR arrival direction cone. Strong and coherent galactic \\cite{Protheroe and Biermann 1997} or extra-galactic magnetic fields \\cite{Farrar et al.2000}, able to bend such UHECR (proton, nuclei) directions, are not really at hand. The needed coherent lengths and strength are not easily compatible with known cosmic magnetic fields \\cite{Elbert et all.1995}. Finally in this scenario the $ZeV$ neutrons born, by photo-pion proton conversions on BBR, may escape the magnetic fields bending and should keep memory of the arrival direction, leading to (unobserved)  clustering toward the primary source (Fargion et all 2001a \\cite{Fargion et al.2001a}, \\cite{Fargion et al.2001b}2001b). Secondaries EeV photons (by neutral pion decays) should also abundantly point and cluster toward the same nearby $AGN$ sources (P.Bhattacharjee et all \\cite{Bhattacharjee et al.2000}2000),(J.W.Elbert, P.Sommers \\cite{Elbert et all.1995}, 1995), in disagreement with $AGASA$ data (for any direct UHECR. Therefore Galactic origin for UHECR might be imagined as a simplest solution (by Micro-Quasars sources), but it contradicts the absence of any evident quadruple (galactic plane) or dipole (galactic halo) UHECR an-isotropy. A often revived solution of the present GZK puzzle, the Topological defects ($TD$), assumes as a source, relic heavy particles (GUT masses) of early Universe; they are imagined diffused as a Cold Dark Matter component, in galactic halo, but therefore they are unable to explain the growing evidences of clustering in $AGASA$~ $UHECR$ arrival data and their self-correlation  with far Compact Blazars (BLac) at cosmic distance  (Tinyakov P.G.et Tkachev \\cite{Tinyakov-Tkachev2001}2001; D.S.Gorbunov, P.G.Tinyakov, I.I.Tkachev, S.V.Troitsky \\cite{Gorbunov et al.2002} 2002). In this frame work it is important to remind the Fly's Eye event ($300$ EeV) whose association with Seyfert Galaxy MCG 8-11-11 (or Quasar 3C147), inspired  earliest articles (\\cite{Fargion Salis 1997}D.Fargion, B.Mele, A.Salis \\cite{Fargion Mele Salis 1999}1997-99) to solve GZK by Z-Shower. Therefore the solution of UHECR puzzle based on primary Extreme High Energy (EHE) neutrino beams (from AGN) at ZeV $E_{\\nu} > 10^{21}$ eV and their undisturbed propagation from cosmic distances up to nearby calorimeter, made by relic light $\\nu$ in dark galactic or local dark halo (\\cite{Fargion Salis 1997}D.Fargion, B.Mele, A.Salis \\cite{Fargion Mele Salis 1999}1997-99,Weiler \\cite{Weiler1999} 1999,  S.Yoshida, G. Sigl, S. Lee \\cite{Yoshida et al.1998}1998) remains the most favorite convincing solution for the GZK puzzle. New complex scenarios for each neutrino mass spectra are then opening and important signatures of Z,WW showering must manifest in observed an-isotropy, composition, spectra shape and space-time clustering of present and future UHECR data. \\subsection{ UHE $\\nu$  Astronomy Energy Windows} Rarest TeVs gamma signals are at present the most extreme and rarest trace of  High Energy Astrophysics. The TeVs signals have shown new power-full Jets blazing to us from Galactic or extragalactic edges. At PeVs energies astrophysical Gamma cosmic rays should also be present, but, excluded a very rare and elusive Cyg$X3$ event, they are  not longer being observed. While the corresponding PeVs charged cosmic rays are abundantly hitting the atmosphere, these missing PeVs gamma sources are very probably mostly absorbed by their photon interactions (photo-pion productions, electron pairs creation) at the source environment and/or along the photon propagation into the cosmic Black Body Radiation (BBR) or into other diffused (radio,infrared,optical) background radiation. Unfortunately PeVs charged cosmic rays, bend and bounded in a  random walk by Galactic magnetic fields, loose their original directionality and their astronomical relevance; their resident time in the galaxy is much longer ($\\geq 10^{3}$-$10^{5}$) than neutral ones, as gamma rays, making the charged cosmic rays more probable to be observed by nearly a comparable ratio. On the contrary astrophysical UHE neutrino signals at $10^{13}$eV-$10^{19}$eV (or higher GZK energies) are unaffected by any radiation cosmic opacity and may open a very new exciting window to High Energy Astrophysics. Lower energy astrophysical UHE $\\nu$ at $10^{9}$eV-$10^{12}$eV should also be present, but their signals are (probably) drowned  by the dominant diffused atmospheric $\\nu$ secondaries noises produced by the same charged (and smeared) UHE cosmic rays (while hitting terrestrial atmosphere), the so called atmospheric neutrinos. In a very far corner, at lowest (MeVs) energy  windows, the abundant and steady solar neutrino flux  and the prompt (but rarer) neutrino burst from a nearby Super-Novae (SN 1987A), have been in last twenty years, already successfully explored. The UHE $10^{13}$eV-$10^{16}$eV $\\nu$ 's  astronomy, being weakly interacting and rarer, may  be captured mainly inside huge volumes, bigger than Super-Kamiokande ones; at present  most popular detectors consider underground ones (Cubic Kilometer Size like AMANDA-NESTOR) or (at higher energy $10^{19}$eV-$10^{21}$eV) the widest Terrestrial atmospheric sheet volumes (Auger-Array Telescope or EUSO atmospheric Detectors). Underground $km^3$ detection is based mainly on $\\nu_{\\mu}$ (above hundred TeVs energies, after their interaction with matter) leading to $\\mu$ kilometer size lepton tracks \\cite{Gandhi et al. 1998} . Rarest atmospheric horizontal shower are also expected by $\\nu$ interactions in air (and, as we shall discuss, in the Earth Crust). While $km^3$ detectors are optimal for PeVs neutrino muons, the Atmospheric Detectors (AUGER-EUSO like) exhibit a minimal threshold at highest ($\\geq 10^{18} eV$) energies. ",
        "conclusions": "The discover of the expected UHE neutrino Astronomy is urgent and just behind the corner. Huge volumes are necessary. Beyond underground $km^3$ detectors a new generation of UHE neutrino calorimeter lay on front of mountain chains and just underneath our feet: The Earth itself  offers huge Crown Volumes as Beam Dump calorimeters observable via upward Tau Air Showers, UPTAUs and HORTAUs. Their effective Volumes as a function of the quota $h_1$ has been derived by an analytical function variables in equations above  and Appendix. These Volumes and Masses are discussed below and summirized in the last column of the above Table; they are  large enough to offer  an ideal calorimeter for future UHE neutrino detection."
    },
    "0208/astro-ph0208179_arXiv.txt": {
        "abstract": "{We have observed the region of the hard X-ray transient EXS~1737.9$-$2952 near the Galactic Centre using the Narrow Field Instruments (NFI) of the BeppoSAX X-ray satellite. In this second part of our investigation we report on our spectrum analysis and time variability study of the field. The main results are the MECS spectra of each of the 10 identified sources in the interval 1.65$-$10~keV and spectral fits of the source data. The fluxes obtained with the spectral fits are $1.7-4.8 \\cdot 10^{-12} \\mbox{erg} \\, \\mbox{cm}^{-2} \\, \\mbox{s}^{-1}$. The absorption for the sources with powerlaw and Raymond-Smith thermal plasma models is in the range $N_H=0.5-6.7 \\cdot 10^{22} \\, \\mbox{cm}^{-2}$. The low number of counts and lack of source identifications in the simultaneous 0.1$-$2~keV LECS data of the same field supports high absorption. This indicates that these sources are at least at the distance of the Galactic Centre. From the distance estimate a lower limit for the X-ray luminosity   $L_x \\approx 2-5 \\cdot 10^{34}$ erg/s (2$-$10 keV) is obtained. A powerlaw with a photon index in the range $\\alpha=1.1-1.8$ generally gives a fair fit to the data, but strong line contribution (iron line at 6$-$7~keV) is evident for 5 sources and exists at lower confidence also in the other 5 sources. The fits indicate  differences in line position in the range 6.1$-$7.0~keV suggesting that the ionisation state and/or emission mechanism may not be the same in all sources. The Raymond-Smith model for the 5 sources with reasonable spectral fit yields $ \\mathrm{k}T \\approx 8 - 10 $~keV. Due to the low S/N of the data, other line parameters are not used in our analysis. A $\\chi^2$-analysis of the time-binned data indicates that two of the sources are variable on a time scale of hours at very high confidence ($> 99.99$ \\%), and one source with lower confidence (99.67\\%). The scale ($\\sim$ hours) of the time variability indicates that these sources could be low-mass X-ray binaries. The other sources are most probably high or low mass X-ray binaries or supernova remnants. We also extracted and analysed spectra from larger subfields in the observed MECS region. A subfield including 8 of the new sources and a major contribution of diffuse emission between them yielded a fairly good fit to a power-law spectrum with photon index  $\\alpha=1.3$ and a strong iron line at 6.8 keV, but a poor fit to a Raymond-Smith and bremsstrahlung model for a single source. A spectral fit to another field with only residual emission and no point sources yielded spectral parameters close to the diffuse emission near GC observed by other investigators, except for the high interstellar absorption ($N_H \\approx 2.0 \\cdot 10^{22} \\, \\mbox{cm}^{-2}$). The PDS spectrum at harder X-rays centred on the same position was also observed. Due to lack of spatial resolution, and a FOV larger than that of MECS, this spectrum was more difficult to interpret. The largest contribution of the spectrum is probably by 1E1740.7$-$2942. It is reasonably close to the centre of PDS collimator field, and also the observed flux matches the prediction. The source for the hard X-ray transient EXS~17137.9$-$2952 cannot be identified from the present observations. ",
        "introduction": "The existence of black hole candidates near the Galactic Centre (GC), such as 1E1740.7$-$2942 (see Sunyaev et al. \\cite{Sun91}), has lead us to predict an increasing number of such X-ray objects with improving observing sensitivity. The discovery of 10 previously undetected objects within the SAX MECS field of view centred at EXS~1737.9$-$2952 (Huovelin et al. 1999, \\cite{Huo99}) was nevertheless a surprise, since the region had revealed no clues as to their  existence at any wavelength, except one transient in hard X-rays observed by the EXITE balloon experiment (Grindlay et al. \\cite{Gri93}).  The analysis of the new sources is not straightforward, since we have only one energy region (MECS, 1.65$-$10.5 keV) where we can utilise spatially resolved BeppoSAX data to derive source spectra and study their variability (see \\cite{Huo99}). Lack of spatial resolution and the large field of view of PDS and HPGSPC make the interpretation of the hard X-ray data difficult.  As pointed out in \\cite{Huo99}, it can  not be taken for granted that the new sources really are physically near the GC. It would, however, strengthen this scenario, if $N_H$~ were clearly enhanced for the X-ray spectra of sources overlapping with a molecular cloud near the GC. There is, indeed, another source of potentially useful data of this sky area, provided by our submillimeter (SEST) observations at CO emission lines by Vilhu et al. (\\cite{Vil94}). They found a dense molecular cloud moving at a peculiar velocity near EXS~1737.9$-$2952. In 1990's, there have also been several other studies at different wavelengths, which are summarized in Durouchoux et al. (\\cite{Dur98}).  Most of the observed SAX spectra contain the iron K-shell line (at 6-7 keV). K-shell emission can be produced by either by fluorescence or radiative recombination. The line strength, width and position depend on several parameters of the emitting gas, e.g. temperature, ionization state, and iron abundance. Thus the iron line provides a valuable diagnostic of the physical properties of the gas.  In this paper we report on the spectral and time variability analysis of the new BeppoSAX sources identified in \\cite{Huo99}, discuss alternatives as to their nature, and also study the emission of the region divided in larger subfields. ",
        "conclusions": "On the basis of the X-ray spectral fitting of our MECS observation, the new BeppoSAX sources are characterised by  high hydrogen column density. Thus they are probably close to the Galactic Centre or behind it. Deriving from the nondetection of sources in the LECS observation of the same field, the hydrogen column density may be so high that even extragalactic origin might be possible. Using the assumption that the sources are near the GC, the (absorbed) X-ray luminosities of the sources are in the range $L_x \\approx 2-5 \\cdot 10^{34}$ erg/s (2$-$10 keV), and the maximum unabsorbed luminosities are $\\sim$40\\% higher.  The rough indications of X-ray spectrum slope and iron line properties from the observation do not allow to find conclusive evidence for any specific interpretation as to the nature of the sources. It is, however, reasonable to consider case by case different types of sources as possible explanations for our observation. The following conclusions include first our considerations of a common explanation (diffuse emission near the GC, and a SNR) for the whole observed field.  The average emission spectrum of the region including 8 of the 10 new sources is hard (powerlaw photon index 1.3) and it includes a strong iron line at 6.8 keV (subfield 1 in Table 1, and Fig. 4). It also shows a strong interstellar absorption component (N$_H \\approx 10^{22}$). These features do not match the diffuse emission near the GC observed by Koyama et al. (\\cite{Koy94}), and thus the diffuse emission scenario is improbable.  Taking into account the observed time variability of sources 1, 2, and 10, and the differences between the spectra of individual sources, an explanation of the whole field consisting of a supernova remnant is also very improbable. The minimum distance derived from the hydrogen column density would lead to to a size scale for the source, which is far too large for any single SNR. The spatial and temporal fluctuations of the X-ray  spectrum also speaks for existence of a group of independent sources superimposed by a spectrum component due to the radiation and absorption by interstellar medium.  Considering explanations for the nature of individual sources in the field, one source category can be readily ruled out. The X-ray luminosities of the new sources are well above those of chromospherically active stars and cataclysmic variables.  From the remaining possible source categories, X-ray binaries are the most favourable explanation for all 10 sources. The observed luminosities of the new sources, assuming that they are near the GC or further away from us, are well within the range of X-ray binaries. X-ray binaries also provide a  plausible explanation for the variability in the order of hours for three of the new sources (1,2, and 10), and the fairly hard X-ray spectra with 6$-$7~keV iron line emission suit well for typical XRBs.  The remaining seven sources (3-9), where variability could not be verified, might as well be supernova remnants or extragalactic sources (AGN), if just the spectrum characteristics are considered. However, the low probability of finding ten AGNs in a SAX MECS field makes the extragalactic explanation too far fetched. Judging from the expected number (0.2) of extragalactic sources with the observed luminosity of our sources, and the lack of identifications of extragalactic sources with any previous observations  of our MECS field, we strongly suggest that the new SAX sources are all galactic.  Looking at the remaining exlanations, any of the sources 3$-$9 could be either a SNR or an X-ray binary. For the SNR scenario, it is not plausible to make further suggestions, since the soft X-ray emission, which would be a good diagnostic of different types of SNR, is strongly depressed, very probably by interstellar absorption. Also, the angular resolution of our observation is not good enough to distinguish any spatial features of a SNR at the GC distance.  The identification of the source for the hard X-ray transient (Grindlay et al. \\cite{Gri93}) is still not possible on the basis of the presented observations, and the nature of EXS~1737.9-2952 remains an enigma.  Even a short observing time of the EXS region with large X-ray satellites like  Chandra or XMM-Newton would improve significantly the classification of these  new X-ray sources. Unfortunately we have not been successful enough to get observing time for our project  with these facilities."
    },
    "0208/astro-ph0208209_arXiv.txt": {
        "abstract": "The initial results of the Deep Lens Survey (http://dls.bell-labs.com) to identify possible brown dwarfs and extremely metal poor red halo subdwarfs near the hydrogen burning limit are presented. Individual deep CCD high galactic latitude survey fields appear to offer a low probability of discovering field BD's, but taken collectively offer an opportunity to begin addressing questions regarding the scale height and distribution of these objects.  In all likelihood, the very depth of such surveys will greatly increase our knowledge of the coolest extreme halo objects, which currently are known in far fewer numbers than T dwarfs. Ultimately, the large volume surveyed by the Large Synoptic Survey Telescope will identify vast numbers of such objects, providing a more complete picture of their spatial distribution. ",
        "introduction": "Since the early 1980's, CCD's have been employed on large telescopes to obtain deep images of small sections of the sky.  With better than $60\\%$ quantum efficiency ranging from the atmospheric UV cutoff to the drop in their sensitivity at $1\\mu$, these first surveys readily reached a limiting magnitude of 26 in all of the broad bandpasses encompassed by this wavelength range.  Until the mid-1990's the 1 cm size of the CCD chip limited the area on the sky covered with each exposure to a few sq. arcmin.  Consequently, the actual volume surveyed per field was very small, typically a few or tens of $pc^{3}$, up to the scale height of 325 pcs for late type disk population dwarf stars.  As such, they proved an ineffective method to discover the latest M dwarfs, let alone L or T brown dwarfs with unknown scale height and local volume densities estimated to be in the range of $10^{-2}$ per L or T class of objects (Liu et al. 2002). Complicating the picture was the fact that these untargeted surveys had multipurpose science drivers, often were focused at higher galactic latitudes, and employed many different filter systems.  More recent developments leading to larger CCD chips and mosaics have increased the sizes of individual deep survey fields to tens of arc minutes across, but have not significantly increased survey depth. With the success of the large area SDSS optical survey in discovering BD's, and the marked absence of similar results from the many deeper pencil beam surveys over two decades, the latter do not appear to be a very efficient method to discover such objects.  Yet despite continued limited spatial coverage, because of their increased depth, serendipitous discoveries of BD's by existing pencil beam surveys may help to address questions surrounding the uncertainties regarding the spatial distribution of field BD's. ",
        "conclusions": "Very deep multicolor pencil beam surveys provide a rich data base which can be utilized to provide information on the scale height distribution of field brown dwarfs and more accurately define the luminosity function of lowest mass stellar halo population.  Though each individual mosaiced CCD subfield in itself offers a low probability of detection, the combined data from larger abutted areas, such as those covering 4 sq. deg. in the DLS, should allow for a meaningful sample of objects to be evaluated. Using $\\sim0.01/pc^3$ each for the local volume density of L and T dwarfs as estimated by Liu (2000), we might expect to find $\\raisebox{-0.5ex}{$\\stackrel{<}{\\scriptstyle \\sim}$}3$ T dwarfs and $\\raisebox{-0.5ex}{$\\stackrel{<}{\\scriptstyle \\sim}$}33$ L dwarfs in each 4 sq. degree DLS field when our survey in completed.  This assumes detection in $R-z^{\\prime}$ and a uniform spatial distribution with a vertical scale height of no more than 200 pcs.  Reducing the scale height to 100 pcs lowers the expected number of L dwarfs by a factor of 8. If we simply search for stellar objects which drop out of all bands except z', then the number of T dwarfs per field should drop from $\\sim30$ (indicating a scale height $\\sim200$ pc) to about 4 (scale height $\\sim100$ pcs).  Whereas only a relative handful of the coolest halo subdwarfs have been identified to date using other techniques, carefully analyzed deep CCD observations including B band data should discover in one DLS field numbers of very red esdM's similar to those currently known.  Extrapolating from the luminosity function of Dahn et al. (1995, 1996), we expect to find at least 4 extreme subdwarfs similar to LHS 1742a (esdM 5.5) or LHS 1826 (esdM6) per each LHS field and 15 stars similar to LHS 453 (esdM3.5).  The observations for esdM's should prove equally important as the search for additional L and T dwarfs since we know far less about the coolest halo population.  When the 30 sq. degree DLS is finished, more will be known about these subluminous halo objects and field BD's. The problem with current pencil beam surveys is that the numbers of halo stars found do not permit cutting the sample in multiple ways, testing their distribution with galactic coordinates, etc.  Enabled by a confluence of optics and microelectronics advances, the proposed Large Synoptic Survey Telescope ($LSST$, http://lsst.org), will produce a large volume optical multi-color survey of $14,000$ sq. degrees to $27^{th}$ magnitude. Completion of the facility is expected by 2012, and the data will be public.  Image quality is expected to be as good or better than the current generation of new technology telescopes ($0.4^{\\prime\\prime}$ fwhm), and each field of the sky will be covered hundreds of times, permitting proper motion studies and enhanced rejection of galaxies due to the low surface brightness achieved.  Moreover, due to higher quantum efficiency at $\\sim1\\mu$, the $LSST$ filter set will most likely include an additional filter near this spectral region, complementing the SDSS $ugriz$ filter set.  The combination of broad area coverage, survey depth and increased near IR sensitivity will allow for a direct determination of the local disk distribution of field brown dwarfs.  Furthermore, the $LSST$ will create a sample of thousands of subluminous halo stars, probing deep into the halo as well as spanning the full range of galactic coordinates. The role of these stars in the evolution of the structure of our galaxy will then be more apparent."
    },
    "0208/gr-qc0208078_arXiv.txt": {
        "abstract": "We present a class of general relativistic soliton-like solutions composed of multiple minimally coupled, massive, real scalar fields which interact only through the gravitational field.  We describe a two-parameter family of solutions we call ``phase-shifted boson stars'' (parameterized by central density $\\rho_0$ and phase $\\delta$), which are obtained by solving the ordinary differential equations associated with boson stars and then altering the phase between the real and imaginary parts of the field.  These solutions are similar to boson stars as well as the oscillating soliton stars found by Seidel and Suen [E. Seidel and W.~M. Suen, Phys. Rev. Lett. {\\bf 66},  1659  (1991)]; in particular, long-time numerical evolutions suggest that phase-shifted boson stars are stable.  Our results indicate that scalar soliton-like solutions are perhaps more generic than has been previously thought. ",
        "introduction": "\\vspace{-0.2cm} The nature of dark matter in the universe is currently an open question in physics, with many models being proposed to fill this gap in our understanding, some of which resort to the use of exotic matter.  Of interest to us is one class of models composed of massive scalar fields coupled to the general relativistic gravitational field, from which compact, star-like solutions can be formed, solutions which go by the names of ``oscillating soliton stars'' (or ``oscillatons'') \\cite{OSS,Lopez} for real fields and ``boson stars'' \\cite{Kaup,RB,Colpi,Jetzer,Mielke,Bizon} for complex fields. These star-like solutions have received renewed attention recently, and a substantial body of evidence has been advanced in an effort to show that these fields may be key players on both galactic \\cite{JWLee, Francisco, Miguel} and cosmological \\cite{Matos} scales.   Boson stars have been suggested as alternatives to primordial black holes \\cite{Mielke2} as well as supermassive black holes in the centers of galaxies \\cite{Torres1}, and their gravitational lensing properties have been explored \\cite{Dabrowski}, further developing the treatment of these solutions as objects of astrophysical interest.  Apart from the possible astrophysical relevance of these star-like objects, we find them interesting to study from a mathematical standpoint as well, for their properties as soliton-like solutions in the nonlinear dynamics of general relativity.  The ``solution space'' of general relativity is still largely unexplored, and these scalar objects comprise simple systems with which to conduct investigations.  It is from this viewpoint that we will proceed in this paper.  In 1991, Seidel and Suen \\cite{OSS} considered the model of a real massive scalar field, minimally coupled to the general relativistic gravitational field, with the additional simplifying assumption of spherical symmetry.  These authors were interested in the existence of ``nontopological solitons'' in the model: that is, whether the equations of motion admitted stable, localized, non-singular distributions of matter which could be interpreted as ``scalar stars''.  A theorem due to Rosen~\\cite{Rosen} suggested that, should such solutions exist, they could not be static.  Thus, Seidel and Suen looked for {\\em periodic} configurations by substituting a particular Fourier ansatz into the equations of motion, and solving the resulting hierarchy of ODEs via a generalized shooting technique. The authors found strong evidence that periodic star-like solutions {\\em did} exist, and, via direct numerical simulation, demonstrated that their ``oscillating soliton stars'', if not absolutely stable, had lifetimes many orders of magnitude longer than the stars' intrinsic dynamical times.\\footnote{More precisely, the oscillating stars constitute a one-parameter family which may be parametrized by the mean (period-averaged) central density, $\\rho_0$.  As with other relativistic stellar models, a plot of total (ADM) mass versus $\\rho_0$ exhibits a maximum at $\\rho_0^\\star$, which seems to coincide with a transition from stable to unstable configurations. As expected, only stars with $\\rho_0 < \\rho_0^\\star$ could be stably evolved for long times.}  These results were surprising to some researchers, particularly since the model has no conserved Noether current, which, it had been argued, was responsible for the existence of solitonic solutions in other non-linear field theories involving scalar fields~\\cite{Coleman,Lee}. However, at least heuristically, we can understand the existence of the oscillating stars as arising from a balance between the attractive gravitational interaction and the effective repulsive self-interaction generated by the mass of the scalar field ({\\it i.e.} via the dispersion relation of the Klein-Gordon equation).  Recently it was shown by Ure\\~na-L\\'opez \\cite{Lopez} that approximate solutions for boson stars and oscillating soliton stars, or ``oscillatons'' as he calls them, can both be derived from a single set of equations in a sort of ``stationary limit''.  The similarities seen between boson stars and oscillating soliton stars in terms of their curves relating mass, radius and central density can thus be related formally.  In this paper, we build on the works of Seidel and Suen and Ure\\~na-L\\'opez by considering a matter content consisting of {\\em multiple} scalar fields, and we find some further ways in which boson stars and oscillating soliton stars are similar. For the specific case of two scalar fields, we find evidence for a new family of quasi-periodic, solitonic configurations.\\footnote{Although the oscillations appear to be periodic, we do not have a proof of their periodicity and so we use the term ``quasi-periodic'' to describe their temporal behavior.  Small departures from strict periodicity over long times scales are visible in the simulation results; however, these departures become smaller as we decrease the radial mesh spacing $\\Delta r$, and thus it is plausible that in the limit $\\Delta r \\rightarrow 0$, the solutions are truly periodic.} Together with previous results, this suggests that solitonic solutions are generic to models which couple massive scalar fields through the Einstein gravitational field. We should note, however, that we make no attempt to address the question of whether these multiple scalar fields actually exist in nature; rather we are interested in their existence as valid mathematical solutions in the Einstein-Klein-Gordon system.  We begin by considering $n$ real, massive Klein-Gordon fields $\\phi_i, \\, i = 1, 2, \\ldots n$, without additional self-interaction, minimally coupled to the general relativistic gravitational field.  Specifically, choosing units such that $c=1$  and $G = 1$, the Lagrangian density for the coupled system is  \\begin{equation} {\\cal L} = \\sqrt{-g} \\left( R - \\sum_{i=1}^{n} \\left( {\\phi_i}^{;a}{\\phi_i}_{;a} - m_i^2\\phi_i^2 \\right)\\right) \\end{equation} where $g\\equiv\\det{g_{\\mu\\nu}}$, $R$ is the Ricci scalar, and $m_i$ is the mass of the $i$-th scalar field.  We now restrict attention to spherical symmetry, and adopt the ``polar/areal'' coordinate system, so that the metric takes the form: \\begin{equation} ds^2 = -\\alpha^2(t,r)\\,dt^2 +  a^2(t,r)\\,dr^2 + r^2\\,d\\Omega^2 \\label{metric} \\end{equation} The complete evolution of the scalar fields and spacetime can then be given in terms of a Klein-Gordon equation for each of the $\\phi_i$, and two constraints derived from the Einstein field equations and the coordinate conditions used to maintain the metric in the form~(\\ref{metric}).  We solve these equations using the same scheme adopted for the critical phenomena study described in \\cite{HawleyChop}, and only briefly review that scheme here.  We define the following auxiliary scalar field variables:  \\begin{equation}\\Phi_i\\equiv\\phi_i',\\ \\ \\ \\ \\ \\ \\ \\ \\ \\Pi_i\\equiv { a\\over\\alpha}\\dot{\\phi}_i\\end{equation} where all variables are functions of $t$ and $r$, $\\dot{}\\equiv\\partial/\\partial t$ and $'\\equiv\\partial/\\partial r$. The Klein-Gordon equation is written as the following system: \\begin{eqnarray} \\dot{\\Pi}_i &=& {1\\over r^2}\\left( {r^2\\alpha\\over a}\\Pi_i \\right)' - m_i^2\\alpha a\\phi_i  \\cr \\dot{\\Phi}_i &=& \\left( {\\alpha\\over a}\\Pi_i \\right)'  \\cr \\phi_i(t,r) &=& \\phi_i(t,r_{\\rm max}) - \\int_{r_{\\rm max}}^{r} \\, \\Phi(t,{\\tilde r}) \\, d{\\tilde r} \\label{eq:kg} \\end{eqnarray} where $r=r_{\\rm max}$ is the outer boundary of the computational domain.  The constraint equations are the ``Hamiltonian constraint''  \\begin{equation}  a' =  a{1- a^2\\over 2r} + 2\\pi r a \\sum_{i=1}^{n} \\left( {\\Pi_i}^2 + {\\Phi_i}^2 + a^2 m_i^2\\phi_i^2 \\right), \\label{eq:ham} \\end{equation} and the ``slicing condition''  \\begin{equation} \\alpha' = \\alpha \\left( { a^2-1\\over r} + { a'\\over  a} - 4\\pi r a^2\\sum_{i=1}^{n}m_i^2\\phi_i^2  \\right). \\label{eq:slice} \\end{equation}  For diagnostic purposes, we have also found it useful to compute and monitor the quantities, $M_i(t,r_{\\rm max})$, defined by \\begin{equation} M_i(t,r_{\\rm max}) \\equiv 4\\pi \\int_0^{r_{\\rm max}} \\tilde{r}^2\\, \\rho_i(t,\\tilde{r})\\, d\\tilde{r}, \\label{eq:defMi} \\end{equation} where \\begin{equation} \\rho_i(t,r) =\\frac{ \\Pi_i{}^2+ \\Phi_i{}^2 +  {a^2}\\phi_i{}^2}{2a^2}. \\end{equation} \\label{eq:defrhoi} Loosely speaking, we can interpret $M_i(t,r_{\\rm max})$ as the total contribution of field $i$ to the ADM mass of the spacetime.  In particular, so long as no matter out-fluxes through $r=r_{\\rm max}$, we have \\begin{equation} \\sum_{i=1}^{n} M_i(t,r_{\\rm max}) = \\hbox{{\\rm const.}} \\label{eq:conssumMi} \\end{equation}  We solve Eqs. (\\ref{eq:kg})-(\\ref{eq:slice}) subject to the the boundary conditions $a(t,0) = 1$ (local flatness at the origin) and $\\alpha(t,r_{\\rm max}) = 1/a(t,r_{\\rm max})$ (so that $t$ measures proper time as $r\\to\\infty$).  As in \\cite{HawleyChop}, we use the Sommerfeld condition for a {\\em massless} field to set the values $\\phi(t,r_{\\rm max}), \\Phi(t,r_{\\rm max})$ and $\\Pi(t,r_{\\rm max})$. Since the Sommerfeld condition is not ideal for a massive field, we ran our simulations with different values of $r_{\\rm max}$, testing for any periodicity or other effect which might be a function of $r_{\\rm max}$, and usually ran with an $r_{\\rm max}$ which was large compared to the time for which we ran the simulation.  Even with smaller $r_{\\rm max}$, we found our results to be essentially independent of $r_{\\rm max}$ and attribute this to the fact that there is very little scalar radiation emitted from the soliton-like objects considered here. Our results are also essentially independent of the resolution of the finite differencing algorithm and the Courant-Friedrichs-Levy factor, $\\Delta t/\\Delta r$, and we confirm that our results converge in a second-order-accurate manner using independent residual evaluations.   \\vspace{-0.4cm} ",
        "conclusions": "We have strong numerical evidence for the existence of a two-parameter family of soliton-like solutions to the Einstein-Klein-Gordon system (parameterized by central density $\\rho_0$ and phase $\\delta$) for the case of two scalar fields. We speculate that extending the system to more scalar fields will yield similar results (especially if one uses a trivial extension like Eq. (\\ref{trivdecomp}) ).  The solutions we refer to as ``phase-shifted boson stars'' consist of boson star initial data for which the phase difference $\\delta$ between the real and imaginary components of the field have been altered.  These solutions oscillate in a seemingly periodic manner for very long times; thus they appear to be stable.  For the case of $\\delta=0$, we obtain close approximations to the oscillating soliton stars of Seidel and Suen. For other values of $\\delta$, we find solutions which also appear to be stable and periodic; furthermore we can see mass-energy being exchanged between the two fields."
    },
    "0208/astro-ph0208515_arXiv.txt": {
        "abstract": "The magnification effects of clustered matter produce variations in the image sizes and number density of galaxies across the sky. This paper advocates the use of these effects in wide field surveys to map large-scale structure and the profiles of galaxy and cluster sized halos. The magnitude of the size variation as a function of angular scale is computed and the signal-to-noise is estimated for different survey parameters. Forthcoming surveys, especially well designed space-based imaging surveys, will have high signal-to-noise on scales of about 0.1 arcminute to several degrees. Thus the clustering of matter could be measured on spatial scales of about 50 Kpc to 100 Mpc. The signal-to-noise is dominated by sample variance rather than shot-noise due to the finite number density of galaxies, hence the accuracy of the measurements will be limited primarily by survey area, sampling strategy and possible systematics. Methods based on magnification are compared with the use of shape distortions and the contrasts and complementarities are discussed. Future work needed to plan survey strategy and interpret measurements based on magnification is outlined. ",
        "introduction": "Gravitational lensing refers to the distortions in images of distant galaxies due to the deflection of light rays by mass concentrations. The distortion on a circular image can be decomposed into an amplification of the size of the image and an anisotropic stretching of its shape into an ellipse. The size amplification is called magnification and the anisotropic stretching is the shear. Gravitational lensing due to galaxy clusters and large-scale structure typically leads to distortions of order 1-10\\% (e.g. Gunn 1967; Miralda-Escude 1991; Blandford et al 1991; Kaiser 1992; Bernardeau, van Waerbeke \\& Mellier 1997; Jain \\& Seljak 1997; Kaiser 1998). In this regime of weak lensing the magnification $\\mu$ is given by \\begin{equation} \\mu = \\left((1-\\kappa)^2 - |\\bm{\\gamma}|^2\\right)^{-1} \\simeq 1 + 2 \\kappa . \\end{equation} where $\\kappa$ is the convergence and $\\bm{\\gamma}$ the complex shear. So far observational studies of weak lensing have largely used the measured ellipticities to estimate the shear and thus the projected mass distribution. %  This paper makes the case for using effects of magnification in addition to the shear in mapping dark matter. Magnification leads to fluctuations in the sizes and number densities (in a flux limited survey) of galaxies (e.g. Bartelmann \\& Narayan 1995; Broadhurst, Taylor \\& Peacock 1995; Schneider, King \\& Erben 200). In the context of galaxy clusters the change in number density has been used to constrain the mass distribution, but with less accuracy than the shape measurements (Taylor et al 1998). We argue that for forthcoming blank field surveys the prospects are much better than for clusters to measure both effects of magnification, on sizes and number densities.  (i) Magnification effects, unlike the shear, require control fields to estimate the mean, unlensed size distribution. This had been a limitation for small, arcminute sized, cluster fields, but is automatically done in a blank field survey.  (ii) The signal-to-noise (henceforth ${\\cal S/N}$) due to shot noise is somewhat lower for magnification effects than the shear, and this has proven critical in cluster lensing, since a factor of 2 in ${\\cal S/N}$ is hard to make up. However field lensing surveys with areas larger than 10 square degrees are mostly in the regime where sample variance or systematics dominate the errors. It is therefore feasible that from forthcoming imaging surveys with good control of systematics (photometric calibration for number density, resolution for sizes, and psf anisotropy for shear) all three lensing measurements can be made. Consistency checks on the different systematics can then be made, the ${\\cal S/N}$ on the measured dark matter clustering improved, and new information on halo properties can be extracted.  (iv) Space based imaging surveys will make possible the measurements of sizes with an accuracy hard to achieve from the ground. Such surveys will become feasible over small areas with the Advanced Camera for Surveys on the HST, and over substantial fractions of the sky with a wide field imaging satellite telescope.  The main goal of this paper is to propose that measurements of magnification effects, in particular the effect on galaxy sizes, be an integral part of the lensing agenda for forthcoming imaging surveys: wide area, multi-color ground based surveys like the CFHT Legacy Survey (see www.cfht.hawaii.edu/Science/CFHLS/), the proposed LSST (www.lss.org) and WFHRI (www.ifa.hawaii.edu/ $\\tilde{}$kaiser/wfhri) surveys, and especially a space based imaging survey as proposed for the SNAP satellite (http://snap.lbl.gov) which will have the key requirements of small psf and pixels $\\sim 0.1$ arcsecond, photometric redshifts, and survey area exceeding 100 square degrees (G. Bernstein, private communication). The formalism for computing statistical measures of magnification is presented in Section 2. Section 3 provides estimates of the ${\\cal S/N}$ expected for measurements of size fluctuations. We conclude in Section 4. ",
        "conclusions": "What kind of survey would be suitable for measuring the magnification effects discussed in this paper? For the effect of magnification on galaxy sizes, a wide area space based multi-color imaging survey would be ideal. It is challenging for a ground based telescope to overcome the effect of psf smearing on the size distribution, unless one has the luxury of a large enough sample of galaxies with sizes larger than the psf. With appropriate multi-color imaging one can obtain photometric redshifts which can help reduce the scatter in measuring the size variance induced by lensing, allow one to check for intrinsic correlations in sizes and eliminate their contribution if needed. It also allows for the possibility of measuring the evolution of matter clustering by binning the source galaxies in redshift (Jain \\& Seljak 1997; Hu 1999). With a psf of order 0.1 arcseconds and deep imaging, it is feasible to make size measurements on of order a million galaxies over a 10 square degree area (based on the size vs. magnitude measurements in the Hubble Deep Field by Gardner \\& Satyapal 2000). This would give adequate ${\\cal S/N}$ to measure the variance of the size distribution over a few bins in angle ranging from 1 to 10 arcminutes.  With an area coverage of 100s of square degrees, which would probably be feasible only with a dedicated imaging satellite such as SNAP, one can measure the projected matter power spectrum to a precision of a few percent, measure higher order correlations, and ideally in combination with shear information, get useful constraints on cosmological parameters. On the smallest scales, cross-correlation statistics would probe galaxy halos on scales of a few 10s of Kpc. By combining the magnification measurements with the shear, the density profile of halos can be measured far more accurately than with just galaxy-galaxy lensing, which probes only the integrated mass within radii. Further work is needed to quantify this, explore how small the scales that one can probe are, and check the validity of the approximation of equation 1 on these scales. Magnification effects make possible other useful measures of the non-Gaussian lensing field that have proven difficult to obtain from shear data, such as the skewness of $\\kappa$ which probes $\\Omega_m$ (Bernardeau et al 1997) and peak statistics which probe the mass function of halos (Jain \\& van Waerbeke 2000).  The cross-correlation effects of magnification on the number densities of galaxies, and of foreground galaxy position with background galaxy sizes, are in principle easier to measure. This is because these statistics are first order in the lensing convergence whereas the size variance is of second order. The interpretation is more complex in that it involves the relation of a foreground galaxy population with the mass. The main requirements for accurate measurements are photometric redshifts for a large sample of galaxies (to separate the foreground and background galaxies), and high imaging quality as discussed above. For deep imaging data that has a redshift distribution peaked at $z\\gsim 1$, an adequate dataset would encompass 10 square degrees, while an ideal dataset would cover more than a 100 square degrees. The southern strip of SDSS fulfills the requirements outlined above, as do other smaller imaging surveys that are in progress or being planned.  It is hoped that the discussion and results presented here motivate the integration of magnification measurements as part of the scientific agenda of wide area imaging surveys. The precise requirements for a given survey that will enable useful magnification measurements to be made need careful consideration. At the same time work is needed on survey strategy, techniques for combining magnification and shear information, and appropriate statistical measures that can be extracted from the data.  Acknowledgments: I would like to thank Matthias Bartelmann, Andrew Connolly, Gary Bernstein, Jason Rhodes, David Rusin, Alex Szalay, Masahiro Takada and Andy Taylor for helpful discussions. Comments from an anonymous refereee led to improvements in the paper. This work was supported in part by a NASA-ATP grant and a Keck foundation grant."
    },
    "0208/astro-ph0208319_arXiv.txt": {
        "abstract": "The far-red portion of the spectrum offers bright prospects for an accurate classification of cool stars, like the giant components of symbiotic stars. The 8480--8740 \\AA\\ region, free from telluric absorptions and where the GAIA Cornerstone mission by ESA will record spectra for $3\\times 10^8$ stars, is investigated on the base of available observed and synthetic spectral atlases. We have identified and calibrated diagnostic line ratios useful to derive the effective temperature (spectral type) and gravity (luminosity class) for cool stars observed at spectral resolutions 10,000 $\\leq \\lambda/\\Delta\\lambda \\leq$ 20,000, bracketing that eventually chosen for GAIA. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0208/astro-ph0208296_arXiv.txt": {
        "abstract": "We have observed the black hole candidate 1E 1740.7 --2942, the brightest persistent hard X-ray source within a few degrees of the Galactic centre, for 10 ksec with \\emph{Chandra} (ACIS-I) on August 2000. Attempting to compensate for pile-up effects we found the spectra were well-fit by an absorbed power law, with photon indices $\\Gamma=1.54 ^{+0.42}_{-0.37}$ (readout streak) and $\\Gamma =1.42^{+0.14}_{-0.14}$ (annulus), consistent with a black hole low/hard state. We have analysed a public observation performed by \\emph{Chandra} which utilised short frames in order to avoid severe pile-up effects: subtracting the core point spread function from the whole image, we did not find evidence for any elongated feature perpendicular to the radio jet axis, as reported in a recent analysis of the same data. Moreover, comparing the radial profiles with those of an unscattered X-ray point source, we found indication of an extended, previously undetected, X-ray scattering halo. The measured halo fractional intensity at 3 keV is between 30 and 40 percent within 40 arcsec but drops below detectable levels at 5 keV. Finally, by placing a limit on the X-ray flux from the radio emitting lobe which has been identified as the hot spot at the end of the northern jet of 1E 1740.7 -- 2942, we are able to constrain the magnetic energy density in that region. ",
        "introduction": "The Black Hole Candidate (BHC) 1E 1740.7 -- 2942 is the brightest hard X-ray source close to the centre of our Galaxy, with a hard X-ray spectrum and luminosity comparable to Cygnus X-1 (Liang \\& Nolan 1984; Sunyaev \\etal 1991a; Liang 1993). Interest in this source increased following the discovery of its association with a double sided radio-emitting jet by Mirabel \\etal (1992), since when  it has been classified as a \\emph{``microquasar''}. Unfortunately 1E 1740.7 -- 2942 suffers from extremely high galactic extinction: intense searches, both in the optical and IR band, have failed to identify a counterpart (Prince \\& Skinner 1991; Mereghetti \\etal 1992; Djorgovski \\etal 1992; Mirabel \\& Duc 1992; Marti \\etal 2000; Eikenberry \\etal 2001). It has been proposed that 1E 1740.7 -- 2942 is accreting from a nearby molecular cloud in which the source is likely to be embedded (Bally \\& Leventhal 1991; Mirabel \\etal 1991; Phillips \\etal 1995; Yan \\& Dalgarno 1997). Recently, Smith \\etal (2001) reported the detection of a weak periodic modulation in the long-term X-ray lightcurve: a period of about 12.5 days has been estimated.  In addition much effort has been devoted to the possible identification of 1E 1740.7 -- 2942 with a source of 511 keV annihilation radiation, but any definitive confirmation has been not reported so far (Bouchet \\etal 1991; Sunyaev \\etal 1991b; Anantharamaiah \\etal 1993; Jung \\etal 1995).  Due to its very high hydrogen column density ($N_H \\sim 10^{23}$ cm$^{-2}$ -- e.g. Sheth \\etal 1996; Churazov \\etal 1996; Sakano \\etal 1999) this object is in principle an ideal candidate to investigate the properties of the X-ray scattering halos which are known to be generated by dust grains along the line of sight and which provide an unique opportunity to probe the dust component of the interstellar medium (see Predehl \\& Schmitt 1995) as well as potentially measuring the distance (Predehl \\etal 2000). X-ray halos have been observed by \\emph{Einstein, ROSAT} and \\emph{ASCA}, but thanks to the \\emph{Chandra}'s angular and energy resolution it is now possible to detect and investigate them in far greater detail than ever before. More recently Cui \\etal (2001) observed 1E 1740.7 -- 2942 with \\emph{Chandra} HETGS and reported the detection of an elongated X-ray feature in the zeroth-order image, with an extension of about 3 arcsec, orientated roughly perpendicular to the radio-jet axis. \\\\ This paper is structured as follows: in section 2 we describe the observation and present our results focusing on the spectral analysis and on the overall structure of the source, which has been reconstructed both from our observation and from another public observation (see Cui \\etal 2001), in which the effects of the pile-up do not distort the central region of the image. In section 3 we show some preliminary results on the X-ray halo investigation, comparing the radial profiles of 1E 1740.7 -- 2942 with those of an X-ray point-like source which is supposed to be unscattered, while in the fourth section we provide an estimate of X-ray and radio fluxes coming from the radio emitting region which as been identified as the hot spot at the end of the northern jet of 1E 1740.7 -- 2942. The last section is devoted to our summary and conclusions. \\begin{figure*} \\epsfig{figure=fig_1a.eps,height=5.2cm} \\epsfig{figure=fig_1b.eps,height=5.0cm} \\caption{\\label{fig01}\\footnotesize{\\emph{Left panel}: 2-10 keV energy spectrum of 1E 1740.7 -- 2942 extracted from the readout streak. It is well fitted by an absorbed power law with $\\Gamma=1.54 ^{+0.42}_{-0.37}$, N$_{H}=11.8^{+2.3}_{-1.9} \\times 10^{22}$ cm$^{-2}$} and $\\chi^{2}$/d.o.f.=82/80. \\emph{Right panel}: 2-10 keV spectrum extracted from an annulus region with an internal radius of 4 arcsec. In this case we obtained $\\Gamma =1.42^{+0.14}_{-0.14}$ and $N_{H}=10.45^{+ 0.73}_{-0.70} \\times 10^{22} cm^{-2}$ with $\\chi^{2}$/d.o.f.=430/331.} \\end{figure*} ",
        "conclusions": "We have performed a detailed analysis of a 10 ksec {\\em Chandra} ACIS-I observation of the black hole candidate 1E 1740.7 -- 2942. The conclusions of this work are as follows:  \\begin{itemize} \\item We have utilised two approaches to measuring the X-ray spectrum without suffering the effects of pile-up, which are very strong in the core of our image. Using the readout streak, the 2-10 keV spectrum is well fitted by an absorbed power law with $\\Gamma=1.54 ^{+0.42}_{-0.37}$ and N$_{H}=11.8^{+2.3}_{-1.9} \\times 10^{22}$cm$^{-2}$. Using an annulus from 4--30 arcsec, providing more counts but potentially more prone to pile-up effects, we also fit an absorbed power law, with $\\Gamma =1.42^{+0.14}_{-0.14}$ and $N_{H}=10.45 ^{+ 0.73}_{-0.70} \\times 10^{22} cm^{-2}$ with $\\chi^{2}/{ d.o.f.}=430/331$. Both results are consistent, as expected, with a black hole low/hard state, and furthermore indicate that the annulus approach does not suffer too badly, if at all, from pile-up. \\item We do not confirm the presence of an elongated X-ray feature (about 3 arcsec) reported by Cui \\etal 2001.  Analysing the same data set, now public, we did not find evidence for any asymmetric X-ray structure within about 4 arcsec from the centre. On larger angular scales there is no obvious asymmetry -- in deeper images from our new data -- to several arcmin. \\item We calculated and compared the surface brightness distributions of 1E 1740.7 -- 2942 to those of an unscattered X-ray source in different energy bands. Despite complications due to pile-up effects and uncertain PSF, which lead the comparison between the profiles to be very sensitive to the normalisation factor, we found clear evidence for scattered X-rays in the energy range 2.5-3.5 keV at an angular separation $\\lsim$ 40 arcsec from the core. At higher energies and/or larger angular separations, the scattering halo is not clearly detectable. \\item We provide an upper limit on the ratio between X-ray and radio fluxes in the region which corresponds to the hot spot at the end of the northern radio jet emitted by 1E 1740.7 -- 2942, from which we can crudely constrain $B_{lobe} \\geq 1\\mu$G.  \\end{itemize}  Even though 1E 1740.7 -- 2942 is in principle an ideal candidate for the study of X-ray scattering halos because of its huge column density, which is proportional to the amount of dust grains along the line of sight, such an investigation is complicated by the fact that the source is almost completely absorbed below about 3 keV. Theoretically, the optical depth due to scattering scales as $E^{-2}$, making the softer, \\emph{i.e. the absorbed}, band more suitable for investigating X-ray halos. This demonstrates that a limit exists to the study of halos which can be perfomed with \\emph{Chandra} for such strongly absorbed sources like 1E 1740.7 -- 2942, since the absorbing $N_{H}$ column is also proportional to the amount of dust scattering. Comparing the halo brightness with X-ray absorption would allow to directly quantify the amount and the density of that molecular clouds; a detailed analysis of the relation between halo and scattering dust properties is however beyond the aim of this paper.  Finally we would like to stress that, in order to correctly compare the radial profile of a source which is supposed to show an X-ray scattering halo or other extended features with that of a point-like source it will be necessary a detailed comparison between simulated and measured PSF wings profiles. However, until a complete calibration of the PSF wings are available, using real observations as a template remains our best option."
    },
    "0208/astro-ph0208543_arXiv.txt": {
        "abstract": "We present the high-resolution spectrum of the accretion-powered millisecond pulsar \\J\\/ during its 2002 outburst, measured using the Low Energy Transmission Grating Spectrometer onboard the {\\em Chandra X-ray Observatory}.  The 1.5--25.3~\\AA\\/ (0.5--8.3~keV) \\Ch\\/ spectrum is well fit by a power-law $+$ blackbody model with photon index $\\Gamma=1.55\\pm$0.03, blackbody temperature $kT_{\\rm bb}=0.65\\pm$0.03~keV, and blackbody normalization $R_{\\rm bb,km}/d_{\\rm 10kpc}=7.6\\pm0.8$.  No emission or absorption features are found in the high-resolution spectrum, with a 3$\\sigma$ equivalent width upper limit of $<0.007$~\\AA\\/ at 1.5~\\AA\\/ and $<0.12$~\\AA\\/ at 24~\\AA.  The neutral absorption edge depths are consistent with the estimated interstellar absorption along the line of sight to the source.  We found no orbital modulation of the 2--10~keV X-ray flux, to a 3$\\sigma$ upper limit of 1.1\\%, which implies an upper limit on the binary inclination angle of $i\\lesssim 85^{\\circ}$ for a Roche-lobe--filling companion.  We also present the broadband spectrum measured over the course of the outburst by the {\\em Rossi X-ray Timing Explorer} (\\XTE).  The \\XTE\\/ spectrum of \\J\\/ is also well fit with a power-law $+$ blackbody model, with average values of $\\Gamma=1.76\\pm$0.03, $kT_{\\rm bb}=0.66\\pm$0.06~keV, and $R_{\\rm bb,km}/d_{\\rm 10kpc}=5.9\\pm1.3$ in the 2--50~keV energy range.  The blackbody flux remained constant over the course of the outburst, while the power-law flux was strongly correlated to the (decreasing) flux of the source.  We find that the difference in power-law photon indices measured from \\Ch\\/ and \\XTE\\/ spectra can be explained by a change in the power-law photon index at low energies. ",
        "introduction": "Millisecond pulsars (MSPs) have long been considered one of the possible endpoints of low-mass X-ray binary (LMXB) evolution.  The neutron star (NS) is thought to be spun-up to a millisecond period by accretion from its low-mass companion.  After the accretion phase has ended, the NS may turn on as a radio MSP.  In the last five years, this theory has been confirmed with the identification of three accretion-powered MSPs, SAX J1808.4$-$3658, XTE~J1751$-$305 and \\J\\/ \\citep{wv98,cm98,mss+02,gcm+02}.  Interestingly, all three accretion-powered MSPs are in short period binaries, with the two most recently discovered having binary periods $\\approx$43 min \\citep{mss+02,gcm+02}.  These short periods place the MSPs XTE~J1751$-$305 and \\J\\/ in the class of ultracompact binaries, defined as having orbital periods $\\lesssim$80 min.  Ultracompact binaries require hydrogen-deficient or degenerate donors \\citep*{nrj86}.  Recently, O and Ne emission and absorption features were discovered in two known and three suspected ultracompact systems \\citep*[][]{scm+01,jpc01,jc02}.  These results led the authors to conclude that the donor stars are degenerate C-O WDs.  On the other hand, the {\\em XMM-Newton} spectrum of XTE~J1751$-$305 did not show emission or absorption features \\citep{mwm+02}.  In addition, no unusual abundances were required to fit the neutral absorption edges. This result is consistent with the suggestion of \\citet{mss+02} that the donor in XTE~J1751$-$305 is a He WD \\citep[see also,][]{b02}.  \\J\\/ was discovered in April 2002 by the All Sky Monitor onboard the {\\em Rossi X-ray Timing Explorer} \\citep[\\XTE;][]{r02}.  Further \\XTE\\/ observations detected 185~Hz pulsations modulated by a 43.6-min binary orbit \\citep*{rss02,gcm+02}.  Radio and optical counterparts with positions consistent with the X-ray position were also detected \\citep*{rdm02,ggh02,c02}.  An optical spectrum of \\J\\/ revealed emission lines from \\ion{C}{3}/\\ion{N}{3} $\\lambda$4640-4650 and H$\\alpha$ $\\lambda$6563 \\citep{ccg+02}.  Given the optical detection of emission lines, and the ultracompact nature and high-Galactic latitude of the source, \\J\\/ is an ideal candidate for a high-resolution X-ray spectroscopic study to search for emission and absorption features similar to those seen in other ultracompact systems.  In this letter, we present results from a Director's Discretionary Time observation of \\J\\/ with the {\\em Chandra X-ray Observatory}, as well as spectral results from the \\XTE\\/ pointed observations throughout the outburst.  \\vspace{0.2in} ",
        "conclusions": "We have found that the spectrum of \\J\\/ is well described by a power-law $+$ blackbody model over limited energy ranges, with interstellar absorption consistent with $N_{\\rm H}=7.6\\times 10^{20}$~cm$^{-2}$, the expected hydrogen column density along the line of sight to the source.  The \\Ch\\/ spectrum shows no prominent emission or absorption lines in the spectrum \\citep[similar to XTE~J1751$-$305;][]{mwm+02}, with an EW limit which varies from $<0.007$~\\AA\\/ at 1.5~\\AA\\/ to $<0.12$~\\AA\\/ at 24~\\AA.  No orbital modulation of the X-ray flux was detected in either the \\Ch\\/ or \\XTE\\/ data.  From the lack of X-ray eclipses, we set an upper limit on the binary inclination of $i\\lesssim 85^{\\circ}$ for a Roche-lobe--filling companion.  In addition, the absence of dips in the X-ray lightcurve suggests that the source is most likely at an inclination of $i\\lesssim60^{\\circ}-70^{\\circ}$ \\citep*[see, e.g.,][]{fkl87,wnp95}.  The blackbody spectral component in LMXBs is generally attributed to the neutron star, possibly from a ``hot spot'' where the accretion column meets the neutron star surface.  From our fits, we find a blackbody radius of $R_{\\rm bb,km}=(5.3-7.7)\\ d_{\\rm 10kpc}$.  Given the lower limit on the distance of 5~kpc \\citep{gcm+02}, we find a lower limit on the blackbody radius of $R_{\\rm bb}>2.7$~km.  We note that this lower limit does not include a correction for the conversion between the color temperature and the effective temperature of the neutron star atmosphere \\citep*[see,][for a discussion]{lvt93}.  This correction could increase the inferred radius by a factor of $\\approx2$.  But even after this correction, the inferred radius is not consistent with canonical NS models for a distance of 5~kpc suggesting a ``hot spot'' emission region.  During the outburst, we found that the blackbody flux remained constant, while the power-law flux declined.  In addition, when the combined \\Ch\\/ and \\XTE\\/ spectrum is fit with a Comptonization model, we find that the blackbody temperature, $kT_{\\rm bb}$, is significantly greater than the seed photon temperature of the Comptonization model, $kT_0$.  This evidence suggests that the power-law and blackbody components are associated with different emission processes.  We compared our results with two physical models that were developed to explain the spectral properties of the accretion-powered MSP SAX~J1808.4$-$3658 \\citep{gdb02,tcw02}.  Both models were formulated using data from the 1998 outburst of SAX~J1808.4$-$3658, but come to very different conclusions regarding the relationship between the power-law or Comptonization component and the blackbody. \\citet{gdb02} assumed that the blackbody was the source of the seed photons (i.e, $kT_0 = kT_{\\rm bb}$) and proposed that the Comptonization component comes from the shock heated accretion column with soft photon input from the ``hot spot'' blackbody component. Such a model implies that the blackbody and power-law/Comptonization component fluxes should be correlated, which was the case for SAX~J1808.4$-$3658 but was not seen for \\J.  On the other hand, \\citet{tcw02} found that $kT_0 << kT_{\\rm bb}$ and suggested that the accretion disk provided the majority of the soft photon input to a spherical Compton cloud, of radius $R\\approx~60$~km, from which the Comptonization component originates.  The \\citet{tcw02} model allows for the the blackbody and power-law/Comptonization component to vary independently. \\citet{tcw02} also suggested that the low ($\\tau_{0}\\approx4$) optical depth of the Comptonizing region found for SAX~J1808.4$-$3658 allows for the pulsations of the NS to be detected, while the higher optical depths ($\\tau_{0}>5$) reported in more luminous LMXBs suppress the pulsation amplitudes.  We find a best-fit $\\tau_{0}\\approx2.7$ for \\J, consistent with this explanation.  However, we note that \\citet{mwm+02} did not find an acceptable fit of the {\\em XMM} spectra of the millisecond X-ray pulsar XTE~J1751$-$305 using a Comptonization model.  While this model seems more appropriate for the spectral results of \\J, \\citet{tcw02} do not discuss what emission component is responsible for the pulsations.  It is important to realize that both models are based on data with a low-energy bound of 2~keV, which is above the peak flux of the thermal components.  The lower energy range of \\Ch\\/ allows us to make a more robust measurement of $kT_0$ for \\J\\/ than was possible for the \\XTE\\/ spectrum of SAX~J1808.4$-$3658.  We conclude that neither model is completely appropriate to explain both our results.  More observations of these sources throughout outburst are necessary in order to fully understand their phase-averaged spectral properties.  We suggest that in-depth data analysis, including pulse-phase spectroscopy, will provide more insight into the relationship between the various spectral components.  The pulse-phase resolved spectral analysis of XTE~J0929$-$314 will be presented in a later paper.  We find that the best-fit photon index was significantly different between the \\Ch\\/ and \\XTE\\/ fits with $\\Gamma=1.55-1.62$ for the \\Ch\\/ fits, dependent on the assumed instrumental C abundance, and $\\Gamma=1.76$ for the \\XTE\\/ fits.  Since fits with and without the pileup model give consistent parameter values, we reject the possibility that pileup causes the lower \\Ch\\/ photon index.  The combined \\Ch\\/ and \\XTE\\/ spectrum of \\J\\/ is well fit by either a Comptonization $+$ blackbody model (which implicitly includes a low-energy turnover from the Comptonization component), or a broken power-law $+$ blackbody model, with break energy 1.4--4.4~keV.  The high-energy ($E>3$~keV) \\Ch\\/ spectrum is consistent with the \\XTE\\/ spectral results, indicating that instrumental differences alone do not give rise to the spectral turnover.  Based on these results, we suggest that the difference in best-fit photon indices between \\Ch\\/ and \\XTE\\/ arises from spectral turnover at low energies.  While a power-law model is a reasonable approximation to Comptonization at high energies, extrapolating the power law to low energies leads to a predicted flux that diverges at zero energy.  Therefore, the power-law must turnover at some energy for the integrated to flux to be finite.  In addition, the difference in the photon index between the \\Ch\\/ and \\XTE\\/ spectral fits is similar to that noted by \\citet{mwm+02} between the \\XTE\\/ and {\\em XMM} spectral fits of XTE J1751$-$305. From their {\\em XMM} EPIC spectrum, \\citet{mwm+02} found a best-fit power-law photon index of 1.44, while \\citet{mss+02} found a best-fit photon index of 1.7--1.9 using \\XTE.  We interpret the difference between the {\\em XMM} and \\XTE\\/ spectral results for XTE J1751$-$305 as evidence of low energy turnover in the broadband spectrum of that source, providing independent support that the turnover is astrophysical in origin."
    },
    "0208/astro-ph0208069_arXiv.txt": {
        "abstract": "We have examined the soft X-ray plus optical/UV spectrum of the nearby isolated neutron star \\rxj, comparing with detailed models of a thermally emitting surface. Like previous investigators, we find the spectrum is best fit by a two-temperature blackbody model. In addition, our simulations constrain the allowed viewing geometry from the observed pulse fraction upper limits. These simulations show that \\rxj\\ is very likely to be a normal young pulsar, with the non-thermal radio beam missing Earth's line of sight. The SED limits on the model parameter space put a strong constraint on the star's $M/R$. At the measured parallax distance, the allowed range for $\\Mstar=1.5\\Msun$ is $\\Rstar=13.7\\pm0.6\\km$. Under this interpretation, the EOS is relatively stiff near nuclear density and the `Quark Star' EOS posited in some previous studies is strongly excluded. The data also constrain the surface $T$ distribution over the polar cap. ",
        "introduction": "\\rxj, discovered by \\citet{walt96}, is the nearest and brightest known neutron star not showing emission dominated by non-thermal magnetospheric processes. As such it offers a unique opportunity to study the bare thermal surface emission. Measurements of the spectrum can probe the neutron star mass (\\Mstar) and radius (\\Rstar), constraining the high density equation of state (EOS). Since the discovery, there have been several intensive observing campaigns covering the optical-UV (\\HST) and most recently the detailed soft X-ray spectrum (\\CXO). An initial $50\\ks$ Low-Energy Transmission Grating (LETG) spectrum showed a broad band spectrum remarkably consistent with a simple blackbody \\citep{bur01}, although hints of spectral features were suggested \\citep{vanKer02}. A deeper observation using $450\\ks$ of Director's Discretionary Time (DDT) was made. This unique data set has been the subject of prompt study; several authors show that lines in the spectrum are undetectable, while pulse searches have placed stringent limits on the observed soft X-ray pulse fraction \\citep{ran02,dra02}. These data have been variously interpreted, including the widely reported speculation (based on the X-ray spectrum alone) that \\rxj\\ might be a bare quark star \\citep[esp.][]{dra02}.  Despite the very stringent constraints placed on the X-ray pulse fraction, \\citep[$<4.5\\%$ at $99\\%$ confidence, including a ${\\dot P}$ search][]{ran02}, there is strong evidence that \\rxj\\ is a rotation-powered pulsar. \\citet{vanKer01b} discovered an H$\\alpha$ nebula surrounding the neutron star, concluding that it could be best interpreted as a bow-shock nebula powered by a relativistic wind of $e^\\pm$ generated by pulsar spindown. The bow shock geometry then provides an estimate of the spindown power $\\dot{E} = I \\Omega {\\dot \\Omega} = 8\\times 10^{32}\\es~d_{140}^3$ \\citep{vanKer01b}. Adopting magnetic dipole braking at constant $B$, this gives $\\dot{E} = 10^{34} (B_{12} \\tau_6)^{-2}\\es$ for a surface dipole field $10^{12}B_{12}\\Gauss$ and characteristic age $10^6\\tau_6\\y$, suggesting $B_{12} \\tau_6 \\sim 3$.  A critical parameter in the discussion of this source is the distance, which has been the subject of some controversy. Initial estimates from \\HST\\ astrometry gave a parallax distance of $61\\pc$ \\citep{wal01}. \\citet{kap02}, however re-analyzed these data, deriving $d=143\\pc$. A fourth \\HST\\ observation appears to have resolved this discrepancy, giving an overall measurement of $d=117\\pm 12\\pc$ \\citep{wal02}; we adopt this value. ",
        "conclusions": ""
    },
    "0208/astro-ph0208225_arXiv.txt": {
        "abstract": "We have identified seven independent pulsation modes in the helium-atmosphere variable (DBV) white dwarf star CBS 114, based on 65 hours of time-resolved CCD photometry from the 0.75-m telescope at SAAO. We interpret these pulsations as non-radial g-modes with the same spherical degree $\\ell$=1, as suggested by the mean period spacing of $37.1\\pm0.7$ seconds. We use a genetic-algorithm-based fitting method to find the globally optimal model parameters, including the stellar mass ($M_*=0.73\\ M_{\\odot}$), the effective temperature ($T_{\\rm eff}=21,000$~K), the mass of the atmospheric helium layer ($\\log[M_{\\rm He}/M_*]=-6.66$), and the central oxygen mass fraction ($X_{\\rm O}=0.61$). The latter value implies a rate for the \\cago reaction near $S_{300}=180$~keV~b, consistent with laboratory measurements. ",
        "introduction": "Almost every star in our galaxy will eventually become a white dwarf. Since they are relatively simple compared to main-sequence stars, white dwarfs provide one of the best opportunities for learning about stellar structure and evolution. The helium-atmosphere variable (DBV) white dwarfs are among the simplest of all, with a surface helium layer surrounding a degenerate C/O core. The internal composition and structure of white dwarfs formed through single-star evolution is largely determined by the relative rates of two nuclear reactions that compete for the available helium nuclei during core helium burning in the red giant phase of evolution: the $3\\alpha$ and \\cago reactions. The $3\\alpha$ rate is well-constrained from laboratory measurements, but the \\cago rate is still very uncertain. Recent advances in the analysis of asteroseismological data on DBV stars \\cite{mwc01} now make it possible to obtain precise measurements of the central C/O ratio, providing a more direct way to determine the \\cago rate at stellar energies. The first application of this new method to the star GD~358 implied a reaction rate that is significantly higher than most extrapolations from laboratory data \\cite{msw02}. Fortunately, each pulsating white dwarf can provide an independent measurement of the reaction rate, so the analysis of asteroseismological data for additional DBV stars would be useful. ",
        "conclusions": "Our optimal model for CBS~114 has a higher mass and a lower central oxygen mass fraction than the optimal model for GD~358 \\cite{mwc01}.  Turning these values into a measurement of the \\cago rate requires the calculation of evolutionary internal chemical profiles like those of \\inlinecite{sal97}. A model of the internal chemical profile with the same mass as our fit to CBS~114 requires a rate for the \\cago reaction of $S_{300}=177\\pm3$~keV~b (internal uncertainty) to match the derived central oxygen abundance within the 1$\\sigma$ limits (M.~Salaris, private communication). This value is consistent with (but much more precise than) the rate derived from recent high-energy laboratory measurements ($S_{300}=165\\pm50$~keV~b; \\opencite{kun02}). By contrast, the rate previously derived from a similar treatment of the white dwarf GD~358 was significantly higher ($S_{300}=370\\pm40$~keV~b). However, \\inlinecite{met02b} identified a systematic difference in the analysis of GD~358 and CBS~114. After correcting this difference in treatment, the two sets of data yielded reaction rates that are both consistent with laboratory measurements, and marginally consistent with each other.  \\inlinecite{bf02} have proposed an alternative model to explain the pulsation spectrum of GD~358, involving two composition transition zones in the surface helium layer, as suggested by time-dependent diffusion calculations similar to those done by \\inlinecite{dk95}. \\citeauthor{bf02} note that their fit, which has a pure carbon core, is actually worse with a core of either pure oxygen or a uniform C/O mixture. An important test of their alternative model will be whether it can match the pulsation spectrum of CBS~114 with the same set of assumptions used in the analysis of GD~358, as \\inlinecite{met02b} has done with the model used here."
    },
    "0208/astro-ph0208013_arXiv.txt": {
        "abstract": "Supernova remnants may exhibit both thermal and nonthermal X-ray emission.  In a previous study with ASCA data, we found that the middle-aged supernova remnant RCW 86 showed evidence for both processes, and predicted that observations with much higher spatial resolution would distinguish harder X-rays, which we proposed were primarily synchrotron emission, from softer, thermal X-rays.  Here we describe {\\it Chandra} observations which amply confirm our predictions.  Striking differences in the morphology of X-rays below 1 keV and above 2 keV point to a different physical origin.  Hard X-ray emission is correlated fairly well with the edges of regions of radio emission, suggesting that these are the locations of shock waves at which both short-lived X-ray emitting electrons, and longer-lived radio-emitting electrons, are accelerated.  Soft X-rays are spatially well-correlated with optical emission from nonradiative shocks, which are almost certainly portions of the outer blast wave.  These soft X-rays are well fit with simple thermal plane-shock models.  Harder X-rays show Fe K$\\alpha$ emission and are well described with a similar soft thermal component, but a much stronger synchrotron continuum dominating above 2 keV, and a strong Fe K$\\alpha$ line. Quantitative analysis of this line and the surrounding continuum shows that it cannot be produced by thermal emission from a cosmic-abundance plasma; the ionization time is too short, as shown both by the low centroid energy (6.4 keV) and the absence of oxygen lines below 1 keV. Instead, a model of a plane shock into Fe-rich ejecta, with a synchrotron continuum, provides a natural explanation.  This requires that reverse shocks into ejecta be accelerating electrons to energies of order 50 TeV. We show that maximum energies of this order can be produced by radiation-limited diffusive shock acceleration at the reverse shocks.  In an Appendix, we demonstrate that an explanation of the continuum as due to nonthermal bremsstrahlung is unlikely. ",
        "introduction": "While most shell supernova remnants (SNRs) show thermal X-ray spectra dominated by emission lines of highly ionized species of C, N, O, Ne, Mg, Si, S, and Fe, a few show featureless spectra which can be well understood as synchrotron emission: SN 1006 \\citep{Koyama95, dyer01}; G347.3-05 \\citep{koyama97, slane99}, G266.2-1.2 \\citep{slane01}, and G28.6-0.1 \\citep{bamba01}. The absence of lines means that a synchrotron interpretation is almost unavoidable; impossibly low abundances and/or peculiar physical conditions would be required to suppress lines completely in a thermal plasma \\citep[e.~g.,][]{hss86, laming98}, while nonthermal bremsstrahlung implies the presence of electrons that should excite lines as well, and inverse-Compton scattering of any photon population would result in far too hard a spectrum.  If synchrotron emission can dominate the spatially-integrated X-ray spectrum of a few remnants, it is likely to play some role in many more, e.~g., by totally dominating thermal emission in selected locations within them, or by overwhelming thermal continua (but not lines) in thermally-emitting regions.  The signature of an X-ray spectrum with both thermal and synchrotron radiation would be unusually weak lines, diluted by the synchrotron continuum.  We interpreted weak X-ray lines in the remnant RCW 86 by such combination of thermal and synchrotron emission, based on ASCA observations \\citep[hereafter BRRD]{brrd01}. A very similar interpretation of RCW 86 spectra was also proposed by \\citet{bamba00}, based on the ASCA Performance Verification data.    RCW 86 (G315.4--2.3) is a large (42\\arcmin\\ in diameter) shell-like SNR with a moderate blast wave speed in the range of 400--900 km s$^{-1}$, determined from observations of Balmer-dominated shocks which almost completely encircle the remnant \\citep{lobl90,smith97,ghrs01}. At a kinematic distance of 3 kpc, its large size and the moderate blast wave speed imply a mature ($\\sim 10^4$ yr old) remnant \\citep{rosado96}.  The bright complex of optical emission filaments at its SW corner, to which the optical designation ``RCW86'' actually refers, has been generally interpreted as a region where the blast wave impacted a dense ($\\sim 10$ cm$^{-3}$) interstellar cloud. This is also where X-ray and radio emission are the brightest.  Radio observations by the MOST telescope at 843 MHz \\citep{wg96}, with an angular resolution of about $45^{\\prime\\prime}$, and by the Australia Telescope Compact Array (ATCA) at 1.3 GHz \\citep{dsm01}, with $8^{\\prime\\prime}$ resolution, show a complete shell of varying brightness.  The radio morphology in the SW region of the remnant is somewhat unusual, consisting of a bright ridge of emission mostly interior to optical and soft X-ray filaments which mark the location of the blast wave. ASCA Performance Verification X-ray spectra showed strikingly weak X-ray lines, much weaker than expected from a normal (solar) abundance plasma, with the exception of the Fe K$\\alpha$ line at 6.4 keV \\citep{vink97}. These results were confirmed by deeper ASCA observations (BRRD) and by more recent BeppoSax observations \\citep{bvfms00}. A substantial Fe K$\\alpha$ equivalent width in the SW suggests enhanced Fe abundance with respect to solar, irrespective of the origin of the underlying high-energy continuum, which led \\citet{bvfms00} to consider a possibility that we are observing metal-enriched supernova (SN) ejecta.    In our earlier work (BRRD), we used long ASCA observations to deduce the presence of three spectral components: a soft thermal component ($kT \\sim 0.8$ keV), with a relatively long ionization timescale ($\\tau \\equiv n_e t \\sim 2 \\times 10^{11}$ s cm$^{-3}$) and with approximately solar abundances, modeled as a plane shock (XSPEC model {\\it pshock}); a synchrotron component, modeled as the high-energy tail of the radio spectrum with XSPEC model {\\it srcut}; and a hot thermal component ($kT \\sim 5$ keV, $\\tau \\sim 5 \\times 10^{8}$ s cm$^{-3}$) required to account for Fe K$\\alpha$ emission.  These components varied in relative strength in three regions in the southwest corner of RCW 86.  The {\\it srcut} model is just the synchrotron emissivity of a power-law electron distribution with an exponential cutoff, calculated numerically \\citep{rk99}; it is characterized by a single free parameter, the peak frequency $\\nu_{\\rm rolloff}$ radiated by electrons with the $e$-folding energy of the exponential cutoff.  The fitted values of $\\nu_{\\rm rolloff}$ varied from 1 to $4 \\times 10^{16}$ Hz, implying (for $B \\sim 10 \\ \\mu$G) maximum electron energies of order 20 TeV.  The synchrotron fitting used as inputs the radio fluxes for each region measured from the MOST image \\citep{wg96}, and an assumed constant spectral index of 0.6.    This analysis explained the ASCA data reasonably well, but with ASCA's spatial resolution of several arcminutes, we were unable to perform several more rigorous tests of the hypothesis of synchrotron X-ray emission.  Our spectral fits suggested that the soft thermal emission is due to nonradiative shocks as identified by H$\\alpha$ emission, while the synchrotron X-rays should show a spatial distribution similar to radio emission.  In an attempt to verify these predictions, we observed RCW 86 with the {\\it Chandra} X-ray Observatory in February 2001. We report here on these observations, which not only confirm our predictions, but allow us to associate synchrotron X-ray emission with Fe-rich SN ejecta seen through Fe K$\\alpha$ line emission, and identify all high-energy continuum as synchrotron emission. ",
        "conclusions": "The major conclusion of our paper is that the detailed morphology of soft and hard X-rays in RCW 86 strikingly supports the hypothesis of different origins.  Our spectral analysis confirms that the hard X-rays are best described as a synchrotron continuum with Fe K$\\alpha$ emission that must come from a thermal component with a substantial overabundance of iron with respect to solar.  This result implies that some of the original supernova ejecta are still unmixed after $\\sim 10^4$ yr, and that the reverse shock into those ejecta can also accelerate electrons to X-ray synchrotron-emitting energies of order 50 TeV.   Both these conclusions substantially complicate the task of understanding the X-ray spectra of middle-aged supernova remnants. Apparently even after ages of order $10^4$ years, remnants are not in the simple Sedov phase, and the possibility of synchrotron X-ray emission confusing the analysis of thermal emission cannot be neglected.  The only ultimate solution to this problem requires X-ray spectroscopy with much higher energy resolution.  (Grating resolution is adequate, but grating observations of large extended sources with {\\it Chandra} or {\\it XMM-Newton} are virtually uninterpretable.)  The {\\sl Astro-E2} microcalorimeters hold out the possibility of spatially resolved spectroscopy with enough resolution to allow the inference of temperatures from line complexes alone without requiring any assumptions about the continuum.  In that case, one can immediately determine the extent of synchrotron contributions to the continuum.   With {\\it Chandra's} spatial resolution, we can for the first time discern discrete shock features and minimize projection effects, which should aid substantially in the detailed thermal modeling that will be the next step in understanding RCW 86.  The hard X-ray emission appears to delineate particular shocks, where both radio and X-ray emitting electrons are accelerated.  The greater extent of radio emission behind such features is expected from preferential energy losses on the much more energetic X-ray emitting electrons.  The width of the X-ray emission is quite consistent with this interpretation, for expected post-shock velocities of hundreds of km s$^{-1}$ and magnetic field strengths of order 10 $\\mu$Gauss.   The observed rolloff frequencies in areas containing synchrotron continua can be understood in terms of radiative-loss-limited shock acceleration, but simple estimates require favorable conditions to reach the inferred electron energies.  In particular, nearly perpendicular shocks reach higher energies more readily, and this may be one factor that distinguishes shocks that produce synchrotron X-ray continua from those which do not.  We repeat, however, that the {\\it srcut} model is highly simplified, and fits with more elaborate models might produce somewhat different values for $\\nu_{\\rm rolloff}$.   Progress in understanding RCW 86, and the general problems it raises for the understanding of supernova remnants, requires further observations in various bands.  {\\it XMM-Newton} observations can obtain significantly higher signal-to-noise ratios at high energies, allowing testing in other regions of our claim of an association of Fe K$\\alpha$ emission with regions of hard continuum.  Optical and infrared observations in principle should be able to detect synchrotron continuum at a level exactly predictable from our fits, but distinguishing it from the confusing emission might prove extremely challenging.  Since our fitted values of $\\nu_{\\rm rolloff}$ are far above optical frequencies, we expect that there should be little difference between the optical or IR morphology of synchrotron emission and the radio morphology.  A strategy to search for extremely faint diffuse emission then might center on sharp features in the radio image, such as the edge coincident with a hard X-ray filament in the middle of chip S3."
    },
    "0208/astro-ph0208363_arXiv.txt": {
        "abstract": "We determine the torque exerted in a steady state by an external potential on a three-dimensional gaseous disk at a non-coorbital corotation resonance.  Our model accounts for the feedback of the torque on the surface density and vorticity in the corotation region, and assumes that the disk has a barotropic equation of state and a nonzero effective viscosity.  The ratio of the torque to the value given by the formula of Goldreich \\& Tremaine depends essentially on a single dimensionless parameter, which quantifies the extent to which the resonance is saturated.  We discuss the implications for the eccentricity evolution of young planets. ",
        "introduction": "The forcing of a gaseous disk by a gravitational perturber at a resonance can result in a strong response and an interchange of energy and angular momentum between the perturber and the disk.  A corotation resonance arises where the pattern speed of the forcing matches the local angular speed of disk material.  This occurs in the context of galaxies where a stellar bar or spiral arms force motions in the interstellar medium (Binney \\& Tremaine 1987).  Corotation resonances also arise when satellites orbit within a disk, as occurs with young planets in protoplanetary disks, or moons in planetary ring systems (Goldreich \\& Tremaine 1979, 1980; hereafter GT79 and GT80 respectively).  For planets, two types of corotation resonance need to be distinguished.  A coorbital corotation resonance occurs where the orbital period of the planet matches the orbital period of disk material. For a planet with a circular orbit, this is the only type of corotation resonance that can arise.  The second type of corotation resonance is non-coorbital and occurs if the planet executes an eccentric orbit.  The forcing due to a planet with an eccentric orbit can be decomposed into a series of rotating components of various strengths and pattern speeds (GT80).  Bar or spiral galaxies, on the other hand, give rise to non-coorbital resonances.  The analyses of the coorbital and non-coorbital resonances are different.  In the coorbital case, which has recently been considered by Masset (2001, 2002) and Balmforth \\& Korycansky (2001), the forcing is stronger and involves multiple Fourier components.  There is more likely to be a reduction in density caused by the tendency of the planet to open a gap in the disk.  In this paper, however, we are concerned with non-coorbital corotation resonances, which are critical in determining the eccentricity evolution of planets resulting from planet--disk interactions (GT80), and are therefore likely to be important in explaining the eccentricities of many of the observed extrasolar planets.  For a disk in which a gap is opened by a planet having a small orbital eccentricity, corotation resonances tend to damp the planet's eccentricity, while Lindblad resonances cause it to grow.  The effects of each type of resonance are strong, but are nearly equal in magnitude, to the extent that they nearly cancel.  The balance is slightly in favor of eccentricity damping, if the corotation resonances operate at maximal efficiency (i.e., are unsaturated).  The final outcome of eccentricity evolution depends on the details.  The disturbances of the disk caused by a corotation resonance remain localized to the corotation region (GT79). They are unable to propagate away from the resonance, as occurs in the case of Lindblad resonances.  Instead, they act back locally on the disk and may change its density and vorticity in such a manner as to reduce (saturate) the corotation torque (e.g., Ward 1991).  However, the effective turbulent viscosity of the disk could lessen the saturation by limiting the back-reaction on the disk (Ward 1992).  Full saturation occurs when the torque is reduced to zero, but even a small degree of saturation (5\\%) could change the sense of eccentricity evolution in the GT80 model from decay to growth (Goldreich \\& Sari 2002).  In this paper, we develop a detailed model for the saturation of a non-coorbital corotation resonance in a viscous accretion disk. We determine how the strength of the corotation torque varies with the strength of the tidal forcing and the effective viscosity of the disk. ",
        "conclusions": "\\label{Summary and discussion}  We have determined the torque exerted in a steady state by an external potential on a three-dimensional gaseous disk at a non-coorbital corotation resonance.  Our model accounts for the feedback of the torque on the surface density and vorticity in the corotation region, and assumes that the disk has a barotropic equation of state and a nonzero effective viscosity.  The torque formula of Goldreich \\& Tremaine (1979) (eq.~[\\ref{tgt}]) must be modified by a reduction factor $t_{\\rm c}(p)$, plotted in Figure~1, which quantifies the extent to which the resonance is saturated.  This factor depends essentially on a single dimensionless parameter $p$, defined in equation (\\ref{p}), which measures the strength of the potential $\\Psi$ relative to the viscosity $\\nu$ (i.e., the nonlinearity relative to the dissipation).  In Section~\\ref{Numerical solution}, we determined that the characteristic radial width of the resonance in the limit of large $p$ (low viscosity) is $\\delta\\sim\\sqrt{\\Psi}/\\Omega$.  (This refers to the scale of Reynolds and viscous stresses; the tidal torque is spread over a region of characteristic width $c/\\kappa$.)  We found that the torque is reduced by a factor $\\sim p^{-3/2}$, which can be regarded as the ratio of the viscous diffusion rate $\\sim \\nu/\\delta^2$ over the libration region to the libration rate $\\sim m|d\\Omega/dr|\\delta$. In this regime, our results are broadly consistent with the saturation model of Ward (1992), which involves the same ratio of rates, but applied to the coorbital region.  Earlier, Goldreich \\& Tremaine (1981) had argued that, in a disk of collisional particles, saturation would occur when (in our notation) $p\\gg1$.  Goldreich \\& Sari (2002) have recently developed an evolutionary model for planetary eccentricity in which eccentricity growth occurs through a finite-amplitude instability.  For infinitesimal eccentricity, the damping caused by corotation resonances just overcomes the eccentricity growth due to Lindblad resonances.  Above a critical level of eccentricity, however, the corotation resonances become sufficiently saturated that growth occurs.  Using our evaluation of the saturation function $t_{\\rm c}(p)$, Goldreich \\& Sari (2002) were able to determine the critical value of eccentricity required within the context of a specific disk model.  The results in Section~\\ref{Application to eccentric resonances} and Table~1 show that for a Jupiter-mass planet orbiting within a typical protoplanetary disk, eccentricities of a few percent are adequate to saturate all first-order eccentric corotation resonances except those of the lowest $m$-values ($m\\la4$).  To achieve a $5\\%$ reduction in torque at such resonances, as may be sufficient to change the balance from eccentricity damping to growth (GT80), requires only eccentricities of $1\\%$ or less.  Exactly which resonances contribute most to eccentricity evolution depends on the extent of the gap cleared by the planet, which in turn depends on the mass ratio and the properties of the disk.  As seen in Table~1, the largest corotation torques are those associated with the highest $m$-values, and they are the easiest to saturate.  A lower-mass planet that opens a smaller gap may excite many resonances so that the dominant torque comes from approximately the cut-off $m$-value, of order $r/H$ (GT80).  In that case, $63\\%$ saturation is achieved optimistically at the torque cut-off when $e\\ga1.4\\,q^{-1}\\alpha^{2/3}(H/r)^3$.  Note that, although the large value of $m$ is advantageous for saturation, the low mass of the planet is unfavorable (see eq.~[\\ref{e_063}]). On the other hand, if the planet mass is small enough so that there is no gap in the disk, then the coorbital resonances cause eccentricity decay (Ward 1988; Artymowicz 1993).  We have neglected a number of potential complications.  Numerical simulations of a Jupiter-mass planet in a circular orbit with the disk parameters adopted in Section~\\ref{Application to eccentric resonances} suggest that the disk edge is not sharp (i.e., its radial extent is much greater than $H$; see Figure~1 of Lubow, Seibert, \\& Artymowicz 1999).  It is evident, however, that within the radial extent of the planet's Roche lobe, material is captured by the planet and the resonant effects considered here do not play a role. It is not clear which eccentric corotation resonances are excited in the presence of an eccentric planet.  The complications are due at least in part to nonlinear effects other than those considered here. For example, shocks associated with the planet's wake cause a non-closure of streamlines in the vicinity of the disk edge, leading to a drift of material through the corotation regions.  In addition, material within the Roche lobe of the planet exerts torques that may contribute to the eccentricity balance.  As usual in accretion disk theory, our knowledge of the effective viscosity of the disk is limited.  In Section~\\ref{A simplifying assumption} we adopted a convenient assumption in order to make analytical progress.  Although we have succeeded in analyzing a three-dimensional disk, thereby generalizing existing theories, we assumed that the disk was barotropic.  The effects of buoyancy or baroclinicity on the corotation region remain uninvestigated.  In recent work we were able to verify analytical theories of the torques exerted at Lindblad and vertical resonances through direct numerical simulations (Bate et al. 2002).  It would be valuable to conduct simulations of a non-coorbital corotation resonance to test the findings of the present paper.  After eccentric corotation resonances saturate at small eccentricity, the eccentricity growth may be limited at intermediate values.  The limitation may be due to the overlap of resonances or alternatively through the excitation of higher order Lindblad resonances, some of which cause eccentricity damping.  A contribution to eccentricity damping occurs at an eccentric inner (outer) Lindblad resonance that lies outside (inside) the planet's orbit.  SPH simulations of eccentric-orbit binary stars suggest that little eccentricity growth via resonances occurs for eccentricities in the range of $0.5-0.7$ or higher (Lubow \\& Artymowicz 1993), and similar limits may occur for planets.  In conclusion, we have found a simple quantitative measure of the saturation of a corotation resonance in a gaseous disk.  This analysis suggests that planets may plausibly experience a net growth of eccentricity through their interaction with the disk in a variety of circumstances, provided that the eccentricity is not extremely small to begin with."
    },
    "0208/astro-ph0208155_arXiv.txt": {
        "abstract": "Many nearby stars are surrounded by a bright ring or disk of cold dust. Our calculations show that these disks and rings of dust are signposts of recent planet formation. Bright rings appear because dust associated with the formation of a planet absorbs and scatters light from the central star. The calculations explain the rings observed so far and predict that all nascent solar systems have dusty rings. ",
        "introduction": "Every planetary system forms from a thin disk of gas and dust in orbit around a young star.  In the planetesimal theory, planets grow from collisions and mergers of smaller bodies, planetesimals, embedded in the disk.  Protoplanets with radii of 100 km or more stir up the remaining planetesimals along their orbits.  A cascade of collisions among rapidly moving planetesimals produces a ring of dust grains, which slowly disappears as protoplanets grow into planets.  This entire process can lead to a solar system similar to our own \\citep{lis93,man00}.  Recent observations support this general picture.  The dusty disks around many nearby stars are as large or larger than our solar system \\citep{koe01,wei02}.  The dusty ring around the nearby 10 Myr old star, HR 4796A, has a thickness of $\\sim$ 15 AU and lies $\\sim$ 70 AU from its central star \\citep{jay98,koe98,sch99,gre00,tel00}.  The rings or partial rings around $\\epsilon$ Eridani, Vega, and other older stars have similar dimensions \\citep{den00,koe02,wil02}.  There have been few detailed numerical calculations of planet formation for comparison with these modern observations. Some calculations explore the early stages of planetesimal growth in a small range of disk radii \\citep{gre84,ws93}.  These models follow the development of a single large planet, typically the Earth or Jupiter.  Others use $n$-body simulations to investigate the last stages of planet formation, when large bodies coalesce to form a few planets \\citep{lis96,lev98,cha01}.  To simulate the formation of an entire planetary system, the calculation must span a large range of disk radii. Extending the calculations over a large fraction of the disk allows simulation of the diversity of observable phenomena in extrasolar planetary systems. Supercomputers now allow such {\\it multiannulus} planetesimal calculations covering a decade or more in disk radius \\citep{spa91,wei97,kor01}. The multiannulus calculations we discuss predict the behavior of a planet-forming disk during the early and intermediate stages of planet formation. The models yield images for comparison with observations of disks around nearby stars. ",
        "conclusions": "In our calculations, multiple rings tend to form when planets grow rapidly. Rapid growth occurs when the mass density of colliding bodies $\\rho_p$ is small and when the total mass in planetesimals $M_0$ is large. Bodies with low tensile strength also promote rapid growth and multiple ring production.  Calculations with $S_0 \\le 10^4$ erg g$^{-1}$ produce multiple rings more often than do calculations with $S_0 \\sim 10^6$ erg g$^{-1}$.  Multiple ring production is insensitive to $e_0$, the initial mass distribution, and other initial conditions.  Our calculations demonstrate that planet formation produces copious amounts of dust.  The dust production rate ranges from $\\sim 10^{18}$ g yr$^{-1}$ to $10^{21}$ g yr$^{-1}$. This dust absorbs and reradiates stellar energy with a relative luminosity of $L_{\\rm dust}/L_{\\star}$ $\\sim 10^{-5}$ to $10^{-3}$, comparable to observed luminosities for dusty disks surrounding nearby stars \\citep{hab01,spa01}. Our dust formation timescales of 10--100 Myr are comparable to the ages of nearby stars with dusty disks \\citep{hab01,son00,spa01}.  Dust first forms in large quantities when the largest bodies reach sizes of $\\sim$ 1000 km.  Dust disappears when disruptive collisions exhaust the supply of $\\sim$ 1 km bodies and radiative processes remove dust from the ring.  Thus, dust is concentrated in concentric rings which propagate outward through the disk as a function of time.  The outer edge of each ring marks the location where 1000 km objects are just starting to form; the inner edge marks the location where collisions have exhausted the supply of $\\sim$ 1 km bodies and dust has disappeared.  Thus, dusty rings are signposts for recent formation of 1000 km or larger planets surrounding a star.  \\vskip 6ex  We acknowledge a generous allotment, $\\sim$ 500 cpu days, of computer time on the Silicon Graphics Origin-2000 `Alhena' at the Jet Propulsion Laboratory through funding from the NASA Offices of Mission to Planet Earth, Aeronautics, and Space Science.  Advice and comments from J. Brauman and M. Geller greatly improved our presentation.  R. Mackey of the JPL supercomputing group assisted with the animations of Figure 3 and Figure 4."
    },
    "0208/astro-ph0208188_arXiv.txt": {
        "abstract": "This is the first of a series of three papers devoted to the calibration of a few parameters of crucial importance in the modeling of the evolution of intermediate-mass stars, with special attention to the amount of convective core overshoot. To this end we acquired deep $V$ and $R$ photometry for three globular clusters of the Large Magellanic Cloud (LMC), namely NGC~2173, SL~556 and NGC~2155, in the age interval 1--3 Gyr.  In this first paper, we describe the aim of the project, the VLT observations and data reduction, and we make preliminary comparisons of the color-magnitude diagrams with both Padova and Yonsei-Yale isochrones. Two following papers in this series present the results of a detailed analysis of these data, independently carried out by members of the Yale and Padova stellar evolution groups. This allows us to compare both sets of models and discuss their main differences, as well as the systematic effects that they would have to the determination of the ages and metallicities of intermediate-age single stellar populations. ",
        "introduction": "\\label{intro}  In an epoch of extraordinary discoveries on the high-z Universe, we still have gaps in our understanding of stellar evolution, which is in fact essential for the correct interpretation of the light of distant galaxies. Galactic star clusters have traditionally provided a major way to study stellar evolution.  However, our Galaxy only contains stars in age and chemical composition domains which reflect its particular history, and these are the domains which have been relatively well explored.  Extrapolation outside the limits of the well explored age and metallicity range is not always safe. For instance, it has become increasingly apparent in recent years that metal content can affect stellar evolution in unexpected ways. The UV upturn in elliptical galaxies is a case in point.  Studies of very metal rich stellar systems revealed that, contrary to simple expectations, old metal rich stellar populations do not simply become redder in all wavebands as they evolve, but rather produce a population of UV bright stars (Greggio \\& Renzini, 1990; Horch, Demarque \\& Pinsonneault, 1992; Fagotto et al. 1994; Yi, Demarque \\& Oemler 1998). On the other hand, understanding the evolution of extremely metal-poor and metal-free (Population III) stars will similarly be essential for the interpretation of primordial, high-z stellar populations.  In intermediate-age stellar populations, the color-magnitude diagram (CMD) and the luminosity function are affected by convective core overshoot.  While there is not a general agreement on the efficiency of this process, and its extent is not well established, most researchers agree that this parameter affects significantly the morphology of the CMD of these clusters and naturally the determination of their ages as well (e.g. Rosvick \\& VandenBerg 1998; Keller, Da Costa \\& Bessell 2001; Meynet, Mermilliod \\& Maeder 1993; Carraro et al. 1993; Demarque, Sarajedini \\& Guo 1994).  This uncertainty is thus also a problem for the spectral dating of distant stellar systems from their integrated light (Heap et al.  1998; Yi et al. 2000).  In addition, deep CMDs of intermediate-age clusters are essential to help disentangle the relative importance of other poorly determined parameters, such as mass loss during red giant branch (RGB) evolution, internal rotation (important for the more massive objects), and helium content. In particular, the adopted value of the parameter $\\delta Y/\\delta Z$ of helium enrichment significantly affects the mass luminosity relation of stellar models, and consequently their evolutionary lifetimes.  The papers in this series deal mainly with the problem of convective core overshoot in intermediate-age stellar populations, and therefore we will briefly review some of the recent work in the subject. Early observational arguments in favor of convective core overshoot come from clusters like Pleiades (Maeder \\& Mermilliod 1981; Mazzei \\& Pigatto 1989) and from other Galactic clusters, as discussed in great detail by Maeder \\& Mermilliod (1981) and Mermilliod \\& Maeder (1986). Barbaro \\& Pigatto (1984) and Chiosi \\& Pigatto (1986) argued for overshoot in stars with masses in the range $1.5 - 2.2 M_{\\odot}$ by pointing out that the base of the RGB is not well populated in clusters with age 1 -- 2 Gyr (whereas it is in older clusters), as if degenerate He-ignition and He-flash were avoided for this mass range, in contrast with classical models.  More recent studies include those of Aparicio et al. (1990), Carraro et al. (1993), Meynet et al. (1993); Rosvick \\& VandenBerg (1998); Demarque et al. (1994); Dinescu et al. (1995); Kozhurina-Platais et al. (1997) which deal with Galactic open clusters of ages 1.0 to 6.0 Gyr. All conclude that a certain amount of convective core overshoot is preferred to reproduce the CMDs. Being Galactic, all these clusters have metallicity close to solar, and therefore do not allow the possibility of testing the dependence of convective core overshoot on metallicity.  The Magellanic Clouds (MC) offer an unusual opportunity to test stellar evolution since populous clusters of a wide variety of ages and chemical compositions can be observed and compared at effectively the same distance. In particular, they contain young and intermediate-age metal-poor stellar populations that are absent in the Galaxy.  A number of young MC clusters (age $\\le$ 500 Myr) have been used by Lattanzio et al (1991), Vallenari et al. (1991, 1994), Stothers \\& Chin (1992), Brocato, Castellani \\& Piersimoni (1994), Chiosi et al. (1995), Testa et al. (1999) and Barmina et al. (2002) to constrain evolutionary models of young stars. It is interesting to note that these different studies do not always agree on the preferred amount of convective core overshoot to best reproduce the observed CMDs.  In particular there is a long lasting debate about the young LMC cluster NGC 1866, where turnoff stars have about $4 - 5 M_{\\odot}$: Chiosi et al. (1989a,b) and Brocato \\& Castellani (1988), Lattanzio et al. (1991) and Brocato et al. (1994), Testa et al. (1999) and Barmina et al. (2002), are pairs of papers presenting systematically opposite conclusions (for or against overshooting).  To briefly summarize only the most recent ones, Testa et al. (1999) obtained deep photometry ($V<24$) of a wide region of this cluster, and concluded that the best fit to both the integrated luminosity function and the magnitude of the horizontal branch is achieved by classical models, without overshooting, provided that a 30$\\%$ fraction of binaries is allowed. However, a new analysis of the same data with a more accurate completeness correction and normalization, by Barmina et al. (2002), showed that models with overshoot give a better fit to both the overall morphology of the CMD and to the integrated luminosity function of main sequence stars, reproducing the correct ratio of main sequence to post-main sequence stars.  Thanks to the brightness of the stars in the young MC clusters, the above mentioned observations were feasible using medium size telescopes. Instead, our aim is to explore the less studied regime of intermediate-age, low metallicity clusters, which turnoff is by definition fainter, and therefore only 8-meter size telescopes under excellent seeing conditions, or the HST, can give the necessary high quality photometry.  Our intention is also to independently analyze the data with two of the most recent sets of stellar evolutionary models widely in use, namely, those of Padova (Girardi et al. 2000) and Yonsei-Yale (Yi et al. 2001, $Y^2$ thereafter).  In this paper, we discuss the project setup (Sec. 2), and we present the data obtained with VLT: the observations, photometry and crowding tests are presented in Sec. 3, while the procedure used to statistically substract the LMC field stars from the cluster CMDs is discussed in Sec. 4. In Sec. 5, we briefly discuss the main differences between the stellar evolution models of Girardi et al. (2000) and Yi et al. (2001). In Sec. 6, a preliminary comparison of the cluster CMDs with these stellar evolutionary models is performed. In two forthcoming papers that follow in this same issue, these data are independently analyzed in detail with both the $Y^2$ (Woo et al. 2002) and Padova (Bertelli et al. 2002) models, in such a way that some feedback on the stellar evolution models used can be provided. ",
        "conclusions": "\\label{summ}  This preliminary investigation on three intermediate age LMC clusters, namely NGC 2173, SL 556 and NGC 2155 is the starting point for a very detailed analysis aiming at a test on input physics for stellar models.  In principle there is the opportunity to check the efficiency of convective core overshoot for star masses between 1.1 and 1.5 $M_{\\odot}$ by comparison of these CMDs with synthetic CMDs and isochrones.  The isochrone fitting to the CMDs of the clusters gives some information on their age and metallicity, but we must take into account that there are also uncertainties in the LMC distance modulus and interstellar reddening.  Only a more refined analysis, considering also the star distribution in several regions of the CMD, will give reliable results.  In the following two papers of this series, the synthetic CMD technique will be used independently by members of the Padova and Yale groups to determine the cluster characteristics, taking into account the uncertainties in the observations and in stellar evolutionary models, and to give some feed-back into the stellar evolutionary models themselves."
    },
    "0208/astro-ph0208141_arXiv.txt": {
        "abstract": "We fit elliptical isophotes  to the Hubble Deep Field-North  WFPC-2 and NICMOS data to study the rest-frame $(UV_{218}-U_{300})_o$ color profiles and rest-frame B surface brightness profiles of 33 intermediate redshift galaxies ($0.5 \\leq z \\leq 1.2$) with $I_{814}$ $<$ 25 and 50 high redshift galaxies ($2.0 \\leq z \\leq 3.5$) with $H_{160}$ $<$ 27. From the weighted least-squares fit to the color profiles we find that, at intermediate redshifts, the galaxies possess negative color gradients ($\\langle$ $\\Delta(UV_{218}-U_{300})_o$/$\\Delta$log(r) $\\rangle $= $-0.091$ $\\pm$ 0.007 mag dex$^{-1}$) indicating a reddening towards the center of the profile similar to local samples whereas, at high redshifts, the galaxies possess positive color gradients ($\\langle$ $\\Delta(UV_{218}-U_{300})_o$/$\\Delta$log(r) $\\rangle $= 0.272 $\\pm$ 0.007 mag dex$^{-1}$) indicating that star formation is more centrally concentrated.  Although the presence of dust can cause some reddening to occur towards the centers of the profiles seen at intermediate redshifts, it can not explain the strong central blueing of light seen at high redshifts. Thus, we are witnessing a population of galaxies with strong positive color gradients at high redshifts which do not seem to exist in large numbers at lower redshifts.  This indicates that star formation is more centrally concentrated in the distant galaxy sample which differs from the prevalent mode of extended disk star formation that we observe in the local universe. Additionally, we find that it is critical to correct for PSF effects when evaluating the surface brightness profiles since at small scale lengths and faint magnitudes, an $r^{1/4}$ profile can be smoothed out substantially to become consistent with an exponential profile. After correcting for PSF effects, we find that at higher look-back time, the fraction of galaxies possessing exponential profiles have slightly decreased while the fraction of galaxies possessing $r^{1/4}$ profiles have slightly increased. Our results also suggest a statistically insignificant increase in the fraction of peculiar/irregular type galaxies. We compare our results with recent semi-analytical models which treat galaxy formation and evolution following the cold dark matter hierarchical framework. ",
        "introduction": "Two of the most fundamental and intriguing questions in astronomy are how galaxies form and how they evolve with time.  In order to answer these questions, we must be able to compare and contrast the properties of galaxies at different redshifts. Whereas we have a wealth of information about galaxies at z $\\leq$ 1.0, our knowledge of the properties of galaxies at higher redshifts is limited. However, with the advent of the Hubble Deep Field-North (HDF-N) project \\citep{wil96,thom98,dic99,dic00}, we can probe to fainter surface brightness limits and smaller angular scales than before, the sizes, shapes, and colors of distant galaxies which will ultimately yield important clues to understanding their structure, formation, and subsequent evolution.  The near-infrared data from NICMOS combined with the optical data from WFPC-2 give us the unique opportunity to compare the surface brightness properties of galaxies at the same rest-frame wavelengths over a range of redshifts allowing us to better understand galaxy formation and evolution from an observational standpoint.  One observational test of galaxy evolution is the study of color profiles which will give us an idea of the distribution of stellar populations in the galaxies and whether this distribution changes with time. Due to limits in resolution, the study of the color profiles of galaxies have been predominantly restricted to the low and intermediate redshift regimes. However, the deep, high resolution, multi-wavelength data from the HDF-N probed in this study provides us with the rare opportunity to study the color profiles of galaxies at higher redshifts than those studied in the past. Previous studies of early-type galaxies at z $\\leq$ 1.0 have shown that they tend to have redder colors in their central regions and gradually become bluer outwards \\citep{tamu00,tam00,vad88,fran89,pel90}. This trend in the color profiles can be explained by either a stellar age or metallicity gradient which become degenerate at $z = 0$ \\citep{silva94}. Most studies concur that models involving the metallicity gradient best reproduce the color gradients seen in early-type galaxies at z $\\leq$ 1.0. On the other hand, the gradients observed in late-type galaxies may be due to both age and metallicity effects, i.e., the central parts of the galaxies in general have older stars and higher metallicity than the outer parts making them have redder colors towards the centers. In their study of the near-IR and optical color profiles of 86 face-on disk dominated galaxies, \\citet{deJ96} concluded that their color gradients were best reproduced by models involving both stellar age and metallicity gradients. The existence or lack of dust in galaxies can also complicate matters and must be addressed when interpreting color profiles since, in theory, dust may also be responsible for the central reddening in galaxies if we assume that dust generally tends to be more concentrated in the center and consequently would produce more extinction there \\citep{deJ96,evans94,buy94}.  A study of the color profiles and surface brightness profiles of galaxies spanning a wide range of redshifts will help us place constraints on galaxy formation and evolution by enabling us to compare what we learn from observations with what we predict from theoretical models. Currently, the hierarchical structure formation model \\citep{bau98,cole94,kauf97,rouk97,whit91} represents the popular framework for how structure formed and evolved in the universe.  This model assumes that the  universe is dominated by nonbaryonic dark matter which only interacts with visible matter through its gravitational influence and ultimately determines where galaxies will form.  It predicts that the gravitational perturbations in the early universe will cause the smallest mass fluctuations to collapse first and then to subsequently merge into progressively larger structures until they form the mature galaxies we observe today. \\citet{bau98} (hereafter BCFL) analyzed the properties of the high redshift Lyman break galaxies in the context of this model.  In their semi-analytic treatment of galaxy formation in hierarchical clustering theories, they generated mock catalogs of the high redshift Lyman break galaxies (LBGs) using the color criterion imposed by \\citet{steid93} and modeled the growth of dark matter haloes by the accretion of matter through mergers taking into account the cooling of gas into stars and feedback.  \\citet{SPF00} (hereafter SPF) also applied semi-analytic models of galaxy formation within the hierarchical clustering framework to analyze the population of Lyman break galaxies at high redshift. Since they conveniently address the properties of Lyman break galaxies (the sample of significant interest to us) within the current popular framework for galaxy formation and evolution, we will interpret our results within the context of the BCFL and SPF models. ",
        "conclusions": "From our surface brightness and color analysis of galaxies at a range of redshifts probed at the same rest-frame wavelengths in the Hubble Deep Field-North, we conclude:  1). The color profiles of galaxies reveal that at earlier epochs, there are fewer galaxies with old red stellar populations in their centers and most galaxies have centrally concentrated star formation.  As in local samples of galaxies, our intermediate redshift galaxies (0.5 $\\leq$ z $\\leq$ 1.2) have redder central regions due to a combination of age, metallicity and dust gradients.  Since age gradients are the only viable explanation for the central blueing of our high redshift sample, we conclude that the majority of high redshift galaxies contain centrally concentrated starbursts.  2). From the galaxy models, we have demonstrated the importance of taking into account the PSF effects before attempting to classify morphologies. After correcting for the PSF effects in the surface brightness profiles, we find that the population of galaxies possessing $r^{1/4}$ profiles have slightly increased with look-back time while those with exponential profiles have slightly decreased.  Our results also suggest a statistically insignificant increase in the fraction of galaxies with peculiar/irregular structure at higher redshifts. The results should be viewed with a note of caution, however, due to the small sample size and large uncertainties in the galaxy models especially at small scale lengths and faint magnitudes.  3). If, as predicted by hierarchical galaxy formation models,  mergers and interactions played an important role in the lifetime of these galaxies, such processes would be responsible for the positive color gradients indicative of centrally concentrated star formation seen in the high redshift galaxies. SPF show that their \"collisional starburst\" model, in which bursts are triggered by mergers, best reproduces the observed properties of Lyman break galaxies and of the Universe in general. Such centrally condensed nuclear star bursts are consistent with our observed trends in the color gradients. A good way to test whether the \"collisional starburst\" model holds is by applying the color-asymmetry diagram developed by \\citet{con00} to distinguish between starbursts driven by mergers and interactions from starbursts which were ignited by some other method.  Since the Hubble Deep Field  represents only a small volume of space, we must be cautious in generalizing our results to the universe as a whole.  We need to obtain more observations of high redshift galaxies to measure critical quantities such as their virial masses and estimation of dust content. We also need to better understand how and to what degree observational selection comes into play when we compare galaxies at different redshifts.  Given that we are now detecting fainter galaxies at higher redshifts and that instruments with multi-object spectroscopic capabilities in the near-IR and 8-m class telescopes are now readily available, much information will undoubtedly be unraveled in the near future concerning the nature of LBGs which will allow us to formulate a more robust theory on the formation and evolution of galaxies."
    },
    "0208/astro-ph0208507.txt": {
        "abstract": "I discuss recent advances being made in the physics and astrophysics of cosmic rays and cosmic \\grays\\ at the highest observed energies as well as the related physics and astrophysics of very high energy cosmic neutrinos. I also discuss the connections between these topics. ",
        "introduction": "In these lectures, I will discuss physics and astrophysics at the highest energies using current astrophysical observations. By taking a synoptic view of ultrahigh energy hadrons, photons and neutrinos, one can gain insights into the profound connections between different fields of observational astronomy and astrophysics which use different experimental techniques. Observations have been made of cosmic \\grays\\ up to 50 TeV energy and of ultrahigh energy cosmic rays up to 300 EeV ($3 \\times 10^8$ TeV). As of this date, no very high or ultrahigh energy cosmic neutrinos have been detected, however, the {\\it AMANDA} (Antarctic Muon and Neutrino Detector Array) experiment, now in operation, is searching for neutrinos above 1 TeV energy (Wischnewski 2002).%\\cite{wi02}  The subjects of these lectures concern some of the deepest questions in frontiers of cutting-edge astrophysics. They also involve physics at the highest energy frontiers. The new physics which has been and may be invoked to explain the highest energy observations comprise such questions as the violation of Lorentz invariance (special relativity), grand unification of the electroweak and strong interactions, quantum gravity theory and the question of whether we live in a universe containing new large extra dimensions. ",
        "conclusions": ""
    },
    "0208/astro-ph0208231_arXiv.txt": {
        "abstract": "{Using a sample of 57 VLT FORS spectra in the redshift range $1.37<z<3.40$ (selected mainly from the FORS Deep Field survey) and a comparison sample with 36 IUE spectra of local ($z \\approx 0$) starburst galaxies we derive \\CIV\\ and \\SiIV\\ equivalent width values and estimate metallicities of starburst galaxies as a function of redshift. Assuming that a calibration of the \\CIV\\ equivalent widths in terms of the metallicity based on the local sample of starburst galaxies is applicable to high-$z$ objects, we find a significant increase of the average metallicities from about $ 0.16\\,Z_{\\odot}$ at the cosmic epoch corresponding to $z \\approx 3.2$ to about $ 0.42\\,Z_{\\odot}$ at $z\\approx 2.3$. A  significant further increase in metallicity during later epochs cannot be detected in our data. Compared to the local starburst galaxies our high-redshift objects tend to be overluminous for a fixed metallicity. Our observational results are in good agreement with published observational data by other authors and with theoretical predictions of the cosmic chemical evolution. ",
        "introduction": "\\label{Introduction}  \\begin{figure*} \\includegraphics{ms2802f1.eps} \\caption{ Examples of the low-resolution galaxy spectra obtained with FORS. The redshift of the objects is increasing from panel {\\bf a)} to {\\bf c)}. The dotted line indicates the noise level which  (due to the night sky spectrum and the wavelength dependent instrumental efficiency) varies with wavelength. The S/N of these spectra is about $\\le$\\,16, 43 and 12 for object 7029, 5903 and 6063, respectively. The object designations correspond to the catalog of Heidt et al. (2002). For orientation the expected positions of selected spectral lines are indicated by vertical dashed lines. The two apparent emission features (marked by arrows) longward of the \\CIV\\ absorption feature are artifacts and resulted from the increased noise at this wavelengths due to telluric OH band edges and atmospheric absorption. For the same reason Fig. 1c becomes unreliable beyond 7000 \\AA . \\label{fig:exa_spectra} } \\end{figure*} Among the unsolved questions of cosmology is the beginning and evolution of the star formation process at early cosmic epochs. From the presence of heavy nuclei in high-z quasars and galaxies it is clear that star formation started rather early (see e.g. Hamann and Ferland 1999, Dietrich et al. 1999). Moreover, galaxy counts and emission line studies of high-z galaxies indicate that the star formation rate (SFR) declined rapidly since at least the epoch corresponding to redshifts of $z \\approx 2$ (see e.g. Madau et al. 1996, Madau 2000). On the other hand intrinsic interstellar extinction (which may affect strongly  the rest-frame UV of distant galaxies) and uncertainties concerning the relation between line emission and the star formation introduce considerable uncertainties into the derivation of the global star formation rate at early cosmic epochs from visual galaxy counts and emission line studies, while investigations using IR and mm-wave data are hampered by small samples.  An alternative approach to investigate the early star formation history of the universe is the evaluation of the chemical enrichment history of the universe, as star formation and the rapid evolution of massive stars results in a production of heavy nuclei more or less proportional to the SFR. Steidel et al. (1996a, 1996b) and  Lowenthal et al. (1997) have demonstrated that galaxies with redshifts up to about $z \\approx 5$ can be observed with optical photometry and spectroscopy during periods of high star formation activity (resulting in high rest-frame UV emission which is redshifted into the optical wavelength range in the observers frame). Basic properties of these objects, such as number densities, luminosities, colors, sizes, morphologies, star formation rates, overall chemical abundances, dynamics and clustering have been investigated in various recent papers (cf. e.g. Steidel et al. 1996b, Yee et al. 1996, Lowenthal et al. 1997, Pettini et al. 2000, and Leitherer et al. 2001). Steidel et al. (1996a, 1996b) already noted that the high-z galaxies show, on average, relatively weak metallic absorption lines, and they ascribe this finding tentatively to a lower metal content (a suggestion which has later been reiterated in several subsequent papers). On the other hand most of the published spectra of high-z galaxies are not of sufficient S/N to provide qualitative information on the metal content. Therefore, we obtained new high S/N spectra of galaxies with $z \\le 3.5$ with the aim of  studying the chemical evolution of starburst galaxies at high redshifts.  In the present paper we describe results on the \\CIV\\ absorption line strength and their interpretation in terms of chemical evolution with cosmic age at redshifts $0<z<3.5$. Most of the new spectra were obtained with the FORS instruments at the ESO VLT in the course of a photometric and spectroscopic study of distant galaxies in the FORS Deep Field (FDF) (Appenzeller et al. 2000, Heidt et al. 2001, Bender et al. 2001). Due to its combination of depth and a (compared to the HDFs) relatively large area the FDF is particularly well suited for statistical studies of high-redshift galaxies. In the present investigation we restrict ourselves to redshifts $z \\leq 3.5$ since at higher redshifts the position of the redshifted \\CIV\\ resonance lines tends to coincide with strong OH night sky lines. Hence, for an accurate sky subtraction a higher spectral resolution or longer exposure times than we could achieve so far would be needed. In Sect.~2 we describe the sample selection and the observations. In Sect.~3 we present and discuss the measurement of the \\CIV\\ (and \\SiIV ) equivalent widths, from which we estimate the metallicities of the investigated objects in Sect.~4. In Sect.~5 we compare our results with data available in the literature, in Sect.~6 we draw our conclusions. ",
        "conclusions": "\\label{Conclusions}  Our study shows that the FDF starburst galaxies at $z \\approx 3.2$ have on average significantly lower \\CIV\\ equivalent widths than starburst galaxies at lower redshifts. In view of the known close relation of the \\CIV\\ strength to the metallicity in local starburst galaxies, it appears likely that this effect is due to a significant evolution of the average metallicity in such objects at high redshifts. Using data from well studied local starburst galaxies we calibrated the \\CIV\\ strength in terms of the heavy element content of these objects. Assuming that this calibration is applicable to high redshift starburst galaxies we find that the mean cosmic metallicity as observed in starburst galaxies has increased significantly between the cosmic epochs corresponding to $z \\approx 3.2$ and $\\approx 2.3$ (when the universe was between 2 and 3 Gyrs old). If this interpretation of the increase of the \\CIV\\ absorption strength  with decreasing redshift is correct, an intense phase of star formation and evolution of massive stars must have occurred during this period. At lower redshifts ($z <2.5$) our data indicate little further increase of the average metallicity of starburst galaxies. Hence, the further cosmic chemical enrichment seems to have  been insignificant during the last 11 Gyrs. The metallicity evolution indicated by our data is in reasonable agreement with published theoretical chemical enrichment models due to star formation at early cosmological epochs. Our results are also in agreement with a metallicity evolution found in high-z radio galaxies (de Breuck et al. 2000) and the tentative evidence for a gradual chemical enrichment of the gas producing the damped Ly$\\alpha $ lines (Pettini et al. 1997, Savaglio et al. 2000).  Since the FDF observations provide a fairly complete sample of the bright starburst galaxies in the observed direction and redshift ranges, the observed chemical evolution should be characteristic of the cosmic volumes with the most intense star formation at the corresponding epochs. However, it is also clear, that our results do not apply to all objects and volumes at a certain redshift or epoch. It is, e.g., well known that the BLR gas of high-redshift QSOs is characterized by high metallicities and that no significant chemical evolution is observable in these objects up to at least $z=5$ (see e.g. Hamann and Ferland, 1999, Dietrich et al. 1999, Dietrich and Wilhelm-Erkens, 2000). Obviously, in the environment of these early QSOs much star formation and stellar evolution must have taken place at epochs corresponding to even higher redshifts. It is also known, however, (and confirmed from the FDF survey) that high-z starburst galaxies are much more frequent than bright high-z QSOs. Hence, the starburst galaxies are expected to be a more representative tracer of the history of the overall cosmic chemical evolution than the QSOs.  From our data we cannot determine whether at high redshift a metallicity-luminosity relation does exist, since we do not have any faint objects in our high-$z$ sample. But it is evident that the high-redshift galaxies are on average overluminous for their metallicities compared with local starburst galaxies. This trend is also found by Pettini et al. (2001) and Kobulnicky \\& Koo (2000) for Lyman break galaxies. From tests on various subsamples we find that the observed dependence of $W_{0}(\\CIV )$ on redshift is not caused by a luminosity effect. Furthermore we showed that this dependence is also not cause by the possible selection effect of preferentially observing galaxies with low dust content and hence low metallicity at high redshifts.  Differences in the \\SiIV\\ to \\CIV\\ ratios between local and high-$z$ galaxies in our sample suggest differences in the population and star formation history in the galaxies with $z>3$. Short bursts of star formation may have been more important (relative to periods of ``continuous star formation'') at these early epochs."
    },
    "0208/astro-ph0208411_arXiv.txt": {
        "abstract": "At energies higher than the brane tension, the dynamics of a scalar field rolling down a potential are modified relative to the predictions of General Relativity. The modifications imply, among other things, that steeper potentials can be used to drive an epoch of slow--roll inflation. We investigate the evolution of entropy and adiabatic modes during inflation driven by two scalar fields confined on the brane. We show that the amount of entropy perturbations produced during inflation is suppressed compared to the predictions made by General Relativity. As a consequence, the initial conditions do not matter in multiple field inflation in brane worlds if inflation is driven at energies much higher than the brane tension. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0208/astro-ph0208483_arXiv.txt": {
        "abstract": "The burning speed of a thermonuclear supernova front could be described by the fractal model of combustion. We have  examined the effects of magnetic fields on the fractalization of the front considering a white dwarf with a nearly dipolar magnetic field and found an intrinsic asymmetry on the velocity field of the expanding plasma. For white dwarf's magnetic fields of  $10^8-10^9$ G at the surface, and assuming a field roughly 10 times greater near the center, we have found asymmetries in the velocity field $> 10-20 \\%$ at $\\rho \\sim 10^8\\,{\\rm g \\,cm}^{-3}$ , produced between the magnetic polar and the equatorial axis of the remnant. This effect may be related to the asphericities inferred from spectro-polarimetric observations of very young SN Ia remnants (for example: the SN 1999by). In the present work, we analyse the dependence of the asymmetry with the composition of the white dwarf progenitor. ",
        "introduction": "The explosion of a type Ia supernova begins with the combustion at the center of a Chandrasekhar mass white dwarf of carbon-oxygen (C+O)  or oxygen-neon-magnesium (O+Ne+Mg) fuels. The heat is transported mainly by conduction due to degenerate and completely relativistic electrons as a subsonic deflagration wave propagatating outwardly of the star. The deflagration front born laminar is subject to several hydrodynamic instabilities such as Landau-Darrieus (LD) and Rayleigh Taylor (RT) instability (Arnett \\& Livne 1994, Khokhlov 1993) that produce an increment of the area at which the nuclear reactions take place. This causes an increase of the nuclear energy generation rate and consequently an acceleration of the front.  The combustion front is stabilized by the merging of cells, the formation of cusps, and the expansion of the exploding star. This leads to the formation of a cellular structure at microscopic scales. The bubbles that grow due to RT instability are also subject to Kelvin-Helmoltz (KH) or shear instability when nonlinear stabilization fails. The onset of the KH instability marks the transition to the fully developed turbulence regime at the lower scales. During this, fluid motions are characterized by the formation of a turbulent cascade in the inertial scales where viscous dissipation is not important. This turbulence can be described by the Kolmogorov's scaling law.  The fractal model for the combustion (e.g., Timmes \\& Woosley 1992, Niemeyer \\& Woosley 1997) has achieved some success on describing the acceleration of wrinkled flames both in experiments and also in numerical simulations. The velocity of the flame in this case is given by:  \\begin{equation} v_{frac} \\, = v_{lam} (L/l_{min})^{D-2} \\end{equation}  \\noindent Where $v_{frac}$ is the effective fractal velocity and $v_{lam}$  is the laminar velocity of the flame; $L$ and $l_{min}$ are the greatest and minimum scales, respectively, of preturbations which are R-T unstable; and D is the fractal dimension of the front. The derivation of the value of D for a turbulent combustion regime has given $D =  2.25 -2.5$ (see Ghezzi, de Gouveia Dal Pino \\& Horvath 2001), which is in agreement with previous numerical results  (Blinnikov, Sasorov \\& Woosley 1995). ",
        "conclusions": "An asymmetry in the velocity field is developed by the presence of a dipolar magnetic field during the fractal growth of the deflagration front of a type Ia supernova  that can lead to the formation of a prolate remnant. The magnetic field introduces an effective surface tension in the equator of the white dwarf progenitor that reduces the velocity of the combustion front at the equator, $v_{eq}$, with respect to the velocity at the poles, $v_{pol}$, so that $v_{pol} > v_{eq}$. The asymmetry is larger for heavier progenitors\\footnote{In this work we calculated the \"instantaneous asymmetry values\" at a given density, as we will show (see Ghezzi et al. 2002, and Ghezzi 2002) integrating the effect over the explosion leads to higher asphericity of the remnant.}. In particular, for progenitors with a composition $X(^{12}{\\rm C})=0.2\\,\\,X(^{16}{\\rm O})=0.8\\,$, $\\Delta \\rho /\\rho=0.415$, and surface magnetic fields $\\sim 10^{8}$ G, a $10\\,$ to $20\\,\\%$ asymmetry has been found at a middle distance from the center of the star (see Fig. 2).  As only a small fraction of the observed white dwarfs are inferred to have magnetic fields higher than about $10^{8}$ G, asymmetries are not expected to occur very frequently. Nonetheless, recent spectropolarimetric observations have revealed a linear polarization component in the radiation of very young SN Ia remnants, which suggests that prolate atmospheres with asymmetries $> 17\\,\\%$ are producing it (see Leonard, Filippenko, \\& Matheson 1999, Wang , Wheeler \\& H\\\"oflich 1997 and Howell, Hoeflich, Wang \\& Wheeler 2001). The model presented here offers a plausible explanation for such observations."
    },
    "0208/astro-ph0208210_arXiv.txt": {
        "abstract": "We outline a method by which the angular radii of giant and main sequence stars located in the Galactic Bulge can be measured to a few percent accuracy.  The method combines comprehensive ground-based photometry of caustic-crossing bulge microlensing events, with a handful of precise ($\\sim 10 \\muas$) astrometric measurements of the lensed star during the event, to measure the angular radius of the source, $\\theta_*$.  Dense photometric coverage of one caustic crossing yields the crossing time scale $\\dt$.  Less frequent coverage of the entire event yields the Einstein timescale $\\te$ and the angle $\\phi$ of source trajectory with respect to the caustic.  The photometric light curve solution predicts the motion of the source centroid up to an orientation on the sky and overall scale.  A few precise astrometric measurements therefore yield $\\thetae$, the angular Einstein ring radius.  Then the angular radius of the source is obtained by $\\theta_*=\\thetae(\\dt/\\te) \\sin{\\phi}$.  We argue that the parameters $\\te, \\dt, \\phi$, and $\\thetae$, and therefore $\\theta_*$, should all be measurable to a few percent accuracy for Galactic bulge giant stars using ground-based photometry from a network of small (1m-class) telescopes, combined with astrometric observations with a precision of $\\sim 10\\muas$ to measure $\\thetae$. We find that a factor of $\\sim 50$ times fewer photons are required to measure $\\thetae$ to a given precision for binary-lens events than single-lens events.  Adopting parameters appropriate to the {\\it Space Interferometry Mission} (SIM), we find that $\\sim 7$ minutes of SIM time is required to measure $\\thetae$ to $\\sim 5\\%$ accuracy for giant sources in the bulge. For main-sequence sources, $\\thetae$ can be measured to $\\sim 15\\%$ accuracy in $\\sim 1.4$ hours.  Thus, with access to a network of 1m-class telescopes, combined with 10 hours of SIM time, it should be possible to measure $\\theta_*$ to $5\\%$ for $\\sim$80 giant stars, or to $15\\%$ for $\\sim$7 main sequence stars. We also discuss methods by which the distances and spectral types of the source stars can be measured.  A byproduct of such a campaign is a significant sample of precise binary-lens mass measurements. ",
        "introduction": "Although of fundamental importance to stellar astrophysics, precise measurements of angular radii are generically difficult to acquire routinely and in a model-independent way.  Classical direct methods of measuring stellar radii include lunar occultations, interferometry, and eclipsing binaries.  Lunar occultation measurements yield precise angular radii (see Richichi et al.\\ 1999 and references therein), but the number of stars to which this technique can be applied is limited. The number of direct measurements using interferometers has recently increased dramatically with advent of, e.g.\\ the Palomar Testbed Interferometer (van Belle et al.\\ 1999a, Colavita et al.\\ 1999), and the Navy Prototype Optical Interferometer (Armstrong et al.\\ 1998, Nordgren et al.\\ 1999), and is likely to continue to increase as more technologically advanced interferometers come online. Unfortunately, both lunar occultation and interferometric angular diameter measurements have traditionally been primarily limited to nearby, evolved stars.  Angular radii of main-sequence stars can be determined using detached eclipsing binaries (i.e.\\ Popper 1998), however the large amount of data (both photometric and spectroscopic) required to yield accurate radii determinations makes this method prohibitive.  Thus, of the $\\sim300$ direct, precise angular diameter measurements compiled by van Belle (1999), the overwhelming majority, $\\sim 85\\%$, are of evolved stars. Finally, it will be difficult to acquire a large sample of angular radii determinations of stars with metallicity considerably smaller than solar using these methods, due to the paucity of metal-poor stars in the local neighborhood.  Here we present a method, based on a suggestion by Paczy\\'nski (1998), of measuring angular radii of stars that overcomes some of the difficulties inherent in the classical methods.  This method employs the extraordinary angular resolution provided by caustics in gravitational microlensing events, and as such is yet another in the growing list of applications of microlensing to the study of stellar astrophysics (see Gould 2001 for a review).  The original suggestion of Paczy\\'nski (1998) was to invert the method of Gould (1994) for measuring the relative source-lens proper motion $\\murel$ in microlensing events.  If the lens transits the source in a microlensing event, precise photometry can be used to determine the time it takes for the lens to transit one source radius, $t_*=\\theta_*/\\murel$, where $\\theta_*$ is angular radius of the source.  An estimate of $\\theta_*$, using an empirical color-surface brightness relation, together with a measurement of the flux of the source, can then be used to estimate $\\murel$, which Gould (1994) argued could be used to constrain the location of the lens.  However, as Paczy\\'nski (1998) pointed out, it is possible to independently measure the angular Einstein ring radius of the lens, \\begin{equation} \\thetae=\\sqrt{{{4 G M}\\over \\drel c^2} }, \\label{eqn:thetae} \\end{equation} by making precise astrometric measurements of the centroid shift of the source during the microlensing event using, i.e., the {\\it Space Interferometry Mission} (SIM).\\footnote{http://sim.jpl.nasa.gov} Here $M$ is the mass of the lens, $\\drel$ is defined by, $\\drel\\equiv \\dos\\dol/\\dls$, and $\\dos$, $\\dol$, and $\\dls$ are the distances between the observer-source, observer-lens, and lens-source, respectively.  Since $\\murel = \\thetae/\\te$, by combining the measurement of $\\thetae$ with the Einstein timescale $\\te$ of the event determined from the light curve, it is possible to measure $\\theta_*$ for the source stars of microlensing events. We show that, with reasonable expenditure of resources, it should be possible to measure angular radii of a significant sample ($\\sim 80$) of giant stars in the bulge to an accuracy of $\\la 5\\%$, or $\\sim 7$ main-sequence stars to an accuracy of $\\la 15\\%$.   Limb-darkening determinations should also be possible for the majority of the sources, and most will be relatively metal poor as compared to those for which angular radii determinations are currently available.  Although measurements of $\\theta_*$ can be made using single-lens events, in \\S\\ref{sec:bvs} we argue that this method is better suited to caustic-crossing binary-lens events, which are more common, easier to plan for, and considerably less resource-intensive than source-crossing single-lens events. We describe in some detail the basic method of measuring $\\theta_*$ for the source stars of caustic-crossing binary-lens events in \\S\\ref{sec:method}, including a discussion of the expected errors on the individual parameters that enter into the measurement.  We discuss various subtleties, complications, and extensions to the method in \\S\\ref{sec:discussion}, and also present an estimate of the number of $\\theta_*$ measurements that might be made in this way.  Finally, we summarize and conclude in \\S\\ref{sec:conclusion}. ",
        "conclusions": "\\label{sec:conclusion}  We have outlined a method to measure the angular radii $\\theta_*$ of giant and main sequence source stars of fold caustic-crossing binary microlensing events toward the Galactic bulge.  Our method to measure $\\theta_*$ consists of four steps.  First, survey-quality data can be used to discover and alert caustic-crossing binary-lensing events. Such data is sufficient to characterize the event timescale $\\te$ and the angle $\\phi$ of source trajectory with respect to the caustic. Dense sampling of one of the caustic crossings yields the caustic-crossing timescale $\\Delta t$.  The global solution to the binary-lens light curve yields a prediction for the trajectory of the centroid of the source up to an unknown angle $\\alpha$, and the scale, $\\thetae$.  Thus a few, precise astrometric measurements during the course of the event yield $\\thetae$.  The angular source radius is then simply given by $\\theta_*=\\thetae(\\dt/\\te) \\sin{\\phi}$.  We argued, based on past experience with modeling binary-lens events, that the parameters $\\Delta t$, $\\phi$, and $\\te$ should be measurable to a few percent accuracy, provided one caustic-crossing is densely and accurately sampled, and the entire event is reasonably well-covered.  We then performed a series of Monte Carlo experiments that demonstrated that astrometric measurements during the course of the binary-lens event should allow for the determination of $\\thetae$ to $\\sim2\\%$ accuracy, assuming photon-limited statistics and a total of 60,000 photons per event.  This is a factor of $\\sim 50$ fewer photons than are required to measure $\\thetae$ to the same precision in single-lens events and corresponds to an exposure time of $T=1.6$~hour with SIM on an $I=18$ source. Therefore, it should be possible to measure $\\theta_*$ for a significant sample of giant and main-sequence stars in the bulge with reasonable expenditure of resources.   \\bigskip We would like to thank Neal Dalal for helpful conversations.  We would also like to thank the anonymous referee for useful comments and suggestions. This work was supported by NASA through a Hubble Fellowship grant from the Space Telescope Science Institute, which is operated by the Association of Universities for Research in Astronomy, Inc., under NASA contract NAS5-26555, by JPL contract 1226901, and by the Science Research Center (SRC) of the Korean Science and Engineering Foundation (KOSEF)."
    },
    "0208/astro-ph0208026_arXiv.txt": {
        "abstract": "We investigate the environments of quasars such as number distribution of galaxies using  a semi-analytic model which includes both galaxy and quasar formations based on the hierarchical  clustering scenario. We assume that a supermassive black hole is fueled by accretion of cold gas and that it is a source of quasar activity during a  major merger of the quasar host galaxy with another galaxy. This major merger causes spheroid formation of the host galaxy. Our model can reproduce not only general form of the galaxy luminosity functions in the local Universe but also the observed relation between a supermassive black hole mass and a spheroid  luminosity, the present black hole mass function and the quasar luminosity functions at different  redshifts. Using this model, we predict the mean number of quasars per halo, bias parameter of quasars  and the probability distribution of the number of galaxies around quasars. In our model, analysis of the mean number of quasars per  halo shows that the spatial distribution of galaxies is different from that of quasars. Furthermore, we found from calculation of the probability distribution of galaxy numbers that at $0.2 \\lesssim z \\lesssim 0.5$, most quasars are likely to reside in galaxy groups. On the other hand, at $1 \\lesssim z \\lesssim 2$ most quasars seem to reside in more varied  environments than at a lower redshift; quasars reside in environments ranging from small groups of galaxies to clusters of galaxies. Comparing these predictions with observations in future will enable us to constrain our quasar formation model. ",
        "introduction": "The environments of quasars provide important clues to the physical processes of their formation and also yield important information about the relations between the distribution of quasars and the large-scale structure of the universe. For more than three decades, we have known that quasars are associated with enhancements in the spatial distributions of galaxies (\\cite{BSG69}). Studies of the environments of quasars in the nearby universe ($z \\lesssim 0.4$) have shown that quasars reside in environments ranging from small to moderate groups of galaxies rather than in rich clusters (e.g. \\cite{BC91}; \\cite{FBK96}; \\cite{MD01}). In order to interpret the observational results of the environments of quasars at  low redshifts and predict the environments of quasars at high redshifts, a physical model of quasar formation based on cosmological context is required.  It has become widely accepted that quasars are fueled by accretion of gas onto  supermassive black holes (SMBHs)  in the nuclei of  host galaxies since \\citet{Lyn} proposed this idea on quasars. Recent  observations of galactic centers suggest that a lot of nearby galaxies  have central black  holes and their estimated masses correlate with the  luminosities of spheroids\\footnote{Throughout this paper, we refer to bulge or elliptical galaxy as {\\it spheroid}.} of their host galaxies (e.g. \\cite{KR95}; \\cite{Mag98}; \\cite{MF00}). The connection between SMBHs and their host spheroids suggests that the formation of SMBHs physically links the formation of the spheroids which harbor the SMBHs. Thus, this implies that the formation of quasars is closely related to the formation of galaxies, especially of spheroids. Therefore, in order to study the formation and evolution of quasars, it is necessary to construct a unified model which includes both galaxy formation and quasar formation.  Recently, some authors have tried to construct galaxy formation models on the basis of the theory of hierarchical structure formation in  cold dark matter (CDM) universe. These efforts are  referred to as  semi-analytic models (SAMs) of galaxy formation. In the CDM universe, dark matter halos cluster gravitationally and merge together in a manner that depends on the adopted power spectrum of initial density fluctuations.  In each of the merged dark halos, radiative gas cooling, star formation, and  supernova feedback occur.  The cooled dense gas and stars constitute {\\it galaxies}.  These galaxies sometimes merge together in a common dark halo and more massive galaxies form. In SAMs, the merger trees of dark matter halos are constructed using a Monte-Carlo algorithm and simple models are adopted to describe the above gas processes. Stellar population synthesis models are used to calculate the luminosities and colors of model galaxies. It is therefore straightforward to understand how galaxies form and evolve within the context of this model. SAMs successfully have reproduced a variety of observed features of local galaxies such as their luminosity functions, color distribution, and so on (e.g. \\cite{KWG93}; \\cite{CL94}, \\yearcite{CL00}; \\cite{SP99}; \\cite{NTGY01}, \\yearcite{NYTG02}).  In these models, it is assumed that disk stars are  formed by cooling of gas in the halo. If two galaxies of comparable mass merge, it is assumed that starbursts occur and  form the spheroidal component in the center of the galaxy. $N$-body simulations have shown that a merger hypothesis for the origin of spheroids can explain their detailed internal structure (e.g. \\cite{Barn}; \\cite{Her92}, \\yearcite{Her93}; \\cite{HHS94}). Kauffmann and Charlot (\\yearcite{KC98}) have demonstrated that the merger scenario for the formation of elliptical galaxies is consistent with the color-magnitude relation and its redshift evolution (see also \\cite{NG01}). On the other hand, hydrodynamical simulations have shown that a merger of galaxies drives gas to fall rapidly to the center of a merged system and to fuel nuclear starburst (\\cite{NW83}; \\cite{MH94}, \\yearcite{MH96}; \\cite{BH96}). Moreover, observed images of quasar hosts show that many quasars reside in interacting systems or elliptical galaxies (\\cite{BKSS}). Therefore, it has often been thought that the major merger of galaxies would be a possible  mechanism for quasar and spheroid formation.  So far, a lot of studies on quasar evolution based on the hierarchical clustering scenario have been carried out with the assumption that the formation of quasars is linked to the first collapse of dark matter halos with galactic mass and that these models can explain the decline of quasar number density at $z \\gtrsim 3$ (e.g. \\cite{ER88}; \\cite{HR93}) and properties of luminosity functions of quasars (e.g. \\cite{HL98}; \\cite{HNR98}; \\cite{HMKYU}). However, if quasars are directly linked to spheroids of host galaxies rather than to dark matter halos, the approximation of a one-to-one relation between quasar hosts and dark matter halos would be very crude, especially at low redshift. Therefore, it is necessary to construct a model related to spheroid formation and SMBH formation directly. Kauffmann and Haehnelt (\\yearcite{KH00}) introduced a unified model of the evolution of galaxies and quasars within the framework of SAM (see also \\cite{Cat01}). They assumed that SMBHs are formed and fueled  during major galaxy mergers and  their model reproduces quantitatively the observed relation between spheroid luminosity and black hole mass  in nearby galaxies, the strong  evolution of the quasar population with redshift, and the relation between the luminosities of nearby quasars and those of their host galaxies.  In this paper, we investigate properties of quasar environments, using a SAM incorporated simple quasar evolution model. We assume that SMBHs are formed and fueled  during major galaxy mergers  and the fueling process leads quasar activity. While this assumption is similar to the model of  Kauffmann and Haehnelt (\\yearcite{KH00}), our galaxy formation model and the adopted model of fueling process are different from their model. Here we focus on optical properties of quasars and attempt to consider the number of quasars per halo, effective bias parameter of quasars and the number of galaxies around quasars as  characterizations of environments of quasars, because a) these quantities provide a direct measure of bias in their distribution with respect to galaxies and b) comparing results of the model with observations will enable us to constrain our quasar formation model.  The paper is organized as follows: in \\S \\ref{model} we briefly review our SAM for galaxy formation; in \\S \\ref{qsomodel} we introduce the quasar formation model; in \\S \\ref{env} we calculate the galaxy number distribution function around quasars; in \\S \\ref{disc} we provide a summary and discussion.  In this study, we use a low-density, spatially flat cold dark matter ($\\Lambda$CDM) universe with the present density parameter $\\Omega_0=0.3$, the cosmological constant $\\lambda_0=0.7$, the Hubble constant in units of $100 {\\rm km \\ s^{-1}\\ {Mpc^{-1}}}$ $h=0.7$ and the present rms density fluctuation in spheres of $8 h^{-1} {\\rm Mpc}$ radius $\\sigma_8=1.0$. ",
        "conclusions": "\\label{disc} We have constructed a unified semi-analytic model for galaxy and quasar formation and  have predicted the mean number of quasars per halo with mass $M$, $\\langle N_{\\rm QSO}(M) \\rangle$, the effective bias parameter of quasars $b_{{\\rm eff, QSO}} (z)$ and probability distribution of the number of galaxies around quasars, $P(N_{\\rm gal}|N_{\\rm QSO})$, as characterizations of the environments of quasars. These quantities reflect the processes of quasar formation such as the amount of cold gas available for fueling, the galaxy merger rate and the quasar life timescale. Therefore, by comparing these predictions with observations, one  will be able to constrain quasar formation models.  Our model can reproduce not only general form of the galaxy luminosity functions in the local Universe but also the observed relation of the SMBH mass to spheroid  luminosity, and the quasar luminosity functions at different  redshifts (Fig.\\ref{fig:bulge-bh} and Fig.\\ref{fig:qso-lum}). Using this model, we have shown $\\langle N_{\\rm QSO}(M) \\rangle$ and $P(N_{\\rm gal}|N_{\\rm QSO})$. The ratio of $\\langle N_{\\rm QSO}(M) \\rangle$ to $\\langle N_{\\rm gal}(M) \\rangle$ varies with halo mass in our model (Fig\\ref{fig:number-gal}). These results of our model suggest that the clustering of galaxies is not the same as the clustering of quasars and the effective bias parameter of quasars and its evolution are different from these of galaxies (Fig.\\ref{fig:bias}). Furthermore, we predict  the galaxy number distribution function around quasars, $P(N_{\\rm gal}|N_{\\rm QSO})$ (Fig\\ref{fig:gnd}). At lower redshifts ($0.2 \\lesssim z \\lesssim 0.5$), most halos which have quasars have at most several galaxies. This indicates that most quasars reside in groups of galaxies. On the other hand, at higher redshift ($1 \\lesssim z \\lesssim 2$), the number of galaxies in the halo with  quasars is from several to dozens; quasars reside in ranging from small groups of galaxies to clusters of galaxies. These results show that most quasars at higher redshift reside in more varied environments than at lower redshift. This model prediction is checkable by statistics of galaxies around quasars which will be obtained in future.  It is  still controversial whether the environments of quasars depend on their optical and radio luminosities. Some authors have claimed that radio-loud quasars were located in richer environments than radio-quiet quasars at at $z<0.6$ (e.g. \\cite{YG84}; \\cite{YG87}; \\cite{EYG91}; \\cite{HRV91}). However, other people obtained a different result. For example, \\citet{HCJ95} observed the galaxy environment of radio-loud quasars and radio-quiet quasars and concluded that there is no significant difference in the richness. Recent studies support this conclusion (e.g. \\cite{WLLS01}). The discrepancies between different studies may be caused partly by too small quasar samples and by differences in sample selection of quasars. This situation will soon improve with the availability of a new generation of very large quasar surveys such as the 2dF quasar redshift survey (\\cite{Cr01a}) and the Sloan Digital Sky Survey (\\cite{SDSS}). Although we do not deal with radio properties of quasars in this paper, our investigation of quasar environments will also provide a clue for understanding the radio character of quasar environments.  The mean number of quasars per halo, $\\langle N_{\\rm QSO}(M) \\rangle$, and probability distribution of the number of galaxies around quasars, $P(N_{\\rm gal}|N_{\\rm QSO})$, used in this study can provide some useful features of the quasar environments. Furthermore, the spatial galaxy-quasar correlation function is used in order to quantify the galaxy environments around a quasar. Therefore, for the further investigation of environments and clustering of quasars and in order to constrain the quasar formation model, it is also necessary to predict  spatial distribution of galaxies and quasars. We will show the results using  the combination of cosmological $N$-body simulation and SAM for formation of galaxy and quasar in the near future.   \\bigskip We would like to thank T. T. Takeuchi for providing us with the reanalyzed data of the quasar luminosity functions derived from the 2dF 10k catalogue. We are also grateful to K. Okoshi, H. Yahagi and S. Yoshioka for useful comments and discussions. We also thank to the anonymous referee for a thorough reading of the manuscript and for his valuable suggestions and comments, which improved our paper very much. Numerical computations in this work were partly carried out at the Astronomical Data Analysis Center of the National Astronomical Observatory, Japan. This work has been supported in part by the Grant-in-Aid for the Scientific Research Funds (13640249) of the Ministry of Education, Culture, Sports, Science and Technology of Japan.   \\appendix"
    },
    "0208/astro-ph0208576_arXiv.txt": {
        "abstract": "The physics during the inflationary stage of the universe is of quantum nature involving extremely high energy densities. Moreover, it is out of equilibrium on a fastly expanding dynamical geometry. We complement here the 1999 Chalonge Lectures on out of equilibrium fields in self-consistent inflationary dynamics [astro-ph/0006446] investigating inflation driven by the evolution of highly excited {\\bf quantum states}. These states are characterized by a non-perturbatively large number of quanta in a band of momenta and with zero or nonzero expectation value of the inflaton scalar field. They represent the situation in which initially a non-perturbatively large energy density is localized in a band of high energy quantum modes and are  coined tsunami-waves. The self-consistent evolution of this quantum state and the scale factor is studied analytically and numerically. It is shown that the time evolution of these  quantum states lead to two consecutive stages of inflation under conditions that are the quantum analogue of slow-roll. The evolution of the scale factor during the first stage has new features that are characteristic of the quantum state.  During this initial stage the quantum fluctuations in the highly excited band build up an effective homogeneous condensate with a non-perturbatively large amplitude as a consequence of the large number of quanta. The second stage of inflation is similar to the usual classical chaotic scenario but driven by this effective condensate. The excited quantum modes are already superhorizon in the first stage and do not affect the power spectrum of scalar perturbations.  Thus, this tsunami quantum state provides a field theoretical justification for chaotic scenarios driven by a classical homogeneous scalar field of large amplitude. ",
        "introduction": "A wealth of observational evidence from the temperature anisotropies in the cosmic microwave background strongly points towards inflation as the mechanism to produce the primordial density perturbations\\cite{turner,cmb}. Thus, inflationary cosmology emerges as the basic theoretical framework to explain not only the long-standing shortcomings  of standard big bang cosmology but also to provide a testable paradigm for structure formation\\cite{revius}-\\cite{infl}. The recent explosion in the quantity and quality of data on temperature anisotropies elevates inflation to the realm of an experimentally testable scenario that leads to robust predictions that withstand detailed scrutiny\\cite{turner,cmb}.  However at the level of implementation of an inflationary proposal, the situation is much less satisfactory. There are very many different models for inflation motivated by particle physics and most if not all of them invoke one or several scalar fields, the inflaton(s), whose dynamical evolution in a scalar potential leads to an inflationary epoch\\cite{revius}-\\cite{infl}. The inflaton field is a scalar field that provides an effective description for the fields in the grand unified theories. Furthermore there is the tantalizing prospect of learning some aspects of the inflationary potential (at least the part of the potential associated with the last few e-folds) through the temperature anisotropies of the cosmic microwave background\\cite{reconstruction}.  Most treatments of inflation study the evolution of the inflaton via the {\\em classical} equations of motion for the chosen scalar potential instead of making a quantum field treatment of the dynamics. That is, the effect of quantum fluctuations is neglected in the dynamics of the inflaton. Furthermore, since inflation redshifts inhomogeneities very fast, the classical evolution is studied in terms of a {\\em homogeneous classical scalar} field. The quantum field theory interpretation is that this classical, homogeneous field configuration is the expectation value of a quantum field operator in a translational invariant quantum state. While the evolution of this coherent field configuration (the expectation value or order parameter) is studied via classical equations of motion, quantum fluctuations of the scalar field around this expectation value are treated in a linear approximation for the high wavenumbers that lead to the  seeds for scalar density perturbations of the metric\\cite{revius}-\\cite{infl}.  The large amplitude modes that dominate the energy  of the universe during inflation are mimic by the classical homogeneous field. Moreover, the slow roll approximation is used in all cases.  A fairly broad catalog of inflationary models based on scalar field dynamics labels these either as `small field' or `large field'\\cite{reconstruction}. In the `small field' category  the scalar field begins its evolution with an initial value very near the origin of the scalar potential and rolls down towards larger values, an example is new inflation\\cite{revius,coles,linde}. In the `large field' category, the scalar field begins very high up in the potential hill and rolls down towards smaller values, an example is chaotic inflation\\cite{revius,coles,linde}.  It is only recently that the quantum dynamics of the scalar fields in the coupled evolution of matter and geometry has been studied self-consistently\\cite{noscos,asam}. This work is associated with the dynamics of non-equilibrium phase transitions in models that fall, broadly, in the `small field' category.  This subject is reviewed in the 1999 Chalonge Lectures on out of equilibrium fields in self-consistent inflationary dynamics. The conclusion of these works is that a  treatment of the quantum fluctuations that couple self-consistently to the dynamics of the metric provides a solid quantum field theoretical framework that justifies microscopically the picture based on classical inflation. At the same time these studies provide a deeper understanding of the quantum as well as classical aspects of inflation and inflationary perturbations. They clearly reveal the classicalization of initial quantum fluctuations\\cite{noscos,asam}, and furnish a microscopic explanation (and derivation) of the effective, homogeneous classical inflaton\\cite{asam}.  Very recently, a quantum dynamical  treatment of models whose classical counterpart are large field models is proposed in ref.\\cite{tsuinf}. The {\\em classical} description in these models begins with a homogeneous inflaton scalar with very large amplitude $\\phi \\sim M_{Pl}$\\cite{revius,coles,linde,infl}, i.e, very high up  in the scalar potential well.  The initial quantum state may be mixed (described by a density matrix in Fock space) or a pure state (described by a vector in Fock space). The initial state is characterized by the expectation value of the inflaton (order parameter) and the spectrum of excitations (initial particle distribution). The order parameter corresponds to the classical inflaton field in the classical limit whereas the initial particle distribution describes the spectrum of excitations in the initial state.  It is not obvious whether an initial particle distribution can give rise to inflation. As shownin ref.\\cite{tsuinf}, there exists classes of initial particle distributions leading to efficient inflation. In particular, initial states with zero order parameter and a band of excited modes can lead to an inflationary epoch. This extends the set of possible initial states that leads to chaotic inflation making it a more natural description of the early universe.  In the customary treatment of chaotic inflation (classical chaotic inflation \\cite{revius,infl}) all the energy is contained in the classical (space independent) field, so the field modes do not contribute neither to the energy nor to the background dynamics. This is clearly a very special choice of initial conditions. We shall call this scenario classical chaotic inflation.  We review in these lectures the dynamics that results from the evolution of a  {\\bf quantum state} which drives the dynamics of the scale factor through  the expectation value of the energy momentum tensor\\cite{tsuinf}. The main idea behind this approach is akin to the experimental situation in ultrarelativistic heavy ion collisions, wherein an initial highly excited state (heavy ions with very large energy) lead to the formation of a plasma that expands and cools\\cite{harris}. Recently  this situation has been modeled  by considering the evolution of highly excited quantum states  coined tsunami-waves in\\cite{tsurob,tsu1,tsu2}.  {\\em The goals of the Tsunami Inflation:}  The ideas and concepts in refs.\\cite{tsurob,tsu1,tsu2} were adapted in ref.\\cite{tsuinf} to study the self-consistent dynamics of the metric and the evolution of a highly excited quantum state with the goal of providing a quantum description of large field inflationary models {\\em without assuming an expectation value for the scalar field}. Novel quantum states were considered, which are the cosmological counterpart of the  tsunami-waves introduced in\\cite{tsurob,tsu1,tsu2} with the following properties\\cite{tsuinf}:  {\\em i)} a quantum state with non-perturbatively large number of quanta in a momentum band, i.e, a large number of high energy excitations. We first solve explicitly the case of a narrow band and then we consider more general distributions of quanta.  {\\em ii)} vanishing expectation value of the scalar field.  The rationale behind considering these quantum states is that they provide a natural description of a situation in which a state of large energy density composed mainly of energetic particles evolves in time. The main idea is to model the situation conjectured to drive inflationary cosmology:  that is an initial state with a large energy density and pressure which in turn couples to the metric leading to inflation. Efficient inflation should follow independently of the details of the initial state. In our case, this means that efficient inflation is to be obtained for different shapes of the initial particle distribution. That is, we want to avoid a {\\bf fine tuning} of the initial conditions.  [In the usual studies (classical chaotic inflation) all of the initial energy is in the classical zero mode while the quantum fluctuations are taken to be in the ground state].  After inflation, energy transfer from the heavier to lighter particles results in reheating and particle production that eventually excites the light sector and leads to a radiation or matter dominated phase.  The conditions under which a Tsunami quantum state leads to inflationary dynamics is established and  the self-consistent evolution of this quantum state and the space-time metric is studied in detail\\cite{tsuinf}.  We emphasize that we are {\\bf not} proposing here yet a new model of inflation. Instead we focus on  inflation driven by the evolution of a {\\bf quantum} state, within the framework of familiar models based on scalar fields with typical quartic potentials. This is  in contrast with the usual approach in which the  dynamics is driven by  the evolution of a homogeneous {\\bf classical} field of large amplitude.  {\\em Brief summary:} We find that inflation occurs under fairly general conditions that are the {\\em quantum } equivalent of slow-roll. There are {\\em two} consecutive but distinct inflationary stages: the first one is completely determined by the quantum features of the state. Even when the expectation value of the scalar field {\\em vanishes at all times } in this quantum state, the dynamics of the first stage gives rise to the emergence of an {\\em effective classical homogeneous condensate}. The amplitude of the effective condensate is non-perturbatively large as a consequence of the non-perturbatively large number of quanta in the band of excited wavevectors. The second stage is similar to the familiar classical chaotic scenario, and can be interpreted as being driven by the dynamics of the effective homogeneous condensate\\cite{tsuinf}. The band of excited quantum modes, if not superhorizon initially they cross the horizon during the first stage of inflation, hence they do not modify the power spectrum of scalar density perturbations on wavelengths that are of cosmological relevance today. Actually, in the explicit examples worked out in ref.\\cite{tsuinf}, the excited modes are initially superhorizon due to the generalized slow-roll condition. Therefore, in a very well defined manner, tsunami quantum states provide a quantum field theoretical justification, a microscopic basis, for chaotic inflation, explaining the classical dynamics of the homogeneous scalar field.  In section II we introduce the quantum state, obtain the renormalized equations of motion for the self-consistent evolution of the quantum state and the scale factor. In section III we  provide detailed  analytic and numerical studies of the evolution and highlight the different inflationary stages. In section IV we discuss generalized scenarios. The summary of results is presented in the conclusions. An appendix is devoted to technical details and the equations of motion for mixed states. ",
        "conclusions": "We have presented here inflation in typical scalar field theories as a consequence of the time evolution of a novel quantum state. This quantum state is characterized by a {\\em vanishing} expectation value of the scalar field, i.e, a vanishing zero mode, but  a non-perturbatively large number of quanta in a momentum band, thus its name--tsunami-wave state.  This state leads to a non-perturbatively large energy density which is localized in the band of excited quantum modes. We find that the self-consistent equations for the evolution of this quantum state and the scale factor lead to inflation under conditions that are the quantum analog of slow-roll.  The self-consistent evolution was studied analytically and numerically in a wide range of parameters for the shape and position of the distribution of excited quanta. The numerical results confirm all the features obtained from the analytic treatment.  Under the conditions that guarantee inflation,  there are two consecutive but distinct inflationary epochs. The first stage features a rapid fall-off of the Hubble parameter and is characterized by the quantum aspects of the state. During this first stage the large number of quanta in the excited band are redshifted and build up an {\\em effective homogeneous classical condensate}. The amplitude of this condensate is non-perturbatively large, of ${\\cal O}(1/\\lambda)$, as a consequence of the non-perturbatively large number of quanta in the band of excited modes.  The second stage is similar to the classical chaotic scenario and it is driven by the dynamics of this effective classical condensate, with  vanishing  expectation value of the scalar field. Under the tsunami slow-roll conditions  on the quantum state, the total number of e-folds is more than enough to satisfy the constraints of inflationary cosmology. The band of excited wave-vectors if not initially outside the causal horizon, becomes superhorizon during the first inflationary stage, therefore these excited states do not modify the power spectrum of scalar density perturbations  on wavelengths that are of cosmological relevance today.  Therefore, these tsunami-wave quantum states provide a quantum field theoretical justification of chaotic (or in general large field) inflationary models and yield to a microscopic understanding of the emergence of classical homogeneous field configurations of large amplitude as an effective collective mode built from the large number of quanta in the excited band.  In addition, we recall that it is necessary to choose an initial state that breaks the $ \\Phi \\to - \\Phi $ symmetry in classical chaotic scenarios \\cite{revius,coles,linde}. This is {\\em not} the case here. We have inflation with {\\em zero} expectation value of the scalar field.  For completeness we have also studied more general states and established the important difference between tsunami (pure or mixed) quantum states leading to inflation, and thermal mixed states which do not lead to inflation."
    },
    "0208/astro-ph0208430_arXiv.txt": {
        "abstract": "Dynamical studies of superbubbles and Wolf-Rayet ring nebulae show discrepancies from the standard, adiabatic model for wind-blown bubbles.  We therefore study the physical properties and kinematics of three candidate bubbles blown by single O stars, to evaluate whether these discrepancies are also found in these simpler objects.  Our sample candidates are N44\\,F, N44\\,J, and N44\\,M, in the outskirts of the \\hii\\ complex N44 in the Large Magellanic Cloud. We have obtained ground-based and {\\it HST} emission-line images and high dispersion echelle spectra for these objects. From the \\ha\\ luminosities and the \\oiii/\\ha\\ ratios of these nebulae, we estimate the spectral types of the ionizing stars to be O7V, O9.5V and O9.5V for N44\\,F, N44\\,J, and N44\\,M, respectively.  We find that the observed expansion velocity of 12 $\\rm km\\ s^{-1}$ for N44\\,F is consistent with the stellar wind luminosity expected from the central ionizing star, as predicted by the standard bubble model.  The observed upper limits for the expansion velocities of N44\\,J and N44\\,M are also compatible with the expected values, within the uncertainties. We also report the discovery in N44\\,F of strongly-defined dust columns, similar to those seen in the Eagle Nebula.  The photoevaporation of these dense dust features may be kinematically important and may actually govern the evolution of the shell. The inclusion of photoevaporation processes may thus undermine the apparent agreement between the observed bubble dynamics and the simple adiabatic models. ",
        "introduction": "Mechanical feedback from massive stars is a fundamental driver of galaxy evolution.  The supernovae (SNe) and supersonic stellar winds from these stars generate bubbles and shells in the interstellar medium (ISM), which may dominate the ISM structure formation in many star-forming galaxies.  The standard, adiabatic model for these shells and superbubbles \\citep[e.g.,][]{Wetal77} predicts the production of hot, $10^6 - 10^7$ K, low-density gas within these shells, which bear the products of SN and massive star nucleosynthesis.  Therefore, these bubble structures are thought to be the source of the diffuse, hot ionized component of the ISM.  Vigorous star-forming regions may generate supergiant shells that blow out of galaxy disks, thereby dispersing metals and mass far from the parent star formation event.  In addition, this mechanical energy contributes to ISM kinematics and is thought to be a primary source of turbulence \\citep[e.g.,][]{NF96,Go00}.  The effects of mechanical feedback have been clearly demonstrated, for example, in starburst galaxies \\citep{Jetal00,Setal00} and in superbubbles around OB associations \\citep[e.g.,][]{Setal92,Oey96}. However, several troubling discrepancies remain unsolved.  For example, almost all clean examples of bubbles blown by Wolf-Rayet stars and of superbubbles present shells that are too small for their observed parent stellar population \\citep[e.g.,][]{TC82,Detal95}.  Many superbubbles also show anomalous kinematics and X-ray emission \\cite[e.g.,][]{ro82,CM90,WH91}. To resolve these puzzles and more clearly understand the mechanical feedback process, it is therefore necessary to examine simpler systems.  In active star-forming regions, high concentrations of OB stars collectively produce superbubbles. But in their peripheries, where stars are loosely distributed,  discrete bubbles may be produced by individual stars. \\ha\\ images of \\hii\\ regions ionized by OB associations in the Large Magellanic Cloud (LMC) have revealed large (50--150 pc) gas shells around the OB associations and numerous small ($<$ 15 pc), ring-like nebulae outside the large shell \\citep[see, e.g.,][] {DEM}. These small nebulae are candidate wind-blown bubbles of single O or B stars \\citep{WD98}, and offer perhaps the simplest test of the standard, adiabatic bubble model.  We have investigated the kinematics and physical nature of three ring-like nebulae in N44 \\citep{hen56} in the LMC.  N44 is a bright \\hii\\ complex with a 44~pc$\\times$67~pc superbubble at the center and several compact \\hii\\ regions along the shell rim.  In the surrounding area, small dense \\hii\\ regions as well as extended diffuse nebulae are also present at distances up to 160 pc. Figure~\\ref{n44tot} shows these features and their identifications given by \\citet{hen56}. Three OB associations exist in N44 \\citep{LH70}: LH~47 in the central shell, LH~48 in the compact \\hii\\ region N44\\,I, and LH~49 partially embedded in the bright \\hii\\ region N44\\,D.  The three nebulae that we have studied are N44\\,F, N44\\,J, and N44\\,M.  N44\\,F is a bright, circular \\hii\\ region at the northwest rim of the superbubble and is adjacent to a bright filament connected to the main nebula. In contrast, N44\\,J and N44\\,M, appearing respectively on the  northern and eastern parts of N44, are fainter and relatively isolated. None of these three nebulae have been studied since their original identification by \\citet{hen56}.  We have obtained ground-based CCD images and high-dispersion echelle spectra of N44\\,F, N44\\,J, and N44\\,M to study the morphology and kinematics of these nebulae.  Additionally, we have used {\\it Hubble Space Telescope} ({\\it HST}) WFPC2 images of N44\\,F in the \\ha\\ and \\sii\\ lines to analyze its ionization structure.  In this paper, we first describe these observations in \\S 2, then present the analysis and results for N44\\,F in \\S 3 and N44\\,J and N44\\,M in \\S 4. We present our conclusions in \\S 5. ",
        "conclusions": "Massive stars inject a large amount of energy into the ISM through their ionizing radiation, fast stellar winds, and SNe.  The stellar winds, which dominate the early mechanical feedback, are able to sweep the ambient medium into an expanding shell and create a cavity around the parent stars.  \\hii\\ regions ionized by massive stars are therefore thought to evolve from compact \\hii\\ regions to ring-like nebulae.  We have studied three candidate wind-blown bubbles in the N44 complex: N44\\,F, N44\\,J, and N44\\,M. They all have diameters of $\\sim$10~pc. Their ionizing fluxes and \\oiii/\\ha\\ ratios suggest that their ionizing stars have spectral types of roughly O7V, O9.5V and O9.5V, respectively. N44\\,F has an expansion velocity of 12 \\kms, while the other two have expansion velocities $<$ 8\\kms. The stellar wind luminosity implied by the ionizing stars and the observed bubble dynamics appear consistent with that expected in the standard, adiabatic bubble model within the limits of observational uncertainties.  Similar results were found by \\citet{OM94} for two bubbles around single O star in M33.  This is in sharp contrast to the anomalous kinematics seen in superbubbles and Wolf-Rayet bubbles. We caution, however, that our observed expansion velocities may not provide the most stringent constraints when comparing with models, because the expected expansion velocities of N44\\,J and N44\\,M are comparable to or smaller than the convolution of the isothermal sound velocity ($\\sim$10 \\kms) and the instrumental FWHM (13 \\kms).  Deeper, higher-resolution spectroscopic observations of the \\nii\\ or \\oiii\\ lines are needed to pin down the expansion velocity for a more definitive test of models.  We also report the discovery of several dust pillars in N44\\,F, which are similar to those found in the Eagle Nebula \\citep{hes96} and 30 Doradus \\citep{sco98}.  These structures are indicative of the evaporation of molecular clouds by the ionizing source.  The examples in N44\\,F are especially well-suited to investigating the process and role of photoevaporation in star formation and shell evolution, because of their location within such a well-defined, wind-blown bubble.  We found that photoevaporation, as evidenced by the pillar structure, may be dynamically important for the shell evolution in N44\\,F, slowing down the expanding shell and inhibiting its growth. These photoevaporation processes may thus undermine the seeming agreement between the observed bubble dynamics and the simple adiabatic bubble model."
    },
    "0208/hep-ph0208261_arXiv.txt": {
        "abstract": "\\singleandabitspaced  In models in which all of the Standard Model fields live in extra ``universal'' dimensions, the lightest Kaluza-Klein (KK) particle can be stable.  Calculations of the one-loop radiative corrections to the masses of the KK modes suggest that the identity of the lightest KK particle (LKP) is mostly the first KK excitation of the hypercharge gauge boson.  This LKP is a viable dark matter candidate with an ideal present-day relic abundance if its mass is moderately large, between $600$ to $1200$ GeV\\@. Such weakly interacting dark matter particles are expected to become gravitationally trapped in large bodies, such as the Sun, and annihilate into neutrinos or other particles that decay into neutrinos. We calculate the annihilation rate, neutrino flux and the resulting event rate in present and future neutrino telescopes.  The relatively large mass implies that the neutrino energy spectrum is expected to be well above the energy threshold of AMANDA and IceCube.  We find that the event rate in IceCube is between a few to tens of events per year. ",
        "introduction": "The premiere astrophysical conundrum is the nature and identity of dark matter.  The accumulated body of evidence in favor of the existence of dark matter is by now overwhelming: Studies of the cosmic microwave background \\cite{cmb}, high redshift supernovae \\cite{sn}, galactic clusters and galactic rotation curves \\cite{rotation} indicate that the matter density of the universe is $\\Omega_M \\simeq 0.3-0.4$. Constraints from big-bang nucleosynthesis, however, limit the baryonic matter density to a small fraction of this number \\cite{bbn}.  Furthermore, the observed density of luminous matter is also very small, $\\Omega_L < 0.01$ \\cite{luminous}.  Therefore, the vast majority of the mass in the universe is dark.  Additionally, cosmic microwave background studies and large scale structure formation requires that the majority of the dark matter be cold (non-relativistic) \\cite{cmb,structure}.  Dark matter could exist in several forms. Perhaps the most interesting possibility is a neutral, stable, weakly interacting particle arising from physics beyond the Standard Model.  Candidates for such an animal abound, including the lightest supersymmetric particle, the axion, etc.  The possibility of a stable Kaluza-Klein (KK) excitation as particle dark matter was raised many years ago \\cite{KolbSlansky} and more recently \\cite{DDG2}.  Models in which all of the Standard Model fields propagate in ``universal'' extra dimensions \\cite{ACD} (for earlier work, see \\cite{antoniadis}) provide the most natural home for KK dark matter \\cite{CMS,ST,CFM}. This is because bulk interactions do not violate higher dimensional momentum conservation (KK number), and in these models all of the couplings among the Standard Model particles arise from bulk interactions.  To generate chiral fermions at the zero mode level, the extra compact dimension(s) must be modded out by an orbifold.  For five dimensions this is $S^1/Z_2$, while in six dimensions $T^2/Z_2$ is suitable and has other interesting properties \\cite{ACD} including motivation for three generations from anomaly cancellation \\cite{DP} and the prevention of fast proton decay \\cite{ADPY}.  An orbifold does, however, lead to some of the less appealing aspects of the model.  Brane-localized terms can be added to both orbifold fixed points that violate KK number. If these brane localized terms are symmetric under the exchange of the two orbifold fixed points, then a remnant of KK number conservation remains, called KK parity.  All odd-level KK modes are charged under this discrete symmetry thereby ensuring that the lightest level-one KK particle (LKP) does not decay.  This is entirely analogous to exactly conserved $R$-parity in supersymmetric models which ensures the lightest supersymmetric particle is stable.  The stability of the LKP suggests it could well be an interesting dark matter candidate.  The identity of the lightest KK particle crucially depends on the mass spectrum of the first KK level.  At tree-level, the mass of each excitation is simply \\begin{equation} \\left(m^1_i\\right)^2 = \\frac{1}{R^2} + \\left(m^0_i\\right)^2 \\end{equation} where $R$ is the compactification radius that could be as large as $1/(300 \\; \\mathrm{GeV})$ without conflict with experiment \\cite{ACD}. However, brane-localized terms can be added on the orbifold fixed points that significantly modify the masses and higher dimensional wavefunctions.  The tree-level (matching) contributions at the cutoff scale of the higher dimensional theory are not calculable, but can be estimated using naive dimensional analysis. The result is that the size of these terms are suppressed by a volume factor, of order $\\Lambda R$ where $\\Lambda$ is the cutoff. More importantly, these brane-localized terms are renormalized upon evolving from the matching scale to the mass scale of the light KK modes \\cite{GGH}.  For universal extra dimensions, Cheng, Matchev, and Schmaltz showed that the tree-level mass formula receives significant one-loop radiative corrections from log-enhanced brane-localized terms on the orbifold fixed points \\cite{CMS}. These radiative corrections are, in many cases, larger than the shifts in the tree-level masses resulting from the masses of the zero modes. Indeed, here we will generally assume that these contributions dominate over the tree-level volume-suppressed matching contributions. Then, with the further assumption that the KK excitation of the Higgs does not receive a (significant) brane-localized negative contribution to its mass, the identity of the lightest KK state is identified as the first KK excitation of the photon.  Like the ordinary photon, the KK photon is an admixture between the first KK hypercharge gauge boson and the first KK neutral SU(2) gauge boson.  However, this identification is somewhat misleading, as \\cite{CMS} point out, since the mixing angle for the level-one KK gauge bosons is generally much smaller than the Weinberg angle.  A leading order approximation to the mass of the lightest $\\B$ state is \\begin{equation} m_{\\B}^2 \\simeq \\frac{1}{R^2} \\left[ 1 + \\frac{g'^2}{16 \\pi^2} \\left( - \\frac{39 \\zeta(3)}{2 \\pi^2} - \\frac{1}{3} \\ln \\Lambda R + (2 \\pi R v)^2 \\right) \\right] \\end{equation} neglecting higher order ${\\cal O}(\\sin \\theta^1 v^2)$ corrections. This approximation is equivalent to identifying LKP $\\equiv \\gamma^1 \\simeq \\B$, which we do for the remainder of the paper.  The relic density of the $\\B$ has been calculated in a recent paper by Servant and Tait \\cite{ST}.  Assuming the LKPs were once in thermal equilibrium, they found that the relic density is in the favorable region for providing the cold dark matter of the universe, $\\Omega_{\\B} h^2 = 0.16 \\pm 0.04$, when the mass is moderately heavy, between $600$ to $1200$ GeV\\@. The range of mass depends on the relative importance of coannihilation with KK modes near in mass to the LKP\\@. We shall see later that coannihilation is active throughout the parameter space when the KK mass spectrum is obtained using the radiative corrections arising from renormalized brane-localized terms. Direct detection of the LKP as dark matter has been considered by Cheng, Feng, and Matchev \\cite{CFM}.  They emphasized that unlike the case of a neutralino LSP, the bosonic nature of the LKP means there is no chirality suppression of the annihilation signal into fermions.  The annihilation rate of the LKP is therefore roughly proportional to the (hypercharge)$^4$ of the final state, leading to a large rate into leptons.  The large mass of the LKP suggests that dark matter detection experiments sensitive to very heavy mass relics should be among the most promising methods of detection.  In this paper, we explore the signal expected at present and future high energy neutrino telescopes that can probe precisely this type of heavy dark matter.\\footnote{Note that \\cite{ST,CFM} also remarked on the potential importance of indirect detection at neutrino telescopes.} ",
        "conclusions": "The prospects for indirect detection of $\\B$ dark matter is very promising at kilometer scale neutrino telescopes.  Using the one-loop radiative corrections to the KK mass spectrum, we showed that if the $\\B$ lies in the mass range in which it has an acceptable present-day relic density to be the dark matter of the universe, then a $1$ km$^2$ neutrino telescope is expected to detect between a few to tens of $\\B$ annihilation events in the Sun per year. This mass range of the $\\B$ is between about $600$ to $800$ GeV where coannihilations with right-handed KK leptons plays a significant role.  The relatively large signal relies on the Sun reaching equilibrium between $\\B$ capture and annihilation, which we explicitly verified for this range of $\\B$ masses.  Although we have focused on the $\\B$ mass range that results in the appropriate dark matter relic density as currently consistent with cosmology, it is straightforward to extrapolate to other masses. In fact, there are two particle physics effects that are expected to lead to a \\emph{lowering} of the $\\B$ mass for a fixed relic density. The first effect is the inclusion of coannihilation with left-handed KK leptons.  The second effect, in a six dimensional model, is that there are typically multiple LKPs corresponding to multiple conserved parities.  In both of these cases the lowering of the mass of the LKP(s) for a fixed relic density implies a larger event rate at neutrino telescopes.  We are therefore optimistic that future detectors will find or exclude this fascinating possibility."
    },
    "0208/hep-th0208013_arXiv.txt": {
        "abstract": "width} \\setlength{\\abstractwidth}{\\textwidth} \\addtolength{\\abstractwidth}{-6pc}        \\thispagestyle{empty} \\pagestyle{plain}  \\newcommand{\\onefigure}[2]{\\begin{figure}[htb] \\begin{center}\\leavevmode\\epsfbox{#1.eps}\\end{center}\\caption{#2\\label{#1}}  \\end{figure}} \\renewcommand{\\thefootnote}{\\fnsymbol{footnote}} \\renewcommand{\\thanks}[1]{\\footnote{#1}} % \\newcommand{\\starttext}{ \\setcounter{footnote}{0} \\renewcommand{\\thefootnote}{\\arabic{footnote}}} \\renewcommand{\\theequation}{\\thesection.\\arabic{equation}} \\newcommand{\\be}{\\begin{equation}} \\newcommand{\\bea}{\\begin{eqnarray}} \\newcommand{\\eea}{\\end{eqnarray}} \\newcommand{\\beq}{\\begin{equation}} \\newcommand{\\ee}{\\end{equation}} \\newcommand{\\eeq}{\\end{equation}} \\newcommand{\\N}{{\\cal N}} \\newcommand{\\<}{\\langle} \\newcommand{\\oc}{{\\cal O}_C} \\renewcommand{\\a}{\\alpha} \\renewcommand{\\b}{\\beta} \\newcommand{\\half}{{1\\over 2}} \\renewcommand{\\>}{\\rangle} \\def\\ba{\\begin{eqnarray}} \\def\\ea{\\end{eqnarray}} \\newcommand{\\PSbox}[3]{\\mbox{\\rule{0in}{#3}\\special{psfile=#1}\\hspace{#2}}}  \\def\\C{Complementarity} \\def\\c{complementarity} \\def\\hp{Holographic Principle} \\def\\14{{1\\over4}} \\def\\12{{1 \\over 2}} \\def\\eq{&=&} \\def\\tro{\\tilde{\\rho}} \\def\\d{\\partial} \\def\\dt{\\partial_{\\tau}} \\def\\ds{\\partial_{\\sigma}} \\def\\h3{h^{3\\over 2}} \\def\\R{\\bar{R}} \\def\\qft{quantum field theory} \\def\\>{\\rangle} \\def\\<{\\langle} \\def\\sc {Schwarzschild} \\def\\ls{\\sqrt{\\alpha'}} \\def\\des{de Sitter space} \\def\\f{\\Phi} \\def\\pr{Poincare recurrence} \\def\\prs{Poincare recurrences} \\def\\eu{e^{u\\over2}}   \\begin{document} \\renewcommand{\\theequation}{\\thesection.\\arabic{equation}} \\begin{titlepage} \\bigskip \\rightline{SU-ITP 00-25} \\rightline{MIT-CTP-3295} \\rightline{hep-th/0208013}  \\bigskip\\bigskip\\bigskip\\bigskip  \\centerline{\\Large \\bf {Disturbing Implications of a Cosmological Constant}}  \\bigskip\\bigskip \\bigskip\\bigskip  \\centerline{\\it L. Dyson$^{a,b}$, M. Kleban$^a$, L. Susskind$^a$  } \\medskip \\medskip \\centerline{$^a$Department of Physics} \\centerline{Stanford University} \\centerline{Stanford, CA 94305-4060} \\medskip \\medskip \\centerline{$^b$Center for Theoretical Physics} \\centerline{Department of Physics}\\centerline{Massachusetts Institute of Technology} \\centerline{Cambridge, MA 02139} \\medskip \\medskip  \\bigskip\\bigskip \\begin{abstract} In this paper we consider the implications of a cosmological constant for the evolution of the universe, under a set of assumptions motivated by the holographic and horizon complementarity principles. We discuss the ``causal patch\" description of spacetime required by this framework, and present some simple examples of cosmologies described this way. We argue that these assumptions inevitably lead to very deep paradoxes, which seem to require major revisions of our usual assumptions. \\medskip \\noindent ",
        "introduction": "As emphasized by Penrose many years ago, cosmology can only make sense if the world started in a state of exceptionally low entropy. The low entropy starting point is the ultimate reason that the universe has an arrow of time, without which the second law would not make sense. However, there is no universally accepted explanation of how the universe got into such a special state. In this paper we would like to sharpen the question by making two assumptions which we feel are well motivated from observation and recent theory. Far from providing a solution to the problem, we will be led to a disturbing crisis.  Present cosmological evidence points to an inflationary beginning and an accelerated de Sitter end. Most cosmologists accept these assumptions, but there are still major unresolved debates concerning them. For example, there is no consensus about initial conditions. Neither string theory  nor quantum gravity  provide a consistent starting point for a discussion of the initial singularity or why the entropy of the initial state is so low. High scale inflation postulates an initial de Sitter starting point with Hubble constant roughly $10^{-5}$ times the Planck mass. This implies an initial holographic entropy of about $10^{10}$, which is extremely small by comparison with today's visible entropy. Some unknown agent initially started the inflaton high up on its potential, and the rest is history.  Another problem involves so-called transplanckian modes. The quantum fluctuations which seed the density perturbations at the end of inflation appear to have originated from modes of exponentially short wave length. This of course conflicts with everything we have learned about quantum gravity from string theory.  The same problem occurs when studying black holes.  In the naive free field theory of Hawking radiation, late photons appear to come from exponentially small wavelength transplanckian modes \\cite{uglum}. We now know that this is an artifact of trying to describe the complex interacting degrees of freedom of the horizon by quantum field theory defined on both the interior and exterior of the black hole. A consistent approach based on black hole complementarity \\cite{stretch} describes the black hole in terms of  strongly interacting degrees of freedom of Planckian or string scale, and restricts attention to the portion of the space--time outside the horizon.     The late time features of  a universe with a cosmological constant are also not well understood. The conventional view is that the universe will end in a de Sitter phase with all matter being infinitely diluted by exponential expansion. All comoving points of space fall out of causal contact with one another. The existence of a future event horizon implies that the objects that string theory normally calculates, such as S--Matrix elements, have no meaning \\cite{quint}. In addition, there are questions of stability of de Sitter space which have been repeatedly raised in the past \\cite{tsamis}. The apparent instabilities are due to infrared quantum fluctuations which seem to be out of control. Thus the final state is also problematic.  In our opinion both the transplanckian and the late time   problems have a common origin. They occur because we try to build a quantum mechanics of the entire global spacetime--including regions which have no operational meaning to a given observer, because they are out of causal contact with that observer. The remedy suggested by the black hole analogue is obvious;  restrict all attention to a single causal patch \\cite{tom,lisa,birthday}. As in the case of black holes, the quantum description of such a region should satisfy the usual principles of quantum mechanics \\cite{stretch}. In other words, the theory describes a closed isolated box bounded by the observer's horizon, and  makes reference to no other region. Furthermore,  as in the case of black holes, the mathematical description of this box should satisfy the conventional principles of linear unitary quantum evolution.  Perhaps the most important conceptual lesson that we have learned from string theory is that quantum gravity is a special case of quantum mechanics. Thus far every nonperturbative well-defined formulation of string theory involves a hamiltonian, and a  space of states for it to  act on. This includes matrix theory and all the versions of AdS/CFT. The question of this paper is whether the usual rules apply to cosmology, and can they explain, or at least allow, the usual low entropy starting point. In the following  we will  assume the usual connections between quantum statistical mechanics and thermodynamics.\\footnote{After completion of this work, we received \\cite{dongsu}, which considered a similar scenario.}  These assumptions--together with the existence of a final cosmological constant--imply that the universe is eternal but finite. Strictly speaking, by finite we mean that the entropy of the observable universe is bounded, but we can loosely interpret this as saying the system is finite in extent. On the average it is in a steady state of thermal equilibrium.  This is a very weak assumption, because almost any  large but finite system, left to itself for a long enough time, will equilibrate (unless it is integrable) \\cite{mark}. However, intermittent fluctuations occur which temporarily disturb the equilibrium. It is during the return to equilibrium that interesting events and objects form\\footnote{ After completion of this work, we became aware of \\cite{vilenkin}, in which the authors consider a related scenario where the inflaton can repeatedly tunnel to a false vacuum, which can result in an infinite cycle of new inflationary periods.}.  The essential point can be illustrated with an analogy. Instead of the universe, let's consider a sealed box full of gas molecules. Start with a particular low entropy initial condition with all the molecules in a very small volume in one corner of the box. The molecules are so dense that they form a fluid. When released the molecules flow out from the corner and eventually fill the box uniformly with gas. For some time the system  is far from equilibrium. During this time, the second law insures that the entropy is increasing and interesting things can happen. For example, complex ``dissipative structures\" such as eddy flows, vortices, or even life can form. Eventually the system reaches equilibrium, and all structures disappear. The system dies an entropy death. This is the classical hydrodynamic  description of the evolution of a ``universe\". But this description is only correct for  time intervals which are not too long.  Let $S$ be the final thermodynamic entropy of the gas. Then on time scales of order \\be T_r \\sim \\exp{S} \\label{trec} \\ee the system will undergo Poincare recurrences \\footnote{ See the appendix for a discussion on the quantum Poincare recurrence theorem.} \\cite{lisa,birthday}. Such a recurrence can bring all the particles back into the corner of the room. On such long time scales the second law of thermodynamics does not prevent rare events, which effectively reverse the direction of entropy change. Obviously, the recurrence allows the entire process of cosmology to begin again, although with a slightly different initial condition. What is more, the sequence of recurrences will stretch into the infinite past and future. The question  then is whether the origin of the universe can be a naturally occurring fluctuation, or must it be due to an external agent which starts the system out in a specific low entropy state?  We will discuss this in greater detail in Section 6.   \\setcounter{equation}{0} ",
        "conclusions": ""
    },
    "0208/astro-ph0208424_arXiv.txt": {
        "abstract": "Images obtained with the Gemini Multi-Object Spectrograph (GMOS) are used to investigate the stellar content and distance of the dwarf irregular galaxy Kar 50. The brightest object is an HII region, and the bright stellar content is dominated by stars with $g'-r' < 0$. The tips of the main sequence and the red giant branch are tentatively identified near $r' = 24.9$ and $i' \\sim 25.5$, respectively. The galaxy has a blue integrated color with no significant color gradient, and we conclude that Kar 50 has experienced a recent galaxy-wide episode of star formation. The distance estimated from the brightest blue stars indicates that Kar 50 is behind the M81 group, and this is consistent with the tentative RGB-tip brightness. Kar 50 has a remarkably flat central surface brightness profile, even at wavelengths approaching $1\\mu$m, although there is no evidence of a bar. In the absence of another large star-forming episode, Kar 50 will evolve into a very low surface brightness galaxy. ",
        "introduction": "The Local Group contains over 30 dwarf galaxies that span a range of morphological types (e.g. van den Bergh 2000). While such nearby systems are fundamental laboratories for investigating dwarf galaxy evolution, many are satellites of M31 and the Milky-Way, and hence have been affected by hierarchal interactions (e.g. Mayer et al. 2001). In order to develop a more comprehensive understanding of dwarf galaxies, including the effects of environment, it is necessary to study objects outside of the Local Group.  Located at a distance of $\\sim 3.5$ Mpc, the M81 group contains a number of dwarf galaxies, some of which are systematically different from those in the Local Group (Caldwell et al. 1998). Karachentseva, Karachentsev \\& B\\\"{o}rngen (1985, hereafter KKR) list possible dwarf members of the M81 group. Many of these have yet to be studied in detail, and some will likely turn out not to be members of the M81 group. This being said, the census of galaxies in the M81 group is likely far from complete, and Froebrich \\& Meusinger (2000) report the detection of additional possible low surface brightness members.  One of the faintest, and hence intrinsically most interesting, systems in the KKR compilation is Kar 50. KKR noted that Kar 50 is `unusually filamentary with barely outlined knots', and they partially resolved Kar 50 into stars using images recorded during 1 arcsec seeing conditions. Various observational properties of Kar 50 are summarized in Table 1.  Motivated by the possible intrinsic faintness of Kar 50, coupled with its unusual appearance, we decided to include this galaxy in a background field obtained as part of a program to study the outer regions of NGC 2403. The observations and the data reduction procedures are discussed in \\S 2, and these data are used to investigate the integrated photometric properties of Kar 50 in \\S 3. Our $i'$ and $z'$ images are of particular interest for work of this nature, as they sample redder wavelengths than previous studies, and hence are better probes of any cool stellar component, such as might arise from an old population. The properties of the brightest stars in Kar 50 are investigated in \\S 4. A distance based on the brightest stars is calculated in \\S 5, and we conclude that Kar 50 is located well behind the M81 group. A brief summary and discussion of the results follows in \\S 6. ",
        "conclusions": "Deep multicolor images obtained with the GMOS on the Gemini North telescope have been used to investigate the structure and stellar content of the dwarf irregular galaxy Kar 50. The presence of a significant population of bright blue stars, coupled with the blue integrated colors of the galaxy and the flat color profiles, which indicate that the stellar content of the galaxy is well mixed, indicate that Kar 50 has experienced a recent galaxy-wide star-forming episode. Kar 50 may thus be one of the small percentage of dwarf irregular galaxies that at any given time host a strong burst of star formation (Dohm-Palmer et al. 1998). In the remainder of this Section we discuss (1) the stellar content of Kar 50, with emphasis on checking the distance estimated from the brightest blue stars, and (2) the evolution of the galaxy.  \\subsection{The Distance and Stellar Content of Kar 50}  While contamination from foreground stars and background galaxies complicates efforts to study the resolved stellar content of Kar 50 when $r' < 25$, in \\S 4 it is demonstrated that a secure population of bright galaxy members can be identified using $g'-r'$ color. The application of a color criterion to reject contaminating objects will of course cause some objects, in this case those that have red colors, to be excluded from the analysis. The brightest red supergiants (RSG)s in dwarf galaxies have M$_V \\sim -7$ (Rozanski \\& Rowan-Robinson 1994), and thus would be expected to occur near $r' \\sim 23.7$ at the distance of Kar 50. While this is well within the detection limits of our data, the agreement between the numbers of objects with $g'-r' < 0$ in the Kar 50 and background fields indicates that RSGs do not occur in great numbers in Kar 50.  The brightest object in Kar 50 has an SED suggesting that it is an HII region, and if this is the case then the $r'$ flux will be dominated by H$\\alpha$ emission. With an assumed distance modulus of 30.4, this object has a luminosity of $1.3 \\times 10^{37}$ ergs/sec in $r'$, which is consistent with the peak H$\\alpha$ luminosity of HII regions in other irregular galaxies (Youngblood \\& Hunter 1999). With a distance modulus of 27.5 then the luminosity of the HII region drops to $8.9 \\times 10^{35}$ erg/sec, which is near the lower end of what is seen in irregular galaxies. Thus, the luminosity of this object is consistent with it being an HII region. Nevertheless, this object is very compact, as the maximum size of 40 parsecs is a factor of 2 smaller than what is seen among HII regions in other irregular galaxies (Youngblood \\& Hunter 1999).  The distance of $12_{-4}^{+7}$ Mpc measured from the brightest blue stars places Kar 50 well behind the M81 group. There are significant uncertainties inherent to the use of the brightest blue stars as standard candles, and so in the remainder of this section we check if this distance is consistent with the properties of other resolved objects in Kar 50.  The RGB-tip is a prime distance indicator for old stellar systems, as it produces a clear break in luminosity functions and CMDs. The RGB-tip occurs near M$_{i'} \\sim -4$ in old metal-poor populations (e.g. Davidge et al. 2002), and hence would occur at $i' \\sim 23.5$ if Kar 50 were in the M81 group. Given the evidence for moderately recent star formation, a well populated AGB sequence might also be anticipated above the RGB-tip in Kar 50. There is no evidence in the CMDs of Kar 50 of either a well-defined RGB peaking near $i' \\sim 23.5$ or a statistically significant AGB sequence when $i' < 23.5$, which would produce an excess red population with respect to the background at this brightness. Moreover, it is evident from Figure 7 that many of the sources on the $(i', r'-i')$ CMD have $(r'-i') \\sim 0.0$, whereas RGB stars will have $r'-i' \\geq 0.3$. There is an apparent onset of stars in the $(i', r'-i')$ CMD with $i' \\sim 25.4$ and $r'-i' \\sim 0.3$. If this is the RGB-tip then this supports a distance modulus $\\sim 29.4$, which falls within the $\\pm 1$ mag errors in the distance modulus computed from the brightest blue stars. We are reluctant to assign a firm distance estimate based on the RGB-tip because it is at the faint limit of our data, and AGB stars can complicate distances measured from this feature.  The brightest stars at visual wavelengths in star-forming galaxies tend to have spectral types A -- F (e.g. Humphreys \\& McElroy 1984), and the stars used to estimate the distance to Kar 50 have colors that are consistent with this range of spectral types (\\S 5). Humphreys \\& McElroy (1984) list the effective temperatures and brightnesses of supergiants and upper main sequence stars in the Milky-Way and Magellanic Clouds, and we have used these data to locate these sequences on the $(r', g'-r')$ CMD. Colors were calculated using the relation between effective temperature and $g'-r'$ predicted by the solar metallicity log(g)=4.5 atmosphere models of Lenz et al. (1998), while M$_{r'}$ was computed from the Krisciunas et al. (1998) transformation relations. The Lenz et al. (1998) models have surface gravities appropriate for main sequence stars, but not supergiants, and this likely introduces systematic errors in $g'-r'$ on the order of a few hundredths of a magnitude in the placement of the supergiant sequence.  The loci defined by Ia and Ib supergiants, as well as the main sequence, are shown in Figure 8 based on the adopted distance modulus of 30.4. The type Ia supergiant sequence loosely matches the upper envelope of stars in Kar 50, while the Ib sequence passes through the main body of the CMD. In addition, the bright end of the main sequence is in reasonable agreement with the clump of stars we identified in \\S 4 as the upper main sequence tip. Based on these comparisons we predict that spectroscopic studies will reveal that 1) the brightest stars in Kar 50 are Ia supergiants, and (2) the brightest main sequence stars in Kar 50 are early-type O stars.  \\subsection{Some Comments on the Structure and Evolution of Kar 50}  The light and color profiles of Kar 50 offer clues about the past evolution of the galaxy. The light profile of Kar 50 is flat over a linear scale of $\\sim 1$ kpc, while the color profile suggests that the stellar content is well mixed throughout the galaxy. Light and color profiles similar to those in Kar 50 are not uncommon among Magellanic Irregular galaxies in the Virgo (e.g. Figures 8 and 9 of Binggeli \\& Cameron 1993) and M81 (Bremnes, Binggeli, \\& Prugniel 1998; Froebrich \\& Meusinger 2000) groups.  Bar instabilities can mix gas throughout a system (e.g. Friedli, Benz, \\& Kennicutt 1994), and produce flat light profiles. The disruption of bars may play an important role in the morphological evolution of dwarf galaxies, with the final system having a lower density and different kinematical properties (Mayer et al. 2001). However, despite the flat light and color profiles of Kar 50, there is no evidence of a bar in Figure 1.  Most surveys of the structural characteristics of nearby dwarf systems have relied on observations at blue and visible wavelengths. The $z'$ filter samples a wavelength interval that is less affected by recent star formation than filters covering shorter wavelengths, although the relatively blue $i'-z'$ color of Kar 50 suggests that a significant fraction of the light near $1\\mu$m still comes from a young component. $JHK$ observations of Kar 50 will be of interest, as they will allow firmer constraints to be placed on (1) the size of any old stellar subtrate, (2) the degree of central concentration of this component, and (3) the presence of a bar, given that bars are more easily detected in $H$ than at optical wavelengths (Eskridge et al. 2000).  In the absence of subsequent star-forming episodes, Kar 50 will evolve into a very low surface brightness system. To estimate the surface brightness of Kar 50 after fading we adopt, based on the integrated colors of the galaxy, a M/L = 0.5 solar, and further assume that the final M/L = 1.0, based on the integrated color of a `typical' low surface brightness system in the Virgo cluster, for which $\\overline{B-V} = 0.6$ (e.g. Impey, Bothun, \\& Malin 1988). Kar 50 will then fade by 0.8 mag per square arcsec if the population ages passively, and the central surface brightness will drop to $\\sim 25$ mag arcsec$^{-2}$ in $g'$. This is within the range of central surface brightnesses seen among low surface brightness galaxies in Virgo, some of which also have flat central light profiles (e.g. Impey et al. 1988)."
    },
    "0208/astro-ph0208338_arXiv.txt": {
        "abstract": "A detailed kinematic analysis of ionized gas in the nearby irregular galaxy \\objectname{NGC 4449} is presented. Observations are conducted in the spectral lines of H{$\\alpha$} and [\\ion{S}{2}]. Our scanning Fabry--Perot interferometric observations are presented from both a global as well as a local perspective. We have analysed the global velocity field, the spatially extended diffuse gaseous component (DIG), the \\ion{H}{2} region populations, and, furthermore, have determined the rotation curve based on the heliocentric radial velocities of the global H{$\\alpha$} spatial distribution.  Our results for \\objectname{NGC 4449} show that the optical velocity field has a decreasing value in radial velocity along the optical bar from NE to SW, presenting an anticorrelation relative to the outer velocity field of the \\ion{H}{1} component. This is in agreement with previous studies. The DIG component that permeates the entire galaxy was analysed (up to a limiting surface brightness of $\\sim 3.165\\times10^{-5}$ ergs cm$^{-2}$ s$^{-1}$ steradian$^{-1}$) in terms of its radial velocity field as well as its velocity dispersions. We find that the diffuse gas component presents peculiar kinematical features such as abrupt velocity gradients and highly supersonic velocity dispersions ($\\sigma\\sim$ 4 times the values of the nearest \\ion{H}{2} regions) but that its kinematical and dynamical influence is important on both global and local scales. The optical rotation curve of this nearby irregular shows that the NE sector rotates like a solid body ($V_{rot}\\sim$40 km s$^{-1}$ at $R=$2 kpc). In contrast, for the SW side, our results are not conclusive; the behavior of the gas at those locations is chaotic. We conclude that the origin of such complex kinematics and dynamics is undoubtedly related to the aftermath of an interaction experienced by this galaxy in the past. ",
        "introduction": "In terms of frequency of occurrence, the most common morphology for galaxies in the Universe are the Irregular (Irr) strain \\citep{gallagher84}. Irregulars may serve as morphological Rosetta stones bridging the low and high redshift Universe \\citep{block01}. Historically, drawings of Irrs may be found as early as 1847 (the Large Magellanic Cloud, classified by Hubble as an irregular, was sketched by Sir John Herschel in that year). When imaged in the near--infrared, a handful of irregulars betray a magnificent -- albeit weak -- spiral density wave; examples are NGC 5195 and NGC 922 (the latter galaxy even shows spiral arm modulation). Generally, however, there is no decoupling of the Population I and Population II disks; many optical irregulars remain irregular in the near--infrared. One of the most intriguing questions yet to be answered is exactly what triggers/drives the formation of stars in galaxies.   Here the Irr species offer unprecedented opportunities, for they are excellent laboratories in which one can examine the star formation process --and its influence on the interstellar medium (ISM)-- often in the complete absence of spiral density waves. Furthermore, the existence of fossil interacting features in some of these galaxies makes them worthy of detailed study \\citep{hunter00}. The global study of the kinematical behavior of the gas in such environments is dynamically important. The presence of embedded superbubbles, giant \\ion{H}{2} regions, filaments, supernova remnants (SNRs) and their combined effects, makes the global velocity field of irregular galaxies both stimulating and important to analyze.   The face of the galaxy \\objectname{NGC 4449}, a nearby Magellanic type Irr galaxy, is highly peculiar, with H${\\alpha}$ streamers and filaments everywhere. Not only does \\objectname{NGC 4449} present giant \\ion{H}{2} regions, but also a SNR, superbubbles, and a diffuse emission (which embraces almost the entire optical boundary). The faint H${\\alpha}$ filament population alluded to above is intriguing; the filaments delineate the inner edge of a 2 kpc diameter \\ion{H}{1} supershell \\citep{hunter97} located at the northwest side of the main optical body. At other wavelengths, there is a body of evidences suggesting a close connection between the diffuse nebulae, the diffuse emission and the X--ray emission from the center to the west of the main body \\citep{dellaceca97,vogler97,bomans97}. As in the Large Magellanic Cloud, the galaxy \\objectname{NGC 4449} presents a  bar--like morphology; it is located at a distance of 3.7 Mpc \\citep{bajaja94}. \\objectname{NGC4449} is not isolated in the sky; it is a member of the CVn I Cloud which includes a number of Im dwarfs and galaxies such as NGC 4214, NGC 4244 and IC 4182. For a list of relevant parameters, the reader is referred to Table~\\ref{parameters}.   \\objectname{NGC 4449} shows an unusually rich distribution of blue supergiant stars at its northern periphery \\citep{hunter97b}; this conclusion has been enhanced by ultraviolet (UV) imaging \\citep{hill98}. The UV results indicate that the UV--emitting populations (OB complexes) lie in the northern part of \\objectname{NGC 4449}, while direct propagation of star formation appears possible south of the bar. The latter conclusion was reached using the H$\\alpha$/FUV(2231\\AA) ratio, from which age estimations of the OB complexes of this galaxy were derived, indicative of gradients in stellar age. The direct propagation of star formation in \\objectname{NGC 4449} places this galaxy as one of a handful --besides our \\objectname{Galaxy}, \\objectname{M82} \\citep{satyapal97} and a couple of \\objectname{LMC} OB associations and nebulae \\citep{parker92,laval92,rosado96}-- where sequential star formation has been discerned and measured \\citep{hill98}.      The morphological and dynamical peculiarities in \\objectname{NGC 4449} are also reflected in its \\ion{H}{1} and ionized gas contents. The neutral gas in the outer parts of \\objectname{NGC 4449} \\citep{vanwoerden75, hunter98} seems to be counter rotating relative to the ionized gas distributed in the inner disk \\citep{sabbadin84,valdez00}. This counter rotation implies the co--existence of two physical systems rotating in opposite directions in this galaxy. In Irrs, this behavior has only be seen in a few examples: \\objectname{IC 10} \\citep{shostak89} and \\objectname{NGC 4449} \\citep{sabbadin84}. As does the LMC, \\objectname{NGC 4449} has a significant magnetic field, a fact which has only recently emerged \\citep{otmianowska00,chyzy00}. One of the conclusions reached in these studies is that the existence of a global magnetic field is possible even for small values of galactic rotation (v$_{rot}$ $<$ 30 km s$^{-1}$).    The relatively small distance of \\objectname{NGC 4449} earmarks this specimen for detailed kinematical studies. The first Fabry--Perot (FP) interferometric analysis of this galaxy was performed photographically with a fixed gap FP etalon by \\citet{crillon69}, who determined radial velocities for the more prominent complexes of \\ion{H}{2} regions in a spatial field of 5$\\times$5 arcmin. Other researchers who employed scanning FPs to study the kinematics of the gas in \\objectname{NGC 4449} are \\citet{malumuth86}, \\citet{arsenault88} and \\citet{fuentes00}. The observations in \\citet{malumuth86} were spatially limited (3 arcmin square area) and kinematical interpretations were aggravated due to a loss of parallelism in the etalon plates. \\citet{arsenault88} and \\citet{fuentes00} focused their work on the study of supersonic velocity dispersions in the brightest \\ion{H}{2} regions and those \\ion{H}{2} regions belonging exclusively to the bar. Finally, conventional kinematic studies by means of classic, long slit spectroscopy for \\objectname{NGC 4449} abound in the literature; many projects have been devoted to the most prominent \\ion{H}{2} regions as well as to the filament and superbubble populations \\citep[and references therein]{sabbadin84,hartmann86,hunter97,martin98}.    In the present analysis, we revisit the analysis of the kinematics of the ionized gas in \\objectname {NGC 4449} as a part of a long term project focused on the study of the interrelationship between gas and stars in irregular galaxies \\citep[and references therein]{valdez01, rosado01, borissova00}. This study presents, for the first time, a tridimensional kinematical analysis of the entire, optical distribution of the ionized gas in \\objectname{NGC 4449}. Earlier studies had been undertaken by means of longslit spectroscopy and FP interferometry, but focused only on localised objects (such as ionised hydrogen regions) or otherwise, in a smaller field of view compared to the present work.   In Section 2 we describe the FP observations and data reduction. The global, optical morphology in four optical lines are presented in Section 3, as is a statistical analysis of the \\ion{H}{2} region population. The kinematics of local (\\ion{H}{2} regions) and global scale features (diffuse ionized gas, global velocity field) are discussed in Section 4. Section 5 is devoted to dynamical results -- as deduced from our optical rotation curve. In Section 6 we compare our kinematical results with published \\ion{H}{1} ones. Finally, in Section 7, we present our conclusions. ",
        "conclusions": "We have successfully performed a detailed kinematical and dynamical analysis of the entire ionized gas content in the nearby irregular galaxy \\objectname{NGC 4449}, by means of Fabry--Perot interferometry. The analysis has been accomplished separately on both global as well as local scales and several of our conclusions are new.   On local scales, we focused our attention on the \\ion{H}{2} region population, extracting radial velocities and velocity dispersions. In this way, we present, for the first time, the most extensive kinematical catalog of the \\ion{H}{2} regions in this galaxy yet published. We find that the \\ion{H}{2} region population shows similar integrated properties when compared to kinematical data previously obtained in the relatively isolated irregular \\objectname{IC 1613} \\citep{valdez01}. The latter statement also applies to the velocity dispersions in the star forming regions: in both galaxies, they are approximately identical. Moreover, in many cases we did detected velocity profiles indicative of a complex structure not fitted by {\\it single} gaussian curves. These profiles are undoubtedly related to the high perturbed radial velocity field, which shows traces of a plethora of superbubbles, filaments and a very active (and widespread) diffuse ionized gas spanning the entire optical disk. Statistically, using the \\ion{H}{2} region diameter distribution and the \\ion{H}{2} region luminosity function, we find that the star forming complexes are typical of those obtained for larger samples of irregulars.  We furthermore analysed the kinematical properties of the diffuse ionized gas (DIG) on global scales by means of radial velocity and velocity dispersion fields; locally, we explored the kinematics, making use of profile decomposition. As far as we are aware, this is the first time that such an analysis has been performed in this object. Our results point towards a highly chaotic medium, especially at those locations close to the brightest \\ion{H}{2} complexes in the bar. We postulate that the chaotic radial velocity field induces local compression ensuring the propagation of stochastic star formation --as also discussed by \\citet{hartmann86}.   On global scales, we find a relatively well behaved radial velocity field, showing a decreasing gradient in radial velocity along the optical bar from NE to SW, suggesting solid body rotation. This result agrees well with previous kinematical results \\citep[and references therein]{valdez00}. We believe that \\objectname{NGC 4449} presents solid body rotation (albeit at low rotation velocities) north of the optical center. Reliable fits are not possible in the southern sector of the disk, due to the chaotic motions at those locales. The inner \\ion{H}{1} kinematics (inner 13$\\farcm$4) also shows a good correspondence with the optical radial velocities we find in this study.   In general, our kinematical and dynamical findings for the ionised gas in \\objectname{NGC 4449} are fully consistent within the framework of a past interaction with \\objectname{DDO 125}."
    },
    "0208/astro-ph0208562_arXiv.txt": {
        "abstract": "We carry out a comprehensive study of the dynamics of large-scale perturbations in quintessence scenarios. We model the contents of the Universe by a perfect fluid with equation of state $\\wf$ and a scalar field $Q$ with potential $V(Q)$. We are able to reduce the perturbation equations to a system of four first-order equations. During each of the five main regimes of quintessence field behaviour, these equations have constant coefficients, enabling analytic solution of the perturbation evolution by eigenvector decomposition. We determine these solutions and discuss their main properties. ",
        "introduction": "Recent observations seem to indicate that the Universe is undergoing a period of accelerated expansion~\\cite{acc}. Whereas cosmologists initially introduced a cosmological constant in order to explain this, a range of different models have emerged, amongst which quintessence has been particularly prominent in the literature~\\cite{RP,qui}. It is defined as a scalar field rolling down its potential and presently dominating the dynamics of the Universe. An important class of quintessence models are known as tracking models~\\cite{RP,qui}, where the late-time evolution of the field has an attractor behaviour rendering its evolution fairly independent of initial conditions. In contrast to a cosmological constant, which is by definition perfectly homogeneous, the quintessence field can, and indeed must, have perturbations.  The evolution of perturbations in quintessence models have been studied by many authors \\cite{RP,quintperts,BMR,perts2}.  In this paper we carry out an exhaustive and elegant analysis of those in the large-scale approximation.  We model the contents of the Universe by a perfect fluid with equation of state $\\wf$ and a scalar field $Q$ with potential $V(Q)$.  We assume a flat Universe throughout. ",
        "conclusions": "We have derived four first-order equations to describe large-scale perturbations in quintessence scenarios. During each of the five main regimes of quintessence behaviour, these equations have constant coefficients, enabling analytic solution of the perturbations by eigenvector decomposition.  We have seen that during the kinetic regime there is a growing isocurvature mode which then remains constant until tracking begins. However, if the quintessence field undergoes a long period of tracking, there remain only adiabatic perturbations which, in case of a sub-dominant tracker, are independent of its initial conditions. A low initial quintessence energy density~\\cite{ML}, or a long kinetic period, may prevent the non-adiabatic modes disappearing completely.  It is possible in principle to carry out the same analysis without using the large-scale approximation, although the equations may then be too complicated to be useful."
    },
    "0208/astro-ph0208081_arXiv.txt": {
        "abstract": "Water masers around an AGB star, IRC+60169, were observed at four epochs using the Japanese VLBI networks. The distribution of the maser features is limited in a thick-shell region, which has inner and outer expansion velocities of 7 km s$^{-1}$ and 14 km s$^{-1}$ at radii of 25 mas and 120 mas, respectively. The distribution of the red-shifted features exhibits a ring-like structure, the diameter of which is 30 mas, and corresponds to the inner radius of the maser shell. This implies that dense gas around the star obscures red-shifted emission. Although a position--radial velocity diagram for the maser features is consistent with a spherical shell model, the relative proper motions do not indicate an expansion  motion of the shell. A remarkable property has been found that is a possible periodic change of the alignment pattern of water maser spots. ",
        "introduction": "A low-mass star passes the Asymptotic Giant Branch (AGB) phase in the HR diagram, in which it undergoes heavy mass-loss processes. After the AGB phase, the star forms a planetary nebula (PN) by ionizing its circumstellar envelope (e.g., Kwok 1993).  It has been stated that many young PNe exhibit a bipolar morphology or large deviation from spherical symmetry (e.g., Aaquist, Kwok 1991; Sahai et al. 1998). There still exists a missing link in the mass-loss history between AGB stars and central stars of PNe.  Most AGB stars show maser emission from water molecules (Reid, Moran 1981; Elitzer 1992). Masers have been used for probes of the mass-loss process in the circumstellar envelope of the AGB star. VLBI (Very Long Baseline Interferometry) observations have revealed that relatively young AGB stars (i.e., Mira variables and semi-regular variables) already exhibit bipolarity in their spherically expanding flows (e.g., Marvel 1997; Ishitsuka et al. 2001). On the other hand, for more evolved AGB stars (i.e., IRC/AFGL objects and OH/IR stars) and proto-PNe,  little is known about the detailed mass-loss process by using the maser kinematics (cf., Marvel, Boboltz 1999; Imai et al. 2002).  IRC+60169 has been classified as an IRC/AFGL object using the IRAS two-color diagram (van der Veen, Habing 1988; Takaba et al. 1994). It shows a clear double-peaked spectrum of water masers, which is a typical signature of IRC/AFGL objects and OH/IR stars. The shape of the spectrum exhibited a violent time variation. In 1977 the red-shifted component was stronger than the blue-shifted one, and in 1982 an intensity change between these components occurred (Engels et al. 1988). This fact is thought to be related to the temporary variation in the mass-loss flow.  We present the water-maser observations of IRC+60169 using Japanese VLBI networks in order to reveal the origin of the complicated time-variation in the intensity and distribution of the water masers and to investigate the kinematics of the circumstellar envelope by measuring the proper motions of the masers. ",
        "conclusions": "In the present work,  multi-epoch VLBI observations of the water masers in IRC+60169 were carried out. The main results from our data are described as follows: 1) The blocking effect is suggested by the ring-shaped distribution of the red-shifted maser features. The blocking size was estimated to be $\\sim 20$ AU, which almost equals to the inner edge of the maser shell. This phenomena is believed to be seen only in evolved AGB stars. 2) The maser-shell size ($\\sim$80 AU) suggests that IRC+60169 is more evolved than Mira variables. Although random motions are likely to dominate the kinematics of the envelope, radial acceleration also occurs in the maser shell. 3) We found that the alignment pattern of the maser spots seems to change periodically with the stellar pulsation. This implies that the spots pattern is stationary during a year, and that the environment in which masers occur moves to the radial direction. 4) Using the statistical parallax method, we obtained a weighted-mean distance to IRC+60169 of 310$\\pm$40 pc. This value is probably underestimated because of the blocking effect and/or the intrinsic asymmetrical velocity field.  Since we have measured only eleven proper motions, the kinematics of the circumstellar envelope in IRC+60169 is still unclear. Further systematic monitoring observations every few months will reveal the kinematics of the envelope, the distance, and the environment of the maser emitting region. In addition,  SiO maser observations of IRC+60169 will be useful in revealing properties of the blocking gas, because it is considered to exist in it.  \\bigskip We would like to express our gratitude to each member of the Japanese VLBI networks for their support during observations, correlations, and data reduction. We also wish to thank an anonymous referee for useful comments. HI and HS were supported by the Grant-in-Aid for JSPS Fellows by Ministry of Education, Culture, Sports, Science  and Technology."
    },
    "0208/gr-qc0208075_arXiv.txt": {
        "abstract": "We provide an analytic method to discriminate among different types of black holes on the ground of their strong field gravitational lensing properties. We expand the deflection angle of the photon in the neighbourhood of complete capture, defining a strong field limit, in opposition to the standard weak field limit. This expansion is worked out for a completely generic spherically symmetric spacetime, without any reference to the field equations and just assuming that the light ray follows the geodesics equation. We prove that the deflection angle always diverges logarithmically when the minimum impact parameter is reached. We apply this general formalism to Schwarzschild, Reissner-Nordstrom and Janis-Newman-Winicour black holes. We then compare the coefficients characterizing these metrics and find that different collapsed objects are characterized by different strong field limits. The strong field limit coefficients are directly connected to the observables, such as the position and the magnification of the relativistic images. As a concrete example, we consider the black hole at the centre of our galaxy and estimate the optical resolution needed to investigate its strong field behaviour through its relativistic images. ",
        "introduction": "Gravitational lensing is one of the first applications of General Relativity ever studied \\cite{Einstein}. Firstly it was recognized in the light deflection by sun, secondly in lensing of quasars by foreground galaxies, then in the formation of giant arcs in galaxy clusters and finally in galactic microlensing. Now it is an ordinary phenomenon in the panorama of astronomical observations (see \\cite{SEF} for a complete treatment and references therein).  The full theory of gravitational lensing has been developed following the scheme of the weak field approximation and, in this formulation, it has been successfully employed to explain all the physical observations.  In the last years, however, the scientific community is starting to look at this phenomenon from the opposite point of view, opening a strong field perspective. Viergutz \\cite{Vie} made a semi-analytical investigation about geodesics in Kerr geometry; in Ref. \\cite{Bar} the appearance of a black hole in front of a uniform background was studied; Falcke, Melia and Agol \\cite{FMA} considered the emission of the accretion flow as source. Virbhadra \\& Ellis \\cite{VirEll} showed that a source behind a Schwarzschild black hole would produce one set of infinite relativistic images on each side of the black hole. These images are produced when a light ray with small impact parameter winds one or several times around the black hole before emerging. Later on, by an alternative formulation of the problem, Frittelli, Kling \\& Newman \\cite{FKN} attained an exact lens equation, giving integral expressions for its solutions, and compared their results to those by Virbhadra \\& Ellis. The same problem has been investigated by Bozza et al. in Ref. \\cite{BCIS}, where a strong field limit was first defined in Schwarzschild black hole lensing and used to find the position and the characteristics of all the relativistic images analytically. Eiroa, Romero and Torres \\cite{ERT} applied the same technique to a Reissner - Nordstrom black hole. Recently, in another work \\cite{VirEll2}, Virbhadra \\& Ellis distinguished the main features of gravitational lensing by normal black holes and by naked singularities, analyzing the Janis, Newman, Winicour metric. They remarked the importance of these studies in providing a test for the cosmic censorship hypothesis.  The reason for such an interest in gravitational lensing in strong fields is that by the properties of the relativistic images it may be possible to investigate the regions immediately outside of the event horizon. High resolution imaging of black holes by VLBI \\cite{VLBI} could be able to detect the relativistic images and retrieve information about strong fields stored within these new observables. Moreover, since alternative theories of gravitation must agree with GR in the weak field limit, in order to show deviations from GR it is necessary to probe strong fields in some way. Indeed, deviation of light rays in strong fields is one of the most promising grounds where a theory of gravitation can be tested in its full form.  Of course, the study of null geodesics in strong fields is not easy and up to now it has always been carried out using numerical techniques. An analytical treatment would enlighten the dependence of the observables on the parameters of the system, allow easy checks about the detectability of the images and open the way to comparisons between the results in different metrics. In Ref. \\cite{BCIS}, a new way to expand the deflection angle in the Schwarzschild metric was suggested. The deflection angle near its divergence was approximated by its leading order and its first regular term and then plugged into the lens equation. In this way, very simple and reliable analytical formulae were derived for the relativistic images and their main features.  So, as the weak field limit takes the first order deviation from Minkowski, the strong field limit starts from complete capture of the photon and takes the leading order in the divergence of the deflection angle.  The strong field limit of Ref. \\cite{BCIS} was only developed in Schwarzschild spacetime. In this paper, we provide a general method to extend the strong field limit to a generic static spherically symmetric spacetime. Our method is universal and can be applied to any spacetime in any theory of gravitation, provided that photons satisfy the standard geodesics equation. The parameters of the strong field limit expansion are directly connected with the observables, providing an effective tool to discriminate among different metrics. In Sect. 2, we state the problem and carry out the strong field limit of the deflection angle. In Sect. 3, we apply the method to some simple metrics: Schwarzschild, Reissner-Nordstrom and Janis-Newman-Winicour black hole, discussing their differences with reference to the gravitational lensing phenomenology. In Sect. 4, we establish a connection between the strong field limit coefficients and the relativistic images, analyzing the case of the black hole at the center of our galaxy as a concrete example where our results can be tested. Finally, Sect. 5 contains the summary. ",
        "conclusions": "Gravitational lensing is undoubtedly  a potential powerful tool for the investigation of strong fields. By general arguments we have shown that the deflection angle diverges logarithmically as we approach the photon sphere. We have drawn a general method to compute the coefficient of the leading order divergent term and the first regular term. When the latter cannot be calculated analytically, we have seen that it can be well approximated by a simple series expansion starting from Schwarzschild spacetime.  We have applied our method to Schwarzschild, Reissner-Nordstrom and Janis-Newman-Winicour black hole, explicitly calculating and plotting the strong field limit coefficients.  Of course, it is possible to apply the strong field limit, in the form given in this paper, to any spherically symmetric metric representing a black hole. In this way, it is possible to compare the gravitational lensing behaviour of these objects in different theories of gravitation. In principle, the extension to non-spherically symmetric and rotating black holes is possible. However, the dependence of the deflection angle on more than one variable can put severe obstacles in the way of analytic solutions of the problem. Nevertheless, this is indeed another important point which needs to be investigated to complete the picture of black hole lensing.  Differences in the deflection angle are immediately reflected on the relativistic images. If the mass and the distance of the lens is known, then the detection of any set of relativistic images would immediately check the Schwarzschild geometry. VLBI should be able to provide an observational answer, if the relativistic images are not hidden behind environmental noise.  Furthermore, if the outermost image is resolved from the others, it is then possible to fully reconstruct the strong field limit coefficients and select a precise black hole model. Our present observational facilities are not so far from the required resolutions, which, for the galactic black hole, are of the order of 0.01 microarcsecs. As a long term project the detection of the outermost image stands as a very interesting, non-trivial challenge for future technology.  Strong field limit represents an important step in the construction of a robust theoretical scheme connecting the gravitational lensing with the strong field properties. By a simple and reliable expansion, it clarifies the whole phenomenology and the differences between various models that should be expected in the appearance of relativistic images. If these so elusive features will be detected, we will finally have a way to effectively discriminate between alternative theories of gravitation and grow in our knowledge of spacetime.   \\bigskip"
    },
    "0208/astro-ph0208032_arXiv.txt": {
        "abstract": "We estimate the growth of matter perturbations in a class of recently proposed dark-energy models based on the (loop-corrected) gravi-dilaton string effective action, and characterized by a global attractor epoch in which dark-matter and dark-energy density scale with the same effective equation of state. Unlike most dark-energy models, we find that the accelerated phase might start even at redshifts as high as \\( z\\approx 5 \\) (thus relaxing the coincidence problem), while still producing at present an acceptable level of matter fluctuations. We also show that such an early acceleration is not in conflict with the recently discovered supernova SN1997ff at \\( z\\approx 1.7 \\). The comparison of the predicted value of \\( \\sigma _{8} \\) with the observational data provides interesting constraints on the fundamental parameters of the given model of dilaton-dark matter interactions. ",
        "introduction": "}  {\\small \\label{Sec1}}{\\small \\par}  {\\small Independent cosmological observations have recently pointed out the existence of a significant fraction of critical density in the form of unclustered matter, possibly characterized by a negative pressure \\cite{rie,perl,lee,net}. Such a component of the cosmic fluid, conventionally denoted dark energy, is believed to drive the accelerated evolution of our present Universe, as suggested by the study of the Supernovae Ia (SNIa) Hubble diagram. While the simplest explanation of the dark-energy component is probably a cosmological constant, it might be neither the most motivated, nor the best fit to the data. }{\\small \\par}  {\\small The existence of a (presently dominating) dark-energy component raises at least three important questions:} \\emph{\\small a}{\\small ) what is the physical origin of such a new component?} \\emph{\\small b}{\\small ) why its energy density today is just comparable to the matter energy density?} \\emph{\\small c}{\\small ) did the acceleration start only very recently in the cosmological history? }{\\small \\par}  {\\small The first question refers to the fundamental physical mechanism able to generate a cosmic and universal dark-energy distribution. So far, most models of dark energy have adopted a purely phenomenological approach, even if a few proposal concerning the possible role of the dark-energy field in the context of fundamental physics have already appeared \\cite{fri,wet90,albre,MG01,GPV01,pie}. }{\\small \\par}  {\\small The other questions concern two distinct (although possibly related) aspects of the so-called {}``coincidence problem{}'' \\cite{Ste}. The first aspect refers to the density of the dark-energy and of the dark-matter component. Since in most models the two components have different equations of state, they scale with time in a different way, and their densities should be widely different for essentially all the cosmological history: by contrast, current data are strongly pointing at a comparable proportion of the two components just at the present epoch. }{\\small \\par}  {\\small The second aspect of the coincidence problem may arise if the epoch of accelerated evolution started only very recently, say at \\( z\\approx 1 \\). There are reasons to suspect that this may be the case because, in most models, the growth of structures is forbidden after the beginning of the accelerated expansion, so that the acceleration cannot be extended too far into the past. It has also been recently claimed that the supernova SN1997ff \\cite{super,super1} at \\( z\\approx 1.7 \\) provides an additional indication that the acceleration is a relatively recent phenomenon. }{\\small \\par}  {\\small A promising approach for a possible simultaneous answer to the above questions has been recently provided by a dilatonic interpretation of the dark energy based on the infinite bare-coupling limit of the superstring effective action \\cite{GPV01}, whose cosmological solutions are characterized by a late-time global attractor where dark-matter and (dilatonic) dark-energy densities have an identical scaling with time. A very similar cosmological scenario has been studied also in \\cite{amtoc,bias}. The respective amount of dark-matter and dark-energy density is eventually determined by the fundamental constants of the model, and it is expected to be of the same order, so that their ratio keeps frozen around unity for the whole duration of such an asymptotic regime. In such a context, the today approximate equality of dark-matter and dark-energy density would be no longer a coincidence of the present epoch, but a consequence of the fact that our Universe has already entered the asymptotic regime (actually, for a region of parameter space allowed by observations, it is also possible that the ratio of dark-energy to dark-matter density is close to unity not only in the asymptotic regime, but already after the equivalence epoch). Finally, the cosmic field responsible for the observed large-scale acceleration would be no longer introduced} \\emph{\\small ad hoc}{\\small , being identified with a fundamental ingredient of superstring/M-theory models of high-energy physics. }{\\small \\par}  {\\small In this paper we will focus our discussion on the possible constraints imposed by structure formation on the above model of dilatonic dark energy, by studying the growth of matter perturbations in the two relevant post-equivalence epochs: a first decelerated epoch, dubbed} \\emph{\\small dragging phase}{\\small , and a second accelerated epoch, dubbed} \\emph{\\small freezing phase}{\\small . We will then reconstruct the behavior of the cosmological gravitational potential, evaluate the Sachs-Wolfe and integrated Sachs-Wolfe contributions to the spectrum of CMB anisotropies, and estimate the present level of matter fluctuations (in particular, the so-called variance \\( \\sigma _{8} \\), smoothed out over spheres of radius 8 Mpc \\( h^{-1} \\)). Finally, we will compare it with observations. }{\\small \\par}  {\\small The results that we find are interesting, and might help answering the third question posed at the beginning of this section: among the range of parameters compatible with a phenomenologically acceptable \\( \\sigma _{8} \\) we find indeed values allowing a long epoch of acceleration, starting as far in the past as at \\( z\\approx 5 \\). By contrast, the models of dark energy uncoupled to dark matter, with frozen (or slowly varying) equation of state, cannot accelerate before \\( z\\approx 1 \\) . It seems appropriate to anticipate here that the production of an acceptable level of fluctuations even in the case of an early start of the accelerated epoch is due to two concurrent factors: the first is that, during the freezing phase, perturbations do not stop growing like in other models of accelerating dark energy; the second is that the horizon at equivalence in our model shifts at larger scales with respect to a standard \\( \\Lambda  \\)CDM model. It is also important to stress that such an early acceleration is by no means in contrast with the recently observed supernova SN1997ff at \\( z\\approx 1.7 \\). }{\\small \\par}  {\\small For the model of dilatonic dark energy considered in this paper the coincidence problem can thus be alleviated by the fact that the ratio of dark-matter to dark-energy density is of order one not only at present, not only in the course of the future evolution but (at least in principle) also in the past, long before the present epoch. It should be clear, however, that this possibility could be strongly constrained by future observations of supernovae at high redshift. }{\\small \\par}  {\\small The paper is organized as follows. In Sect. \\ref{Sec2} we present the details of our late-time, dilaton-driven cosmological scenario, and define the (theoretical and phenomenological) parameters relevant to our computation. In Sect. \\ref{Sec3} we discuss the growth of matter perturbations and compute the variance \\( \\sg _{8} \\). In Sect. \\ref{Sec4} we impose the observational constraints on our set of parameters, and determine the maximal possible extension towards the past of the phase of accelerated evolution. We also find, as a byproduct of our analysis, interesting experimental constraints on the fundamental parameters of the string effective action used for a dilatonic interpretation of the dark-energy field. In Sect. \\ref{Sec4bis} we compare our model with the constraints provided by the farthest observed supernova at \\( z\\approx 1.7 \\). Our conclusions are finally summarized in Sect. \\ref{Sec5}. }{\\small \\par} ",
        "conclusions": "}  {\\small \\label{Sec5}}{\\small \\par}  {\\small In this paper we have considered a phenomenological model of dark-energy-dark-matter interactions based on the infinite bare-coupling limit of the superstring effective action. The dilaton, rolling down an exponentially suppressed potential, plays the role of the cosmic field responsible for the observed acceleration, and drives the Universe towards a final configuration dominated by a comparable amount of kinetic, potential and CDM energy density. }{\\small \\par}  {\\small The effective dilatonic coupling to dark matter switches on at late enough times (i.e., large enough bare coupling), and affects in a significant way the post-equivalence cosmological evolution. The time-dilution of the dark-matter density, in particular, is first slightly enhanced (during the dragging phase) and then considerably damped (during the freezing phase) with respect to the standard \\( a^{-3} \\) decay law. The large-angle fluctuation scales relevant to the observed CMB anisotropies reenter the horizon during the dragging epoch, and exit the horizon again during the freezing epoch. In spite of this unconventional evolution, the growth of the matter-density perturbations may be large enough to match consistently present observations. }{\\small \\par}  {\\small The predicted value of the (smoothed out) density contrast \\( \\sg _{8} \\), compared with data obtained from cluster abundance, defines a significant allowed region in the parameter space of the given class of dilatonic dark-energy models. The analysis of such an allowed region provides two main results. }{\\small \\par}  {\\small The first is that the bounds on the past-time extension of the accelerated (freezing) epoch are significantly weaker than in conventional dark-energy models (uncoupled to dark matter, with frozen equation of state). The establishment of the freezing regime, in our class of dilatonic models, is allowed long before the present epoch (up to \\( z\\simeq 5 \\)), thus providing (in principle) a further relaxation of the coincidence problem, by extending the present cosmological configuration not only in the far future, but also towards the past. }{\\small \\par}  {\\small The possibility of very early (\\( z>1 \\)) accelerated evolution is indeed a typical signature of such a class of dilatonic models, useful in principle to discriminate it from other (uncoupled) dark energy models, hopefully on the grounds of future observational data. It is important to stress, to this respect, that the farthest type Ia supernova so far observed is at \\( z\\simeq 1.7 \\), and is perfectly compatible with an accelerated Universe already at that epoch, provided the data of the magnitude-redshift diagram are consistently fitted by the accelerated kinematics of dilatonic models. }{\\small \\par}  {\\small The second results concerns the parameters of the (non-perturbative) dilaton potential appearing in the strong-bare-coupling regime of the string effective action. The dilaton mass scale \\( V_{0} \\), for an efficient and realistic dark energy scenario, appears in such a context to be tightly anchored to a value very near to the present Hubble curvature scale. A small deviation of \\( V_{0} \\) from the required value is enough to remove the predictions of the dilatonic model from the region of parameter space allowed by the \\( \\sg _{8} \\) data. }{\\small \\par}  {\\small This means that, under the assumption that the dilatonic models discussed in this paper provide the correct explanation of the observed cosmic acceleration, the measurements of the density contrast \\( \\sg _{8} \\), besides their obvious astrophysical importance, would also acquire an interesting high-energy significance for providing an indirect (parameter-dependent) measurements of the dilaton mass scale. }{\\small \\par}  {\\small We note, finally, that astrophysical observations may provide several additional constraints on dilatonic dark-energy models. For instance, the clustering evolution of sources at high redshifts may constrain directly the freezing growth exponent \\( m_{2} \\); as already mentioned, the baryon bias that develops during freezing is also observable, at least in principle \\cite{bias}; finally, further constraints can be derived from a computation of the full multipole spectrum of the CMB radiation (see e.g. \\cite{aqtp} for a recent study of the CMB constraints on coupled dark-energy models with power-law potentials). Preliminary results seem to confirm the conclusions of this paper, but a detailed discussion of these new constraints is postponed to a future work. }{\\small \\par}  {\\small"
    },
    "0208/astro-ph0208518_arXiv.txt": {
        "abstract": "Observations of the red giant Arcturus ($\\bf \\alpha$ Boo) obtained with the star tracker on the Wide Field Infrared Explorer ({\\em WIRE}) satellite during a baseline of 19 successive days in 2000 July-August are analysed. The power spectrum has a significant excess of power at low-frequencies. The highest peak is at $\\sim$4.1 $\\mu$Hz, which is in agreement with the ground-based radial velocity and photometry study of Belmonte et al. (1990a; 1990b). The variability of Arcturus can be explained by sound waves, but it is not clear whether these are coherent p-mode oscillations. ",
        "introduction": "After the failure of the main mission of the Wide Field Infrared Explorer ({\\em WIRE}) satellite, launched by NASA in 1999 March, its star tracker was used for asteroseismology of bright objects (Buzasi et al. 2000; Buzasi 2002). Here we report on preliminary analysis of observations of the red giant Arcturus ($\\alpha$ Boo), which is the brightest star in the northern hemisphere, and has been observed in velocity several times (Belmonte et al. 1990a; 1990b; Merline 1995).  Nearly 1.2 million 0.1-s exposures were obtained during  19 successive days in 2000 July-August. Fig.~1 presents the light curve. The data from each orbit were binned into a single mean point with a typical error of about 15 ppm. The light curve shows changes on time scales of hours to days.  \\begin{figure} \\vspace{9cm}  %  \\special{psfile=fig1.eps angle=0 vscale=90 hscale=70 voffset=-70 hoffset=-40}  \\caption{The light curve of $\\alpha$ Boo during the 19-d WIRE run in 2002. Each point represents a mean of about 4600 0.1-s exposures obtained during one 96-min orbit. The errors on a single point are smaller than its dimensions in the graph.} \\end{figure}  The upper panel of Fig.~2 displays the amplitude spectrum of the data. There is an excess of power at low frequencies. The highest peak at 4.11 $\\mu$Hz is in agreement with the ground-based radial velocity and photometry study of Belmonte et al. (1990a; 1990b).  \\begin{figure} \\vspace{9cm}  %  \\special{psfile=fig2.eps angle=0 vscale=65 hscale=70 voffset=20 hoffset=-40}  \\caption{Upper Panel -- Amplitude spectrum of the {\\em WIRE} data. There is a significant excess of power at low frequencies. The four highest peaks are at 4.11, 2.26, 3.00 and 0.98 $\\mu$Hz respectively.  Second panel -- The spectral window (after planting a 4.11-$\\mu$Hz sinusoid in the window function with the same amplitude as in the data). Third panel -- The power spectrum after prewhitening the four highest peaks. Lower panel -- after prewhitening the eight highest peaks.} \\end{figure}  Table~1 lists the ten highest peaks in the power spectrum. The typical frequency error is $\\sim$0.1 $\\mu$Hz. Peak 4 (0.98 $\\mu$Hz) is questionable as the data extend over only $\\sim$2.5 cycles of it. Fig.~3 presents the amplitude spectrum at the frequency range of 0-20 $\\mu$Hz. The dotted vertical lines show the locations of the ten highest peaks.  \\begin{table*} \\caption{The ten highest frequencies in the amplitude spectrum}  \\begin{tabular}{rrccc} &           &                &              \\\\ Peak & Frequency & Semi-amplitude & comments     \\\\ & ($\\mu$Hz) & (ppm)          &              \\\\ &           &                &              \\\\ 1    & 4.11      & 980            &              \\\\ 2    & 2.26      & 800            &              \\\\ 3    & 3.00      & 760            &              \\\\ 4    & 0.98      & 620            & questionable \\\\ 5    & 5.73      & 520            &              \\\\ 6    & 8.04      & 420            &              \\\\ 7    & 4.92      & 380            &              \\\\ 8    & 10.47     & 270            &              \\\\ 9    & 7.35      & 270            &              \\\\ 10   & 16.89     & 240            &              \\\\ \\end{tabular} \\end{table*} ",
        "conclusions": "The amplitude and frequency of the power excess in Arcturus are consistent with solar-like oscillations (Kjeldsen \\& Bedding 1995).  The strongest peaks have a frequency spacing of $\\Delta\\nu\\simeq$~0.8$\\,\\mu$Hz, which is in excellent agreement with the expected value of 0.9$\\pm$0.1 (Kjeldsen \\& Bedding 1995), but not with the value of 5.0 $\\mu$Hz reported by Belmonte et al. (1990a; 1990b).  We note, however, that our value of $\\Delta\\nu$ is very close to our frequency resolution ($\\sim$0.55 $\\mu$Hz), which is set by the length of the run. The same was true for the observations by Belmonte et al.  Could granulation noise, rather than oscillations, be responsible for the excess power?  This is unlikely, since the variability is also detected in velocity, with an amplitude of $\\sim$60\\,m/s at 4.3\\,$\\mu$Hz (Belmonte et al. 1990a; 1990b).  The ratio of the photometric amplitude to the velocity amplitude is as expected for sound waves, and about ten times greater than expected for granulation noise.  We conclude that the variations in Arcturus can be explained by sound waves, but it is not clear whether these are coherent p-mode oscillations or perhaps a single mode with a short lifetime.  A longer time series is required to distinguish between the two options.  \\begin{figure} \\vspace{9cm}  %  \\special{psfile=fig3.eps angle=0 vscale=60 hscale=70 voffset=-20 hoffset=-40}  \\caption{Amplitude spectrum of the {\\em WIRE} data in the range 0-20 $\\mu$Hz. The locations of the ten highest peaks are marked by vertical dotted lines.} \\end{figure}"
    },
    "0208/astro-ph0208204_arXiv.txt": {
        "abstract": "We have made deep {\\it Chandra} observations of three powerful FRII radio sources: two quasars (3C\\,263 and 3C\\,351) and one radio galaxy (3C\\,330). X-ray emission from hotspots and lobes, as well as from the active nucleus, is detected in each source.  We model the hotspots' synchrotron spectra using VLA, BIMA and {\\it Hubble Space Telescope} data. In 3C\\,263 and 3C\\,330, the hotspots' X-ray emission is at a level consistent with being synchrotron self-Compton (SSC) emission, with a hotspot magnetic field close to the equipartition value. In the two hotspots of 3C\\,351, however, an SSC origin for the X-rays would require the magnetic field strength to be an order of magnitude below the equipartition value in our models: in addition, there are offsets between the radio, optical and X-ray emission from the secondary hotspot which are hard to explain in a simple SSC model. We discuss the emission mechanisms that may be responsible for these observations.  On our preferred model, the X-ray emission from the radio lobes of the three sources is due to inverse-Compton scattering of the microwave background radiation. If this is the case, the magnetic field strengths in the lobes are typically about a factor 2 below the equipartition values, assuming uniform lobe electron and magnetic field distributions.  We detect extended X-ray emission, which we attribute to a cluster/group environment, around 3C\\,263 and 3C\\,330. This detection allows us to show that the lobes are close to pressure balance with their surroundings, as long as no non-radiating particles contribute to the internal pressure of the lobes. ",
        "introduction": "\\label{intro}  {\\it Chandra} has now detected a large number of X-ray features related to the jets and hotspots of extragalactic radio sources (see \\citealt{hk02} for a recent review). The jet detections have attracted a great deal of interest, but much of the fundamental physics behind them remains unclear. The X-ray jets commonly seen in FRI sources \\citep*{wbh01a} probe the high-energy particle acceleration in these objects \\citep*{hbw01b}, while the few FRII jets that have been detected \\citep[e.g.,][]{smlp00} may be evidence for extremely high bulk speeds on kiloparsec scales \\citep[e.g.,][]{tmsu00}. But the details of the processes responsible for producing the X-rays in both classes of source are still debatable, and the issue is confused by uncertainties about the Doppler boosting factors of the X-ray and radio-emitting material.  In hotspots the situation is more clear-cut. These structures, which are believed to trace a strong terminal shock or shocks at the end of the jet in powerful FRII sources, are generally thought to be unlikely to be moving with respect to the observer at highly relativistic speeds \\citep{bblr95, s95, al00}, although moderately relativistic motions in and beyond the hotspot (with $v \\sim 0.3c$) may be required to explain some observations \\citep{dbls97}. Their radio-to-optical spectra are often well constrained, giving limits on the amount of synchrotron emission expected in the X-ray, and their sizes and structures are often well known. This means that it is possible to make a simple prediction of the X-ray flux densities expected from inverse-Compton emission. The dominant emission process for bright, compact hotspots which show strong spectral steepening or a synchrotron cutoff at high radio/optical frequencies is likely to be synchrotron-self-Compton emission (SSC), in which the radio photons from the synchrotron process are inverse-Compton scattered into the X-ray band by the synchrotron-emitting electrons. A prediction of the SSC flux density can be made if the magnetic field strength in the hotspot, and therefore the electron number density, can be estimated. Conversely, a measured X-ray flux density which is inferred to be SSC can be used to estimate the magnetic field strength in the hotspot.  \\begin{deluxetable}{lllll} \\tablewidth{13cm} \\tablecaption{SSC and unknown high-energy FRII hotspot detections} \\tablehead{\\colhead{Mechanism} & \\colhead{Source} & \\colhead{Source type} & \\colhead{Instrument}&\\colhead{Reference}} \\startdata SSC near&Cygnus A&NLRG&{\\it ROSAT}&1\\\\ equipartition&&&{\\it Chandra}&2\\\\ &3C\\,295&NLRG&{\\it Chandra}&3\\\\ &3C\\,123&NLRG&{\\it Chandra}&4\\\\ &3C\\,196\\tablenotemark{a}&Q&{\\it HST}&5\\\\ &3C\\,207&Q&{\\it Chandra}&6\\\\ &3C\\,263&Q&{\\it Chandra}&This paper\\\\ &3C\\,330&NLRG&{\\it Chandra}&This paper\\\\[2pt] \\tableline Unclear\\tablenotemark{b}&3C\\,351&Q&{\\it Chandra}&7, this paper\\\\ \\tableline Non-SSCE&Pictor A&BLRG&{\\it Einstein}&8\\\\ &&&{\\it Chandra}&9\\\\ &3C\\,390.3&BLRG&{\\it ROSAT}&10, 11\\\\ &3C\\,303\\tablenotemark{c}&Q&{\\it ROSAT}&12\\\\ \\enddata \\tablenotetext{a}{Claimed optical SSC detection.} \\tablenotetext{b}{This source will be discussed in more detail below.} \\tablenotetext{c}{X-ray source may be a background quasar.} \\tablecomments{Source type abbreviations are as follows: NLRG, narrow-line radio galaxy; BLRG, broad-line radio galaxy; Q, quasar. } \\tablerefs{(1) \\citealt{hcp94}; (2) \\citealt{wys00}; (3) \\citealt{hnpb00}; (4) \\citealt{hbw01a}; (5) \\citealt{h01}; (6) \\citealt{bbcs02a}; (7) \\citealt{bbcp01}; (8) \\citealt{rm87}; (9) \\citealt{wys01}; (10) \\citealt{p97}; (11) \\citealt{hll98}; (12) \\citealt{hw99}} \\label{hotspots} \\end{deluxetable}  {\\it ROSAT} observations of the archetypal FRII object Cygnus A \\citep*{hcp94} showed that the X-ray emission from its hotspots was consistent with the SSC process if the magnetic field strength was close to the equipartition or minimum-energy values \\citep{b56}. This result is often taken as evidence for equipartition between magnetic fields and synchrotron-emitting electrons in radio sources in general. Other {\\it ROSAT} observations, however \\citep*[e.g.,][]{hll98}, showed that some hotspots had X-ray emission too bright to be produced by the SSC mechanism with equipartition field strengths, suggesting that a different emission process was responsible. At the time of writing, after a number of new {\\it Chandra} hotspot detections, this dichotomy remains, as shown in Table \\ref{hotspots}. Three interesting facts about this division are immediately apparent. Firstly, the number of objects whose emission processes are not from SSC near equipartition (hereafter `non-SSCE' objects) is still a significant fraction of the total. Secondly, the non-SSCE objects all display broad emission lines, which may, after all, indicate some role for beaming in the hotspots. And, thirdly, the non-SSCE objects all have optical synchrotron hotspots.  The major remaining questions in this area are therefore: \\begin{enumerate} \\item \\label{q1} Are magnetic field strengths close to equipartition typical in hotspots? \\item What emission process is responsible for non-SSC sources, and is it related to their other common properties, or to relativistic beaming? \\end{enumerate}  Question \\ref{q1} can only be addressed by new observations of hotspots, and in the first part of this paper we report on our deep {\\it Chandra} observations of three sources selected on the basis of bright compact radio hotspot emission, describing the emission from their hotspots, lobes, nuclei and cluster environments. In \\S7 we will return to the second question.  The remainder of the paper is organized as follows. In \\S\\ref{obs} we discuss the selection of the sample and the observations we have used. In \\S 3 we discuss the general methods of analysis we have applied: in \\S\\S 4, 5 and 6 we discuss the results of applying these methods to our three target sources. Finally, in \\S\\S 7 and 8, we explore the general conclusions that can be drawn from our results.  Throughout the paper we use a cosmology with $H_0 = 65$ km s$^{-1}$ Mpc$^{-1}$, $\\Omega_{\\rm m} = 0.3$ and $\\Omega_\\Lambda = 0.7$. Spectral indices $\\alpha$ are the energy indices and are defined throughout in the sense $S \\propto \\nu^{-\\alpha}$. J2000.0 co-ordinates are used throughout. ",
        "conclusions": "We have detected X-ray emission from the hotspots and lobes of three powerful FRII radio sources, and low-level extended emission from the hot-gas environments of two of them.  The hotspots in two of the objects we have studied, 3C\\,330 and 3C\\,263, are consistent with being SSC emission with a hotspot magnetic field within a factor 2 below the value for energy equipartition with the synchrotron-emitting electrons. We argue therefore that magnetic field strengths close to equipartition are common in hotspots.  The double hotspots in the third source, 3C\\,351, are difficult to accommodate within an SSC model; they require magnetic field strengths much further from equipartition and, at least in one hotspot, there are offsets between the radio and X-ray centroids that a simple SSC model does not explain. The flat X-ray spectral indices rule out the possibility that the X-rays can be synchrotron emission from the electron population responsible for the radio emission. We are left with several alternative models, none of them particularly attractive: the most plausible condenders are an SSC model with large departures from equipartition, together with unmodeled spatial structure in the electron population in the secondary hotspot, a synchrotron model with a second, high-energy electron population, or strong beaming effects in the hotspot regions. In any case, it is striking that the two hotspots behave in such a similar way, in spite of their large spatial separation (and the expectation that primary and secondary hotspots should have different physics). As we have pointed out, other X-ray hotspot sources, like Cygnus A and 3C\\,390.3, seem to share this tendency for the hotspots to appear to know about each other.  The lobe emission we have detected seems most likely to be due to inverse-Compton scattering of microwave background photons by the synchrotron-emitting electron population. All three sources can be explained by such a model, in which the magnetic field strengths in the lobes are at most a factor 2 below equipartition (on a simple model with uniform field and electron distributions in the lobes). If this model is correct, the pressures in lobes are a factor of a few above the canonical minimum pressures. This factor may be enough to account for the discrepancy between the minimum pressure and the pressure in the external hot gas environment, discussed by \\citet{hw00}, and indeed we suggest that the two sources with detected extended X-ray emission, 3C\\,263 and 3C\\,330, are in approximate pressure balance with the hot gas in their cluster/group environments; such a model is also possible for 3C\\,351."
    },
    "0208/astro-ph0208174_arXiv.txt": {
        "abstract": "We have obtained a new high-resolution $K^\\prime$-band image of the central region of the rich X-ray cluster RX J0848.9+4452 at z=1.26. We found that the brightest cluster galaxy (BCG) in the cluster is clearly separated into two distinct objects. Whereas the optical to near-infrared colors of the objects are consistent with the predictions of passive evolution models for galaxies formed at high redshift, the luminosities of the two galaxies are both considerably fainter than predicted by passive evolution of BCG's in low and intermediate redshift clusters. We argue that this is evidence of an on-going merger of normal cluster ellipticals to form the dominant galaxy in the core of RX J0848.9+4452. The two galaxies appear to point towards the nearby cluster ClG J0848+4453 \\cite{sta97} and are aligned with the outer X-ray contour of their parent cluster, supporting a model of BCG formation by collimated infall along the surrounding large-scale structure. ",
        "introduction": "The cores of rich clusters are often dominated by massive elliptical galaxies: many of these brightest cluster galaxies (BCG's) are conspicuously different from other cluster members, in being at least one magnitude brighter than other giant ellipticals (Sandage 1976, Hoessel et al. 1980) and being located at or close to the bottom of the cluster potential well (Oegerle \\& Hill 2001). It is likely that BCG's arise from merging of massive satellites, drawn towards the cluster center by dynamical friction, as also hinted by the high frequency of multiple nucleation in these objects (e.g., Furuzawa et al. 1999 and  references therein), or by collimated infall along filaments in the early epochs of cluster formation (West 1994; Dubinski 1998).  The $K$-band Hubble diagram provides a measure of BCG evolution but previous results have been contradictory: \\cite{aes98} argued that BCG's dim with redshift according to, or fainter than, no-evolution models, suggesting that their mass has therefore increased in recent times by galaxy merging or cannibalism. On the other hand, \\cite{com98} found evidence that BGC's evolve passively in an X-ray selected sample of clusters. \\cite{bur00} suggested that the discrepancy is due to sample selection, as Aragon-Salamanca et al's (1998) sample consists of optically selected clusters some of which are not yet virialized enough to show strong X-ray emission.  Recently, \\cite{nel01, nel02a, nel02b} have confirmed that (i) BCG's in different environments evolve differently \\citep{bur00} and (ii) there has been significant growth in BCG mass in low X-ray luminosity clusters since at least $z \\sim 1$.  However, \\cite{bro02} recently found that there is no notable difference in the luminosity of brightest cluster galaxies between clusters with high and low X-ray luminosity at $z < 0.1$, suggesting that BCG evolution is nearly complete at the present epoch in cluster environments.  Observations at high redshift allow us to study systems in an earlier stage of dynamical evolution and at a younger age. Among the small sample of $z > 1$ clusters discovered so far, RX J0848.9+4452 \\citep{ros99} is one of the most secure examples of a rich, high concentration system resembling rich Abell clusters today. \\cite{sta01} measured the temperature of the X-ray hot gas as T=5.8 keV and showed the gas has a regular, centrally concentrated spatial distribution. The central galaxy of the cluster is indeed the highest-redshift one in the sample studied in \\cite{bur00}, and therefore provides the strongest leverage on the evolution of BCG's.  We have exploited the good image quality achievable on the Subaru telescope to obtain very deep near-infrared images of this cluster at high spatial resolution. We find that the BCG is clearly resolved into two components in our $K^\\prime$ image and interpret this result in the light of the $K$-band Hubble diagram of BCG's.  In this paper, we focus on the properties of the BCG in the cluster core while our study of the luminosity function and color-magnitude diagram will be presented in a separate paper (Yamada et al. in preparation). We adopt cosmological parameters $q_0=0.5$, $H_0$ = 50 km s$^{-1}$ Mpc$^{-1}$, $\\lambda_0=0$ throughout the paper for consistency with previous work unless noted. ",
        "conclusions": "We have found that the brightest galaxy in the rich X-ray cluster RX J0848.9+4452 at $z=1.26$ consists of two components and is therefore fainter than its local counterparts.   We consider here the colors of the two components in order to constrain their stellar populations. We used a 4800s WFPC2 $F814W$ exposure of this field taken by HST. Figure 4 shows a $4'' \\times 4''$ region of this image including our galaxies. The two components are clearly identified.  We smoothed the $F814W$ image to same resolution as our IRCS data and measured $I-K$ colors of the two components through $0.6''$-diameter apertures centered on the $K^\\prime$-band centroids. Both components have very red colors, $I_{F814W}-K$ = 3.27$\\pm$0.07 and 3.62$\\pm$0.08 (in $AB$ magnitude), respectively, but galaxy A is 0.3 mag bluer than B.  Figure 5 shows the two-color diagram of $K^\\prime$-selected galaxies in our IRCS field. Optical colors are from the Suprime-Cam study of this cluster by \\cite{nak02}. Our IRCS image was degraded to match the resolution and seeing ($1''$) of the Suprime-Cam image. Colors were measured within 2$^{\\prime\\prime}$-diameter apertures. Galaxies A and B are not separated in the image, and we treated them as a single galaxy. We measured  $R-I$ and $I-K$ colors (in Cousins and UKIRT-MK system) of our galaxies. We also plotted passive evolution models for old cluster galaxies from \\cite{kod97}. The composite brightest cluster galaxy (A+B) is very close to predictions of passive-evolution models for a $M_V =-23$ (at $z=0$) galaxy observed at z=1.26 whose predicted $K$-band magnitude is 16.4. It is also one of the reddest galaxies in the field. Thus galaxies A and B of the rich z=1.26 cluster have stellar populations quite consistent with prediction of the passive-evolution model for the most luminous old elliptical galaxies, but are less massive than today's BCG's, having sizes comparable to those of normal ellipticals.  If we assume that the two galaxies lie at the same redshift, their separation is as small as 6 kpc. Numerical simulations \\citep{hau78,rol79} show that these objects should merge on timescales of less than a few $10^8$ years. We therefore argue that we are witnessing the major merger of two massive ($ \\sim L^*$) elliptical galaxies to fuel the hierarchical growth of the BCG in RX J0848.9+4453 although the exact redshifts of the two systems and their relative velocity are necessary to draw definite conclusions on this issue. This result is consistent with models of hierarchical merging growth of BCGs. \\cite{aes98} argued that the typical mass of BCGs at $z=1$ is expected to be about half of the present. \\cite{dub98} also showed in his numerical simulation that the central galaxy forms through the merger of several massive galaxies in early history of cluster formation.  \\cite{yam00} have measured $K$ luminosities of BCG's in three other $z > 1$ clusters: 3C324, B2 1335+28 and ClG J0848+4453 and found these galaxies are all fainter than the predictions of \\cite{bur00} based on passive evolution models. These three clusters, however, are less X-ray luminous than RX J0848.9+4453 \\citep{dic97,sta01} and/or are more irregular and less concentrated in their galaxy distributions \\citep{kaj00,tan00,nak01,sta01}: the present result shows that BCG's are consistently less massive than present-day systems even in well-virialized systems.  It is not uncommon for present-day BCG's to contain multiple nuclei: about half of the BCG's in Abell clusters have multiple nuclei within 19.2 kpc \\citep{hoe85,ton85}; however, objects with small separation are somewhat rare: only 7\\% of the galaxies have multiple nuclei within 5 kpc of each other; the magnitude difference between the nuclei is larger than 1 magnitude in 60\\% of the low-redshift sample of \\cite{hoe85}. Therefore, the nascent BCG in RX J0848.9+4453 is a relatively rare object and may be an example of an equal mass merger as is predicted to occur at the top of the merging hierarchy.  The two component galaxies are aligned to point roughly in the direction of the nearby $z=1.27$ cluster ClG J0848+4453 \\citep{sta97} and are aligned with the outer X-ray contour of RX J0848.9+4452 \\citep{sta01} This may be the high redshift equivalent of the alignment effect of BCGs in nearby clusters \\citep{bin82,ful99}. Galaxies in the $z=1.2$ cluster around the radio galaxy 3C324 may be similarly aligned \\cite{nak01}. \\cite{wes94} proposed that this arises as clusters form by collimated infall along filaments. \\cite{dub98} also showed in his N-body simulation that the shape and orientation of the BCGs become nearly congruent with the galaxy distribution in the host cluster. The observed evidence of the alignment effect at high redshift, if confirmed, implies that such collimated structure formation indeed took place at $z > 1$.  \\vspace{0.5cm}  This work is based on data collected at Subaru Telescope, which is operated by the National Astronomical Observatory of Japan. This work is also based in part on observations with the NASA/ESA Hubble Space Telescope, obtained from the data archive at the Space Telescope Science Institute, U.S.A., which is operated by AURA, Inc.\\ under NASA contract NAS5--26555. The Image Reduction and Analysis Facility (IRAF) used in this paper is distributed by National Optical Astronomy Observatories. U.S.A., operated by the Association of Universities for Research in Astronomy, Inc., under contact to the U.S.A. National Science Foundation.TK acknowledges JSPS postdoctoral research fellowship and the Daiwa-Adrian Prize 2001 for financial support."
    },
    "0208/astro-ph0208497_arXiv.txt": {
        "abstract": "{ We report on the finding of the  strongest H$\\alpha$ emission -- pseudoequivalent width of 705~\\AA~-- known so far in a young, late type dwarf. This object, named as S\\,Ori~71, is a substellar candidate member of the 1--8\\,Myr star cluster $\\sigma$\\,Orionis. Due to its overluminous location in color-magnitude diagrams, S\\,Ori~71 might be younger than other cluster members, or a binary of similar components.  Its mass is in the range 0.021--0.012\\,$M_\\odot$, depending on evolutionary models and possible binarity. The broad H$\\alpha$ line of S\\,Ori~71 appears asymmetric, indicative of high velocity mass motions in the H$\\alpha$ forming region. The origin of this emission is unclear at the present time. We discuss three possible scenarios: accretion from a disk, mass exchange between the components of a binary system, and emission from a chromosphere. ",
        "introduction": "Both mass accretion and chromospheric activity have H$\\alpha$ emission as a signature.  Chromospheric activity appears as a consequence of the low density, the inverted temperature profile and the strong magnetic field present in K- and M-type dwarfs. Other Balmer lines, as well as Ca{\\sc ii} H\\&K and the Ca{\\sc ii} infrared triplet can appear in emission too.  The activity detected in some brown dwarfs (objects incapable of burning hydrogen stably, with masses below 0.075 $M$$_\\odot$, Chabrier at al$.$ \\cite{chabrier00} and references therein) have, probably, this origin (Mart\\'{\\i}n et al$.$ \\cite{martin99a}), although so far there is no theoretical model capable of explaining this phenomenology (e.g., Mohanty et al$.$ \\cite{mohanty02}). In addition, the H$\\alpha$ emission line of these cool objects often has an intrinsic variability, and strong flares are commonly detected in M dwarfs.  On the other hand, accretion can appear in interacting binaries (during the transfer process of material) or in accreting objects with circumstellar disks, such as the young TTauri stars. These pre-main sequence stars are normally classified as classical TTauri (CTT) stars or weak-line TTauri (WTT) stars.  The first group is characterized by strong H$\\alpha$ emission (larger than 10 or 20 \\AA, Appenzeller \\& Mundt \\cite{appenzeller89}; Mart\\'{\\i}n \\cite{martin97}), asymmetric and broad H$\\alpha$ profiles (sometimes with double peaks), presence of forbidden emission lines (arising from shocks produced by jets and outflows), blue/UV and infrared excesses, and a strong Li\\,{\\sc i}\\,6708\\,\\AA~line in absorption (indicative of youth). On the contrary, WTT stars lack most of these properties due to the absence of an active disk, and show smaller H$\\alpha$ emissions while keeping strong lithium lines. In addition, WTT stars display a lower degree of variability (e.g., Herbst et al$.$ \\cite{herbst02}).  The $\\sigma$\\,Orionis cluster (Walter et al$.$ \\cite{walter97}) is a young (1--8\\,Myr, Zapatero Osorio et al$.$ \\cite{osorio02a}) stellar association with low reddening ($E(B-V)$=0.05, Lee \\cite{lee68}), and located at a Hipparcos distance of 352$^{+166}_{-85}$\\,pc. Many substellar members of this cluster show H$\\alpha$ emission in their optical spectra (Barrado y Navascu\\'es et al$.$ \\cite{barrado02}, and references therein).  This paper presents the discovery of a brown dwarf with likely membership in $\\sigma$\\,Orionis, which has the strongest H$\\alpha$ emission line ever detected in a TTauri late-type star or active very cool object.  We discuss the origin of this extraordinary emission.  \\setcounter{table}{0} \\begin{table*} \\caption[]{ Photometric and spectroscopic data of S\\,Ori~71 (S\\,Ori\\,J053900.2--023706).  Photometric errors: $\\pm$\\,0.10\\,mag. } \\begin{tabular}{ccccccccc} \\hline RA (J2000) DEC                            & $I_c$ &$I_c-J$& SpT.            & pW (H$\\alpha$) & $T_{\\rm eff}$ & log\\,$L/L_{\\odot}$ & Mass (single) & Mass (binary) \\\\ ($^{h\\,m\\,s}$)\\hspace{10mm}($^\\circ$ ' '') &      &       & & (\\AA)          & (K)           & & ($M_{\\odot}$) & ($M_{\\odot}$) \\\\ \\hline 05 39 00.2 ~ --02 37 06                   & 20.02 & 2.88  & L0\\,$\\pm$\\,0.5  & 705\\,$\\pm$\\,75 & 2200--2500    & --2.66\\,$\\pm$\\,0.15& 0.014--0.021  & 0.012--0.017  \\\\ \\hline \\end{tabular}  \\end{table*} ",
        "conclusions": ""
    },
    "0208/astro-ph0208348_arXiv.txt": {
        "abstract": "We report on the analysis of a large sample of 744 type 1 Active Galactic Nuclei, including quasars and Seyfert\\,1 galaxies across the redshift range from $0\\la z \\la 5$ and spanning nearly 6 orders of magnitude in continuum luminosity. We discuss correlations of continuum and emission line properties in the rest-frame ultraviolet and optical spectral ranges. The well established Baldwin Effect is detected for almost all emission lines from \\ovi $\\lambda 1034$ to \\ob . Their equivalent widths are significantly anti-correlated with the continuum strength, while they are nearly independent of redshift. This is the well known Baldwin Effect. Its slope $\\beta $, measured as log\\,$\\,W_\\lambda \\propto \\beta ~{\\rm log}\\,\\lambda L_\\lambda (1450 {\\rm \\AA })$, shows a tendency to become steeper towards higher luminosity. The slope of the Baldwin Effect also increases with the ionization energy needed to create the individual lines. In contrast to this general trend, the \\nv $\\lambda 1240$ equivalent width is nearly independent of continuum luminosity and remains nearly constant. The overall line behaviors are consistent with softer UV continuum shapes and perhaps increasing gas metallicity in more luminous Active Galactic Nuclei. ",
        "introduction": "Broad emission lines (BELs) are a defining property of quasar spectra. Nearly 25 years ago Baldwin (1977) discovered an anti-correlation between the equivalent width in \\civ $\\lambda 1549$, $W_\\lambda$(\\civ ), and the continuum luminosity, $L_\\lambda$(1450 \\AA ), measured at 1450 \\AA\\ in the quasar rest-frame. This result was based on a sample of 20 quasars spanning $\\sim2$ orders of magnitude in continuum luminosity in the redshift range $1.24 \\la z\\la 3.53$. It has become known as the ``Baldwin Effect'' (hereafter BEff). The BEff was subsequently confirmed in \\civ $\\lambda 1549$, as well as other BELs such as \\lya\\ and \\ovi $\\lambda 1034$ (V\\'eron-Cetty, V\\'eron, \\& Tarenghi 1983; Baldwin, Wampler, \\& Gaskell 1989; Kinney, Rivolo, \\& Koratkar 1990; Osmer, Porter, \\& Green 1994; Zheng, Kriss, \\& Davidsen 1995; Green, Forster, \\& Kuraszkiewicz 2001; see Osmer \\& Shields 1999 for a recent review).  A major impetus for studying the BEff was that it might be useful for calibrating Active Galactic Nuclei (AGNs) luminosities, e.g. based on $W_\\lambda$(\\civ ). The AGNs could then be used as cosmological standard candles. But in the following years studies of bigger quasar samples revealed a large scatter in the anti-correlation of the continuum luminosity vs. emission line strength (Baldwin, Wampler, \\& Gaskell 1989; Zamorani et al.\\,1992). The BEff is nonetheless important as a diagnostic of AGN structure and, perhaps, metal abundances (Korista, Baldwin, \\& Ferland 1998). The relation between the continuum luminosity and the relative emission line strengths and ratios can be used to study the evolution and physics of the quasar phenomenon (Baldwin 1999). In particular, this correlation can be used to test model predictions for the dependence of the shape of the continuum spectral energy distribution as a function of luminosity (Binette et al.\\,1989; Netzer, Laor, \\& Gondhalekar 1992; Zheng \\& Malkan 1993; Wandel 1999a,b).  The most fundamental problem, however, is that the physical cause of the BEff remains unknown. It was suggested by Mushotzky \\& Ferland (1984) that the observed relation can be explained by an anti-correlation of the ionization parameter, U, and the continuum luminosity, with $U = Q(H) / 4 \\pi r^2 c n_H$, where Q(H) is the number of hydrogen ionizing photons emitted per second by the central continuum source, r is the distance between the continuum source and the emission line region, and $n_H$ is the hydrogen density in the line-emitting cloud. Assuming an additional relation of decreasing covering factor with increasing continuum strength, the observed BEff could be well described for \\civ $\\lambda 1549$. However, this model does not naturally explain the BEff in the full range of measured lines. It predicts, for example, a lack of a BEff for Ly$\\alpha$ even though it is clearly detected (e.g., Kinney, Rivolo, \\& Koratkar 1990; Osmer, Porter, \\& Green 1994; Laor et al.\\,1995; Green 1996).  Another important clue to the physical cause of the BEff is that the strength of the relationship (i.e., the slope of the $W_\\lambda $ -- $L_c$ anti-correlation) seems to depend on the ionization energy of the emission-line species (Zheng et al.\\,1995; Espey \\& Andreadis 1999). In particular, the equivalent widths of high-ionization lines like \\ovi $\\lambda 1034$ decrease more dramatically with $L_c$ than lines with moderate ionization energies like \\civ $\\lambda 1549$ or low ionization energies like \\mgii $\\lambda 2798$ and Balmer emission lines. Our results in the present paper confirm this claim, and improve upon the overall empirical characterization of the BEff correlations.  The BEff might result from a more fundamental correlation between the continuum luminosity, $L_c$, and the shape of the ionizing (EUV\\,--\\,soft X-ray) continuum (Binette et al.\\,1989; Zheng \\& Malkan 1993; Zheng et al.\\,1997; Korista et al.\\,1998). Netzer (1985, 1987) and Netzer et al.\\,(1992) suggested accretion disc models to explain the observed continuum and emission line correlations. Recently, Wandel (1999\\,a,b) added to this the growth in black hole mass by accretion in the accretion disc model. His analysis predicts that the continuum luminosity increases towards higher black hole mass and that the shape of the ionizing continuum becomes softer. Hence, it is suggested that the BEff is driven by a softening of the ionizing continuum towards higher luminosities. This model is attractive because the UV\\,--\\,X-ray spectral softening has been well documented by observations (Tananbaum et al.\\,1986; Wilkes et al.\\,1994; Green et al.\\,1995), and because the spectral softening provides a natural explanation for steeper BEff slopes in higher ionization lines.  An additional possibility is that the BEff is driven, at least in part, by a trend for higher metallicities in more luminous AGNs. Korista et al.\\,(1998) presented theoretical results showing the dependence of the emission line equivalent widths on both the continuum shape and the metallicity of the gas. Their proposal is based on evidence for higher metal abundances in more luminous quasars (Hamann \\& Ferland 1993, 1999; Dietrich et al. 1999; Dietrich \\& Wilhelm-Erkens 2000), and on the suggestion that more luminous quasars reside in more massive host galaxies, which will naturally be more metal rich (Cen \\& Ostriker 1999; Pettini 1999; Kauffmann \\& Haehnelt 2000; Granato et al.\\,2001; see Hamann \\& Ferland 1999 for a review).  Our combination of ground-based and satellite data also provides the opportunity to study evolutionary aspects of the line strengths in detail --- e.g. by measuring the same rest-frame UV lines across a wide range of redshifts. One serious complication affecting previous work is that luminosity and redshift are often correlated in quasars samples, because the quasars are selected from magnitude-limited surveys. Most studies have supported the original claim by Baldwin (1977), that W$_\\lambda$ scales inversely with luminosity and that there is no significant trend with redshift $z$ (e.g., Kinney, Rivolo, \\& Koratkar 1990;  Osmer et al.\\,1994; Francis \\& Koratkar 1995; Wilkes et al.\\,1999). However, some recent work based on the Large Bright Quasar Survey (LBQS) claims that the relationship to redshift is even stronger than with luminosity (Forster et al.\\,2001; Green, Forster, \\& Kuraszkiewicz 2001).  The present paper is the first in a series in which we examine the emission-line properties in a large sample of $744$ type 1 AGNs. The sample includes Seyfert\\,1 galaxies and both radio-loud and radio-quiet quasars spanning an unprecedented wide range in both redshift ($0\\la z\\la 5$) and intrinsic luminosity ($\\sim 6$ orders of magnitude). The database and data processing will be described in detail in a future paper. A major concern regarding BEff studies is that selection effects in AGN samples, e.g. pertaining to emission-line strengths, might bias some measurements of the BEff. Our sample has the advantage of being drawn from a variety of surveys, with objects selected based on radio properties, grism spectroscopy, or broad-band color criteria. Hence, no specific selection criteria were applied for the data. Another key advantage is that our sample includes a wide range of luminosities at various redshifts. We can therefore address the separate redshift and luminosity dependences in the emission line data.  There are several interesting problems we wish to address. The most basic issue is to quantify the nature of empirical correlations with $L_c$ and/or $z$. In an upcoming study we will further examine correlations with other AGN properties such as radio-loudness (Baldwin 1977; Sargent, Steidel, \\& Boksenberg 1989; Steidel \\& Sargent 1991; Francis \\& Koratkar 1995). We will also report on trends with $L_c$ or $z$ among various emission-line metallicity indicators. Here we present an initial analysis on the nature of the BEff. Our approach is to construct composite spectra for specific $L_c$ and $z$ intervals, thus providing high signal-to-noise spectra and allowing us to study trends in both weak and strong emission lines.  The AGN sample is described briefly in \\S2\\ and the calculation and analysis of the composite spectra in \\S3 and \\S4. The main results appear in several figures in \\S5. We compare our results with prior studies and discuss them in the context of suggested models in \\S6. Throughout this paper we use the cosmological parameters H$_o = 65$ km\\,s$^{-1}$\\,Mpc$^{-1}$, $\\Omega _M = 0.3$, and $\\Omega _\\Lambda = 0$ (Carroll, Press, \\& Turner 1992). Introducing $\\Omega_\\lambda = 0.7$ (Netterfield et al.\\,2002) instead of $\\Omega_\\lambda = 0.0$, the luminosities at the highest redshifts would be $\\sim 10$\\,\\%\\ smaller, while at low redshifts they would reach a maximum of $\\sim 20$\\,\\%\\ larger at redshift $z\\simeq 1$. ",
        "conclusions": "We have investigated a large sample of $744$ type\\,1 AGN covering the redshift range from $0 \\leq z \\leq 5$ and nearly 6 orders of magnitude in continuum luminosity. To enhance the signal-to-noise ratio, minimize the influence of peculiarities of individual quasars, and investigate weak as well as strong emission lines, we computed composite spectra representing narrow intervals in redshift and luminosity. The emission line fluxes were derived using multi-component Gaussian fits after removing a powerlaw continuum fit, a Balmer continuum emission template, and a UV and optical iron emission template from the composite spectra. Our main results are the following.  \\begin{itemize} \\item In composite spectra spanning the full redshift range at nearly constant luminosity we detect no strong trend in the line $W_\\lambda $ with redshift i.e., with cosmic time. However, there is a marginal tendency for the highest redshift quasars ($z \\ga 4$) to show slightly stronger emission lines than their counterparts at lower redshift. \\item In the composite spectra ranked by luminosity we find a significant Baldwin Effect in nearly all emission lines in the ultraviolet to optical domain. The only exceptions are \\nv $\\lambda 1240$, H$\\beta $, H$\\gamma $, and optical \\feii\\ which remain constant in $W_\\lambda $ within the uncertanties. The lack of a BEff for the high-ionization feature NV$\\lambda 1240$ suggests that the chemical composition of the gas is an additional parameter that can strongly influence the equivalent width of this and possibly other lines. \\item We detect a strong Baldwin Effect for the prominent NLR emission line [\\oiii ]$\\lambda 5007$. The strength of the BEff for [\\oiii ]$\\lambda 5007$ is very similar to the BEff measured in \\oiii ]$\\lambda 1663$. \\item The slope $\\beta $ of the Baldwin Effect, where log\\,$W_\\lambda \\propto \\beta \\cdot$\\,log\\,$\\lambda\\,L_\\lambda $, shows a significant correlation with the ionization energy, $\\chi _{ion}$, needed to produce the lines. \\item The slope of the Baldwin Effect, $\\beta $, tends to be steeper at higher luminosities, $\\lambda\\,L_\\lambda (1450 {\\rm \\AA }) \\ga 10^{44}$\\,erg\\,s$^{-1}$, compared to the lower luminosity regime. \\item The Baldwin Effect, its steepening towards higher luminosities, and the correlation of the slope $\\beta $ with $\\chi_{ion}$ can all be well explained in the context of a luminosity dependent spectral energy distribution of the ionizing continuum. Assuming that the SED can be described as a combination of a powerlaw continuum and a thermal UV bump, the ionizing continuum becomes softer for increasing luminosity as the UV bump is shifted to longer wavelengths. This behaviour can be explained with accretion disc models as suggested by Netzer et al.\\,(1992) and Wandel (1999a,b). \\end{itemize}"
    },
    "0208/astro-ph0208454_arXiv.txt": {
        "abstract": "{We have identified 7 new magnetic DA white dwarfs in the Early Data Release of the Sloan Digital Sky Survey. Our selection strategy has also recovered all the previously known magnetic white dwarfs contained in the SDSS EDR, KUV\\,03292+0035 and HE\\,0330--0002. Analysing the SDSS fibre spectroscopy of the magnetic DA white dwarfs with our state-of-the-art model spectra, we find dipole field strengths $1.5\\,\\mbox{MG}\\le B_\\mathrm{d}\\le 63$\\,MG and effective temperatures $8500\\le\\Teff\\le 39\\,000$\\,K. As a conservative estimate, we expect that the complete SDSS will increase the number of known magnetic white dwarfs by a factor 3. ",
        "introduction": "The population of magnetic white dwarfs spans an enormous parameter space in magnetic field strength $B$, effective temperature \\Teff, rotational period \\Prot, atmospheric abundances, and mass \\Mwd\\ -- with the number of accurate measurements per parameter dimension decreasing in this sequence. Despite intense spectroscopic and polarimetric surveys carried out over the last two decades \\citep[e.g.][]{schmidt+smith95-1, putney95-1, putney97-1, hagenetal87-1, reimersetal94-1, reimersetal96-1, reimersetal98-1}, only $\\sim65$ magnetic white dwarfs are known at present \\citep{jordan01-1,wickramasinghe+ferrario00-1}. The small size of this sample seriously hampers the progress of our understanding of the origin of the strong magnetic fields found in a small fraction (few \\%) of all white dwarfs, as well as of the evolution of these exotic stars.  The Sloan Digital Sky Survey (SDSS), the largest spectroscopic survey carried out to date, samples a great variety of galactic and extragalactic objects at high galactic latitudes. Due to the partial overlap in colour space between white dwarfs and quasars, it can be expected that the SDSS will result in the identification of a large number of white dwarfs and, hence, in a significantly increased sample of known magnetic white dwarfs. Here we describe the sample of magnetic white dwarfs identified in the Early  Data Release of the SDSS.   \\begin{table*}[t] \\caption[]{\\label{t-sdssobs} Confirmed and candidate magnetic white dwarfs from the SDSS EDR. The equinox/epoch 2000 coordinates are coded in the object designation. The objects are classified in the SDSS EDR as stellar (S) or unknown (U). The spectra are uniquely identified in the SDSS database by the combination of the Plate ID, the modified Julian date, and the fibre ID of the observation. The magnitudes listed here are from the associated imaging data.} \\begin{flushleft} \\setlength{\\tabcolsep}{1.2ex} \\begin{tabular}{llllllccccc} \\hline\\noalign{\\smallskip} MWD         & \\multicolumn{2}{c}{Class. ~~~ Sample} & Spectrum ID & Obs. date   & Exp.        & $u^*$ & $g^*$ & $r^*$ & $i^*$ & $z^*$ \\\\ \\hline\\noalign{\\smallskip} SDSS\\,J030407.40--002541.7 & DA & S & 411-51817-172 & 2000-09-30 10:37 & 2700\\,s & 18.06 & 17.77 & 17.95 & 18.09 & 18.37 \\\\ % SDSS\\,J033145.69+004517.0$^*$ & DA & S & 415-51810-370 & 2000-09-23 08:34 & 3600\\,s & 17.31 & 17.23 & 17.49 & 17.74 & 18.00 \\\\ % SDSS\\,J033320.37+000720.7$^+$ & DB?\\, & U &415-51810-492 & 2000-09-23 08:34 & 3600\\,s & 17.02 & 16.52 & 16.41 & 16.34 & 16.48 \\\\ % SDSS\\,J034511.11+003444.3 & DA & S & 416-51811-590 & 2009-09-24 09:35 & 3600\\,s & 19.11 & 18.63 & 18.52 & 18.49 & 18.50 \\\\ % SDSS\\,J121635.37--002656.2 & DA & U & 288-52000-276 & 2001-04-01 06:24 & 3602\\,s & 19.85 & 19.57 & 19.83 & 20.05 & 20.11 \\\\ % SDSS\\,J122209.44+001534.0 & DA & U & 289-51990-349 & 2001-03-22 06:25 & 3604\\,s & 20.51 & 20.23 & 20.47 & 20.65 & 21.04 \\\\ % SDSS\\,J172045.37+561214.9 & DA & U & 367-51997-461 & 2001-03-22 11:51 & 6302\\,s & 20.00 & 20.11 & 20.49 & 20.76 & 21.29 \\\\ % SDSS\\,J172329.14+540755.8 & DA & S & 359-51821-415 & 2000-10-03 03:59 & 4500\\,s & 19.14 & 18.81 & 18.90 & 19.04 & 19.30 \\\\ % SDSS\\,J232248.22+003900.9 & DA & S & 383-51818-421 & 2000-09-07 08:10 & 3600\\,s & 18.91 & 19.02 & 19.31 & 19.62 & 19.82 \\\\ % SDSS\\,J232337.55--004628.2 & DA? & S & 383-51818-215 & 2000-10-01 04:29 & 3600\\,s & 17.88 & 18.00 & 18.27 & 18.52 & 18.78 \\\\ % \\noalign{\\smallskip}\\hline \\multicolumn{9}{l}{$^*$\\,=\\,KUV\\,03292+0035~~~$^+$\\,=\\,HE\\,0330--0002} \\\\ \\end{tabular} \\end{flushleft} \\end{table*} ",
        "conclusions": "The population of known magnetic white dwarfs presently comprises $\\sim65$ stars \\citep{wickramasinghe+ferrario00-1}. Considering that the first magnetic white dwarf (Grw$+70^{\\circ}8047$) has been discovered nearly 60 years ago, the ``average'' discovery rate of these stars has been $\\sim1$ per year. As outlined in the Introduction, this sample is still far too small to stringently test theoretical models for the origin of the magnetic field and for the evolution of magnetic white dwarfs.  Unfortunately, discovering new magnetic white dwarfs is a tedious process. Figure\\,\\ref{f-colours} shows colour-colour diagrams for all ``stellar'' objects from the EDR spectroscopic database. It is apparent that the magnetic white dwarfs discussed are well distributed over the locus of hot stars --~mainly white dwarfs and subdwarfs. It is, hence, not possible to select magnetic white dwarfs by their colour alone, the availability of a large spectroscopic data set is essential for a significant increase in the number of known magnetic white dwarfs.  In this paper, we have presented fibre spectroscopy of 10 magnetic white dwarf candidates identified in the EDR of the SDSS. We confirm eight of these objects as magnetic DA white dwarfs, of which seven are new discoveries spanning a wide range of magnetic field strengths. This represents a substantial addition to the population of known magnetic white dwarfs. The spectrum of SDSS\\,2323 contains relatively broad \\Hb\\ and \\Hg\\ absorption troughs, however, we were not able to confirm the magnetic nature of this star from the SDSS data alone. Finally, we recovered the peculiar magnetic white dwarf HE\\,0330--0002.  Our eye ball inspection of the SDSS fibre spectra is certainly biased in a number of ways.  In white dwarfs with weak magnetic field strengths ($B\\la1-2$\\,MG) the resolution and the signal-to-noise (S/N) ratio of the SDSS spectra becomes insufficient to detect the Zeeman splitting of the Balmer lines.  Insufficient S/N of the SDSS also lowers the probability of detecting the weak Zeeman absorption components in both hot or high field white dwarfs ($\\Teff\\ga40\\,000$\\,K, $B\\ga80$\\,MG). Finally we have restricted or analysis to magnetic white dwarfs with a pure hydrogen atmosphere~--~and discarded a handful of SDSS spectra with absorption features that could not be identified either with typical stellar transitions or with hydrogen Zeeman components for a wide range of field strengths. We are, however, confident that our selection of magnetic white dwarfs is complete for stars with pure hydrogen atmospheres, magnetic fields of a few MG to a few tens MG, and effective temperatures $10\\,000\\la\\Teff\\la40\\,000$\\,K. An \\textit{a posteriori} check of the EDR database using the lists of \\citet{jordan01-1} and \\citet{wickramasinghe+ferrario00-1} confirmed that KUV\\,\\,03292+0035 and HE\\,0330--0002 are the only previously known magnetic white dwarfs with spectroscopic coverage.  The Early Data Release represents $\\sim$5\\% of the final SDSS data set. Accounting for our selection bias, we conservatively estimate that the complete SDSS will increase the number of known magnetic white dwarfs by at least a factor 3. Scaling the number of white dwarfs identified by \\citet{stoughtonetal02-1} in the EDR sample (734) to the complete SDSS, and assuming $\\sim2-4\\%$ as ratio of magnetic to non-magnetic white dwarfs \\citep{jordan01-1, wickramasinghe+ferrario00-1}, suggests that a careful exploitation of the entire SDSS data base, including intense spectroscopic and polarimetric follow-up observations, may well lead to the discovery of several hundred new magnetic white dwarfs."
    },
    "0208/astro-ph0208512_arXiv.txt": {
        "abstract": "Exploring the recent expansion history of the universe promises insights into the cosmological model, the nature of dark energy, and potentially clues to high energy physics theories and gravitation.  We examine the extent to which precision distance-redshift observations can map out the history, including the acceleration-deceleration transition, and the components and equations of state of the energy density. We consider the ability to distinguish between various dynamical scalar field models for the dark energy, as well as higher dimension and alternate gravity theories.  Finally, we present a new, advantageous parametrization for the study of dark energy. ",
        "introduction": "The quest to explore the expansion history of the universe has carried cosmology well beyond ``determining two numbers'' -- the present dimensionless density of matter $\\Omega_m$ and the present deceleration parameter $q_0$ \\cite{san61}.  Observations have advanced so that now cosmologists seek to reconstruct the entire function $a(t)$ representing the expansion history of the universe.  A myriad of cosmological observational tests can probe the function $a(t)$ more fully, over much of the age of the universe (see Sandage \\cite{san88}, Linder \\cite{lin89,lin97}, Tegmark \\cite{teg}). This paper concentrates on the most advanced method, the magnitude-redshift relation of Type Ia supernovae.  Just as understanding the thermal history of the early universe has taught us an enormous amount about both cosmology and particle physics (see, e.g., \\cite{kt}), the recent expansion history of the universe promises similarly fertile ground with the discovery of the current acceleration of the expansion of the universe.  This involves concepts of the late time role of high energy field theories in the form of possible quintessence, scalar-tensor gravitation, higher dimension theories, brane worlds, etc.~as well as the fate of the universe. ",
        "conclusions": "The geometry, dynamics, and composition of the universe are intertwined through the theory of gravitation governing the expansion of the universe. By precision mapping of the recent expansion history we can hope to learn about all of these.  The brightest hope for this in the near future is the next generation of distance-redshift measurements through Type Ia supernovae that will reach out to $z\\approx1.7$.  Just as the thermal history of the early universe taught us much about cosmology, astrophysics, and particle physics, so does the recent expansion history have the potential to greatly extend our physical understanding.  With the new parametrization of dark energy suggested here, one can study the effects of a time varying equation of state component back to the decoupling epoch of the cosmic microwave background radiation.  But even beyond dark energy, exploring the expansion history provides us cosmological information in a model independent way, allowing us to examine many new physical ideas.  From two numbers we have progressed to mapping the entire dynamical function $a(t)$, to the brink of a deeper understanding of the dynamics and fate of the universe."
    },
    "0208/astro-ph0208038_arXiv.txt": {
        "abstract": "A contour plot of positive iso-$J$ curves is shown for a gravitational binary  lens $\\epsilon_2 = 0.1883$ and $\\ell = 0.687$. The caustic curve is made of 4-cusped central caustic (so-called ``stealth bomber\") on the lens axis and two triangular caustics off the lens axis. The cusps of the trioids are all negative cusps. The positive image magnification contours outside the positive cusps on the lens axis are elongated along the symmetry axis.  It is in contrast with figure 4 of astro-ph/0206162 (Gaudi and Petters) where the contours appear to contract along the symmetry axis. We apologize if our citation of their figure 4 without the caution played any role of causing confusion with readers. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0208/astro-ph0208274_arXiv.txt": {
        "abstract": "{ We present the first results of an observational project, which addresses the period changing behaviour of a sample of high-amplitude $\\delta$ Scuti stars. In this paper we discuss the double-mode nature of V567~Ophiuchi. It was observed on 15 nights in two consecutive years in order to resolve the long-standing ambiguity related to its secondary period. A frequency analysis of almost 5000 individual single-filtered CCD $V$ measurements resulted in two independent frequencies ($f_1=6.6879$ d$^{-1}$ and $f_2=11.8266$ d$^{-1}$) with a ratio of $f_1/f_2=0.565$. Earlier data taken from the literature were used to refine the dominant period, and the re-analysis supports the existence of the secondary period. Possible asteroseismological implications are briefly discussed. ",
        "introduction": "High-amplitude $\\delta$ Scuti stars (hereafter HADS) form an interesting subgroup of short-period pulsators located inside the classical instability strip near the main-sequence. Their light variation is characterized by relatively large amplitudes (a conventional limit is $A_{\\rm V}\\geq0\\fm30$) which is associated with one or two stable frequencies. The variability of these stars has been interpreted as caused by radial pulsation in fundamental or low-order radial overtone pulsation (Rodr\\'\\i guez et al. 1996, Petersen \\& Christensen-Dalsgaard 1996), although some empirical evidence is present for microvariability due to weakly excited high-order radial or non-radial modes of pulsation (Garrido \\& Rodr\\'\\i guez 1996). It also appears that they exhibit a period-luminosity relation which has been studied by several authors (e.g. McNamara 1997, Petersen \\& H\\o g 1998). Previous studies based on bright field stars have recently been supplemented by analyses of tens of objects discovered by the MACHO and OGLE projects (Alcock et al. 2000, Poretti 2001).  Breger \\& Pamyatnykh (1998) tried to infer evolutionary conclusions from long-term observational records of selected HADS. Although they found evidence of period changes in a significant fraction of HADS, it was not possible to relate them to stellar evolution. The detected period changing behaviours range from continuous period decrease to continuous period increase, and period jumps might also be present in several stars. Furthermore, cyclic period variations due to possible binary light-time effect were inferred in a few cases (see references in Breger \\& Pamyatnykh 1998). However, despite the large amplitudes, short periods and moderate brightnesses of the stars studied, many of them need period updates as the latest observations in the literature were obtained almost two decades ago. That is why we started a photometric monitoring of bright northern HADS. Our target stars include all HADS brigther than $V\\sim11^{\\rm m}$ situated in favourable positions in the northern sky (our sample is partially overlapping with that of Breger \\& Pamyatnikh). The observations started in 1995 (Kiss \\& Szatm\\'ary 1995), and since then we have obtained unfiltered (one star), Johnson (eight stars) and Str\\\"omgren (one star) photometric observations for ten stars in order to get an updated view of their period changes (some early results have already been published in Kiss \\& Szatm\\'ary 1995 and Kiss \\& Derekas 2000). Here we report on results for V567~Ophiuchi, which is the only one that clearly shows double-mode pulsation. The period changes of the remaining stars will be discussed in a companion paper.  The light variations of V567~Oph (=BD+1$^\\circ$3547, $\\langle V\\rangle\\approx 11\\fm2$, $A_{\\rm V}=0\\fm34$, $P=0.1495$ d, spectral type A6--F1, Powell et al. 1990) were discovered by Hoffmeister (1943) giving a period ($\\approx 1/8$ d) that was an alias of the true one ($\\approx 1/7$ d). De Bruyn (1972) was the first who determined the period accurately. The observational record until 1990 is quite numerous and has been summarized in Powell et al. (1990). These authors carried out a detailed photometric and spectroscopic study of V567~Oph, which was based on Str\\\"omgren photometry and medium-resolution optical spectroscopy. Besides determining the fundamental physical parameters of the star, Powell et al. (1990) suspected the existence of a secondary period, although they could not draw a firm conclusion on it. In the same year, Poretti et al. (1990) presented Fourier decomposition of three HADS. The true period of 0.149 d was established for V567~Oph and they concluded that it was monoperiodic. This conclusion was based on four nights of observations distributed in two separate years (1984 and 1986). Rodr\\'\\i guez et al. (1996) studied the phase shifts and amplitude ratios for a large set of stars, including V567~Oph. The radial nature of its pulsation was deduced. Hintz \\& Joner (1997) and McNamara (1997), adopting the suggestion by Powell et al. (1990), mentioned the star in their studies as double-mode variable. Contrary to this, Petersen \\& H\\o g (1998) listed V567~Oph as oscillating in fundamental mode only. Musazzi et al. (1998) went farther, as they refused the secondary periodicity of the star. It was claimed that the stability of the light curve found by Poretti et al. (1990) clearly proves the monoperiodic nature. The final argument so far was presented by Schwendiman \\& Hintz (1999), who gave information on the existence of a second period of V567~Oph. Unfortunately, in their poster abstract they did not go into any further details (neither the period value nor its amplitude was specified).  Other studies dealing at least partly with V567~Oph include Kinman (1998), who presented an analysis of local space densities of HADS. He noted that for four stars (DY~Her, V567~Oph, ZZ~Mic and EH~Lib) it is likely that they are old disk members instead of belonging to the young galactic disk or the halo, as other HADS do. Balona \\& Evers (1999) discussed the mode identification of well-observed $\\delta$ Scuti stars with the use of multicolour photometry. The applied procedure failed to infer plausible mode identification in three HADS (DY~Her, RS~Gru, V567~Oph). The single frequency in each case was identified with an $l=1$ g-mode. Balona \\& Evers (1999) concluded that the most likely explanation of their result is that these stars are evolved radially-pulsating stars outside the range of the models applied. Although the models show unstable g-modes for main-sequence stars, these are unlikely to attain the high amplitudes which are observed.  In this paper, we present new Johnson $V$ photometry of V567~Oph which revealed the secondary period unambiguously. The observations are described in Sect.\\ 2, while the period analysis is discussed in Sect.\\ 3. An interpretation and possible implications are given in Sect.\\ 4. ",
        "conclusions": "What can be said about the nature of this double-mode pulsation? To draw some constraints on mode identification, we have inspected the period ratio and $Q$ values.  The frequency ratio has an intriguing value of $f_1/f_2=0.565$ which is far from the usually found 0.75--0.79 associated with radial fundamental and first overtone pulsation (see Petersen \\& Christensen-Dalsgaard (1996) for a parameter study, while Alcock et al. (2000) for a larger sample of double-mode HADS). As pointed out by McNamara (2000), several candidates for higher overtone pulsation can be found in $\\omega$~Cen, for which recent linear nonadiabatic models (Gilliland et al. 1998) predict $f_0/f_2=0.63$ and $f_0/f_3=0.53$. However, neither of these two values fits the observed one which is fairly accurate (its error is less than $\\pm$0.001). From the observational point of view, it is interesting that there is no other double-mode HADS with similar frequency ratio. Therefore, we searched the literature to find similar stars among the lower amplitude $\\delta$ Scuti stars. Close resemblance was found for AN~Lyn (0.565, Rodr\\'\\i guez et al. 1997, Zhou 2002), V663~Cas (0.591, Mantegazza \\& Poretti 1990) and 63~Her (0.564, Breger et al. 1994). In all cases the authors arrived to the conclusion that a mixture of radial and non-radial modes is needed to explain the ``non-standard'' frequency ratios. Adopting their argumentations, it is reasonable to accept this consideration in the case of V567~Oph, too. Recent theoretical models also support this idea (Bono et al. 1997, Gilliland et al. 1998).  We have calculated the pulsation constant $Q$ for both frequencies with the formula  $$\\log Q=-6.456+\\log P+0.5 \\log g + 0.1 M_{\\rm bol} + \\log T_{\\rm eff}$$  \\noindent in terms of four observables (Breger et al. 1993). The physical parameters of V567~Oph were determined by Powell et al. (1990), McNamara (1997) and Balona \\& Evers (1999). We list the corresponding $Q$ values in Table\\ 4.  \\begin{table} \\begin{center} \\caption{$Q$ values from various physical parameter determinations} \\begin{tabular}{|llllll|} \\hline Ref. & $\\log g$ & $M_{\\rm bol}$ & $\\log T_{\\rm eff}$ & $Q_1$ & $Q_2$ \\\\ \\hline (1) & 3.74 & 1\\fm1 & 3.87 & 0.037 & 0.021\\\\ (2) & 3.76 & 0\\fm85 & 3.87 & 0.035 & 0.020\\\\ (3) & 3.88 & 1\\fm40 & 3.87 & 0.047 & 0.027\\\\ (3) & 3.26 & 1\\fm34 & 3.90 & 0.024 & 0.013\\\\ \\hline \\end{tabular} \\end{center} References: (1) Powell et al. (1990), (2) McNamara (1997), (3) Balona \\& Evers (1999). \\end{table}  While the first two sets of parameters are in good agreement, the last two are fairly contradictory. Balona \\& Evers (1999) used different calibrations. For V567~Oph, they applied calibrations by Balona (1994) and Moon \\& Dworetsky (1985) -- the latter one produced that deviant $\\log g$=3.26 resulting in hardly acceptable $Q$ values. If we keep the first set of parameters of Balona \\& Evers, it still results in uncomfortably high $Q_1$. However, this is likely caused by a systematic error in their Str\\\"omgren calibrations, an error that was pointed out by Rodr\\'\\i guez \\& Breger (2001). From a comparison of photometric and geometric parallaxes they found that for slowly rotating stars Str\\\"omgren calibrations underestimate the absolute magnitude by 0\\fm5 in average. Since V567~Oph is a typical slowly rotating HADS ($v\\sin i<18$ km~s$^{-1}$, McNamara 1985), its absolute magnitude is likely to be affected by this systematic error. A correction by $\\Delta M_{\\rm V}=0\\fm5$ shifts $Q_1$ to 0.036 and $Q_2$ to 0.020, being in good agreement with the other two calibrations.  Assuming 20\\% uncertainty in $Q$ values (that is $\\pm0.007$ and $\\pm0.004$ for $Q_1$ and $Q_2$, respectively) we conclude that the dominant period indeed corresponds to the radial fundamental mode, as assumed for the overwhelming majority of HADS. On the other hand, the secondary period can be identified with a non-radial mode of radial order $n=2$ or 3 (Bono et al. 1997, Gilliland et al. 1998). Presently nothing can be specified unambiguously about the non-radial degree $l$, for which the simpliest assumption is 1 or 2.  In order to perform more secure mode identification, accurate multicolour photometry would be of great importance. Unfortunately, the relative faintness of the star does not make it favourable target object for photoelectric photometry. Therefore, CCD observers with proper instrumentation are expected to gain more insights into the peculiar pulsation pattern of V567~Oph by taking follow-up observations of this interesting HADS."
    },
    "0208/astro-ph0208042_arXiv.txt": {
        "abstract": "Transits of bright stars offer a unique opportunity to study detailed properties of extrasolar planets that cannot be determined through radial-velocity observations.  We propose a new technique to find such systems using all-sky small-aperture transit surveys.  We derive a general formula for the number of stars that can be probed for such systems as a function of the characteristics of the star, the planet, and the survey.  We use this formula to derive the optimal telescope design for finding transits of bright stars: a 5 cm ``telescope'' with a $4k\\times 4k$ camera. ",
        "introduction": "In the past three years, great strides have been made in the detection of extrasolar planets (XSPs).  To date, nearly all of the roughly 100 known XSPs have been discovered using the radial velocity (RV) technique.  However, RV detections, in and of themselves, yield only a few planetary parameters, namely the period $P$, the eccentricity $e$, and $M \\sin (i)$, where $M$ is the mass of the planet and $i$ is the inclination of its orbit.  By contrast, if a planet transits its host star, much more information is available.  First, of course, the $M \\sin (i)$ degeneracy can be broken.  Second, the ratio of the radii of the planet and host star can be measured.  Therefore, provided that the star can be classified well enough to determine its mass and radius, then the planet's radius and hence its density can be determined.  Third, and perhaps most important, if the transits can be  observed with sufficient signal-to-noise ratio (S/N), then one can probe otherwise unobservable details of a planet, such as its oblateness \\citep{hui02}, atmospheric conditions \\citep{char02}, and perhaps satellites and rings. Regardless of how a planet is initially discovered, once it is determined to transit its host star, this wealth of information can in principle be extracted by intensive follow-up observation of these transits.  This fact has been amply demonstrated by the discovery and analysis of the transiting planet HD209458b \\citep{char00, cody02}.  At the moment, all ongoing and proposed transit surveys are carried out in relatively narrow pencil beams.  They make up for their small angular area with relatively deep exposures.  These surveys fall into two basic classes: field stars \\citep{how00, brown99, mal01, udal02}\\footnote{ \\url{http://www.psi.edu/\\~{}esquerdo/asp/asp.html} \\\\ \\url{http://www.hao.ucar.edu/public/research/stare/stare.html} \\\\ \\url{http://bulge.princeton.edu/\\~{}ogle/}}, and clusters (\\citealt{str00}; \\citealt{burke02})\\footnote{ \\url{http://star-www.st-and.ac.uk/\\~{}yt2/WEB\\_GROUP/top.html}}.  These surveys are potentially capable of establishing the frequency of planets in various environments, but they are unlikely to find the kinds of transits of bright stars that would be most useful for intensive follow-up analysis. Although some of the surveys of field stars are considered ``wide field'', their total survey areas are small compared to $4\\pi$ str.  One project that has the potential to cover a very large area is WASP \\citep{str02}, which plans to employ five cameras, each with a $9.^{\\circ}5 \\times 9.^{\\circ}5$ field of view.  An alternative method is to conduct an all-sky survey.  Instead of continuous observation of all targets (which is impossible from a practical standpoint for an all-sky survey), this approach would necessarily involve revisiting each target in the sky at regular, semi-regular, or random intervals throughout the course of the project. This kind of observing strategy will not yield a continuous light curve on any star, as the current transit surveys do.  Rather, this plan will generate an only sporadically sampled light curve.  However, the long time baseline for the survey will eventually generate just as many individual observations of a single star.  Transit-like dips in the data stream will not be visually obvious, but by repeatedly phase-folding the full light curve back on itself over a range of periods, one can detect the dips from the transits. (See \\S~\\ref{sec:randomnoise}.)  This approach is especially relevant given the fact that there are several all-sky surveys already being planned for objectives other than transit detections.  It should be possible, for instance, to utilize the photometric data stream of upcoming astrometric missions for transit detection.  Space-based projects such as GAIA\\footnote{http://astro.estec.esa.nl/GAIA/gaia.html} and DIVA\\footnote{http://www.ari.uni-heidelberg.de/diva/diva.html} would take hundreds of observations of millions of stars over mission lengths of years with the aim of obtaining precise astrometry.  These data could equally well be analyzed for planetary transits.  There are also several existing or proposed ground-based all-sky surveys, including the Large-aperture Synoptic Survey Telescope (LSST)\\footnote{http://www.lssto.org/lssto/index.htm}, the Panoramic Optical Imager (POI), and the All-Sky Automated Survey (ASAS)\\footnote{http://www.astrouw.edu.pl/\\~{}gp/asas/asas.html} \\citep{poj00}. These surveys will be imaging the entire sky every few days, a qualitatively similar cadence to those of GAIA and DIVA.  While LSST and POI will likely be saturated by stars of $V\\la 12$, which are of greatest interest for transit follow-ups, ASAS is of particular interest in the present context because of its very small aperture.  In this paper, we examine the process of analyzing photometric data streams from all-sky surveys to find transits.  We calculate the sensitivity to XSP detection, the distance out to which these detections will be possible, and the number of false-positive detections due to random noise.  We derive a general expression for the number of stars that can be probed by this technique as a function of the total sensitivity of the survey. We apply our analysis to the problem of telescope design and conclude that very small, 5 cm telescopes are optimal for finding transits of bright stars. ",
        "conclusions": ""
    },
    "0208/astro-ph0208332_arXiv.txt": {
        "abstract": "For the classification of rotating compact stars with two high density phases a phase diagram in the angular velocity ($\\Omega$) - baryon number ($N$) plane is investigated. The dividing line $N_{\\rm crit}(\\Omega)$ between configurations with one and two phases is correlated to a local maximum of the moment of inertia and can thus be subject to experimental verification by observation of the rotational behavior of accreting compact stars. Another characteristic line, which also can be measured is the transition line to black holes that of the maximum mass configurations. The positions and the shape of these lines are sensitive to changes in the equation of state (EoS) of stellar matter. A comparison of the regional structure of phase diagrams is performed for polytropic and relativistic mean field type EoS and correlations between the topology of the transition lines and the properties of two-phase EoS are obtained. Different scenarios of compact star evolution are discussed as trajectories in the phase diagram. It is shown that a population gap in the $\\Omega - N$ plane for accreting compact stars would signal a high density phase transition and could be observed in the distribution of so called Z sources of quasi periodic oscillations in low-mass X-ray binaries.  \\pacs{PACS numbers: 04.40.Dg, 12.38.Mh, 26.60.+c, 97.60.Gb} ",
        "introduction": " ",
        "conclusions": ""
    },
    "0208/astro-ph0208104_arXiv.txt": {
        "abstract": "We review the characteristics of the dust continuum emission from normal galaxies, as revealed by the ISOPHOT Virgo Cluster Deep Survey (Tuffs et al. 2002; Popescu et al. 2002, Popescu \\& Tuffs 2002b). ",
        "introduction": "The ISOPHOT Virgo Cluster Deep Survey (Tuffs et al. 2002; Popescu et al. 2002) represents the {\\it deepest survey} (both in luminosity and surface brightness terms) of normal galaxies yet measured in the Far-Infrared (FIR). The survey consists of 63 gas-rich Virgo Cluster galaxies selected from the Virgo Cluster Catalog (VCC; Binggeli, Sandage \\& Tammann 1985; see also Binggeli, Popescu \\& Tammann 1993) and measured using the ISOPHOT instrument (Lemke et al. 1996) on board ISO (Kessler et al. 1996).  The fundamental incentive for choosing the VCC as the basis of a statistical sample for ISOPHOT was that a luminosity- {\\it and} volume - limited sample of cluster periphery and cluster core galaxies representative of the field and cluster environments, respectively, could be observed down to the least luminous dwarf galaxies reachable with ISOPHOT. This should allow an investigation of the \\newline strength and time-dependence of all manifestations of star formation activity and its relation to intrinsic galaxy properties such as Hubble type or sheer overall size. From an observational point of view the Virgo cluster has the advantage that it is situated at high galactic latitude and is close to the ideal distance for the detection of dwarf galaxies with ISOPHOT. The VCC has also a full representation of morphological types of normal gas rich galaxies, including  quiescent systems and even, to some extent, low surface brightness objects, ranging from bright ($B_{\\rm T}\\,\\sim\\,10$) giant spirals down to blue compact dwarfs (BCDs) and irregular galaxies at the completeness level of $B_{\\rm T}\\,\\sim\\,18$. Thus, the VCC is ideal for providing the basis for statistical investigations of the FIR properties of gas rich galaxies in the local universe spanning a broad range in star-formation activity and morphological types, including dwarf systems.  \\begin{figure}[htb] \\includegraphics[scale=0.5]{figure1.eps} \\caption{The Virgo Cluter spiral galaxy VCC~66. Contour plots of the dust emission observed with ISOPHOT at 100\\,${\\mu}$m are overlaid on the R-band image.} \\end{figure} ",
        "conclusions": ""
    },
    "0208/hep-th0208019_arXiv.txt": {
        "abstract": "{We consider scalar Born-Infeld type theories with arbitrary potentials $ V(T)$ of a scalar field $T$. We find that for models with runaway potentials $V(T)$ the generic inhomogeneous solutions after a short transient stage can be very well approximated by the solutions of a Hamilton-Jacobi equation that describes free streaming wave front propagation. The analytic solution for this wave propagation shows the formation of caustics with multi-valued regions beyond them. We verified that these caustics appear in numerical solutions of the original scalar BI non-linear equations. Our results include the scalar BI model with an exponential potential, which was recently proposed as an effective action for the string theory tachyon in the approximation where high-order spacetime derivatives of $T$ are truncated. Since the actual string tachyon dynamics contain  derivatives of all orders, the tachyon BI model with an exponential potential becomes inadequate when the caustics develop because high order spatial derivatives of $T$ become divergent.\\\\ BI type tachyon  theory with   a potential   decreasing at large $T$ could have interesting cosmological applications because the tachyon field rolling towards its ground state at infinity acts as pressureless dark matter. We find that inhomogeneous cosmological tachyon fluctuations  rapidly grow and develop multiple caustics.  Any considerations of the role of the tachyon field in cosmology will have to involve finding a way to predict the behavior of the field at and beyond these caustics.} ",
        "introduction": "In this paper we investigate generic  solutions of scalar Born-Infeld field theories of the form \\begin{equation}\\label{action} S = -\\int d^Dx \\, V(T) \\sqrt{1 + \\alpha' \\partial_\\mu T \\partial^\\mu T} \\ , \\end{equation} where $T(x^{\\mu})$ is a (dimensionless) scalar field and $V(T)$ is its potential. The dimensional parameter  $\\alpha'$ has units of $mass^{-2}$. In string theory $M_s=l_S^{-1}=1/\\sqrt{\\alpha'}$ are the fundamental string mass and length scales. For the remainder of the paper we will work in units where $\\alpha'=1$.  Although this problem has merit by itself, especially as a BI generalization of the sigma model approach to the effective string theory action \\cite{sigma}, the reason to explore this theory now is related to  the rolling tachyon and its application to  cosmology.  Recently Sen conjectured \\cite{Sen1,Sen2} that the qualitative dynamics of a string theory tachyon field in a background of an unstable D-brane system could be approximated with the effective action (\\ref{action}) with $V(T) \\sim e^{-T}$.  However, string theory tachyon calculations are  reliable only in the approximation where derivatives $\\nabla^{\\mu} T$ are truncated beyond the quadratic order \\cite{trun}. For instance, the 4D effective field theory action of the tachyon field $T$ on a D$3$-brane, computed in bosonic theory around the top of the potential, up to higher derivative terms, is given by \\cite{trun} \\begin{equation}\\label{action1} S_B = \\tau_3\\int d^4x\\, \\sqrt{g}\\left(\\alpha' e^{-T}  \\partial_\\mu{T}\\partial^\\mu{T} + (1+T) e^{-T}\\right) + O(  \\partial_\\mu \\partial^\\mu{T}) \\ , \\end{equation} where $\\tau_3$ is the brane tension.  It is far from clear that the model (\\ref{action}) describes actual tachyon dynamics. Still, this relatively simple formulation of tachyon dynamics (\\ref{action}) triggered significant interest in the investigation of the role of tachyons in cosmology \\cite{cosm,FKS,SW,KL,STW} and in the field theory of tachyons \\cite{Sen3}. . Even before the recent works on tachyon cosmology, a model known as ``k-inflation'' \\cite{k}, based on the action (\\ref{action}), had been considered in cosmology in the search for interesting equation of states. Note that  the BI model (\\ref{action}) with a constant potential $V=1$ and a time-like derivative of the tachyon field ($\\partial_{\\mu}T \\partial^{\\mu}T < 0$) is equivalent to the Chaplygin gas model that was proposed as a model for both dark matter and dark energy in the present Universe~ \\cite{Kam}.   In what follows we will continue to call the scalar field $T$ a tachyon field, bearing in mind that the connection of this field to the actual string theory tachyon is uncertain and that it might also have application to a broad potential pool of other theories.  Our investigation combines both analytic approximations and ``heavy duty'' field theory lattice simulations \\cite{latticeasy}, which we adopt here for the theory (\\ref{action}). We found that for BI scalars with runaway potentials $V(T)$, i.e. potentials for which $T \\rightarrow \\infty$ over time, the most interesting features of the inhomogeneous solutions of the equations of motion are the formation of caustics and the folding of the hypersurface $T(t,\\vec{x})$.  In section \\ref{eqofmotion2} we consider the equation of motion for the tachyon field in Minkowskii spacetime in $1+1$ dimensions. In this simple situation we are able to clearly show our main results. (In later sections we will show that these same results apply to $3+1$ dimensions and  an expanding universe.) First we consider an exponential potential $V(T) \\sim e^{-T}$, and we assume some initial inhomogeneous realization of the tachyon field $T(x)$.  In cosmology this is usually a realization of a random Gaussian field. For theories where $\\alpha'$ is a fundamental scale, consistency of the effective field theory action approach requires that the gradients (in fundamental units) not be large, $\\nabla_x T \\lesssim 1$. Assuming initial  spatial gradients are not large, we argue analytically and confirm numerically that after a quick, transient regime, the equation of motion reduces to a simple first order Hamilton-Jacobi equation  that describes  the propagation of free massive  wave fronts. We then show that this same approximation holds for a wide class of runaway potentials.  In section \\ref{rootarg} we solve this Hamilton-Jacobi equation  exactly using the method of characteristics and find that the solution quickly develops caustics. We suggest a couple of geometrical techniques to view the time evolution of $T$ during this process. We also solve the original equation of motion numerically for a realization of a Gaussian field and show the formation of caustics from the regions of small spatial gradients, in agreement with the results of our approximation.  In section \\ref{dim} we extend the results on caustics in  the $T(t, \\vec x)$ field to $3+1$ dimensions. The qualitative results are unchanged.    In section \\ref{eqofmotion1} we discuss the cosmological implications of our results. The cosmological applications of a tachyon field $T$ with an exponential potential are two-fold. First, a time dependent field $T$ with a sufficiently rapidly decreasing potential ($T^2V(T)\\to 0$ at $T \\to \\infty$) has a pressureless ``cold dark matter'' equation of state in the late time asymptotics $t \\gg 1$ with its energy density decreasing as $\\varepsilon \\propto a^{-3}$, where $a(t)$ is the scale factor of a Friedmann-Robertson-Walker (FRW) cosmological model of the Universe \\cite{Sen2,FKS}. If the energy density of the rolling tachyon becomes dominant, it could be a dark matter candidate. Second, the tachyon appears in various models of brane inflation. In models with unstable $ D$-branes or annihilating $ D$ and $\\bar D$ branes where the unstable tachyon mode has a ground state at $ T \\to \\infty$, ironically, it is exactly the above property of the tachyon that poses a problem. Indeed, the energy density of the pressureless tachyon almost always ends up dominating \\cite{KL,STW}.    A consistent investigation of the tachyon field in cosmology must include spatial inhomogeneities. The tachyon potential has a maximum with negative curvature. As it rolls from this maximum inhomogeneities will be generated from its vacuum fluctuations, as in the theory of spontaneous symmetry breaking (tachyonic preheating) \\cite{preheating}.  The evolution of linear fluctuations of a rolling tachyon in an expanding universe was considered in \\cite{FKS,SW}. As would be expected, tachyon fluctuations coupled with small metric fluctuations grow with time due to gravitational instability, just like usual pressureless matter. However, the linear stage of tachyon fluctuations very quickly changes to a strongly non-linear stage.  In the usual cold dark matter scenario cosmologists understand well the non-linear bottom-up evolution of gravitational instability.  In particular, the theory of non-linear structure formation includes caustics and multi-streaming of collisionless dark matter particles. (For CDM, however, early small scale caustic formation is a  transient phenomenon).  Many qualitative aspects of this theory can be understood even without expansion of the universe in the much simpler model of media of cold collisionless particles with a smooth initial potential velocity field \\cite{ASZ}.  We shall address a similar problem for  the tachyon model (\\ref{action}).  Thus, to pursue tachyon cosmology further, we have to understand the non-linear evolution of inhomogeneities, i.e. the generic solution of (\\ref{action}). In section \\ref{eqofmotion1} we write down the equation of motion for the tachyon field in an expanding universe and recall the results of \\cite{FKS} regarding linear growth of tachyon fluctuations. We develop an analysis of non-linear inhomogeneities of the tachyon in an expanding universe. Again, the equation for the tachyon field is reduced to the covariant form of the Hamilton-Jacobi equation.  We argue that the fully non-linear evolution of tachyon dark matter leads to the formation of caustics and multi-valued regions beyond them, similar to what we derived in the simpler case of sections \\ref{eqofmotion2}-\\ref{dim} without expansion of the universe.  In section \\ref{discussion} we discuss the issue of how to physically interpret the development of caustics. Formally, caustics border the regions where the solution becomes multi-valued. The physical interpretation of multi-valued solutions and caustics depends on the scalar field theory  specified by  the potential $V(T)$ and the dimensions of space-time. Second order and higher derivatives blow up on caustics. Therefore the action (\\ref{action}), which is supposed to work in the approximation of truncation of high order derivatives    , is no longer a valid approximation to string theory tachyon dynamics once caustics have formed. It is unclear how to work with multi-valued solutions $T$ in the four dimensional theory derived for a single-valued effective field $T$.   Also, the interaction of the scalar BI field with other fields (like  the dilaton or gauge fields in the tachyon case) may be important at caustics. We discuss some of the issues involved in making such an interpretation, but we do not intend to resolve the question here. ",
        "conclusions": "\\label{discussion}  We considered field theory with the action (\\ref{action}). At first glance  the equation of motion for  the evolution of  the non-linear inhomogeneous field $T(t, \\vec x)$ looks very complicated. Remarkably, we found that in  flat spacetime for several examples of  runaway potentials $V(T)$ with the ground state at infinity, generic solutions of the theory (\\ref{action}) very quickly approach an asymptotic form $T(t, \\vec x) \\to S(t, \\vec x) $, which obeys the Hamilton-Jacobi equation \\begin{equation} \\dot S^2 -({\\vec \\nabla_{\\vec x}}S )^2=1 \\ . \\end{equation} Considering an example of  the theory (\\ref{action}) in an expanding universe with small metric perturbations,  we found that in this more complicated case  the equation of the tachyon  is again very well approximated by the solution of the covariantly generalized Hamilton-Jacobi equation \\begin{equation}\\label{Ham} g^{\\mu\\nu} \\frac{\\partial S}{ \\partial x^{\\mu}} \\frac{\\partial S}{ \\partial x^{\\nu}} =1 \\ . \\end{equation} We conjecture that the solution $S$ of the Hamilton-Jacobi equation (\\ref{Ham}) is an attractor for the equation of motions of the theory (\\ref{action}) with a runaway potentail $V(T)$. It will be interesting to check if this conjecture can be extended to  other non-trivial geometries $g_{\\mu\\nu}$ and how it depends on $V(T)$.  Another question we  would like to address is the character of convergency of the solution of the actual tachyon equation to the approximated function $S$ which obeys (\\ref{Ham}).  Generalizing the convergence of the homogeneous runaway tachyon solution, we may suggest the asymptotic expansion \\begin{equation}\\label{asymp} T = S + f_1 \\, e^{-S}+f_2 \\,  e^{-2S} + ... \\ , \\end{equation} where $f_1, f_2, ...$ are non-exponential functions of $t$ and $x$. The functions $f_1, f_2, ...$ can  be important, when we address the issue about the value of energy density at caustics.    The most important finding of our study is that generic solutions of theory (\\ref{action}) with a runaway potential contain  caustics and multi-valued regions of $T$. The interpretation of the multi-valued field $T$ may depend on the potential $V(T)$ and the dimensionality $D$ of the theory. However, the meaning of the  multi-valued solutions in an effective four dimensional theory with a runaway potential is not clear to us.  The model (\\ref{action}) can be used as an approximation to the string theory tachyon when higher order derivatives can be neglected \\begin{equation}\\label{aaa} S=-\\int d^4x \\sqrt{-g} V(T) \\sqrt{1+\\alpha' \\nabla_{\\mu} T \\nabla^{\\mu}T} + O(\\nabla_{\\mu}\\nabla^{\\mu} T) \\ . \\end{equation} We  found, however,  that the generic solution of the equations where we keep only  low order gradients is not consistent  because at the caustics second and higher derivatives blow up. Therefore the simple theory (\\ref{action}) does not work for the string theory tachyon. As a result, at this point,  we can say little about cosmological  applications of the tachyon.  The situation is different for the theory (\\ref{aaa}) with the potential $V(T)$ which has a ground state at finite $T$. In this case caustics may not develop. However, in this case, as was shown in \\cite{FKS}, the tachyon field model (\\ref{action}) describes neither non-relativistic cold dark matter nor dark energy in the present Universe.  Another aspect that should be included in  a comprehensive  analysis of the tachyon is its interaction with  the dilaton field, gauge fields and  others \\cite{sigma}. It will be interesting to see if the interaction is especially important on caustics or if the backreaction of such  interactions may even affect their formation."
    },
    "0208/astro-ph0208110_arXiv.txt": {
        "abstract": "{We have analysed the full ISO spectrum of the planetary nebula \\object{NGC~6302} in order to derive the mineralogical composition of the dust in the nebula. We use an optically thin dust model in combination with laboratory measurements of cosmic dust analogues.  We find two main temperature components at about 100 and 50 K respectively, with distinctly different dust compositions. The warm component contains an important contribution from dust without strong infrared resonances. In particular the presence of small warm amorphous silicate grains can be excluded.  The detection of weak PAH bands also points to a peculiar chemical composition of the dust in this oxygen-rich nebula. The cool dust component contains the bulk of the mass and shows strong emission from crystalline silicates, which contain about 10 percent of the mass. In addition, we identify the 92 $\\mu$m band with the mineral calcite, and argue that the 60 $\\mu$m band contains a contribution from the carbonate dolomite. We present the mass absorption coefficients of six different carbonate minerals. The geometry of the dust shell around \\object{NGC~6302} is studied with mid-infrared images obtained with TIMMI2. We argue that the cool dust component is present in a circumstellar dust torus, while the diffuse emission from the warm component originates from the lobes. ",
        "introduction": "\\label{sec:intro}  All low and intermediate mass stars end their life on the Asymptotic Giant Branch (AGB) by ejecting their entire H-rich envelope through a slowly expanding, dusty wind. After the AGB, the central star quickly increases its effective temperature to values high enough to begin ionizing its AGB ejecta: a planetary nebula (PN) is born. Depending on the mass and thus luminosity of the star, the transition from AGB to PN can go very fast: for the most massive objects, in less than 1000 years ionization of the nebula begins. These massive objects therefore are characterized by dense AGB remnants and a very hot luminous central star.  \\object{NGC~6302} is probably one of the best studied PNe with a massive progenitor. A recent determination of the mass of the ionized nebula is about 2~$M_{\\odot}$ \\citep{PB_99_NGC6302}, based on a distance determination of 1.6 kpc \\citep{GRM_93_ngc6302}. Dust mass estimates also indicate a very massive shell ($M_{\\mathrm{d}} = 0.02 \\, M_{\\odot}$), using a distance of 2.2 kpc \\citep{GMR_89_NGC6302}. These distance determinations rely on VLA observations of the expansion of the nebula over a 2.75 yr period, assuming an outflow velocity of 13 km s$^{-1}$. In principle, this is a very reliable method, however the epoch over which the nebula is observed, is very short.  Therefore the increase in size is very small and hard to measure, indicating that the error bars on these distance determinations are still very large.  Instead, we adopt a distance of 0.91 kpc, based on emission-line photometry of \\object{NGC~6302}, from which the $B$ and $V$ magnitudes, the luminosity and the distance can be derived \\citep{SK_89_PNe}. The large number of luminous PNe studied gives confidence in the distance determination from this statistical method.  The nebular abundances indicate that \\object{NGC~6302} is a type I nebula, consistent with a massive progenitor. \\citet{CRB_00_ngc6302_ngc6537} estimate that the progenitor mass is 4--5 $M_{\\odot}$.  The morphology of the nebula observed at optical wavelengths is highly bipolar, pointing to non-spherical mass loss on the AGB, resulting in a dusty torus in the equatorial region \\citep{LD_84_ngc6302}. The inclination angle of the system is $\\sim 45^{\\mathrm{o}}$ with respect to the line-of-sight \\citep{B_94_ngc6302}. The temperature of the central star is very high: \\citet{CRB_00_ngc6302_ngc6537} mention a temperature of $\\sim 250\\,000$ K, while \\citet{PBD_96_centralstar} arrive a temperature of $\\sim 380\\,000$ K. Although there is some uncertainty about the distance and therefore masses and luminosities involved, everything points to a rather massive and luminous progenitor.  The Infrared Space Observatory (ISO) spectrum has been presented in several papers \\citep{B_98_LWS_AGB,MLS_01_NGC6302,MWT_02_xsilI} and is characterized by a wealth of narrow solid state features in the 20--70 $\\mu$m spectral range caused by circumstellar dust in the AGB remnant, as well as by strong emission lines from a multitude of fine-structure lines originating from the ionized gas in the nebula. The dust bands have been identified with crystalline silicates and a number of other components \\citep{KTS_00_diopside,MLS_01_NGC6302,MWT_02_xsilII}. \\citet[hereafter referred to as Paper~I]{KJW_02_carbonates} have reported on the detection of carbonates in the dust shell, based on the identification of broad features at $\\sim$60 and $\\sim$92 $\\mu$m.  Optical and near-infrared images have already shown that the dust distribution around \\object{NGC~6302} is rather complex \\citep{LD_84_ngc6302,SCM_92_PNimages}.  The ISO spectroscopy supports this, because a broad dust temperature range is required to explain the shape of the spectral energy distribution. In addition the dust composition seems to be very complex, as there is evidence for a mixed chemistry by the presence of both oxygen-rich dust and carbon-rich dust features, the latter in the form Polycyclic Aromatic Hydrocarbons (PAHs) (see \\citet{MLS_01_NGC6302} for a discussion on the origin of this dichotomy).  In order to reconstruct the mass loss history of \\object{NGC~6302}, including the geometry and composition of the AGB wind, infrared spectroscopy and imaging are needed.  Unfortunately, the ISO data lack spatial information, limiting the analysis to the bulk dust composition.  Mid-infrared imaging can reveal the present-day geometry of the dust envelope, which puts limits on the mass loss history.  This paper is organized as follows: In Sect.~\\ref{sec:TIMMI2} the ground-based mid-infrared imaging and spectroscopy is presented. The ISO spectroscopy is discussed in Sect.~\\ref{sec:ISO}, which describes our laboratory data of carbonates (Sect.~\\ref{sec:carbonates}) and a model fit to the observed spectrum (Sect.~\\ref{sec:modelfit}). In Sect.~\\ref{sec:discussion} we propose a possible geometry of \\object{NGC~6302} and discuss the astronomical relevance of carbonates. Our results are summarized in Sect.~\\ref{sec:summary}. ",
        "conclusions": "\\label{sec:discussion}  \\subsection{Geometry and composition of the circumstellar dust shell} \\label{sec:geometry}  In Sect.~\\ref{sec:fitparameters} we derived that the dust in the circumstellar environment of \\object{NGC~6302} consists of two temperature components. The cold component (30--60 K) has a mass of 0.050 $M_{\\odot}$ and contains silicates, water ice and carbonates, indicative of a oxygen-rich chemistry. The warm component (100--118 K) contains only $1.3 \\cdot 10^{-4}$ $M_{\\odot}$, which is mainly caused by a featureless species like carbon or iron. About 5\\% of the warm component is in the form of silicates. The large difference in chemical composition between the warm and the cold component and the lack of continuity in the temperature distribution suggest that the two components have a different spatial distribution. This then allows a mixed chemistry; regions of C-rich dust could exist in a PN with a strong silicate spectroscopic signature.  The presence of PAHs in \\object{NGC~6302} \\citep[first reported as UIR bands by][]{RA_86_pne} supports the idea of a mixed chemistry, both the warm, diffuse component and the PAHs could reside in the same regions of the nebula.   \\begin{figure} \\resizebox{\\hsize}{!}{\\includegraphics{H3717F10.eps}} \\caption[]{Proposed geometry of NGC 6302. A dense torus of cool dust is present around the central star (light grey) under an inclination of 45$^{\\mathrm{o}}$ with respect to the line-of-sight. More diffuse material in the polar regions, is indicated in dark grey. The torus obscures the polar region on the western side. A part of the torus is located behind the eastern lobe. The emission from the warm dust and PAHs originates from the lobes, as well as scattered light from the central star. The torus contains the cool dust and the OH maser. For clarity the approximate position of the central star is indicated, however in reality it is obscured by the dusty torus and can only be studied in scattered light from the polar regions.} \\label{fig:geometry} \\end{figure}   The large mass of the cold component suggests that it is present in a circumstellar torus, which leaves the polar regions for the warm carbon-rich component. Optical images of \\object{NGC~6302} show that the eastern lobe is brighter than the western lobe \\citep[see Fig.~\\ref{fig:TIMMI2} and also][]{SCM_92_PNimages}, suggesting that the western lobe is partly obscured by the circumstellar torus.  This leads to a geometry as indicated in Fig.~\\ref{fig:geometry}. The dusty torus around the central star is seen under an inclination angle $\\sim 45^{\\mathrm{o}}$, and partially obscures the western lobe. In return, the eastern part of the torus is located behind the eastern outflow. The dusty torus contains a high column density of dust towards the central star, and therefore, most of the dust is effectively shielded from heating by the central star and remains relatively cool. The dust in the polar regions however, is directly radiated by the central star and thus has a higher temperature.  Our observations, as well as previously published work, is in agreement with the proposed geometry. The torus is optically thick in the UV, optical and near-infrared. Therefore, the optical images only show the scattered light from the central star in the polar regions. We have convolved the warm and cool dust spectra modelled in Sect.~\\ref{sec:fitparameters} with the TIMMI2 N11.9- and Q-band transmission profile and corrected for the atmospheric transmission in order to determine the contribution of the dust components in both bands.  We find that in the N11.9-band, only 0.2\\% of the flux comes from the cold component, and consequently 99.8\\% originates from the warm dust component. For the Q-band these numbers are 29.0\\% and 71.0\\% respectively.  Since the peak in intensity in the N-band images and the optical image coincide (see Fig.~\\ref{fig:overlay}) we conclude that indeed the warm dust is located in the polar regions. The Q-band imaging reveals a different picture; one third of the flux originates from the cool dust component, and two third from the warm dust. Both the warm dust in the eastern lobe, and the cool dust in the western part of the torus contribute to the observed image, leading to a more or less spherical structure in this wavelength band (Fig.~\\ref{fig:TIMMI2}).  N-band spectroscopy of the eastern lobe shows PAH emission.  From Fig.~6 of \\citep[][figure is rotated by 180$^{\\mathrm{o}}$]{CRB_00_ngc6302_ngc6537} it is clear that the PAH emission intensity peak coincides with the intensity peak in the optical and N-band, indicating that the PAH emission also arises from the lobes, where UV radiation from the central star can penetrate. The UV radiation field also explains the ionization observed in the lobes (Fig.~\\ref{fig:TIMMI2spec}). \\citet{PCM_99_PN} also notice that the emission of warm dust grains and forbidden line transitions emerges from the same region. The 6cm data presented by \\citet{GMR_89_NGC6302} traces the location H{\\sc ii} region. In their Fig.~3 the region confined by the torus is clearly visible.  On the other hand, the optically thick torus around the central star effectively shields the molecular material in the torus from the high energy radiation. The infrared radiation field in the torus will pump OH molecules present and cause the formation of OH masers.  Indeed, OH maser emission is observed in \\object{NGC~6302}, on the west side of the object \\citep{PPT_88_NGC6302}, indicating that it arises from part of the torus located at the front side of the object. This is consistent with the fact that predominantly the blue-shifted part of the line profile is seen \\citep{PPT_88_NGC6302}.  \\subsubsection{Validity of model assumptions} \\label{sec:assumptions}  The geometry described here is consistent with the assumptions we have used to fit the ISO spectrum, and with the TIMMI2 images. Here we will discuss the validity of the assumptions used in this model.  {\\bf i)} We assumed that the dust is optically thin at large wavelengths. The TIMMI2 images indeed suggest that this is true: the torus is still visible as a dark lane in the N-band images but starts to disappear when \\object{NGC~6302} is observed in the Q-band. At longer wavelengths, where the spectral energy distribution (SED) peaks and the different solid state species are identified, the dust torus will be optically thin.  A simplified calculation verifies this: The dust is thought to be present in a torus, with a scale height $h$ from the equatorial plane and an outer radius $R_{\\mathrm{out}}$. The inner radius is small compared to the outer radius and can be ignored.  The torus is seen under an inclination angle of 45$^{\\mathrm{o}}$ \\citep{B_94_ngc6302}. The length of the line-of-sight through the disk is thus $l = 2h \\sqrt{2}$.  Assuming that the density is constant throughout the disk, we find for the infrared optical depth along this line-of-sight:  \\begin{equation} \\tau_{\\mathrm{IR}} = Q \\,  n \\, \\sigma_{\\mathrm{geom}} \\, 2 h \\sqrt{2} \\end{equation}  where  \\begin{equation} n = \\frac{M_{\\mathrm{d}}}{M_{\\mathrm{grain}}} \\, \\frac{1}{2h \\pi {R_{\\mathrm{out}}}^2} \\end{equation}  By adding all cold dust components listed in Table~\\ref{tab:dustmasses}, the dust contained in the torus is determined to be $M_{\\mathrm{d}} = 0.05$ $M_{\\odot}$.  The mass of a grain follows from the used grain size of 0.1 $\\mu$m, and the average density of silicate grains. Using cgs units, we find that the torus is optically thin in the mid- and far-infrared in case  \\begin{equation} \\frac{Q_{\\mathrm{IR}}}{{R_{\\mathrm{out}}}^2} < 1.1 \\cdot 10^{-36} \\, \\mathrm{cm}^{-2} \\end{equation}  For a typical value of $Q_{\\mathrm{IR}}$ in the mid- and far-infrared of 0.0028 we find that for $R_{\\mathrm{out}} > 5.0 \\cdot 10^{16}$ cm, or $3.4 \\cdot 10^3$ AU, the dusty torus is optically thin at these wavelengths.  This size is not unreasonable for a circumstellar torus, as it is comparable to the sizes of the dust shell shells around AGB stars, the PN progenitors. In addition, we can argue that the density distribution will not be flat, but probably behaves more like a power law.  Then, most lines-of-sight through the torus will be optically thin even in case of a smaller torus. Therefore, we are confident that the torus is indeed optically thin in the mid- and far-infrared.  {\\bf ii)} We assumed that all grains have a size of 0.1 $\\mu$m. In reality, the grains formed in the circumstellar environment will cover a wide range of sizes. However, we believe that we actually get a good estimate of the actual dust masses, because we use the mass absorption coefficients, and is thus independent from grain size. Using the mass absorption coefficients will lead to a direct determination of the dust mass, without calculating the mass of individual grains. The calcite and dolomite data presented in this work are mass absorption coefficients $\\kappa_\\nu$, the data used for diopside, enstatite and forsterite are measured in terms of absorption efficiency $Q_\\nu/a$.  For amorphous olivine, metallic iron and water ice we used optical constants, and derived absorption efficiency for grains with a size of 0.1 $\\mu$m. Only the mass determination of these three species is probably somewhat affected by our assumption on the grain size.  {\\bf iii)} We assumed that all grains are spherical. Again an assumption that is probably not very realistic. The dominant dust component (amorphous olivine) must be in the form of non-spherical particles to explain the infrared-spectrum of AGB stars, and the same is true for metallic iron \\citep{KDW_02_composition}. The similarity in the spectral appearance of the solid state features detected in PNe and their direct progenitors, AGB stars, \\citep[e.g.][]{SKB_99_ohir,MWT_02_xsilI}, indicates that the dust grains in the two types of objects probably have similar properties. Therefore, we may conclude that the grains in \\object{NGC~6302} are very likely not homogeneous spheres, as these particles result in a shift in peak position with respect to non-spherical particles and inhomogeneous spheres.  However, in our model the only purpose of the grain shape is to calculate the total mass of some of the dust components. The effect that grain shape has on light scattering is not taken into account, as our model uses mass absorption coefficients.  {\\bf iv)} We assumed that all grains are of homogeneous composition. Using mass absorption coefficients to calculate the emerging spectrum, we simply derived the total mass of a certain mineral found in the line of sight. It is impossible to discriminate between homogeneous grains, and grains consisting of two or more mineral species, therefore we cannot exclude the presence of grains of mixed composition. For instance, it is well known that volatiles, such as water vapour, condense in layered structures on grains \\citep{JM_85_condensation}. We stress that the derived total masses are independent from the assumed homogeneity of grains.  {\\bf v)} We assumed that the density and temperature gradients are simple power laws of the distance to the central star.  The complex structure of \\object{NGC~6302} indicates that this is probably far besides the truth, but since the optical depth is low, all infrared radiation emitted by the grains is received. Therefore not the density and temperature as a function of distance to the central star, but the mass-temperature relation of the different dust species becomes important, in a similar fashion as described by \\citet{BDV_00_ABAur}. Our simulation of this relation by a power law density and temperature gradient seems to be in agreement with the observed spectrum.  {\\bf vi)} Finally we assumed that all dust species are found in the same temperature range. However, if the different dust components are not in thermal contact with each other, i.e.~if they are present in separate -- but co-spatial -- grain populations, the difference in optical properties, notably in the UV and visual, causes a difference in the temperature profile \\citep{BDV_00_ABAur,KWD_01_xsilvsmdot}. The determination of the temperature range is based on the overall shape of the SED, and we forced the crystalline components -- which are the carriers of the narrow features superposed on the broad continuum -- to have the same temperature profile as the amorphous component. Since the dust emission strongly depends on the temperature of the grains, this will probably be our main source of error in our mass estimates. However, dropping the constraint that all dust species are found in the same temperature range, leads to a large increase in free parameters for our simple model.  Considering all the effects discussed here, we estimate that our absolute mass determinations are accurate within a factor of two, if the distance toward \\object{NGC~6302} is 910 pc. If we take into account the large uncertainty in the distance determinations, the error in the absolute dust mass determination further increases.  The uncertainty in the various laboratory measurements introduces an additional uncertainty of a factor of two, leading to a total error of a factor of $\\sim$3.  This inaccuracy in the laboratory measurements can be inferred from comparison between published data.  The crystalline silicate features published by \\citet{KTS_99_xsils} are intrinsically much stronger (up to 5 times), than the same features in measurements by other groups \\citep{SP_73_mg2sio4,S_74_optprop,JMD_98_crystalline}.  Consequently, the crystallinity, defined by the mass of enstatite and forsterite as a fraction of the total dust mass, lies within 5\\% -- 25\\%, and the values given in Table~\\ref{tab:dustmasses} are thus just lower limits. The error in the relative dust mass fractions, i.e.~the mass ratios between the identified minerals, is much better, as they all depend in the same way on our model assumptions. We therefore estimate the error in the relative dust mass fractions of the order of 50\\%.  The error in the absolute and relative flux levels is small compared to the inaccuracies discussed here, and is therefore not taken into account.  In order to further approach reality, full radiative transfer calculations should be performed.  Using spatial information derived from imaging and measurements of outflow velocities it is possible to reconstruct the geometry of the nebula. With 2-dimensional full radiative transfer calculations over a large wavelength range, from UV to far-infrared, it is possible to resolve the temperature and density distribution throughout the PN, for the different dust components. Also, the effect of grain sizes, grain shapes and mineralogical homogeneity then becomes apparent in the spectroscopy. This will lead to a much more accurate description of the properties of the dust in the nebula. However, we chose not to do these calculations for the following reasons: First, optical constants are not available for some of the identified dust components. Second, our knowledge about the geometry of the nebula is still too limited to justify the huge effort of 2-dimensional radiative transfer calculations. These calculations introduce new free parameters, and it will be very hard to constrain the model parameters in an unambiguous way with the limited information available. Spatially resolved spectroscopy of the mid- and far-infrared regions will certainly improve the image we have of \\object{NGC~6302}. Finally, to obtain reliable mass estimates the distance toward \\object{NGC~6302} should be more accurately known.  \\subsection{The astronomical relevance of carbonates} \\label{sec:literature}  In paper~I and this work, we present the first extrasolar detection of carbonates. The detection is based on far-infrared features of carbonate at 92 $\\mu$m and dolomite, which contributes to the 60 $\\mu$m complex. Due to the low abundances, the strong resonances present at short wavelengths (see Fig.~\\ref{fig:IRtrans}) do not stand out in the spectra, as many other species contribute significantly to the opacity at these wavelengths.  This is not the case at the far-infrared wavelengths, therefore it is possible to detect even very small amounts of carbonates, beyond reasonable doubt.  There have been a number of searches for carbonate features in the mid-infrared. The previously unidentified infrared (UIR) feature at 11.3 $\\mu$m seen in planetary nebula \\object{NGC~7027} was attributed to carbonates \\citep{GFM_73_PNe,BR_75_carbonate}. However, this identification became unlikely with the non-detection of the 6.8 $\\mu$m carbonate feature in \\object{NGC~7027} \\citep{RSW_77_NGC7027}, and the lack of characteristic features due to carbonates in the 22--35 $\\mu$m region of the spectrum \\citep{MFH_78_NGC7027}.  The UIR feature at 11.3 $\\mu$m in \\object{HD~44179}, also known as the Red Rectangle, was attributed to the carbonate magnesite (MgCO$_3$) \\citep{B_77_HD44179}.  However, \\citet{CAT_86_COratio} showed PAHs are the most likely carrier of the UIR feature at 11.3 $\\mu$m seen in various astrophysical environments.  Thus, the sole detection of the 11.3 $\\mu$m feature is no longer considered evidence for the presence of carbonates.  Interesting environments to search for carbonates are star forming regions and young stars. Carbonates form easily on the surface of planets when liquid water is present, but other formation mechanisms cannot be excluded (Paper~I). \\citet{RSP_77_BNKL} have determined an upper limit to the carbonate/silicate mass ratio in the interstellar medium toward Orion. They find that the this mass ratio is at most 0.05, based on the detection limit of the 6.8 $\\mu$m feature. The 6.8 $\\mu$m feature is seen in absorption toward embedded protostars. The resemblance of the spectral appearance of the 6.8 $\\mu$m feature in \\object{W33A} and in Interplanetary Dust Particles (IDPs) has lead to the conclusion that they have the same carrier, most likely carbonates \\citep{SW_85_laboratory,TB_86_IDP}, although \\citet{TAB_84_protostars} argue that it is due to hydrocarbons.  \\citet{C_89_AFGL961} observes the 6.8 $\\mu$m feature towards protostar \\object{AFGL~961} and claims that it is partially due to carbonates along with a contribution of hydrocarbons.  Hints of the 11.3 and 13.4 $\\mu$m carbonate features are also claimed to be present.  The similarity between the 6.8 $\\mu$m features in IDPs and in interstellar lines-of-sight is reported by \\citet{QRB_00_7micron} as well, presumably coming from the same carrier, although carbonates are rejected as a carrier.  From the high resolution ISO SWS spectroscopy it becomes clear that the spectral shape of the 6.8 $\\mu$m absorption feature observed towards protostars cannot be explained by carbonates, but a satisfactory alternative is still lacking \\citep{KTB_01_ice}. Recently, the $\\sim$92 $\\mu$m carbonate feature has been detected towards protostar \\object{NGC~1333-IRAS~4} (Ceccarelli et al., in prep.).  Although extraterrestrial carbonates are quite often found in meteorites and IDPs, they are actually not easy to detect by means of astronomical observations.  The only claim still standing at the moment is the detection of features at 26.5 and 31 $\\mu$m due to dolomite and calcite respectively in the ISO SWS spectrum of \\object{Mars} \\citep{LED_00_Mars}. However, the far-infrared features of dolomite and calcite at $\\sim$60 and $\\sim$92 $\\mu$m are not seen on \\object{Mars} \\citep{BEB_00_marsFIR}.  As pointed out in Paper~I, until now carbonates are believed to be formed through aqueous alteration. In this formation process, carbon dioxide (CO$_2$) dissolves in liquid water and forms carbonate ions (CO$_3$$^{2-}$). If cations like Ca$^{2+}$ or Mg$^{2+}$ are found in the solution as well, carbonates can be formed as a lake sediment when saturation is reached. The presence of carbonates is seen as evidence of planet formation, as an atmosphere containing carbon dioxide is required, as well as liquid water on the surface of the planet. The connection with planet formation is the justification for the search for carbonates in the circumstellar environment of young stars. Carbonates are ubiquitous in our own solar system, as is shown from the composition studies of meteorites and interplanetary dust particles. Carbonate-containing meteorites are considered tracers of the formation history of the solar system \\citep[e.g.][]{EZB_96_aqueous,B_98_aqueous}. However, the detection of carbonates in PNe suggests that alternative formation mechanisms exist and that the presence of carbonates no longer provides direct evidence for planet formation (Paper~I).    We have determined the composition and distribution of the dust in the planetary nebula \\object{NGC~6302}. We found that a warm (100-118 K) and a cool (30-60 K) dust component are present. The cool dust component is located in a circumstellar torus of which the inner part effectively shields the UV/optical and near-infrared radition from the central star.  The torus contains amorphous olivine, forsterite, clino-enstatite, water ice, diopside, dolomite and calcite.  Outside the solar system, the carbonates dolomite and calcite are only seen in \\object{NGC~6302} and \\object{NGC~6537} (see also Paper I), and the formation mechanism of carbonates in these environments is not yet understood.  Diopside is so far only found in \\object{NGC~6302} \\citep{KTS_00_diopside} and in two OH/IR stars \\citep{DDW_00_OHIR}. Water ice, forsterite, enstatite and amorphous olivine are very common in the dust shells of evolved stars \\citep[e.g.][]{MWT_02_xsilII,MWT_02_xsilI,MWT_02_xsilIII}.  The circumstellar torus also contains the OH maser reported by \\citet{PPT_88_NGC6302}.  The optical depth in the polar regions is smaller, and therefore, these regions can be observed in the optical in scattered light from the central star. These regions contain the warm dust component, which thermally emits in the mid-infrared. The N-band images are dominated by thermal emission from the warm dust. From the ISO spectroscopy, constraints on the dust composition could be derived: predominantly metallic iron (or another featureless dust component), amorphous olivine, forsterite and clino-enstatite are detected. For water ice, diopside, calcite and dolomite we have derived upper limits for the mass fraction in the warm component. Only 0.3\\% of the total dust mass is contained in the warm component, which has a different mineralogical composition than the cool component. UV radiation can penetrate the polar regions and thus we are able to explain the observed PAH emission, which is also reported by \\citet{CRB_00_ngc6302_ngc6537}.  In addition, the mass absorption coefficients of carbonate minerals are presented in this work. We studied calcite, dolomite, ankerite, aragonite siderite and magnesite."
    },
    "0208/astro-ph0208326_arXiv.txt": {
        "abstract": "WIMP-nucleon cross sections $\\sigma \\alt 10^{-9}$~pb may be probed by ton-scale experiments with low thresholds and background rates $\\sim 20$ events per year. An array of cryogenic detectors (``CryoArray'') could perform well enough to reach this goal. Sufficient discrimination and background suppression of photons has already been demonstrated.  Reduction of neutron backgrounds may be achieved by siting the experiment deep enough.  Removal of the surface-electron backgrounds alone has not yet been demonstrated, but the reductions required even for this troublesome background are quite modest and appear achieveable. ",
        "introduction": "Direct detection of supersymmetric WIMP dark matter in the coming decade appears possible. As shown in Fig.~\\ref{limitplot}, experiments under construction should probe a large fraction of parameter space allowed by minimal supersymmetric theory and experimental constraints~\\cite{bottino,baltz01,ellis01}. Interestingly, if the sign of $\\mu$, the Higgs mixing parameter in the superpotential, is positive, the WIMP mass must be $< 500$~\\gev, and the WIMP-nucleon cross section must be $\\agt10^{-9}$~pb~\\cite{baltz01,ellis01}, possibly making the entire parameter space accessible to a ton-scale WIMP-detection experiment. CryoArray would be a 1-ton deployment of semiconductor cryogenic detectors, of a type similar to that in use in CDMS II~\\cite{saabthesis}. This experiment would be sensitive to a WIMP-nucleon cross section $\\sigma \\approx 6\\times 10^{-10}$~pb, corresponding to a signal of a few % WIMP interactions % per 100~kg years.  \\begin{figure} \\begin{center} \\includegraphics [width=.403\\textwidth]{Cryo_CrssSecLimitGoal.eps} \\end{center} \\caption{Comparison of theoretical expectations and experimental sensitivities for spin-independent WIMP-nucleon cross section vs WIMP mass. Curves indicate experimental limits for CDMS I~\\cite{r19prlprd} (solid), and projected sensitivities for CDMS II at Soudan (dashed) and for CryoArray (dotted). The region outlined in dashes~\\cite{bottino} and the lightest shaded region~\\cite{baltz01} each shows the results from calculations under an effective supersymmetry theory. The medium-gray and darkest regions~\\cite{ellis01} arise from more constrained frameworks. These and other results and projections are available via an interactive web plotter~\\protect\\cite{dmplotter}. } \\label{limitplot} \\end{figure}  The detectors of CDMS or CryoArray measure phonons and charge carriers separately for each interaction in order to allow rejection of the otherwise dominant electron-recoil background events. The background discrimination of these detectors has already been demonstrated to be so good~\\cite{r19prlprd,saabthesis} that instrumenting a one-ton detector is fully justified. Below, we review in detail the expected performance of CryoArray, and conclude with a short discussion of the challenge to build such an experiment at a reasonable cost.  \\begin{table*}[th] \\caption{ Mean single-detector event rates and counts (\\#) between 15--45~keV recoil energy in the Ge detectors of CDMS I, CDMS II, and CryoArray. Rates are listed in units of mdru ($10^{-3}$~keV$^{-1}$~kg$^{-1}$~day$^{-1}$). Values listed for CDMS I and preliminary rejection efficiencies listed for CDMS II at the Stanford Underground Facility (SUF) have been achieved; values listed for CDMS II at Soudan and for CryoArray (at Soudan or at a potential deeper site) are projections. } \\begin{tabular}{llrrlrrrrl} \\hline &  & & \\multicolumn{1}{c}{Event} & \\multicolumn{1}{c}{Exposure} & \\multicolumn{1}{c}{Raw}  &  & \\multicolumn{1}{c}{After} & \\multicolumn{2}{c}{After}  \\\\ &  & \\multicolumn{1}{c}{Depth} & \\multicolumn{1}{c}{Rate} & \\multicolumn{1}{c}{(1000} & \\multicolumn{1}{c}{Events}  & Rejection & \\multicolumn{1}{c}{Reject} & \\multicolumn{2}{c}{Subtraction} \\\\ &  Site & \\multicolumn{1}{c}{(mwe)} & \\multicolumn{1}{c}{(mdru)} & \\multicolumn{1}{c}{kg day)} & \\multicolumn{1}{c}{(\\#)}  & Efficiency & \\multicolumn{1}{c}{(\\#)} & \\multicolumn{1}{c}{(\\#)} & \\multicolumn{1}{c}{(mdru)} \\\\ \\hline \\multicolumn{5}{l}{Photons}         &        &           &     &    \\\\ \\hline CDMS I    & SUF    &   16 & ~800   &  ~~~0.016 &    384 &   99.96\\% & 0.1 &  0~ &  ~~0 \\\\ CDMS II   & SUF    &   16 & ~800   &  ~~~0.04  &    960 &   99.97\\% & 0.3 &  0~ &  ~~0 \\\\ CDMS II   & Soudan & 2080 & ~260   &  ~~~2.50  &  19500 &   99.97\\% & 6.5 &  5~ &  ~~0.07 \\\\ CryoArray &        &      & ~~13   &  500      & 195000 &   99.97\\% & 65~~  & 15~ &  ~~0.001 \\\\ \\hline \\multicolumn{5}{l}{Electrons}      &        &           &     &   \\\\ \\hline CDMS I    & SUF    &   16 & ~~300   &  ~~~0.016 &    145 &   95.00\\% & 7~~  &  7~ &  15 \\\\ CDMS II   & SUF    &   16 &  ~~80   &  ~~~0.04  &     96 &   95.00\\% & 5~~  &  5~ &  ~~4 \\\\ CDMS II   & Soudan & 2080 &  ~~20   &  ~~~2.50  &   1500 &   95.00\\% & 75~~  & 15~ &  ~~0.2 \\\\ CryoArray &        &      &  ~~~1   &  500      &  15000 &   99.50\\% & 75~~  & 15~ &  ~~0.001 \\\\ \\hline \\multicolumn{5}{l}{Neutrons produced in shield} &     &           &     &   \\\\ \\hline CDMS I    & SUF    &   16 & 2200  &  ~~~0.016 &   1000 &   99.90\\% & 1~~   & 0~ &  ~~0 \\\\ CDMS II   & SUF    &   16 & 1000  &  ~~~0.04  &   1200 &   99.95\\% & 0.5 & 0~ &  ~~0 \\\\ CDMS II   & Soudan & 2080 & ~~0.5   &  ~~~2.50  &     38 &   99.90\\% & 0~~   & 0~ &  ~~0 \\\\ CryoArray & Soudan & 2080 & ~~0.5   &  500      &   7500 &   99.90\\% & 8~~   & 6~ &  ~~0.0004 \\\\ Cryoarray & NUSL   & 4500 & ~0.020 &  500      &    300 &   99.90\\% & 0~~   & 0~ &  ~~0 \\\\ \\hline \\multicolumn{5}{l}{Neutrons produced in cavern rock} &     &             &     &   \\\\ \\hline CDMS I    & SUF    &   16 &  50 &  ~~~0.016 &    24 & $\\sim50$\\% & 12~~ & 8~ &  17 \\\\ CDMS II   & SUF    &   16 &  22 &  ~~~0.04  &    26 & $\\sim50$\\% & 13~~ & 8~ &  ~~7 \\\\ CDMS II   & Soudan & 2080 &  0.22 &  ~~~2.50  &    16 & $\\sim50$\\% & 8~~ & 6~ &  ~~0.08 \\\\ CryoArray & Soudan & 2080 &  0.22 &  500      &  3300 & $\\sim50$\\% & 1650~~ & 80~ &  ~~0.005 \\\\ CryoArray & NUSL   & 4500 &  0.01 &  500      &   150 & $\\sim50$\\% & 75~~ & 18~ &  ~~0.001 \\\\ CryoArray & NUSL   & 7200 & 0.0004 & 500      &     6 & $\\sim50$\\% & 3~~ & 3~ &  ~~0.0002 \\\\ \\hline \\end{tabular} \\label{backgrounds} \\end{table*} ",
        "conclusions": "CryoArray should work well enough to probe WIMP-nucleon cross sections $\\sigma \\alt 10^{-9}$~pb once it is built. The most significant challenge at this point is determining how to build it at a reasonable cost. For CDMS~II, the 42 detectors are being built using university facilities and technicians under the close supervision of CDMS physicists. Detectors are tested at least twice in batches of $\\leq3$ in university dilution refrigerators. This time-consuming and physicist-intensive procedure is not practical for CryoArray's $\\sim$ 500--2000 detectors. An industrial approach is needed.  In order to ``out-source'' this effort, we will need to engineer and specify a stable and reliable high-yield process. Since it is not feasible to test each individual module prior to deployment in the array, the yield must be \\agt90\\%. This goal is ambitious; current detector yields are $\\sim$50\\%. We expect to continue to make gains in understanding pathologies in the processing steps during the remaining CDMS~II construction period, which is scheduled for completion at the end of 2003. In 1--2 years we will have substantially more experience in fabricating and operating CDMS detectors. The knowledge gained will be essential for establishing a realistic plan for building CryoArray and assessing its likely performance. This knowledge will likely point the way to additional laboratory work to be carried out before a full-scale proposal is considered."
    },
    "0208/gr-qc0208011_arXiv.txt": {
        "abstract": "We compute  the overlap function between Post-Newtonian (PN) templates and gravitational signals emitted by   binary systems composed of one neutron star and one point mass, obtained by a perturbative approach. The calculations are performed for different stellar models and for different detectors, to estimate how effectual and faithful the PN templates are, and to establish whether effects related to the internal structure of neutron stars may possibly be extracted  by the matched filtering technique. ",
        "introduction": "The detection of gravitational waves  emitted  during the late inspiral and merger phases of coalescing compact binaries is one of  the main targets of the ground based interferometric  detectors that are currently in the final stage of construction or in the commissioning phase (LIGO, VIRGO, GEO600, TAMA). To detect these signals, a detailed knowledge of the emitted waveforms is fundamental; indeed, the performances of the matched filter which will be used to extract the signal from the detectors noise, largely depend on the capability of the theoretical templates to reproduce the true waveforms.  Although the population of  neutron star-neutron star (NS-NS) binaries is expected to double that of black hole-black hole (BH-BH) binaries, and to be several times larger than that of (NS-BH) \\cite{binaries}, BH-BH binaries having a total mass  of $\\sim(20-40)$ M$_\\odot$ will likely    be detected first by the initial ground-based interferometers, because, due to their larger mass, the signal is more intense in the frequency region where the detectors are more sensitive \\cite{grishchuk}. In the future, however, as the detectors  sensitivity in the high frequency region improves, NS-NS coalescence should become  detectable as well. According to recent investigations \\cite{kojima,holai,tutti1,tutti2}, the signal emitted during the latest phases preceeding coalescence differs from that emitted by two black holes essentially in one respect: the modes of oscillation of the stars  could be ``marginally excited\". A mode is resonantly excited if the system moves on an orbit such that the Keplerian orbital frequency, $\\omega_k,$ is in a definite ratio with the mode frequency $\\omega_i,$ i.e. if $\\ell\\omega_k=\\omega_i,$ where $\\ell$ is the harmonic parameter. In general the frequency of the fundamental mode is too high to be excited directly, because the stars merge before reaching the corresponding  orbit \\cite{holai}; however, the width of the resonance of the fundamental mode (especially for $\\ell=2$) is large enough to allow the mode to be  marginally excited before  the resonant frequency is reached. As a consequence, more energy is  emitted with respect to that due to the orbital motion,  the process of inspiralling is accelerated, and this changes the phase of the emitted signals during the last orbits before merging. This effect is stronger for stiffer equations of state (EOSs), or for low mass NSs, for which the frequency of the  fundamental mode is lower, and the  width of the resonance is larger \\cite{tutti2}. Thus, an accurate detection of the signal emitted in a NS-NS binary coalescence besides probing  the theory of gravity, as BH-BH signals would do, would also give an insight into the equation of state of matter at high density regimes unreachable in a terrestrial laboratory.  The aim of this paper is to investigate whether the templates that are being constructed to extract the signal emitted by  inspiralling binaries  from the detectors noise are well suited to detect a NS-NS coalescence. ",
        "conclusions": "In Fig. \\ref{FIG2} we show the contour levels of constant  ambiguity function, as a function of ${\\cal M}$ and $\\eta$. We compute the {\\it minimax} overlap, which is physically more relevant for detection than the {\\it best} overlap, as explained in \\cite{DIS1}. The {\\it true} waveform is that emitted by the stellar model D excited by  an orbiting point particle of $1.4 M_\\odot$, while the PN template  is Pad\\'e-6. The noise curves are, respectively,  those of VIRGO I, EURO and EURO-Xylophone, indicated as EURO-X. We do not plot the curves for LIGO I, because they are very similar to those of VIRGO I. The true parameters of the  binary system are $\\eta=0.25$ and ${\\cal M}=1.2188 M_\\odot$. Model D is one with a soft EOS, and it is very compact. The frequency of the fundamental mode is too high to be excited at a significant level, and indeed in Fig. \\ref{fig1}b) we see that its GW-luminosity is very close  to that of a black hole. Thus, we expect that in this case both the chirp mass and $\\eta$ will be accurately determined by the PN templates, that are constructed for black hole coalescence. This is confirmed in Fig. \\ref{FIG2} for all the considered  detectors. From the three panels of the figure we see that the most important parameter in determining the overlap function is the chirp mass, which can be inferred with an error smaller than one part in a thousand (notice the scale on the $x$-axis).  This fact was already noted in refs. \\cite{cutlerflanagan,grishchuk}.  The dependence on the symmetric mass ratio is found to be somewhat weaker, and the relative error in its estimation is about 2-3 \\%.  It is interesting to plot a similar figure  for the stellar model B. This model has a stiff EOS, and the marginal excitation of the fundamental mode before merging is visible in Fig. \\ref{fig1}b). From Fig. \\ref{fig3} we see that if the detector is VIRGO, the chirp mass and the mass ratio where  ${\\cal A}$ has a maximum are ${\\cal M}= 1.2185 M_\\odot$ and $\\eta=0.234;$ if the detector is EURO, ${\\cal M}=1.2178  M_\\odot$ and $\\eta=0.225,$ and if the detector is EURO-Xylophone  ${\\cal M}= 1.2161 M_\\odot$ and $\\eta=0.213$. Thus, whereas the chirp mass would still be determined to a very  good accuracy the determination of $\\eta$ would be less accurate if the detectors are EURO-type, i.e. very sensitive at high frequency, and  if the templates remain tuned to the black hole signal. This can be understood also by looking at Table \\ref{table3}, where we give  the values of the  effectualness  for LIGO I, VIRGO I, EURO and EURO-X. The {\\it true} signal is that emitted by the five models of NSs given in Table \\ref{table1}; as a template we use the Pad\\'e-6 approximant (column 1 for each detector), and the signal obtained by integrating the BPT equation for a Schwarzschild black hole perturbed by an orbiting particle (column 2), because the Pad\\'e approximant is nothing but an approximation of this signal; thus we wanted to check what is the change if we use as a template the exact signal emitted by a black hole. We see that the performances of $P$-approximants, as well as those of BH approximants, degrade if the detector is very sensitive in the high frequency region. In this case, the use of templates which account for effects of stellar structure would be needed. In Table \\ref{table4} we show the analogous results for the faithfulness.  The required effectualness threshold  of 0.965 is always achieved in LIGO I for all of the stellar models; for VIRGO I  it is a little lower because the detector is more sensitive at high frequency. Thus, if the coalescing binary system is composed of two neutron stars, both LIGO and VIRGO would be able to detect it by using the standard PN templates, and to determine the masses with a sufficient accuracy, provided the event occurs  close enough to be visible by these instruments.  If the noise curve is that of EURO or EURO-Xylophone, the effectualness is lower and a relevant fraction of events would be missed using the standard PN templates; for instance EURO would miss $\\sim$ 36\\% of the events if the stars have low mass as in the stellar model A, and  $\\sim$ 18\\% if the EOS is that of model B. For EURO-Xylophone it would be worse: $\\sim$ 78\\%  events missed for model A,  $\\sim$ 57\\% for model B and $\\sim$ 33\\% for model C. We would like to emphasize that this difference between neutron star models is what makes the situation more interesting. By constructing  filters which  include resonant effects, we would both increase the chances to detect NS-NS events and  be able to estimate the oscillation frequencies of NSs.  Knowing the mass of the stars, these could be used to infer their radius, as suggested by recent investigations \\cite{asterosysm,vallisneri}, and set constraints on the EOS of nuclear matter in the supranuclear density regime \\cite{lattimer}. In addition, it should be stressed that the imprint that the internal structure of the stars leaves on the GW signal may be enhanced by rotation, the effect of which is to  lower some of the mode frequencies; this would shift  the  effect of the mode excitation toward lower frequencies and amplify the signal in the region where EURO-type detectors  would be more sensitive."
    },
    "0208/astro-ph0208056_arXiv.txt": {
        "abstract": "{As a continuation of our previous work, which concerned the radial abundance distribution in the galactic disc over the distances 4--10 kpc this paper presents the first results on the metallicicty in the outer disc (R$_{\\rm G} > 10$ kpc). Based on high-resolution spectra obtained for 19 distant Cepheids we sampled galactocentric distances from 10 to 12 kpc. Combined with the results of our previous work on the inner and middle parts of the galactic disc, the present data enable one to study the structure of the radial abundance distribution over a large baseline. In particular, we find indications of a discontinuity in the radial abundance distribution for iron as well as a number of the other elements. The discontinuity is seen at a galactocentric distance R$_{\\rm G} = 10$ kpc. This finding supports the results reported earlier by Twarog et al. (\\cite{twaet97}). ",
        "introduction": "As shown in our previous papers (Andrievsky et al. \\cite{andret02a} -- Paper~I, \\cite{andret02b} -- Paper~II) the radial abundance distribution within the region of galactocentric distances from 4 to 10 kpc is best described by two distinct zones. One of them (inner: 4.0 kpc $<$R$_{\\rm G} < 6.5$ kpc) is characterized by a rather steep gradient, while in the mid part of galactic disc (6.5 kpc $<$R$_{\\rm G} < 10.0$ kpc), the distribution is essentially flat (e.g. for iron the gradient is d[Fe/H]/dR$_{\\rm G} \\approx -0.03$ dex/kpc).  As discussed in Paper~I and Paper~II, such a bimodal character in the distribution may result from the combined effect of the radial gas flow in the disc and the radial distribution of the star formation rate. We note here that there are conflicting models of the galactic structure, and that possibly the metallicity gradients can help to decide which are the more likely ones. According to Sevenster (\\cite{sev99a}, \\cite{sev99b}) and others (see references in Paper I) the bar extends its influence to a co-rotation radius at about 4--6 kpc. In contrast, according to Amaral \\& L\\'epine (\\cite{amle97}) and others, the spiral arms extend from the Inner Lindblad Resonance which is at about 2.5 kpc, to the Outer Lindblad Resonance, at about 12 kpc, and the co-rotation of the spiral pattern is close to the solar galactic orbit. In the vicinity of a bar we expect to see elongated orbits of stars, and consequently, a small metallicity gradient. On the other hand, according to L\\'epine, Mishurov \\& Dedikov (\\cite{lmd01}) an interaction between the gas and spiral waves in the disc forces the gas to flow in opposite directions inside and outside the Galactic co-rotation annulus. This mechanism produces a cleaning effect in the middle part of the disc and consequently a flattening of the metallicity distribution. At the same time, a decreased star formation rate  in the vicinity of the galactic co-rotation, where the relative velocity of the spiral arms and of the gas passing through these arms is small, should also result in some decrease in the abundances.  As the next step of this study we have begun to investigate the radial abundance distribution in the outer disc. The region of primary interest is at a galactocentric radius R$_{\\rm G} \\approx 10$ kpc, where according to Twarog et al. (\\cite{twaet97}) there exists a discontinuity in the metallicity distribution. Such a discontinuity can be suspected from earlier works of Janes (\\cite{jan79}), Panagia \\& Tosi (\\cite{pantos81}), and Friel (\\cite{fri95}). However, Twarog et al. (\\cite{twaet97}) were the first to clearly stress this result. Twarog et al. used photometic metallicities (interpreted to imply [Fe/H]) for a large sample of open clusters, and they found that galactic disc breaks into two distinct zones. Between R$_{\\rm G} \\approx 6.5-10.0$ kpc they found a mean iron abundance $<$[Fe/H]$>$ of $\\approx 0$ (i.e., the slope is very small, if present). Beyond R$_{\\rm G} \\approx 10.0$ kpc the mean $<$[Fe/H]$>$ is $\\approx-0.3$. This implies a sharp discontinuity at R$_{\\rm G} \\approx 10$ kpc.   Recently, Caputo et al. (\\cite{capet01}) reported a similar result. Those authors calibrated BVI data for a large sample of galactic Cepheids (galactocentric distances from 6 to 19 kpc) as a function of metallicity using non-linear pulsation models. Their results (although not very reliable on a per star basis) suggest that the derived metallicity distribution in the galactic disc can be represented either by a single gradient of $-0.05$ dex~kpc$^{-1}$, or by a two-zone distribution with a slope of $-0.01 \\pm 0.05$ dex~kpc$^{-1}$ within 10 kpc and $-0.02 \\pm 0.02$ dex~kpc$^{-1}$ in the outer region of the galactic disc. In other words, within each region the metallicity gradient is weak to non-existent, while between these regions a significant change of the metallicity/gradient does occur.  These results require an independent check, as well as further specification of the quantitative characteristics of the radial abundance distribution in the outer part of the galactic disc. With this aim we have organized a separate study of distant Cepheids covering the span of distances from 10 to 12 kpc. A short description of the observational material is given in the next section. ",
        "conclusions": "\\subsection{Iron abundance gradient}  The specific aim of this paper is to extend the current work to greater distances towards the galactic anticenter, and to search for possible irregularities in the abundance distributions. Distances can be separated into three zones: The inner part of the galactic disk (gradient d[Fe/H]/dR$_{\\rm G} \\approx -0.13\\pm 0.03$ dex~kpc$^{-1} $), the mid part of the disk (gradient $\\approx -0.02\\pm 0.01$ dex~kpc$^{-1} $), and a piece of outer disc. For the latter we derive a gradient $-0.06\\pm 0.01$ dex~kpc$^{-1}$ and a mean [Fe/H] $\\approx -0.19\\pm 0.08$ dex. The gradient for each zone was derived from a least-squares fit using the weighted data (for the weights assigned to the stars of the inner and mid parts of the disc see Paper~I and Paper~II). Each star of the present sample was assigned unit weight (W = 1). Two stars (TV Cam and YZ Aur) from the outer zone were excluded from the present determination. For those Cepheids we have analyzed only photographic spectra (see Paper~I), therefore we consider the abundance results (only iron) for TV Cam and YZ Aur as being much less reliable than for other stars where we have the CCD spectra. We also excluded from the statistics the very distant Cepheid EE Mon, and the anomalous Cepheid TZ Mon (about the latter see below). The former will be included in the statistics consideration after we sample the region of galactocentric distances from 12 kpc to 15 kpc (next papers of this series).  The transition zone at 10 kpc can be easily identified in Fig. 1. After this point the metallicity drops by approximately 0.2 dex. All the stars in the bin beyond 10 kpc are iron-deficient. The same result is seen in Fig.3ab of Twarog et al. (\\cite{twaet97}) which shows their open cluster metallicity values as a function of galactocentric radius. It should be noted that the sample of open clusters used by Twarog et al. consists of the clusters with ages spanning from 1 to 5 Gyr. Thus, the youngest clusters used for the gradient study are approximately 10 times older than Cepheids. By comparing the iron abundance gradient from Cepheids with that from open clusters one would, in principle, estimate how the abundance gradient evolved with time. Nevertheless, in practice it is difficult to realize, because any conclusion will suffer from the rather high uncertainty of the open cluster data. The only we can state confidently is that the discontinuity of the metallicity distribution has really survived over several Gyrs, until now.  Using a linear fit to the [Fe/H] data over the entire baseline (4--12 kpc) produces a gradient of $-0.06$ dex~kpc$^{-1}$ which is surprisingly close to the canonical literature value of $-0.06/-0.07$ dex~kpc$^{-1}$ (see Papers I and II and references therein).  \\subsection{Radial distribution of other species}  In this paper we will not discuss in detail the radial distributions of the elements other than iron. This will be done in a future paper using the whole observed baseline 4--15 kpc after the region of galactocentric distances between 12 and 15 kpc is well sampled.  As is well known, the carbon abundance is altered during the Cepheid evolution, specifically as a result of the first dredge-up. Therefore we do not attach a great significance to the radial distribution obtained for this element. Nevertheless, interestingly enough, the [C/H]--R$_{\\rm G}$ dependence is rather clear. Two remarkable features inherent to carbon abundance distribution should be stressed. First, the extremely low carbon content in FN Aql (see Paper~I and Fig. 2, R$_{\\rm G} \\approx 6.9$ kpc), and second, the well expressed \"dip\" in carbon abundance at galactocentric distances near 10 kpc. The presence of a coherent behavior in carbon can be interpreted as the original carbon abundances show a similar gradient, and as the mixing event at the first dredge-up modifies all of the abundances by similar amounts; i.e., that the mixing process has similar efficiency (depth) for all stars in this mass range.  We believe that rather large scatter in the radial distribution of oxygen abundance at distances 10--12 kpc is probably caused by some unreliability in oxygen abundance derived for distant Cepheids. This scatter probably does not reflect the real inhomogeneity of oxygen abundance in the interstellar media of the outer disc. The same can be said about elements such as Sc, Cu, Zn (only a few spectral lines for each of these elements including O were analyzed). Those elements whose abundances were based on the analysis of a significant number of lines (e.g. Si, Ti, V, Ni), show rather tight distributions.  The step distribution and apparent discontinuity (similar to iron) is also seen for Si, Ti, Mn, Co, Ni, Zr and Nd.  \\subsection{A possible explanation for the observed features}  If we consider the ensemble of our results, paying special attention to the elements for which the data present smaller scatter because they present a large number of lines in the spectra, like Si, Ti, V, Ni, in addition to Fe and C, it is very clear that the metallicity distribution is flat between about 6 and 10 kpc, and that there is a discontinuity of about $-0.2$ dex for R$_{\\rm G} > 10$ kpc, and perhaps a minimum at about 10.5--11.0 kpc which can be traced from some distributions (let us recall that we adopt R$_{\\odot} = 7.9$ kpc). This is a new result, although the discontinuity at about 10 kpc was already observed by Twarog et al. (\\cite{twaet97}), in a sample of open clusters. An attractive explanation for the minimum at 10.5 kpc (if it truly exists) is the minimum of star formation rate that is expected at co-rotation, as mentioned in the Introduction.  Recently Mishurov et al. (\\cite{mishet02}) produced a detailed model of chemical evolution of the galactic disc, taking into account the effect of co-rotation, to explain our data presented in Papers I and II. The new data presented in this paper, suggest that the model of Mishurov et al. (\\cite{mishet02}) is basically correct, but that the co-rotation radius should be slightly shifted to about 10.5 kpc. The discontinuity in the metallicity distribution at 10 kpc is possibly explained by the gap in the gas density distribution that is associated with co-rotation (see L\\'epine, Mishurov \\& Dedikov \\cite{lmd01}). If we divide the Galactic disc in a large number of concentric rings, the gas from neighbouring rings tends to mix due to supernova explosions, stellar winds, cloud collisions, etc., that do not respect the frontiers between concentric rings. This mixing is equivalent to a diffusion term, and tends to smooth out metallicity gradients in the gas (and therefore, in recently formed stars). However, the gas density gap associated with co-rotation, which is observed in the 21 cm hydrogen line as discussed by L\\'epine, Mishurov \\& Dedikov (\\cite{lmd01}) is possibly a barrier that avoids contact between the gas at R$_{\\rm G} >$ R$_{\\rm c}$ and R$_{\\rm G} <$ R$_{\\rm c}$ and allows the existence of two distinct zones.   \\subsection{Some notes about TZ Mon}  This star differs from other Cepheids by its very low V$_{\\rm t}$ value (1.5~km~s$^{-1}$, see Table 1) which is not appropriate for supergiants. Nevertheless, only by using this small value can one avoid any dependence between iron abundance from individual iron lines and their equivalent widths. With the adopted microturbulent velocity this distant Cepheid shows solar-like abundances for the great majority of elements. For Mn we detected a strong overabundance. Another interesting feature, which is seen in the TZ Mon spectrum, is the 6707 \\AA~Li~I line. Normally this line is not present in supergiant spectra. The equivalent width of W=10 m\\AA~ results in an absolute lithium abundance for TZ Mon of about 0.84.  This is consistent with Li abundances previously determined for some non-variable G/K supergiants by Luck (\\cite{luck77})."
    },
    "0208/astro-ph0208260_arXiv.txt": {
        "abstract": "In this paper, we limit the stellar content of 13 high-velocity clouds (HVCs; including 1 compact HVC)  using the detection limits of a new survey for resolved Milky Way satellite galaxies in  the Sloan Digital Sky Survey Early Data Release (EDR).  Our analysis is sensitive to stellar associations within the virial radius of the Milky Way that are up to 50$\\times$ fainter than the faintest known Milky Way satellites. Statistically, we find no stellar overdensity associated with any of the clouds.  These non-detections suggest lower limits of M$_{HI}$/$L_V$ that range between $\\sim 0.2$ and $250$, assuming cloud distances of 50 and 300 kpc and using fiducial optical scale lengths of 3$'$ and 7$'$.  We explore the implication of these non-detections on the origin of the HVCs. ",
        "introduction": "For decades, astronomers have struggled to determine the properties of the high-velocity clouds (HVCs), a population of objects identified by  21 cm HI emission  with velocities inconsistent with Galactic rotation.  Recent improvements in available radio survey data (e.g. the HI Parkes All-Sky Survey and the Leiden Dwingeloo Survey; Barnes et al.\\ 2001 and Hartmann \\& Burton 1997 respectively) and  UV instrumentation (e.g. STIS and FUSE), have substantially increased our understanding of these numerous ($\\sim$ 2000 in the South alone, Putman et al.\\ 2002) clouds.  However, the origin of the vast majority of the HVC population has remained controversial.  Several explanations for their presence have been suggested, including: a Galactic fountain; tidal or ram pressure stripping from dwarf galaxies; and low mass dark matter halos predicted by currently favored cosmologies but not observed in optical surveys (Blitz et al. 1999, Braun \\& Burton 2000, among others).  A combination of these, and other, scenarios is likely necessary to explain the wide range of observed HVC properties (see Wakker \\& van Woerden 1997 for overview of suggested origins and discussion).   One of the principle limitations in distinguishing between these scenarios is the unknown distances to the clouds (e.g. Gibson et al. 2001).   The most reliable distance estimates to date have resulted from absorption line studies toward background stars at known distances placing concrete upper and lower limits on the distances to a handful of nearby HVCs (Wakker 2001). Unfortunately, this method cannot be applied to clouds at extragalactic (i.e. $>$ 100 kpc) distances due to a lack of background stars.  Alternatively, if the HVCs host a detectable stellar population, their distances could be measured directly  using the location of the tip of the Red Giant Branch or RR Lyrae stars. Identifying stellar populations associated with HVCs would therefore provide the measurements of both stellar content and distance necessary for constraining their possible origins. In particular, stellar associations would provide strong support for the theory that some HVCs may be extremely low mass galaxies and, as such, may likely host stars (see \\S4.2 for discussion).  Several groups are currently conducting deep, pointed surveys toward compact HVCs (CHVCs) in an attempt to identify resolved stars associated with the clouds (Grebel, Braun \\& Burton 2000, Gibson et al. 2001). Although these observations are excellent probes for stars close to the position of the cloud, there are notable drawbacks to this method. Observations of Local Group dwarf spheroidals (dSphs) suggest that HI associations with dSphs are not necessarily coincident with the galaxies' positions ( Carignan et al 1998, St-Germain et al. 1999, Blitz \\& Robishaw 2000) . Furthermore, although only CHVCs have been targeted in existing searches, Putman \\& Moore (2001) have shown that the CHVC distribution is not consistent with the predicted distribution of dark matter halos around the Milky Way and that other subsets of the HVC family possess properties more similar to those predicted for the halos.  The only existing large-area search for stellar counterparts used digitized POSS plates to examine one square degree areas around 264 clouds (Simon \\& Blitz 2002). Although their survey is sensitive to stellar associations throughout the Local Group, all 264 clouds were classified as non-detections.  However, the survey used a nonuniform data set with a poorly characterized surface brightness limit that is no fainter than the known Local Group dwarf galaxies.  Substantial improvements in survey sensitivity and uniformity can be made using the Sloan Digital Sky Survey (SDSS). SDSS is a unique tool with which to search for stellar counterparts to HVCs, providing the means both to survey sufficiently large areas to detect displaced stars and to efficiently sample a large number of clouds. We are currently using SDSS data to conduct a survey for resolved low surface brightness satellites out to the Milky Way's virial radius (Willman et al. 2002, hereafter Paper I).  Our survey also provides a way to statistically investigate the possibility of stellar associations with the high-velocity clouds on very large scales.  Our detection limits (calculated in Paper I) provide a template with which to interpret the results, whether positive or null, in a quantitative framework.  In this paper, we present an analysis of 325 square degrees of the SDSS Early Data Release (EDR; Stoughton et al. 2002) imaging database. This zero-declination strip contains 13 zero-declination HVCs, as cataloged by Putman et al. (2002).  We briefly summarize our survey technique in \\S\\ref{sec:tech}, discuss the analysis of the SDSS Early Data Release in \\S\\ref{sec:application}, use the analysis results and survey detection limits to constrain the stellar nature of the HVCs in \\S\\ref{sec:constraints}, and discuss our findings in \\S4. ",
        "conclusions": "We have used our non-detections of stars associated with 13 HVCs to place new constraints on their stellar content.  We have shown that many of the 13 clouds have HI masses and (minimum) HI mass-to-light ratios similar to those of known dSphs, if the clouds are closer than 350 kpc.  Also, 3 of the HVCs have minimum M$_{HI}$/L that would place them among the most gas-rich galaxies, if shown to reside in a dark matter halo.  Given our results, and those of Simon and Blitz (2002), stellar populations appear unlikely to be a promising distance determination technique for a large number of HVCs.  However, the continued pursuit of stellar associations with HVCs is a worthwhile endeavor, as strong circumstantial evidence suggests that some fraction, however small, of the HVCs are likely to host a detectable low surface brightness (LSB) galaxy.  As there are only 1-3 known gas-rich Milky Way dSph satellites, the discovery of {\\it any} nearby gas-rich stellar association would substantially increase the sample and provide insights into extremely low-mass galaxy evolution.  We are currently pursuing the prospect that some of our 20 detections may actually be a new LSB HI poor Milky Way satellite.  As our present survey volume is too small to unquestionably contain any of the ``missing'' low mass, dark matter halos predicted by $\\Lambda$CDM cosmology, we cannot place limits on missing galaxy explanations for the HVCs. However, the analysis performed in this work will be applied to survey quality SDSS data as it becomes available, and then correlated with the current and future HVC catalogs of Putman et al. (HIPASS) and de Heij et al. (2002) (Leiden/Dwingeloo HI Survey). The completed Sloan Survey will cover $\\sim 125\\times$ more area than we have presented in this paper, guaranteeing that ``missing'' dark matter halos would be well represented in our survey volume (provided $\\Lambda$CDM is correct).  Our future analysis of the SDSS will enable us to apply the arguments presented here to hundreds of clouds, a sample large enough for us to do a concrete statistical assessment of their nature."
    },
    "0208/astro-ph0208440_arXiv.txt": {
        "abstract": "Analytical expressions for the non-relativistic and relativistic Sunyaev-Zel'dovich effect (SZE) are derived by means of suitable convolution integrals. The establishment of these expressions is based on the fact that the SZE disturbed spectrum, at high frequencies, possesses the form of a Laplace transform of the single line distortion profile (structure factor). Implications of this description of the SZE related to light scattering in optically thin plasmas are discussed. ",
        "introduction": "}  Distortions to the cosmic microwave background radiation (CMBR) spectrum arise from the interaction between the radiation photons and electrons present in large structures such as the hot intracluster gas now known to exist in the universe. Such distortions are only a very small effect changing the brightness of the spectrum by a figure of the order of 0.1 percent. This effect, excluding the proper motion of the cluster, is now called the thermal Sunyaev-Zel'dovich effect ~\\cite{SZ1}~\\cite{SZ2} and its detection is at present a relatively feasible task due to the modern observational techniques available. Its main interest lies on the fact that it provides information to determine important cosmological parameters such as Hubble's constant and the baryonic density ~\\cite{rev}-\\cite{Steen1}.  The kinetic equation first used to describe the SZE was one derived by Kompaneets back in 1956 ~\\cite{Peebles}~\\cite{Komp}-\\cite{Weymann}. This approach implies a photon diffusion description of the effect that works basically when the electrons present in the hot intracluster gas are non-relativistic. Many authors, including Sunyaev and Zel'dovich themselves, were very reluctant in accepting a diffusive mechanism as the underlying phenomena responsible for the spectrum distortion. Later on, Rephaeli and others~\\cite{rel} computed the distorted spectrum by considering Compton scattering off relativistic electrons. Those works show that, at high electron temperatures, the distortion curves are significantly modified. Useful analytic expressions that describe the relativistic SZE can be easily found in the literature ~\\cite{Sazonov-RevF}.  One possible physical picture of the effect is that of an absortion-emission process in which a few photons happen to be captured by electrons in the optically thin gas. An electron moving with a given thermal velocity emits (scatters) a photon with a certain incoming frequency $\\nu _{o}$ and outgoing frequency $\\nu $. The line breath of this process is readily calculated from kinetic theory taking into account that the media in which the process takes place has a small optical depth directly related to the Compton parameter $y$. When the resulting expression, which will be called \\emph{structure factor}, is convoluted with the incoming flux of photons obtained from Planck's distribution, one easily obtains the disturbed spectra. For the non-relativistic SZE, a Gaussian structure factor has been successfully established~\\cite{yo}, but the obtention of a simple analytic relativistic structure factor is a more complicated task~\\cite{Birk}. Nevertheless, this work shows that a relativistic structure factor can be obtained with the help of the expressions derived in Refs.~\\cite {Sazonov-RevF} and by the use of simple mathematical properties of the convolution integrals that describe the physical processes mentioned above. In both cases, extensive use has been made of the intracluster gas.  To present these ideas we have divided the paper as follows. In section II, for reasons of clarity we summarize the ideas leading to the SZ spectrum for a non-relativistic electron gas emphasizing the concept of structure factor. In section III we discuss the relativistic case using an appropriate equation for the structure factor whose full derivation is still pending but leads to results obtained by other authors using much more complicated methods. Section IV is left for concluding remarks. ",
        "conclusions": "It is known that, for the case of a beam of photons of intensity $ I_{o}\\left( \\nu \\right) $ incident on a hot electron gas regarded as an ideal gas in equilibrium at a temperature $T_{e}$, the full non-relativistic SZE distorted spectrum may be computed from the convolution integral given in Eq. (\\ref{trece}), which defines the joint probability of finding an electron scattering a photon with incoming frequency $\\bar{\\nu}$\\thinspace and outgoing frequency $\\nu $ multiplied by the total number of incoming photons with frequency $\\bar{\\nu}$. $1-2y$ is a function that corrects the outgoing photon frequency due to the thermal effect~\\cite{imp1}. $\\frac{1}{ \\sqrt{\\pi }}W\\left( \\nu \\right) $ is the normalizing factor of the Gaussian, $W\\left( \\nu \\right) $ is the width of the spectral line at frequency $\\nu $ and its squared value is defined in Eq. (\\ref{catorce}).  Thus, it is clear that an accurate convolution integral between the undistorted Planckian and a \\emph{regular function} exists and can be obtained from strictly physical arguments in the non-relativistic SZE. It is interesting, however, that the corresponding relativistic expression does not seem to be a simple physical generalization of Eq.~(\\ref{trece}). Yet, Eqs.~(\\ref{diecisiete},\\ref{veintitres}) give the correct mathematical description of the relativistic distortion by means of Dirac delta functions and its derivatives. Two questions can now be posed. One, of a rather mathematical fashion, states under what conditions a structure factor such as that appearing in Eq. (\\ref{diecisiete}) can be written in a representation involving Dirac's delta functions and its derivatives. This subject seems to be related to the mathematical theory of distributions. The second one, of more physical type, concerns with the possible description of the thermal SZE as a light scattering problem in which CMBR photons interact with an electron gas in a thermodynamical limit that allows simple expressions for the structure factors. If this were the case, the thermal relativistic SZE physics would not need of Montecarlo simulations or semi-analytic methods in its convolution integral description.  Emphasis should be made on the fact that the main objective pursued in this work is to enhance the physical aspects of the SZE by avoiding complicated numerical techniques in many cases. This is achieved, as shown, by resorting the concept of structure factors, or scattering laws, widely used in statistical physics.  \\bigskip  \\textbf{APPENDIX}  \\bigskip \\qquad The starting point here is Eq.(\\ref{diecisiete}). In the RJ limit, $I_{oRJ}\\,\\left( \\bar{\\nu}\\right) =\\frac{2kT}{c^{2}}\\bar{\\nu}^{2}$. Introducing $G(\\bar{\\nu},\\nu )=\\delta (\\bar{\\nu}-\\nu (1-ay))$ one obtains \\[ I_{RJ}(\\nu )=\\frac{2kT}{c^{2}}\\int_{0}^{\\infty }\\bar{\\nu}^{2}\\delta (\\bar{\\nu }-\\nu (1-ay))d\\bar{\\nu}=\\frac{2kT}{c^{2}}\\nu ^{2}(1-ay)^{2} \\] keeping terms linear in $y$ we find that \\[ \\frac{I_{RJ}(\\nu )-I_{oRJ}\\,\\left( \\nu \\right) }{I_{oRJ}\\,\\left( \\nu \\right) }=-2ay \\] This result is consistent with Eq. (8) setting $a=1$.  Now, in the Wien limit, $I_{oW}\\,\\left( \\bar{\\nu}\\right) =\\frac{2h}{c^{2}} \\bar{\\nu}^{3}e^{-\\frac{h\\bar{\\nu}}{kT}}$, in this case the introduction of $ \\delta (\\bar{\\nu}-\\nu (1-ay))$ leads to \\[ I_{W}(\\nu )=\\frac{2h}{c^{2}}\\int_{0}^{\\infty }\\bar{\\nu}^{3}e^{-\\frac{h\\bar{ \\nu}}{kT}}\\delta (\\bar{\\nu}-\\nu (1-ay))d\\bar{\\nu}=\\frac{2h}{c^{2}}\\nu ^{3}(1-ay)^{3}e^{-\\frac{h\\nu }{kT}}e^{\\frac{h\\nu }{kT}ay} \\]% Expanding $e^{-\\frac{h}{kT}ay}$ and keeping terms up to first order in $y$ we obtain that% \\[ \\frac{I_{W}(\\nu )-I_{oW}\\,\\left( \\nu \\right) }{I_{oW}\\,\\left( \\nu \\right) } =-6ay+axy \\] Finally, setting $a=1$ and considering $x>>6$ in the Wien limit, we obtain the desired result, Eq. (\\ref{nueve}).  This work has been supported by CONACyT (Mexico), project 41081-F."
    },
    "0208/astro-ph0208489_arXiv.txt": {
        "abstract": "Neutron stars, asteroids, comets, cosmic-dust granules,  spacecraft, as well as whatever other freely spinning body dissipate energy when they  rotate  about  any  axis  different from principal. We discuss the internal-dissipation-caused  relaxation  of  a freely precessing rotator  towards  its minimal-energy mode  (mode that corresponds to the spin about the maximal-inertia axis).  We show  that this simple system contains in itself  some quite unexpected physics.  While the body nutates at some rate,  the internal stresses and strains within the body oscillate at frequencies  both  higher and  (what is especially surprising)  lower than  this rate.  The internal dissipation  takes place  not so much  at the frequency  of nutation but rather at the second and higher harmonics.  In other words, this mechanical system provides  an example  of an  extreme non-linerity. Issues  like chaos and separatrix  also come into play.  The earlier estimates,  that ignored non-linearity,  considerably underestimated the efficiency  of the internal relaxation of wobbling asteroids and comets.  At the same time, owing to the non-linearlity  of inelastic relaxation,  small-angle  nutations  can  persist for very long time spans.  The latter circumstance is important  for  the analysis  and interpretation  of NEAR's data on  Eros'  rotation state.  Regarding the comets,  estimates  show  that  the currently available  angular resolution  of  spacecraft-based instruments  makes it  possible  to observe wobble damping  within year- or  maybe even month-long spans of time. Our review also covers pertinent topics from the cosmic-dust astrophysics;  in particular,  the role played by precession damping in the dust alignment. We show that this damping provides coupling of the grain's rotational and vibrational degrees of freedom; this entails occasional flipping of dust grains due to thermal fluctuations. During such a flip, grain preserves its angular momentum, but the direction of torques arising from H$_2$ formation reverses. As a result, flipping grain will not rotate fast in spite of the action of uncompensated H$_2$ formation torques. The grains get ``thermally trapped,'' and their alignment is marginal. Inelastic relaxation competes with the nuclear and Barnett relaxations, so we define the range of sizes for which the inelastic relaxation dominates. ",
        "introduction": "}  \\subsection{Motivation}   On the 14-th of February 1967 the Soviet Union launched artificial spacecraft Kosmos 142, to carry out some ionospheric research. The sputnik had the shape of a cross constituted by four 15-meter-long rods. A separate container, shaped as a cylinder with hemispheres on its ends, was attached in elastic manner to the cross, in a position orthogonal to its plane. This block had dimensions of about 1.6 $m$ $\\times$ 0.8 $m$, and carried in itself all the scientific equipment. It was connected to the cross frame by a joint, and it turned out that this perpendicular position of the container was not secured with a sufficient strength. The mission planners wanted the satellite to rotate in the plane of the cross at a rate of 2 revolutions per second. At a certain point, when the spacecraft was yet gaining rotation, deformation started. The cylindrical container overpowered the locking device in the joint, and bent towards the plane of the cross-shaped frame. This phenomenon was addressed by Vasil'ev \\& Kovtunenko (1969) who pointed out that the intensity of the effect depends, among other things, upon the angular velocity of rotation of the cross frame. In 22 months after that event, on the 14-th of December 1968, a similar sputnik, Kosmos 259, was launched. Its rotation rate was not so swift: less that one revolution per second. This time no deformations of the spacecraft was observed, and the mission succeeded.  The misadventure of Kosmos 142 resulted from the first principles of mechanics: a freely rotating top must end up in the spin state that minimises the kinetic rotational energy, for a fixed angular momentum. This spin mode can be achieved by one or both of the following means: adjustment of shape or/and alteration of the rotation axis. Since the Russian spececraft was easily deformable, it ``preferred'' the first option. Things would go differently if the satellite's construction were more rigid. The latter effect was observed back in 1958, when the team operating the first American artificial satellite was surprised by some unexpected maneuvres that the spacecraft suddenly began to carry out. The satellite, called Explorer I, was a very elongated body with four flexible antennas on it. After launching and getting to the orbit, it was set to perform steady rotation about its longest dimension. However, the flight operators never managed to keep the spacecraft in the designed spin state: Explorer persistently deviated from the simple rotation and went into a wobble, exhibiting slowly changing complex spin. Naturally, the rotation state was evolving toward that of minimal kinetic energy (the angular momentum being fixed). We would remind that the state of rotation about the maximal-inertia axis is the one minimising the kinetic energy, while spin about the least-inertia axis corresponds to the maximal energy. Hence, one should expect that the body will (through some dissipative processes) get rid of the excessive energy and will change the spin axis.  Another example of unsupported rotator subject to internal dissipation is a cosmic-dust granule. Due to various spin-up mechanisms (the main of which is catalytic formation of $H_2$ moleculae on the granule surface (Purcell 1979)), these particles spend most part of their life in rotation. This circumstance gives birth to a whole sequence of subtle effects, which determine alignment of the dust relative to the interstellar magnetic field. This alignment can be indirectly observed through measuring the polarisation degree of the starlight passing through the dust cloud (Lazarian 2000, 1994). It turns out that theoretical description of alignment in based on one's knowledge of the granules' typical rotation state: it is important whether the dust particles are, predominantly, in their principal spin states or not (Lazarian \\& Efroimsky 1999; Efroimsky 2002).  Similar to spacecraft and interstellar grains, a comet or an asteroid in a non-principal rotation mode will dissipate energy and will, accordingly, return to the stable spin (Prendergast 1958, Burns \\& Safronov 1973, Efroimsky \\& Lazarian 1999). Nevertheless, several objects were recently found in excited states of rotation. These are asteroid 4179 Toutatis (Ostro et al. 1993, Harris 1994, Ostro et al. 1995, Hudson and Ostro 1995, Scheeres et al. 1998, Ostro et al. 1999) and comet P/Halley (Jewitt 1997; Peale \\& Lissauer 1989; Sagdeev et al. 1989; Peale 1991; Wilhelm 1987). Quite possibly, tumbling are also comet 46P/Wirtanen (Samarasinha, Mueller \\& Belton 1996; Rickman \\& Jorda 1998), comet 29P/Schwachmann-Wachmann 1 (Meech et al 1993). The existing observational data on asteroid 1620 Geographos may, too, be interpreted in favour of wobble (Prokof'eva et al. 1997; Prokof'eva et al. 1996; Ryabova 2002).  The dynamics of a freely rotating body is determined, on the one hand, by the initial conditions of the object's formation and by the external factors forcing the body out of its principal spin state. On the other hand, it is influenced by the internal dissipation of the excessive kinetic energy associated with wobble. Two mechanisms of internal dissipation are known. The so-called Barnett dissipation, caused by the periodic remagnetisation, is relevant only in the case of cosmic-dust-granule alignment (Lazarian \\& Draine 1997). The other mechanism, called inelastic relaxation, is, too, relevant for mesoscopic grains, and plays a primary role in the case of macroscopic bodies. Inelastic relaxation results from alternating stresses generated inside a wobbling body by the transversal and centripetal acceleration of its parts. The stresses deform the body, and inelastic effects cause energy dissipation.  The external factors capable of driving a rotator into an excited state are impacts and tidal interactions, the latter being of a special relevance for planet-crossers. In the case of comets, wobble is largely impelled by jetting. Even gradual outgassing may contribute to the effect because a spinning body will start tumbling if it changes its principal axes through a partial loss or redistribution thereof. Sometimes the entire asteroid or comet may be a wobbling fragment of a progenitor disrupted by a collision (Asphaug \\& Scheeres 1999, Giblin \\& Farinella 1997, Giblin et al. 1998) or by tidal forces. All these factors, that excite rotators, compete with the inelastic dissipation that always tends to return the rotator to the minimal-energy state.  Study of comets' and asteroids' rotation states may provide much information about their recent history and internal structure. However, theoretical interpretation of the observational data will become possible only after we understand quantitatively how inelastic dissipation affects rotation. The kinetic energy of rotation will decrease at a rate equal to that of energy losses in the material. Thus, one should first calculate the elastic energy stored in a tumbling body, and then calculate the energy-dissipation rate, using the material quality factor $\\, Q $. This empirical factor is introduced for a phenomenological description of the overall effect of the various attenuation mechanisms (Nowick \\& Berry 1972; Burns 1986, 1977; Knopoff 1963; Goldreich \\& Soter 1965). A comprehensive discussion of the $Q$-factor of asteroids and of its frequency- and temperature-dependence is presented in Efroimsky \\& Lazarian (2000).   \\subsection{Complex Rotation of a Rigid Body}  Our review addresses unsupported rotation of rigid and not-entirely-rigid objects. Stated differently, we intend to describe behaviour of unsupported rotators of two sorts: ideal (i.e., those that are exempt from internal dissipation) and realistic (i.e., those subject to dissipation). While the role of dissipative phenomena in the rotating top has become an issue only less than half a century ago, complex rotation of an ideal (absolutely rigid) top has been on the scientific agenda since, at least, the mid of XVIII-th century. This problem generated some of the major mathematical advances carried out by Jacobi, Poinsot and other eminent scholars. However, the founding father of this line of study was Euler whose first notes on the topic date back to 1750's.  Leonhard Euler, the most prolific scientist of all times, will forever retain an aura of mistery in the eyes of historians. Very few researchers, if any, shared his power of insight and his almost superhuman working ability. His life in science consisted of three major periods: the first Russian period (which began in 1730, when young Euler retired from the Russian navy for the sake of academic career), the Berlin period (that started in 1741, when Euler assumed a high administrative position at the Berlin Academy), and the second Russian period (which began in 1765, when major disagreements with King Frederich the Second moved Euler to accept an invitation from Empress Catherine the Great, to return to St.Petersburg). Each of these three periods in Euler's life was marked by numerous scientific achievements in all areas of mathematics known at that epoch.  One of the fields, that grossly benefitted from Euler's attention during his tenure in Berlin, was mechanics of an unsupported top. Euler wrote ca 1760 his pivotal result on the topic (Euler 1765), his famous equations: \\ba \\frac{d}{dt} \\left(I_i \\; {{\\Omega}}_i\\right) \\; - \\; \\left(I_j \\; - \\; I_k \\protect\\right) \\; \\Omega_j \\; \\Omega_k \\; = \\; \\tau_i \\; \\; \\; , \\label{1.1} \\ea where $\\;I_{1,2,3}\\;$ are the eigenvalues of inertia tensor of the body. The tensor is defined through \\ba I_{ij}\\;\\equiv\\; \\int dm\\,\\left\\{{\\robold}^2 \\delta_{ij}\\;-\\;\\rho_i \\rho_j \\right\\}\\;\\;\\;, \\label{1.2} \\ea $\\;\\robold\\;$ being the position of mass element $\\;dm\\;$ relative to the centre of mass of the body. Equations (\\ref{1.1}) are merely a reformulation of the simple fact that the torque equals the rate of change of the angular momentum. They express this law in {\\bf a} body frame. Among the body frames, there exists one (called principal) wherein the inertia tensor is diagonal. In (\\ref{1.1}), $\\;\\Omega_{1,2,3}\\;$ are the angular-velocity components as measured in that, principal, coordinate system. Quantities $ \\; \\tau_i \\;$ are principal-axes-related components of the total torque acting on the body. As ever, we shall assume without loss of generality that $\\;I_3\\,\\geq\\,I_2\\, \\geq\\,I_1\\;$. Hence the third axis will always be that of major inertia.  In the body frame, the period of angular-velocity precession about the principal axis $\\;3\\;$ is: $\\;\\tau\\;=\\;2\\,\\pi/\\omega\\,.$ Evidently, \\begin{eqnarray} {\\dot{\\Omega}}_i/{\\Omega}_i \\; \\approx \\; {\\tau}^{-1} \\; \\; \\; , \\; \\; \\; \\; \\; {\\dot{I}}_i/{I}_i \\; \\approx \\; {\\tau}^{-1} \\, \\epsilon \\; \\; \\;,\\;\\;\\; \\label{22.3} \\end{eqnarray} $\\epsilon\\;$ being a typical value of the relative strain that is several orders less than unity. These estimates lead to the inequality  $\\; \\dot{I_i} \\, \\Omega_i\\;\\ll \\;I_i\\, \\dot{\\Omega_i}\\;$, thereby justifying the commonly used approximation to Euler's equations\\footnote{~Rigorously speaking, in the case when the approximation (\\ref{22.3}) is not satisfied, not only (\\ref{22.4}) fail but even equations (\\ref{1.1}) must be somewhat amended. The problem arises from the ambiguity in the choice of the body frame. For example, if we prefer to choose the coordinate system wherein the inertia tensor always remains diagonal, then the angular momentum will be different from zero in the body frame (and will be of order $\\epsilon$). If, though, we choose the coordinates in which the angular momentum vanishes, then the inertia matrix will no longer be diagonal. With this choice of the body frame, (\\ref{1.1}) should rather be written down not in terms of $I_i$ but in terms of all $I_{ij}$. We shall not elaborate on this issue in our review.}: \\be I_i \\; {\\dot{\\Omega}}_i \\; - \\; \\left(I_j \\; - \\; I_k \\protect\\right) \\; \\Omega_j \\; \\Omega_k \\; = \\tau_i \\; \\; \\; \\; . \\label{22.4} \\ee Naturally, this elegant system of equations has carried since its birth the name of its author. It took scientists some more  years to understand that the system deserves its given name for one more reason: formulae (1.1) are exactly the Euler-Lagrange equations for the Lagrangian of an unsupported rigid body. Proof of this fact demands a certain effort. A brief (but still not trivial) derivation offered in 1901 by Poincare can be found in the textbook by Marsden (2000).  The Euler equations simplify considerably when two of three moments of inertia $\\;I_i\\;$ are equal. This is called dynamic symmetry, to distinguish it from the full geometric symmetry. Further on, whenever we refer to symmetric top, we shall imply only the dynamic symmetry, not the geometric one. This case was addressed by Euler (1765) himself, and later by Lagrange (1813) and Poisson (1813). For prolate symmetric rotators (i.e., when $\\;I_3\\,=\\,I_2\\,>\\,I_1\\;$), in the absence of external torques, the solution is simple: \\be {\\Omega}_1 \\; \\; = \\; \\; const\\;\\;,\\;\\;\\; {\\Omega}_2 \\; \\; = \\; \\; {\\Omega}_{\\perp} \\sin {\\omega}t~~,~~~ {\\Omega}_3 \\; \\; = \\; \\; {\\Omega}_{\\perp} \\cos {\\omega}t~~,~~~ \\label{1.3} \\ee where $\\omega =(I_1/I_3  - 1) \\Omega_1 $. We see that, from the viewpoint of an observer placed on the rotating body, the vector of inertial angular velocity $\\;\\bf \\vec \\Omega\\;$ describes a circular cone about the minor-inertia axis (1) of the body. So does the angular-momentum vector $\\,\\bf \\vec J\\,$. Both $\\;\\bf \\vec \\Omega\\;$ and $\\,\\bf \\vec J\\,$ precess about the least-inertia axis at the same rate $\\omega =(I_1/I_3  - 1) \\Omega_1 $, though at different angles from the axis.  In an inertial coordinate system, the angular momentum $\\,\\bf \\vec J\\,$ will not precess, because it must keep constant for a free rotator.  Instead, it is the least-inertia axis (1) and the angular velocity $\\;\\bf \\vec \\Omega\\;$ that precess about $\\,\\bf \\vec J\\,$, in an inertial observer's opinion. (For brevity, we denote each vector by one letter, though we, of course, imply that every vector transforms appropriately whenever the coordinate system is changed.)  In the case of oblate (dynamic) symmetry ($\\;I_3\\,>\\,I_2\\,=\\,I_1\\;$) free precession will be expressed, in the body frame, by solution \\be {\\Omega}_1 \\; \\; = \\; \\; {\\Omega}_{\\perp} \\cos {\\omega}t~~,~~~ {\\Omega}_2 \\; \\; = \\; \\; {\\Omega}_{\\perp} \\sin {\\omega}t~~,~~~ {\\Omega}_3 \\; \\; = \\; \\; const \\label{1.4} \\ee where $\\,\\omega = (I_3/I_1 - 1) \\Omega_3\\;$ is the mutual rate of circular precession of $\\;\\bf \\vec \\Omega\\;$ and $\\;\\bf \\vec J\\;$ about the major inertia axis (3).  The general case of $I_3>I_2 \\geq I_1$ is quite involved and demands numerics (see Mitchell \\& Richardson (2001); Richardson \\& Mitchell (1999), and references therein). Still, in the absence of external torques the problem can be solved analytically, and Euler coped with it (Euler 1765), though to that end he had to introduce functions similar to what we now call elliptic integrals. The solution much simplifies when expressed through the elliptic functions of Jacobi {\\textit {sn, cn, dn}}. These were defined and studied by Karl Jacobi (1829) and used by him (Jacobi 1849, 1882) to describe free rotation. Jacobi's functions are generalisations of the trigonometric ones, in the following sense: while for symmetric prolate and oblate rotators the circular precession is expressed by (\\ref{1.3}) and (\\ref{1.4}) correspondingly, in the general case $I_3 \\geq I_2 \\geq I_1$ precession is expressed by very similar formulae that contain Jacobi's finctions instead of {$\\;\\sin\\;$} and {$\\;\\cos\\;$}: \\begin{eqnarray} \\Omega_1\\;=\\;\\gamma\\;\\,{\\it{dn}}\\left(\\omega t , \\; k^2 \\protect\\right)\\;\\;, \\;\\;\\;\\;\\Omega_2 \\; = \\; \\beta \\, \\; sn\\left(\\omega t , \\; k^2 \\protect\\right) \\;\\;,\\;\\;\\;\\;\\Omega_3 \\; = \\; \\alpha\\;\\,{\\it{cn}}\\left(\\omega t,\\;k^2 \\protect\\right)\\;\\;,\\;\\;\\; \\label{1.5} \\end{eqnarray} for $\\;{\\bf{J}}^2 \\; < \\; 2\\;I_2 \\; T_{\\small{kin}} \\; $, and \\begin{eqnarray} \\Omega_1\\;=\\;{\\tilde \\gamma}\\;\\,{\\it{cn}}\\left({\\tilde \\omega} t,\\;{\\tilde k}^2 \\protect\\right)\\;\\;,\\;\\;\\;\\;\\Omega_2 \\; = \\;{\\tilde \\beta}\\,\\;{\\it sn}\\left( {\\tilde \\omega} t ,\\;{\\tilde k}^2 \\protect\\right)\\;\\;,\\;\\;\\;\\;\\Omega_3 \\; = \\; { \\alpha}\\;\\,{\\it{dn}}\\left({\\tilde \\omega} t,\\;{\\tilde k}^2 \\protect\\right) \\label{1.6} \\end{eqnarray} for $\\;\\;{\\bf{\\vec J}}^2\\;>\\;2\\;I_2\\;T_{\\small{kin}}\\;$. Here the precession rate $\\;\\it \\omega\\;$ and the parameters $\\;\\alpha ,\\;\\beta ,\\;{\\tilde \\beta}, \\; \\gamma ,\\;{\\tilde \\gamma}, \\;{\\tilde \\omega},\\;k\\;$ and $\\;{\\tilde k}\\;$ are certain combinations of $\\;I_{1,2,3}, \\;T_{\\small {kin}}\\;$ and $\\;{\\bf \\vec J}^2\\;$. We see that (\\ref{1.5}) is a generalisation of (\\ref{1.3}), while (\\ref{1.6}) is that of (\\ref{1.4}). Solution (\\ref{1.5}) approaches (\\ref{1.3}) in the limit of prolate symmetry, $\\;(I_3\\,-\\,I_2)/I_1\\,\\rightarrow\\,0\\;$, while solution (\\ref{1.6}) approaches (\\ref{1.4}) in the limit of oblate symmetry, $\\;(I_2\\,-\\,I_1)/I_1\\,\\rightarrow\\,0\\;$. This situation is illustrated by Figure 1. To understand this \\begin{figure} \\plotone{Efroimsky_review_1.ps} \\caption{The constant-angular-momentum ellipsoid, in the angular-velocity space. The lines on its surface are its intersections with the kinetic-energy ellipsoids corresponding to different values of the rotational energy. The quasi-stable pole A is the maximal-energy configuration, i.e., the state wherein the body spins about its minimal-inertia axis. The stable pole C symbolises the minimal-energy state, i.e., rotation about the maximal-inertia axis. The angular-velocity vector describes the constant-energy lines, and at the same time slowly shifts from one line to another, approaching pole C. The picture illustrates the case of an elongated body: $I_3 \\stackrel{>}{\\sim}I_2>I_1$. The trajectories are circular near A and remain (in the case of an elongated body) virtually circular almost up to the separatrix. The trajectories will regain a circular shape only in the closemost proximity of C.} \\end{figure} picture, one should keep in mind that two quantities are conserved for a freely-spinning body: the angular momentum \\be {\\bf{\\vec J}}^2 \\;=\\;I_1^2\\,{\\Omega}_1^2\\;+\\;I_2^2\\,{\\Omega}_2^2\\;+\\;I_3^2\\, {\\Omega}_3^2 \\;\\;,\\;\\;\\; \\label{1.7} \\ee (which is conserved exactly) and the kinetic energy \\be T_{\\small{kin}}\\;=\\;\\frac{1}{2}\\;\\left\\{I_1\\,{\\Omega}_1^2\\;+\\;I_2\\,{\\Omega}_2^2 \\;+\\;I_3\\,{\\Omega}_3^2 \\right\\}\\;\\;.\\;\\;\\; \\label{1.8} \\ee (which is conserved only approximately because of the internal dissipation). Expressions (\\ref{1.7}) and (\\ref{1.8}) define ellipsoids in the angular-velocity space $\\;(\\Omega_1,\\,\\Omega_2,\\,\\Omega_3)\\;$. Intersection of these gives the trajectory described by the tip of vector $\\;\\bf \\vec \\Omega\\;$ in the angular-velocity space. On the picture we see the angular-momentum ellipsoid (\\ref{1.7}) with lines marked on its surface. These lines are its intersection with ellipsoids (\\ref{1.8}) appropriate to several different values of energy $\\,T_{\\small{kin}}\\;$. For a fixed value of ${\\bf{\\vec  J}}^2$, i.e., for a fixed angular-momentum surface (\\ref{1.7}), there exists an infinite family of kinetic-energy surfaces (\\ref{1.8}) intersecting with it. The largest surface of kinetic energy (corresponding to the maximal value of $\\;T_{kin}\\;$) will be an ellipsoid that fully encloses our angular-momentum ellipsoid and only touches it in point A and its opposite. Similarly, the smallest surface of kinetic energy (corresponding to minimal $\\;T_{kin}\\;$) will be an ellipsoid escribed by our angular-momentum ellipsoid and only touching it from inside, at point C and its opposite. For a fixed $\\;\\bf \\vec J\\;$, the maximal and minimal possible values of the kinetic energy are achieved during rotations about the minimal-inertia and maximal-inertia axes, appropriately. In the case of a non-dissipative torque-free rotation, the tip of vector $\\;{\\bf{\\vec \\Omega}}\\;$ will be describing, on Fig. 1, a curve along which the angular-momentum and energy ellipsoids intersect (Lamy \\& Burns 1972). Solution (\\ref{1.5}) is valid for higher energies, i.e., from point A through the separatrix. In astronomy such rotations are called LAM (~=~ Long-Axis Modes). Solution (\\ref{1.6}) works for lower energies, i.e., from the separatrix through point C. Such rotations are called SAM (~=~ Short-Axis Modes). Wherever the trajectories on Fig.1, i.e, in the space ($\\,\\Omega_1\\,$, $\\,\\Omega_2\\,$, $\\,\\Omega_3\\,$), are almost circular\\footnote{Be mindful that the trajectory in the space ($\\,\\Omega_1\\,$, $\\,\\Omega_2\\,$, $\\,\\Omega_3\\,$) being almost circular does not necessarily mean that the precession cone of the major-inertia axis about $\\bf \\vec J$ is circular or almost circular.}, the solutions (\\ref{1.5}) and (\\ref{1.6}) may be approximated by (\\ref{1.3}) and (\\ref{1.4}), correspondingly. In the limit of an oblate rotator, the applicability domain of (\\ref{1.5}) will shrink into a point (or, to be more exact, into two points: A and its opposite). Similarly, in the limit of a prolate body, the applicability region of (\\ref{1.6}) will shrink into two points: C and its opposite.   \\subsection{Realistic Rotators}  The formalism developed by Euler and refined by Jacobi might be a perfect tool for description of rotation of asteroids, comets, cosmic-dust granules, spacecraft and whatever other unsupported rigid rotators, if not for one circumstance, inner dissipation. Because of this circumstance, the Euler-Jacobi theory of precession works only for time spans short enough to neglect kinetic-energy losses.  The necessity of internal dissipation follows from the basic principles of mechanics. A freely spinning body of a fixed angular momentum has kinetic energy whose values are constrained to lie within a certain bounded range. Hence, from the physical viewpoint, it is very natural for this body to be seeking ways of relaxation. In other words, the body must ``do its best'' to get rid of the excessive kinetic energy, in order to approach the minimal-energy configuration. Thence the necessity of some dissipation mechanism.  Two such mechanisms are known. One is relevant only for mesoscopic rotators, like interstellar-dust grains, and therefore plays a certain role in the cosmic-dust alignment. This is the Barnett dissipation, a phenomenon called into being by periodic remagnetisation of a precessing paramagnetic body (Lazarian \\& Draine 1997).  The second mechanism, inelastic dissipation, is, too, relevant for mesoscopic grains (Lazarian \\& Efroimsky 1999), and it plays the decisive role in the macroscopic bodies' relaxation. The effect results from the alternating stresses produced inside a wobbling body by the time-dependent acceleration of its parts. The stresses deform the body, and the inelastic effects cause dissipation of the rotational energy.  Dissipation entails relaxation of the precession: the major-inertia axis of the body and its angular-velocity vector $\\;\\bf \\vec \\Omega \\;$ tend to align along the angular momentum $\\bf \\vec J$. In other words, the precession cone described by $\\;\\bf \\vec \\Omega \\;$ about $\\bf \\vec J$ will be narrowing until $\\;\\bf  \\vec \\Omega \\;$ aligns along $\\bf \\vec J$ completely. A simple calculation (Efroimsky 2001, Efroimsky 2000, Efroimsky \\& Lazarian 2000, Lazarian \\& Efroimsky 1999) shows that in this case the major-inertia axis of the body will align in the same direction, so that, from the body-frame viewpoint, $\\;\\bf \\vec \\Omega \\;$ will eventually be pointing along this axis. This configuration will correspond to the minimal kinetic energy, the angular momentum being fixed.  An inertial observer will thus see the unsupported body miraculously changing its rotation axis. This is exactly what happened in 1958 when, to mission experts' surprise, rotating satellite Explorer I changed its rotation axis and went into wobble (Thomson 1961).  This was, probably, the first example of a practical need for a further development of the Eulerian theory of a free top, a development that would address an unsupported top with dissipation. However, Chandrasekhar realised this already in  mid-50s, after having been alerted by Kuiper, and asked a postdoc, Kevin Prendergast, to look into that\\footnote{~The authors are grateful to Tom Gehrels for providing these historical facts.}. The most general question was (and still is): how many asteroids in the Solar System can be in non-principal (i.e., nutating) spin states, and how can this evidence of the impact frequency in the main belt? Prendergast in his paper (1958) implied that it is collisions\\footnote{~The collisions within the main belt became a popular topic much later, in 90-s. See, for example, (dell'Oro, Paolicchi, Cellino, Zappala, Tanga, \\& Michel 2001).} that drive asteroids out of the principal state and make them wobble. An important point made by Prendergast (1958) was generation of the second harmonic in a symmetrical oblate rotator: if a body is precessing at an angular rate $\\;\\omega\\; $, then the dissipation is taking place not only at this frequency but also at double thereof. Prendergast considered only the deceptively simple case of symmetrical rotator, and therefore failed to notice the emergence of harmonics higher than the second. Besides, the mathematical treatment of the problem, offered in his paper, was erroneous in several other aspects. Nevertheless, the fact that he noticed the second harmonic was by itself an important contribution for which Prendergast should be credited. His paper was published much ahead of time and, therefore, was forgotten. Independently from Prendergast, Lazarian and Efroimsky (1999) came across the second harmonic some 40 years later. Generation of the higher harmonics was pointed out only in (Efroimsky 2000). The reason why the important work by Prendergast was not fully appreciated by his contemporaries is that back in 50-s the observational astronomy lacked any reliable data on wobbling asteroids. So, Prendergast's paper went almost unnoticed, and his successors had to start up from scratch.  The interest in the asteroidal precession re-emerged in 70-s, after the publication of the milestone work by Burns \\& Safronov (1973) that suggested estimates for the relaxation time, based on the decomposition of the deformation pattern into bulge flexing and bending, and also on the conjecture that ``the centrifugal bulge and its associated strains wobble back and forth relative to the body as the rotation axis {\\bf $\\; \\bf \\omega\\;$} moves through the body during a wobble period.'' As turned out later, the latter conjecture was a too strong statement, because the inelastic dissipation, for the most part of it, is taking place not near the surface but in the depth of the body, i.e., not right under the bulge but deep beneath it. Thus, the bulge is much like an iceberg tip. This became clear when the distribution of precession-caused stresses was calculated, with improved boundary conditions (Efroimsky \\& Lazarian 2000), (Lazarian \\& Efroimsky 1999)\\footnote{~This topic will be discussed in section 4. Our treatment of stresses, demonstrated there, is not mathematically rigorous either. It is a polynomial approximation which satisfies the boundary conditions only approximately. From the physical viewpoint, it is not worth further refining that treatment, because the slight mishandling of the boundary conditions ``spoils'' the solution much less than the irregularity and inhomogeneity of a the realistic body.}. Burns \\& Safronov's treatment neglected the nonlinearity, i.e., generation of frequencies higher and lower than the nutation rate. The nonlinearity, in fact, is essential. Its neglect leads to a large underestimation of the damping rate, because in many spin states a considerable input comes from the harmonics (Efroimsky \\& Lazarian 2000), (Efroimsky 2000). The neglect of nonlinearity leads to up to a two-order underestimate of the precession-damping rate.  In the same year, Peale published an article dealing with inelastic relaxation of nearly spherical bodies (Peale 1973), and there he did take the second harmonic into account.  In 1979 Purcell addressed a similar problem of interstellar-grain precession damping. He ignored the harmonics and mishandled the boundary conditions upon stresses: in (Purcell 1979) the normal stresses had their maximal values on the free surfaces and vanished in the centre of the body (instead of being maximal in the centre and vanishing on the surfaces). These oversights lead to a several-order underevaluation of the dissipation effectiveness and, thereby, of the relaxation rate.  \\subsection{~~Precession ~damping}  The dynamics of precession relaxation is described by the angular rate of alignment of the maximal-inertia axis (3) along the angular momentum $\\bf \\vec J$, i.e., by the decrease in angle $\\;\\theta\\;$ between these. In the case of oblate symmetry (when $\\;I_3\\;>\\;I_2\\;=\\;I_1\\;$), this angle remains adiabatically unchanged over the precession period, which makes $d{\\theta}/dt$ a perfect measure of the damping rate (Efroimsky \\& Lazarian 2000). However, in the general case of a triaxial body angle $\\;\\theta\\;$ evolves periodically through the precession cycle. To be more exact, it evolves $\\it{almost}$ periodically, and its value at the end of the cycle is only slightly different from that in the beginning of the cycle. The relaxation is taking place through accumulation of these slight variations over many periods. This is called adiabatic regime, i.e., regime with two different time scales: we have a ``fast'' process (precession) and a ``slow'' process (relaxation). Under the adiabaticity assumption, one may average  $\\;\\theta\\;$, or some function thereof, over the precession cycle. Then the damping rate will be described by the evolution of this average. Technically, it is convenient to use the average of its squared sine  (Efroimsky 2000). One can write for a triaxial rotator: \\be \\frac{d\\;<\\sin^2 \\theta >}{dt}\\;=\\;\\frac{d\\;<\\sin^2\\theta>}{dT_{kin}}\\;\\; \\frac{dT_{kin}}{dt}\\;\\;\\;,\\;\\;\\;\\; \\label{2.1} \\ee while for an oblate one the expression will look simpler: \\be \\left(\\frac{d\\, \\theta }{dt}\\right)_{(oblate)}\\;=\\;\\left(\\frac{d\\,\\theta}{ dT_{kin}}\\right)_{(oblate)}\\;\\;\\frac{dT_{kin}}{dt}\\;\\;\\;.\\;\\;\\;\\; \\label{2.2} \\ee The derivatives $\\;d\\;<\\sin^2\\theta>/dT_{kin}\\;$ and $\\;\\left(d\\,\\theta/dT_{kin}\\right)_{(oblate)}\\;$ appearing in (\\ref{2.1}) and (\\ref{2.2}) indicate how the rotational-energy dissipation affects the value of $\\;<\\sin^2 \\theta>\\;$ (or simply of $\\;\\theta\\;$, in the oblate case). These derivatives can be calculated from the equations of motion (see Efroimsky \\& Lazarian (2000) and Efroimsky (2000)). The kinetic-energy decrease, $\\; dT_{kin}/dt\\;$, is caused by the inelastic dissipation: \\be d{T}_{kin}/dt \\; = \\; < d{W}/dt > \\;\\; \\;,\\;\\;\\; \\label{2.3} \\ee $W\\;$ being the energy of the alternating stresses, and $\\;<...>\\;$ denoting an average over a precession cycle. (This averaging is justified within the adiabatic approach. For details see section III below.) Finally, in the general case of a triaxial top, the alignment rate will read: \\be \\frac{d\\,<\\sin^2\\theta>}{dt}\\;=\\;\\frac{d\\,<\\sin^2\\theta>}{dT_{kin}}\\;\\; \\frac{d\\,<W>}{dt}\\;\\;\\;,\\;\\;\\; \\label{2.4} \\ee and for a symmetrical oblate top: \\be \\left(\\frac{d\\, \\theta }{dt}\\right)_{(oblate)}\\;=\\;\\left(\\frac{d\\,\\theta}{ dT_{kin}}\\right)_{(oblate)}\\;\\;\\frac{d\\,<W>}{dt}\\;\\;\\;.\\;\\;\\;\\; \\label{2.5} \\ee Now we are prepared to set out the strategy of our further work. While calculation of $\\;d {\\small \\left. \\langle \\right.} \\sin^2 \\theta {\\small { \\left. \\ket \\right.}} / dT_{kin}\\;$ and $\\;\\left(d\\theta /dT_{kin}\\right)_{oblate}\\;$ is an easy exercise\\footnote{See formula (\\ref{5.18}) below and also formulae (A12 - A13) in Efroimsky 2000.}, our main goal will be to find the dissipation rate $\\;d\\,<W>/dt\\;$. This quantity will consist of inputs from the dissipation rates at all the frequencies involved in the process, i.e., from the harmonics at which stresses oscillate in a body precessing at a given rate $\\, \\omega\\,$. The stress is a tensorial extension of the notion of a pressure or force. Stresses naturally emerge in a spinning body due to the centripetal and transversal accelerations of its parts. Due to the precession, these stresses contain time-dependent components. If we find a solution to the boundary-value problem for alternating stresses, it will enable us to write down explicitly the time-dependent part of the elastic energy stored in the wobbling body, and to separate contributions from different harmonics: \\be <W>\\;=\\;\\sum_{n}\\;\\,<W(\\omega_n)>\\;\\;\\;\\;\\;.\\;\\; \\label{4.9} \\ee $W(\\omega_n)\\;$ being the elastic energy of stresses alternating at frequency $\\,\\omega_n$. One should know each contribution $W(\\omega_n)$, for these will determine the dissipation rate at the appropriate frequency, through the frequency-dependent empirical quality factors. The knowledge of these factors, along with the averages $\\,<W(\\omega_n)>\\,$, will enable us to find the dissipation rates at each harmonic. Sum of those will give the entire dissipation rate due to the alternating stresses emerging in a precessing body.  \\subsection{Inelastic dissipation caused by complex rotation}  Equation (\\ref{4.9}) implements the most important observation upon which all our study rests: generation of harmonics in the stresses inside a precessing rigid body. The harmonics emerge because the acceleration of a point inside a precessin body contains centrifugal terms that are quadratic in the angular velocity $\\bf{ \\vec \\Omega}$. In the simpliest case of a symmetrical oblate body, for example, the body-frame-related components of the angular velocity are given in terms of $ \\,\\sin \\omega t\\,$ and $\\,\\cos \\omega t\\,$ (see formulae (\\ref{1.3}) - (\\ref{1.4})). Evidently, squaring of $\\bf{\\vec \\Omega}$ will yield terms both with $\\,\\sin \\omega t\\,$ or $\\,\\cos \\omega t\\,$ and with $\\,\\sin 2 \\omega t\\,$ or $\\, \\cos 2 \\omega t\\,$. The stresses produced by this acceleration will, too, contain terms with frequency $\\,\\omega t\\,$ as well as those with the harmonic $\\,2\\omega t$. In the further sections we shall explain that a triaxial body precessing at rate $\\,\\omega\\,$ is subject, in distinction from a symmetrical oblate body, to a superposition of stresses oscillating at frequencies $\\;\\omega_n\\,=\\,n\\,\\omega_1\\, $, the ''base frequency'' $\\,\\omega_1\\,$ being lower than the precession rate $\\, \\omega$. The basic idea is that in the general, non-oblate case, the time dependence of the acceleration and stresses will be expressed not by trigonometric but by elliptic functions whose expansions over the trigonometric functions will generate an infinite number of harmonics. In subsection 4.3 we shall explain this in more detail.  The total dissipation rate will be a sum of the particular rates (Stacey 1992) to be calculated empirically. The empirical description of attenuation is based on the quality factor $\\,Q(\\omega)\\,$ and on the assumption of attenuation rates at different harmonics being independent from one another: \\be \\dot{W}\\;=\\;\\sum_{n} \\; \\dot{W}{({\\omega_n})}\\;=\\;-\\;\\sum_{n}\\; \\frac{\\omega_n\\;W_0({\\omega}_n)}{Q({\\omega_n})}\\;=\\;-\\;2\\;\\sum_{n}\\; \\frac{\\omega_n\\;\\,<W({\\omega}_n)>}{Q({\\omega_n})}\\;\\;\\;\\;\\\\ \\label{4.10} \\ee $\\;Q(\\omega)\\;$ being the quality factor of the material, and $\\;W_0({\\omega}_n)\\;$ and $\\;\\,<W({\\omega}_n)>\\,\\;$ being the maximal and the average values of the appropriate-to-$\\omega_n\\;$ fraction of elastic energy stored in the body. This expression will become more general if we put the quality factor under the integral, implying its possible coordinate dependence\\footnote{In strongly inhomogeneous nutating bodies attenuation may depend on location.}: \\be \\dot{W}\\;=\\;-\\;2\\;\\sum_{\\omega_n}\\;\\int\\;dV\\;\\left\\{\\frac{\\omega_n}{ Q({\\omega_n})}\\;\\,\\frac{d\\,<W(\\omega_n)>}{dV}\\;\\right\\}\\;\\;\\;,\\;\\; \\label{4.11} \\ee The above assumption of attenuation rates at different harmonics being mutually independent is justified by the extreme smallness of strains (typically, much less than $\\;10^{-6}$) and by the frequencies being extremely low ($10^{-5}\\,-\\,10^{-3}\\; Hz$). One, thus, may say that the problem is highly nonlinear, in that we shall take into account the higher harmonics in the expression for stresses. At the same time, the problem remains linear in the sense that we shall neglect any nonlinearity stemming from the material properties (in other words, we shall assume that the strains are linear function of stresses). We would emphasize, though, that the nonlinearity is most essential, i.e., that the harmonics $\\;\\omega_n\\;$ come to life unavoidably: no matter what the properties of the material are, the harmonics do emerge in the expressions for stresses. Moreover, as we shall see, the harmonics interfere with one another due to $\\; W\\;$ being quadratic in stresses. Generally, all the infinite amount of multiples of $\\; \\omega_1\\;$ will emerge. The oblate case, where only $\\;\\omega_1\\;$ and $\\;2\\omega_1\\;$ show themselves, is an exception. Another exception is the narrow-cone precession of a triaxial rotator studied in Efroimsky (2000): in the narrow-cone case, only the first and second modes are relevant (and $\\,\\omega_1\\,\\approx \\,\\omega$).  Often the overall dissipation rate, and therefore the relaxation rate is determined mostly by harmonics rather than by the principal frequency. This fact was discovered only recently (Efroimsky \\& Lazarian 2000, Efroimsky 2000, Lazarian \\& Efroimsky 1999), and it led to a considerable re-evaluation of the effectiveness of the inelastic-dissipation mechanism. In some of the preceding publications, its effectiveness had been underestimated by several orders of magnitude, and the main reason for this underestimation was neglection of the second and higher harmonics. As for the choice of values of the quality factor $\\;Q\\,$, Prendergast (1958) and Burns \\& Safronov (1973) borrowed the terrestial seismological data for $Q$. In Efroimsky \\& Lazarian (2000), we argue that these data may be inapplicable to asteroids.  To calculate the afore mentioned average energies $\\,<W(\\omega_n)>\\,$, we use such entities as stress and strain. As already mentioned above, the stress is a tensorial generalisation of the notion of pressure. The strain tensor is analogous to the stretching of a spring (rendered in dimensionless fashion by relating the displacement to the base length). Each tensor component of the stress consists of two inputs, elastic and plastic. The former is related to the strain through the elasticity constants of the material; the latter is related to the time-derivative of the strain, through the viscosity coefficients. As our analysis is aimed at extremely small deformations of cold bodies, the viscosity may well be neglected, and the stress tensor will be approximated, to a high accuracy, by its elastic part. Thence, according to Landau \\& Lifshitz (1976), the components of the elastic stress tensor $\\sigma_{\\it{ij}}$ are interconnected with those of the strain tensor $\\epsilon_{\\it{ij}}$ like: \\be \\epsilon_{ij} \\; \\; = \\; \\; \\delta_{ij} \\; \\; \\frac{Tr \\; \\sigma}{9 \\; K} \\; \\; + \\; \\; \\left( \\; \\sigma_{\\it{ij}} \\; \\; - \\; \\; \\frac{1}{3} \\; \\; \\delta_{ij} \\; \\; Tr \\; \\sigma \\right) \\; \\frac{1}{2 \\; \\mu} \\; \\; \\; ~~~, \\label{4.6} \\ee $\\mu$ and $K$ being the {\\it{adiabatic}} shear and bulk moduli, and $Tr$ standing for the trace of a tensor.  To simplify the derivation of the stress tensor, the body will be modelled  by a rectangular prism of dimensions $\\,2\\,a\\,\\times\\,2\\,b\\,\\times\\,2\\,c\\,$ where $\\,a\\,\\ge\\,b\\,\\ge\\,c$. The tensor is symmetrical and is defined by \\be \\partial_{i}\\sigma_{ij}\\;=\\;\\rho\\;a_j\\;\\;,\\;\\; \\label{4.12} \\ee $a_j$ being the time-dependent parts of the acceleration components, and $\\,\\rho\\,a_j$ being the time-dependent parts of the components of the force acting on a unit volume\\footnote{Needless to say, these acceleration components $\\,a_j\\,$ are not to be mixed with $\\,a\\,$ which is the longest dimension of the prism.}. Besides, the tensor $\\,\\sigma_{ij}$ must obey the boundary conditions: its product by normal unit vector, $\\,\\sigma_{ij}n_j\\,$, must vanish on the boundaries of the body (this condition was not fulfilled in Purcell (1979)).  Solution to the boundary-value problem provides such a distribution of the stresses and strains over the body volume that an overwhelming share of dissipation is taking place not near the surface but in the depth of the body. For this reason, the prism model gives a good approximation to realistic bodies. Still, in further studies it will be good to generalise our solution to ellipsoidal shapes. The first step in this direction has been made by Molina, Moreno \\& Martinez-L{\\'o}pez (2002).  Equation (\\ref{4.12}) has a simple scalar analogue\\footnote{This example was kindly offered to us by William Newman.}. Consider a non-rotating homogeneous liquid planet of radius $\\,R\\,$ and density $\\,\\rho\\,$. Let $\\;g(r)\\;$ and $\\,P(r)\\,$ be the free-fall acceleration and the self-gravitational pressure at the distance $\\,r\\,\\le \\,R\\,$ from the centre. (Evidently, $\\,g(r)\\,=\\,(4/3) \\, \\pi \\, G \\, \\rho \\, r\\,$.) Then the analogue to (\\ref{4.12}) will read: \\be \\rho \\; g(r) \\;=\\;-\\;\\frac{\\partial P(r)}{ \\partial r} \\;\\;\\;\\;\\;\\;,\\;\\;\\;\\; \\label{444} \\ee the expression $\\;\\rho \\, g(r)\\;$ standing for the gravity force acting upon a unit volume, and the boundary condition being $\\;P(R)\\,=\\,0$. Solving equation (\\ref{444}) reveals that the pressure has a maximum at the centre of the planet, although the force is greatest at the surface. Evidently, the maximal deformations (strains) also will be experienced by the material near the centre of the planet.  In our case, the acceleration $\\,\\bf \\vec a\\,$ of a point inside the precessing body will be given not by the free-fall acceleration $\\,g({\\bf{ \\vec r}})\\,$ but will be a sum of the centripetal and transversal accelerations: $\\,{\\bf{ \\vec \\Omega}}\\,\\times\\,( {\\bf{ \\vec \\Omega}}\\,\\times\\,{\\bf{ \\vec r}} )\\,+\\,{\\bf {\\dot{ \\vec \\Omega}}}\\,\\times\\,{\\bf{ \\vec r}}\\;$, the Coriolis term being negligibly small. Thereby, the absolute value of $\\,{\\bf{ \\vec a}}\\,$ will be proportional to that of $\\,\\bf  \\vec r\\,$, much like in the above example. In distinction from the example, though, the acceleration of a point inside a wobbling top will have both a constant and a periodic component, the latter emerging due to the precession. For example, in the case of a symmetrical oblate rotator, the precessing components of the angular velocity $\\,\\bf \\vec \\Omega\\,$ will be proportional to $\\,\\sin \\omega t\\,$ and $\\,\\cos \\omega t\\,$, whence the transversal acceleration will contain frequency $\\,\\omega\\,$ while the centripetal one will contain $\\,2 \\omega$. The stresses obtained through (\\ref{4.12}) will oscillate at the same frequencies, and so will the strains. As we already mentioned, in the case of a non-symmetrical top an infinite amount of harmonics will emerge, though these will be obertones not of the precession rate $\\,\\omega\\,$ but of some different ''base frequency'' $\\, \\omega_1\\,$ that is less than $\\,\\omega.$  Here follows the expression for the (averaged over a precession period) elastic energy stored in a unit volume of the body: \\ba \\frac{d\\;\\,<W>}{dV}\\;=\\;\\frac{1}{2}\\;\\,<\\epsilon_{ij}\\;\\sigma_{ij}>\\;= \\;\\frac{1}{4 \\mu}\\; \\left\\{ \\left(\\frac{2 \\; \\mu}{9\\; K} \\; - \\; \\frac{1}{3} \\protect\\right) \\; \\,<\\,\\left(Tr\\;\\sigma \\protect\\right)^2\\,>\\;+\\;<\\sigma_{ij}\\,\\sigma_{ij}> \\protect\\right\\} \\; = \\nonumber \\\\ \\nonumber \\\\ \\frac{1}{4\\mu}\\;\\left\\{\\,-\\,\\frac{1}{1\\,+\\,\\nu^{-1}}\\;\\,<\\left(Tr\\;\\sigma \\protect\\right)^2>\\,+\\,<\\sigma_{xx}^2>\\,+\\,<\\sigma_{yy}^2>\\,+\\,<\\sigma_{zz}^2 > \\, + \\, 2 \\,<\\,\\sigma_{xy}^2 \\, + \\, \\sigma_{yz}^2 \\, + \\, \\sigma_{zx}^2 \\,> \\protect\\right\\}\\;\\; \\label{4.7} \\ea where $2\\mu/(9K)-1/3= -\\nu/(1+\\nu) \\approx -1/5$, $\\,\\nu$ being Poisson's ratio (for most solids $\\,\\nu \\approx 1/4$). Naturally\\footnote{Very naturally indeed, because, for example, $\\,\\sigma_{xx} \\epsilon_{xx} dV\\,=\\,(\\sigma_{xx}\\,dy\\,dz)(\\epsilon_{xx}\\,dx)\\,$ is a product of the $\\,x$-directed pressure upon the $\\,x$-directed elongation of the elementary volume $dV$.}, the total averaged elastic energy is given by the integral over the body's volume: \\be <W>\\;=\\;\\frac{1}{2}\\;\\int\\;dV\\;\\sigma_{ij}\\;\\epsilon_{ij} \\;\\;\\;\\;,\\;\\;\\; \\label{4.8} \\ee and it must be expanded into the sum (\\ref{4.9}) of inputs from oscillations of stresses at different frequencies. Each term $\\,\\bra W(\\omega_n) \\ket\\,$ emerging in that sum will then be plugged into the expression (\\ref{4.10}), together with the value of $\\,Q\\,$ appropriate to the overtone $\\,\\omega_n$. ",
        "conclusions": "}  {\\it 1.} In the article this far we have described the present situation in the studies of the dynamics of an unsupported top, and some of its applications to tumbling asteroids and comets and to the cosmic-dust particles. We addressed relaxation of excited (out of principal state) rotators through energy dissipation resulting from nutation-caused stresses.  {\\it 2.} In many spin states of an unsupported rotator, dissipation at frequencies different from the nutation frequency makes a major input into the inelastic-relaxation process. These frequencies are overtones of some \"basic\" frequency, that is LOWER than the precession frequency. This is a very unusual example of nonlinearity: the principal frequency (precession rate) gives birth not only to  higher frequencies but also to lower frequencies.  {\\it 3.} In many spin states, the inelastic relaxation far more effective than believed hitherto.  {\\it 4.} However, if the rotation states that are close to the separatrix on Fig.2, the lingering effect takes place: both precession and precession-damping processes slow down. Such states (especially those close to the homoclinic point) may mimic the principal rotation state.  {\\it 5.} A finite resolution of radar-generated images puts a limit on our ability of recognising whether an object is nutating or not. Nutation-caused changes of the precession-cone half-angle may be observed. Our estimates show that we may be very close to observation of the relaxational dynamics of wobbling small Solar System bodies, dynamics that may say a lot about their structure and composition and also about their recent histories of impacts and tidal interactions. Monitoring of a wobbling comet during about a year after it leaves the 3 AU zone will, most probably, enable us to register its precession relaxation.  {\\it 6.} Measurements of the damping rate will provide us with valuable information on attenuation in small bodies, as well as on their recent histories of impacts and tidal interactions  {\\it 7.} Since inelastic relaxation is far more effective than presumed earlier, the number of asteroids expected to wobble with a recognisable half-angle of the precession cone must be lower than expected. (We mean the predictions suggested in (Harris 1994).) Besides, some of the small bodies may be in the near-separatrix states: due to the afore mentioned lingering effect, these rotators may be ``pretending'' to be in a simple rotation state.  {\\it 8.} Though the presently available theory predicts a much higher relaxation rate than believed previously, this high rate may still be not high enough to match the experimentally available data. In the closemost vicinity of the principal spin state the relaxation rate must decrease and the rotator must demonstrate the \"exponentially-slow finish\". Asteroid 433 Eros is a consolidated rotator whose $Q$-factor should not be too low. It is possible that this asteroid was disturbed sometimes in its recent history by the tidal forces. Nevertheless, it shows no visible residual precession. Hence, there may be a possibility that we shall have to seek even more effective mechanisms of relaxation. One such mechanism may be creep-caused deformation leading to a subsequent change of the position of the principal axes in the body.   {\\it 9.} Inelastic coupling of the cosmic-dust grain's rotational and vibrational degrees of freedom influences randomisation of grain axes when an ultrasmall grain absorbs a UV photon. As a result, the microwave dipole emission arising from ultrasmall grains may be polarised, while the near-infrared emission arising from the same grains may be unpolarised.\\\\  ~\\\\  {\\Large{\\underline {\\bf Acknowledgements}}}\\\\ ~\\\\ The authors wish to express their gratitude to the colleagues who participated, through direct collaboration as well as through advice and discussion, in the afore described research. ME would like to deeply thank William Newman for numerous fruitful discussions and for offering several highly illustrative examples that were included in the text. AL has the pleasure to thank Bruce Draine, Roger Hildebrand and John Mathis for stimulating exchanges. VS wants to acknowledge the contribution from Anatoly Neishtadt, Daniel Scheeres and Alexey Vasiliev. The work of AL was supported by the NSF through grant AST-0125544. The work of VS was supported through the NASA JURRISS Grant NAG5-8715 and INTAS Grant 00-221.  \\pagebreak"
    },
    "0208/astro-ph0208395_arXiv.txt": {
        "abstract": "{High energy emission has been discovered serendipitously by the BeppoSAX/PDS telescope in the $\\sim$1.3$^{\\circ}$ field of view around the Piccinotti source H0917-074.  A re-pointing of BeppoSAX/NFI has allowed the association of this emission with the Seyfert 2 galaxy \\mcg which lies within the original HEAO1/A2 error box of H0917-074.  This is the first PDS serendipitous discovery of a Seyfert 2 galaxy and the first detection of \\mcg in the X-ray domain.  The measured 2-10 keV flux of  \\mcg is $\\sim$1 $\\times$ 10$^{-11}$ erg cm$^{-2}$ s$^{-1}$ compatible with the Piccinotti HEAO-1/A2 observation. This is a factor of $\\sim$6 greater than that observed from EXO0917.3-0722, originally suggested as the counterpart of the Piccinotti source.  The 2-10 keV spectrum of \\mcg shows the presence of Fe K$_{\\alpha}$ emission together with an absorption feature at $\\sim$8.7 keV.  At high energies, the Seyfert 2 still dominates and the observed 20-100 keV flux is $\\sim$4 $\\times$ 10$^{-11}$ erg cm$^{-2}$ s$^{-1}$. ",
        "introduction": "The Piccinotti sample (Piccinotti et al. 1982) is to date  the only statistically complete sample of active galactic nuclei (AGN) in the 2-10 keV energy range. It was obtained using data from the A2 experiment on the HEAO-1 satellite which performed a survey of 8.3 sr of the sky (65.5\\% coverage) with $|b| \\ge20^{\\circ}$, at a limiting flux of 3.1 $\\times$ 10$^{-11}$ erg cm$^{-2}$ s$^{-1}$. The sample was composed of 36 AGN: 30 Seyfert galaxies, one starburst galaxy (M82), 4 BL Lac objects and one QSO (3C273). \\\\ The BATSE instrument on board CGRO provided, for the first time, a systematic coverage of the whole sample at higher energies (20-100 keV range, Malizia et al. 1999) and for the brightest sources OSSE observations in the 50-500 keV band have also been performed (McNaron-Brown et al. 1995, Zdziarski et al. 2000). BeppoSAX has observed almost the whole Piccinotti sample in the broad band 0.1-200 keV energy range, with the exception of the three Seyfert galaxies: IIIZW2, MKN590 and NGC3227. \\\\ We have carried out a multiyear project to observe with BeppoSAX-NFI the poorly studied (i.e. fainter) sources of the Piccinotti sample among which is the galaxy H0917-074 which was identified with the QSO \\exo. Thanks to the BeppoSAX-PDS observation we discovered that \\exo is contaminated by another hard X-ray source subsequently identified with the Seyfert 2 galaxy \\mcg, which is the true X-ray emitter detected by the A2 instrument. \\\\ It is worth noting that the high sensitivity of the PDS instrument (Frontera et al. 1997) on board the BeppoSAX satellite has provided the opportunity to increase the number of Seyfert 2s detected up to 100 keV, improving our understanding of the high energy characteristics of this type of object.  BeppoSAX observations of Seyfert 2 galaxies have demonstrated that these objects can be powerful hard X-ray emitters (above 10 keV) even though their 2-10 keV radiation is severely attenuated by absorption in thick material (Bassani et al. 1999). In fact, hard X-ray spectra are probably the best tool to directly measure the absorption affecting Seyfert 2 nuclei.\\\\ In this paper the BeppoSAX broad band spectrum of the first PDS serendipitous discovery of a Seyfert 2 galaxy, \\mcg, is presented. From this study, which is also the first of this source in the X-ray domain, \\mcg turns out to be a Compton thin Seyfert 2 galaxy (N$_{H}\\sim$7 $\\times$ 10$^{22}$ atoms cm$^{-2}$) with a slightly peculiar spectrum. ",
        "conclusions": "The main result of this work is that the X-ray source corresponding to H0917-074 in the only 2-10 keV complete sample of AGN (Piccinotti et al. 1982) is the Seyfert 2 \\mcg instead of the QSO/Seyfert 1 \\exo. \\mcg was never before observed in the X-ray domain and from the present study it turns out to be a Compton thin Seyfert 2 galaxy (N$_{H}\\sim$7 $\\times$ 10$^{22}$ atoms cm$^{-2}$) with an almost standard spectrum. The only peculiarity in this spectrum is an absorption feature at around 8 keV which cannot be explained by the presence of warm material around the source. Our best fit model is a power law absorbed by uniform cold material and reflected from a standard cold disk. \\\\ With this observation the number of Seyfert 2s belonging to the Piccinotti sample grows to 8 out of 30 Seyfert galaxies: all these objects have column densities  N$_{H}>$10$^{22}$ atoms cm$^{-2}$. In addition a few type 1 Seyferts in the Piccinotti sample have N$_{H}$ exceeding this value (NGC4151, Zdziarski et al. 2001; NGC526A, Landi et al. 2001; NGC3783, De Rosa et al. 2002). This sample is complete down to a flux limit of 3.1 $\\times$ 10$^{-11}$ erg cm$^{-2}$ s$^{-1}$ and therefore can be used to estimate with some confidence the percentage of absorbed sources at these high fluxes: out of 31 objects identified with Seyfert/QSO (Malizia et al. 1999 and this paper), 11 are absorbed above 10$^{22}$ atoms cm$^{-2}$, implying that a consistent fraction (30$\\pm$12\\%) of all AGN in the 2-10 keV band are absorbed."
    },
    "0208/astro-ph0208307_arXiv.txt": {
        "abstract": "The remarkable rapid variations in radio flux density and polarization of the quasar PKS 0405-385 observed in 1996 are subject to a correlation analysis, from which characteristic time scales and amplitudes are derived. The variations are interpreted as interstellar scintillations. The cm wavelength observations are in the weak scintillation regime for which models for the various auto- and cross-correlations of the Stokes parameters are derived and fitted to the observations. These are well modelled by interstellar scintillation (ISS) of a $30 \\times 22\\mu$as source, with about 180 degree rotation of the polarization angle along its long dimension. This success in explaining the remarkable intra-day variations (IDV) in polarization confirms that ISS gives rise to the IDV in this quasar. However, the fit requires the scintillations to be occurring much closer to the Earth than expected according to the standard model for the ionized interstellar medium (IISM). Scattering at distances in the range 3-30 parsec are required to explain the observations. The associated source model has a peak brightness temperature near $2 \\times 10^{13}$K, which is about twenty-five times smaller than previously derived for this source.  This reduces the implied Doppler factor in the relativistic jet, presumed responsible to 10-20, high but just compatible with cm wavelength VLBI estimates for the Doppler factors in Active Galactic Nuclei (AGNs). ",
        "introduction": "\\label{sec:intro}    Fluctuations on times of a day or less in the flux density of compact, flat spectrum  extragalactic sources at GHz frequencies were first discovered by Heeschen (1984). This, so called, {\\em flickering} of radio sources was attributed to the flux density scintillation in the interstellar medium of our Galaxy (Heeschen \\& Rickett 1987). The subsequent discoveries of much stronger and more rapid intraday variability (IDV) in a number of AGN (Witzel et al.\\ 1986; Kedziora-Chudczer et al.\\ 1997 [KCJ]; Dennett-Thorpe and de Bruyn 2000a [DTB]; Quirrenbach, et al.\\ 2000; Qian et al.\\ 2000, 2001), ignited a debate as to their origin. Many AGN radio sources were already known to show strong variability, though a thousand times slower, which was interpreted as intrinsic to the source. Therefore intrinsic processes were originally used to explain IDV as well.  However an intrinsic interpretation of strong IDV in extragalactic radio sources leads to a high brightness temperature, far in excess of the $T_{B}= 10^{12}$ K Compton limit for the synchrotron emission (Kellermann \\& Pauliny-Toth 1969). This problem was temporarily solved by postulating strong beamed emission from a relativistic jet in IDV sources (Quirrenbach et al. 1992). Radiation from a relativistic shock in such a jet was subsequently postulated to reduce the bulk Lorentz factor needed to explain the short time scales (Qian et al., 1991). However the IDV of quasar, PKS 0405-385 (KCJ), with 50\\% flux density changes within an hour, would require a Doppler factor, D$\\sim10^{3}$, to explain the inferred variability brightness temperature $T_{B}^{var}\\sim 10^{21}$K of a source as small as $10^{-10}$~arcsec.  The explanation of IDV in terms of relativistic beaming with a Doppler factor higher than $10^{2}$ is difficult due to the very high energy requirements in the source (Begelman, Rees \\& Sikora 1994). In addition such extreme relativistic beaming is not supported by the VLBI observations of superluminal sources (Vermeulen \\& Cohen, 1994; Kellermann, et al., 2000).  Rickett et al.\\ (1995) [R95] analyzed the IDV of one particular well-studied quasar (B0917+624) and concluded that it was due to ISS in the ionized medium of our Galaxy. Narayan (1992) also noted that the intrinsic interpretation of IDV implies sources so small that they should also show strong effects of scintillation at radio frequencies.   Though the brightness temperature derived from ISS models depends on the variability time scale, it also depends critically on the distance ($L$) to the scattering plasma and on the level of intensity modulation observed. In the two extreme IDV sources (PKS 0405-385 and J1819+385) the modulation index (rms/mean flux density) was found to peak at tens of percent at a frequency near 5 GHz (KCJ and DTB). From this one can derive the angular size of the scintillating component as approximately the same as that of the first Fresnel zone at this critical frequency, which divides the strong and weak scattering region. For PKS 0405-385 KCJ estimated the size of the scintillating component, 5 $\\mu $arcsec, corresponding to a brightness temperature $T_{b} > 5\\times 10^{14}$ K, by assuming the distance to the scattering region to be 500 pc (from the model for the interstellar plasma of Taylor and Cordes, 1993 [TC93]) and assuming a velocity of 50 km s$^{-1}$ relative to the line of sight. Though such a brightness is over six orders of magnitude less than derived under an intrinsic interpretation, a Doppler factor of $\\sim 1000$ is still needed, since the brightness deduced by ISS methods scales linearly with $D$.  Thus the problem remains of how to explain such high $D$ values (see Marscher, 1998).  Conclusive evidence for the ISS interpretation of IDV comes from the observations of the time delay in the variability pattern observed in PKS 0405-385 between the ATCA and VLA (Jauncey et al., 2000) and between Westerbork and the VLA for  J1819+385 (Dennett-Thorpe and de Bruyn, 2002). Further, annual changes of the time scale of variability due to orbital motion of the Earth have been observed for sources J1819+385 (Dennett-Thorpe 2000) and B0917+624 (Rickett et al.\\ 2001; Jauncey and Macquart, 2001).  Neither of these phenomena can be explained as intrinsic variation.   One argument against the ISS interpretation of IDV is based on the misconception that scintillations cannot cause variability in the fractional polarized flux density, which is observed in IDV sources. Although a single isolated point source cannot cause rapid changes in the degree or angle of linear polarization, IDV in the polarized flux density can be caused by two or more polarized components with misaligned position angles, at least one of which scintillates (Quirrenbach et al. 1989, R95). Simonetti (1991) also proposed polarized substructure viewed through a thin (0.04 AU) shock with a high plasma density (500 cm$^{-3}$) as the cause of a single rapid rotation in the 6-cm polarization angle from 0917+624 observed during ongoing IDV-ISS.  In this paper we present the linear polarization observations for the quasar, PKS 0405-385 obtained during a period when the source exhibited its most rapid and strongest IDV (Section~\\ref{sec:obs}). We describe the methods of statistical data analysis in Section~\\ref{sec:analysis}. In Section~\\ref{sec:iss} we introduce the theory of weak scintillation, which requires a detailed model for the source brightness distribution and also for the distance, velocity and density spectrum of the scattering plasma. We compare the shape and time scale of the auto-correlation of total intensity with theory, assuming a single Gaussian scintillating component and an extended non-scintillating component. We conclude that the scattering plasma is highly anisotropic and given its velocity relative to the Earth, we obtain a new observational constraint on the effective source size and scattering distance.  In Section~\\ref{sec:iss.stokes} we use ISS theory for a source with a polarized brightness that differs from that of its total brightness to obtain expressions for the mutual cross-correlations between the Stokes parameters ($I,Q,\\& U$).  In Section~\\ref{sec:fitting}, after fixing the scattering distance and axial ratio of the plasma, we fit the theoretical correlations to the observations and obtain a polarization model of the source on the scale of 10 $\\mu$as. Though the modelling is not unique, its success confirms that ISS can indeed explain the remarkable and complex pattern of variability in this source. We discuss the implications for the source and the IISM in Section \\ref{sec:disc}. ",
        "conclusions": "\\label{sec:disc}   \\subsubsection*{Coupling of the Scattering and Source Models}  In Section \\ref{sec:dist.dia} we analyzed weak ISS at 4.8 and 8.6 GHz, caused in a localized layer of scattering at distance $L$ from the Earth, for a single Gaussian source component. A range  $2 \\simless L \\simless 30$ pc matches the observations, with an associated range of source diameters $\\theta_{\\rm mas}$ (and a generally inverse relationship between these two parameters). A major finding of the paper is a reduction in the peak source brightness temperature to $\\sim 2 \\times 10^{13}$ K, which is a factor 25 less than inferred in the earlier analysis of KCJ.  This brings the relativistic bulk Doppler factor for the presumed jet closer to the inferred values from the superluminal motion of VLBI sources. While we assumed that the IISM velocity was that of the LSR (36 km s$^{-1}$ relative to the Earth in June), a velocity of 75 km s$^{-1}$, which is the upper bound from the intercontinental time delay in the ISS, changes the allowed ranges to $8 \\simless L \\simless 100$ pc, and $0.13 \\simgreat \\theta_{8.6, \\rm mas} \\simgreat 0.013$ mas and $2 \\times 10^{12} \\simless T_b \\simless 6 \\times 10^{13}$ K, which is still a factor 8 smaller than inferred by KCJ.   \\subsubsection*{Degree of Polarization}  The observations provide values for $Q_{\\rm rms}$ and $U_{\\rm rms}$, but these were not included in the fitting, initially, since the normalized correlation functions are unaffected by the overall degree of polarization. This was remedied by adding two data points to the computation of the fitting residual $S^2$.  The squared difference between model and observation of $Q_{\\rm rms}$ and $U_{\\rm rms}$ was added to $S^2$, with a weight such that each would be approximately equivalent to one of the auto-correlations. When included in the fit process, this reduced the peak brightness in the best fitting source models, and raised the maximum degree of polarization.   In the final two and three component models of Figures \\ref{fig:2comp.polb} and \\ref{fig:3comp.polb} the peak polarization degree is about 70\\%, which is close to the theoretical maximum from a uniform synchrotron source (see Gardner and Whiteoak 1966). However, this is not a coincidence, rather it is a result of an upper bound being placed on the maximum degree of linear polarization during the fitting process. The degree of polarization might also be affected by the addition of a polarized component, which is too extended to scintillate; depending on its position angle this would either increase or decrease the maximum degree of polarization. However, our model has about 1.3 Jy in a structure that is larger than $\\simgreat 0.2$ mas, for which the peak brightness is too small to change the polarized brightness significantly. The model shown in Figure \\ref{fig:3compx} has structure in $Q$ and $U$ on significantly finer scales than in $I$. Our model is the sum of three circular Gaussian components, creating a somewhat elongated structure. The polarized flux density of each component was constrained to be less than 70\\% of the total flux density of component 1. Though this allows weak Gaussian components with individual degrees of polarization greater than 100\\% (see Table 3), it keeps the polarization brightness less than 70 \\% of the total brightness at all points across the source, consistent with synchrotron theory. With this constraint the two-component model gave close agreement in $Q_{\\rm rms}$ and $U_{\\rm rms}$, while for the three-component model these quantities were about 25\\% below the observed values.   \\subsubsection*{VLBI Observations and Time Evolution of the ISS}  VLB observations of PKS 0405-385 were made in June 1996, revealing two components, barely resolved at 1 mas resolution (Kedziora-Chudczer et al., 2001).  The flux density at 8.4 GHz in the ``core'' was 1.5 Jy with 0.24 Jy in a NW extension. We can ask how such a model might be reconciled with the episode of fast IDV in June 1996. Our results suggest that the IDV was due to 30 by 22 $\\mu$as compact structure of about 0.5 Jy at 8.6 GHz and with $\\sim 1.3$ Jy in a more extended component (jet?), part of which was picked up as a NW extension in the VLB image. From the observed IDV the compact core has a brightness temperature of $\\sim 2\\times 10^{13}$K, which we interpret as $\\sim 3\\times 10^{11}$K Doppler boosted in a relativistic jet. This implies a minimum Doppler factor of about 14, including the $1+z$ factor ($z = 1.285$) in Marscher's (1998) expression. Depending on the angle of the jet to the line of sight this corresponds to an apparent superluminal expansion of about 0.13 mas in two months.  Thus the ultra compact feature causing the ISS may have expanded from 0.03 mas to 0.13 mas, which could have quenched the ISS substantially on the time scale of two months.  Such behaviour differs considerably from the long-lived scintillation seen in the other two very rapid scintillating sources, J1819-4835 (DTB, Bignall et al., 2002)  Subsequently the total flux density of the source almost doubled over 1.5 years indicating further expansion of the previously scintillating component. In a separate paper we model this long term time evolution of the PKS 0405-385 as expanding compact knots in a more extended jet. However, we cannot yet rule out the alternative hypothesis that the changes in IDV are due to changes in the scattering in the local ISM.  \\subsubsection*{The Local ISM}  By greatly reducing the scattering distance, we have greatly reduced the source brightness temperature over that inferred by KCJ. Having replaced a source puzzle by an interstellar puzzle, we now consider the implications for the local interstellar medium (ISM) and look for any corroborating evidence on the line of sight toward the quasar ($l=119^{\\circ}, b=-48^{\\circ}$).  The $SM$ values in the local ISM responsible for the IDV in PKS 0405-385 are $\\sim 3 \\times 10^{-4}$ m$^{-20/3}$ kpc. This is comparable to value in the TC93 Galactic disk, which extends more than 10 times further than our screen distance. With a screen thickness of say 5 pc the local value of the scattering strength parameter ($C_{N0}^2$) must be about 100 times greater than in the TC93 model. This suggests we have detected a remarkable nearby concentration of very irregular plasma.  We compare this with the enhanced scattering at the edge of the ``local bubble'' (Bhat et al.\\ 1998). They model their pulsar observations by an ellipsoidal shell with $SM= 6\\pm3 \\times 10^{-5}$ m$^{-20/3}$ kpc, at 134 pc in the direction toward PKS 0405-385 (including the factor 3 reduction according to appendix C of Rickett et al., 2000). In the set of pulsars that they observed the closest was 45$^{\\circ}$ from our line-of-sight. Thus taking our observations and theirs together requires a much less regular structure than their ellipsoidal model of uniform $SM$, such that the scattering layer toward PKS 0405-385 has a 5 times greater $SM$ and lies at only 25-100 pc from the Earth.  In an approximately orthogonal direction Rickett et al.\\ (2000) deduced a deficit in scattering on the 310 pc line toward PSR B0809+74 and suggested patchiness in the shell of enhanced scattering. We can only conclude that the local IISM is spatially inhomogeneous in its turbulence, and we now consider other evidence on the local ISM toward PKS 0405-385.   The diffuse ionized component of the ISM mapped in the Southern H$_{\\alpha}$ Sky Survey (Gaustad et al. 2001) does not reveal any distinct ionized region in this direction. The possibility of a turbulent stellar wind from a nearby star crossing the line of sight is also excluded on the basis of the SUPERCOSMOS measurements (Miller et al. 1991) and our optical imaging with the 1 m telescope at the Mount Stromlo and Siding Spring Observatory. The images show no bright star closer than 1.5', and although they do show a nearby galaxy at 0.5', the optical spectrum of the quasar shows an absorption feature at $z = 0.8$ (KCJ), but this is too distant to cause the rapid variability.  Various observers have reported observations of the local interstellar medium (see Breitschwerdt et al., 1998). For example, G\\'{e}nova et al.\\  (1998) report measurements of Na I absorption lines over a wide range of Galactic longitudes, from which they deduced the kinematics of several local interstellar clouds.  They identify an interstellar cloud ``P'' with velocity of 13.8 km~s$^{-1}$ (toward $l=225^{\\circ}, b=5.4^{\\circ}$) covering a large solid angle that includes PKS 0405-385.  Since the degree of ionization and distance to this cloud are not clear, we have no grounds for identifying it as being associated with the enhanced scattering.  Radio observations of the Galactic continuum show the presence of various arcs and spurs (e.g.\\ Haslam et al., 1982). Spoelstra (1972) analyzed the radio data in terms of spherical shells and modelled the ``Cetus Arc'' as a sphere subtending an angular radius of 50$^{\\circ}$ with its center 110 pc from the Earth toward $l=110^{\\circ}, b=-30^{\\circ}$.  The spherical shell has a radius of 84 pc and thickness $\\sim 11$ pc. Though Spoelstra did not explicitly suggest it this gives a shortest distance to the inner edge of the shell as 26 pc. The direction toward the quasar is only $15^{\\circ}$ from the center and hence the distance to the shell is $\\sim 27$ pc, which is remarkably close to the 25 pc used in our model.  This could well be a coincidence but nevertheless it points to a possible cause of the enhanced scattering layer needed to explain the rapid ISS observed.  \\subsubsection*{Polarized IDV at Cm-Wavelength}  Since ISS can explain the very rapid polarized IDV in 0405-385, one can ask if ISS can explain the generally slower polarized IDV at centimeter wavelengths from other sources. Q89 reported rapid polarized IDV from quasar 0917+624, which was subsequently explained as ISS by R95 and Qian et al.\\ (2001).  More recently Gabuzda et al.\\ (2000a,b,c) have reported IDV in polarized flux density and position angle, observed during VLB observations of three compact sources at 5 GHz.  They report changes in polarized flux density on time scales as short as 4 hours, which are faster and of larger fractional amplitude than changes in total flux density. They argue that these changes are intrinsic to particular components found in the VLB images. While intrinsic changes are clearly a viable explanation, we suggest the alternative of polarized ISS from complex very fine structure in the polarized emission (such as due to a rapid position angle rotation, finer than the VLBI resolution) and a smoother distribution in total brightnes that quenches the ISS in $I$. However, we have not considered a quantitative ISS model for their observations.   \\subsubsection*{Conclusions}  \\begin{itemize}  \\item  Weak ISS can explain the rapid IDV at 8.6 GHz and 4.8 GHz, if it is caused by a local enhancement in scattering (and turbulence?) at about 25 pc from the Earth.  We have assumed that the scattering plasma is stationary in the LSR and so we used the velocity of the Earth relative to the LSR at the time of the observations. At most the velocity might be twice our assumed value, which would increase the scattering distance to 100 pc.  \\item The scattering is found to be highly anisotropic with an axial ratio 1:4, in which the narrow dimension of the density micro-structure is within about $25^{\\circ}$ of the effective velocity. This provides evidence in support of strongly anisotropic plasma turbulence as proposed by Goldreich and Sridhar (1995, 1997), see also Nakayama (2001) and Backer and Chandran (2002). Further the high degree of anisotropy implies a well-ordered magnetic field, as might be expected in a relatively thin scattering layer.   \\item The peak total brightness temperature in the scintillating component, which we associate with the compact core of a jet source, is about $2 \\pm 1 \\times 10^{13}$ K,   This is comparable to other highly beamed jet models with Doppler factors in the range 14-20.  \\item The detailed inter-relations of the linear Stokes' parameters at 8.6 GHz can be modelled quantitatively by ISS of a plausible source model in which 0.5 Jy is in an inner compact component with dimensions $30 \\times 22 \\mu$as. The peak degree of polarization is $\\sim 70 $\\% and there is a rapid rotation by about $180^{\\circ}$ of the angle of polarization across the longer axis of the source. The results indicate polarized structure on a linear scale of 0.2 pc, which can be compared with parsec scales recently reported on 12 blazars at 15 and 22 GHz VLBI by Homan et al.\\ (2002). The remaining 1.3 Jy is in a more extended structure larger than, say, 0.2 mas which may be polarized but is too large to scintillate. We emphasize that this model is not uniquely determined, but that other models with similar features can also be found.  \\item The local enhanced scattering poses a puzzle, which may be resolved by observations of nearby pulsars. A possible explanation is enhanced turbulence thought to exist at the edge of the local interstellar bubble.  However, in such a case the boundary of the bubble is quite irregular and far from a simple ellipsoid. An alternative is scattering in the nearside of an expanding shell, identified in continuum radio maps as the Cetus arc, which is presumed to be a remnant of an expanding supernova shell of about 84 pc in radius.  \\item  We have continued regular monitoring of the the flux density and polarization of the source since 1996, and will present the results in a future paper, including the second episode of rapid IDV. Models for the evolution of the source and of the local scattering medium will be discussed. \\end{itemize}"
    },
    "0208/astro-ph0208131_arXiv.txt": {
        "abstract": "{\\small We present results of spectral and timing analysis based on 11 RXTE PCA/HEXTE observations of the microquasar XTE J1550-564 during its last outburst in January 2002. The observed behaviour is comparable to most Black-Hole Candidates in the low/hard state This is unlike the 1998-99 outburst, when it showed a much more complex feature, probably because of the higher luminosity. For each of the 11 observations we extracted energy spectra and power density spectra, finding typical features of a low/hard state.\\\\ } ",
        "introduction": "The microquasar XTE J1550-564 was discovered with the All Sky Monitor on board RossiXTE \\cite{levine96} in September 1998 \\cite{smith98} and to date it showed four outbursts. The compared analysis of all these outbursts allows to relate different behaviours of the source with different values of flux reached during the outburst. The outburst here analyzed is quite different from the first two and similar to the 2001 one. Radio observations were reported by \\cite{corbel02}. A more complete analysis is described in \\cite{belloni2002}. ",
        "conclusions": "From the analysis of energy spectra we can see how the presence of a high energy cutoff in the power law component is consistent with the presence of a thermal distribution of electrons in the inner region. In such a model the decrease of the cutoff energy implies a decrease of electron temperature: since the photon index does not change correspondingly the optical depth of the Comptonizing cloud must have increased at the same time.  Interpreting the two Lorentzian components present in the PDS as L$_b$ and L$_{\\ell}$ (see \\cite{bpk02}), we can verify how the values found are consistent with existing correlations found in other sources, like the one concerning $\\nu_b$ vs. $\\nu_{\\ell}$ \\cite{straaten02} and $\\nu_b$ vs. rms$^2$ \\cite{bellonihasinger90b}. The timing analysis results confirm that XTE J1550-564 is in a low/hard state.  The general behaviour showed by XTE J1550-564 during this outburst is compatible with the one observed in other BHC in their low/hard state, indicating that at lower fluxes (and therefore possibly lower accretion rates) the source behaves like ``normal'' BHC."
    },
    "0208/astro-ph0208241_arXiv.txt": {
        "abstract": "Two newly identified magnetic cataclysmic variables discovered in the Sloan Digital Sky Survey (SDSS), SDSSJ155331.12+551614.5 and SDSSJ132411.57+032050.5, have spectra showing highly prominent, narrow, strongly polarized cyclotron humps with amplitudes that vary on orbital periods of 4.39 and 2.6 hrs, respectively. In the former, the spacing of the humps indicates the 3rd and 4th harmonics in a magnetic field of $\\sim$60MG. The narrowness of the cyclotron features and the lack of strong emission lines imply very low temperature plasmas and very low accretion rates, so that the accreting area is heated by particle collisions rather than accretion shocks. The detection of rare systems like these exemplifies the ability of the SDSS to find the lowest accretion rate close binaries. ",
        "introduction": "The commissioning year of the Sloan Digital Sky Survey (SDSS; York et al. 2000, Stoughton et al. 2002) showed that this survey is highly effective in finding new cataclysmic variables (Szkody et al. 2002). While previous surveys had primarily identified the brightest systems with the highest mass transfer rates, the SDSS photometry in 5 filters to well beyond 20th magnitude (Gunn et al. 1998, Fukugita et al. 1996, Hogg et al. 2001, Pier et al. 2002, Smith et al. 2002) is able to find the population of low mass transfer rate, very short orbital period systems that are predicted to exist in close binary evolution models (Howell, Rappaport \\& Politano 1997). Included in this latter group are some systems which contain highly magnetic white dwarfs with field strengths 6-240 MG which are termed AM Her systems, or polars (see Warner 1995 for a review of all types of cataclysmic variables).  In the polars, the high field prevents the formation of an accretion disk by directing the ballistic flow of material transferred from the late type secondary onto one or both magnetic poles of the white dwarf. Different regimes of accretion rate and magnetic field strength are predicted to result in quantitatively different accretion scenarios (Wickramasinghe \\& Ferrario 2000; WF2000). At the highest specific accretion rates (100 g cm$^{-2}$ s$^{-1}$), high density blobs carry the mass and energy below the surface of the white dwarf. At mid accretion rates (1 g cm$^{-2}$ s$^{-1}$), a standoff shock is formed above the surface and the gas cools primarily by 10-30 keV thermal bremsstrahlung emission. As the accretion lowers by another factor of 100, the cooling becomes dominated by cyclotron emission, until at the lowest rates, a shock does not form at all and the energy of the incoming ions is transmitted directly to the atmosphere of the white dwarf in what is termed a ``bombardment solution\" (Kuipers \\& Pringle 1982; Fischer \\& Beuermann 2001). The magnetic field also affects the results in that higher fields tend to produce weaker shocks and possibly more direct heating by blobs beneath the surface (Ramsay et al. 1994). Complicating this picture further is the fact that the accretion rate of polars can sporadically change. Typically, sytems in high mass-transfer states exhibit strong HeII $\\lambda$4686 and Balmer emission lines which dominate the optical spectrum, and strong X-ray and cyclotron emission. At low states of reduced or absent mass transfer, the line emission disappears (except for some narrow Balmer emission from the irradiated secondary star), the X-ray and cyclotron emission is much reduced or absent, and the photospheres of the white dwarf and late secondary are visible.  During low states, the magnetic field may be identified through Zeeman splitting in the photospheric spectrum, but for most systems, the field is determined in high states of accretion from the presence of cyclotron harmonics in the optical and near-IR. Cyclotron opacity is a rapidly decreasing function of harmonic number, and the high harmonics are strongly angle-dependent, polarized, and broadened with increasing electron temperature (e.g. Chanmugam 1980; Meggitt \\& Wickramasinghe 1982). For T$<<$10 keV, the wavelength of the harmonic number {\\it n} is simply related to the magnetic field $B$: $\\lambda_n = (10,700)/n)(10^{8}/B)$ \\AA.  Within the above theoretical and empirical framework, observation of the soft and hard X-ray fluxes, and of the optical spectral and cyclotron features, can elucidate the magnetic field and accretion regime of identified polars. More than 80\\% of the $\\sim$ 65 known polars were discovered in the ROSAT All Sky Survey (RASS; Voges et al. 1999), with typical count rates for 15-19th optical magnitudes in the range of 0.2-2.5 c/s (Beuermann \\& Burwitz 1995). The selection criteria of high X-ray count rate and strong optical emission lines resulted in the discovery of polars in the middle to high specific accretion rate regimes. Recently, the deep objective prism plates of the Hamburg Quasar survey provided the identification of 2 polars with extremely low accretion rates (WX LMi = HS 1023+3900; Reimers, Hagen \\& Hopp 1999, and HS 0922+1333; Reimers \\& Hagen 2000) and the followup of faint ROSAT sources yielded two more (RX J012851.9-233931; Schwope, Schwarz \\& Greiner 1999, and RX J1007.5-2016; Reinsch et al. 1999). In this paper, we describe 2 new SDSS polars which are among the most extreme cases of the intriguing cyclotron-dominated systems at the lowest accretion rates. As only a small fraction of the eventual SDSS data have been examined, the survey should discover a modest-sized sample of these previously exotic stars. ",
        "conclusions": "Our photometry, spectroscopy and polarimetry of the SDSS source SDSS1553 have revealed a polar system with an orbital period of 4.39 hr. The spectrum is dominated by extreme amplitude, highly polarized cyclotron harmonics near 6200\\AA\\ and 4600\\AA, indicating a white dwarf magnetic field strength of 58 MG, and TiO features from an M5V secondary star, indicating a distance of 100 pc. Similar cyclotron features and photometric variability  in SDSS1324 indicate a polar with an orbital period near 2.6 hr. The narrowness and extreme amplitude of the cyclotron features imply that these systems are in the regime of low plasma temperature and very low specific accretion rate (the bombardment solution) where the accreting area is heated by particle collisions and the accretion luminosity appears as cyclotron radiation. The low count rates in the RASS ($<$0.04 c/s) support this view.  With its ability to probe a wide variety of stellar systems, the SDSS is contributing to a less biased view of the conditions in polars, especially at low mass transfer rates."
    },
    "0208/astro-ph0208288_arXiv.txt": {
        "abstract": "We study the viscous effects on the interaction between the coplanar Be-star disc and the neutron star in Be/X-ray binaries, using a three-dimensional, smoothed particle hydrodynamics code. For simplicity, we assume the Be disc to be isothermal at the temperature of half the stellar effective temperature.  In order to mimic the gas ejection process from the Be star, we inject particles with the Keplerian rotation velocity at a radius just outside the star. Both Be star and neutron star are treated as point masses.  We find that the Be-star disc is effectively truncated if the Shakura-Sunyaev viscosity parameter $\\alpha_{\\rm SS} \\ll 1$, which confirms the previous semi-analytical result.  In the truncated disc, the material decreted from the Be star accumulates, so that the disc becomes denser more rapidly than if around an isolated Be star. The resonant truncation of the Be disc results in a significant reduction of the amount of gas captured by the neutron star and a strong dependence of the mass capture rate on the orbital phase. We also find that an eccentric mode is excited in the Be disc through direct driving due to a one-armed bar potential of the binary. The strength of the mode becomes greater in the case of a smaller viscosity. In a high-resolution simulation with $\\alpha_{\\rm SS}=0.1$, the eccentric mode is found to precess in a prograde sense.  The mass capture rate by the neutron star modulates as the mode precesses. ",
        "introduction": "\\label{sec:intro}  The Be/X-ray binaries represent the largest subclass of high-mass X-ray binaries. About two-thirds of the identified systems fall into this category. These systems consist of a Be star (i.e., a B star with an equatorial disc) and, generally, a neutron star. The orbit is wide (several tens of days $\\la P_{\\rm orb} \\la$ several hundred days) and eccentric ($0.1 \\la e \\la 0.9$).  Most of the Be/X-ray binaries show only transient X-ray activity due to transient accretion of the circumstellar matter of the Be star, while some show persistent X-ray emission. Each Be/X-ray binary exhibits some or all of the following three types of X-ray activity: \\begin{enumerate} \\item periodic (Type~I) X-ray outbursts, coinciding with periastron passage ($L_{\\rm X} \\approx 10^{36-37}\\,{\\rm erg\\, s}^{-1}$), \\item giant (Type~II) X-ray outbursts ($L_{\\rm X} \\ga 10^{37}\\,{\\rm erg\\, s}^{-1}$), which show no clear orbital modulation, \\item persistent low-luminosity X-ray emission ($L_{\\rm X} \\la 10^{34}\\,{\\rm erg\\, s}^{-1}$) \\end{enumerate} (\\citealt*{swr86}; see also \\citealt{neg98}).  These features imply a complicated interaction between the Be-star envelope and the neutron star.  A Be star has a two-component extended atmosphere, a polar region and a cool ($\\sim 10^4\\,{\\rm K}$) equatorial disc. The polar region consists of a low-density, fast ($\\sim 10^3\\,{\\rm km}\\,{\\rm s}^{-1}$) outflow emitting UV radiation. The wind structure is well explained by the so-called line-driven wind model, in which the radiative acceleration results from the scattering of the stellar radiation in an ensemble of spectral lines (\\citealt*{cak75}; \\citealt{abb82}). On the other hand, the equatorial disc, which is geometrically thin and nearly Keplerian, consists of a high-density plasma from which the optical emission lines and the IR excess arise.  The radial velocity of the disc is smaller than a few ${\\rm km\\,s}^{-1}$, at least within $\\sim 10$ stellar radii (\\citealt{han94}; \\citealt{han00}; \\citealt{wm94}). Although there is no widely accepted model for discs around Be stars, the viscous decretion disc model proposed by \\citet*{lso91} explains many of the observed features and thus seems promising (\\citealt{por99}; see also \\citealt{oka01}). In this model, the matter supplied from the equatorial surface of the star drifts outwards because of the viscous effect and forms the disc. The basic equations for viscous decretion discs are the same as those for viscous accretion discs, except that the sign of $\\dot{M}$ (mass decretion/accretion rate) is opposite. The boundary conditions for decretion discs, however, are different from those for accretion discs. Therefore, the decretion disc has a structure different from that of the accretion disc \\citep{pri91}.  Until quite recently, models for Type~I X-ray outbursts in Be/X-ray binaries had assumed a large disc around the Be star so that the neutron star can accrete gas when it passes through the disc near periastron. However, \\citet{no01} and \\citet{on01a} recently performed a semi-analytical study based on the viscous decretion disc model for Be stars and showed that the Be disc in Be/X-ray binaries is truncated at a radius smaller than the periastron distance, as long as $\\alpha_{\\rm SS} \\ll 1$, where $\\alpha_{\\rm SS}$ is the Shakura-Sunyaev viscosity parameter (see Fig.~\\ref{fig:bex} for a schematic view of a Be/X-ray binary). The truncation of the disc is due to the resonant torque exerted by the neutron star, which removes the angular momentum from the disc. The disc radii they obtained for seven particular systems (4U\\,0115+63, V\\,0332+53, A\\,0535+262, EXO\\,2030+375, 2S\\,1417$-$624, GRO\\,J1008$-$57, and 2S\\,1417$-$624) are consistent with the X-ray behaviour of those systems. Moreover, the result is in agreement with the result of \\citet*{rei97} that there is a positive correlation between the orbital size and the maximum equivalent width of H$\\alpha$ ever observed in a system, a measure of the maximum disc size around the Be star in the system.  \\begin{figure} \\resizebox{\\hsize}{!}{\\includegraphics{fig1.eps}} \\caption{Schematic view of a Be/X-ray binary, taken from \\citet{on01b}.} \\label{fig:bex} \\end{figure}  The truncation of the Be disc in Be/X-ray binaries is not surprising. The resonant interaction is important in various contexts even in a fly-by encounter with a perturber. In fly-by encounters between a disc galaxy and a point mass perturber, the energy is always transported from the disc to the perturber through the resonant interaction, except for overhead encounters where the energy transfer is small \\citep{pal83}. In distant encounters between a circumstellar accretion disc and a perturbing mass with $r_{\\rm peri}/r_{\\rm disc} \\ga 2$, where $r_{\\rm peri}$ and $r_{\\rm disc}$ are the periastron distance and disc radius, respectively, the disc material loses energy and angular momentum to the perturber's orbit through a resonance feature \\citep*{hal96}.  In the case of Be/X-ray binaries, the surface density of the Be disc is expected to increase more rapidly than that for isolated Be stars, as a consequence of truncation. This qualitatively agrees with the result found by \\citet{zam01} that the Be discs in Be/X-ray binaries are about twice as dense as those around isolated Be stars.  The disc may finally become optically thick, and become unstable to radiation-driven warping (\\citealt{pri96}; see also \\citealt{por98}). Multi-wavelength, long-term monitoring observations of V635~Cas, the optical counterpart of 4U\\,0115+63, revealed that the Be disc in 4U\\,0115+63 undergoes a quasi-cyclic ($3-5\\,{\\rm yr}$) dynamical evolution \\citep{neg01}: after each disc-loss episode, the disc starts reforming, grows until it becomes unstable to warping, and after that a Type~II outburst occurs. Although a direct link between the warped disc and the Type~II outburst is still missing, the dynamical evolution of the Be disc is likely to be the agent that controls the X-ray behaviour of the system.  This way, the truncated disc model, at least qualitatively, explains many of the observed features of Be/X-ray binaries. The semi-analytical model adopted by \\citet{no01} and \\citet{on01a}, however, only compares the resonant torque integrated over the whole orbit with the viscous torque to determine at which radius the disc is truncated. Hence, it cannot make a quantitative prediction about how perfect or imperfect the truncation is.  Moreover, it predicts nothing about phase-dependent features, such as the disc deformation and the change in the mass capture rate.  Therefore, in order to study the efficiency of the resonant truncation and the orbital-phase dependence of the interaction, we simulate the interaction between the Be-star disc and the neutron star in Be/X-ray binaries, using a 3D SPH code. In a general context, such simulations will also enable us to study the interaction between the viscous disc and the companion in an eccentric orbit. In this paper, which is the first of a series of papers dedicated to understanding the interaction between the Be disc and the neutron star, we study the effects of viscosity on disc truncation in a coplanar system. ",
        "conclusions": "\\end{minipage} \\end{table*}  \\subsection{Disc Evolution under the Influence of the Neutron Star} \\label{sec:sd_binary}  \\citet{al94} investigated the tidal/resonant truncation of circumstellar and circumbinary discs in eccentric binaries and found that a gap is always formed between the disc and the binary orbit. Following their formulation, \\citet{no01} and \\citet{on01a} showed that the Be disc in Be/X-ray binaries is truncated via the resonant interaction with the neutron star as long as $\\alpha_{\\rm SS} \\ll 1$. In the following, we investigate the resonant interaction between the Be disc and the neutron star in more detail, by analysing the results from 3D SPH simulations.  Table~\\ref{tab:summary} lists some characteristic quantities from these simulations, which will be discussed below.  Figs.~\\ref{fig:sd_a1} and \\ref{fig:sd_a01} show the disc evolution under the influence of the neutron star for $\\alpha_{\\rm SS}=1$ and $\\alpha_{\\rm SS}=0.1$, respectively. The truncation of the disc is obviously more evident for $\\alpha_{\\rm SS}=0.1$ than for $\\alpha_{\\rm SS}=1$.  From Fig.~\\ref{fig:sd_a01}, we clearly see how the resonant torque works on a viscous decretion disc.  When the disc size is small ($t < 10P_{\\rm orb}$ for $\\alpha_{\\rm SS}=0.1$), the disc evolution is almost identical to that around an isolated Be star. As the disc grows, however, the effect of the resonant torque from the neutron star becomes apparent and the radial density distribution begins to have a break at a radius around the 4:1 to 5:1 resonance radii (for $\\alpha_{\\rm SS}=0.1$). We call this radius the truncation radius. Since the resonant torque prevents disc material from drifting outwards, the disc density increases more rapidly than in discs around isolated Be stars. Outside the truncation radius, the density decreases rapidly. The wavy patterns seen in the surface density distribution in the left panel and in the radial velocity distribution in the right panel of Fig.~\\ref{fig:sd_a01} are due to the tightly wound spiral density wave excited in the disc.  \\begin{figure} \\resizebox{\\hsize}{!}{\\includegraphics{fig6.eps}} \\caption{Evolution of the viscous decretion disc with $\\alpha_{\\rm SS}=1$ in a Be/X-ray with $P_{\\rm orb}=24.3\\,{\\rm d}$ and $e=0.34$.  The left panel shows the surface density evolution. The time interval between adjacent contours is 5$P_{\\rm orb}$ ($\\sim 1/3\\,{\\rm yr}$) ($t =$ 5$P_{\\rm orb}$, 10$P_{\\rm orb}$, $\\ldots$ from left).  The right panel shows the disc structure at the end of the simulation. The solid, the dashed, and the dash-dotted lines denote the surface density, the azimuthal velocity normalised by the critical velocity of the Be star, and the radial Mach number.  In both panels, the density is integrated vertically and averaged azimuthally, while the velocity components are averaged vertically and azimuthally. The profile of $V_{\\phi}$ for $r \\la 0.7a$ is indistinguishable from one proportional to $r^{-1/2}$. Annotated at the bottom of the right panel is the number of SPH particles, $N_{\\rm SPH}$, at this epoch.} \\label{fig:sd_a1} \\end{figure}  \\begin{figure} \\resizebox{\\hsize}{!}{\\includegraphics{fig7.eps}} \\caption{Same as Fig.\\ref{fig:sd_a1}, but for $\\alpha_{\\rm SS}=0.1$.} \\label{fig:sd_a01} \\end{figure}  In contrast to the low-viscosity simulation shown in Fig.\\ref{fig:sd_a01}, the resonant torque has little effect on the disc structure when the viscosity is very high. As seen in the left panel of Fig.\\ref{fig:sd_a1}, there is an only very modest break in the surface density distribution for $\\alpha_{\\rm SS}=1$. In this simulation, the disc almost reaches an equilibrium state at $t \\sim 15P_{\\rm orb}$ ($\\sim 1\\,{\\rm yr}$) in the sense that the disc structure varies regularly, depending on the orbital phase.  We also performed a simulation with $\\alpha_{\\rm SS}=0.3$. The disc structure obtained is something between those with $\\alpha_{\\rm SS}=0.1$ and $\\alpha_{\\rm SS}=1$. There is a clear break and a wavy feature in the radial surface-density distribution but they are not so strong as those for $\\alpha_{\\rm SS}=0.1$.  For comparison, we present in Fig.~\\ref{fig:sd_copl2} the result from a simulation with $\\alpha_{\\rm SPH}=1$ and $\\beta_{\\rm SPH}=2$, in which $\\alpha_{\\rm SS}$ is variable in time and space.  From Fig.~\\ref{fig:sd_copl2}, we observe that the disc with constant artificial viscosity parameters evolves in a similar way to that of the $\\alpha$-disc with a similar viscosity parameter, except for the presence of the dip in the density distribution close to the star, as was expected from the previous section.  \\begin{figure} \\resizebox{\\hsize}{!}{\\includegraphics{fig8.eps}} \\caption{Same as Fig.~\\ref{fig:sd_a1}, but for $\\alpha_{\\rm SPH}=1$ and $\\beta_{\\rm SPH}=2$.} \\label{fig:sd_copl2} \\end{figure}  In order to study the interaction between the Be disc and the neutron star in more detail, we performed a high-resolution simulation with $\\alpha_{\\rm SS}=0.1$. In this simulation, the number of SPH particles is about an order of magnitude larger than, and so the smoothing length is on average less than a half of, that of other simulations. Fig.~\\ref{fig:sd_ukaff} shows the surface density evolution for $t=0-45P_{\\rm orb}$ and the disc structure at $t=45P_{\\rm orb}$ in this high-resolution simulation (unfortunately, the allocated computing time ran out at $t=47P_{\\rm orb}$). From Fig.~\\ref{fig:sd_ukaff}, we easily see more detailed disc structure than that in the corresponding simulation with a lower resolution shown in Fig.~\\ref{fig:sd_a01}.  The wavy pattern in the surface density distribution caused by the spiral density wave is more remarkable in the high-resolution simulation.  This is because a larger number of particles give a higher resolution of the interacting region, which makes the interaction more localised and stronger. Although the surface density profile has breaks near the 5:1 resonance radius ($r \\sim 0.33a$) and the 4:1 radius ($r \\sim 0.39a$), the steep density decrease already begins at a much smaller radius, which coincides with the outermost density peak of the spiral wave.  \\begin{figure} \\resizebox{\\hsize}{!}{\\includegraphics{fig9.eps}} \\caption{Same as Fig.~\\ref{fig:sd_a01}, but for the high-resolution simulation.} \\label{fig:sd_ukaff} \\end{figure}  In the rest of this paper, we mainly present the results from this high-resolution simulation as the representative ones with $\\alpha_{\\rm SS}=0.1$, because it gives a better understanding of the star-disc interaction.   \\subsection{Phase-Dependent Disc Structure} \\label{sec:phase}  Most Be/X-ray binaries with known orbital parameters have orbital eccentricities in the range from 0.3 to 0.9. In such systems, the star-disc interaction is likely to be strongly phase-dependent. In this subsection, we discuss phase-dependent features except for the mass capture rate by the neutron star, which will be discussed separately.  Figs.\\ref{fig:ss_a1} and \\ref{fig:ss_a01} give snapshots covering one orbital period for $\\alpha_{\\rm SS}=1$ and $\\alpha_{\\rm SS}=0.1$, respectively. Each panel shows the surface density in a range of about two orders of magnitude in the logarithmic scale.  From these figures, we note a remarkable difference in the disc structure between $\\alpha_{\\rm SS}=1$ and $\\alpha_{\\rm SS}=0.1$. For $\\alpha_{\\rm SS}=1$, the disc has a significant density up to the periastron distance and experiences a strong interaction at and after the periastron passage of the neutron star.  On the other hand, for $\\alpha_{\\rm SS}=0.1$, the resonant torque from the neutron star is much more effective at truncating the disc than for the high viscosity disc.  The sharp decline in the disc density outside the 5:1 resonance causes a gap between this radius and the periastron distance, apparently reducing the mass capture rate by the neutron star, as will be seen in Section~\\ref{sec:capture}.  In both cases, the spiral density waves are clearly seen between the periastron passage and the apastron passage. The opening angle of the spirals, which is related to the effective gravity in the disc, is smaller for a larger binary separation.  Through the resonant interaction, the angular momentum is transported from the disc to the binary. We have to admit that the effect of the angular momentum transport on the binary turned out to be much stronger than we had expected. Despite the fact that the mass of each particle is only $10^{-10}M_{\\sun}$ so that the disc mass is only about $10^{-5}-10^{-6}M_{\\sun}$ in our simulations, the increase in the binary orbital period is visible in the late stage of simulations. The computed phase lags behind the correct orbital phase by $\\sim 0.01$ at $t \\sim 30$ in the simulation with $\\alpha_{\\rm SS}=1$ and by $\\sim 0.04$ and $\\sim 0.05$ at $t \\sim 45$ in the high and normal resolution simulations with $\\alpha_{\\rm SS}=0.1$, respectively. In the following, figures should be read taking this phase shift into account.  \\begin{figure*} \\resizebox{\\hsize}{!} {\\includegraphics[angle=-90]{fig10.eps}} \\caption{Snapshots for $\\alpha_{\\rm SS}=1$, which cover one orbital period.  Each panel shows the logarithm of the surface density. The dark spot near the origin is the Be star, while another dark spot orbiting about the Be star denotes the neutron star with the variable accretion radius. Annotated at the bottom of each panel are the number of SPH particles, $N_{\\rm SPH}$, and the integrated number of particles captured by the neutron star, $N_{\\rm acc}$.} \\label{fig:ss_a1} \\end{figure*}  \\begin{figure*} \\resizebox{\\hsize}{!} {\\includegraphics[angle=-90]{fig11.eps}} \\caption{Same as Fig.\\ref{fig:ss_a1}, but for the high-resolution simulation with $\\alpha_{\\rm SS}=0.1$.} \\label{fig:ss_a01} \\end{figure*}  In order to have a measure of the disc radius, i.e., the radius at which the disc density has a major break, we applied a non-linear least-square fitting method to the radial distribution of the azimuthally-averaged surface density $\\Sigma$, adopting the following simple fitting function, \\begin{equation} \\Sigma \\propto \\frac{\\left( r/r_{\\rm d} \\right)^{-p}}{1+\\left( r/r_{\\rm d} \\right)^{q}}, \\label{eq:lsq} \\end{equation} where $p$ and $q$ are constants and $r_{\\rm d}$ is the disc radius.  Fig.~\\ref{fig:rd_copl} shows the phase dependence of the disc radius (thick line) in the simulations shown in Figs.\\ref{fig:ss_a1} and \\ref{fig:ss_a01}. To reduce the fluctuation noise, we folded the data on the orbital period over $25 \\le t \\le 30$ for $\\alpha_{\\rm SS}=1$ and $42 \\le t \\le 47$ for $\\alpha_{\\rm SS}=0.1$.  In the figure, the horizontal solid lines denote some of the lowest $n:1$ resonance radii (the 2:1, the 3:1, $\\ldots$, the 10:1 from top), while the thin sinusoidal line denotes the orbit of the neutron star. The origin of the phase is at the periastron passage of the neutron star.  From Fig.~\\ref{fig:rd_copl}, we note that the disc radius coincides with the 4:1 resonance radius ($r/a = 0.39$) for $\\alpha_{\\rm SS}=1$, whereas the disc has a radius intermediate between the 4:1 radius and the 5:1 radius ($r/a = 0.33$) for $\\alpha_{\\rm SS}=0.1$ (the mean $r_{\\rm d}$ is $0.39a$ for $\\alpha_{\\rm SS}=1$ and $0.36a$ for $\\alpha_{\\rm SS}=0.1$). The latter is a typical feature expected for a disc in which the wave damps locally \\citep{al94}.  We also note that the disc radius modulates around the mean value. The disc shrinks after the periastron passage of the neutron star, which gives a negative torque to the disc. After that it restores its radius by viscous diffusion, so that the amplitude of the modulation is larger for $\\alpha_{\\rm SS}=1$ than for $\\alpha_{\\rm SS}=0.1$.  \\begin{figure} \\resizebox{\\hsize}{!}{\\includegraphics{fig12a.eps} \\quad\\quad \\includegraphics{fig12b.eps}} \\caption{Orbital-phase dependence of the disc radius $r_{\\rm d}$. Averaging is done over $25 \\le t \\le 30$ for $\\alpha_{\\rm SS}=1$ (left) and $42 \\le t \\le 47$ for $\\alpha_{\\rm SS}=0.1$ (right). The horizontal solid lines denote the 2:1, the 3:1, the 4:1, $\\ldots$, the 10:1 resonance radii from top to bottom, and the horizontal dashed line denotes the phase averaged value of $r_{\\rm d}$. The thin sinusoidal line denotes the orbit of the neutron star. The periastron passage of the neutron star, which occurs at phase 0, is denoted by the vertical dashed line.} \\label{fig:rd_copl} \\end{figure}  Since the disc structure depends on the binary phase, the disc emission is expected to exhibit an orbital modulation as well. To see whether this is the case, we calculated the continuum flux from the Be disc for $r \\ge 2r_{\\rm inj}-R_{*}$, assuming that the disc is optically thin and the emissivity is proportional to $\\rho^2$, where $\\rho$ is the local density. For simplicity, we ignored the effect of the obscuration of the disc by the star, which will become important for systems with high inclination angles. We then obtained a base flux curve by performing the cubic-spline fitting of the fluxes at apastron passages. The base flux describes the long-term change in the continuum flux. Finally, we computed the orbital modulation by subtracting the base flux from the instantaneous fluxes.  Fig.~\\ref{fig:lc_copl} shows the orbital modulation in the optically thin continuum from the Be disc for $\\alpha_{\\rm SS}=1$ (left) and $\\alpha_{\\rm SS}=0.1$ (right).  For $\\alpha_{\\rm SS}=0.1$, the modulation is more clearly seen in the high-resolution simulation (thick line) than in the normal one (thin line).  In order to reduce the fluctuation noise, we folded the data on the orbital period over the period annotated in the panel.  Contrary to what is expected from Fig.\\ref{fig:rd_copl}, the continuum exhibits little modulation for $\\alpha_{\\rm SS}=1$, and about one per cent of modulation is seen for $\\alpha_{\\rm SS}=0.1$. The negative result for $\\alpha_{\\rm SS}=1$ suggests that the strongly-perturbed outer disc contributes little to the optically thin emission, because of its low density. On the other hand, the positive result for $\\alpha_{\\rm SS}=0.1$, in particular in the high-resolution simulation, though the amplitude is still very small, seems to come from the region where the density is enhanced by the spiral wave, because the disc radius does not change significantly and the rise and the subsequent decline of disc emission are in phase with the density enhancement and the subsequent damping caused by the spiral density wave.  The difference between the modulation patterns from the high and normal resolution simulations with $\\alpha_{\\rm SS}=0.1$ suggests that resolving the spiral density wave is important to obtain a reliable modulation pattern of such low amplitude.  \\begin{figure} \\resizebox{\\hsize}{!}{\\includegraphics{fig13a.eps} \\quad\\quad \\includegraphics{fig13b.eps}} \\caption{Orbital modulation in the optically thin continuum from the Be disc for $\\alpha_{\\rm SS}=1$ (left) and $\\alpha_{\\rm SS}=0.1$ (right). In the right panel, the thick and thin lines are for the high and normal resolution simulations, respectively. Phase 0 corresponds to the periastron passage of the neutron star.} \\label{fig:lc_copl} \\end{figure}  Although Fig.~\\ref{fig:lc_copl} gives little observability of the orbital modulation in optically thin disc emission, it is still likely that optically thick emission, such as Balmer lines, will show significant orbital modulation. Calculating the optically thick emission from the disc is, however, beyond the scope of this paper.   \\subsection{Excitation of the Eccentric Mode} \\label{sec:ecc_mode}  In a circumbinary disc around an eccentric binary, an eccentric mode is excited through direct driving due to a one-armed bar potential \\citep{al96a}.  In this subsection, we study whether the same mechanism works in Be/X-ray binaries. We analyse the evolution of the eccentricity in the Be disc by decomposing the surface density distribution into Fourier components which vary as $\\exp i(k \\phi - \\ell \\Omega_{\\rm orb} t)$, where $k$ and $\\ell$ are the azimuthal and time-harmonic numbers, respectively, and $\\Omega_{\\rm orb} = [G(M_{*}+M_{X})/a^3]^{1/2}$ is the frequency of mean binary rotation.  Following \\citet{lub91}, we define the mode strength by \\begin{eqnarray} S_{f, g, k, l} &=& \\frac{2}{\\pi M_{\\rm d} (1+\\delta_{k, 0}) (1+\\delta_{\\ell, 0})} \\int_t^{t+2\\pi \\Omega_{\\rm orb}^{-1}} dt' \\nonumber\\\\ &&\\nonumber\\\\ && \\times \\int dr \\int_0^{2\\pi} d\\phi\\ \\Sigma(r, \\phi, t) f(k \\phi) g(\\ell t'), \\label{eq:sfgkl} \\end{eqnarray} where $f$ and $g$ are either $\\sin$ or $\\cos$ functions.  The surface density here is computed by summing up $\\delta$ functions at particle positions, not by taking the kernel into account as has been done in the previous sections.  Then, the total strength of the mode $(k, \\ell)$ is defined by \\begin{eqnarray} S_{k, \\ell}(t) &=& (S_{\\cos, \\cos, k, \\ell}^2 + S_{\\cos, \\sin, k, \\ell}^2 \\nonumber\\\\ && + S_{\\sin, \\cos, k, \\ell}^2 + S_{\\sin, \\sin, k, \\ell}^2)^{1/2}. \\label{eq:skl} \\end{eqnarray}  Fig.~\\ref{fig:mode_a1} shows the excitation and precession of the (1,0) mode (i.e., the eccentric mode) for $\\alpha_{\\rm SS}=1$. The upper panel shows the strengths of several modes, while the lower panel shows the evolution of the angle $\\omega_{\\rm d}$ between the eccentric vector of the disc and that of the binary orbit defined by \\begin{equation} \\tan \\omega_{\\rm d} = \\frac{\\int dr \\int_0^{2\\pi} d\\phi\\ \\Sigma(r, \\phi, t) \\sin \\phi} {\\int dr \\int_0^{2\\pi} d\\phi\\ \\Sigma(r, \\phi, t) \\cos \\phi}. \\label{eq:omega_d} \\end{equation}  The eccentric mode grows initially linearly in time ($t \\la 8$), as is predicted by the theory and was also found by \\citet{la00}. At $t \\sim 15$ the strength of the eccentric mode is saturated at $S_{1,0} \\sim 0.04$. As it is saturated, the mode stops precessing and is locked at $\\omega_{\\rm d} \\sim \\pi$.  \\begin{figure} \\centerline{\\includegraphics[width=0.35\\textwidth] {fig14a.eps}} \\centerline{\\includegraphics[width=0.35\\textwidth] {fig14b.eps}} \\caption{Excitation of the eccentric mode. The upper panel shows the strengths of several modes.  The solid, the dashed, the dash-dotted, and the dotted lines denote the strengths of the (1,0) mode, the (2,3) mode, the (1,1) mode, and the (2,2) mode, respectively. The lower panel shows the angle between the eccentric vectors of the disc and the binary.} \\label{fig:mode_a1} \\end{figure}  In order to study the structure of the eccentric mode in more detail, we analyse the orbits of individual particles. The position and velocity of each particle are instantaneously equal to those of an elliptical Keplerian orbit of semi-latus rectum $\\lambda$ and the eccentric vector $\\bmath{e}_{\\rm SPH}$ with the amplitude $e_{\\rm SPH}$ and the longitude of periastron, $\\omega$, with respect to the stellar periastron (e.g., \\citealt{ogi01}). The radial coordinate $r$ is then given by \\begin{equation} r = \\frac{\\lambda}{1+e_{\\rm SPH} \\cos (\\phi-\\omega)}. \\end{equation} Given $e_{\\rm SPH} \\ll 1$, $\\lambda \\sim r$ in our simulations.  The upper panels of Fig.~\\ref{fig:ecc_lowres} present the distribution of the orbital parameters of disc particles for $\\alpha_{\\rm SS}=1$ at $t=30$. From the left panel, we note that the disc eccentricity increases almost linearly in $\\lambda$ for $\\lambda \\la 0.3a$ and is roughly constant for $\\lambda \\ga 0.3a$. The right panel for the radial dependence of the longitude of periastron shows that the eccentric mode is twisted in a trailing sense.  \\begin{figure} \\centerline{\\includegraphics[width=0.4\\textwidth] {fig15a.eps}} \\centerline{\\includegraphics[width=0.4\\textwidth] {fig15b.eps}} \\caption{Eccentricity $e_{\\rm SPH}$ and the longitude of periastron, $\\omega$, of the particle orbits from the normal-resolution simulations with $\\alpha_{\\rm SS}=1$ at $t=30$ (upper) and $\\alpha_{\\rm SS}=0.1$ at $t=50$ (lower).  The gray-scale plot in each panel shows the particle distribution in linear scale. $<e>$ and $<\\omega/\\pi>$ are mean values of $e_{\\rm SPH}$ and $\\omega/\\pi$.} \\label{fig:ecc_lowres} \\end{figure}  The normal-resolution simulation with $\\alpha_{\\rm SS}=0.1$ showed a similar trend. The eccentric mode grew initially linearly in time ($t \\la 40$). The mode strength was almost saturated at $t \\sim 45$ at $S_{1,0} \\sim 0.1$.  The mode also stopped precessing and was locked at $\\omega_{\\rm d} \\sim 3\\pi/2$.  The distribution of the orbital parameters of disc particles at $t=50$ from this simulation is presented in the lower panels of Fig.~\\ref{fig:ecc_lowres}. From the left panel, we note that the eccentricity has a maximum at $\\lambda \\sim 0.3a$. The right panel shows that the eccentric mode is twisted in a trailing sense, as in the case for $\\alpha_{\\rm SS}=1$.  Recently, \\citet{ogi01} has developed a non-linear theory of the evolution of the shape and surface density of a three-dimensional eccentric accretion disc.  When the eccentricity of the disc is small so that terms of relative order $O(e^2)$ may be neglected, this theory provides a linear evolutionary equation for the complex eccentricity $E(\\lambda,t)=e(\\lambda,t)\\,{\\rm e}^{{\\rm i}\\omega(\\lambda,t)}$ of the disc.  We have reworked this theory for the case of a strictly isothermal decretion disc with no radial velocity, and also evaluated the tidal forcing terms associated with a companion object of small eccentricity.  The governing equation is then of the form \\begin{eqnarray} \\lefteqn{\\Sigma(GM_*\\lambda)^{1/2}{{\\partial E}\\over{\\partial t}}= {{\\partial}\\over{\\partial\\lambda}}(Z_1\\Sigma c_{\\rm s}^2\\lambda)+ Z_2\\Sigma c_{\\rm s}^2}&\\nonumber\\\\ &&+{{1}\\over{4}}\\frac{{\\rm i}GM_X\\Sigma\\beta}{\\lambda_{\\rm b}} \\left[b_{3/2}^{(1)}(\\beta)E-b_{3/2}^{(2)}(\\beta)E_{\\rm b}\\right], \\label{eq:dedt} \\end{eqnarray} where $E_{\\rm b}$ and $\\lambda_{\\rm b}$ are the (constant) eccentricity and semi-latus rectum of the binary orbit, $b_\\gamma^{(m)}$ is the Laplace coefficient of celestial mechanics, and $\\beta=\\lambda/\\lambda_{\\rm b}$.  The dimensionless coefficients $Z_1$ and $Z_2$ are given by \\begin{equation} Z_1=C_1E+C_2\\lambda{{\\partial E}\\over{\\partial\\lambda}},\\qquad Z_2=C_3E+C_4\\lambda{{\\partial E}\\over{\\partial\\lambda}}, \\end{equation} where $C_1,\\dots,C_4$ are dimensionless complex constants that depend only on the shear and bulk viscosity parameters $\\alpha_{\\rm SS}$ and $\\alpha_{\\rm b}$. In the limit $\\alpha_{\\rm SS}, \\alpha_{\\rm b} \\to 0$ these coefficients become purely imaginary and give rise to precession induced by the pressure of the disc. Most importantly, the real part of $C_2$, which has the role of a diffusion coefficient for short-wavelength eccentric perturbations, turns out to be positive when $\\alpha_{\\rm b}/\\alpha_{\\rm SS}>2/3$, as is true in the SPH simulations [see equation~(\\ref{eq:nubsph})].  If this condition were not satisfied, the disc would experience a local eccentric instability and equation (\\ref{eq:dedt}) could not be evolved forward in time.  Equation (\\ref{eq:dedt}) has the character of a dispersive and diffusive linear wave equation for $E(\\lambda,t)$.  The eccentricity of the binary provides a tidal forcing that is independent of time. Starting from an initially circular state $E=0$, the eccentricity of the disc first grows linearly in time and then approaches a steady state.  The steady eccentric shape can be determined either by solving the time-dependent problem until the transient response decays, or by solving the second-order ordinary differential equation obtained by setting the time-derivative to zero.  We have performed both of these calculations, adopting the boundary condition $E=0$ at the stellar surface, a `stress-free' boundary condition $Z_1=0$ at a notional outer edge ($\\beta=0.39$), and a surface density distribution $\\Sigma\\propto\\lambda^{-2}$.  The steady configurations of $e(\\lambda)$ and $\\omega(\\lambda)$ for the cases $\\alpha_{\\rm SS}=1$ and $\\alpha_{\\rm SS}=0.1$ are illustrated in Fig.~\\ref{fig:ecc_model}.  One should not expect a perfect agreement with the SPH simulations because the theory is based on the assumption of small eccentricities and only approximates the actual surface density distribution.  However, the steady eccentric shapes based on linear theory offer a fair explanation of the results of the SPH simulations, especially for the high-viscosity case.  \\begin{figure} \\centerline{\\includegraphics[width=0.4\\textwidth] {fig16a.eps}} \\centerline{\\includegraphics[width=0.4\\textwidth] {fig16b.eps}} \\caption{Model eccentricity $e$ and longitude of periastron $\\omega$ of the particle orbits for $\\alpha_{\\rm SS}=1$ (upper) and $\\alpha_{\\rm SS}=0.1$ (lower).  The notional outer edge is at $\\beta (=\\lambda/\\lambda_{\\rm b})=0.39$.} \\label{fig:ecc_model} \\end{figure}  Fig.~\\ref{fig:mode_a01} shows the evolution of an eccentric mode from the high-resolution simulation with $\\alpha_{\\rm SS}=0.1$.  The detailed structure of the mode is shown in Fig.~\\ref{fig:ecc_a01hi} by snapshots covering one precessional period.  From these figures, we note that the evolution and the structure of the eccentric mode are very similar to those from the corresponding normal resolution simulation until $t \\sim 25$. At $t \\sim 25$, the eccentric mode suddenly begins to precess in a prograde sense, as seen in the lower panel of Fig.\\ref{fig:mode_a01}. As $\\omega_{\\rm d}$ increases from $\\sim 3\\pi/2$ to $2\\pi$, the mode strength increases.  The mode stagnates at $\\omega_{\\rm d} \\sim 0$ for $(1-2)P_{\\rm orb}$ around $t \\sim 30$, suggesting that the eccentric disc at $\\omega_{\\rm d} \\sim 0$ is an unstable configuration. The precession decelerates before reaching $\\omega_{\\rm d} = 0$ and accelerates after that. At the same time, the mode changes its shape. As seen in the lower panel of Fig.~\\ref{fig:ecc_a01hi}, the eccentric mode is twisted in a trailing sense for $\\omega_{\\rm d} < 0$ and in a leading sense after that. At $\\omega_{\\rm d}=0$, apsidal axes of all particle orbits align.  As seen in Fig.~\\ref{fig:ecc_a01hi}, the twist of the mode increases, as the apsidal axis of the disc moves toward the stellar periastron ($\\omega_{\\rm d}=0 \\to \\pi$). The strong radial dependence of the phase caused by the twist reduces the eccentricity of the disc, as shown in the upper panel of Fig.~\\ref{fig:mode_a01}.  The precession is fastest at $\\omega_{\\rm d} \\sim \\pi$, suggesting that there is a stable point at $\\omega_{\\rm d} \\sim \\pi$.  After that, the mode becomes trailing and its strength increases. The precessional period is about $20 P_{\\rm orb}$.  It is important to note that a similar behaviour is found in circumbinary discs around eccentric binaries.  According to \\citet{la00}, this behaviour occurs as follows: When the eccentricity of the disc edge is small, $\\omega_{\\rm d}$ is locked at a stable value $\\omega_{\\rm d} = 3\\pi/2$. However, as the eccentricity grows, the locking action weakens, and the prograde precession due to the quadrupole moment of the binary potential dominates. The disc edge begins to precess when its eccentricity becomes $(0.2-0.7) e$, and afterwards the eccentricity oscillates with a precessional period. The disc typically attains the eccentricity of $(0.5-1) e$.  We note that the above behaviour of the eccentric mode in circumbinary discs around eccentric binaries is strikingly similar to that shown in Fig.~\\ref{fig:mode_a01}, except that, in our simulation for Be/X-ray binaries, the growth and precessional time-scales of the eccentric mode are much shorter and the disc eccentricity attained is significantly smaller than in circumbinary discs.  \\begin{figure} \\centerline{\\includegraphics[width=0.35\\textwidth] {fig17a.eps}} \\centerline{\\includegraphics[width=0.35\\textwidth] {fig17b.eps}} \\caption{Same as Fig.~\\ref{fig:mode_a1}, but for the high-resolution simulation with $\\alpha_{\\rm SS}=0.1$.} \\label{fig:mode_a01} \\end{figure}  \\begin{figure*} \\resizebox{\\hsize}{!}{\\includegraphics{fig18.eps}} \\caption{Change in the eccentricity $e_{\\rm SPH}$ and the longitude $\\omega$ of the disc particle orbits in the high-resolution simulation with $\\alpha_{\\rm SS}=0.1$. Epochs are chosen for illustrative purpose, so the time interval is not constant. The gray-scale plot in each panel shows the particle distribution in linear scale. $<e>$ and $<\\omega/\\pi>$ are mean values of $e_{\\rm SPH}$ and $\\omega/\\pi$ at each epoch, respectively.} \\label{fig:ecc_a01hi} \\end{figure*}   \\subsection{Mass Capture Rate by the Neutron Star} \\label{sec:capture}  As mentioned in Section~\\ref{sec:intro}, most of the Be/X-ray binaries show transient X-ray activities. Among them, some exhibit periodical X-ray outbursts called Type~I, which coincide with the periastron passage, while the others show occasional giant X-ray outbursts called Type~II and little or no detectable X-ray emission in quiescent phase. 4U\\,0115+63, the system we are modelling in this paper, belongs to the latter group.  In this subsection, we first study how much mass is captured by the neutron star in a general context, and then discuss whether our model predicts the mass capture rate consistent with the observed X-ray behaviour of 4U\\,0115+63.  Fig.~\\ref{fig:mdot_a1} shows the change in the mass capture rate by the neutron star, $\\dot{M}_{\\rm acc}$, and the disc mass $M_{\\rm d}$ for $\\alpha_{\\rm SS}=1$. The upper panel shows the evolution of $\\dot{M}_{\\rm acc}$ and $M_{\\rm d}$, while the lower panel shows the orbital-phase dependence of $\\dot{M}_{\\rm acc}$. In the lower panel, we folded $\\dot{M}_{\\rm acc}$ on the orbital period over $25 \\le t \\le 30$ to reduce the fluctuation noise. The horizontal dashed line and dash-dotted line denote the mean mass capture rate by the neutron star and the mean mass-loss rate from the Be star, respectively.  We have already seen that there is little truncation of the Be disc for $\\alpha_{\\rm SS}=1$. Fig.~\\ref{fig:mdot_a1} confirms this.  For $t \\ga 20$ the Be disc is almost in equilibrium in the sense that the disc mass and the mass capture rate only shows regular orbital modulations with constant amplitude.  For $25 \\le t \\le 30$, the mean mass-loss rate from the Be star is $2.4 \\times 10^{-10} \\rho_{-11} M_\\odot{\\rm yr}^{-1}$, while the mean mass capture rate for the same period is $2.1\\times 10^{-10} \\rho_{-11} M_\\odot{\\rm yr}^{-1}$.  Thus, in this high viscosity disc, the neutron star captures the disc mass at about the same rate as the mass-loss rate from the Be star.  \\begin{figure} \\centerline{\\includegraphics[width=0.35\\textwidth] {fig19.eps}} \\caption{Evolution of the disc mass and the mass capture rate by the neutron star (upper) and the orbital-phase dependence of the mass capture rate (lower). In the upper panel, the thin line denotes the mass capture rate $\\dot{M}_{\\rm acc}$ and the thick line denotes the disc mass $M_{\\rm d}$.  In the lower panel, the data are folded on the orbital period over $25 \\le t \\le 30$.  The horizontal dashed line and the dash-dotted line in the lower panel denote the mass capture rate by the neutron star and the mass-loss rate from the Be star averaged over the period annotated at the bottom of the panel.} \\label{fig:mdot_a1} \\end{figure}  The simulation with $\\alpha_{\\rm SS}=0.3$ also gives a high fraction (about 60 per cent for $45 \\le t \\le 50$) of mass capture rate with respect to the mass-loss rate, as listed in Table~\\ref{tab:summary}. This indicates that the resonant truncation effect is not effective for $\\alpha_{\\rm SS}=0.3$, either.  In contrast to the simulations with $\\alpha_{\\rm SS}=1$ and $\\alpha_{\\rm SS}=0.3$, the high-resolution simulation with $\\alpha_{\\rm SS}=0.1$ revealed a subtle feature in the mass capture rate, as shown in Fig.~\\ref{fig:mdot_a01}.  Before the precession of the eccentric mode began, the mass capture rate $\\dot{M}_{\\rm acc}$ increased monotonically: $\\dot{M}_{\\rm acc} = 0$ before $t \\sim 10$ with the resolution of this simulation. Then, the peak mass capture rate showed a gradual increase to a level at $\\dot{M}_{\\rm acc} = (8-9) \\times 10^{-11} \\rho_{-11} M_{\\sun}\\,{\\rm yr}^{-1}$ at $t \\sim 25$.  After the eccentric mode began to precess, $\\dot{M}_{\\rm acc}$ exhibited a long-term modulation, depending on the longitude of the eccentric mode, $\\omega_{\\rm d}$. It gradually increased as $\\omega_{\\rm d}$ increased from 0 to $\\pi$ and decreased as $\\omega_{\\rm d}$ increased from $\\pi$ to $2\\pi$. At the periastron passage at $t \\sim 38$, $\\dot{M}_{\\rm acc}$ was a maximum of $(2-3) \\times 10^{-10} \\rho_{-11} M_{\\sun}\\,{\\rm yr}^{-1}$, which was about a factor of three higher than the level before precession. The eccentricity of the disc was $\\sim 0.1$ at $\\omega_{\\rm d} \\sim 0$ and $0.02-0.03$ at $\\omega_{\\rm d} \\sim \\pi$. It should be noted that even this small eccentricity caused a factor of three enhancement in the mass capture rate. If the disc had a much stronger disturbance, the enhancement in $\\dot{M}_{\\rm acc}$ is likely to be much larger.  In addition to the long-term modulation due to the precession of the eccentric mode, we note that the mass capture rate by the neutron star is much smaller and more strongly phase-dependent for $\\alpha_{\\rm SS}=0.1$ than for $\\alpha_{\\rm SS}=1$. For $\\alpha_{\\rm SS}=0.1$, the phase at which the neutron star captures the disc mass most slightly lags behind the periastron passage, because of the presence of a gap between the disc outer radius and the periastron. In the high resolution simulation, the peak mass capture rate for $42 \\le t \\le 47$ is $\\sim 10^{-10} \\rho_{-11} M_\\odot{\\rm yr}^{-1}$, which is one order of magnitude smaller than that for $\\alpha_{\\rm SS}=1$.  The mass capture rate then decreases by two orders of magnitude by the apastron passage.  Even in the normal resolution simulation, in which the disc density around the truncation radius is significantly higher than that in the high-resolution simulation, the peak mass capture rate is about a factor of three smaller than that for $\\alpha_{\\rm SS}=1$, and decreases by a factor of fifty by apastron. Note that similar, strongly phase-dependent accretion was found in simulations by \\citet{al96b} for circumbinary discs around eccentric binaries.  In the high-resolution simulation with $\\alpha_{\\rm SS}=0.1$, the neutron star, on average, captures the disc mass at the rate of $2.3 \\times 10^{-11} \\rho_{-11} M_\\odot{\\rm yr}^{-1}$ for $42 \\le t \\le 47$, while the mean mass-loss rate from the Be star for the same period is $1.6 \\times 10^{-10} \\rho_{-11} M_\\odot{\\rm yr}^{-1}$. This indicates that, even after three years of disc growth, about 6/7 of the gas lost from the star still accumulates in the disc, making the disc continually denser.  \\begin{figure} \\centerline{\\includegraphics[width=0.35\\textwidth] {fig20.eps}} \\caption{Same as Fig.~\\ref{fig:mdot_a1}, but for the high-resolution simulation with $\\alpha_{\\rm SS}=0.1$. In the lower panel, the data are folded on the orbital period over $42 \\le t \\le 47$. For comparison, the mass capture rate from the normal-resolution simulation with $\\alpha_{\\rm SS}=0.1$ (thin solid line) is overlapped in the lower panel.} \\label{fig:mdot_a01} \\end{figure}  The transient nature in the X-ray activity of Be/X-ray binaries is considered to result from the interactions between the accreted material and the rotating magnetised neutron star \\citep{swr86}. If the accreted material is dense enough to make the magnetospheric radius smaller than the corotation radius, the accretion onto the neutron star causes a bright X-ray emission (the direct accretion regime). Otherwise, the magnetospheric radius is larger than the corotation radius, and the accretion onto the neutron star is prevented by the propeller mechanism. The system is then in quiescence (the propeller regime).  Recently, \\citet{cam01} found that 4U\\,0115+63 showed a dramatic X-ray luminosity variation from $L_{\\rm X} \\sim 2 \\times 10^{34} {\\rm erg\\,s}^{-1}$ to $L_{\\rm X} \\sim 5 \\times 10^{36} {\\rm erg\\,s}^{-1}$ in less than 15\\,hr close to periastron, whereas it showed a nearly constant X-ray luminosity at a level of $L_{\\rm X} \\sim 2 \\times 10^{33} {\\rm erg\\,s}^{-1}$ near apastron. They concluded that the system was in the transition regime between the direct accretion regime and the propeller regime close to periastron and in the propeller regime near apastron, because the direct accretion regime and the propeller regime, respectively, apply for $L_{\\rm X} \\ga \\xi^{7/2} 10^{37} {\\rm erg\\,s}^{-1}$ and $L_{\\rm X} \\la \\xi^{7/2} 2 \\times 10^{34} {\\rm erg\\,s}^{-1}$, where $0.5 \\le \\xi \\le 1$ is a parameter which depends on the geometry and physics of accretion.  Below we try to compare our simulation results with the above criteria by \\citet{cam01}. Although it is likely that the captured material will form an accretion disc around the neutron star, we do not know how the accretion rate is related to the mass capture rate shown in Figs.~\\ref{fig:mdot_a1} and \\ref{fig:mdot_a01}, which is based on the moments at which particles enter the variable accretion radius of the neutron star.  Therefore, we consider two extreme situations, in which $t_{\\rm acc} \\ll P_{\\rm orb}$ or $t_{\\rm acc} \\sim P_{\\rm orb}$, where $t_{\\rm acc}$ is the accretion time-scale. In the former situation, the accretion rate profile is approximately the same as the profile of mass capture rate, whereas in the latter situation, the variation in the accretion rate will be much smaller than that in the mass capture rate. We assume that the X-ray luminosity is given by $L_X = \\eta GM_X \\dot{M}/R_X$ with $\\eta=1$.  For $\\alpha_{\\rm SS}=1$, $L_{\\rm X} \\sim 10^{37} \\rho_{-11} M_{\\sun}\\,{\\rm yr}^{-1}$ at periastron if $t_{\\rm acc} \\ll P_{\\rm orb}$ and all the captured mass accretes onto the neutron star. Note that this level of luminosity enters the direct accretion regime. Since 4U\\,0115+63 is considered to be in the direct accretion regime only in occasional giant X-ray outbursts, we conclude that the $\\alpha_{\\rm SS}=1$ model is inconsistent with the observation if $t_{\\rm acc} \\ll P_{\\rm orb}$. If $t_{\\rm acc} \\sim P_{\\rm orb}$, the neutron star will emit the X-ray luminosity corresponding to the mean mass capture rate of $\\sim 2 \\times 10^{-10} \\rho_{-11} M_{\\sun}\\,{\\rm yr}^{-1}$ (see the bottom panel of Fig.~\\ref{fig:mdot_a1}). The X-ray luminosity is then $L_{\\rm X} \\sim 2 \\times 10^{36} \\rho_{-11} {\\rm erg\\,s}^{-1}$. This level of luminosity enters the transition regime. Hence, the $\\alpha_{\\rm SS}=1$ model is not ruled out by the observed constraint if $t_{\\rm acc} \\sim P_{\\rm orb}$.  We have the same conclusion for the $\\alpha_{\\rm SS}=0.3$ model. Since it gives the mass capture rate about a half of that for $\\alpha_{\\rm SS}=1$, the model is inconsistent with the observation if $t_{\\rm acc} \\ll P_{\\rm orb}$, but is not ruled out if $t_{\\rm acc} \\sim P_{\\rm orb}$.  On the other hand, for $\\alpha_{\\rm SS}=0.1$, $L_{\\rm X} \\sim 10^{36} \\rho_{-11} M_{\\sun}\\,{\\rm yr}^{-1}$ at periastron even if $t_{\\rm acc} \\ll P_{\\rm orb}$ and all the captured material accretes onto the neutron star.  This suggests that the system is in the transition regime even at periastron. It is likely that the low mass capture rate in this simulation put the system into the propeller regime in most of the orbital phases. Thus, the $\\alpha_{\\rm SS}=0.1$ model for 4U\\,0115+63 is consistent with the observed X-ray behaviour, irrespective of the accretion time-scale.     In this paper, we have presented results from 3D SPH simulations of the disc formation around isolated Be stars and of the interaction between the Be star disc and the neutron star in Be/X-ray binaries, based on the viscous decretion disc model for Be stars \\citep{lso91}. In several simulations we adopted constant values of artificial viscosity parameters $\\alpha_{\\rm SPH}$ and $\\beta_{\\rm SPH}$, for which the Shakura-Sunyaev viscosity parameter $\\alpha_{\\rm SS}$ is variable in time and space. In the other simulations, we adopted constant values of $\\alpha_{\\rm SS}$, for which $\\alpha_{\\rm SPH}$ is variable in time and space and $\\beta_{\\rm SPH}=0$. We preferred constant $\\alpha_{\\rm SS}$ simulations because the analysis of the results becomes easier.  In all simulations, the Be disc was nearly Keplerian and the radial velocity was highly subsonic. These features are consistent with those observed for Be discs.  The simulations of isolated Be stars showed that our code is capable of producing results similar to those from the 1D simulations. The simulated mass-loss rate from the Be star in the first several years of disc formation was several $\\times 10^{-10} \\rho_{-11} M_{\\sun}\\, {\\rm yr}^{-1}$ for a wide range of viscosity parameter, which is consistent with the observed equatorial mass-loss rate.  Here, $\\rho_{-11}$ is the highest local density at $t= 1$\\,yr normalised by $10^{-11}{\\rm g}\\,{\\rm cm}^{-3}$, a typical value for Be stars.  In binary simulations, we have studied the effect of viscosity on the star-disc interaction in the case of a coplanar system with $P_{\\rm orb}=24.3\\,{\\rm d}$ and $e=0.34$, the parameters for 4U\\,0115+63, one of the best studied Be/X-ray binaries. We have chosen these orbital parameters because they enable us to compare the simulation results with the observed ones and the short orbital period makes the computing time bearable.  Some of the results from these simulations are summarised in Table~\\ref{tab:summary}.  Our simulations showed that there is a radius outside which the azimuthally averaged surface density decreases steeply. For a smaller $\\alpha_{\\rm SS}$, the slope outside this radius is steeper, giving a stronger truncation effect on the disc. Among the simulations with $\\alpha_{\\rm SS}=1$, 0.3, and 0.1, we found that the resonant truncation of the Be disc is effective only for $\\alpha_{\\rm SS}=0.1$. For $\\alpha_{\\rm SS}=1$ and 0.3, the neutron star captures the disc mass at a rate comparable to the mass-loss rate from the Be star. These results confirm the previous semi-analytical result by \\citet{no01} and \\citet{on01a} on disc truncation that the resonant truncation is effective for a disc with $\\alpha_{\\rm SS} \\ll 1$. The truncation radius for $\\alpha_{\\rm SS}=0.1$ roughly agrees with that derived semi-analytically.  Our simulations, in particular the high-resolution simulation with $\\alpha_{\\rm SS}=0.1$, showed how the disc grows under the influence of the neutron star.  In the initial build-up phase, the disc evolution is similar to that for isolated Be stars. But, later on, the effect of the resonant torque becomes apparent, preventing the disc gas from drifting outwards at several resonance radii. The effect is most remarkable at the 4:1 and 5:1 radii ($r/a \\sim 0.39$ and 0.33, respectively) for $\\alpha_{\\rm SS}=0.1$. As a result, the disc density increases more rapidly than that for isolated Be stars. This feature is consistent with \\citet{zam01}, who found that the discs in Be/X-ray binaries are about twice as dense as those of isolated Be stars.  Since the neutron star orbits in an eccentric orbit about the Be star, the interaction is phase dependent. The disc shrinks a little at periastron and then recovers gradually. Consequently, the disc emission will vary with phase. For our system with $e=0.34$, this effect turned out not to be large enough to cause an observable variation in the disc continuum as long as it is optically thin. It is possible, however, that the orbital modulation in the disc continuum in systems with much higher orbital eccentricity is observable. Moreover, it is likely that the profiles of optically thick emission lines from the disc, such as Balmer lines, will show an orbital modulation even in systems with moderate eccentricity, because the disc structure is made non-axisymmetric by the periodic excitation and damping of the spiral density wave.  In the Be disc in Be/X-ray binaries, an eccentric mode is excited through direct driving due to the (1,0) harmonic of the binary potential. In high-viscosity simulations, the mode grows initially linearly in time and then approaches a steady state, of which steady eccentric shapes based on linear theory of the evolution of a 3D eccentric decretion disc described in Section~\\ref{sec:ecc_mode} offer a fair explanation. The strength of the mode in the steady state is larger for a smaller value of $\\alpha_{\\rm SS}$.  On the other hand, in the high-resolution simulation with $\\alpha_{\\rm SS}=0.1$, the eccentric mode undergoes prograde precession at some point. The precession period is about $20P_{\\rm orb}$. The precession rate is not constant. It accelerates for $0 < \\omega_{\\rm d} < \\pi$ and decelerates for $\\pi < \\omega_{\\rm d} < 2\\pi$, where $\\omega_{\\rm d}$ is the angle between the eccentric vector of the disc and that of the binary orbit. The precession rate is radius-dependent. It is larger at a larger radius. Since the eccentric mode is leading for $0 < \\omega_{\\rm d} < \\pi$ and trailing for $\\pi < \\omega_{\\rm d} < 2\\pi$, the twist of the mode increases with time for $0 < \\omega_{\\rm d} < \\pi$ and decreases for $\\pi < \\omega_{\\rm d} < 2\\pi$. The strength of the eccentric mode, which anti-correlates with the twist of the mode, decreases for $0 < \\omega_{\\rm d} < \\pi$ and increases for $\\pi < \\omega_{\\rm d} < 2\\pi$.  As the mode precesses, the mass capture rate, $\\dot{M}_{\\rm acc}$, by the neutron star modulates.  It is a maximum at $\\omega_{\\rm d} \\sim \\pi$. Even for an eccentricity of 0.1, $\\dot{M}_{\\rm acc}$ at maximum is about a factor of three higher than the level before precession. If the disc had a much stronger disturbance, the enhancement in $\\dot{M}_{\\rm acc}$ is likely to be much larger. Such a system may temporarily show periodic X-ray outbursts.  In our model for Be/X-ray binaries, in which the disc material overflows toward the neutron star via the $L_1$ point, the mass capture rate by the neutron star becomes strongly phase dependent. The dependence is stronger for a smaller value of viscosity. In the disc with $\\alpha_{\\rm SS}=0.1$, the mass capture rate decreases by two orders of magnitude between periastron and apastron passages, and after apastron passage, no disc mass is captured by the neutron star. Note that our model gives a much stronger contrast in the mass capture rate than that expected for the stellar wind accretion.  We have also compared the simulated mass capture rate with the observed X-ray behaviour of 4U\\,0115+63, considering two extreme situations, in which $t_{\\rm acc} \\ll P_{\\rm orb}$ or $t_{\\rm acc} \\sim P_{\\rm orb}$, where $t_{\\rm acc}$ is the accretion time-scale. We found that the disc model for $\\alpha_{\\rm SS}=0.1$ gives a result consistent with the observation for both situations, whereas the higher viscosity models for $\\alpha_{\\rm SS}=1$ and 0.3 are ruled out unless $t_{\\rm acc} \\sim P_{\\rm orb}$.  Analysing multi-wavelength long-term monitoring observations of 4U\\,0115+63, \\cite{neg01} found that the Be star undergoes quasi-cyclic ($\\sim 3-5\\,{\\rm yr}$) activity, losing and reforming its circumstellar disc. They also found that, at some point, the growing disc becomes unstable to warping and then tilts and starts precessing. Type II X-ray outbursts take place after the warping episode. As shown in this paper, our $\\alpha_{\\rm SS}=0.1$ model explains many of the observed features of 4U\\,0115+63 in the phase before the warping occurs. Our simulation, however, showed no dynamical instability, and therefore is incapable of explaining the warping episode. Including the effect of radiation from the Be star may turn out to be essential to have a model that explains the whole cycle of the disc evolution, which is beyond the scope of this paper.  In this paper, we have concentrated our study on the viscous effects on star-disc interaction in a coplanar system with fixed orbital period and eccentricity. This was done not only as a first step to have a comprehensive understanding of the interaction between the Be disc and the neutron star in Be/X-ray binaries, but also to have an archetypal model with which we can compare the results from our future simulations. In the next paper, we will study the effects of the misalignment angles between the Be disc and the binary. We will discuss the effects of the orbital eccentricity in the third paper, and the fourth paper will conclude the series by studying the effects of the orbital period."
    },
    "0208/astro-ph0208077_arXiv.txt": {
        "abstract": "We report subarcsecond-resolution VLA and Keck mid-infrared imaging of the dwarf starburst galaxy II~Zw~40. II~Zw~40 contains a bright, compact thermal radio and infrared source with all the the characteristics of a collection of dense HII regions ionized by at least 14,000 O stars.  The supernebula is revealed to be multiple sources within an envelope of weaker emission. The radio emission is dominated by free-free emission at 2~cm, and the spectrum of this emission appears to be rising.  This suggests that the free-free emission is optically thick at 2 cm, and that the individual HII regions are $\\sim$1 pc in size. This complex of ``supernebulae\" dominates the total infrared luminosity of II~Zw~40, although the radio source is less than $\\sim$150~pc in diameter.   Multiple super star clusters appear to be forming here,  the much larger analogues of large Galactic HII region complexes. ",
        "introduction": "Starburst galaxies are distinguished by the concentration, and not simply the quantity, of star formation. Optical and UV studies with the resolution of the HST have found star formation regions to contain super-star clusters (SSCs), as populous and as dense as globular clusters but with ages of only a few million years. Observations in the radio and infrared are finding the even younger and obscured counterparts of these star clusters. Traced by optically thick free-free emission at centimeter wavelengths, radio/infrared ``supernebulae\" \\citep*{THB98,TBH00,G01} or ``ultra-dense HII regions\" \\citep{KJ99} have been discovered in nearby starburst galaxies. The discovery of these regions has been due to the development of subarcsecond imaging in the infrared and radio. These HII regions are bright and compact, excited by immense clusters of hot young stars, or ``starforming clumps\" \\citep{B01}. The sources have in common all of the signs of young star formation: high gas density \\citep[often $\\rm >10^4~cm^{-3}:$][]{THB98,KJ99,Tar00,M02}, a cm-wave continuum with spectral index from -0.1 to +1.5, substantial extinctions \\citep*[Av $>\\sim 10$ mag:][]{K89,H90}, and very strong mid-infrared emission \\citep*{G01,B01,Dale01,VJC02}. It is not just that these supernebulae are very luminous in the infrared, they may actually dominate the infrared emission of the entire galaxy. In the nearby dwarf galaxy NGC~5253, for example, at least 75\\% of the mid-infrared emission seen by IRAS and 25\\% of the total bolometric luminosity appear to come from one obscured star cluster less than 2~pc in diameter \\citep{G01}. In another starbursting dwarf, He 2-10, essentially all the mid-infrared emission is from star-forming clumps in a 200 pc by 50 pc disk \\citep*{B01,VJC02}. If infrared emission traces and quantifies star formation, as has been thought since IRAS, then since these clusters or supernebula are {\\it the} sources of infrared, they are also {\\it the} starburst.  II~Zw~40 is a starburst dwarf galaxy 10.5 ($\\rm H_o$/75) Mpc away. It appears to be the result of a collision of two smaller galaxies \\citep*{BST,BK88, Van98}. At optical wavelengths it is dominated by one bright star cluster \\citep{S70}, which is also a strong source of $H\\alpha$ \\citep{SS70} and associated nebular lines \\citep{WR}, has the Wolf-Rayet feature \\citep{VC}, and in general has all the signs of a young star formation region.  The infrared spectrum is dominated by high-excitation lines such as [NeIII] and [SIV] \\citep{R91, Thornley,B02}. The radio continuum emission is nearly all thermal emission \\citep{KWB91,KWT84,J78,D93}, and the emission is relatively compact \\citep{WWB,SW,KWB91}. Like the radio continuum, the H$\\alpha$ \\citep{SS70} and Brackett $\\gamma$ lines \\citep{WWB,MO,JL,Ve96,DSW} are bright in this galaxy . Molecular gas is conspicious by its absence; \\cite{S92} remark on its unusually high star formation efficiency. In these respects, II~Zw~40 resembles NGC 5253 \\citep{Cr99,THB98,TBH00}, probably the youngest, most massive, and most extreme example of an obscured young super star cluster or supernebula in the local universe.  Motivated by the  galaxy's similarity to NGC 5253, we imaged II~Zw~40 at 6, 3.6 and 2 cm using the Very Large Array\\footnote{The National Radio Astronomy Observatory is a facility of the National Science Foundation operated under cooperative agreement by Associated Universities, Inc.} in its A configuration for the maximum angular resolution, and at $11.7\\mu$m with the Long Wavelength Spectrometer (LWS) at the W. M. Keck Observatory.\\footnote{The W. M. Keck Observatory is operated as a scientific partnership among the California Institute of Technology, the University of California, and the National Aeronautics and Space Administration. The Observatory was made possible by the generous financial support of the W. M. Keck Foundation.}  The observations and results are described in the next section and the nature of the source in $\\S3$, in which we also compare the supernebula in II~Zw~40 to other dwarf starbursts and discuss the significance of the structure seen in the radio maps. ",
        "conclusions": "Our high-resolution radio and infrared observations of the starburst dwarf galaxy II~Zw~40 have found that the radio emission at 6, 3.6, and 2~cm is thermal free-free emission with $\\sim$70-75\\%, of the cm-wave emission confined to a region less than 3\\arcsec\\ in diameter. Within this region, there is a bright core of diameter $\\sim 1.5$\\arcsec\\ (75 pc) containing at least four sub-sources. The spectrum of the compact emission appears to be rising between 3.6 cm and 2 cm, suggesting that the free-free emission is optically thick at 2~cm, with implied emission measure of $\\rm EM\\sim 10^9~cm^{-6}\\,pc$ and electron density $\\rm n_e\\sim 3-4\\times 10^4~cm^{-3}$. For optically thick emission, the observed brightnesses require that the emitting sources are $\\sim$1 pc diameter HII regions, with $\\rm N_{Lyc}\\sim 6\\times 10^{51}~s^{-1}$. The 75 pc ``core\" starburst region  is a complex of four compact ``supernebulae\", each powered by $\\sim$600 O stars, with an estimated $\\rm \\sim 10^5~M_\\odot$ in young stars in each cluster. The supernebulae/clusters are separated by $\\sim$12~pc. The total ionization of II~Zw~40 is at least $\\rm N_{Lyc}=1.4\\times10^{53}~s^{-1}$, or 14,000 O7 star equivalents. For a Salpeter ZAMS IMF, we estimate that roughly 1-10 million young stars and a young stellar mass of $\\rm 2\\times10^6~M_\\odot$ are present within II~Zw~40. The total young stellar luminosity is $\\rm L_{OB}\\sim 3\\times 10^9~L_\\odot$. The OB luminosity is in good agreement with or perhaps even higher than the observed infrared luminosity of $\\rm L_{IR}\\sim 1.9\\times 10^9~L_\\odot$. The radio-infrared ratios in this galaxy differ from the usual starburst values; we attribute this to the extreme youth of the star formation activity, the low metal content of the gas, and a very weak cool dust component."
    },
    "0208/hep-ph0208012_arXiv.txt": {
        "abstract": "The latest $g_{\\mu}-2$ measurement by Brookhaven confirms the earlier measurement with twice the precision. However, interpretation of the result requires specific assumptions regarding the errors in the hadronic light by light (LbL) correction and in the hadronic vacuum polarization correction. Under the assumption that the analysis on LbL correction of Knecht and Nyffeler and the revised analysis of Hayakawa and Kinoshita are valid the new BNL result implies a deviation between experiment and the standard model of $1.6 \\sigma -2.6 \\sigma$ depending on the estimate of the hadronic vacuum polarization correction. We revisit  the $g_{\\mu}-2$ constraint for mSUGRA and its implications for the direct detection of sparticles at colliders and for the search for supersymmetric dark matter in view of the new evaluation. ",
        "introduction": "The BNL $g_{\\mu}-2$ Collaboration has announced\\cite{bnl2002} an improved result for $a_{\\mu}=(g_{\\mu}-2)/2$ with twice the precision of their 2001 analysis\\cite{bnl2001}. Here we interpret these results in the context of  supersymmetry. It has been recognized for some time that the muon anomalous moment can act as an important probe of physics beyond the standard model especially of supersymmetry. This is so because the new physics contribution to the leptonic anomalous magnetic moment $a_{\\it l}$ scales as $ m_l^2/\\Lambda^2$ and is proportional to the square of the lepton mass. Thus  $a_{\\mu}$  provides a more sensitive probe of  new physics than $a_e$ even though $a_e$ is more sensitively determined while the $a_{\\tau}$ determination is far less sensitive to be competitive. Because of the above any improvement in the determination of $a_{\\mu}$ has important implications for new physics beyond the standard model. Thus the improved measurement of BNL experiment in 2001\\cite{bnl2001} which gave $a_{\\mu}^{exp}=11659203(15)\\times 10^{-10}$ resulted in  a large amount of theoretical activity to explore the implications of the  results.  At the time of the BNL2001 result the standard model prediction consisting of the qed, electro-weak and hadronic corrections was estimated  to be $a_{\\mu}^{SM}= 11659159.7(6.7)\\times 10^{-10}$\\cite{czar1} which gave $a_{\\mu}^{exp}- a_{\\mu}^{SM}=43(16)\\times 10^{-10}$ implying a $2.6\\sigma$ difference between experiment and theory. The above difference was based on a  light by light (LbL) hadronic correction of \\cite{hayakawa,bijnens}  $a_{\\mu}^{had}(LbL)=-8.5(2.5)\\times 10^{-10}$ which was later found to be in error\\cite{knecht,hkrevised,Bijnens2,Blokland}. Thus the analysis of Knecht et.al.\\cite{knecht} gives \\cite{knecht} $a_{\\mu}^{had}(LbL)=8.3(1.2)\\times 10^{-10}$ while the revised analysis of Hayakawa and Kinoshita gives\\cite{hkrevised} $a_{\\mu}^{had}(LbL)=8.9(1.5)\\times 10^{-10}$. A partial analysis of $a_{\\mu}^{had}(LbL)$ has also been given in Refs.\\cite{Bijnens2,Blokland}. Thus the partial analysis of Ref.\\cite{Bijnens2} finds $a_{\\mu}^{had}(LbL)=8.3(3.2)\\times 10^{-10}$  and the work of Ref.\\cite{Blokland}  which computed the pion pole part finds $a_{\\mu}^{had}(LbL:\\pi^0 pole)=5.6\\times 10^{-10}$. Again the sign of these corrections agree with the sign of the reevaluation of this quantity in Refs.\\cite{knecht,hkrevised}. Using the average of the first two\\cite{knecht,hkrevised} which are the more complete calculations and using the hadronic vacuum polarization correction of Ref.\\cite{davier} one finds $a_{\\mu}^{SM}=11659176.8(6.7)\\times 10^{-10}$ and the BNL2001 result gives $a_{\\mu}^{exp}-a_{\\mu}^{SM}= 26(16)\\times 10^{-10}$. This difference corresponds only to a $1.6\\sigma$ deviation between theory and experiment.   Before proceeding further we wish to discuss a bit further the issue of errors in the hadronic corrections specifically the LbL correction and the $\\alpha^2$ vacuum polarization correction. Regarding the LbL hadronic correction, in addition to the analyses mentioned above, i.e., Refs.\\cite{knecht,hkrevised,Bijnens2,Blokland} there is also the analysis of Ref.\\cite{Ramsey} based on chiral perturbation theory. This analysis finds $a_{\\mu}^{had}(LbL) =(1.3^{+5}_{-6}+3.1\\tilde C)\\times 10^{-10}$ where $\\tilde C$ is a correction arising from the sub leading contributions. These contributions are either of $O(\\alpha^3 p^2/\\Lambda^2)$, where p is a mass of order $m_{\\mu}$ or $m_{\\pi}$ and $\\Lambda$ is a hadronic scale $\\sim 1$ GeV, which are not enhanced by a factor of $N_C$ (the number of quark colors) or of  $O(N_C\\alpha^3 p^2/\\Lambda^2)$ but are not enhanced by large logarithms. The coefficients $\\tilde C$ in this analysis is an  unknown low energy constant (LEC) which contains the contributions of nonperturbative physics at short distance. The authors of Ref.\\cite{Ramsey} view the parameter $\\tilde C$ as basically unconstrained  except through the measurement of the anomalous moment itself. In our analysis we assume the validity of the analyses of Refs.\\cite{knecht,hkrevised}. Aside from the LbL contribution, the other source of error  in the hadronic correction is the contribution from the $\\alpha^2$ hadronic vacuum polarization correction\\cite{davier,narison,yndurain,hadronic} and this error is an important component in extracting the deviation between experiment and the standard model.   The new  BNL result gives the world average on $a_{\\mu}$ as \\cite{bnl2002}. \\beq a_{\\mu}^{exp}=(11659203)(8)\\times 10^{-10} \\eeq With the LbL correction of Refs.\\cite{knecht,hkrevised} and assuming the leading hadronic correction to lie in the range $692(6)\\times 10^{-10}$ to $702(8)\\times 10^{-10}$\\cite{davier,narison,yndurain} the standard model prediction lies in the range $a_{\\mu}^{SM}=11659177(7)\\times 10^{-10}$ to $a_{\\mu}^{SM}=11659186(8)\\times 10^{-10}$ which leads to \\beqn a_{\\mu}^{exp}-a_{\\mu}^{SM}=(26)(10)\\times 10^{-10}\\nonumber\\\\ ~~~~~~~~~~~~~~~~~~~~to~~(17)(11)\\times 10^{-10} \\eeqn which corresponds to a difference between experiment and theory of about $1.6\\sigma$ to $2.6\\sigma$. ",
        "conclusions": "The new Brookhaven measurement of $g_{\\mu}-2$ has confirmed the earlier measurement with  twice the precision. However, interpretation of what this implies for physics beyond the standard model is very sensitive to the errors in the standard model prediction. The main surprise at the end of last year was the switch in sign of the LbL contribution. There is now a general agreement that the sign of this contribution is positive and also an agreement between two independent evaluations\\cite{knecht,hkrevised} on its size. An important exception to this is the result of chiral perturbation theory, which although a more fundamental approach is beset by the appearance of unknown low energy constants which require a nonperturbative approach such as lattice gauge calculation for its evaluation. In the present analysis we have assumed the validity of the analysis  on LbL  of  Knecht and Nyffeler and the revised analysis of  Hayakawa and Kinoshita. Taking account of all the errors the difference between experiment and the standard model is now predicted to lie between $1.6\\sigma$ and $2.6\\sigma$. We have revisited the implications of the $g_{\\mu}-2$ constraint for supersymmetry in view of the new result. Using the $2.6\\sigma$ difference and a $2\\sigma$ error corridor our analysis points to significant regions in the mSUGRA parameter space consistent with the $b\\rightarrow s+\\gamma$, relic density and $g_{\\mu}-2$ constraints. Within $1\\sigma$ error corridor all of the sparticles are accessible at the Large Hadron collider.  Further, for low values of $\\tan\\beta$ sparticles may also be  accessible at Fermilab Tevatron. We also carried out an analysis of the  neutralino-proton cross section $\\sigma_{\\chi^0-p}$ and find that a significant part of the parameter space would be explored by the future dark matter detectors such as CDMS (Soudan), GENIUS and ZEPLIN. Finally, the $g_{\\mu}-2$ experiment could lead to an even more stringent constraint if there was a reduction in the error associated with the standard model prediction. This could come about by eliminating the ambiguity on the LbL correction through a lattice gauge calculation analysis, and through improved low energy data on $e^+e^-\\rightarrow hadrons$ in the analysis of hadronic vacuum polarization correction.   ~\\\\ We thank William Marciano for an interesting discussion. This work is  supported in part by NSF grant PHY-9901057.\\\\"
    },
    "0208/astro-ph0208527_arXiv.txt": {
        "abstract": "We present a model of an outburst of the soft X-ray transient A0620-003. A two-dimensional time-dependent smoothed particle hydrodynamics scheme is used to simulate the evolution of the accretion disc through a complete outburst. The scheme includes the full tidal potential of the binary and a simple treatment of the thermal-viscous disc instability. In the case where the mass accretion rate onto the primary determines the fraction of the disc that can be kept in a hot, high viscosity state by the resulting X-ray emission, we find that the shape of the X-ray light curve is ultimately determined by the relative sizes of the irradiated and unirradiated parts of the disc and the growth time-scale of the tidal instability. The model accounts for the rebrightening that has been observed in the light curves of A0620-003 and several other transients. The primary maximum and subsequent decline are due to the accretion of gas within the irradiated portion of the disc, while the secondary maximum is caused by the accretion of gas in the outer part of the disc that is initially shadowed from the central X-rays, but subject to tidal forces. We propose that tidal effects at the disc edge can be sufficient to drive accretion on a time-scale shorter than that expected for a standard alpha-viscosity disc. The final decay is subsequently controlled by the gradual retreat of the irradiated portion of the disc. If the entire disc is kept in the high-viscosity state by the irradiation, no rebrightening is possible. ",
        "introduction": "The soft X-ray transients (SXTs) are close interacting binary systems consisting of a black hole or neutron star primary and a low-mass Roche-lobe-filling secondary star. The outbursts of soft X-ray transients are quite distinct from those seen in other binary systems such as dwarf novae. The duration and recurrence times are both strikingly longer, suggesting that a large fraction of the mass of the accretion disc is consumed in each outburst and that a different mechanism from the standard disc instability picture is required (comprehensive reviews of the disc instability picture are given by Cannizzo, 1993 and Lasota, 2001).  Typical ionization instability-driven dwarf nova outbursts last a few days and consume a few percent of the disc mass, leading to a recurrence time of the order of weeks. In comparison, a typical SXT outburst lasts for 100 days or more and recurs on a time-scale of several years. The profiles of the X-ray light curves of SXT outbursts vary greatly but they do share common characteristics - the initial peak is often followed by an exponential decline and, in many cases, a rebrightening several tens of days after the first maximum. Smaller rebrightenings later on in the decay from outburst have also been observed. These features can be seen quite clearly in the X-ray light curve of the well-known transient A0620-003 (Elvis et al. 1975, Whelan et al. 1977). A general review of SXT outbursts is given by Tanaka \\& Lewin (1996).  Several authors have proposed models to explain the form of the SXT light curves. The long recurrence time of the outbursts is a problem which has been addressed in many ways. It can be explained by the standard ionization instability model if the Shakura-Sunyaev viscosity parameter $\\rm \\alpha$ is greatly reduced from that typical of a cataclysmic variable disc. Such an explanation is not confined to SXTs; it has been invoked to explain the outbursts of WZ Sagittae (Smak, 1993). However, there is no obvious physical reason why $\\rm \\alpha$ should vary so much from one accretion disc to another. The more successful models involve the removal of the inner part of the disc where the viscous time-scale is short. This can been achieved by a hot, advection-dominated accretion flow (ADAF) onto the hole : Meyer \\& Meyer-Hofmeister (1994) showed that gas from the cool outer accretion disc in a binary system may evaporate into a hot corona and subsequently feed an ADAF onto the comapct object. Dubus et al. (2001) have performed one-dimensional simulations of an SXT including the effects of both irradiation and an ADAF. In this work we do not address the question of recurrence time, but concentrate on the observed outburst profiles. Cannizzo, Chen \\& Livio (1995) found that the exponential decay could be reproduced by parameterising $\\rm \\alpha$ in terms of the aspect ratio $\\rm \\frac{H}{R}$. Augusteijn, Kuulkers \\& Shaham (1993) sought to explain the rebrightenings by considering irradiation of the secondary as a mechanism for increasing the mass transfer rate. However, it is unlikely that the radiation can penetrate the photosphere of the companion to produce a variation in mass transfer rate of more than a few percent (King, 1989).  We consider the effect of disc irradiation by the X-ray emission from the compact object. King \\& Ritter (1998) showed analytically that the assumption that the central X-ray  source can efficiently irradiate the disc leads automatically to long exponential X-ray decays with the observed time-scales. Observations of SXTs suggest that the optical/X-ray flux ratio for these objects is simply too large for the optical emission to result from intrinsic viscous dissipation in the disc alone (van Paradijs \\& McClintock, 1994; Shahbaz \\& Kuulkers, 1998). Moreover, the optical brightness is strongly correlated with the irradiating X-ray flux. However, detailed one-dimensional modelling of the disc structure (e.g. Cannizzo 1994) suggests that the response of the disc to the irradiation is for the surface to become convex rather than concave, thus shielding the outer regions from the irradiating flux. The observations seem to contradict this picture, and for the purposes of this paper we assume that the disc surface remains concave.  In section 2 we describe the irradiated disc model of King and Ritter in more detail, and outline the importance of tidal forces in systems with extreme mass ratios in section 3. The numerical method used to implement the model is then described in section 4. In section 5, we show that the observed peak luminosity of the 1975 outburst of A0620-003 almost certainly implies that the irradiation radius (the outer radius of the region kept heated by the central X-rays) was larger than the accretion disc. We then show in simulations that no rebrightening occurs if the whole disc is kept in the hot viscous state by the irradiation; however, if only a very small area of the outer disc is shielded in some way (by a radiation-induced warp or self-shadowing, for example), the observed rebrightening is produced. ",
        "conclusions": "We have simulated complete outbursts of soft X-ray transient systems, for the first time  in two dimensions. The  rebrightenings seen in the X-ray light curves of several systems can be explained by competition between the ionization instability, the effects of disc irradiation and tidal forces from the secondary star. Their appearance in the light curve is ultimately determined by the irradiated area of the disc, the location within the disc where the disc instability is first triggered and the growth time-scale of the tidal instability. The main findings of this work can be summarized as: \\begin{itemize} \\item{A rebrightening can occur if a small portion of the disc is shielded from the central X-rays in some way. This complements the conclusions of Dubus et al. (1999), who showed that a planar disc cannot be kept heated by irradiation from the central source, in contradiction to observations.} \\item{Tidal torques can be a dominant factor in the energy balance over a large area of the disc. Such an enahnced dissipation in a disc with the classical $\\rm \\alpha$-model structure will drive accretion on a shorter than expected time scale.} \\item{The time delay between the onset of the outburst and the start of the rebrightening is determined by the growth time-scale of the eccentricity and spiral waves in the disc associated with the tidal instability.} \\item{The rise-time of the rebrightening is governed by the disc viscosity - but tidal eccentricity effects cause this time-scale to be much shorter than expected for the arrival of gas from the disc edge.} \\item{No rebrightening is reproduced if the entire disc is irradiated.} \\end{itemize}  The main limitation of this approach is that the viscous evolution of the disc is governed by the local surface density and not a local mid-plane temperature. The ultimate goal is to merge the current 1D thermodynamic state-of-the-art  with our 2D approach. The eventual extension to three dimensions is also particularly desirable in examining the beahviour of the tidal forces and the dynamics of the spiral arms.  There already exists a wealth of X-ray observations of transient sources, and we stress that although some aspects of the X-ray light curves are found to be common to many SXTs, they are all different to some degree. In this work we have presented a mechanism for the outbursts which explains the origin of the rebrightenings observed in the X-ray light curves of some of these systems.  Although one would not expect to observe superhumps in the X-ray light curve, superhumps have been seen in optical light curves of SXT outbursts (O'Donoghue \\& Charles 1996). The superhump mechanism must be slightly different to that operating in cataclysmic variables, as the optical emission from the disc is completely dominated by the reprocessed X-rays. Rather, the superhump modulation can be attributed to the change in visible area of an eccentric disc over an orbital period (Haswell et al. 2001). This is exactly the scenario which we find in this work, and is a clear demonstration of the importance of using a two- (or preferably a three-) dimensional code able to take the tidal effects into account in a natural way."
    },
    "0208/astro-ph0208461_arXiv.txt": {
        "abstract": "We investigate the possibility to find evidence for planets in circumstellar disks by infrared and submillimeter interferometry. Hydrodynamical simulations of a circumstellar disk around a solar-type star with an embedded planet of 1 Jupiter mass are presented. On the basis of 3D radiative transfer simulations, images of this system are calculated. These intensity maps provide the basis for the simulation of the interferometers VLTI (equipped with the mid-infrared instrument MIDI) and ALMA. While ALMA will provide the necessary basis for a direct gap and therefore indirect planet detection, MIDI/VLTI will provide the possibility to distinguish between disks with or without accretion on the central star on the basis of visibility measurements. ",
        "introduction": "Hydrodynamical simulations concerning the evolution of protoplanets in protoplanetary disks have shown that giant protoplanets may open a gap and cause spiral density waves in the disk (see, e.g., Kley~1999, Kley et al.~2001, D'Angelo et al.~2002). Depending on the hydrodynamical properties of the planet and the disk, the gap may extend up to several AU in width. We investigate the possibility to find such a gap as an indicator for the presence of a protoplanet with present-day or near-future techniques. For this reason, we use hydrodynamical simulations of a protoplanetary disk with an embedded planet and compute the expected brightness distributions. We show that ALMA will provide the necessary basis to detect gaps in circumstellar disks in the millimeter/submillimeter wavelength range. ",
        "conclusions": "\\label{concl}  We studied the possibility of the indirect planet detection by observation of the resulting gap in a protoplanetary disk. Based on hydrodynamical simulations and subsequent 3D continuum radiative transfer calculations we generated images of a circumstellar disk with an embedded Jupiter mass planet surrounding a solar-type star.  We found that the gap can be seen very clearly in the simulated images but an extremely high angular resolution is required ($\\approx$\\,10\\,mas). Furthermore, because of the extreme brightness contrast in the innermost region of the disk in the near to mid-infrared, the gap can hardly be detected in this wavelength range with imaging/interferometric observations. However, we found that it will be possible to distinguish between a disk with or without accretion onto the central star with MIDI. In contrast to this, the (sub)millimeter interferometer ALMA will provide the basis for the reconstruction of an image of a gap. Thus, the search for massive protoplanets in circumstellar disks can be based on the indication of a gap."
    },
    "0208/astro-ph0208182_arXiv.txt": {
        "abstract": "s{Antiprotons and antideuterons are considered as probes to look for primordial black holes in our Galaxy. I give a brief overview of the latest developments on the subject.} ",
        "introduction": "Primordial black holes (PBHs) could have formed in the early universe from the collapse of overdense regions through significant density fluctuations. Their detection nowadays is a great challenge as it could allow both to check the Hawking evaporation mechanism and to probe the early universe on very small scales that remain totally out of the range investigated by CMB or LSS measurements. They have recently been searched by their gamma-ray radiation \\cite{MacGibbon2} \\cite{Carr3}, extremely high-energy cosmic-ray emission \\cite{Barrau3}, and antiproton emission \\cite{Orito}. This brief paper gives the latest improvements obtained with antiprotons and antideuterons. Such antinuclei are very interesting as the background due to spallation of cosmic protons and helium nuclei on the interstellar medium is expected to be very small. ",
        "conclusions": "Primordial black holes have been used to derive interesting limits on the scalar fluctuations spectrum on very small scales studies \\cite{Kim2} \\cite{Po}. It was also found that {\\sc pbh}s are a great probe of the early Universe with a varying gravitational constant \\cite {Carr2}. Significant progress has been made in the understanding of the evaporation mechanism itself, both at usual energies \\cite{Parikh} and in the near-planckian tail of the spectrum \\cite{Barrau2} \\cite{Stas2}. Looking for PBHs or improving the current upper limit is therefore a great challenge for the forthcoming years."
    },
    "0208/astro-ph0208475_arXiv.txt": {
        "abstract": "Using the results of our first paper on the {\\em Chandra} HRC observation of the Orion Nebula Cluster (ONC), here we explore the relation between the coronal activity of its 1-Myr-old pre-main sequence population and stellar parameters.  We find that median X-ray luminosities of low mass stars ($M/M_{\\odot} \\lesssim 3$) increase with increasing mass and decreasing stellar age. Brown dwarfs ($0.03 \\lesssim M/M_{\\odot} \\lesssim 0.08$) follow the same trend with mass.  From $M \\sim 0.1$ to $M \\sim 0.5~M_{\\odot}$, median $L_X/L_{bol}$ values increase by about half an order of magnitude and then remain constant at $\\sim 10^{-3.5}$ for the mass range from 0.5 to 3.0 $M/M_{\\odot}$. In these same two mass ranges, $L_X/L_{bol}$ remains roughly constant with age, until it drops by more than two orders of magnitudes at the epoch when $\\sim 2-4 M_\\odot$ stars are expected to become fully radiative.  We find a dependence of $L_X$ and $L_X/L_{bol}$ on circumstellar accretion indicators and suggest three possible hypotheses for its origin. In spite of improved X-ray and rotational data, correlations between activity indicators and rotation remain elusive for these stars, possibly indicating that stars for which rotational periods have been measured have reached some saturation level.  Our study of X-ray activity vs. stellar mass leads us to propose that the few HRC X-ray sources not associated with any optical/infrared counterpart trace a yet to be discovered stellar population of deeply embedded, relatively massive ONC members. ",
        "introduction": "The present work focuses on the study of the X-ray activity of Orion Nebula Cluster (ONC) members: from a purely observational standpoint, we explore the relationship between X-ray activity and stellar characteristics, searching for the physical mechanisms responsible for coronal X-ray activity in pre-main sequence (PMS) stars. In spite of the previous observational work in this area \\citep[e.g.,][]{fei93,cas95,gag95a,fla00a}, no definitive picture has yet emerged.  Our investigation is based on the original {\\em Chandra} High Resolution Camera \\citep[HRC, ][]{mur00} X-ray data and literature optical data  presented in \\citet[][ hereafter Paper~I]{papI} and \\citet{fla02} for objects in the ONC area. For the ONC region under study, Paper~I defined an {\\em optical sample} of 696 optically selected, extinction limited, and well characterized ONC members. Here we address the controversial questions of the relationship of X-ray activity with convection and rotation (i.e., the classical $\\alpha-\\omega$ dynamo parameters), with stellar mass and age, and with circumstellar accretion and/or the presence of circumstellar disks. Taking advantage of our detailed description of activity as a function of stellar mass, we then explore the nature of the few detected X-ray sources not associated with any optical/infrared (IR) counterparts.  The structure of this paper is as follows: In \\S \\ref{sect:XvsOpt} we study the relation of activity with rotation, mass, age and circumstellar accretion strength. In \\S \\ref{sect:unid} we speculate on the nature of X-ray sources with no optical/IR counterpart. Finally we discuss our findings and summarize our conclusions in \\S \\ref{sect:disc}. An Appendix details our technique for investigating possible sources below our X-ray sensitivity limit. ",
        "conclusions": "}  Using {\\em Chandra} X-ray data and recent optical/IR data from the literature, both presented in Paper~I, we have studied relationships among X-ray activity indices and various stellar parameters for an optically-selected sample of ONC members. Our ultimate goal is to explain on physical grounds the mechanisms responsible for X-ray emission and magnetic activity in PMS stars.  We do not find any correlation of either $L_X$ or $L_X/L_{bol}$ with stellar rotational period. This result has sometimes been regarded as evidence against the $\\alpha$ - $\\omega$ dynamo explanation of activity for PMS stars. We note however that: 1) Although the sample of stars for which we have a measure of rotational period is larger than ever considered in similar past studies, it is still biased toward more active stars, as judged from distributions of X-ray activity indicators; 2) Many of the sources with known rotational properties may thus have saturated activity levels and the intensity of their X-ray emission may therefore be insensitive to the rotational period; 3) 90\\% of these rotational periods are shorter than 10 days; 4) As discussed by \\citet{fla02}, periods up to $\\sim 16$ days might correspond to saturated activity for $\\sim$1~Myr-old stars, according to the {\\em consolidated} activity~--~Rossby~number relation.  We therefore do not regard the present lack of evidence of a rotation-activity relationship as conclusive for the ONC and look forward the realization of more complete rotational period databases.  We also note that even if the measured rotational periods were representative of the whole ONC population (as recently indicated by \\citealt{rho01}), a lack of correlation does not necessarily require a new explanation of activity.  As previously noted for other star forming regions \\citep[e.g.,][]{fei93}, we find a significant correlation of $L_X$ with mass for $M \\lesssim 3.0~M_{\\odot}$. The logarithm of the median $L_X$ rises quite linearly on a logarithmic mass scale from $Log(M/M_{\\odot}) \\sim -0.9$ to $\\sim 0.4$, but for higher masses, drops sharply before rising again above $Log(M/M_{\\odot}) \\gtrsim 1.0$. The drop in activity at $M \\sim 3.0~M_{\\odot}$ is even more dramatic in $L_X/L_{bol}$ (more than two orders of magnitude) and may be further enhanced if proper account for multiplicity were taken. Supported by recent PMS evolutionary models (SDF) we interpret this drop as the effect of the disappearance of a convective envelope for stars more massive than $\\sim 3~M_{\\odot}$. In general we note a convincing correlation between the presence of a convective layer and activity for $M < 10~M_{\\odot}$.   For $0.5 \\lesssim M/M_{\\odot} \\lesssim 3.0$, the median $Log(L_X/L_{bol})$ is observed to be quite stable at about -3.5, lower than (though close to) the saturation level. At lower masses however we observe a decrease of the median $L_X/L_{bol}$ with decreasing mass, as it falls by a factor of $\\sim 2.8$ ($0.45$ dex) between $M=0.5-1.0$ and $M=0.1~M_{\\odot}$. We are unsure of the origin of this mass dependence but note a rough coincidence between the turning point in the $L_X/L_{bol}$ - mass relation, at $\\sim 1~M_{\\odot}$, and the mass at which 1~Myr stars cease to be fully convective (cf. Figure \\ref{fig:LXLbvsM4}).  We suggest this may be due to a transition from an $\\alpha$--$\\omega$ dynamo to a fully-convective dynamo.  It may be even more relevant to note that the median value of the Ca~II line equivalent width, which can be used as a proxy of disk accretion (cf. \\S \\ref{sect:CaII}), is significantly lower for stars with masses below 0.5$M_\\odot$ (median $EW$=-0.15) than for stars with mass between 0.5 and 3.0$M_\\odot$ (median $EW$=0.94), possibly indicating a larger fraction of accreting stars at low masses\\footnote{There is indeed growing evidence that disk lifetime is longer for lower mass stars. Two examples of such a result, based on detection of disks in the IR, can be found in \\citet{hil98a} for the ONC and \\citet{hai01} for IC~348.}. We therefore propose (cf.  \\citealt{fla02}) that the increase of $L_X/L_{bol}$ at low masses {\\em might} be the effect of a decrease, with increasing stellar mass, of the fraction of stars that are accreting and/or are surrounded by disks. We indeed found such stars to have significantly lower activity levels (\\S \\ref{sect:CaII}).  Although we do not detect any securely identified brown dwarfs, we are able to estimate the mean X-ray luminosity for the 9 spectroscopically confirmed BDs of the BD bin in Figure \\ref{fig:BD}. The mean $L_X$ of $\\sim 10^{28.5}$~erg$\\cdot$s$^{-1}$ appears to be on the same $L_X$ vs. mass relationship we observe for higher mass stars. This is not particularly surprising because at this evolutionary stage BDs do not fundamentally differ from low mass fully convective PMS stars.  We have evidence of the presence of a population of very weak emitters at the low mass end. In the $0.25-0.50~M_{\\odot}$ mass bin, for example, we observe that the mean X-ray luminosity of stars individually not detected in the X-ray data (Paper~I) is $\\gtrsim 1.5$ orders of magnitude lower with respect to the mean of the whole sample, indicating either a very wide or a bimodal X-ray luminosity distribution. At least part of this dispersion can be explained by a dependence of activity on circumstellar accretion rate. By discriminating stars with high and low circumstellar accretion on the basis of the Ca~II ($\\lambda = 8542\\AA$) equivalent width, we prove with high statistical significance that the X-ray activity of the two groups differs by as much as an order of magnitude. The relationships of median $L_X$ and $L_X/L_{bol}$ with mass appear to hold exclusively for the low accretion sample.  We also note that the scatter in activity at a given mass appears to be larger for the high accretion sample than for the low accretion sample. Although the origin of these differences is presently unknown, we put forward three possible scenarios:   1) Accreting stars are rotating slowly due to disk breaking and their dynamo efficiency depends on rotation, while those accreting less have broken free of their disks to spin-up and thus saturate their activity.  This hypothesis seems to be contradicted by two facts: low and high accretion stars have statistically indistinguishable rotational periods (cf. Figure \\ref{fig:P_dist}) and, as  reported in \\S \\ref{sect:prot}, we do not observe a correlation   between activity and $P_{rot}$. However, the sample of stars for which we have rotational information is not complete and may be subject to biases (see \\S \\ref{sect:prot}), so that we cannot rule out this  possibility until a more representative sample of rotational periods becomes available.   2) Accretion and/or the presence of a disk, and/or  outflows influences coronal geometry, for example, by decreasing the fraction of the stellar surface available for the closed magnetic structures from which X-ray emission is thought to emanate. Other scenarios of altered geometries have been proposed; e.g., \\citet{mon00} argue that coronal structures might extend to the inner part of disk. Because of the inhomogeneous and time variable nature of accretion, variability studies and simultaneous X-ray/optical observations could help clarify this matter.  3) Accreting stars have higher X-ray extinction than assumed, so that the difference in inferred $L_X$ and $L_X/L_{bol}$ is only apparent. Although in Paper~I our ACIS spectra analysis was not able to exclude this possibility with high confidence, due mainly to the low photon statistics of  sources associated with high accretion stars, we also found no supporting evidence for this scenario. Moreover, an empirical determination of the $N_H$ vs. $A_V$ relation for the $\\rho$ Ophiuchi population (cf. \\citealt{ima01}) also does not support this hypothesis. The reason for the departure from the assumed $A_V$-$N_H$ relation could be that the gas to dust ratio may differ for accreting stars with respect to the average interstellar value, resulting in an underestimation of the X-ray absorption. One could imagine a scenario in which  accretion gas columns that cross the line of sight would obscure X-rays and let optical/IR radiation through. The most obvious test for this hypothesis is a better determination of hydrogen absorbing columns through deeper (i.e., longer exposures) medium spectral resolution X-ray observations. As for the previous hypothesis, variability studies might also give useful clues.  Finally we have investigated the nature of sources detected in our HRC X-ray data and not identified with any optical/IR object. On the basis of their X-ray spectra and location in the cloud we conclude that, as a class, they are subject to high extinction. We exclude an extragalactic nature for most of these objects by considering the sky density of such objects as a function of X-ray flux and of the molecular cloud total extinction and conclude that most unidentified X-ray sources are likely associated with deeply embedded, high intrinsic X-ray luminosity, ONC members. Given the relationship between $L_X$ and stellar mass we infer that these stars are most likely fairly massive ($M \\gtrsim 1.0~M_\\odot$), thus indicating the likely presence of a significant number of lower mass, as yet undetected ONC members.   In light of these results, we propose this scenario:  1) Activity in $\\sim 1~Myr$ old low mass PMS stars is ultimately due to a dynamo mechanism that requires a convective layer in order to function. Rotation may or may not be a fundamental ingredient of this dynamo.  2) The same mechanism responsible for X-ray emission of low mass stars is probably also at work in brown dwarfs.  3) Many of our stars appear to be saturated. The fraction of saturated stars, or alternatively the saturation level, depends on stellar mass. At low masses we might be seeing the effect of decreasing accretion/disk fraction with increasing mass, or the transition between two kinds of dynamos:  one that functions on fully convective stars and brown dwarfs and in which the saturation level depends on mass, and another at work on partially convective - partially radiative stars, characterized by a constant saturation level.  4) Highly accreting  stars are seen to be less active with respect to low accretion counterparts, possibly because of (a) different rotational properties, (b) disk/accretion induced modifications of the coronal magnetic field geometry, or (c) anomalous interstellar X-ray absorption due to the presence of circumstellar material.  Additional deep X-ray and IR observations are needed to advance our understanding and enable us to identify which scenario(s) Nature prefers."
    },
    "0208/astro-ph0208019_arXiv.txt": {
        "abstract": "We analyze a high-resolution, high signal-to-noise spectrum of the carbon-rich, extremely metal-poor star CS~29498--043, obtained with the Subaru Telescope High Dispersion Spectrograph. We find its iron abundance is extremely low ([Fe/H] = $-3.7$), placing it among the few stars known with [Fe/H] $\\le -3.5$, while Mg and Si are significantly overabundant ([Mg/Fe] = +1.8, and [Si/Fe] = +1.1) compared with stars of similar metallicity without carbon excess.  Overabundances of N and Al were also found. These characteristics are similar to the carbon-rich, extremely metal-poor star CS~22949--037. Though the sample is small, our discovery of CS~29498--043 suggests the existence of a class of extremely metal-poor stars with large excesses of C, N, Mg, and Si. ",
        "introduction": "\\label{sec:intro}  The most metal-poor stars in the Galactic halo are believed to contain the material ejected from the first generations of stars.  To investigate the nucleosynthetic yields of supernovae, the chemical compositions of a number of extremely metal-poor stars have been studied recently \\citep[e.g., ][]{mcwilliam95, ryan96}. Some objects show distinct chemical characteristics considered to result from nucleosynthesis in a single massive star and its supernova.  An extreme example is CS~22949--037. \\citet{mcwilliam95} found this object to be an extremely metal-poor ([Fe/H] $\\sim -4.0$)\\footnote{[A/B] = $\\log(N_{\\rm A}/N_{\\rm B})- \\log(N_{\\rm A}/N_{\\rm B})_{\\odot}$, and $\\log \\epsilon_{\\rm A} = \\log(N_{\\rm A}/N_{\\rm H})+12$ for elements A and B.} giant with $\\alpha$--element excesses. \\citet{norris01} confirmed the excesses of C, Mg and Si, compared with iron, and discovered an extremely large enhancement of nitrogen ([N/Fe] = +2.7). \\citet{depagne02} found significantly large excesses of oxygen and sodium ([O/Fe] = +2.0, and [Na/Fe] = +2.1). Comparisons with model predictions for the yields of supernovae \\citep[e.g., ][]{woosley95,fryer01,heg02} with zero heavy elements have been made to explain its abundance characteristics.  During our study of very metal-poor stars using the University College London coud\\'e \\'echelle spectrograph (UCLES) at the Anglo-Australian Telescope (AAT) (e.g., Norris et al. 1996), we found that the carbon-rich object CS~29498--043 exhibited strong Mg lines. To conduct a more detailed study, we obtained a high-resolution spectrum with the High Dispersion Spectrograph (HDS) of the Subaru Telescope \\citep{noguchi02}. Our analysis shows that CS~29498--043 is an extremely iron-deficient giant ([Fe/H] $=-3.7$) which, contrary to most stars of similar metallicity, exhibits large excesses of C, N, Mg, and Si compared with Fe. These characteristics are similar to, but more extreme than, those of CS~22949--037.  In this Letter we report on the composition of the carbon-rich, extremely metal-poor star CS~29498--043. Its CH and CN bands are as strong as another carbon-enhanced, extremely metal-poor star CS~22957--027 ([Fe/H]$\\sim -3.4$, Norris et al. 1997b; Bonifacio et al. 1998), which we also analyzed using a spectrum obtained with HDS for comparison. ",
        "conclusions": "\\label{sec:disc}  Our analysis shows that CS~29498--043 is an extremely metal-poor ([Fe/H]$=-3.75$) star with a large excess of C, N, Mg, and Si. These characteristics are similar to those of CS~22949--037 \\citep{mcwilliam95, norris01, norris02}. In Figure \\ref{fig:mgfe}, their [Mg/Fe] values are shown as a function of [Fe/H], along with others from previous works compiled by Norris et al. (2001). The [Mg/Fe] of CS~29498--043 is clearly much higher than the average for other stars with similar [Fe/H], and is even higher than the already extreme CS~22949--037. The [Mg/Fe] of CS~22957--027 follows the trend of the stars with similar [Fe/H], as studied by \\citet{norris97b} and \\citet{bonifacio98}.  Figure \\ref{fig:abpat} shows the abundance differences ([X/Fe]$-<$[X/Fe]$>$) for CS~29498--043 and CS~22949--037 (Norris et al. 2001) relative to the average values of the four stars CD$-24^{\\circ}$17504, CD$-38^{\\circ}$254, CS~22172-002, and CS~22885-096 studied by \\citet{norris01}, as a function of atomic number. Excesses of C, N, Mg, and Si clearly appear in both CS~29498--043 and CS~22949--037, while there is no clear evidence of departure of the relative abundances from zero for Sc-Ni. This suggests that similar nucleosynthesis processes contributed to the abundance patterns of these two stars.  The high [Mg/Fe] and [Si/Fe] values presumably indicate that the heavy elements from which these two stars formed came from supernovae whose outer layers (where C and Mg are produced) escaped, but where relatively little material escaped from nearer the iron core. Clearly, the nucleosynthesis mechanism is distinct from that in the Asymptotic Giant Branch (AGB) stars responsible for high C and s-process abundances in some, but not all, carbon-enhanced, metal-poor stars.  While the excesses of Si in CS~29498--043 and CS~22957--027 are similar, the excess of Mg in CS~29498--043 is noticeably larger than that of CS~22949--037; the difference in [Mg/Fe] is 0.59~dex, significant at a 2.1-sigma level when we adopt the results of \\citet{norris01} for CS~22949--037. (The significance is 1.8-sigma if we adopt the result of Depagne et al. 2002.) The Al abundance of CS~29498--043 is also higher than that of CS~22949--037, although the uncertainty of the Al abundance is large due to the contamination from the CH line and the strong (but uncertain) NLTE effect for this element \\citep[e.g., ][]{baumuller97}. In addition to the small excess of the relative Al abundance, a large enhancement of Na was also found in CS~22949--037 \\citep{depagne02}. Studies of odd-$Z$ elements like Na and K in CS~29498--043 are desirable to investigate the nucleosynthesis processes which produced its abundance pattern.  Measurement of its oxygen abundance, which has a large excess in CS~22949--037 \\citep{depagne02}, is also of importance.  The fraction of stars which are carbon-enhanced increases with decreasing metallicity \\citep{rossi99}. High-resolution spectroscopy has been carried out for more than 10 carbon-rich stars with $-3.0 < ${\\rm [Fe/H]}$ < -2.0$. While some show large excesses of $s$-process elements \\citep[e.g., ][]{aoki02c}, others show no such excess \\citep{aoki02b}. No other carbon-rich object with such large excesses of Mg and Si as CS~29498--043 is known (Aoki, et al. 2002, in preparation).  However, only 4 objects with [C/Fe]$\\gtrsim 1.0$ are known in the metallicity range of [Fe/H] $<-3.0$. One is the carbon- and nitrogen-enhanced star CS~22957--027 studied by \\citet{norris97b}, \\citet{bonifacio98}, and in the present work. Another is the $r$-process-enhanced star CS~22892--052 \\citep{sneden96}. The excesses of carbon and nitrogen of that object ([C/Fe]$\\sim$[N/Fe]$\\sim$1.0) are much smaller than those of CS~22957--027, while the abundance pattern of the elements with $12\\lesssim Z\\lesssim 28$ is similar to those of the other objects with normal carbon abundances (McWilliam et al. 1995; Norris et al. 1997a).  The other two known carbon-rich stars with [Fe/H]$\\lesssim -3$ are CS~22949--037 and CS~29498--043. These objects have lower iron abundances ([Fe/H] $< -3.5$) than CS~22957--027 and CS~22892--052 ([Fe/H] $\\sim -3.0$), and also have large excesses of Mg and Si, as shown here. Though the sample is too small to permit definitive conclusions, the similarity of CS~22949--037 and CS~29498--043 suggests that other extremely metal-poor ([Fe/H]$\\lesssim -3.5$) stars with carbon excess may also show a large enhancement of Mg and Si. Since the ratio of carbon-rich objects seems to increase with decreasing metallicity, a number of objects similar to these two stars perhaps exist.  Further spectroscopic studies of candidate extremely metal-poor stars with strong CH and CN features are essential to investigate the nucleosynthesis processes in zero-(or very low-) metallicity stars in the early Galaxy.  Comparisons with theoretical predictions of yields ejected from supernovae will be presented separately in a future paper; here we emphasize the importance of the discovery of CS~29498--043, which suggests the existence of a class of extremely metal-poor star with excesses of C, N, Mg, and Si."
    },
    "0208/astro-ph0208533_arXiv.txt": {
        "abstract": "{% Smoothing is omnipresent in astronomy, because almost always measurements performed at discrete positions in the sky need to be interpolated into a smooth map for subsequent analysis.  Still, the statistical properties of different interpolation techniques are very poorly known.  In this paper, we consider the general problem of interpolating discrete data whose location measurements are distributed on the sky according to a known density distribution (with or without clustering).  We derive expressions for the expectation value and for the covariance of the smoothed map for many interpolation techniques, and obtain a general method that can be used to obtain these quantities for any linear smoothing. Moreover, we show that few basic properties of smoothing procedures have important consequences on the statistical properties of the smoothed map.  Our analysis allows one to obtain the statistical properties of an arbitrary interpolation procedure, and thus to optimally choose the technique that is most suitable for one's needs. ",
        "introduction": "\\label{sec:introduction}  A common problem in Astronomy is the smoothing of irregularly sampled data.  In general, this happens when one can make measurements on discrete points of a quantity that has some astrophysical relevance. In most cases, the ``quantity'' is a \\textit{continuous field}, i.e.\\ a smooth function on the sky.  In this case it is reasonable to try to reconstruct the field by interpolating the discrete measurements obtained.  Suppose, for example, that we are interested in measuring the column density of a nearby molecular cloud.  A good estimate of the cloud column density can be obtained, for example, using the infrared color of background stars observed through the cloud \\citep[see, e.g.,][]{2001A&A...377.1023L}.  This way, we can obtain reliable measurements of the ``column density field'' at the discrete points corresponding to the star positions.  Finally, we interpolate the various measurements and obtain a smooth map.  Similar situations are often encountered in different fields (e.g., weak gravitational lensing, peculiar velocity field of galaxies).  Interpolation and smoothing are ubiquitous in Astronomy, and indeed many papers have been devoted to the study of the effects of interpolation in particular analyses \\citep[see, e.g.][]{1992ApJ...398..169R}.  However, it is important to observe that many different approaches to the study of interpolation are possible, depending on the type of problem considered.  In this paper, in particular, we will study the statistical properties of several interpolation techniques.  Since the locations on the sky where measurements are performed cannot generally be chosen in Astronomy (cf.\\ the example of reddening of stars above), we will carry out an \\textit{ensemble average\\/} over the measurement locations.  More precisely, we will assume that the measurements are randomly distributed on the sky following a known scheme and we will carry out a statistical analysis on this sample.  Note that there is a complete freedom on the spatial distribution of measurements on the sky, that can be correlated (for example, clustered) or can follow a non-uniform density $\\rho(x)$.  The ensemble average, which has already been carried out under simplifying hypotheses of uncorrelated points and uniform density in earlier papers \\citetext{e.g., \\citealp{1998A&A...335....1L}; \\citealp{2000MNRAS.313..524W}; \\citealp{2001A&A...373..359L}, hereafter Paper~I; \\citealp{LPM}; \\citealp{MSCov}, hereafter Paper~II}, let us to derive general results that generalize the particular configuration considered.  In previous works (see in particular Paper~I and II) we have focused on a widely used smoothing method.  Here, in contrast, we will keep the discussion much more general and consider some wide classes of interpolating techniques.  This alternative approach has a number of advantages: (i)~The results obtained are very general and can be applied to several smoothing techniques (``proof reusability''); (ii)~The analytic discussion is kept at a very simple level; (iii)~The results obtained are more general, since allow for a non-uniform density of locations and correlation on the locations; (iv)~The properties of a smoothing technique can be predicted in advance without the need of long calculations.  This last point, in our opinion, is particularly important for practical applications.  In fact, with the aid of the results obtained in this paper, several smoothing techniques can be easily compared, which allows one to choose the interpolation method more suitable.  We stress, however, that some specific results cannot be derived using the techniques described in this article and need a more specific analysis as done in Paper~I and II.  The paper is organized in two main parts.  In the first part, Sect.~\\ref{sec:classifications}, we classify the smoothing methods and we show the general results associated to each interpolation family. In the second part, Sect.~\\ref{sec:kern-some-smooth}, we illustrate the results obtained earlier by considering several common smoothing techniques.  Finally, in Sect.~\\ref{sec:conclusions}, we briefly draw the conclusions of this analysis.   \\subsection{Notation} \\label{sec:notation}  Let us suppose to have a set of pairs $\\bigl\\{ (x_i, y_i) \\bigr\\}$, where $x_i \\in X$, called ``locations,'' belong to a real vector space $X$ (typically, $X = \\R^n$, with $n=1$, $2$, or $3$), and $y_i \\in Y$, called ``values,'' belong to a field $Y$ (in this paper, we will assume for simplicity $Y = \\R$).  The spatial interpolation problem consists in finding a way to obtain approximated values for a generic $x \\in X$.  Formally, an interpolation procedure is a function $S$ that maps a point $x \\in X$ and an \\textit{unordered\\/} set of couples $(x_i, y_i) \\in X \\times Y$ into a value $y \\in Y$: \\begin{equation} \\label{eq:1} y = S\\bigl( x; \\bigl\\{ (x_i, y_i) \\bigr\\} \\bigr) \\; . \\end{equation} Note that the number $N$ of couples $(x_i, y_i)$ is not fixed \\textit{a priori\\/} (in particular $N$ could vanish).  In the following we will use as synonymous interpolation procedure, smoothing technique, interpolator.  As an example, let us consider a simple interpolator where the smoothed value at $x$ is obtained as a weighted sum of the values $\\{ y_i \\}$.  More precisely, we define $S$ as \\begin{equation} \\label{eq:2} S \\bigl( x; \\bigl\\{ (x_i, y_i) \\bigr\\} \\bigr) = \\frac{\\sum_{j=1}^N y_j / | x - x_j |^\\alpha}{\\sum_{j=1}^N 1 / | x - x_j |^\\alpha} \\; , \\end{equation} where $\\alpha$ is a fixed real number.  This interpolation technique will be often used in this paper to illustrate with an example some general, abstract results.  In this paper we will study some statistical properties of various interpolation procedures assuming that the locations $\\{ x_i \\}$ are random variables distributed with spatial density $\\rho(x)$ (with or without clustering), and that the values $\\{ y_i \\}$ are associated to the locations through the relation \\begin{equation} \\label{eq:3} y_i = f(x_i) + \\epsilon_i \\; . \\end{equation} Here $f \\colon X \\rightarrow Y$ is a known function and $\\{ \\epsilon_i \\}$, representing measurement errors, are random variables with vanishing mean and covariance matrix (taking $\\{ \\epsilon_i \\}$ as a multidimensional random variable) proportional to the identity: \\begin{align} \\label{eq:4} \\langle \\epsilon_i \\rangle & {} = 0 \\; , & \\langle \\epsilon_i \\epsilon_j \\rangle & {} = \\delta_{ij} \\sigma^2 \\; . \\end{align} Note that the second relation of Eq.~\\eqref{eq:4} states the so-called ``statistical orthogonality'' of the measurement errors; this property is trivially satisfied if $\\{ \\epsilon_i \\}$ are independent random variables with fixed variance $\\sigma^2$ (this is a good approximation for many astronomical observations).  The subject of our study will be the average value $\\bigl\\langle \\tilde f(x) \\bigr\\rangle$ of $\\tilde f(x)$, where $\\tilde f(x)$ is the interpolated value of $f$ at $x$: \\begin{equation} \\label{eq:5} \\tilde f(x) = S \\bigl( x; \\bigl\\{ (x_i, y_i) \\bigr\\} \\bigr) = S \\bigl( x; \\bigl\\{ (x_i, f(x_i) + \\epsilon_i) \\bigr\\} \\bigr) \\; . \\end{equation} We will also investigate the covariance (or two-point correlation) of the map $\\tilde f(x)$, defined as \\begin{align} \\label{eq:6} \\Cov(x, x'; \\tilde f) & {} = \\Bigl\\langle \\bigl[ \\tilde f(x) - \\bigl\\langle \\tilde f(x) \\bigr\\rangle \\bigr] \\bigl[ \\tilde f(x') - \\bigl\\langle \\tilde f(x') \\bigr\\rangle \\bigr] \\Bigr\\rangle \\notag\\\\ & {} = \\bigl\\langle \\tilde f(x) \\tilde f(x') \\bigr\\rangle - \\bigl\\langle \\tilde f(x) \\bigr\\rangle \\bigl\\langle \\tilde f(x') \\bigr\\rangle\\; . \\end{align}  A word of explanation is needed regarding averages.  Averages of $\\tilde f(x)$ and $\\tilde f(x) \\tilde f(x')$ are carried out both with respect to $\\{ \\epsilon_i \\}$ and to $\\{ x_i \\}$.  As we have anticipated above, averages with respect to $\\{ x_i \\}$ are carried out assuming that the locations are random variables distributed with density $\\rho(x)$ over the set $X$.  For example, if we assume that the locations are uncorrelated and that the density is constant over the field (this is usually referred to as a homogeneous Poisson process or as compete spatial randomness), then the number $N$ of object locations inside a field $A \\subset X$ of finite area $\\mu(A)$ follows a Poisson distribution with average $\\rho \\mu(A)$: \\begin{equation} \\label{eq:7} p_N(N) = \\e^{-\\rho \\mu(A)} \\frac{ \\bigl[\\rho \\mu(A) \\bigr]^N}{N!} \\; . \\end{equation} Under the same hypotheses, the $N$ locations $\\{ x_i \\}$ are, then, uniformly distributed inside $A$.  In more general cases (in particular, in case of correlation), it can be non-trivial to specify exactly the probability distribution of points.  However, fortunately for the following discussion we need only the density $\\rho(x)$ and an algorithm to randomly generate the locations (as discussed by \\citealp{1954ApJ...119...91S}; see also, e.g., the hierarchical model described by \\citealp{1978AJ.....83..845S}).  The probability distribution for the other random variables considered here, namely $\\{ \\epsilon_i \\}$, need not to be fully specified. Actually, for our purposes, we just need to specify that these variables, representing measurement errors, have vanishing mean, fixed variance, are orthogonal [i.e., satisfy Eq.~\\eqref{eq:4}], and are independent of the locations $\\{ x_i \\}$.  In the following calculations, we will carry out first the average over the measurement errors $\\{ \\epsilon_i \\}$, and then the one over the locations $\\{ x_i \\}$.  We finally note that in this study we allow for cases where the smoothing procedure $S$ cannot be defined.  For example, if the density $\\rho$ is small, we might end up with a configuration without locations $x_i$ close enough to $x$.  Some interpolation procedures then might not be applied.  In the following, we adopt the convention of discarding in the ensemble average the configurations $\\{ x_i \\}$ for which the interpolation procedure \\eqref{eq:1} is not \\textit{locally\\/} defined.  For example, when evaluating the average $\\bigl\\langle \\tilde f(x) \\bigr\\rangle$, we discard configurations that make $S$ not defined at $x$; similarly, for expressions such as $\\bigl\\langle \\tilde f(x) \\tilde f(x') \\bigr\\rangle$ we discard configurations for which $S$ is not defined at $x$ or at $x'$.  Note that we assume that the applicability of the smoothing technique depends only on the locations $\\{ x_i \\}$ and \\textit{not\\/} on the values $\\{ y_i \\}$.  We call $P_0(x)$ the probability of having a configuration $\\{ x_i \\}$ for which the estimator is not defined at $x$.  For example, some smoothing techniques (see Sect.~\\ref{sec:local-interpolators}) are not defined if there are no locations inside a given subset $\\pi$; in this case we find (in case of vanishing correlation, i.e. for a Poisson distribution of locations) \\begin{equation} \\label{eq:8} P_0(x) = \\exp \\biggl( -\\int_\\pi \\rho(x) \\, \\diff x \\biggr) \\; . \\end{equation} Similarly, we define $P_0(x, x')$ as the probability that the smoothing is not defined either at $x$ or at $x'$ (or at both points). If, again, the smoothing technique is defined at $x$ only if there is at least a location inside a given subset $\\pi$, and is defined at $x'$ if there is a location inside the subset $\\pi'$, then \\begin{align} \\label{eq:9} & P_0(x,x') = \\exp \\biggl( - \\int_{\\pi \\cap \\pi'} \\rho(x) \\, \\diff x \\biggr) \\notag\\\\ & \\qquad \\times \\biggl[ \\exp \\biggl( - \\int_{\\pi \\setminus \\pi'} \\rho(x) \\, \\diff x \\biggr) + \\exp \\biggl( - \\int_{\\pi' \\setminus \\pi} \\rho(x) \\, \\diff x \\biggr) \\notag\\\\ & \\qquad \\phantom{{} \\times \\biggl[} - \\exp \\biggl( - \\int_{(\\pi \\setminus \\pi') \\cup (\\pi' \\setminus \\pi)} \\rho(x) \\, \\diff x \\biggr) \\biggr] \\; . \\end{align} Note that $P_0(x, x) = P_0(x)$ and that $P_0(x) P_0(x') \\le P_0(x, x') \\le P_0(x) + P_0(x')$.  If an estimator is always defined, as for the interpolator \\eqref{eq:2}, we just set $P_0(x) = P_0(x, x') = 0$. ",
        "conclusions": "\\label{sec:conclusions}  In this paper we have considered the statistical properties of interpolation techniques.  The discussion has originally been kept at a high level of abstraction, and this has given us the ability to characterize smoothing methods in terms of simple properties.  In particular, we have made a classification of interpolation techniques and we have derived several statistical results associated with each class of interpolators.  A comparison of this analysis with a more technical one carried out in a separate paper for a specific smoothing method \\citep{2001A&A...373..359L} clearly shows the advantages of using an abstract approach to the problem.  In the second part of this paper we have applied the results obtained to several commonly used smoothing methods."
    },
    "0208/hep-ph0208140_arXiv.txt": {
        "abstract": "Weakly Interacting Massive Particle (WIMP) direct detection experiments are closing in on the region of parameter space where relic neutralinos may constitute the galactic halo dark matter. We discuss two issues in the interpretation of data, in particular the calculation of exclusion limits, from these experiments. Firstly we show that the technique that has been used for calculating exclusion limits from binned data without background subtraction produces erroneously tight limits, and discuss alternative methods which avoid this problem. We then argue that the standard maxwellian halo model is likely to be a poor approximation to the dark matter distribution and examine how halo models with triaxiality, velocity anisotropy and small scale clumping affect exclusion limits. ",
        "introduction": "Arguably the best motivated non-baryonic dark matter candidate is the neutralino (the lightest supersymmetric particle), and current direct detection experiments are just reaching the sensitivity required to probe the relevant region of parameter space~\\cite{lars}.  The most stringent exclusion limits on Weakly Interacting Massive Particles (WIMPs) in general currently come from the Edelweiss~\\cite{edelnew} and Cryogenic Dark Matter Search (CDMS) experiments~\\cite{CDMS1,CDMS2}, with competitive constraints also having been produced by experiments which have been optimized for double-beta decay such as Heidelberg-Moscow (HM)~\\cite{HM} and IGEX~\\cite{IGEX}. The exclusion limits from these experiments, calculated assuming a standard maxwellian halo, are plotted in Fig. 1 along with the region of parameter space corresponding to the DAMA collaboration's annual modulation signal~\\cite{DAMA}. Given the experimental progress, and the tension between the DAMA collaboration's annual modulation signal and the exclusion limits from other experiments, it is crucial to examine the assumptions involved in interpreting data from these experiments.   We focus on two separate issues: the calculation of exclusion limits with the correct coverage (i.e. which correspond to the stated degree of confidence) and the effect of halo modeling on exclusion limits. In Sec.~2 we show that the method previously used to calculate confidence limits from experiments without background subtraction and binned data (such as HM and IGEX) produces erroneously tight exclusion limits, and discuss alternative criteria for calculating exclusion limits.  In Sec.~3 we turn our attention to the dependence of the theoretical differential event rate on the WIMP speed distribution. We discuss the properties of galactic halos and examine how models which reproduce these properties affect the exclusion limits from the HM and IGEX experiments.    \\begin{figure} \\begin{center} \\includegraphics{green1.ps} \\includegraphics{green2.ps} \\end{center} \\caption{Exclusion limits, and the DAMA annual modulation region, from the WIMP direct detection experiments discussed in the text, assuming a standard maxwellian halo model, plotted using the interactive limit plotter at http://dmtools.berkeley.edu/limitplots/ .} \\end{figure} ",
        "conclusions": "We have examined two aspects of the interpretation of data from WIMP direct detection experiments. We have seen that care needs to be taken when calculating exclusion limits from experiments without background subtraction so as to produce correct limits. Yellin's optimal interval method provides a sophisticated and robust solution to this problem for experiments with unbinned data and relatively small numbers of events~\\cite{yellin}, in other cases Poisson statistics can be used to formulate criteria which produce correct limits~\\cite{statpap}.  We have also seen that even if the local WIMP distribution is smooth its speed distribution may deviate significantly from the standard maxwellian, and this has a non-negligible effect on exclusion limits and, crucially, effects the limits from different experiments differently. Constraints (and in the future possibly best fits) calculated assuming a standard maxwellian halo could be erroneous~\\cite{uk}. The derivation of reliable constraints on WIMP parameters and comparison of results for different experiments requires a theoretical framework for dealing with the uncertainty in the WIMP speed distribution. On the other hand, more optimistically, it might be possible to derive useful information about the local velocity distribution, and hence the formation of the galactic halo, if WIMPs were detected~\\cite{swf,hws}."
    },
    "0208/astro-ph0208580_arXiv.txt": {
        "abstract": "Planet evolution is tightly connected to the dynamics of both distant and close disk material. Hence, an appropriate description of disk-planet interaction requires global and high resolution computations, which we accomplish by applying a Nested-Grid method. Through simulations in two and three dimensions, we investigate how migration and accretion are affected by long and short range interactions. For small mass objects, 3D models provide longer growth and migration time scales than 2D ones do, whereas time lengths are comparable for large mass planets. ",
        "introduction": "Migration has entered the puzzling scenario of planetary formation as the favorite mechanism advocated to explain the extremely short orbital period of many extrasolar planets.  It is known that any planet-like body is forced to adjust its distance from the central star because of gravitational interactions with the circumstellar material.  However, the dispute about how fast migration proceeds is far from being over.  Numerical methods have been employed to evaluate gravitational torques exerted on embedded planets. We have performed a series of simulations modeling both two and three dimensional disks, varying the mass $M_p$ of the protoplanet in the range from 1 Earth-mass to 1 Jupiter-mass.  The physics of the problem demands that the flow in the protoplanet's neighborhood should be accurately resolved. In order to achieve sufficient resolution, even for very small planetary masses, we use a \\textit{Nested-Grid} technique (D'Angelo, Henning, \\& Kley 2002a). This paper addresses the issues of flow circulation around protoplanets, orbital migration, and mass accretion. ",
        "conclusions": "Circumplanetary disk forms around protoplanets. The spiral wave pattern which marks such disks is less accentuated when the full 3D structure is simulated. Vertical shock fronts develop outside the Hill sphere of the planet.  The estimated values of $\\tau_{\\mathrm{M}}$ in 3D are longer than those predicted by analytical linear theories because of non-linearity effects. When $M_p\\la 30\\;M_{\\earth}$, both $\\tau_{\\mathrm{M}}$ and $\\tau_{\\mathrm{G}}$ are longer in 3D computations then they are in 2D ones."
    },
    "0208/astro-ph0208063_arXiv.txt": {
        "abstract": "In the preceding chapters, the effects of lensing were so strong as to leave an unmistakable imprint on a specific source, allowing a detailed treatment.  However, only the densest regions of the universe are able to provide such a spectacular lensing effect.  To study more representative regions of the universe, we must examine large numbers of sources statistically.  This is the domain of weak lensing. ",
        "introduction": "\\subsection{Motivation}  Weak lensing enables the direct study of mass in the universe. Lensing, weak or strong, provides a more direct probe of mass than other methods which rely on astrophysical assumptions (\\eg\\ hydrostatic equilibrium in a galaxy cluster) or proxies (\\eg\\ the galaxy distribution), and can potentially access a more redshift-independent sample of structures than can methods which depend on emitted light with its $r^{-2}$ falloff.  But strong lensing can be applied only to the centers of very dense mass concentrations. Weak lensing, in contrast, can be applied to the vast majority of the universe.  It provides a direct probe of most areas of already-known mass concentrations, and a way to discover and study new mass concentrations which could potentially be dark. \\index{lens!dark} With sources covering a broad redshift range, it also has the potential to probe structure along the line of sight.  Specifically, we might expect weak lensing to answer these questions: \\begin{itemize} \\item Where are the overdensities in the universe ? \\item Are they associated with clusters and groups of galaxies ?  Does light trace mass in these systems ? \\index{galaxy!cluster} \\index{galaxy!groups} \\item How much do these systems contribute to \\Om, the mean density of matter in the universe ? \\item What is their mass function and how does that function evolve with redshift~? What does that imply for the dark energy equation of state ? \\item What are the structures on larger scales (walls, voids, filaments) ? \\item Is this structure comparable to that seen in cosmological simulations?  Which cosmology matches best ? \\item What is the nature of dark matter ? \\index{dark matter} \\item Can observations of lensing put any constraints on alternative theories of gravity ? \\end{itemize}  Until recently, deep imaging on the scale required to answer the above questions with weak lensing was simply impractical.  The development of large mosaics of CCDs has expanded the field greatly.  The large data volume leads to ever-decreasing statistical errors, which means that very close attention must be paid to systematic errors and calibration issues.  Weak lensing results must be carefully scrutinized and compared with those of other approaches with this in mind.  We start with a review of the basic concepts, the limits of weak lensing, and observational hurdles, and then address the above astrophysical questions.  \\subsection{Basics}  The transition from strong to weak lensing can be seen at a glance in the simulation shown in Figure~\\ref{fig-demo}.  Over most of the field, no one galaxy is obviously lensed, yet the galaxies have a slight tendency to be oriented tangentially to the lens. We seek to exploit this effect to derive information about the lens, and perhaps about the weakly lensed sources as well.  \\begin{figure} \\centerline{\\resizebox{2.4in}{!}{\\includegraphics{wlfig01.ps}} \\resizebox{2.4in}{!}{\\includegraphics{wlfig01b.ps}}} \\caption{Simulated effects of a lens: source plane (left) and image plane (right).  Most regions of the lens can be probed only with weak lensing.  Real sources are not in a plane, but this does not dramatically affect the appearance. Real lenses, such as galaxy clusters, would obscure much of the strong-lensing region.} \\label{fig-demo} \\end{figure}  We start with the {\\it inverse magnification matrix} (see also Chapter on quasar lensing) \\begin{equation} M^{-1} = (1-\\kappa) \\left(\\begin{array}{cc} 1 & 0\\\\ 0 & 1\\\\ \\end{array} \\right) + \\gamma \\left(\\begin{array}{cc} \\cos 2\\phi & \\sin 2\\phi \\\\ \\sin 2\\phi  &  -\\cos 2\\phi \\\\ \\end{array} \\right), \\end{equation} so called because it describes the change in source coordinates for an infinitesmal change in image coordinates, the inverse of the transformation undergone by the sources.  This is Equation 16 of the Quasar Lensing chapter, which derives $M^{-1}$ and defines the quantities within.  We repeat here that the {\\it convergence} $\\kappa$ represents an isotropic magnification, and the {\\it shear} $\\gamma$ \\index{convergence} \\index{shear} represents a stretching in the direction $\\phi$.  They are both related to physical properties of the lens as linear combinations of derivatives of the deflection angle.  However, $\\kappa$ can be interpreted very simply as the projected mass density $\\Sigma$ \\index{density!surface} divided by the critical density \\index{density!critical} $\\Sigma_{\\rm crit}$, while $\\gamma$ has no such straightforward interpretation.  In fact, $\\gamma$ is {\\it nonlocal}: its value at a given position on the sky depends on the mass distribution \\index{mass!distribution} everywhere, not simply at that position.  We will see this fact rear its ugly head in several places throughout this chapter.  Shear is often written as a vector $\\gamma_i = (\\gamma \\cos 2\\phi,\\gamma \\sin 2\\phi)$ or more succinctly as a complex quantity $\\gamma e^{i2\\phi}$.  Without multiple images of a source (as in the strong lensing case), we must have some independent knowledge of the sources if we are to measure magnification or shear. \\index{magnification} For example, if one source were a standard candle or ruler, the apparent magnitude or size of its image would immediately yield the magnification at that point.  Of course, standard candles or rulers occur only in very special cases \\cite{Blakeslee}, so in practice we must analyze source {\\it distributions}. \\index{source!distribution} We no longer get much information from a single source, and thus lose resolution; this is the tradeoff we must make for probing regions with weak tidal fields.  One source distribution that could be used in this way is $n(m)$, the number of galaxies as a function of apparent magnitude.  In practice, this is difficult, because the measured slope of this distribution does not differ greatly from the critical slope at which equal numbers of galaxies are magnified into and out of a given magnitude bin, with no detectable change ($n \\propto m^{0.4}$).  There is enough difference to make some headway, but we would prefer to measure departures from zero rather than small changes in a large quantity.  The distribution of galaxy {\\it shapes}, \\index{galaxy!shape} properly defined, does allow us to measure departures from zero.  Approximate each source as an ellipse with position angle $\\phi$ \\index{source!position angle} and (scalar) ellipticity \\index{source!ellipticity} $\\epsilon = {a^2 - b^2 \\over a^2 + b^2}$, where $a$ and $b$ are the semimajor and semiminor axes.  Define a {\\it vector ellipticity} \\index{ellipticity!vector} $e_i = (\\epsilon \\cos 2\\phi,\\epsilon \\sin 2\\phi)$, or equivalently a {\\it complex ellipticity} \\index{ellipticity!complex} \\index{polarization} $\\epsilon e^{i2\\phi}$ (also called {\\it polarization}).  This encodes the position angle and scalar ellipticity into two quantities which are comparable to each other; the dependence on $2\\phi$ indicates invariance under rotation by 180$^\\circ$.  Figure~\\ref{fig-ellip} gives a visual impression of ellipses in this space.  We can now quantify the visual impression of Figure~\\ref{fig-demo}. In the absence of lensing, as in the left panel, galaxies are randomly oriented: The observed distribution of $e_i$ is \\index{source!distribution} roughly Gaussian with zero mean and an rms of $\\sigma_e \\sim 0.3$. \\begin{figure} \\centerline{\\resizebox{2.4in}{!}{\\rotatebox{-90}{\\includegraphics{wlfig02.ps}}}} \\caption{A sequence of ellipses with various amounts of each ellipticity component.  Inspired by the appearance of these ellipses, the two components \\index{source!ellipticity} are often labeled $e_+$ and $e_\\times$.} \\label{fig-ellip} \\end{figure} In the presence of lensing, as in the right panel, this distribution is no longer centered on zero, as long as we consider an appropriately-sized patch of sky.  In fact, we will {\\it assume} that any departure from zero mean must be due to lensing.  We will examine the limits of this assumption in some detail later, but for now let us accept that on large enough scales, the cosmological principle demands it, and as a practical matter, we average over sources at a wide range of redshifts, which are too far apart physically to influence each other's alignment.  The effect of the magnification matrix on the complex ellipticity can be computed if $M$ is constant over a source. \\index{magnification} This is obviously not valid for very large sources or those near caustics, \\index{caustics} but it is valid for the vast majority of the sky and for typical sources with sizes of a few arcseconds.  The result is that $\\epsilon^I = \\epsilon^S + {\\gamma \\over 1-\\kappa}$, where superscripts indicate image and source planes  \\cite{Blandford91}.  We don't know any of these quantities for a single source, but we do know (or assume for now) that $\\langle \\epsilon^S \\rangle = 0$, where brackets indicate averaging over many sources. Hence \\begin{equation} \\label{eq-shear} \\langle \\epsilon^I \\rangle =  \\langle {\\gamma \\over 1-\\kappa}\\rangle. \\end{equation} The quantity on the right is called the {\\it reduced shear} $g$. \\index{shear!reduced} A second approximation we can often make is that $\\kappa \\ll 1$, so that $  \\langle \\epsilon^I \\rangle = \\langle \\gamma \\rangle$.  This is called the {\\it weak lensing limit}.  \\index{weak lensing!limit}  The fundamental limit to the accuracy with which we can measure $\\gamma$ in the weak lensing limit is {\\it shape noise}, \\index{shape noise} or the width of the source ellipticity distribution $\\sigma_e \\sim 0.3$.  Averaging over $n$ sources should decrease the uncertainty to ${\\sigma_e \\over \\sqrt{n}}$, but $n$ is limited by the depth of the observations and the area over which we are willing to average $\\gamma$; these tradeoffs are discussed below.  Also note that knowledge of the shear alone is not strictly enough to infer mass distributions because of the {\\it mass sheet degeneracy}  \\cite{FGS85,SS95}, \\index{degeneracy!mass sheet} introduced in a different context in the Quasar Lensing chapter.  This degeneracy arises because a uniform sheet of mass induces only magnification, \\index{magnification} not shear. \\index{shear} Because the equations are linear, we could therefore add or subtract a mass sheet without affecting the shear.  In practice, we can still answer many questions with shear alone, as discussed below.   \\subsection{Cosmology dependence} \\index{cosmology}  Both convergence and shear \\index{convergence} \\index{shear} scale as the combination of angular diameter \\index{distances!angular diameter} distances ${\\Dls \\Dl \\over \\Ds}$, or as the {\\it distance ratio} ${\\Dls \\over \\Ds}$ for a given lens.  (Recall from the Quasar Lensing chapter that $\\Dls$, $\\Dl$, and $\\Ds$ are the angular diameter distances from lens to source, observer to lens, and observer to source, respectively.  Note that $\\Ds \\ne \\Dl + \\Dls$; see \\cite{Hogg} for a quick review and  \\cite{Peebles} for a thorough treatment of distance measures in cosmology).  This cosmology-dependent quantity is plotted as a function of source redshift in Figure~\\ref{fig-dratio} for several lens redshifts and two different cosmologies.  In principle, this could be used to constrain the cosmology if source redshifts \\index{source!redshift} are known, and if the lens mass is known independently (the effects of a larger lens mass and a larger universe are degenerate).  But this remains an unused cosmological test because lens parameters and source redshifts are usually poorly known.  Usually, a cosmology is assumed and lens parameters are estimated using any available knowledge of source redshifts. Less often, a well-characterized lens is used to explore the source redshift distribution.  However, source redshift distributions are usually quite broad, and weak lensing can only be used to estimate the {\\it mean} distance ratio to a group of sources, which is not same as the distance ratio corresponding to the mean redshift.  Section~\\ref{sec-srd} deals with ways of estimating the mean distance ratio or otherwise accounting for a broad source redshift distribution.   \\begin{figure} \\centerline{\\resizebox{3in}{!}{\\includegraphics{wlfig03.ps}}} \\caption{${{\\Dls \\Dl} \\over \\Ds}$ as a function of source redshift, for several lens redshifts (indicated by the intersections of the curves with the horizontal axis) and several cosmologies.  The cosmologies are $\\Lambda$-dominated (solid lines, $H_0= 70$ km s$^{-1}$ Mpc$^{-1}$, $\\Omega_m = 0.3$, $\\Omega_\\Lambda = 0.7$) and open (dashed lines, $H_0= 70$ km s$^{-1}$ Mpc$^{-1}$, $\\Omega_m = 0.4$, $\\Omega_\\Lambda = 0$).  Each solid line is higher than its dashed counterpart, reflecting the larger size of the $\\Lambda$-dominated universe.  Although this quantity appears to be a sensitive test of the cosmology, it is degenerate with the lens mass.} \\label{fig-dratio} \\end{figure}  Another way of viewing the same information is to fix the source redshift and plot this ratio as a function of lens redshift (Figure~\\ref{fig-efficiency}).  This reveals the relative importance of different structures along the line of sight and is often called the {\\it lensing kernel} or {\\it lensing efficiency}. \\index{lensing!efficiency}\\index{lensing!kernel}  \\begin{figure} \\centerline{\\resizebox{3in}{!}{\\includegraphics{wlfig04.ps}}} \\caption{Same as for Figure~\\ref{fig-dratio}, but as a function of lens redshift, \\index{lens!redshift} for several values of source redshift \\index{source!redshift} (which correspond to the right-hand end of each curve).  The lensing efficiency is a very broad function, making it difficult to separate unrelated structures along the line of sight. } \\label{fig-efficiency} \\end{figure}   \\subsection{Applicability of weak lensing}  As with all astrophysical tools, we must be aware of the limitations of weak lensing before plunging into results.  They include the weak lensing approximation itself; \\index{weak lensing!approximation} mass sheet degeneracy if only shear is used; poor angular resolution because of its statistical nature; source redshift difficulties; and possible departures from the assumption of randomly oriented sources.  We now examine these limits and ways of dealing with them.  \\subsubsection{Weak lensing approximation}  The approximations that $M$ is constant over each source and that $\\kappa \\ll 1$ cannot be applied when dealing with the centers of massive clusters and galaxies. \\index{galaxy!cluster} Of course, analysis of such regions is not lacking---it is the topic of most of this book.  Here we merely wish to mention work that has been done on combining weak and strong lensing information  \\cite{ASW1998}.  We also note that, where only the second approximation fails, Equation (\\ref{eq-shear}) can be solved iteratively for $\\kappa$.  \\subsubsection{Mass sheet degeneracy} \\index{degeneracy!mass sheet}  Mass sheet degeneracy was a serious concern when fields of view were small and lens mass distributions extended well beyond the edges. Modern imagers now deliver fields of view $\\sim 0.5^\\circ$ on a side ($>3$ Mpc radius for any lens at $z>0.15$), so this concern has diminished.  The degeneracy may also be broken by adding magnification information, which may come from strong lensing, or from a method called the depletion curve. \\index{depletion curve}  Magnification \\index{magnification} imposes two effects of opposite sign on the areal density of sources. \\index{source!density} Galaxies fainter than the detection limit (or any chosen brightness threshold) are amplified above the threshold, increasing the density of sources, but at the same time the angular separation between galaxies is stretched, decreasing the density of sources.  The net effect depends on the slope of $n(m)$, the (unlensed) galaxy counts as a function of magnitude.  A logarithmic slope less than 0.4 (usually the case at visible wavelengths, but barely) will not provide enough ``new'' sources to overcome the dilution effect, so the source density decreases as $\\kappa$ increases toward the center of a cluster.  This {\\it depletion curve} reveals lens parameters, as shown in Figure~\\ref{fig-depletion}  \\cite{MS2000}. Despite the name, the method need not be restricted to one-dimensional information \\cite{BTP95}; \\cite{MS2000} includes a lens ellipticity \\index{lens!ellipticity} \\index{lens!position angle} and position angle estimate based on a crude depletion map.  In practice, measuring magnification is quite difficult, because the slope of $n(m)$ is perilously close to 0.4, and there are few published depletion curve measurements \\cite{MS2000,D2001}. For the remainder of this work, we shall concentrate on algorithms and results using shear, not magnification.  \\begin{figure} \\centerline{\\resizebox{2.4in}{!}{\\includegraphics{wlfig05a.ps}} \\resizebox{2.4in}{!}{\\includegraphics{wlfig05b.ps}} } \\caption{Left: theoretical depletion \\index{depletion curve} curves for a variety of lens velocity dispersions (lens mass $\\propto \\sigma_v^2$).  Right: depletion curve observed for MS1008-1224 in V band.  From \\cite{MS2000}.} \\label{fig-depletion} \\end{figure}  \\subsubsection{Angular resolution} \\index{source!density}  The angular resolution of weak lensing is limited by the areal density of sources.  With a shape noise \\index{shape noise} of $\\sigma_e \\sim 0.3$ and $\\sqrt{n}$ statistics, a shear measurement accurate to $p$ percent requires $\\sim 1000 p^{-2}$ sources.  Angular resolution is then set by the area of sky over which these sources are scattered.  This in turn depends on the depth and wavelength of the observations; in $R$ there is one source per square arcminute in a one-magnitude wide bin at $R \\sim 21.4$, increasing by a factor of $\\sim 2.5$ for every magnitude deeper \\cite{Tyson88}.  A medium deep observation capable of shape measurements to $R \\sim 25$ thus yields about 20 galaxies arcmin$^{-2}$ (assuming a bright cutoff $R>23.5$ to eliminate largely foreground sources), implying that 2 arcmin$^{2}$ are required for 5\\% accuracy in shear.  \\index{source!density}  Getting more sources per unit area requires much more telescope time. Source density will ultimately be limited by confusion \\index{source!confusion} --- when sources are so numerous that they overlap and hinder shape measurements --- around $\\sim 1000$ sources arcmin$^{-2}$ for ground-based data.  This implies $\\sim$ 20\\arcs shear resolution, or better for space-based data, if galaxy counts keep rising at the same rate.  However, such depth is hard to come by and must compete against area and wavelength coverage (useful for constraining source redshifts) when planning for a given amount of telescope time.  Another tradeoff commonly used is to sacrifice resolution in one dimension to achieve better resolution in the other.  Clusters are commonly analyzed in terms of a radial profile, which assumes they are axisymmetric and allows all sources at a given radius from the cluster center to be averaged together.  Less massive clusters and groups can be ``stacked'' to yield an average profile with reasonable resolution, just as in galaxy-galaxy lensing \\cite{Hoekstra_etal2001,Sheldon}.  \\subsubsection{Source redshift distribution} \\label{sec-srd} \\index{source!redshift distribution}  Lack of knowledge of the source redshift distribution is often a limit in calibrating weak lensing measurements.  The root of this problem is that deep imaging quickly outruns the ability of even the largest telescopes to provide spectroscopic redshifts for a fair sample of sources.  The recent development of photometric redshift techniques, \\index{photometric redshift} in which multicolor imaging provides enough spectral information for a reasonable redshift estimate cite{Connolly1995,Hogg1998}, has brought hope that source redshifts may be estimated to sufficient accuracy from imaging alone.  For example, the Hubble Deep Field \\index{HDF} yielded photometric redshifts accurate to $\\sim 0.1$ per galaxy in the redshift range $0-1.4$ with seven filters extending through the near-infrared ($UBVIJHK$) \\cite{Hogg1998}.  A look at Figure~\\ref{fig-dratio} shows that this provides a reasonable accuracy in distance ratio in most situations. The accuracy improves with the number of filters used, resulting in a tradeoff between accuracy and telescope time.  Deep $U$ and infrared imaging are much more expensive than $BVRI$ in terms of telescope time, but it is difficult to effectively cover a large redshift range with only $BVRI$.  Few spectral features are to be found in the observed $BVRI$ bandpasses for sources in the redshift range $\\sim 1.5-3$, which greatly increases uncertainties there.  However, these problems are not fundamental, and photometric redshifts will become routine.  They will do much more than help estimate the mean distance ratio required for calibrating lenses.  Because sources lie at a range of redshifts, they will provide the opportunity to probe structure along the line of sight (albeit with resolution limited by the width of the lensing kernel).  The ultimate goal is {\\it tomography} \\index{tomography} --- building up a three-dimensional view of mass in the universe from a series of two-dimensional views at different redshifts.  The combination of weak lensing and photometric redshifts thus promises to be very powerful, but as yet there are not many published examples of combining the two, and little theoretical work on optimal ways of doing so.  Although we can expect photometric redshifts \\index{photometric redshift} to be a routine part of future lensing work, we must be aware of alternative ways of confronting the source redshift problem. \\index{source!redshift}   First, some questions can be answered without calibration of source redshifts.  The two-dimensional morphology of a cluster lens is one example  --- the source redshift distribution should not depend on position (as long as magnification is negligible and cluster members do not contaminate the source sample).  Similarly, source redshifts are not required for discovery of mass concentrations in surveys, but without them, the volume probed is unknown.  Clearly, the questions which can be answered this way are limited.  A more general calibration strategy is through additional, identical observations of a ``control lens'' of known redshift and mass ({\\it e.g.} a cluster with a dynamical, X-ray, \\index{cluster!X-ray} and/or strong lensing mass estimate).  This does allow estimation of the mean distance ratio to a population of sources much too faint to reach with spectroscopy, but it certainly has its limits.  It is difficult to obtain identical observations, and the (probably considerable) uncertainty in the mass of the control lens becomes a systematic for the rest of the data. But more fundamentally, shear from the control lens samples only that part of the source distribution which is behind the control lens, so that strictly speaking, a control lens must be at the same redshift as the target.  For weak lensing by large-scale structure, \\index{large-scale structure} the distribution, not simply the mean distance ratio, is required.  This would require control lenses at a range of redshifts, which is impractical.  Photometric redshifts should do a much better job with more realistic data requirements.  Even in the age of photometric redshifts, though, this method will have its role.  The shear induced by calibrated lenses will provide a check on photometric redshift estimates, which may not be checkable with spectroscopy if applied to very faint sources.  Another strategy is keeping the imaging as shallow as current redshift surveys, which go to $R \\sim 24$.  One can then look up the median redshift for any magnitude cut; for $23<R<24$, for example, the median redshift is 0.8 \\cite{redshiftsurvey}.  Even the redshift distribution is known to some extent, with 120 sources in that magnitude slice in the survey cited.  Shallow imaging need not probe a small volume, as a large area can be covered with a reasonable amount of telescope time. But it does limit the distance probed and the angular resolution of the mass reconstruction (because the areal density of sources is low at $R \\le 24$).  This strategy also limits selection of sources based on color, which is very useful for limiting contamination by galaxies in a cluster being studied (or for de-emphasizing foreground contamination in general) because the median redshift of a color-selected sample is not yet something that can be looked up in a redshift survey.     \\subsubsection{Intrinsic alignments} \\index{source!alignment}  The crucial assumption in weak lensing is that the sources have random intrinsic orientations, so that any departure from randomness is due to lensing.  This assumption is worth examining before proceeding further.  We will concentrate on potential damage to measurements of weak lensing by large-scale structure (cosmic shear), because the lensing signal from clusters is usually at a much higher level. However, it is worth keeping in mind that all applications of weak lensing could be affected at some level.  The first detections of cosmic shear \\index{shear!cosmic} in 2000 motivated several analytic \\cite{CKD2000} and computational \\cite{HRH2000,CM2000} studies of intrinsic alignment \\index{source!alignment} mechanisms, and the field is still sorting itself out.  There are several mechanisms which could produce such intrinsic alignments, including tidal stretching of galaxies in a gravitational potential, and coupling of the potential to the spin vectors of galaxies \\cite{CNPT2000a}.  The amount of alignment predicted as a function of angular scale varies greatly depending on the mechanism and the strength of the coupling; it remains unknown which model, if any, is correct.  However, in most scenarios, intrinsic alignments would represent a \\raisebox{-0.5ex}{$\\stackrel{<}{\\scriptstyle \\sim}$} 10\\% contamination of the cosmic shear measurements.  While the situation is still evolving, one rule is certain: the effect of any intrinsic alignment is diluted when sources lie at a large range of redshifts, as is naturally the case in deep imaging.  As we shall see, the signal from lensing by large-scale structure \\index{large-scale structure} increases with source redshift.  Hence, lensing must dominate at high enough source redshift, and intrinsic alignments at low enough source redshift.  This is illustrated by Figure~\\ref{fig-cnpt1}, which shows predicted intrinsic alignment and cosmic shear levels for several source redshifts.  For shallow surveys like the Sloan Digital Sky Survey \\cite{SDSS} (SDSS), \\index{SDSS} intrinsic alignment may frustrate attempts to measure cosmic shear, but deeper surveys specifically designed to measure cosmic shear are safe.  Indeed, the roughly one million spectroscopic redshifts SDSS plans to acquire will be invaluable in measuring intrinsic alignments precisely, and their measurements, after scaling to higher redshifts, in turn may facilitate estimation and even removal of intrinsic alignment effects from the deeper surveys.  Deep surveys may also be able to provide a lensing signal using only sources which cannot be physically associated, as indicated by their photometric redshifts.  Density reconstruction methods in the presence of intrinsic alignments are already being investigated \\cite{LP2000}.  \\begin{figure} \\centerline{\\resizebox{2.2in}{!}{\\includegraphics{wlfig06a.ps}} \\resizebox{2.2in}{!}{\\includegraphics{wlfig06b.ps}}} \\caption{ The importance of intrinsic alignments depends strongly on \\index{source!redshift} \\index{source!alignment} source redshift.  The expected levels of ellipticity correlation (defined in Section~\\ref{sec-cosmicshearstats}) due to intrinsic alignments and to weak lensing by \\index{large-scale structure} large-scale structure are shown for low-redshift (median source redshift $z_m = 0.1$) surveys in the left panel and for high-redshift ($z_m = 1$) surveys in the right panel. In each case, the straight line indicates the expected signal from weak lensing, and the curves indicate the expected signal from intrinsic alignments, for two different values of a spin-coupling parameter, giving some idea of the modeling uncertainties.  The right panel also contains cosmic shear \\index{shear!cosmic} measurements from the literature, which all happen to have $z_m \\sim 1$.  Adapted from \\cite{CNPT2000a}.} \\label{fig-cnpt1} \\end{figure}  The two or three detections of intrinsic alignments in real data are indeed at low redshift.  Ellipticity correlations have been reported in SuperCOSMOS data \\cite{BTHD2000}, but because there is no redshift information, intrinsic alignments can only be inferred (the median source redshift is estimated to be $<0.1$, so that the inferred cause is intrinsic alignments rather than lensing).  Spin alignments have been found in the Tully catalog, which consists of several thousand nearby (within a few Mpc) spirals \\cite{PLS}, and in the PSCz, a redshift survey of 15500 galaxies detected by the IRAS infrared \\index{IRAS} satellite mission \\cite{LP2001}.  Because these catalogs have redshift information, these represent solid detections.  However, spin correlations are one step removed from ellipticity correlations, which are the relevant quantity for lensing.  It may also be possible to extract intrinsic alignments from the lensing data itself.  To first order, lensing produces a curl-free, or E-type (in analogy with electromagnetism) shear field.  (Multiple scattering can produce a weak divergence-free, or B-type field, but that can be safely ignored for the moment.)  Therefore, decomposition of a measured shear field into E-type and B-type fields might allow separation of the lensing and intrinsic alignment effects \\cite{CNPT2000b}.  This decomposition is difficult, but a crude indicator of the B-type field is the traditional 45$^\\circ$ test.  In this test, a lensing signal should vanish when one component of the shear is exchanged for the other (equivalent to rotating each source by 45$^\\circ$), and nonzero results would indicate a systematic error. All published cosmic shear results were vetted using this test among others, with no indication of contamination.  However, not all intrinsic alignment mechanisms produce B-type power; an example is tidal stretching (tidal fields are the basic mechanism for both stretching and lensing, after all).  Based on other astrophysical arguments, tidal stretching is not likely to be significant \\cite{CNPT2000b}, but even angular momentum coupling models can produce much more E-type than B-type correlations \\cite{MWK2001}. Passing the 45$^\\circ$ test is a necessary but not sufficient condition for confidence in the results.  In summary, intrinsic alignments are not to be dismissed.  They must be addressed and may even dominate the lensing signal in certain low-redshift applications.  However, the dilution effect of a broad source redshift distribution \\index{source!redshift distribution} means that none of the conclusions of weak lensing to this point can be called into doubt.  Ongoing and future weak lensing studies may have to apply small corrections for this effect, but how small is still uncertain.  Accurate corrections will probably be available by the completion of the SDSS, \\index{SDSS} which will do much to increase our knowledge of intrinsic alignments in the nearby universe.   \\subsection{Measuring shear} \\index{shear!measurment}   In most weak lensing work, a source galaxy is approximated as an ellipse fully described by its quadrupole moments \\index{source!ellipticity} \\begin{eqnarray} I_{xx} \\equiv { \\Sigma Iwx^2 \\over \\Sigma Iw } \\nonumber \\\\ I_{yy} \\equiv { \\Sigma Iwy^2 \\over \\Sigma Iw } \\\\ I_{xy} \\equiv { \\Sigma Iwxy \\over \\Sigma Iw } \\nonumber \\end{eqnarray} where $I(x,y)$ is the intensity distribution above the night sky level, $w(x,y)$ is a weight function, the sum is over a contiguous set of pixels defined as belonging to the galaxy, and the coordinate system has been translated so that the first moments vanish ({\\it i.e.} the centroid of the galaxy is chosen to be the origin of the coordinate system).  Early work used intensity-weighted moments ($w=1$), but it was realized that this produces ellipticity measurements with noise properties that are far from optimal or even divergent.  Now, $w$ is usually chosen to be a circular \\cite{KSB} or elliptical \\cite{BJ2001} Gaussian, which deweights the outer pixels which have a big lever arm but low signal-to-noise.  The two ellipticity components \\index{ellipticity!components} can be defined as \\cite{VJT83} \\begin{eqnarray} e_+ = { I_{xx} - I_{yy} \\over I_{xx} + I_{yy}} \\nonumber \\\\ e_\\times = { 2I_{xy} \\over I_{xx} + I_{yy}} \\end{eqnarray} These are related to the scalar ellipticity $\\epsilon$ and position angle $\\phi$ by \\index{source!position angle} \\begin{eqnarray} \\epsilon = (e_+^2 + e_\\times^2)^{1 \\over 2} \\nonumber \\\\ \\phi = {1 \\over 2} tan^{-1}({e_\\times \\over e_+}) \\end{eqnarray} Then a simple estimate of the shear in the weak lensing limit \\index{weak lensing!limit} is $\\gamma_i = \\langle e_i \\rangle/2$, where the brackets denote averaging over many sources (perhaps with weighting of the sources based on estimated measurement errors, redshift, etc.) to beat down shape noise. \\index{shape noise} Note that this definition of ellipticity differs from that in Equation (\\ref{eq-shear}) by a factor of two; both definitions are presented here because both are common in the literature.  This latter estimator sometimes called the {\\it distortion} statistic. Also, there are alternative formulations in terms of octupole moments \\cite{octupole}, Laguerre expansions \\cite{BJ2001} and shapelets \\cite{shapelets1,shapelets2,shapelets3}. Before applying any of these estimators, we must account for the effects of point-spread function (PSF) anisotropy \\index{PSF!anisotropy} and broadening.   \\subsubsection{PSF anisotropy} \\index{PSF!anisotropy} \\index{PSF!ellipticity}  No optical system is perfect, and PSFs on real telescopes tend to be $\\sim 1-10\\%$ elliptical.  This constitutes a huge systematic error, of the order of the shear induced by even a massive cluster, and it must be removed as completely as possible before analyzing any galaxy shapes, and monitored afterward.  The removal can be done {\\it after} measuring shapes, by essentially subtracting the moments of the PSF from the galaxy moments, but a more computationally stable method is to remove this effect from the image, {\\it before} measuring shapes. This is done by convolving the image with a kernel with ellipticity components \\index{ellipticity!components} opposite to that of the PSF \\cite{FT97}.  The raw PSF is almost certainly position-dependent; therefore the circularizing kernel is also, but the convolved PSF is everywhere round.  A round PSF is called {\\it isotropic}, but keep in mind that this does not imply homogeneous: The convolved PSF may vary somewhat in size, because of the position dependence of the original PSF and of the small broadening introduced by the kernel.  A more sophisticated scheme would introduce more broadening in the right places, leading to a PSF which is homogeneous as well as isotropic. It is also possible to choose a sharpening kernel, but this would amplify the noise in the image.  Figures~\\ref{fig-circ} and~\\ref{fig-escatter} illustrate the effectiveness of the convolution procedure.  Although there are low-level residuals in the convolved image, their lack of spatial correlation means that they will have difficulty masquerading as a weak lensing effect.  Note that these PSF anisotropies change with time, as telescope temperature, focus, and guiding drift, so that each exposure must be treated separately.  A possible benefit here is that if the anisotropies are really uncorrelated temporally, coaddition of multiple exposures will beat down the shape errors.  Also, each CCD in a mosaic must be treated separately, as some discontinuities may arise from small differences in piston between devices.  \\begin{figure}[ht] \\centerline{\\resizebox{3in}{!}{\\includegraphics{wlfig07.ps}}} \\caption{Point-spread function correction in one 2k$\\times$4k CCD. Shapes of stars, which as point sources should be perfectly round, are represented as sticks encoding ellipticity and position angle.  Left panel: raw data with spatially varying PSF ellipticities up to 10\\%. Right panel: after convolution \\index{PSF!convolution} with a spatially varying asymmetric kernel, ellipticities are vastly reduced (stars with $\\epsilon<0.5\\%$ are shown as dots), and the residuals are not spatially correlated as a lensing signal would be.} \\label{fig-circ} \\end{figure}  \\begin{figure}[ht] \\centerline{\\resizebox{3in}{!}{\\includegraphics{wlfig08.ps}}} \\caption{Another way of plotting the efficacy of the PSF correction, \\index{PSF!correction} often seen in the literature.  For the same dataset shown in Figure~\\ref{fig-circ}, the ellipticity components of point sources are shown in a scatterplot, before and after correction.  This type of plot hides any spatial correlation which may exist among the residuals, but Figure~\\ref{fig-circ} shows that the residuals are uncorrelated in this case.} \\label{fig-escatter} \\end{figure}   \\subsubsection{PSF broadening} \\index{PSF!broadening}  Any effect which broadens the PSF will reduce the measured ellipticities of source galaxies which are not much larger than the PSF---generally including the distant galaxies most appropriate for lensing---because they will be broadened relatively more along their minor axes than along their major axes.  In ground-based data, the dominant effect is ``seeing'', the broadening of the point-spread function due to turbulence in the atmosphere.  Note that this is distinct from PSF anisotropy, which is caused by the telescope and camera optics.  In fact, seeing produces a circular PSF as long as the integration time is much longer than the coherence time of the atmosphere (very roughly 30 ms at visible wavelengths); anyone who has observed in terrible seeing has probably noticed that at least the PSF is nicely round !  The effects of PSF anisotropy and broadening are sometimes called ``shearing'' and ``smearing'', respectively.  The former has the effect of introducing a spurious weak lensing signal if uncorrected, while the latter has the effect of reducing any weak lensing signal.  There are several ways of correcting for smearing.  The first is measuring the dilution of a simulated weak lensing signal relative to an unsmeared image, either simulated or perhaps from the Hubble Deep Fields. \\index{HDF} The seeing-free images are sheared by a known amount, convolved with the point-spread function of the real data, repixelized, and the shear measured.  This has the advantage of including some effects which cannot be accounted for analytically, such as the coalescing of separate objects into an apparently elliptical single object.  On the other hand, a global correction is rejected by those who prefer to tailor the corrections to individual sources; after all, a large galaxy is smeared relatively less by seeing than is a small one.  The advocates of this approach tend to use the KSB method, an analytical approach which takes into account the size of the PSF and of each source \\cite{KSB}.  The KSB method also accounts for PSF shearing, but it can just as well be applied to a convolved image.  See \\cite{Kaiser1999,Kaiser2000,Kuijken,BJ2001} for limitations of and possible successors to the KSB method.  These approaches also weight each source according to its ellipticity uncertainty when computing a shear estimate.  \\subsubsection{Source selection}  Not every source in a deep image should be included in a shear measurement.  A typical deep image includes stars, other unresolved sources, foreground galaxies, cluster members if the target lens is a cluster, and spurious objects, such as bits of scattered light around very bright stars.  Getting rid of these unwanted sources is something of an art, which must reflect the particular data set, but generally there are four kinds of cuts.  Magnitude cuts help get rid of stars (for reasonable galactic latitude, stars outnumber galaxies for $R$ \\raisebox{-0.5ex}{$\\stackrel{<}{\\scriptstyle \\sim}$}~$22$ while galaxies greatly outnumber stars for $R > 23$) and bright foreground galaxies.  Galaxies have a broad luminosity function, so such a cut is never completely effective at eliminating the foreground, but it helps.  Color cuts seek to emphasize the faint blue galaxies at $z \\sim 1$  \\cite{TysonSeitzer1988}.  If the target is a cluster, the cut should be blueward of the cluster's color-magnitude ridge.  Even so, some cluster members and other foreground galaxies will survive.  Size cuts eliminate unresolved objects, which at the relevant magnitudes include some stars, but mostly unresolved galaxies.  Finally, cuts designed to insure that an object is not spurious must depend on the type of data available.  Examples include rejecting objects that appear on only one of a multicolor set of images, and rejecting high ellipticity objects which are likely to be unsplit superpositions of two different objects.  \\subsubsection{Sanity checks} There are a number of sanity checks that should be performed before believing any weak lensing result.  In addition to the 45$^\\circ$ test mentioned above, randomizing source positions while retaining their shapes should result in zero signal.  Another good sanity check is correlating the source shapes with an unlensed control population, such as a set of stars.  Finally, there are checks on the basic integrity of the catalog, such as the position angle distribution of sources, which might reveal spurious objects aligned with the detector axes.  Because setting the source selection criteria can be somewhat subjective, it is also good to check that the results do not depend crucially on the exact magnitude or color cut. ",
        "conclusions": ""
    },
    "0208/astro-ph0208549_arXiv.txt": {
        "abstract": "The results of a spectroscopic analysis of 3CR and 6C radio galaxies at redshift $z \\sim 1$ are contrasted with the properties of lower redshift radio galaxies, chosen to be matched in radio luminosity to the 6C sources studied at $z \\sim 1$, thus enabling the redshift--radio power degeneracy to be broken.  Partial rank correlations and principal component analysis have been used to determine which of redshift and radio power are the critical parameters underlying the observed variation of the ionization state and kinematics of the emission line gas.  \\oohb is shown to be a useful ionization mechanism diagnostic. Statistical analysis of the data shows that the ionization state of the emission line gas is strongly correlated with radio power, once the effects of other parameters are removed. No dependence of ionization state on cosmic epoch is observed, implying that the ionization state of the emission line gas is solely a function of the AGN properties rather than the host galaxy and/or environment.  Statistical analysis of the kinematic properties of the emission line gas shows that these are strongly correlated independently with both redshift and radio power.  The correlation with redshift is the stronger of the two, suggesting that host galaxy composition or environment may play a role in producing the less extreme gas kinematics observed in the emission line regions of low redshift galaxies.  For both the ionization and kinematic properties of the galaxies, the independent correlations observed with radio size are stronger than with either radio power or redshift.  Radio source age is clearly a determining factor for the kinematics and ionization state of the extended emission line regions. ",
        "introduction": "\\footnotetext{E-mail: kji@mrao.cam.ac.uk}  Powerful radio sources are typically associated with massive ellipticals, and are often observed to have extended emission line regions aligned along the radio source axis.  The luminosity and kinematic properties of the emission line regions are generally observed to be more extreme for radio sources at high rather than at low redshifts. It is important to establish exactly how the properties of the extended emission line regions (ionization mechanism, kinematics and physical extent) are functions of redshift, radio power and radio size.  This will add to our understanding of these complex systems.  Baum \\& McCarthy (2000) carried out a spectroscopic study of a sample of 52 radio galaxies covering a large range of redshifts, most of which are selected from the 3CR sample. Their analysis of the kinematic and morphological properties of these objects demonstrated a number of important correlations.  The kinematic properties of the emission line gas in these sources were found to vary strongly with redshift and/or radio power, the higher redshift galaxies generally displaying greater line widths and velocity amplitudes.  The inferred mass of ionized gas and the enclosed dynamical gas were also seen to increase with redshift.  Although jet-cloud interactions are not excluded by the data, Baum \\& McCarthy preferred a gravitational origin for the observed kinematics.  Best, R\\\"{o}ttgering \\& Longair (2000a,b) carried out a study of the emission line regions of high redshift 3CR galaxies at $z \\sim 1$. Small radio sources (those with a radio size $D_{rad} < 120\\,\\rm{kpc}$) were observed to exist in a lower ionization state, and to possess more distorted velocity profiles and boosted low-ionization (e.g. [O\\textsc{ii}]) emission line luminosities in comparision with larger radio galaxies ($D_{rad} > 120\\,\\rm{kpc}$).  Their spectra were consistent with small radio sources being predominantly shock ionized, and large radio sources photoionized by the AGN.  Emission line regions were generally larger in spatial extent in the smaller radio sources.  The kinematic and ionization properties of these emission line regions were also found to be strongly correlated.  Best {\\it et al} also compared the properties of the emission line regions of high redshift 3CR galaxies at $z \\sim 1$ with a sample of low redshift 3CR galaxies with z $\\lta 0.2$ (from Baum {\\it et al} 1992).   Similar radio size trends were observed in the low redshift sample, although some trends with either radio luminosity and/or redshift were also observed.  Line luminosities and equivalent widths were generally less extreme in the low redshift sources.  Tadhunter {\\it et al} (1998) showed that a similar correlation exists between line luminosity and radio luminosity in a subsample of 2\\,Jy sources from the sample of Wall \\& Peacock (1985).   Assuming a simple quasar illumination model, the observed ionization state of the emission line gas implied that this correlation could not be explained by a simple increase in the flux of ionizing photons associated with more powerful AGN at higher redshifts.  The observations could however be reconciled with the simple photoionization model if emission line region cloud density is enhanced at higher redshifts, either due to changes in the host galaxy environment or radio source shocks increasing in importance.  The flux--limited 3CR sample suffers, however, from Malmquist bias: the strong correlation between redshift and radio power in a flux limited sample prevents the effects of changes in these two parameters from being disentangled from each other.  This degeneracy needs to be broken, in order to determine to what extent changes in redshift and/or the host galaxy environment, in particular any possible variations in the size and structure of the emission line regions and the density of the IGM, influence the ionization and kinematics of the extended emission line regions of these galaxies. Higher radio power galaxies generally have intrinsically more powerful jets; the bulk kinetic power of the radio jets has been shown to be strongly correlated with narrow line region luminosities (Rawlings \\& Saunders 1991).  It is important that we fully understand the effects of changes in radio power on the ionization state and kinematic properties of the extended emission line regions (EELRs).  Over recent years we have been involved in an ongoing programme of multiwavelength observations investigating a subsample of 11 6C radio galaxies, (Eales 1985; Best {\\it et al} 1999; Inskip {\\it et al (hereafter Paper 1)}). These sources were selected in order to be well matched to a previously well studied subsample of 28 3CR galaxies with redshift $z \\sim 1$, (Longair, Best \\& R\\\"{o}ttgering 1995; Best, Longair \\& R\\\"{o}ttgering 1996, 1997; Best {\\it et al} 1998, 2000a,b). In Paper 1 the results of deep spectroscopic observations of eight of these sources using the William Herschel Telescope (WHT) were presented.  We refer the reader to Paper 1 for details of sample selection, observations and data reduction, reduced 1- and 2-dimensional spectra, tabulated line fluxes and composite spectra. Also included in Paper 1 is a discussion of the kinematic and ionization properties of the 6C galaxies, contrasting them with those of the spectroscopic sample of 3CR galaxies at $z \\sim 1$.  The major results of Paper 1 are briefly summarized below: \\vspace{-10pt} \\begin{enumerate} \\item The kinematical properties of the EELRs of 6C galaxies are similar to those of the more powerful 3CR sources studied at the same redshift.  Small radio sources generally possess more extensive emission line regions with a more distorted velocity profile and more extreme kinematics than those of larger radio sources. \\item \\oo emission line luminosity is anticorrelated with the size (or age) of the radio source. \\item Ionization state varies similarly with radio size for both subsamples, despite the decrease in radio power of the 6C sources. This is interpreted as being due to a changing contribution of ionizing photons from the shock front associated with the expanding radio source.  The optical/UV spectra of the EELRs associated with large radio sources are dominated by photoionization by the AGN and those of small sources by shocks. \\end{enumerate}  \\vspace{-7pt} There is a great deal of evidence that shocks associated with the expanding radio source play a major role in creating the extreme gas kinematics observed in the spectra of radio galaxies. In addition to the results outlined above, the extreme line widths observed in the extended emission line regions are coincident with the radio source structures.   This argues against the gravitational origin for the observed kinematics of the gas preferred by Baum \\& McCarthy, at least for the more extreme sources.  In this paper we compare the results of the 3CR and 6C $\\sim 1$ subsamples with lower redshift 3CR sources, selected to have similar radio powers to those in the 6C subsample.  By considering samples of galaxies covering a larger region of the $P$--$z$ plane (where $P$ is the radio luminosity of the galaxies) we are able to break the degeneracy between redshift and radio power.  The structure of the paper is as follows.  In section 2, we discuss the selection of low redshift galaxies for comparison with the $z \\sim 1$ 3CR and 6C subsamples. The results of our comparison between low and high redshift sources are presented in section 3, and discussed in section 4. Conclusions are drawn in section 5.  Values for the cosmological parameters $\\Omega_0=0.3$, $\\Omega_\\Lambda=0.7$ and $H_{0}=65\\,\\rm{km\\,s^{-1}\\,Mpc^{-1}}$ are assumed. ",
        "conclusions": "Our conclusions are as follows: \\begin{enumerate} \\item A comparison between the emission line diagnostic plots in Paper 1 and the current paper suggests that the shock model with a precursor model best explains the spectra of small double radio sources.   The emission line regions of larger radio sources are well explained by photoionization by the UV flux from the AGN.  An accurate description of the spectra of small radio sources is likely to require a combination of both shock ionization and photoionization, even where shock ionization is likely to be the most important ionization mechanism. \\item In addition to the known variation in EELR gas kinematics with radio size, the kinematic properties are strongly correlated with redshift and radio power independently.   The more extreme gas kinematics observed at earlier cosmic epochs suggest that the structure and/or composition of the EELR gas clouds may vary with redshift.  Evolution in the environment of the radio source (e.g. changes in IGM/ICM density profile) may also be occuring.  The dependence of the EELR kinematics on radio power is likely to be due to the influence of shocks. \\item The ionization state of the gas is strongly independently correlated with radio size, as well as less strongly correlated with radio power.  The data show no dependence on the redshift of the source once radio power effects have been removed.  The ionization state, and hence the dominant source of ionizing photons, is therefore only dependent on the properties of the AGN (i.e. radio size and radio power) and not on cosmic epoch. \\item Whilst there is a fairly tight distribution of scaled $L\\rm{_{[OII]}}$ with radio size in the low redshift subsample, the weak positive correlation of this data suggests that unlike the strong anticorrelation of emission line luminosity with source age observed in the redshift $z \\sim 1$ sources,  this correlation is much weaker or non-existent for the lower redshift radio galaxies.  This indicates that shocks are less important for the radio galaxies at low redshifts, and is consistent with our other results. \\end{enumerate}"
    },
    "0208/astro-ph0208255_arXiv.txt": {
        "abstract": "Based on a simple thermodynamical argument, we proposed a cloudy model with a warm dust cloud deep in the photosphere. We showed, for the first time, that a single grid of model photospheres ($ 800 \\la T_{\\rm eff} \\la 2600$\\,K) offers a natural explanation not only for the division of dwarfs cooler than M into the two distinct types L and T but also for the changes of the spectra and colors along the L -- T spectral sequence. ",
        "introduction": "The new spectral type L is characterized by the red color possibly due to the dust extinction while type T by the volatile molecules such as methane and water.  Initially, we considered a fully dusty model (model B) and a dust-segregated model (model C) for the L and T type dwarfs, respectively. However, it was difficult to explain why such different cases are realized in different types of cool dwarfs. Also, increasing observations on cool dwarfs could not be explained even by the use of these different kinds of models. We then showed that all these difficulties can be resolved by  models with a warm dust cloud deep in the photosphere (Tsuji 2001), and we  extended this idea to a grid of unified cloudy models (UCMs)(Tsuji 2002). In this contribution, we reexamine  some observed data with our UCMs being updated with the use of the new solar carbon and oxygen abundances by Allende Prieto, Lambert, \\& Asplund (2001, 2002). ",
        "conclusions": "We showed that the observed spectra and colors of L and T dwarfs can be accounted for consistently with our UCMs. On the other hand, our previous models such as the fully dusty models B and the fully dust-depleted models C (and, by implication, more or less similar models  by other authors) are far too short to explain the observed colors (Fig.1a), spectra (Fig.2) and spectral indices (Fig.3) of cool dwarfs, especially from middle L to early T. Thus the presence of the dust cloud deep in the photosphere should be an essential feature of L and T dwarfs, as also noted by Marley et al.(2002) from a different approach based on the planetary theory. We followed stellar approach throughout and pursued a simple model as far as possible, but it is clear that this is only an initial step towards understanding the complicated phenomena in ultracool  dwarfs."
    },
    "0208/astro-ph0208125_arXiv.txt": {
        "abstract": "{\\small We summarize the results of VLBA and global VLBI observations of SS433 between 1995 and 2000. With these observations we resolve the inner jet of the source and identify an absorption region ($\\sim$ 25 AU), the ``central radio gap\". The radio gap is caused by free-free absorption of the jet radio emission by a flattened outflow from the binary system. Radio emission is detected on 100 AU scales perpendicular to the normal jets (``equatorial emission region\"; ``equatorial outflow\"). At some epochs the emission is smooth but compact features are frequently detected. We suggest that equatorial outflows may be common in microquasars. }  {\\small The equatorial outflow in SS433 has been observed in seven VLBI experiments using the VLBA and the EVN. Emission is generally detected at only 1.6 GHz, but there are two epochs when bright features are observed also at 5 GHz. One of these radio components is detected at three epochs, and appears to move outward from the central region at a projected velocity of $\\sim1200\\pm500$~km/s. }  {\\small We present multi-frequency VLBA observations of SS433 during an outburst. Our data suggests that electron acceleration is taking place in the radio plasmons ejected from the central source. } ",
        "introduction": "The radio inner jets resemble those in AGN in that the jet length is roughly inversely proportional to the frequency, and a ``core-shift\" is observed between 1.6 and 5 GHz --- indicative of self-absorption \\cite{Paragi_99}. The differences are that the SS433 jet is ballistic, it contains protons, and a large fraction of the jet material is probably thermal.  The approaching and receding jet sides are separated by a radio gap, as was predicted by Stirling, Spencer \\& Watson \\cite{SSW97}. This is caused by free-free absorption due to a disk-like outflow from the central binary system. The brightness ratio of the two jet sides increases with frequency, and at 22 GHz the receding jet is almost completely absorbed (Fig.~\\ref{fig:bp42d}). The frequency dependence of the size of the mas-scale jet is demonstrated in Fig.~\\ref{fig:logd-lognu}. The Gaussian FWHM of the main jet feature on the approaching jet side is proportional to $\\nu^{-0.93}$ at an epoch when the inner jets were detected up to 22 GHz. The apparent brightness temperature does not show a strong frequency dependence; it is $\\sim$10$^9$~K on the approaching, and $\\sim$10$^8$~K on the receding jet side. At 1.6 GHz the brightness ratio reflects only the Doppler-boosting effect and there seems to be no significant free-free absorption on scales larger than $\\sim$50~AU (assuming a distance of 5 kpc).  During large flares the inner jet region might disappear, and pairs of plasmons are ejected from the system. There is indication for ongoing electron acceleration in these jet components (Paragi, Stirling \\& Fejes, these proceedings).  \\begin{figure}[htb] \\centering \\psfig{file=paragi1_1,width=13cm } \\caption{VLBA image of SS433 at 22 GHz on 16 June 1998 (natural weighting). Contour levels increase by a factor of square root 2, the lowest contour is $\\pm$1\\% of the peak brightness (60.1 mJy/beam). The beam FWHM is 1.61$\\times$1.15~mas, its major axis is oriented at $33.3$ degrees.} \\label{fig:bp42d} \\end{figure}  \\begin{figure}[htb] \\centering \\psfig{file=paragi1_2,width=8cm, angle=-90 } \\caption{Frequency dependence of the deconvolved Gaussian FWHM size of the main jet component on the approaching jet side on 16 June 1998, when SS433 was detected up to 22~GHz. The fitted ``core size\" is proportional to $\\nu^{-0.93\\pm0.20}$. The brightness temperature remains roughly constant, it is order of $10^9$~K.} \\label{fig:logd-lognu} \\end{figure}  \\begin{figure}[htb] \\centering \\psfig{file=paragi1_3,width=12.5cm } \\caption{Global VLBI image of SS433 at 1.6 GHz on 6 June 1998 (natural weighting). The Northern equatorial component (N1) was also detected at 5 GHz at two epochs (Paragi, Fejes \\& Szab\\'o, these proceedeings). Its brightness temperature at 1.6 GHz is $2\\times10^{8}$~K. Contour levels increase by a factor of square root 2, the lowest contour is $\\pm$1\\% of the peak brightness (38.3 mJy/beam). The beam FWHM is 10.7$\\times$3.76~mas, its major axis is oriented at $-5.8$ degrees.} \\label{fig:gp017} \\end{figure}  \\begin{figure}[htb] \\centering \\vskip 0.5cm \\psfig{file=paragi1_4,width=12.5cm } \\caption{Global VLBI image of SS433 at 1.6 GHz on 27 May 2000. We used uniform weighting for this image, which results in the highest possible resolution (and an increasement in the image noise at the same time). The inclusion of Hartebeesthoek in the array improved the N--S resolution significantly. The equatorial component is separate from the jet, but appears very close to it. Contour levels increase by a factor of square root 2, the lowest contour is $\\pm$4\\% of the peak brightness (14.9 mJy/beam). The beam FWHM is 4.85$\\times$2.82~mas, its major axis is oriented at $-8.4$ degrees.} \\label{fig:gp025c} \\end{figure}    Since the Very Long Baseline Array observed SS433 in a multi-frequency experiment in 1995, it has been known that microquasars may produce radio emission not only in the well-studied jets, but also in their equatorial region \\cite{Paragi98,Paragi99}. % Because the presence of an outflow had been suggested based on observations in the optical and in X-ray \\cite{ZWI91,KOT96}, the radio emission was naturally attributed to this equatorial flow. Recent works in the UV % \\cite{DRG02} and IR regimes (Fuchs et al., these proceedings) also support this scenario. We initiated VLBA and global VLBI (EVN+Hartebeesthoek+VLBA) observations to monitor changes in this region on milliarcsecond scales. ",
        "conclusions": ""
    },
    "0208/astro-ph0208313_arXiv.txt": {
        "abstract": "We present high resolution observations of the giant extragalactic \\ion{H}{2} regions NGC 604, NGC 2363, NGC 5461 and NGC 5471, based on observations taken with the ISIS spectrograph on the William Herschel Telescope. We have detected -by the first time- C II and O II recombination lines in these objects. We find that recombination lines give larger C$^{++}$ and O$^{++}$ abundances than collisionallly excited lines, suggesting that temperature variations can be present in the objects. We detect  [\\ion{Fe}{4}] lines in NGC 2363 and NGC 5471, the most confident detection of optical lines of this kind in \\ion{H}{2} regions. Considering the temperature structure we derive their H, He, C, N, O, Ne, S, Ar, and Fe abundances. {From} the recombination lines of NGC 5461 and NGC 5471 we determine the presence of C/H and O/H gradients in M101. We calculate the  $\\Delta Y$/$\\Delta O$ and $\\Delta Y$/$\\Delta Z$ values considering the presence of temperature variations and under the assumption of constant temperature. We obtain a better agreement with models of galactic chemical evolution by considering the presence of temperature variations than by assuming that the temperature is constant in these nebulae. ",
        "introduction": "The analysis of the spectra of \\ion{H}{2} regions allows to determine the abundances of He, C, N, O, Ne, S, Ar and Fe in the ionized phase of the interstellar medium. This is useful to trace the chemical evolution of the interstellar gas, to compute the radial abundance gradients in spiral galaxies and even to estimate the primordial helium abundance. Due to the surface brightness of distant extragalactic \\ion{H}{2} regions it is possible to measure their line intensities with reasonable accuracy. Therefore, it is essential and feasible to have confident determinations of their chemical composition. The possibility to obtain deep spectra of \\ion{H}{2} regions with large telescopes allows us to detect and measure important faint emission lines. Among these, recombination lines (hereafter RLs) of heavy element ions are of special interest. The brightest RLs of heavy element ions in the optical domain are \\ion{C}{2} 4267 \\mbox{\\AA}\\ and those of multiplet 1 of \\ion{O}{2} around 4650 \\mbox{\\AA}. These are in fact very faint lines that have an intensity of the order of 0.001$\\times$ $I(H\\beta)$. These lines can give us a more complete view of the physics and chemical content of nebulae and can be used to test if the standard methods for deriving chemical abundances --based on the intensity of bright collisionally excited lines (hereafter CELs)-- are valid.  The ionic abundances of elements heavier than He are usually derived from the intensity of CELs, which depend exponentially on the electron temperature ($T_e$) of the nebular gas. This fact makes necessary to have a very precise determination of $T_e$ to obtain reliable ionic abundances. \\citet{pei67} found that in the presence of inhomogeneities or stratification in the spatial distribution of $T_e$ (the so-called temperature fluctuations, defined by the mean square temperature variation over the observed volume: $t^2$) the ionic abundances obtained from the intensity of CELs are systematically underestimated. In comparison, ionic abundances determined from RLs are almost independent on $T_e$ and are not sensible to the effects of possible temperature structure inside the nebula. However, the faintness of these lines makes very difficult their measurement and even their detection. \\citet{est98,est99a,est99b} have obtained high resolution observations of the Galactic \\ion{H}{2} regions Orion nebula, M8, and M17, obtaining good measurements of \\ion{C}{2} and \\ion{O}{2} lines in the three objects. These authors have found that ionic abundances derived from those RLs are systematically larger than the values obtained from CELs. A similar result has been obtained by \\citet{tsa03} who present measurements of \\ion{C}{2}, \\ion{N}{2}, and \\ion{O}{2} lines for Orion nebula, M17, NGC 3576, and three Magellanic Clouds \\ion{H}{2} regions (30 Doradus, LMC N11, and SMC N66).  The main aim of the observations reported in this paper was to detect and measure \\ion{C}{2} and \\ion{O}{2} lines in bright giant extragalactic \\ion{H}{2} regions (hereafter GEHRs) of the northern hemisphere. These observations will permit to compare the O$^{++}$ abundances obtained by both CELs and RLs from the same spectrum as well as to derive the C$^{++}$ abundance and compare them with the values derived by other authors from space observations of the  UV [\\ion{C}{3}] 1907 $+$ \\ion{C}{3}]  1909 \\mbox{\\AA}\\ lines.  \\footnotetext{Based on observations made with William Herschel Telescope operated on the island of La Palma by the Isaac Newton Group of Telescopes in the Spanish Observatorio del Roque de Los Muchachos of the Instituto de Astrof\\'\\i sica de Canarias.} ",
        "conclusions": "Recombination lines of C II and O II are detected by the first time in some of the most interesting GEHRs of the Northern hemisphere. The comparison of ionic abundances obtained from these lines with those obtained from CELs suggests the presence of temperature variations inside these nebulae.  [\\ion{Fe}{4}] optical  lines are detected in NGC 2363 and NGC 5471, this is the clearest identification of these lines in \\ion{H}{2} regions.  The presence of values of $t^2$ $>$ 0 in GEHRs has important consequences in various  fields of astrophysics, mainly because most of our knowledge about the chemical content of extragalactic objects comes from the spectra of GEHRs. The usual assumption of $t^2 = 0$ leads to underestimate the heavy element abundances derived from CELs. Moreover the usual assumption of $t^2 = 0$ leads to underestimate $Y_p$ the primordial helium abundance \\citep{pei00,pei02}.  {From} the RLs of C and O we confirm the existence of C/H and O/H gradients in M101 derived from CELs.  We find a poor agreement between the predicted $\\Delta Y$/$\\Delta O$ and $\\Delta Y$/$\\Delta Z$ values by galactic chemical evolution models and those derived from observations under the assumption that $t^2 = 0$. Alternatively we find a good agreement between the predicted values and those derived from the RLs of C and O and the CELs of N, Ne, S, Ar, and Fe under the assumption that $t^2$ $>$ 0."
    },
    "0208/astro-ph0208035_arXiv.txt": {
        "abstract": "The neutrino flux and spectra formation in a supernova core is studied by using a Monte Carlo code. The dominant opacity contribution for $\\nu_\\mu$ is elastic scattering on nucleons $\\nu_{\\mu}\\rN\\to\\rN\\nu_{\\mu}$, where $\\nu_\\mu$ always stands for either $\\nu_\\mu$ or $\\nu_\\tau$.  In addition we switch on or off a variety of processes which allow for the exchange of energy or the creation and destruction of neutrino pairs, notably nucleon bremsstrahlung $\\rN\\rN\\to\\rN\\rN\\nu_{\\mu}\\bar\\nu_{\\mu}$, the pair annihilation processes $\\re^+\\re^-\\to\\nu_{\\mu}\\bar\\nu_{\\mu}$ and $\\nue\\bar\\nue\\to\\nu_{\\mu}\\bar\\nu_{\\mu}$, recoil and weak magnetism in elastic nucleon scattering, elastic scattering on electrons $\\nu_{\\mu}\\re^\\pm\\to\\re^\\pm\\nu_{\\mu}$ and elastic scattering on electron neutrinos and anti-neutrinos $\\nu_{\\mu}\\nue\\to \\nue\\nu_{\\mu}$ and $\\nu_{\\mu}\\bar\\nue\\to \\bar\\nue\\nu_{\\mu}$. The least important processes are neutrino-neutrino scattering and $\\re^+\\re^-$ annihilation. The formation of the spectra and fluxes of $\\nu_\\mu$ is dominated by the nucleonic processes, i.e.\\ bremsstrahlung and elastic scattering with recoil, but also $\\nue\\bar\\nue$ annihilation and $\\nu_{\\mu}\\re^\\pm$ scattering contribute significantly.  When all processes are included, the spectral shape of the emitted neutrino flux is always ``pinched,'' i.e.\\ the width of the spectrum is smaller than that of a thermal spectrum with the same average energy.  In all of our cases we find that the average $\\bar\\nu_{\\mu}$ energy exceeds the average $\\bar\\nue$ energy by only a small amount, 10\\% being a typical number.  Weak magnetism effects cause the opacity of $\\nu_\\mu$ to differ slightly from that of $\\bar\\nu_\\mu$, translating into differences of the luminosities and average energies of a few percent. Depending on the density, temperature, and composition profile, the flavor-dependent luminosities $L_{\\nue}$, $L_{\\bar\\nue}$, and $L_{\\nu_\\mu}$ can mutually differ from each other by up to a factor of two in either direction. ",
        "introduction": "}  \\label{sec:Introduction}  In numerical core-collapse supernova (SN) simulations, the transport of $\\mu$- and $\\tau$-neutrinos has received scant attention because their exact fluxes and spectra are probably not crucial for the explosion mechanism.  However, the recent experimental evidence for neutrino oscillations implies that the flavor-dependent fluxes and spectra emitted by a SN will be partly swapped so that at any distance from the source the actual fluxes and spectra can be very different from those originally produced.  In principle, this effect can be important for the SN shock revival (Fuller et~al.\\ 1992) and r-process nucleosynthesis (Qian et~al.\\ 1993, Pastor \\& Raffelt 2002), although the experimentally favored small neutrino mass differences suggest that this is not the case.  On the other hand, in view of the large-mixing-angle solution of the solar neutrino problem flavor oscillations are quite relevant for the interpretation of the SN~1987A neutrino signal (Jegerlehner, Neubig, \\& Raffelt 1996, Lunardini \\& Smirnov 2001a, Kachelriess et~al.\\ 2002, Smirnov, Spergel, \\& Bahcall 1994).  More importantly, the high-statistics neutrino signal from a future galactic SN may allow one to differentiate between some of the neutrino mixing scenarios which explain the presently available data (Chiu \\& Kuo 2000, Dighe \\& Smirnov 2000, Dutta et~al.\\ 2000, Fuller, Haxton, \\& McLaughlin 1999, Lunardini \\& Smirnov 2001b, 2003, Minakata \\& Nunokawa 2001, Takahashi \\& Sato 2002).  Even though the solution of the solar neutrino problem has been established, the magnitude of the small mixing angle $\\Theta_{13}$ and the question if the neutrino mass hierarchy is normal or inverted will remain open and can be settled only by future precision measurements at dedicated long-baseline oscillation experiments (Barger et~al.\\ 2001, Cervera et~al.\\ 2000, Freund, Huber, \\& Lindner 2001) and/or the observation of a future galactic SN.  The usefulness of SN neutrinos for diagnosing flavor oscillations depends on the flavor dependence of the fluxes and spectra at the source.  Very crudely, a SN core is a black-body source of neutrinos of all flavors which are emitted from the surface of the proto-neutron star that was born after collapse.  It is the flavor-dependent details of the neutrino transport in the neutron-star atmosphere which cause the spectral and flux differences that can lead to interesting oscillation effects.  The $\\nue$ and $\\bar\\nue$ opacity is dominated by the charged-current processes $\\nue\\rn\\to\\rp\\re^-$ and $\\bar\\nue\\rp\\to\\rn\\re^+$, reactions that allow for the exchange of energy and lepton number between the medium and the neutrinos.  Therefore, it is straightforward to define an energy-dependent neutrinosphere where this reaction freezes out for neutrinos of a particular energy. This sphere yields a thermal contribution to the neutrino flux at the considered energy.  The atmosphere of the proto-neutron star is neutron rich, providing for a larger $\\nue$ opacity than for $\\bar\\nue$ so that for a given energy the $\\bar\\nue$ flux originates at deeper and thus hotter layers than the $\\nue$ flux. In other words, a larger fraction of the $\\bar\\nue$ flux emerges with high energies.  This simple observation explains the usual hierarchy $\\langle \\epsilon_{\\bar\\nue}\\rangle>\\langle \\epsilon_{\\nue}\\rangle$ of the mean energies.  The spectra are found to be ``pinched'', meaning that the high-energy tail is suppressed relative to that of a thermal spectrum with the same mean energy (Janka \\& Hillebrandt 1989a,b).  This numerical result can be understood analytically by constructing the neutrino spectrum from the fluxes emitted by the energy-dependent neutrinospheres which are at different temperatures (Myra, Lattimer, \\& Yahil 1988, Giovanoni, Ellison, \\& Bruenn 1989).  The formation of the $\\nu_\\mu$, $\\bar\\nu_\\mu$, $\\nu_\\tau$, and $\\bar\\nu_\\tau$ spectra is far more complicated.  The opacity is dominated by the neutral-current scattering on nucleons, $\\nu_\\mu\\rN\\to\\rN\\nu_\\mu$, a process that prevents neutrino free streaming, but is unable to change the neutrino number and is usually considered to be inefficient at exchanging energy.  (Here and in the following $\\nu_\\mu$ stands for either $\\nu_\\mu$ or $\\nu_\\tau$.) Neutrino pairs can be created by nucleon bremsstrahlung, $\\rN\\rN\\to\\rN\\rN \\nu_\\mu \\bar\\nu_\\mu$, and pair annihilation, $\\re^-\\re^+\\to \\nu_\\mu\\bar\\nu_\\mu$ or $\\nue\\bar\\nue\\to \\nu_\\mu\\bar\\nu_\\mu$, while $\\nu_\\mu\\bar\\nu_\\mu$ pairs are absorbed by the inverse reactions.  In addition, energy is exchanged by elastic scattering on leptons, notably $\\nu_\\mu\\re^-\\to \\re^-\\nu_\\mu$, by the recoil in nucleon scattering, $\\nu_\\mu\\rN\\to\\rN\\nu_\\mu$, and by inelastic scattering on nucleons $\\nu_\\mu\\rN\\rN\\to\\rN\\rN\\nu_\\mu$, a channel that is the ``crossed process'' of bremsstrahlung.  For a given neutrino energy these processes freeze out at different radii so that one can define a ``number sphere'' for the pair processes, an ``energy sphere'' for the energy-exchange processes, and a ``transport sphere'' for elastic nucleon scattering with $R_{\\rm number}<R_{\\rm energy}<R_{\\rm transport}$ (Suzuki 1990).  The region between the number sphere and the transport sphere plays the role of a {\\em scattering atmosphere\\/} because neutrinos can not be created or destroyed. They propagate by diffusion and can still exchange energy with the background medium.  Usually the $\\nu_{\\mu}$ transport sphere is deeper than the $\\bar\\nue$ sphere so that numerical simulations find $\\langle\\epsilon_{\\nu_\\mu}\\rangle>\\langle\\epsilon_{\\bar\\nue}\\rangle >\\langle\\epsilon_{\\nue}\\rangle$.  This hierarchy is the main motivation for the proposed use of SN neutrinos as a diagnostic for neutrino oscillations.  However, the quantitative statements found in the literature range from $\\langle\\epsilon_{\\nu_\\mu}\\rangle$ being 20\\% to nearly a factor of 2 larger than $\\langle\\epsilon_{\\bar\\nue}\\rangle$; for a review see Janka (1993) and Sec.~\\ref{sec:PreviousLiterature}.  Of course, the mean energies and their ratios change significantly between the SN bounce, accretion phase, and the later neutron-star cooling phase.  Therefore, one must distinguish carefully between instantaneous fluxes and spectra and the time-integrated values.  While for the analysis of the sparse SN~1987A data only time-integrated values make sense, a future galactic SN may well produce enough events to study the instantaneous fluxes and spectra (Barger, Marfatia, \\& Wood 2001, Minakata et~al.\\ 2001).  The overall energy emitted by a SN is often said to be equipartitioned among all six neutrino degrees of freedom.  In some numerical simulations the neutrino luminosities are indeed astonishingly equal for all flavors (Totani et~al.\\ 1998), while other simulations easily find a factor of two difference between, say, the $\\bar\\nu_\\mu$ and $\\bar\\nue$ luminosities, at least during the accretion phase (Mezzacappa et~al.\\ 2001).  Therefore, it is by no means obvious how precisely equipartition can be assumed for the purpose of diagnosing neutrino oscillations.  Another important feature is the neutrino spectral shape, notably the amount of pinching.  If one could assume with confidence that the instantaneous spectra of all flavors are pinched at the source, and if the measured SN neutrino spectra were instead found to be anti-pinched, this effect would be a powerful diagnostic for the partial spectral swapping caused by flavor oscillations (Dighe \\& Smirnov 2000).  Unfortunately, the existing literature does not allow one to develop a clear view on these ``fine points'' of the neutrino fluxes and spectra, largely because not enough attention has been paid to the $\\nu_\\mu$ and $\\nu_\\tau$ emission from a SN core.  The published full numerical SN collapse simulations have not yet included the bremsstrahlung process or nucleon recoils (but see first results of state-of-the-art models in Rampp et al.\\ 2002), even though it is no longer controversial that these effects are important (Janka et~al.\\ 1996, Burrows et~al.\\ 2000, Hannestad \\& Raffelt 1998, Raffelt 2001, Suzuki 1991, 1993, Thompson, Burrows, \\& Horvath 2000).  Moreover, some of the interesting information such as the spectral pinching was usually not documented.  Another problem with self-consistent hydrodynamic simulations is that the models with the most elaborate neutrino transport usually do not explode so that even the most recent state-of-the-art simulations do not reach beyond the accretion phase at a few hundred milliseconds after bounce (Rampp \\& Janka 2000, Mezzacappa et~al.\\ 2001, Liebend\\\"orfer et~al.\\ 2001), thus not providing any information on the neutron-star cooling phase. Successful multi-dimensional models of the explosion (e.g., Fryer \\& Warren 2002, Fryer 1999 and references therein) were also not continued to the neutron-star cooling phase. These simulations, moreover, treat the neutrino transport only in a very approximate way and do not provide spectral information. The calculations performed by the Livermore group also yield robust explosions (Totani et~al.\\ 1998).  They include a mixing-length treatment of the phenomenon of neutron-finger convection in the neutron star, that increases the early neutrino luminosities and thus enhances the energy transfer by neutrinos to the postshock medium (Wilson \\& Mayle 1993). Whether neutron-finger convection actually occurs inside the neutrinosphere and has effects on a macroscopic scale, however, is an unsettled issue.  We will follow here an alternative approach to full hydrodynamic simulations, i.e.\\ we will study neutrino transport on the background of an assumed neutron-star atmosphere.  While this approach lacks hydrodynamic self-consistency, it has the great advantage of allowing one to study systematically the influence of various pieces of microscopic input physics and of the medium profile.  The goal is to develop a clearer picture of the generic properties of the SN neutrino spectra and fluxes and what they depend upon.  To this end we have adapted the Monte Carlo code of Janka (1987, 1991) and added new microphysics to it. We go beyond the work of Janka \\& Hillebrandt (1989a,b) in that we include the bremsstrahlung process, nucleon recoils and weak magnetism, $\\nue \\bar\\nue$ pair annihilation into $\\nu_\\mu\\bar\\nu_\\mu$, and scattering of $\\nu_{\\mu}$ on $\\nue$ and $\\bar\\nue$. With these extensions we investigate the neutrino transport systematically for a variety of medium profiles that are representative for different SN phases.  One of us (Raffelt 2001) has recently studied the $\\nu_\\mu$ spectra-formation problem with the limitation to nucleonic processes (elastic and inelastic scattering, recoils, bremsstrahlung), to Maxwell-Boltzmann statistics for the neutrinos, and plane-parallel geometry.  Our present study complements this more schematic work by including the leptonic processes, Fermi-Dirac statistics, and spherical geometry.  In addition we apply our Monte Carlo code to the transport of $\\nue$ and $\\bar\\nue$ and thus are able to compare the flavor-dependent fluxes and spectra.  In Sec.~2 we first assess the relative importance of different processes in terms of their energy-dependent ``thermalization depth''. In this context we introduce a number of stellar background models. In Sec.~3 we perform a Monte Carlo study of $\\nu_\\mu$ transport on the previously introduced background models in order to assess the importance of different pieces of input physics.  In Sec.~4 we compare the $\\nu_\\mu$ fluxes and spectra with those of $\\nue$ and $\\bar\\nue$. We conclude in Sec.~5 with a discussion and summary of our findings. ",
        "conclusions": "}  We have studied the formation of neutrino spectra and fluxes in a SN core.  Using a Monte Carlo code for neutrino transport, we varied the microscopic input physics as well as the underlying static proto-neutron star atmosphere. We used two background models from self-consistent hydrodynamic simulations, and several power-law models with varying power-law indices for the density and temperature and different values for the electron fraction $\\Ye$, taken to be constant.  The $\\nu_\\mu$ transport opacity is dominated by neutral-current scattering on nucleons. In addition, there are number-changing processes (nucleon bremsstrahlung, leptonic pair annihilation) and energy-changing processes (nucleon recoil, $\\nu_\\mu\\re^\\pm$ scattering). The $\\nu_\\mu$ spectra and fluxes are roughly accounted for if one includes one significant channel of pair production and one for energy exchange in addition to $\\nu_\\mu{\\rN}$ scattering. For example, the traditional set of microphysics (iso-energetic $\\nu_\\mu{\\rN}$ scattering, $\\re^+\\re^-$ annihilation, and $\\nu_\\mu\\re^\\pm$ scattering) yields comparable spectra and fluxes to a calculation where pairs are produced by nucleon bremsstrahlung and energy is exchanged by nucleon recoil. The overall result is quite robust against the detailed choice of microphysics.  However, in state-of-the-art simulations where one aims at a precision better than some 10--20\\% for the fluxes and spectral energies, one needs to include bremsstrahlung, leptonic pair annihilation, neutrino-electron scattering, and energy transfer in neutrino-nucleon collisions.  Interestingly, the traditional $\\re^+\\re^-$ annihilation process is always much less important than $\\nue\\bar\\nue$ annihilation, a point that we previously raised with our collaborators (Buras et~al.\\ 2002). None of the reactions studied here can be neglected except perhaps the traditional $\\re^+\\re^-$ annihilation process and $\\nu_\\mu\\nue$ and $\\nu_\\mu\\bar\\nue$ scattering.  The existing treatments of the nuclear-physics aspects of the $\\rN\\rN\\to\\rN\\rN\\nu\\bar\\nu$ bremsstrahlung process are rather schematic. We find, however, that the $\\nu_\\mu$ fluxes and spectra do not depend sensitively on the exact strength of the bremsstrahlung rate.  Therefore, while a more adequate treatment of bremsstrahlung remains desirable, the final results are unlikely to be much affected.  The transport of $\\nu_\\mu$ and $\\bar\\nu_\\mu$ is usually treated identically.  However, weak-magnetism effects render the $\\nu_\\mu\\rN$ and $\\bar\\nu_\\mu\\rN$ scattering cross sections somewhat different (Horowitz 2002), causing a small $\\nu_\\mu$ chemical potential to build up.  We find that the differences between the average energies of $\\nu_\\mu$ and $\\bar\\nu_\\mu$ are only a few percent and can thus be neglected for most purposes.  Including all processes works in the direction of making the fluxes and spectra of $\\nu_\\mu$ more similar to those of $\\bar\\nue$ compared to a calculation with the traditional set of input physics.  During the accretion phase the neutron-star atmosphere is relatively expanded, i.e.\\ the density and temperature gradients are relatively shallow. Our investigation suggests that during this phase $\\langle\\epsilon_{\\nu_\\mu}\\rangle$ is only slightly larger than $\\langle\\epsilon_{\\bar\\nue}\\rangle$, perhaps by a few percent or 10\\% at most. This result agrees with the first hydrodynamic simulation including all of the relevant microphysics except $\\nue\\bar\\nue$ annihilation (Accretion-Phase Model II) provided to us by M.~Rampp. For the luminosities of the different neutrino species one finds $L_{\\bar\\nue}\\sim L_{\\nue}\\sim 2\\,L_{\\nu_\\mu}$.  The smallness of $L_{\\nu_\\mu}$ is not surprising because the effective radiating surface is much smaller than for~$\\bar\\nue$.  During the Kelvin-Helmholtz cooling phase the neutron-star atmosphere will be more compact, the density and temperature gradients will be steeper. Therefore, the radiating surfaces for all species will become more similar. In this situation $L_{\\nu_\\mu}$ may well become larger than $L_{\\bar\\nue}$. However, the relative luminosities depend sensitively on the electron concentration. Therefore, without a self-consistent hydrostatic late-time model it is difficult to claim this luminosity cross-over with confidence.  The ratio of the spectral energies is most sensitive to the temperature gradient relative to the density gradient. In our power-law models we used $\\rho\\propto r^{-p}$ and $T\\propto r^{-q}$. Varying $q/p$ between 0.25 and 0.35 we find that $\\langle\\epsilon_{\\bar\\nue}\\rangle:\\langle\\epsilon_{\\nu_\\mu}\\rangle$ varies between $1:1.10$ and $1:1.22$. Noting that the upper range for $q/p$ seems unrealistically large we conclude that even at late times the spectral differences should be small; 20\\% sounds like a safe upper limit. We are looking forward to this prediction being checked in a full-scale self-consistent neutron-star evolution model with a Boltzmann solver.  The statements in the previous literature fall into two classes. One group of workers, using the traditional set of microphysics, found spectral differences between $\\bar\\nue$ and $\\nu_\\mu$ on the 25\\% level, a range which largely agrees with our findings in view of the different microphysics.  Other papers claim ratios as large as $\\langle\\epsilon_{\\bar\\nue}\\rangle:\\langle\\epsilon_{\\nu_\\mu}\\rangle =1:1.8$ or even exceeding $1:2$.  We have no explanation for these latter results. At least within the framework of our simple power-law models we do not understand which parameter could be reasonably adjusted to reach such extreme spectral differences.  In a high-statistics neutrino observation of a future galactic SN one may well be able to discover signatures for flavor oscillations. However, when studying these questions one has to allow for the possibility of very small spectral differences, and conversely, for the possibility of large flux differences. This situation is almost orthogonal to what often has been assumed in papers studying possible oscillation signatures. A realistic assessment of the potential of a future galactic SN to disentangle different neutrino mixing scenarios should allow for the possibility of very small spectral differences among the different flavors of anti-neutrinos. The spectral differences between $\\nu_e$ and $\\nu_{\\mu,\\tau}$ are always much larger, but a large SN neutrino (as opposed to anti-neutrino) detector does not exist.  The diffuse neutrino flux from all past SNe in the universe is difficult to detect, although Super-Kamiokande has recently established an upper limit that touches the upper end of theoretical predictions (Malek et al.\\ 2002).  If our findings are correct, neutrino oscillations will not much enhance the high-energy tail of the spectrum and thus will not significantly enhance the event rate."
    },
    "0208/astro-ph0208203_arXiv.txt": {
        "abstract": "Results of an international campaign to photometrically monitor the unique pre-main sequence eclipsing object KH 15D are reported. An updated ephemeris for the eclipse is derived that incorporates a slightly revised period of 48.36 d. There is some evidence that the orbital period is actually twice that value, with two eclipses occurring per cycle. The extraordinary depth ($\\sim$3.5 mag) and duration ($\\sim$18 days) of the eclipse indicate that it is caused by circumstellar matter, presumably the inner portion of a disk. The eclipse has continued to lengthen with time and the central brightness reversals are not as extreme as they once were. V-R and V-I colors indicate that the system is slightly bluer near minimum light. Ingress and egress are remarkably well modeled by the passage of a knife-edge across a limb-darkened star. Possible models for the system are briefly discussed. ",
        "introduction": "Planets are believed to form during the first $\\sim$100 My of a star's life from matter in a circumstellar disk \\citep{cw98}. The initial stages of that process have been viewed with increasing clarity in high-resolution images of young stars obtained with the Hubble Space Telescope. These have revealed complex disk structures on scales of tens to hundreds of AU around $\\sim$1 My old objects \\citep{o01,ksw}. Probing the inner parts of disks, where terrestrial planets formed in our solar system and where giant planets are found in many exo-solar systems \\citep{mb00,v02}, however, is still well beyond the reach of the current generation of telescopes. Here we report observations of a remarkable eclipsing solar-like star at an age of $\\sim$3 My, which is providing a glimpse of the structure and possibly evolution of circumstellar matter on scales as fine as 0.01 AU. This is possible because of its unique geometry that results in the star being periodically eclipsed by extended non-luminous matter in its vicinity. No other object in the history of astronomy has been found to behave in quite the same way.  The star in question, KH 15D ($\\alpha = 6^{h}41^{m}10.18^{s}$, $\\delta = 9\\degr 28\\arcmin 35.5\\arcsec$, epoch 2000; see \\citet{h01} for an identification chart), was first noticed in 1997 as a unique and potentially important object during a photometric monitoring program of young clusters undertaken with a small telescope at Van Vleck Observatory (VVO) on the campus of Wesleyan University in Middletown, CT \\citep{kehm,kh98}. It undergoes a very deep ($\\sim$3.5 mag) eclipse every 48.3 days and remains in the faint state for about 18 days, suggesting that the eclipsing body is non-luminous circumstellar matter. Remarkably, there is a central reversal of variable height that sometimes returns the star to its out-of-eclipse level (and once to an even brighter level) for a brief time near mid-eclipse. A follow-up study by \\citet{h01} revealed that the star has a mass of 0.5 - 1 solar masses, and a radius of about 1.3 solar radii, indicating that it is still in its contraction phase towards the main sequence and has not yet initiated hydrogen burning in its core. Its pre-main sequence status is confirmed by the presence of Li I in its spectrum and its membership in the young open cluster NGC 2264, which has an age of 2-4 My and a distance of 760 pc \\citep{sbl}. No other star, among the thousands monitored in the VVO program or elsewhere over more than a decade \\citep{smmv,hrhc,reb01,hbjm} has behaved in similar fashion. The closest analogues we can find in the astronomical literature are $\\epsilon$ Aur, an F-type supergiant which is eclipsed every 27 years by dark circumstellar matter \\citep{l96}, and BM Ori, an early B star which is eclipsed every 6.5 days by something dark orbiting it \\citep{pp76}. Neither star is solar-like nor are the eclipses even remotely similar to those of KH 15D. ",
        "conclusions": "KH 15D is a unique and amazing object that promises to tell us much about conditions in the inner circumstellar region of a solar-like star of planet-forming age. The principle purpose of this contribution is to summarize the available observations of the star with the hope of stimulating additional observations, and to support those by providing an updated ephemeris and description of the phenomena. However, a brief discussion of possible models is in order, based on the information currently available. These fall into two categories, depending on whether the K7 star is the dominant mass in the system or not. If it is, then the basic model involves occulting matter orbiting the star. If it is not (i.e. if KH 15D is a binary star system and the unseen component has a mass comparable to or larger than the K7 star), then the observed phenomena could be caused wholly or in part by the motion of the visible star with respect to a circumbinary disk. We discuss the ``single star'' interpretation first, noting that it also applies to binary models in which the K7 star is the dominant mass.  Assuming that the eclipse is caused by a feature or features orbiting the K7 star with a period of 48 or 97 days, we can apply Kepler's third law to derive a semi-major axis for the orbit of the occulting matter of 0.21 or 0.32 AU, respectively. Therefore, if this model is correct, we are probing a region of the circumstellar environment closer to the star than Mercury is to the Sun. The occulting feature appears to have a very sharp edge, such that a knife-edge model fits the data well. However, the transit time for an object orbiting at 0.2-0.3 AU from the K7 star is only $\\sim$0.5 d, much less than the time of ingress or egress ($\\sim$2-2.5 d). Therefore, if the star is eclipsed by a sharp-edged orbiting feature, its occulting edge must be inclined by about 15$\\degr$ to its direction of motion. A sketch of what this would look like during the early stages of ingress is shown in Fig. \\ref{model}. Such an eclipsing structure could be caused by density waves or a corrugation of a disk driven by an embedded planet or brown dwarf, as \\citet{b00} for example, have modeled. Alternatively, the features could be associated with a resonance of a yet undetected mass. During the eclipse, the system is seen primarily or entirely by reflected light from the circumstellar matter. Models such as these have many attractive features, including the potential to explain the central reversal as back scattering off the wave on the far side of the star or as a local minimum in the opacity near the location of an embedded planet or proto-planet.  If KH 15D is a binary in which the K7 star moves substantially with respect to the center of mass of the system, then an entirely different explanation of the observed phenomena is possible. Namely, one could imagine the K7 star passing above (and, possibly, alternately below) the plane of an occulting (presumably circumbinary) disk. In other words, it would be primarily the motion of the star, in this model, that was causing the eclipse, not the motion of the occulting matter. As noted previously, this appears less likely at present for two reasons. First, there is only evidence for a single stellar spectrum from the system at any phase, even when the photosphere of the K7 star is completely occulted. Second, there is very little difference in the radial velocity of the K7 stars on two dates separated by 21 d. Neither of these arguments is sufficient, however, to rule out a binary model. For example, if the visible K7 star were on an eccentric orbit about a slightly more massive star and the orbit was properly inclined to the plane of a circumbinary disk, one could reproduce most or all of the observations, including the central peaks and the ingress/egress time scales. A comprehensive radial velocity study, as is under way, is clearly required to make progress. We note that, regardless of whether KH 15D proves to be a single or binary star, its unique orientation provides us with a powerful tool for studying the structure and evolution of circumstellar matter close to a young star on an unprecedented fine scale."
    },
    "0208/astro-ph0208490_arXiv.txt": {
        "abstract": " ",
        "introduction": "The Kuiper belt\\cite{LuJ02} is the best solar system laboratory for studies of the early stages of runaway accretion. Runaway accretion in the Kuiper belt terminated when the velocity dispersion of its members was increased by an as yet undetermined process\\cite{MJP02}. Unlike the asteroid belt, its largest members have suffered little collisional evolution since.  The discovery that a substantial fraction of its largest members are in binaries with wide separations and order unity mass ratios\\cite{EKO01,KPG+01,BrT02,TrB02,NSG+02,NSG+02b,VPG02} is the latest of many surprises provided by the Kuiper belt. Collisions coupled with tidal evolution, mechanisms that may explain other solar system binaries, fail to account for the formation of the Kuiper belt binaries.  The low frequency of collisions among the large Kuiper belt objects\\cite{Ste02} implies that binaries formed by collisionless interactions mediated by gravity. These were most effective earlier when dynamical friction due to small bodies limited the velocity dispersion of the large ones. This situation pertains during runaway accretion.  In the following section \\S \\ref{sec:prelim}, we outline a simple model for runaway accretion to set the stage for binary formation. We are guided by more detailed formalisms implemented in numerical simulations.  Values of relevant parameters are estimated based on the numbers and sizes of objects deduced from Kuiper belt surveys, and by the extrapolation of the surface density in the minimum mass solar nebula. We estimate the binary formation rate and derive the semi-major axis distribution in section \\S \\ref{sec:binary}. In the final section \\S \\ref{sec:discuss}, we discuss the observational implications of our results and mention several open issues. ",
        "conclusions": "\\label{sec:discuss}  We propose that the wide binaries observed in the Kuiper belt formed during runaway accretion. A fraction of the large bodies that entered each other's Hill spheres became bound as the result of energy lost to small bodies by dynamical friction.\\footnote{Interaction with a third large body was a less important channel for binary formation.} The time scale for a large body to become bound to a similar companion was \\begin{equation} \\left(\\rho R \\over \\sigma\\right)^2 \\left(\\Sigma\\over\\sigma\\right) {\\theta_\\odot^2\\over \\Omega_\\odot} \\sim 3 \\times 10^5 \\yr, \\end{equation}  Further dynamical friction hardened the binaries. The timescale to achieve contact was that during which isolated large bodies grew; \\begin{equation} {\\rho R\\over \\sigma}\\left(\\Sigma\\over\\sigma\\right){1\\over \\Omega_\\odot} \\sim 10^6\\yr. \\end{equation} Initially the inspiral was at a constant rate, but it slowed down inside $r_u$ where the small bodies' number density and velocity dispersion were enhanced above their background values. We deduce that the probability that a large body is part of a binary with angular separation greater than $r_u/a_\\odot\\sim 3''$ is \\begin{equation} {\\Sigma\\over \\rho R} \\left(1\\over \\theta_\\odot\\right)^2 \\sim 3\\times 10^{-3}. \\end{equation} Inward of $r_u$, the binary probability per logarithmic interval of semimajor axis increases inversely with semimajor axis. Close to contact, the probability exceeds unity for $R<a_{crit}\\sim 100\\km$, which implies that systems with higher multiplicities would exist if such are stable. For the resolution of the HST survey $\\sim 0.2''$, we predict a binary fraction of about $5\\%$, roughly compatible with observations (Brown, private communication). Our prediction that close binaries are common could be tested by monitoring the brightness of Kuiper belt objects for evidence of eclipses and/or fast rotation. The latter is a consequence, but not a unique signature, of binary mergers.  In our analysis we estimate the surface density ratio of large to small bodies from the observational census of Kuiper belt objects together with an extrapolation of the surface density of the minimum mass solar nebula. It would be an improvement to have a theoretical understanding of how this ratio evolves during runaway accretion. Also, we elaborated the binary evolution scenario with a given set of parameters, specifically $\\theta_\\odot$ and $\\Sigma /\\sigma$. Different scalings that apply for other parameter regimes remain to be worked out.  Additional predictions suitable for testing by future observations of Kuiper belt binaries could be made by extending our model in a variety of ways. Relaxing the restriction to two mass groups would allow us to investigate the mass ratio distribution of binary components and its dependence on semi-major axis. We predict the formation of multiple systems for bodies with $R<a_{crit}$. Numerical experiments can establish what types of systems survive disruption due to mutual interactions of their members. We could also predict eccentricity and inclination distributions. This would require a more detailed look at the anisotropy of the velocity distribution of small bodies, especially for $a<r_u$. Anisotropy leads to tensorial dynamical friction in which the force is not antiparallel to the velocity. Binney\\cite{Bin77} has investigated a related problem, the evolution of the orbits of galaxies in anisotropic clusters. We can already say that the eccentricities are likely to be large.  They begin large following binary capture. Moreover, since the magnitude of the dynamical friction force increases outward for $a<r_u$, they should remain large in this region.  Simple numerical simulations of the Hill problem that incorporate dynamical friction would help pin down the dependence of the capture probability on the velocity dispersion and on the strength of dynamical friction. Computations involving three massive bodies are needed for evaluation of capture probabilities via that channel. Three body capture cross sections have been computed for stars in globular clusters \\cite{HuB83}, but in our regime the Hill sphere must be taken into account.  The two largest known Kuiper belt objects, Pluto and Charon, form a binary. Collision coupled with tidal evolution is the generally accepted mode of formation for this system. Our scenario for binary formation provides another candidate. It may also apply to the formation of some binaries in the asteroid belt.  In closing, we caution that our numerical results are crude. Dynamical processes have been estimated to order of magnitude. We have ruthlessly discarded numerical coefficients and logarithmic factors. More credence should be accorded to the functional dependences of our results than to the precise numerical predictions."
    },
    "0208/astro-ph0208459_arXiv.txt": {
        "abstract": "We present numerical magnetohydrodynamic  (MHD) simulations of a magnetized accretion disk launching trans-Alfv\\'enic jets.  These simulations, performed in a 2.5 dimensional time-dependent polytropic resistive MHD framework,  model a resistive accretion disk threaded by an initial vertical magnetic field. The resistivity is only important inside the disk, and is prescribed as $\\eta=\\alpha_m V_AH\\exp(-2Z^2/H^2)$ , where  $V_A$ stands for Alfv\\'en speed, $H$ is the disk scale height and the coefficient $\\alpha_m$ is smaller than unity.   By performing the simulations over several tens of dynamical disk timescales, we show that the launching of a collimated outflow occurs self-consistently and  the ejection of matter is continuous and quasi-stationary. These are the first ever simulations of resistive accretion disks launching non-transient ideal MHD jets. Roughly 15 \\% of accreted mass is persistently ejected.  This outflow is safely characterized as a jet since the  flow becomes super-fastmagnetosonic, well-collimated and reaches a quasi-stationary state. We present  a complete illustration and explanation of the `accretion-ejection' mechanism that leads to jet formation from a magnetized accretion disk. In particular, the magnetic torque inside the disk brakes the matter azimuthally and allows for accretion,  while it is responsible for an effective magneto-centrifugal acceleration in the jet.  As such, the magnetic field channels the disk angular momentum and powers the jet acceleration and collimation. The jet originates from the inner disk region where equipartition between thermal and magnetic forces is achieved. A hollow, super-fastmagnetosonic shell of dense material is the natural outcome of the inwards advection of a primordial field. ",
        "introduction": "\\subsection{Observational data}\\label{s:obs}  Jets are ubiquitous phenomena in astrophysics since they are observed around young stellar objects (YSOs), galactic compact objects (neutron stars, X-ray binaries) and in some active galactic nuclei \\citep{Livi97}. These outflows are called jets because they display a very good collimation far from the central object and reach high velocities. Indeed, jets have terminal velocities of roughly $400 \\,\\, {\\rm km/s}$ around YSOs  (e.g. \\citet{Garc01} and references therein), or reach a fraction of the light speed for X-ray binaries \\citep{Mira99} and Fanaroff-Riley class I (FRI) jets. Relativistic motions are also detected in FRII jets launched from active galactic nuclei \\citep{Verm94}. Any model to explain the creation of such outflows must provide a mass source for the jets, achieve their collimation and contain a mechanism to accelerate mass within the jet.  All astrophysical systems mentioned have accretion disks  associated with them. Observed correlations between emission from the accretion disk and from the jet provide evidence that the  jets are launched from the disks directly.  In YSOs, since the work of \\citet{Cabr90},  it is clear that the disk luminosity is correlated with the light coming from optical forbidden lines emitted in the jets (see also \\citet{Hart95}). In galactic systems, a similar correlation occurs, as for example in the first detected microquasar GRS 1915+105 \\citep{Mira98}. In these systems the disk radiation is characterized by X-ray emission while an infrared and a delayed radio emission seem to be the signature of a periodic ejection phenomenon \\citep{Eike98}.  As the energy emitted from the jets is a synchrotron emission, the presence of a magnetic field has to be taken into account for the ejection.  A recent study done by \\citet{Serj98} has shown that for AGNs, a link exists between an emission in the optical range (believed to originate from the disk) and a radio-synchrotron emission (associated with the jet).  Summarizing, observational evidence is mounting that independent of the precise nature of the central accreting object (YSO to AGN), magnetized jets are propelled from surrounding disks.  \\subsection{Accretion-ejection models}  The most promising model for such `accretion-ejection' structures is based on a scenario where a large scale magnetic field threads an accretion disk. The presence of the magnetic field can be explained by the advection of interstellar magnetic field \\citep{Mous76} and/or by the local production of a magnetic field thanks to an effective disk dynamo \\citep{Reko00}. This model has been developed since the seminal work of \\cite{Blan82}, using a magnetohydrodynamic (MHD) approach. It was shown that the magnetic field can azimuthally brake the matter inside the disk (carrying off angular momentum allowing accretion) and accelerate matter above the disk surface. The collimation of the flow is achieved via magnetic tension due to the presence of a \\tor component of the magnetic field \\citep{Love76,Blan76,Heyv89,Saut02}. The magnetic field provides an effective alternative to the radially outward transport of disk angular momentum by viscosity. The interaction of the magnetic structure with the disk plasma can create a MHD Poynting flux leaving the disk along the magnetic surface \\citep{Ferr95}. This energy flux can then be converted into kinetic energy of the matter within the jet. Because the mass density in the jet is smaller than in the disk, it is thereby possible to reach high terminal velocities for a given amount of angular momentum removed from the disk. In this `accretion-ejection' model, the mass of the jet is fed from the accretion disk. This mechanism requires the disk to be in an equilibrium where the vertical thermal pressure gradient overcomes both the magnetic pinching of the disk (if one considers a bipolar topology of the magnetic field) and the gravitational compression. \\citet{Ferr95} have shown that the only stable equilibrium configuration meeting that requirement is an accretion disk where equipartition between thermal and magnetic pressure prevails, in order to avoid both a too strong magnetic compression and the magneto-rotational instability (e.g. see the review by \\citet{Balb98}). In this model, the outflow does not require an external mechanism to ensure a good collimation, since it is achieved by the magnetic field itself. Moreover, the presence of a central object only plays a role through its gravitational field alone and there is no specific interaction between its disk and its magnetospheric or its radiative environment presumed. Obviously, this scenario can be applicable to every system owning a magnetized accretion disk, such as the systems mentioned in section~\\ref{s:obs}.  \\noindent The first observation of a jet has been done in 1918 by Curtis for the optical jet seen around M87 \\citep{Curt18}. Since this observation, many other jets have been observed and monitored over several decades without drastic changes seen in their structures. Does this mean that a model to describe Magnetized Accretion-Ejection Structures (MAES) has to be completely stationary? The answer is not so easy to give. We know that the ejection has been present over several decades in most of the systems (except in X-ray binaries where the phenomenon seems to be transient or periodic). This time can be compared with the dynamical time of a MAES, as characterized by the rotation period of the matter at the inner radius of the MAES. This dynamical time is  \\be \\tau_{dyn} = \\frac{2\\pi}{\\Omega_K} = 2\\pi\\sqrt{\\frac{R^3_i}{GM_*}} \\label{intro1} \\ee \\noindent where $R_i$ is the inner radius of the MAES with a Keplerian rotation profile around a central object of mass $M_*$. In YSOs, where $M_*\\sim 1M_{\\odot}$ and $R_i\\sim 0.1 AU$, the dynamical time is of the order of ten days while in AGN, where $M_*\\sim 10^8 M_{\\odot}$ and $R_i\\sim 10 R_S$ ($R_S$ being the Schwarzschild radius), the dynamical time is typically two days. For the microquasar-type systems, the mass of the stellar black hole is typically $10M_{\\odot}$ and then the dynamical time becomes 2 milliseconds with the same $R_i$. Since these times are much shorter than the observed existence time of the associated jets, the models producing jets have to be close enough to a steady-state to yield permanent outflows over many dynamical time periods.   \\noindent The approaches developed so far to explain persistent jet launching can roughly be classified in two classes. One class uses a semi-analytical formulation assuming stationarity while the other class tries  to model at least a part of the structure using a time-dependent numerical MHD code. The former contains all stationary self-similar studies as well as the work by \\citet{Ogil01} which uses Taylor expansions of physical quantities  to study the dynamics of the launching region in detail.  Some self-similar studies model both the accretion disk and the super-Alfv\\'enic jet in a stationary framework  using a variable separation method with various levels of assumptions \\citep{Ward93,Ferr93a,Li95,Ferr95,Li96,Ferr97,Cass00a,Cass00b}. These studies require some dissipative mechanisms (ambipolar diffusion or magnetic resistivity) to occur inside the disk so that matter can cross the magnetic surfaces and achieves an accretion motion. Although these semi-analytical studies bring deep insight and reveal analytical relations that apply generally to  the accretion-ejection mechanism,  the assumption of self-similarity introduces geometric restrictions on the solutions.  The other class of studies groups all work done using 2.5D MHD time-dependent simulations (three dimensional assuming axisymmetry). Within this class, two categories of study can be distinguished: one which aims at a numerical calculation of the jet  alone \\citep{Usty95,Ouye97,Kras99} while the second one tries to model both the disk and the resulting outflow \\citep{Ushi85,Mats96,Kuwa00,Kato02}.  Stationary calculations of pure ideal MHD jets necessarily treat the disk as a boundary condition. This makes it difficult to determine whether  the prescribed quantities at the base of the jet are in agreement with conditions prevailing in a magnetized accretion disk.  Works that have tried to model  thick \\citep{Mats96,Kuwa00} or thin disks \\citep{Kato02} launching outflows have been done using either ideal MHD or uniformly resistive calculations throughout the computational domain.  While these simulations demonstrate outflows from the disk, the structure is very unstable and ejection only occurs during very few dynamical times. This is in conflict with the observed stability of jets mentioned above. The discrepancy in ideal MHD studies arises from modeling of the magnetized accretion disk subject to the frozen-in condition, so that the magnetic field is continuously advected towards the central object. This gives rise to a central magnetic field accumulation which will ultimately halt the accretion process itself. Moreover, in order to be stationary, it would assume an infinite magnetic reservoir feeding the process by some ad hoc outer boundary conditions. On the other hand, calculations assuming uniform resistivity throughout the computational domain have to justify the presence of such a dissipative phenomenom outside the accretion disk. Our simulations will require neither an infinite reservoir of magnetic field (if the magnetic structure reaches a steady-state) nor resistivity outside the accretion disk.  \\noindent The aim of the present paper is to be a first convergence point between the two classes of works mentioned previously. Indeed, in this paper we present 2.5D time-dependent MHD computations starting from an initial configuration close to a self-similar one. Most importantly, we aim to model continuous ideal MHD jet launching from a {\\em resistive} accretion disk threaded by a large scale magnetic field.  Treating the accretion disk magnetic structure in a resistive manner is the key point to achieving a robust ejection process. Instead of an unrealistic continuous magnetic reservoir, this model only requires  the presence of a primordial field. The magnetic field can attain a  quasi-stable configuration by achieving  a balance between inwards advection, outward diffusion and tensional retraction.  The organization of the paper is as follows: in the next section, we present the MHD formalism used in this work and also the initial conditions of our simulations. In the third section, we present the numerical code we have used, and comment on the computational grid and the boundary conditions. In the fourth section, we present our simulations and give a complete description of the accretion-ejection mechanism. In the last section we conclude and give an outlook to forthcoming work. ",
        "conclusions": "We present in this work the first MHD simulations of a magnetized, resistive accretion disk launching an ideal trans-Alfv\\'enic jet. The simulations are performed in a 2.5  dimensional, time-dependent and polytropic framework. Our MHD simulations involve a magnetic resistivity that is only triggered within the accretion disk. The shape of the resistivity coefficient is derived from \\citet{Shak73} since we express it as $\\eta=\\alpha_m V_A|_{Z=0} H\\exp(-2Z^2/H^2)$ where $V_A$ stands for Alfv\\'en velocity, $H$ disk scale-height and $\\alpha_m$ is a dimensionless parameter controlling the amplitude of the resistivity, smaller than unity). The temporal behaviour of our simulations display the launching of an outflow that, while propagating, becomes collimated. Note that the jet reaches super-fastmagnetosonic velocities within our computational domain.  The long term robustness of the structure is demonstrated by several means.  In particular, we check the angle between \\pol velocity and \\pol magnetic field (which is zero in a pure ideal MHD stationary configuration) and show that, after the outflow has been launched, this angle reaches maximum values in the jet typically smaller than three degrees. Moreover, the ejection rate (mass flux through the disk surface) of the disk, once initiated, remains near-constant over  several tens of the accretion disk dynamical timescale.  These diagnostics characterize a robust system that has reached a  quasi-stationary state.  We also present in this paper, by a detailed scrutiny of one snapshot of our simulation in the near-stationary phase, a complete illustration of the so-called `accretion-ejection' mechanism that is responsible for the jet launching. The key point of this model is that disk resistivity enables the structure to reach a near-static equilibrium where accreted matter inside the disk can pass through the \\pol magnetic surface. Such resistivity would mimic the effects of turbulence occuring inside the disk, provoked by MHD instabilities affecting equipartion accretion disks \\citep{apjl}. Moreover, we confirm by self-consistent numerical simulations that the two necessary conditions to achieve such a jet creation are (1) a vertical disk equilibrium which allows some disk matter streamlines to reach the disk surface (typically an equipartition disk with $B^2/P\\sim 1$, \\citet{Ferr95}) and (2) a turbulence configuration that obeys a simple relation Eq.~(\\ref{init6}) already presented in \\citet{Cass00a}. In our simulations, since we do not take viscosity into account, the angular momentum of disk matter is transferred by the action of a magnetic torque acting on the disk plasma. We highlight the fact that the sign reversal of the magnetic torque at the disk surface, obtained in self-similar analytic models if the turbulence follows the relation \\ref{init6}, is the core of the magneto-centrifugal effect responsible for jet acceleration. By displaying the radial force balance in the jet, we illustrate the role of the \\tor component of the magnetic field which collimates the outflow by means of magnetic tension (also called ``hoop'' stress, see \\citet{Heyv89}).  Forthcoming work will focus on two major points that were not taken into account here. Indeed, future work should be devoted to the implementation of a viscous torque inside the disk. This approach will validate the knowledge on complex angular momentum transport by both magnetic and viscous torques \\citep{Cass00a,Ogil01} combined with jet launching. The second point is the relevance of the energy equation in this kind of system.  Specifically, it has been demonstrated in a self-similar framework that the existence of a hot corona located at the disk surface is relevant to the astrophysical outflow \\citep{Cass00b,Garc01}. So the use of a full energy equation is desirable.  The simulations presented here consider a time-dependent resistivity (through the value of the Alfv\\'en speed in the disk). This could be of great interest for scenarios looking at time-dependent disk models launching jets, in particular for microquasars \\citep{Tagg99,Naya00}. Indeed, a suitable prescription of the transport coefficients could be able to produce structures where a periodic jet launching occurs."
    },
    "0208/astro-ph0208345_arXiv.txt": {
        "abstract": "{ We present a detailed empirical assessment of how the galaxy luminosity function and stellar luminosity density evolves over the last half of the universe's age ($0.2<z<1.2$) for galaxies of different spectral energy distributions (SED). The results are based on $\\sim 25000$ galaxies ($R\\la24$) with redshift measurements ($\\sigma_z\\approx 0.03$) and SEDs across $\\lambda_\\mathrm{obs}\\approx 350\\ldots 930$~nm. The redshifts and SEDs were derived from medium-band photometry in 17 filters, observed as part of the COMBO-17 survey (``Classifying Objects by Medium-Band Observations in 17 Filters'') over three disjoint fields with a total area of 0.78 square degrees. Luminosity functions (LF), binned in redshift and SED-type, are presented in the restframe passbands of the SDSS r-band, the Johnson B-band and a synthetic UV continuum band at 280~nm. \\\\ We find that the luminosity function depends strongly on SED-type at all redshifts covered. The shape of the LF, i.e. the faint-end power-law slope, does depend on SED type, but not on redshift. However, the redshift evolution of the characteristic luminosity $M^*$ and density $\\phi^*$ depends strongly on SED-type: (1) Early-type galaxies, defined as redder than a present-day reference Sa spectrum, become drastically more abundant towards low redshift, by a factor of 10 in the number density $\\phi^*$ from z=1.1 to now, and by a factor of 4 in their contribution to the co-moving r-band luminosity density, $j_\\mathrm{r}$. (2) Galaxies resembling present-day Sa- to Sbc-colours show a co-moving number density and contribution to $j_\\mathrm{r}$ that does not vary much with redshift. (3) Galaxies with blue spectra reflecting strong star formation decrease towards low redshift both in luminosity and density, and by a factor of 4 in their $j_\\mathrm{r}$ contribution. Summed over all SED types and galaxy luminosities, the comoving luminosity density decreases towards low redshift, between z=1.1 and now, by a small amount in restframe r and B, but by a factor of $\\sim6$ in restframe 280~nm. At $z=1.1$, galaxies redder than Sbc's, contribute 40\\% to the total $j_\\mathrm{r}$, which increases to 75\\% by z=0. For $\\lambda_\\mathrm{rest}=$280~nm, this increase is from 12\\% to 25\\% over the same redshift interval. \\\\ Comparison of the three independent sight-lines shows that our results are not significantly affected by large-scale structure. Our lowest redshift bin at $z= [0.2,0.4]$ largely agrees with the recent assessment of the present-day galaxy population by SDSS and 2dFGRS and deviates only by an excess of ``faint blue galaxies'' at $z\\sim 0.3$ compared to very local samples. Overall our findings provide a set of new and much more precise constraints to model the waning of overall star formation activity, the demise of star-bursts and the strong emergence of ``old'' galaxies, with hardly any young population, over the last 6-8 Gigayears. ",
        "introduction": "The formation and subsequent evolution of galaxies are determined both by the overall gravitational growth of structure and by the physics of gas cooling, star formation and feed-back which determine the successive conversion of gas into stars. Hierarchical structure formation within cold dark matter scenarios and their various extension to address star formation, now provide a comprehensive, but parameterized, framework for galaxy formation (e.g. Cole et al. 2000).  The onset of galaxy formation seems to take place at such high redshifts that it has so far escaped direct observation. Instead, most observations of galaxies have so far concentrated on obtaining large samples at lower, more easily accessible redshifts.  At the high redshift end, pioneering galaxy survey work has meanwhile reached redshifts $z\\sim 5$ (e.g. Ouchi etal. 2002). In the local universe, two large surveys are currently characterizing in detail the luminosity function of the galaxy population, based on large samples with $>10^5$ objects, the 2dF Galaxy Redshift Survey (2dFGRS, e.g. Madgwick et al. 2002) and the Sloan Digital Sky Survey (SDSS, e.g. Blanton et al. 2001).  During the past ten years several studies have aimed to map out the evolution of the luminosity function and the total luminosity density from the local universe to a redshift of $\\sim 1$ \\cite{Lilly95,Mad98,Lin99,Fri01}. But samples sizes well in excess of a few thousand objects are only becoming available now or in the near future, e.g. with the 17-colour survey COMBO-17 presented here and with the large spectroscopic campaigns DEEP \\cite{Koo01,Im02} and VIRMOS \\cite{LeF01}. The scientific inferences from existing, faint surveys out to $z \\ga 1$ have been limited mainly by their sample sizes, aggravated by the strong influence of large-scale structure when observing small co-moving volumes. The present survey, and other ongoing initiatives, aim at improving the measurement of the luminosity function by smoothing over structure and increasing the volume.  The COMBO-17 project (``Classifying Objects by Medium-Band Observations in 17 Filters'') was designed to provide a sample of $\\sim$50,000 galaxies and $\\la$1,000 quasars with rather precise photometric redshifts based on 17 colours. In practice, such a filter set provides a redshift accuracy of $\\sigma_\\mathrm{z,gal} \\approx 0.03$, $\\sigma_\\mathrm{z,QSO} \\la 0.1$, smoothing the true redshift distribution of the sample only slightly and allowing the derivation of luminosity functions.  The foremost data analysis goal of the COMBO-17 approach is to convert the photometric observations into a very-low-resolution {\\it spectrum} that allows simultaneously a reliable spectral classification of stars, galaxies of different types and QSOs as well as an accurate redshift (or SED) estimation for the latter two. The full survey catalogue should contain about 75,000 objects with classifications and redshifts on 1.5~$\\sq\\degr$ of area. This {\\it fuzzy spectroscopy} consciously compromises on redshift accuracy ($\\sigma_z\\approx 0.03$) in order to obtain very large samples of galaxies with a reasonable observational effort. While both characteristics are well suited for the analysis of an evolving population, they understandably do not permit dynamical or chemical studies which require quite detailed spectroscopic information.  While the photometric redshift technique has already been applied to galaxy samples about 40 years ago \\cite{Baum63,But83}, we have optimized the technique by increasing the number of filters and narrowing their bandwidth to obtain better spectral resolution and more spectral bins. Therefore, COMBO-17 also provides identifications and reasonably accurate redshifts for quasars (see also Koo 1999 for a nice overview on photometric redshifts and SubbaRao et al. 1996 on applying the technique in the context of luminosity functions).  The goal of the present paper is the use of COMBO-17 redshifts and SEDs for 25,000 galaxies over 0.78~$\\sq\\degr$ to draw up a detailed, empirical picture of how the population of galaxies evolved over the last half of the universe's age.  Our paper is organized as follows: in Sect.~2 we present the observations that have led to the current sample of $\\sim 25$,000 galaxies. Our techniques for obtaining their redshifts, SED classification, luminosities and completeness are described in Sect.~3. The resulting sample properties are discussed in Sect.~4. In Sect.~5 we derive our ``quasi-local'' luminosity function, drawn from the redshift interval of $z=[0.2,0.4]$ and compare it with the results from more local samples obtained by the 2dF Galaxy Redshift Survey and the Sloan Digital Sky Survey. Finally, we show the evolution of the luminosity function and the luminosity density out to $z<1.2$. ",
        "conclusions": "Based on photometry in 17 (mostly medium-band) filters obtained by the COMBO-17 survey on three independent fields of 0.26~$\\sq\\degr$ each, we have derived redshifts and SED classifications for $\\sim 25,000$ galaxies to $R\\la 24$. The redshifts may be viewed either as high-precision photometric redshift estimates or very low resolution (R$\\sim12$) spectral redshifts. They have a precision of $\\sigma_z\\approx 0.03$ and lie mostly in the range $0.2<z<1.2$ with a median redshift of 0.55. The upper limit of galaxy redshifts considered here is set by the availability of spectroscopic cross-checks confirming their precision and accuracy. Our results are unaffected by random redshift errors of $\\sim0.03$, and are robust even for somewhat larger, systematic errors.  Compared to previously published large-sample surveys with better redshift accuracy, COMBO-17's flux limit is nearly two magnitudes fainter and the number of galaxies is more than an order of magnitude larger. Taken together, the properties of the survey allow us to draw up a comprehensive picture of how the luminosity function and the SEDs of galaxies have evolved over the last half of cosmic evolution. The survey is deep enough that the bulk of the stellar light over this redshift range is directly observed. As far as possible, we complement our data by ``local'' information from low-redshift surveys (e.g. 2dFGRS and SDSS), as our survey volume is too small for $z<0.2$.  The observed galaxy SEDs cover a wavelength range of $\\lambda_\\mathrm{obs} = 350 \\ldots 930$~nm and allows us to determine restframe properties of the population at 280~nm and at $B$ with hardly any extrapolation in the range $0.2<z<1.2$. We also derived restframe properties in the SDSS r-band, by applying a type-dependent extrapolation of the observed SED at $z\\ga0.5$.  The goal of this present paper is to present an empirical picture of the evolution of the galaxy population as a function of luminosity, SED and redshift, as a solid constraint for any models of statistical galaxy evolution. Direct comparison with cosmolgical models will be reserved for future papers. Our main findings are as follows: \\begin{enumerate}  \\item Survey areas approaching an area of 1~$\\sq\\degr$ are necessary to reduce field-to-field variations at L$_*$ below a factor of two in redshift bins of $\\Delta z \\sim 0.2$.  \\item The faint end slope of the luminosity function depends quite strongly on SED-type at all observed redshifts. However, within a given (non-evolving) SED type there is no evidence for evolution of the luminosity function shape with redshift (to $z\\sim1$).  \\item A comparison of our ``quasi-local'' sample at $z=[0.2,0.4]$ with 2dFGRS and SDSS shows consistent results around the knee of the luminosity function, but a steeper faint end in COMBO-17, probably reflecting the evolution of faint blue galaxies between the median redshifts $\\sim0.05$ and $\\sim0.34$.  \\item If we define SED type via a fixed, present-day spectral sequence for galaxies, from ellipticals to  young starbursts, then the evolution of the luminosity function from $z\\sim 1.2$ to now depends strongly on SED type, both in its density normalization, $\\phi^*$, and in its characteristic luminosity, $M^*$:  (a) The faint end slope is $\\alpha = 0.50\\pm 0.20$ for early type galaxies (SED \\tp 1), steepening to $\\alpha = 1.50\\pm 0.06$ for galaxies with the bluest starburst colours (SED \\tp 4).  (b) For early type galaxies (defined by z=0 colours bluer than Sa galaxies) $\\phi^*$ increases by more than an order of magnitude from $z=1.2$ to now, while M$_*$ gets fainter by more than a magnitude.  (c) For the latest type galaxies $\\phi^*$ remains roughly constant over the redshift range probed, while M$_*$ gets fainter by almost two magnitudes.  \\item When averaging over all SED types, the evolution of the luminosity density depends considerably on the rest-wavelength considered:  (a) In the restframe B- and r-bands, the integrated luminosity density remains constant from $z=1.1$ to $z=0.5$, dropping subsequently to the present epoch by perhaps 30\\%.  (b) In the restframe near-UV (280~nm) the integrated luminosity density drops by a factor of six from $z\\sim1$ to now, where it appears that much of that drop occurs at redshifts below $z\\sim 0.6$. Our more accurate estimates of the near-UV luminosity density imply a considerably shallower evolution than indicated by the data of Lilly et al. (1995), but is consistent within their 1.5$\\sigma$ confidence limits.  \\item The ratio of the stellar luminosity density contributed by early and late SED type galaxies at different wavelengths changes quite drastically with redshift: At redshift $z\\sim 1$ strongly star-forming galaxies (\\tp 3 and 4) make up the bulk of the radiation in all three restframe bands with a contribution near 90\\% at 280~nm. At the present epoch, in particular the extreme starbursts (\\tp 4) play only a significant role at 280~nm, providing about half of the luminosity there. At the same time, the fraction of light from galaxies with early-type spectra (\\tp 1) have increased from below 10\\% to 40\\% in the B- and r-bands over the same period.  \\end{enumerate}  We reserve any rigorous modeling of these results for a later paper, as it appears that the data warrant a cosmologically motivated interpretation, such as a comparison with semi-analytic or N-Body/SPH models rather than any empirical speculation about ``luminosity evolution\" or ``density evolution\" of various sub-populations. Nonetheless, it seems approriate to make a few qualitative remarks in closing: our results confirm earlier findings that the epoch of galaxy formation, or star formation in galaxies, has been waning over the second half of the present cosmic age, leading to a strong decrease of star-burst galaxies (both luminous and faint) accompanied by the strong emergence of ``old\" galaxies, whose overall SED is redder than an Sa template spectrum. The COMBO-17 data set permits to study this evolution differentially for different SED types and still provides much smaller error bars, due to the enlarged survey area and sample increase. Compared to earlier work, the decline of cosmic star formation activity appears gentler, i.e. the near-UV luminosity density as a proxy for the density of young stars, is found to drop only by a factor of five from $z=1.1$ to now.  Beyond questions of interpretation, there are a number of observational issues that still need to be addressed. Calibration of our 17-band redshifts through spectra for samples of many hundred objects is forthcoming in the course of the ESO-GOODS key project, and will increase the reliability of our estimates even further. The addition of several other fields will reduce the impact of the field-to-field variation, now still a limiting factor in several aspects of the analysis. Finally, the full analysis of the SEDs and the addition of near-IR data will allow us to extend the analysis to include direct observational estimates of the stellar mass."
    },
    "0208/astro-ph0208109_arXiv.txt": {
        "abstract": "{\\small We present the results of a detailed analysis of RXTE observations which show an unusual state transition between high-soft and low-soft states in the microquasar GRS~1915+105. The RXTE pointed observations reveal that these events appeared as a series of quasi-regular dips when hard X-ray and radio flux were very low. The X-ray light curve and colour-colour diagram of the source during these observations are found to be different from any reported so far. The X-ray spectral and timing properties of the source during these dips are distinctly different from those seen during the other various dips and low intensity states seen in this source. There is, however, a remarkable similarity in the properties during the dip and non-dip regions in these set of observations. This indicates that the basic accretion disk structure is similar, but differ only in intensity. We invoke a model, to explain these observations, in which the viscosity is very close to a critical viscosity and the shock wave is weak or absent.} ",
        "introduction": "In spite of the bewildering types of variability classes seen in GRS~1915$+$105, the spectral and temporal characteristics at any given time can broadly be classified into three distinct spectral states: a low-hard state with invisible inner accretion disk (C), a high-soft state with visible inner accretion disk (B), and a low-soft state with spectrum similar to the high-soft state and with much lower intensity (A). All the variability classes are understood as transition among these three basic states.  In GRS~1915$+$105, the low-soft state appears briefly during the rapid state transition between states C and B as well as during the soft dips seen during the variability classes $\\beta$ and $\\theta$ (Belloni et al. 2000). On rare occasions, long stretches of state A are also seen in this source (the variability class $\\phi$). Though transition from low-hard to high-soft states are seen in many Galactic black hole candidate sources, a transition between two different intensity states with similar physical parameters of the accretion disk was not observed in GRS~1915$+$105 or in any other black hole binaries. We have tried to explain the observed peculiar state transition which is seen for the first time in GRS~1915$+$105 on the basis of the presence of an accretion disk with critical viscosity which causes appearance and disappearance of sub-Keplerian flows out of Keplerian matter.   \\vspace{-0.5cm} ",
        "conclusions": "We now discuss a possible scenario to understand this new and exciting behavior in terms of TCAF model of Chakrabarti \\& Titarchuk (1995). A disk with completely free (Keplerian) disk and (sub-Keplerian) halo accretion rates, does exhibit multiplicity in spectral index or multiplicity in disk accretion rate when the spectral index is similar. In the observation described here, there is no evidence for a large  variation of the spectral index. This signifies that though there is not enough time for a change in the total rate, individually, Keplerian and sub-Keplerian rates may have been modified. This is possible if the viscosity parameter is very close to the critical value ($\\alpha \\sim \\alpha_c \\sim 0.015$; Chakrabarti 1996). When $\\alpha$ of the entire disk is well above $\\alpha_c$, the entire disk is pretty much \\end{minipage} \\hfil\\hspace{\\fill} \\begin{minipage}{60mm} \\includegraphics[width=51mm]{naiks_2.ps} {\\small Fig.~2.~The observed count rate spectrum of GRS~1915+105 during non-dip and dip regions of the new class obtained from RXTE/PCA and HEXTE are shown in upper and lower panels respectively. A best-fit model consisting of a disk blackbody, a power-law, and a thermal Compton spectrum is shown as histogram with the data.} \\end{minipage} Keplerian, except very close to the black hole ($r<3r_g$). Similarly when $\\alpha$ of the entire flow is well below $\\alpha_c$, the flow is sub-Keplerian with a possible standing or oscillating shock wave. Since viscosity in a disk can change in a very small time scale (convective/turbulent time-scale in the vertical direction) it is not unlikely that the viscosity near the Keplerian-disk surface is very close to the critical value during the time when this new class is exhibited. Temperature and intensity of the radiation from the Keplerian disk drops to the point that a thin outflow develops from the sub-Keplerian flow. This is the low intensity state. This wind is cooled down and is fallen back on the Keplerian disk, increasing the Keplerian rate, intensity and viscosity, thereby cutting off the wind and bringing the flow to the high intensity state again. As the disk is cooler, no shock oscillation, and therefore no QPO is present. One of the tests that our model is perhaps correct is that because of super-critical viscosity this new class ($\\omega$) should be in association with other soft classes ($\\delta, \\ \\phi $ and $\\gamma$) as is indeed the case. Only when the net accretion rate or viscosity goes down, the spectrum may become partly harder and $\\theta$ and $\\beta$ states could occur. If viscosity goes up, then the state goes to $\\delta$, $\\gamma$ or $\\phi$ and the disk is basically Keplerian in these cases.  \\vspace{-0.4cm}"
    },
    "0208/astro-ph0208423_arXiv.txt": {
        "abstract": "Using a large quantity of Rossi X-ray Timing Explorer  data presented in the literature I  offer a detailed investigation into the  accuracy of quasiperiodic oscillations (QPO) frequency determination. The QPO phenomenon seen in X-ray binaries is possibly a result of the resonance of the  intrinsic (eigen) oscillations and harmonic driving forces of the system. I show that the resonances, in the presence of the  damping of oscillations, occur at the frequencies which are systematically and randomly shifted with respect to the eigenfrequencies of the system. The shift value strongly depends on the damping rate which is measured by the halfwidth of the QPO feature. Taking into account this effect I analyze the QPO data for four  Z-sources: Sco X-1, GX 340+0, GX 5-1, GX 17+2  and two atoll sources: 4U 1728-34, 4U 0614+09. The transition layer model (TLM)  predicts the existence of the invariant quantity: $\\delta$, an inclination angle of the magnetospheric axis with respect to the normal to the disk. I calculate $\\delta$ and  the error bars of $\\delta$ using the resonance shift and I find that the inferred  $\\delta-$values are consistent with  constants for these four  Z-sources, where horizontal branch oscillation and  kilohertz frequencies have been detected and correctly  identified. It is shown that the inferred $\\delta$ are in the range between 5.5 and 6.5 degrees. I conclude that the TLM seems to be compatible with data. ",
        "introduction": "Kilohertz quasi-periodic oscillations (QPOs) have been discovered by the Rossi X-ray Timing Explorer (RXTE) in a number of low mass X-ray binaries (Strohmayer et al.  1996, van der Klis et al. 1996). The presence of two observed peaks with frequencies $\\nu_1$ and $\\nu_2$ in the upper part of the power spectrum became a natural starting point in modeling the phenomena.  Attempts have been made to relate $\\nu_1$ and $\\nu_2$ and the peak separation $\\Delta \\nu= \\nu_2-\\nu_1$ with the neutron star (NS) spin. In the sonic point beat frequency (SPBF) model  by Miller, Lamb \\& Psaltis (1998) the kHz peak separation $\\Delta \\nu$ is considered to be close to the NS spin frequency and thus $\\Delta \\nu$ is predicted to be constant [see also the updated version of the SPBF (Lamb \\& Miller 2001) in which $\\Delta \\nu$ is allowed to be variable]. Observations of kHz QPOs in a number of binaries (Sco X-1, 4U 1728-34, 4U 1608-52, 4U 1702-429 and etc) show that the peak separation decreases systematically when kHz frequencies increases (see a review by van der Klis 2000).  %  There are two other  models in the literature which infer the relations between $\\nu_1$, $\\nu_2$ and $\\nu_{HBO}$. The relativistic precession (RP) model  involves  high speed particle motion in strong gravitational fields, leading to oscillations of the particle orbits. Stella et al. (1999) studied precession of the particle orbit under influence of a strong gravity due to General Relativity (GR) effects. The correlation between high frequency (lower kHz frequency) and low frequency (broad noise component) QPOs previously found by Psaltis, Belloni \\& van der Klis (1999) for black hole (BH) and neutron star (NS) systems has been recently extended over two orders of magnitude by Mauche (2002) to white dwarf (WD) binaries. With the assumption that the same mechanism produces the QPO in WD, NS and BH binaries, Mauche argues that the data exclude RP, SPBF models (as well as any model requiring either the presence or absence of a stellar surface or a strong magnetic field).  The transition layer model (TLM) was introduced by Titarchuk, Lapidus \\& Muslimov (1998), hereafter TLM98, to explain the dynamical adjustment of a Keplerian disk to the innermost sub-Keplerian boundary conditions (e.g. at the NS surface). The low branch frequency $\\nu_L$, the Kepler frequency $\\nu_{\\rm K}$ and the hybrid frequency $\\nu_h$, as they are introduced by OT99 are eigenfrequencies of the oscillator.  In the TLM framework  the observed horizontal branch oscillation (HBO) frequencies $\\nu_{HBO}$, $\\nu_1$ and $\\nu_2$ can be interpreted  as the {\\it resonance} frequencies near $\\nu_{L}$, $\\nu_{\\rm K}$ and $\\nu_{h}$  which are broadened  as a result of the (radiative) damping in the oscillator (see TLM98, Eq. 15). Furthermore, the resonance frequencies $\\nu_{HBO}$, $\\nu_1$, $\\nu_2$ are shifted with respect to the eigenfrequencies $\\nu_L$, $\\nu_{\\rm K}$, $\\nu_h$.  The frequency shift and random errors of the eigenfrequencies depend on the damping rate of oscillations $\\lambda$ (see details in \\S 3).  {\\it One should keep in mind that the systematic and random errors in the centroid frequency determination due to this resonance shift can be a factor of a few larger than the statistical error in the determination of the centroid frequency.} It is worth noting that an interpretation of the observed QPO phenomena in the framework of any oscillatory (or wave propagation) model requires the aforementioned corrections due to resonance in the presence of damping. For the observed Q-values ($Q=\\nu/2\\lambda)$ of a few, the damped intrinsic  harmonic oscillations decay very quickly (less than $10^{-2}$ s) and thus only the forced oscillations can exist for time periods of order  $10^{2-3}$ seconds and longer (for details  see \\S 3.1).  There remains the question of what kind of the observational arguments can support  the resonance mechanism  for QPO phenomena. I address to this issue in \\S 4. The goal of this {\\it Letter} is to demonstrate how the resonance shift  corrections are important for  the accurate interpretation of the QPO data.  The {\\it Letter} reports the results of the QPO data interpretation from four Z-sources: Sco X-1, GX 340+0, GX 5-1, GX 17+2  and two atoll sources: 4U 1728-34, 4U 0614+09 and two atoll sources,  collected by RXTE during 6 years of observations.  In \\S 2, I present the origins of the  RXTE QPO data  used in the present investigation.   In \\S 3 I  provide details of the resonance effect on the eigenfrequency restoration using the observed QPO frequencies and in \\S 4 I offer comparisons of the predictions of the TLM model with the RXTE observations. Summary and conclusions are drawn in \\S 5. ",
        "conclusions": "In this {\\it Letter}, I present the  consequences and predictions of the TLM as an oscillatory model for the QPO observations. It is fair to ask what kind of  arguments one can put forth for the resonance effect in QPO phenomena in general.  I argue that they  are (i) the Lorentzian shape of the QPO feature (e.g. van der Klis 2000), (ii) $1/\\omega$ resonance dependence of the rms amplitude for lower kHz QPO found by Titarchuk, Cui \\& Wood (2002) using Jonker's et al. (2000b) data; (iii) observational evidence for 2:3 resonance in the ``kHz'' QPO sources (Abramovicz \\& Klu\\`zniak 2001; Abramowicz et al. 2002), (iv) harmonics and sidebands of low kHz QPOs (Jonker et al. 2000b). The whole issue of the observational evidence for the resonance effect in the QPO phenomena is beyond the scope of this short {\\it Letter}.  Using the previously cited thorough analysis of  RXTE QPO data for Sco X-1, Sco X-1, GX 340+0, GX 5-1, GX 17+2  and two atoll sources: 4U 1728-34, 4U 0614+09 I a detailed investigation of the accuracy of the QPO frequency determination in the framework  of a weakly  nonlinear oscillator is offered. The TLM  as an example of a weakly nonlinear oscillator  is used to  extract the model parameters and to test the model.    The TLM predicts the existence the  invariance of the $\\delta$. It is established that: (1) The errors of the eigenfrequency extraction are significantly affected  by the errors of the QPO halfwidth. (2)  The inferred $\\delta-$values are consistent with constants for the four  Z-sources and 4U 0614+19 where kHz and HBO frequencies have been detected and  correctly identified. I also put forth  arguments to explain the QPO phenomena as a result of the resonance effect in anharmonic oscillatory systems.   L.T. acknowledges  fruitful discussions with Vladimir Krasnopolsky, Chris Shrader, Paul Ray and with referee. In particular, I am grateful to Charlie Bradshaw, Michiel van der Klis, Peter Jonker and Mariano Mendez for the data  which enable me to make detailed comparisons of the model predictions to the observations."
    },
    "0208/astro-ph0208565_arXiv.txt": {
        "abstract": "{ The stability of rotating isotropic spherical stellar systems is investigated by using N-body simulations. Four spherical models with realistic density profiles are studied: one of them fits the luminosity profile of globular clusters, while the remaining three models provide good approximations to the surface brightness of elliptical galaxies. The phase-space distribution function $f(E)$ of each one of these non-rotating models satisfies the sufficient condition for stability $df/dE<0$.  Different amounts of rotation are introduced in these models by changing the sign of the $z$-component of the angular momentum for a given fraction of the particles. Numerical simulations show that all these rotating models are stable to both radial and non-radial perturbations, irrespective of their degree of rotation. These results suggest that rotating isotropic spherical models with realistic density profiles might generally be stable. Furthermore, they show that spherical stellar systems can rotate very rapidly without becoming oblate. ",
        "introduction": "Dynamical instabilities in spherically symmetric stellar systems have been investigated for more than four decades. In a seminal work, Antonov (\\cite{antonov60,antonov62}) used a variational principle to demonstrate that non-rotating spherical models with a phase-space distribution function $f$ depending only on the energy $E$ are stable to non-radial perturbations if $df/dE<0$. Subsequent works showed that this condition is also a sufficient condition for stability to radial perturbations (Dor\\'emus et al. \\cite{doremus71}; Sygnet et al. \\cite{sygnet84}; Kandrup \\& Sygnet \\cite{kandrup85}). In general, non-rotating spherical stellar systems are described by distribution functions $f$ that depend on both the energy $E$ and the magnitude of the angular momentum $L$. In such systems only the stability to radial modes can be tested by using the sufficient condition $\\partial f/\\partial E<0$ (Dor\\'emus \\& Feix \\cite{doremus73}; Dejonghe \\& Merritt \\cite{dejonghe88}). For this reason, numerical simulations have been an indispensable tool to investigate the stability of anisotropic spherical models.  Several classes of instabilities have been discovered in non-rotating spherical models with anisotropic velocity distributions (e.g., H\\'enon \\cite{henon73}; Merritt \\& Aguilar \\cite{merritt85a}; Barnes et al. \\cite{barnes86}; see Merritt \\cite{merritt99} for a recent review). For example, models dominated by stars on radial or eccentric orbits can be unstable to forming a triaxial bar (Polyachenko \\cite{polyachenko81}; Merritt \\& Aguilar \\cite{merritt85a}; Meza \\& Zamorano \\cite{meza97}), while models composed mainly of stars on circular orbits can exhibit quadrupole-mode oscillations (Barnes et al. \\cite{barnes86}).  In contrast to non-rotating spherical models, comparatively little work has been done to investigate the stability of rotating spherical stellar systems. Miller \\& Smith (\\cite{miller80}) employed several methods to introduce rotation in a spherical $n=3$ polytrope with isotropic velocity distribution. In all cases, they found that the addition of rotation does not affect the stability of their models; indeed, they found that even rapidly rotating systems remain spherically symmetric. More recently, Alimi et al. (\\cite{alimi99}) showed that rotating isotropic spherical $n=4$ polytropes, which were made to rotate by changing the sign of the $z$-component of the angular momentum for a fraction of the particles, were dynamically stable.  On the other hand, Allen et al. (\\cite{allen92}) reported the existence of a ``tumbling bar instability'' in a set of rotating spherical models with different degrees of velocity anisotropy, ranging from completely circular to entirely radial (see also Papaloizou et al. \\cite{papaloizou91}; Palmer \\cite{palmer94a,palmer94b}). In particular, they suggested that the introduction of a small amount of rotation in the isotropic spherical $n=2$ polytrope induces a bar instability in this otherwise stable system. However, Sellwood \\& Valluri (\\cite{sellwood97}) using the same files of initial conditions but a different N-body code, showed that these models are actually stable; the instability appears do not exist. They suggested that the evolution observed by Allen et al. (\\cite{allen92}) was, probably, caused by an improper treatment of variable time steps in their N-body code.  Most of the previous results have been obtained for rotating spherical models with somewhat unrealistic properties. Indeed, most of these works have employed spherical polytropes with finite radius (e.g., Alimi et al. \\cite{alimi99}) or models with unrealistic velocity distributions (e.g., Allen et al. \\cite{allen92}). Therefore, they cannot be considered as general. To investigate the influence of rotation on the stability of spherical stellar systems, it is necessary to consider models with more realistic density profiles and velocity distributions. In this paper, the results of a series of N-body simulations for four of such models are presented. These isotropic spherical models are: the Plummer (\\cite{plummer11}) model, the Hernquist (\\cite{hernquist90}) model, the Jaffe (\\cite{jaffe83}) model, and the $\\gamma=0$ model (Dehnen \\cite{dehnen93}; Tremaine et al. \\cite{tremaine94}). Different degrees of rotation were introduced in these models by using the so-called Lynden-Bell's (\\cite{lynden-bell60,lynden-bell62}) demon to reverse the sense of rotation along the $z$-axis of a given fraction of the particles. Numerical simulations show that all these rotating models are stable, regardless of their degree of rotation.  This paper is organized as follows. The method employed to introduce net rotation in these models and the N-body code used to follow their dynamical evolution are described in Sect.~2. The main results of the numerical simulations are summarized in Sect.~3. Finally, a brief discussion of the results is given in Sect.~4. ",
        "conclusions": "The stability of four rotating isotropic spherical models was investigated by using N-body simulations. The density profiles of these models provide good approximations to the mass distribution of globular clusters and elliptical galaxies. Different degrees of rotation were introduced in these models by reversing the sense of rotation along the $z$-axis of a given fraction of the particles (Lynden-Bell \\cite{lynden-bell62}). Simulations show that all these rotating models are dynamically stable, irrespective of their degree of rotation. No signs of radial, lopsided, or bar instabilities were observed. If some of them exist their growth rate is larger than $50t_h$, the time elapsed for these simulations.  These simulations show that spherical stellar systems can rotate very rapidly without becoming oblate. This result contrasts with the suggestion of Alimi et al. (\\cite{alimi99}) that there do not exist spherical stellar systems in fast rotation (i.e., with $\\mu\\ga 0.1$). However, their conclusions were based on a series of simulations for spherical $n=4$ polytropes with isotropic velocity distributions, which were made to rotate by using a procedure that modifies their initial velocity anisotropy. Therefore, their conclusions are not strictly applicable to the isotropic models studied in this paper. But, they could still be valid for anisotropic spherical models. In fact, there are several other ways to construct rotating spherical models. For example, one could construct a system containing only circular orbits and then reverse a half of them (see Lynden-Bell \\cite{lynden-bell60}). Clearly, such a model would exhibit stronger rotation that the models here analyzed and surely might be unstable."
    },
    "0208/astro-ph0208086_arXiv.txt": {
        "abstract": "{\\small In the hard X-ray spectra of some X-ray binaries line features around 500~keV are detected. We interpret these as arising from pair annihilation in relativistic outflows leading to a significant Doppler shift of the frequencies of the lines. We show that a small fraction of pairs escaping the annihilation region may give rise to the radio synchrotron emission observed in Nova Muscae 1991.} ",
        "introduction": "The composition of the jets observed in Galactic X-ray transients, either electron-positron or electron-proton plasma, is still not fully established.  Here we review the arguments presented in \\cite{Kaiser} where we proposed that the annihilation line features observed in the hard X-ray spectra of some X-ray transients arise from pairs in a bipolar outflow. At the time of annihilation, this outflow is already accelerated to relativistic bulk speeds causing a significant Doppler shift of the frequency of the annihilation lines. We also showed that the subsequent emission of radio synchrotron radiation from the outflow may be caused by only a small fraction of pairs escaping from the annihilation region. We apply this idea to radio and hard X-ray observations of Nova Muscae 1991. The energy requirements for this source rule out a large contribution of protons to the outflow. ",
        "conclusions": ""
    },
    "0208/astro-ph0208279_arXiv.txt": {
        "abstract": "The morphology of the $\\epsilon$ Eridani dust ring is reproduced by a numerical simulation of dust particles captured into the 5:3 and 3:2 exterior mean-motion resonances with a 0.3 eccentricity $10^{-4}$ solar mass planet at periastron at a semi-major axis of 40 AU. The morphology will differ when the planet is at apastron, in about 140 years.  Moderate eccentricity planets in outer extra-solar systems will cause observable variations in the morphology of associated dusty rings. ",
        "introduction": "The Infrared Astronomical Satellite (IRAS) measured far infrared fluxes from the vicinity of main sequence stars \\citep{aumann}, including the nearby systems Vega and Formalhaut which are of moderate age, $10^8$ years, at the end of the era when rocky planets are expected to form.  Recently, submillimeter wavelength images of stars with infrared excesses have revealed structure in some of these disks. Formalhaut has a dust ring with a central cavity \\citep{holland}, Vega, $\\beta$ Pictorus and HR 4796A have secondary dust emission peaks and central cavities \\citep{wilner,holland, jaya,koerner}, HR 4796A has a ring evident in scattered light from the central star \\citep{schneider}, and $\\epsilon$ Eridani has a 60 AU dusty ring with peculiar 4 peaks of emission \\citep{greaves}. HR 4796A and $\\beta$ Pictorus are younger, at ages of only $\\sim 10^7$ years, whereas $\\epsilon$ Eridani (HR 1084) is a nearby 0.5-1 Gyr old K2V ($0.8M_\\odot$) star \\citep{soderblom,song}.  Because of the short lifetime of the dust particles in these disks, and the scarcity of gas, the dust must continually be replenished from collisions of larger orbiting bodies, possibly analogous to Kuiper belt objects.  Hence the disks in the older systems are commonly called debris disks. Dust particles in our solar system spiral toward the sun due to radiation drag forces, known as Poynting-Robertson (P-R) drag and interactions with the solar wind \\citep{burns}.  Gravitational perturbations from planets can scatter the particles or capture them into orbital resonances with planets (e.g., \\citealt{dermott}). Since the lifetime of captured particles can greatly exceed the lifetime of particles which are not captured, resonant trapping can sculpt the dusty disk.  Structure predicted in the dust distribution in the Kuiper Belt is primarily a result of resonance trapping \\citep{liou99,amaya}. \\citet{wilner} suggested that the two peaks seen at 1.3mm in the Vega disk could be explained by the capture of dust particles into the 2:1 exterior mean-motion resonance with a high eccentricity three Jupiter mass giant planet at a semi-major axis of 40AU. \\citet{ozernoy} proposed that the morphology of the submillimeter images of the $\\epsilon$ Eridani system could be caused by dust particles trapped in high libration orbits associated with the 3:2 mean motion resonances with a $\\sim 0.2$ Jupiter mass planet in a circular orbit. While \\citet{ozernoy} considered the morphology caused by dust particles carefully placed in individual resonances, here we initially distribute dust particles over a range of semi-major axes and eccentricities and search for stable dust configurations which can account for the four peaks observed in the submillimeter images of the $\\epsilon$ Eridani system. ",
        "conclusions": "We have found that the morphology of dusty material trapped in exterior resonances with a planet can be strongly dependent upon the eccentricity and orbital phase of the planet. The morphology of the $\\epsilon$ Eridani dust ring seems to be reproduced by dust particles captured into the 5:3 and 3:2 exterior resonances with a moderate eccentricity, $e_p \\sim 0.3$, planet near periastron. When the planet was more massive ($\\mu > 5 \\times 10^{-4}$) or more highly eccentric (greater than 0.4), the resonances closest to the planet did not capture and hold particles as efficiently.  When the planet mass was below a Saturn mass, the 4:3 resonance also contributed to the particle distribution resulting in dominant symmetric four peaks.  The asymmetries observed in the $\\epsilon$ Eridani image were not observed in the dust distribution so we consider planets less massive than Saturn unlikely to explain the particle distribution.  When the planet eccentricity was below 0.15, the azimuthal density variations in the dust distribution were too low to account for the morphology in the $\\epsilon$ Eridani disk.  Our model differs from that proposed by \\citet{ozernoy} in a number of ways.  The period of the orbital planet in our model is 280 years which should cause the pattern to revolve by about the star by $\\sim 1.3^\\circ {\\rm yr}^{-1}$, faster than the $\\sim 0.7^\\circ {\\rm yr}^{-1}$ estimated by \\citet{ozernoy}, where the model planet semi-major axis is $\\sim 60$AU rather than at $\\sim 40$AU. Our model planet is located to the north of the star, rather than to the west of the star.  Our model planet mass is similar to theirs but at a moderate eccentricity. Our model dust concentrations are a result of segregation in the phase of a resonant angle, rather than caused by a large libration amplitude.  Furthermore, because of the eccentricity of the planet, our model predicts that the morphology of the dusty ring will vary, as well as revolve as the planet orbits about the star.  The initial conditions of our simulations cause many of the particles to begin trapped in the resonances and at fairly low eccentricity. It is possible that low eccentricity planetesimals exist in the system and that they are the source of the dust particles that we see.   Alternatively the planetesimals in the system are further out and the dust particles become trapped in the resonances as they spiral in toward the planet. Further simulations would be required to differentiate between these possibilities. In either case, resonant capture into these resonances is less likely and less prolonged when the planet mass or eccentricity is high. If the planet mass is too low then it cannot be responsible for clearing a gap or central region in the dust distribution.  Additional and more massive planets would be required to do this.  It is possible that high eccentricity planets are common in the outskirts of extra-solar systems.  If so then the resulting dust distributions would not only revolve \\citep{ozernoy}, but will also be dependent upon the orbital phase of the planets. This is an exciting prospect because there would be variations in the dust morphology on observable timescales.  \\begin{figure*} \\vspace{16.0cm} \\special{psfile=\"h3_y.ps\" angle=-90 vscale=44 hscale=44 hoffset=-20 voffset=430} \\special{psfile=\"h4_y.ps\" angle=-90 vscale=44 hscale=44 hoffset=200 voffset=430} \\special{psfile=\"h3_s.ps\" angle=-90 vscale=46 hscale=46 hoffset= 35 voffset=290} \\special{psfile=\"afig.ps\" angle=0 vscale=56 hscale=56 hoffset=200 voffset=-180} \\caption[]{ a) The dust distribution for a planet at periastron with ratio of planet mass to stellar mass $\\mu=10^{-4}$, and eccentricity of $e_p=0.3$. The dust particles have $\\beta=0.1$. The star and planet are denoted as black dots at the origin and at a radius of 0.7 from the origin, respectively. Axes are given in units of the planet semi-major axis. b) Same as a) but the planet is at apastron. c) Semi-major axis distribution for the particle distribution shown in a. d) Particle distributions are shown for limited ranges of semi-major axes for the simulations displayed in Figures 1a,b.  Particles in the 3:2 resonance are shown for the planet at periastron (top left) and at apastron (bottom left).   Particles in the 5:3 resonance are shown for the planet at periastron (top right) and at apastron (bottom right). } \\end{figure*}    \\begin{figure*} \\vspace{10.0cm} \\caption[]{ On the right is shown the simulated intensity distribution  based on the simulation shown in Figure 1a, for a 0.3 eccentricity planet at periastron of planet to stellar mass ratio $10^{-4}$. For comparison we show on the left the $850\\mu$m submillimeter emission map of the $\\epsilon$ Eridani system. (this is Figure 1 by \\citealt{greaves}). The location of the star is denoted in both panels as a star. For the simulated intensity distribution we have chosen a disk inclination of $40^\\circ$ where $0^\\circ$ is face on. The dust distribution has been weighted by $r^{-2}$ to simulate the intensity field from the star and then smoothed to the approximate resolution of the submillimeter images. We estimate that the planet is currently located to the north of the star (denoted as a filled circle in the right hand panel) at a declination of about $-9^d27^m20^s$ (epoch 1998 as from Figure 1 of \\citealt{greaves}). Axes for the simulation are given in units of the semi-major axis of the planet. } \\end{figure*}"
    },
    "0208/astro-ph0208092_arXiv.txt": {
        "abstract": "{\\small We have performed a first fully 3-D GRMHD simulation with Schwarzschild black hole with a free falling corona. The initial simulation results show that a jet is created as in the previous simulations using the axisymmetric geometry with the mirror symmetry at the equator. However, the time to generate the jet is longer than in the 2-D axisymmetric simulations. We expect that due to the additional freedom in the azimuthal dimension without axisymmetry with respect to the $z$ axis and reflection symmetry with respect to the equatorial plane, the dynamics of jet formation can be modified. Further simulations are required for the study of instabilities along the azimuthal direction such as accretion-eject instability} ",
        "introduction": "Relativistic jets have been observed in active galactic nuclei (AGNs) and microquasars in our Galaxy, and it is believed that they originate from the regions near accreting black holes or neutron stars \\cite{mku01}. To investigate the dynamics of accretion disks and the associated jet formation, we use a full 3-D GRMHD code.  One of the most promising models for jet formation is the magnetic-acceleration model \\cite{bp82}. The magnetic-acceleration mechanism has been proposed not only for AGN jets, but also for protostellar jets (see \\cite{mku01}). %  Recently, Koide, Shibata, \\& Kudoh (1999) \\cite{ksk99} have investigated the dynamics of an accretion disk initially threaded by a uniform poloidal magnetic field in a non-rotating corona (either in a state of steady fall or in hydrostatic equilibrium) around a non-rotating black hole. The numerical results show that the disk loses angular momentum by magnetic braking, then falls into the black hole.  The infalling motion of the disk, which is faster than in the non-relativistic case because of the general-relativistic effect below $3 r_{\\rm S}$ ($r_{\\rm S}$ is the Schwarzschild radius), is strongly decelerated at the shock formed by the centrifugal force around $r = 2 r_{\\rm S}$ by the rotation of the disk. Plasmas near the shock are accelerated by the ${\\bf J} \\times {\\bf B}$ force, which forms bipolar relativistic jets. Inside this {\\it magnetically driven jet}, the gradient of gas pressure also generates a jet over the shock region ({\\it gas-pressure-driven jet}). This {\\it two-layered jet structure} is formed both in a hydrostatic corona and in a steady-state falling corona. Koide et al.~(2000) \\cite{kmsk00} have also developed a new GRMHD code in Kerr geometry and have found that, with a rapidly rotating ($a = 0.95$) black-hole magnetosphere, the maximum velocity of the jet is 0.9 c and its terminal velocity 0.85 c. All of the previous 2-D GRMHD simulations described here were made assuming axisymmetry with respect to the $z$-axis and mirror symmetry with respect to the plane $z = 0$; the axisymmetric assumption suppressed the azimuthal instabilities. ",
        "conclusions": "Recently, review articles on magnetohydrodynamic production of relativistic jets have reported unsolved questions related to jet formation and its mechanisms which determine the velocity of jets and time variations of jet flux \\cite{mku01,bld01,meier02}.  This simulation result is initial and we will perform more simulations and investigate effects of the third dimension. This simulation study will be extended to understand the different (high/soft and low/hard) states \\cite{fender01} % and different combination of accreting rate $\\dot{m} \\equiv \\dot{M}/\\dot{M}_{Edd}$ and the angular momentum $j \\equiv J/J_{\\rm max}$ \\cite{meier02}. % Further results will be reported elsewhere."
    },
    "0208/astro-ph0208437_arXiv.txt": {
        "abstract": "We have identified counterparts to two submillimeter (submm) sources, SMM\\,J09429+4659 and SMM\\,J09431+4700, seen through the core of the $z=0.41$ cluster A\\,851. We employ deep 1.4-GHz observations and the far-infrared/radio correlation to refine the submm positions and then optical and near-infrared imaging to locate their counterparts.  We identify an extremely red counterpart to SMM\\,J09429+4659, while GMOS spectroscopy with Gemini-North shows that the $R=23.8$ radio source identified with SMM\\,J09431+4700 is a hyperluminous infrared galaxy (L$_{FIR}\\sim 1.5\\times 10^{13}$L$_\\odot$) at $z=3.35$, the highest spectroscopic redshift so far for a galaxy discovered in the submm. The emission line properties of this galaxy are characteristic of a narrow-line Seyfert-1, although the lack of detected X-ray emission in a deep {\\it XMM-Newton} observation suggests that the bulk of the luminosity of this galaxy is derived from massive star formation. We suggest that active nuclei, and the outflows they engender, may be an important part of the evolution of the brightest submm galaxies at high redshifts. ",
        "introduction": "Sensitive surveys in the submm and millimeter wavebands have identified a population of distant dusty, active galaxies which may represent the formation phase of massive spheroidal galaxies (Smail, Ivison \\& Blain 1997; Hughes et al.\\ 1998; Bertoldi et al.\\ 2000; Scott et al.\\ 2002; Webb et al.\\ 2002).  Irrespective of the precise mechanism responsible for the prodigious luminosity of these galaxies, either star formation or dust-reprocessed radiation from an AGN, several have been confirmed as high redshift massive, gas-rich galaxies (Frayer et al.\\ 1998, 1999).  One of the most pressing issues for study is to identify the most distant examples.  These provide a critical test of theoretical models of galaxy formation and evolution, which already struggle to produce sufficiently large gas masses in galaxies at $z\\sim 2$ (Granato et al.\\ 2002).  Identifying similarly luminous, gas- and dust-rich mergers at even higher redshifts, $z>3$, will provide even stronger constraints.  In this letter we discuss the identification and spectroscopic follow-up of two recently discovered submm galaxies in the field of the $z=0.41$ cluster A\\,851.  Exploiting very deep radio and optical/near-infrared images we identified counterparts to both submm sources and subsequently targetted them in spectroscopic observations with the GMOS spectrograph on Gemini-North.  One of these galaxies has the highest spectroscopic redshift for a SCUBA galaxy to date, at $z=3.35$.  We adopt a cosmology with $\\Omega_{\\rm m}=0.3$, $\\Omega_\\Lambda=0.7$ and H$_{\\rm o}=70$\\,km\\,s$^{-1}$\\,Mpc$^{-1}$.  \\begin{figure*}[tbh] \\centerline{\\psfig{file=f1.ps,angle=270,width=6.0in}} \\caption{$BRK$ views of the fields containing the submm sources SMM\\,J09429+4659 and SMM\\,J09431+4700.  The large circles show the nominal $6''$-diameter error circle for the SCUBA sources, while the smaller circles show the positions of the radio counterparts.  Note the strong contrast between the optical/near-infrared colors of H6 and the Extremely Red Object, H8. Each panel is $12''\\times 12''$ with North top and East left.} \\end{figure*} ",
        "conclusions": "The spectroscopic identification of a second HyLIRG in the SCUBA population provides the opportunity for a detailed comparison of the properties of these two galaxies: SMM\\,J02399$-$0136 and SMM\\,J09431+4700.  Detailed study of SMM\\,J02399$-$0136 has shown that it is a massive, gas-rich system (Ivison et al.\\ 1998; Frayer et al.\\ 1998), identified with a pair of galaxies: L1 and L2, at $z=2.80$, and has a far-infrared luminosity of L$_{FIR}\\sim 1\\times 10^{13}$\\,L$_\\odot$.  L1 hosts a partially obscured AGN, which has recently been classified as a Broad Absorption Line (BAL) QSO (Vernet \\& Cimatti 2001).  The second component L2 is separated from L1 by about 10\\,kpc in projection (compared to 50\\,kpc for H6--H7), and may be tidal debris rather than an independent galaxy.  Apart from their apparent binary morphologies, the most striking similarity between these two HyLIRGs is that both systems host AGN. As highlighted by Ivison et al.\\ (2000): 80\\% of the SCUBA galaxies with known redshifts show some signs of an AGN.  The close relationship of AGN and QSO activity to the growth of supermassive black holes (SMBH) and the apparent ubiquity of SMBH in local, massive spheroids suggests that this is a natural consequence if SCUBA galaxies are the progenitors of the most massive galaxies in the local Universe (e.g.\\ Sanders et al.\\ 1988; Silk \\& Rees 1998; Granato et al.\\ 2001; Archibald et al.\\ 2001).  One measure of the importance of the AGN in these systems is to quantify its contribution to their total emission. X-ray observations of the two HyLIRG submm galaxies suggests that in neither does the AGN dominate the bolometric emission (Bautz et al.\\ 2000), which instead comes from an intense starburst which also produces the substantial masses of dust in these galaxies.  Although they do not dominate the energetics, the AGN may still have a profound effect on the evolution of these galaxies: the AGN in SMM\\,J02399$-$0136 is driving a substantial wind which may in time sweep the central regions of the galaxy clear of gas and dust.  Can we find any signs of similar AGN-induced feedback in SMM\\,J09431+4700? There are several hints: the contrast between the narrow, spatially-extended Ly$\\alpha$ emission and the broader, but spatially-unresolved high excitation lines, may indicate that the former arises from emission in an outflow.  This situation is very similar to that seen in some high-redshift radio galaxies, such as 53W002 at $z=2.4$ (Windhorst, Keel \\& Pascarelle 1998).  Moreover, the structured broad emission lines seen in H6 are reminiscent of the structures seen in broad emission lines of some radio galaxies and radio-loud quasars (Eracleous \\& Halpern 1994), although those show much larger velocity ranges.  These structured emission lines are interpreted as resulting from scattering of radiation from the AGN by outflowing conical winds (Corbett et al.\\ 1998) and the same mechanism may be operating in SMM\\,J09431+4700/H6.  The final connection is from the spectral classification of this galaxy as a NLSy1, where analysis of examples at $z\\sim 0$ have led to the identification of strong nuclear winds and a suggested link to BALQSOs (Lawrence et al.\\ 1997; Leighly et al.\\ 1997; Laor et al.\\ 1997).  If both systems do show signatures of massive outflows -- this suggests that winds powered by the AGN (as well as starbursts) must be central to our understanding the growth of the spheroidal components in these massive, young galaxies (Granato et al.\\ 2002).  The feedback on the system from energy injected by the AGN provides one possible mechanism for creating the observed SMBH to bulge mass correlation seen in local galaxies (Maggorian et al.\\ 1998).  The future evolution of these submm sources will be determined by the ability of the AGN to clear the bulk of the gas and dust from the nuclear regions -- if they can, then they may evolve into the population of less-obscured QSOs.  With regard to the wider environment of H6, we note a striking coincidence: H6 is just 400\\,km\\,s$^{-1}$ and less than 1\\,Mpc ($64''$) from an optically-selected galaxy, DG\\,433 ($z=3.3435$), (Trager et al.\\ 1997), suggesting that the two galaxies inhabit a single structure.  DG\\,433 has a UV spectrum dominated by absorption lines and an estimated star formation rate of $\\ls 10^2$\\,M$_\\odot$\\,yr$^{-1}$, indicating it is a much less active system than the hyperluminous galaxy H6.  The relationship between the highly obscured and very active submm-luminous galaxies and the less obscured populations which cluster in the same environments will be one of the most important questions to address in the next few years."
    },
    "0208/hep-ph0208272_arXiv.txt": {
        "abstract": "We study the generation of magnetic fields during the stage of particle production resulting from spinodal instabilities during phase transitions out of equilibrium. The main premise is that long-wavelength instabilities that drive the phase transition lead to strong non-equilibrium charge and current fluctuations which generate electromagnetic fields. We present a formulation based on the non-equilibrium Schwinger-Dyson equations that leads to an {\\bf exact} expression for the spectrum of electromagnetic fields valid for general theories and cosmological backgrounds and whose main ingredient is the transverse photon polarization out of equilibrium. This formulation includes the dissipative effects of the conductivity in the medium. As a prelude to cosmology we study magnetogenesis in Minkowski space-time in a theory of $N$ charged scalar fields to lowest order in the gauge coupling and to leading order in the large $N$ within two scenarios of cosmological relevance. The long-wavelength power spectrum for electric and  magnetic fields at the end of the phase transition is obtained explicitly. It follows that  equipartition between electric and magnetic fields {\\bf does not hold} out of equilibrium.   In the case of a transition from a high temperature phase, the conductivity of the medium severely hinders the generation of magnetic fields, however the magnetic fields generated are correlated on scales of the order of the domain size, which is much larger than the magnetic diffusion length. Implications of the results to cosmological phase transitions driven by spinodal unstabilities are discussed. ",
        "introduction": "There is compelling astrophysical evidence for the existence of magnetic fields in the Universe~\\cite{Parker}. Recent advances in observational techniques mainly through Faraday rotation (RM) complemented with an independent measurement of the electron column density via pulsar dispersion measurements(DM) or X-ray emission from clusters\\cite{Kronberg:1993vk} indicate that the strength of these astrophysical magnetic fields is of order of $\\mu\\mathrm{G}$ and are coherent on scales up to that of galaxy clusters $100~\\mathrm{Kpc}-1~\\mathrm{Mpc}$\\cite{Parker,Kronberg:1993vk}. These magnetic fields are now deemed to play an important role in the evolution and dynamics of galaxies but their origin is still largely unknown\\cite{Grasso:2000wj,Dolgov:2001ce,widrev,giovannini1}.  The galactic dynamos  are some of the promising mechanisms to amplify pre-existing seeds\\cite{Kronberg:1993vk,malyshkin,plasmabook,widrev}. In its simplest conception a linear or kinetic dynamo transfers energy from differential rotation to the build-up of a magnetic field with a typical growth rate determined by the rotation period of a protogalaxy $\\sim \\mathrm{Gyr}^{-1}$. There are many alternative versions of dynamo theory, and some of the most promising require turbulent flows\\cite{Kronberg:1993vk,malyshkin}. Dynamos amplify seeds but an initial seed must be present. The proposals to explain the initial seeds can be classified as astrophysical or of primordial cosmological origin. An important astrophysical mechanism is the Biermann battery (for a recent discussion see~\\cite{malyshkin}) which relies on gradients in the electron number density and pressure which are in different directions that can arise behind hydrodynamic shocks\\cite{widrev}. The primordial cosmological mechanisms purport the generation of seed magnetic fields during different stages of the Early Universe\\cite{TurnerWidrow}. As originally observed by Turner and Widrow\\cite{TurnerWidrow,widrev} the coupling of the electromagnetic field to gravity is conformally invariant, resulting in that cosmological expansion \\emph{per se} will not generate primordial magnetic fields, the coupling of charged fields to gravity is in general not conformally invariant and can lead to magnetogenesis through the minimal coupling to electromagnetic fields. These authors studied a wide range of possibilities for primordial magnetogenesis with encouraging results. More recently a thorough study of the generation of (hyper) magnetic fields during inflation concluded that the amplitude of the seeds on cosmologically relevant scales is probably too small\\cite{Giovannini:2000dj}. These authors included dissipative effects in the medium via a kinetic approach that includes the conductivity.  The amplification of electromagnetic fluctuations during inflation, by the end of inflation and from inflation to reheating has been studied by several  authors in the context of different inflationary models\\cite{muchos} (see criticisms in ref.\\cite{giovashapo}). Dolgov~\\cite{dolgovanomaly} suggested that the breaking of conformal invariance through the trace anomaly may also lead to significant magnetogenesis (see also \\cite{joy}) and non minimal electromagnetic couplings has also been considered\\cite{TurnerWidrow,ratra,Grasso:2000wj}.  Other proposals suggest that primordial magnetic fields can be generated much later during the electroweak phase transition\\cite{otros} or in connection with dark energy during the epoch of large scale structure\\cite{Lee:2001hj}. For a pioneering proposal for magnetogenesis from cosmological first order phase transitions see ref.\\cite{hogan}.  While certainly there is no dearth of proposals an important aspect that is clear from the present body of work is that a consistent framework to study the generation of primordial magnetic fields must include consistently (or semi phenomenologically) the dissipative effects associated with the high conductivity of the medium. This aspect was already highlighted in the seminal work of Turner and Widrow\\cite{TurnerWidrow} and again by Giovannini and Shaposhnikov\\cite{Giovannini:2000dj,giovashapo}.  This is clearly more relevant if the medium is a hot and dense plasma as prevailed during reheating, preheating and most likely any phase transition associated with particle physics scales.  \\vspace{1mm}  {\\bf The goals of this article:} In this article we begin a program to assess the generation of magnetic fields through the non-equilibrium processes associated with supercooled (second order) phase transitions. Here we focus on the process of spinodal decomposition, namely the process of phase separation, domain formation and growth resulting from a non-equilibrium phase transition as the mechanism that produces strong charge fluctuations which lead to magnetogenesis. Although we focus primarily on spinodal decomposition, similar arguments will apply for very weak first order phase transitions with small latent heat and nucleation barriers, since in this case nucleation will be almost indistinguishable from spinodal decomposition. We study models in which there is a charged scalar sector coupled to an abelian (hypercharge) gauge field \\emph{without} breaking the $U(1)$ symmetry associated with the (hyper) charge. The charged scalar and abelian gauge fields are supposed to be part of a larger multiplet of fields pertaining to a larger underlying theory.  We develop the framework to calculate the power spectrum of magnetic fields generated out of equilibrium directly from the underlying quantum field theory. An {\\bf exact} expression for the spectrum of magnetic fields generated through non-equilibrium processes is obtained directly from the non-equilibrium Schwinger-Dyson equations for the transverse gauge field propagator [see eq.(\\ref{S.fund})]. This expression is general and valid for all types of charged matter fields, the main ingredient being the non-equilibrium transverse polarization for the (hypercharge) gauge field.   This framework allows to include the effects of  a conductivity in the medium and is a generalization of a previous study of photon production out of equilibrium\\cite{Boyanovsky:1999jh}.  After establishing a connection between the dynamics in cosmological space times and the simplified case of flat Minkowski space time with a model for the phase transition, we apply these methods to study the power spectrum of magnetic fields during quenched phase transitions. In order to study the non-perturbative aspects of spinodal decomposition, such as the growth of unstable modes below the phase transition we invoke a model of $N$ charged fields in the large $N$ limit. Most particle physics models (GUT's, SUSY, etc.) contain a large number of scalar fields, thus warranting the use of the large $N$ approach.   The purpose of this study is to extract the important and robust features that lead to the generation of magnetic fields from these non-equilibrium processes in a simpler setting and to build intuition into the main physical aspects.  We postpone the application of these methods to the full cosmological situation to forthcoming articles.  As mentioned above, many mechanisms for primordial magnetogenesis have been studied previously~\\cite{Grasso:2000wj}-\\cite{ratra}. Our main goal here is to provide {\\bf reliable quantitative estimates} for primordial magnetogenesis during out of equilibrium phase transitions. In particular we focus on the dynamics of magnetogenesis in a minimally coupled scalar-gauge theory where the non-equilibrium dynamics is a consequence of spinodal instabilities. The formulation presented here is very general, it includes consistently absorptive effects such as the conductivity and can be implemented in other scenarios.  In section II we present the main physical picture behind the mechanism of magnetogenesis studied. In section III we introduce the model under consideration and discuss the  issue of gauge invariance.   Section IV discusses the similarity  between Friedmann-Robertson-Walker cosmologies and flat Minkowsky space time that warrants a preliminary study of magnetogenesis within the simpler setting. As a prelude to cosmology, section V  briefly reviews the main aspects of non-equilibrium dynamics of spinodal decomposition in the scalar sector in Minkowsky space time, which are present in the cosmological setting. In section VI we obtain the {\\bf exact} expression for the power spectrum of magnetic fields directly from the non-equilibrium Schwinger-Dyson equations for the transverse photon propagator. We compare the non-equilibrium result to the more familiar equilibrium setting.  In section VII we implement this framework to study magnetogenesis in two cosmologically relevant settings that purport to model a phase transition during inflation and a radiation dominated era. We discuss in detail how the conductivity of the plasma emerges in this formulation and obtain the energy density on long-wavelength magnetic fields extracting features that will remain in the full cosmological problem. ",
        "conclusions": "In this article we have provided a framework to study consistently the generation of primordial magnetic fields produced by non-equilibrium phase transitions. The main premise is that during the process of spinodal decomposition which is the hallmark of non-equilibrium phase transitions (without free energy barriers) the instabilities that drive the process of phase separation and domain formation would lead to strong charge and current fluctuations if the scalar fields carry (hyper) charge. One of the main results of this article is a calculational framework derived from the underlying non-equilibrium Schwinger-Dyson equations. The result is an {\\bf exact} expression for the spectrum of magnetic and electric fields which is valid in all generality  for scalar and spinor charged fields and in an arbitrary cosmological background[see eq.(\\ref{S.fund})]. The main ingredient is the transverse photon polarization and we have confirmed that this expression has the correct equilibrium limit.  Separating the contribution from short wavelengths that are in local thermodynamic equilibrium from the long-wavelength fluctuations which fall out of equilibrium during the phase transition allows us to include the dissipative effects associated with the conductivity in a high temperature plasma.  We studied the generation of magnetic fields during quenched phase transitions in Minkowski space time for a theory of $N$ charged scalar fields coupled to an abelian gauge field (hypercharge) and one neutral scalar field (the inflaton). This is a reliable prelude to cosmology since the dynamics in Friedmann-Robertson-Walker backgrounds is similar to that in Minkowski space-time with a time dependent mass. Symmetry breaking occurs along the neutral direction and gauge symmetry is not spontaneously broken. We explicitly obtained the spectrum of electric and magnetic fields to leading order in $\\alpha$ (hypercharge) and to leading order in the large $N$ [see eqs.(\\ref{potmag})-(\\ref{specratio})].  Two cases of cosmological relevance were studied, a transition in vacuum that models an inflationary phase transition, and one from a high temperature phase to a low temperature phase that models a transition in a radiation dominated era. In the second case the high temperature plasma has a large conductivity. Separating the contribution from short wavelength fluctuations to the photon polarization we include the effects of the conductivity, while long wavelength fluctuations of the charged fields fall out of equilibrium during the transition and lead to the generation of magnetic and electric fields. We have provided explicit expressions  for the magnetic and electric power spectra for long wavelengths [see eq.(\\ref{hiTelec})-(\\ref{ratiorhos})].  These studies have revealed several robust features which are expected to survive in cosmological spacetimes:  \\begin{itemize} \\item{The magnetic fields are correlated over length scales comparable to the size of the scalar field domains produced during the spinodal stage and in a high temperature plasma,  this scale is much larger than the magnetic diffusion length. }  \\item{Equipartition between electric and magnetic fields does {\\bf not} hold out of equilibrium. For rolling down from an initial vacuum state the ratio of electric to magnetic power behaves  as $ \\frac{\\mu^2}{k^2} \\, \\log \\lambda $ for long wavelengths [see eq.(\\ref{specratio})].  Electric photons overwhelm magnetic ones in this case. In the case of a conducting plasma, the ratio between electric and magnetic power behaves as $k^2/\\sigma^2$ for small $k$ with $\\sigma$ the DC conductivity [see eq.(\\ref{hiTelec})]} therefore magnetic photons dominate over electric ones in this case.  \\item{The conductivity severely hinders the generation of long-wavelength magnetic fields.   }  \\item{The ratio $r(L)$ between the energy density of magnetic fields on scales larger than $L$ and the energy density of background radiation features an ubiquitous factor $(LT)^{-4}$ which leads to large suppression factor which must be overcome in order for primordial seed fields to be amplified by the galactic dynamo mechanism. This point was originally made in ref.\\cite{TurnerWidrow} and is one of the formidable roadblocks that must be overcome towards the explanation of galactic and extragalactic magnetic fields from primordial seeds. }  \\end{itemize}  We will report on our study of magnetogenesis in inflationary and radiation dominated cosmologies in a forthcoming article\\cite{magfieldII}."
    },
    "0208/astro-ph0208398_arXiv.txt": {
        "abstract": " ",
        "introduction": "\\label{sec:1}  %  One of the greatest challenges of modern astrophysics is understanding how galaxies,  such as our Milky  Way, form within the  framework of the Big  Bang cosmology. The  current theory  of structure  formation, the extension  of the Big  Bang model  called the  Cold Dark  Matter (CDM) scenario,  predicts  that galaxies  form within extended massive dark matter halos built from  smaller pieces that  collided and  merged, resulting in the hierarchy of galaxies, groups, and clusters observed today.  The entire  sequence  of  events  is  thought  to  be  seeded  by  quantum fluctuations  in the very  early Universe  and governed  by mysterious \"dark  matter\"  which constitutes  about  85\\%  of  all matter  in  the universe.  Although the accurate properties of galaxies depend on complicated baryonic processes (radiative cooling, formation and evolution of stars, etc.) operating on small scales, we expect that overall spatial distribution of dark matter halos is closely related to the observed galaxy distribution. Here we present numerical simulations designed to study the formation, evolution and present day properties of such dark matter halos in different cosmological environments.  In all simulations the spatially flat cold dark matter model with a cosmological constant ($\\Lambda$CDM with $\\Omega_{\\rm M}=0.3$, $\\Omega_{\\Lambda}=0.7$, $\\sigma_8=0.9$, and $h=0.7$), favored by most current observations, has been assumed.  \\smallskip ",
        "conclusions": "The adaptive refinement tree code is a useful tool to study cosmological structure formation on different scales and with different resolutions. It runs well on a variety of platforms with shared and distributed memory. The simulations described here have been performed on the Hitachi of LRZ Munich, the small development Hitachi at the AIP, the IBM SP of the Potsdam Institute for Climate Impact Research, and the IBM SP of NERSC Berkeley.  \\smallskip"
    },
    "0208/astro-ph0208484_arXiv.txt": {
        "abstract": "Massive black holes appear to be present in the nuclei of almost all galaxies, but their genesis and evolution are not well understood.  As astrophysical black holes are completely characterized by their masses and spins, the observed joint distribution of these quantities contains important clues to their history.  We examine the coevolution of mass and spin in binary merger growth scenarios.  We find that holes are typically spun {\\it down} by mergers.  Rapid rotation results only if the binary's larger member already spins quickly and the merger with the smaller hole is consistently near prograde; or, if the binary's mass ratio approaches unity.  If, as some observations have suggested, observed black holes spin rapidly, then this limits the importance of merger scenarios for the growth of black holes. ",
        "introduction": "Black holes span a wide spectrum of masses: the case for stellar mass holes ($M\\sim 10\\,M_\\odot$) in the field [e.g., {\\citet{bailyn98}}] and supermassive holes ($M\\sim 10^6 - 10^9\\,M_\\odot$) in galactic bulges [e.g., {\\citet{ferr2002,korm_geb2001}}] is extremely strong; tantalizing evidence suggests middleweight holes ($M\\sim 10^2 - 10^4\\,M_\\odot$) as well {\\citep{cm1999,cp2002,gebhardt2000,vdm2001}}. Stellar mass holes likely form in stellar collapse; the origins of more massive holes remains mysterious.  Such holes could form in the collapse of massive gas accumulations; they could grow from smaller holes by accretion; they could grow by capturing stellar mass bodies; and they could grow by repeatedly merging with holes of comparable mass.  Any or indeed all of these mechanisms could contribute to the growth of a given hole.  A black hole's spin may help identify which scenario most strongly impacted its recent history.  Since spin likely drives outflows and jets in active galaxies, and since jets are presumed to align with black hole spin {\\citep{rees1978}}, spin may provide an observational probe of a hole's recent growth {\\citep{m2002}}.  We examine how spin and mass coevolve in mergers.  Binaries will form following galaxy mergers {\\citep{bbr1980}}, and may harden to the point that gravitational-wave (GW) emission drives its members together. Eventually, they encounter the last stable orbit (LSO), and then plunge and coalesce into a single hole.  Our goal is to understand the mass and spin of this remnant hole.  For nearly equal mass holes, this is extremely difficult: we must model the spacetime dynamics of the transition from a binary to a single black hole, accounting for both holes' spins and the radiated energy and angular momentum.  A proper analysis requires mature numerical relativity codes [see, e.g., {\\citet{lehner2001}}].  The problem is simpler for small mass ratio, $q\\equiv m_2/m_1\\equiv m/M\\ll 1$.  This binary is well described as a test particle orbiting a black hole.  GW emission shrinks the small hole's orbit to the LSO, whereupon it plunges into the large hole.  Neglecting the final emission of radiation after the plunge, the hole evolves simply: its mass adds the small body's energy at the LSO, its spin adds the LSO angular momentum.  Because we only need global ``conserved'' quantities, this description works surprisingly well even for rather large mass ratio. Post-Newtonian analyses {\\citep{blanchet2002,bd1999,damour2001}} show that finite mass ratio typically changes the LSO and its orbital constants by a factor of order $\\eta\\equiv mM/(m + M)^2 = q/(1 + q)^2\\le 0.25$.  The error due to the test particle description is $\\lesssim 0.3$ for $q\\lesssim 0.5$.  We also may safely neglect the energy and angular momentum radiated in the final merger: although its GW luminosity may be large, its duration will be very short.  The mass carried off in this phase, for example, is $\\Delta M\\simeq (0.01 - 0.1)M q^2$ {\\citep{drpp,sasnak}}.  Neglecting this radiation incurs an error that is less important than other errors built into our approximations, and rapidly becomes negligible for small mass ratio. Likewise, the small hole's spin can be neglected: since a hole's spin scales with its mass squared, spin will be less important than the orbital angular momentum, provided we exclude $q\\gtrsim 0.5$.  We set the speed of light $c$ and Newton's constant $G$ to 1; a useful conversion factor is $1\\,M_\\odot = 1.5\\,\\mbox{km}$.  Our binary has masses $M$ and $m$; the larger hole has mass $M$ and spin $|{\\bf S}| = aM = aMc/G$; $0\\le a\\le M$.  Vectors are written in boldface; hatted quantities have been made dimensionless by dividing out powers of mass --- e.g., $\\hat a = a/M$. ",
        "conclusions": "We find that the remnant of a major merger is rarely rapidly rotating: rapid rotation follows only if the larger binary member spun rapidly before merger {\\it and} the plunge was nearly prograde; or, if the binary's mass ratio $q\\simeq1$.  Given the variety of black hole masses seen in galaxies, mergers with $q\\simeq1$ should be rare; given the small volume of parameter space leading to rapid rotation, serendipitous configurations leading to a rapid rotation should also be rare.  Mounting evidence, mostly from spiral galaxies, suggests that in many cases massive black holes nonetheless rapidly rotate [e.g., {\\citet{wilmsetal2001,erz2002}}].  Our results strongly suggest that this spin cannot come from mergers [e.g.\\ \\citet{wc1995, kauff00}] but, instead, is consistent with the view that black hole mass (and spin) is assembled radiatively [e.g.\\ \\citet{small92,yut02}].  The spin evolution of a hole that grows by repeated minor mergers is neatly described by a Fokker-Planck equation [Eq.\\ (\\ref{eq:FP})], taking a particularly simple form if the mergers arise from an isotropic cluster.  In this case, the evolution has a secular component, which is approximately described by a ``doctrine of original spin'' [$S = {\\hat a}M^2$ remains roughly constant while $M$ grows; cf.\\ Eq.\\ (\\ref{eq:secularevolve})], and a diffusive component, with a spectrum of fluctuations governed by the coefficient $D_{||}$ [cf.\\ Eq.\\ (\\ref{eq:Dparterm})].  This is in accord with recent work on models to grow intermediate mass black holes in clusters {\\citep{cole}}.  Finally, we predict the typical angle $\\langle\\delta\\theta\\rangle$ by which a hole's orientation changes following merger [Eq.\\ (\\ref{eq:delta_theta})].  These results may be of particular observational interest.  Jets launched by a black hole's spin should track its inclination: if the hole is kicked into a new orientation, the jet will ``kink''.  The angle of the kink should equal the hole's change in orientation.  Applying our predicted dependence for the kink angle on the binary's parameters may provide some insight into the conditions of the hole's last merging, untangling a bit of its recent growth history.  An abrupt change in inclination [such as discussed in {\\citet{me2002}}] requires a comparatively rare major merger.  This work grew out of discussions at the KITP conference {\\it Black Holes: Theory Confronts Reality, Three Years Later}; we thank the participants of that meeting for their stimulating input.  We are also extremely grateful to Xinwu Cao for pointing out that the figure used in an earlier version of this manuscript contained important errors. SAH is supported by NSF Grant PHY-9907949; RDB is supported by NASA grants NAGW 5-2837, 5-7007."
    },
    "0208/astro-ph0208217_arXiv.txt": {
        "abstract": "We compute the structure function scaling of a 2MASS extinction map of the Taurus molecular cloud complex. The scaling exponents of the structure functions of the extinction map follow the Boldyrev's velocity structure function scaling of super--sonic turbulence. This confirms our previous result based on a spectral map of $^{13}$CO J=1-0 covering the same region and suggests that supersonic turbulence is important in the fragmentation of this star--forming cloud. ",
        "introduction": "Stars are formed predominantly from very large clouds of cold interstellar gas containing up to millions of solar masses of material. The dynamics of such clouds is therefore a crucial ingredient in the process of star formation. Observations of emission spectra of molecular transitions have shown that the kinematics of star--forming clouds is best described as supersonic random motions, often referred to as supersonic turbulence. Numerical simulations of supersonic turbulence have indeed been compared successfully with the observations, in the sense that many statistical properties of numerical supersonic turbulent flows are also found in the observational data (e.g. Padoan et al. 1998, 1999, 2001).  \\nocite{Padoan+98cat} \\nocite{Padoan+99per} \\nocite{Padoan+2001cores}  The cold gas in star--forming clouds, especially their molecular component, cools very rapidly down to a typical equilibrium temperature of approximately 10~K. The shocks caused by the observed random supersonic velocity field are therefore roughly isothermal, which allows them to compress the gas effectively. Expansions are also favored by the isothermal behavior of the gas and large voids of very low density can be generated. The result is a very large contrast between the highest and the lowest densities, as commonly found in numerical simulations with an isothermal equation of state. We usually refer to this effect of the turbulent velocity on the density field as {\\it turbulent fragmentation}, to stress the fact that star--forming clouds are more likely fragmented into dense prestellar cores directly by the turbulent velocity field, rather than by a hierarchical process of gravitational fragmentation.  The traditional way to study the dynamics of star forming clouds is to probe their kinematics by the Doppler shift in spectral lines of molecular transitions. However, the dynamics can also be studied through the density field, as the gas density is so strongly affected by the supersonic velocity. The investigation of the cloud spatial structure may be used to test predictions of numerical and analytical models more directly than using the velocity field. Projected density is in fact easier to measure than the velocity field. This is especially true if stellar extinction measurements are available, since they provide the most reliable estimate of column density.  Thanks to the recently completed ``Two Micron All Sky Survey'' (2MASS; Cutri et al. 2001), it is now possible to generate extinction maps of several extended giant molecular cloud complexes with a dynamical range in both column density and spatial resolution that is not matched by any molecular line survey. In this work we derive new extinction maps of the Taurus region using 2MASS point source data, and show that we can probe values of dust column density over more than two orders of magnitude and achieve a spatial resolution higher than in IRAS 100~$\\mu$m images (\\S~2). In \\S~3 we present the results of the structure function analysis of the extinction map and obtain a scaling that is indistinguishable from that of the velocity structure functions in supersonic turbulence. Conclusions are drawn in \\S~4. ",
        "conclusions": "Boldyrev (2002) has recently proposed an analytic model for the velocity structure function scaling in supersonic turbulence. The model is an extension of the scaling of incompressible turbulence proposed by She \\& Leveque (1994) and has already been successfully tested with numerical simulations of supersonic turbulence (Boldyrev, Nordlund \\& Padoan 2002a). An equivalent analytic model for the scaling of the structure functions of projected density in supersonic turbulence is not available yet. Only the slope of the second order structure function has been derived from the velocity structure functions, under certain approximations (Boldyrev, Nordlund \\& Padoan 2002b). However, the fact that the projected density follows the same scaling as the velocity field in supersonic turbulence suggests that the density field in the Taurus region is the result of a multiplicative process with a Log-Poisson statistics (Dubrulle 1994), very likely the result of the turbulent fragmentation.  \\nocite{Boldyrev+2002scaling} \\nocite{Boldyrev+2002structure} \\nocite{She+Leveque94} \\nocite{Dubrulle94}  The importance of supersonic turbulence in the fragmentation of star-forming regions has been established in previous works (e.g. Padoan \\& Nordlund 1999; Padoan et al. 2001; Padoan \\& Nordlund 2002). The purpose of the present work is primarily to determine the statistical properties of the fragmentation process, independent of its origin. Such statistical properties may be universal, for example if they are mainly the consequence of turbulence, or depend on several physical parameters, such as gas density, temperature, turbulent velocity dispersion and star formation activity. We plan to compute and analyze 2MASS extinction maps of different extended star-forming regions in order to establish the properties of the fragmentation process that leads to the formation of stars in different environments.  \\nocite{Padoan+Nordlund99mhd} \\nocite{Padoan+2001cores} \\nocite{Padoan+Nordlund2002imf}"
    },
    "0208/astro-ph0208021_arXiv.txt": {
        "abstract": "s{We present a scenario for the spin-up and evolution of binary millisecond pulsars.  This can explain the observational properties of the recently discovered binary millisecond pulsar \\psr, with orbital  period 32.5 hrs, in the Globular Cluster  NGC 6397. The optical counterpart of this system is a star  as luminous as the cluster turnoff stars,  but with a lower \\Teff\\ (a  larger radius) which we  model with a star  of initial mass compatible  with the  masses evolving  in the cluster  ($\\simeq 0.85$ \\Msun). This star has  suffered Roche lobe overflow while  evolving off the main sequence, spinning up the neutron star to the present period  of 3.65 ms.  There are evidences that at  present, Roche lobe overflow is still going on. Indeed Roche lobe deformation of the mass losing component is necessary to  be compatible with the  optical light curve. The presence of matter around the system is also consistent with the long lasting irregular radio eclipses seen in the system. We propose that this system is presently in a phase of `radio--ejection' mass loss. The radio--ejection phase can be initiated only if the system is subject to intermittency in the mass  transfer during the spin--up phase. In fact, when the system is detached the pulsar radio emission is not quenched, and may be able to prevent further mass accretion due to the action  of the pulsar pressure  at the inner  Lagrangian point. } ",
        "introduction": "The widely accepted scenario for the formation of a millisecond radio pulsar (hereafter MSP) is the  recycling of an old  neutron star (hereafter NS) by a spin-up process  driven by  accretion of  matter and  angular momentum from a Keplerian disc, fueled {\\it via}  Roche lobe overflow of a binary late--type companion (see  Bhattacharya  \\&  van den  Heuvel  1991 for  a review). If  the NS has a  magnetic dipole moment  (typical values are $\\mu \\sim 10^{26}\\--10^{27}$ G cm$^{3}$)  the disc is truncated at the magnetosphere,  where   the  disc  pressure  is balanced by the magnetic pressure, $P_{\\rm MAG}$, exerted by the NS magnetic  field. Once  the accretion  and spin-up  process ends, the NS is visible as a MSP. Indeed,  a  common requirement  of  all  the  models of  the  emission mechanism  from   a  rotating  magnetic  dipole  is   that  the  space surrounding the NS  is free of matter up to  the light cylinder radius $R_{\\rm LC}$ (at  which the speed of a body  rigidly rotating with the NS equals the speed of light) .  An interesting evolutionary phase  can occur if the mass transfer rate drops below the  level required to allow the expansion of the magnetosphere beyond $R_{\\rm  LC}$, switching--on the emission from the rotating magnetic dipole (e.g. Illarionov \\& Sunyaev, 1975; Ruderman, Shaham \\& Tavani 1989; Stella et al., 1994). The pressure exerted by the radiation field of the radio pulsar may overcome the pressure of the accretion disk, thus determining the ejection of matter from the system. Once the disk has been swept away, the radiation pressure stops the infalling matter as it overflows the inner Lagrangian point. ",
        "conclusions": "% \\label{sec:system} We have considered the evolution of possible progenitors of the binary MSP \\psr\\ in the Globular Cluster NGC 6397. We can reproduce the HR diagram location of the optical companion, starting mass transfer to the NS from a hypothetical secondary of mass 0.85 \\Msun, slightly evolved off the main sequence when mass transfer begins. In conclusion we propose that:  i) Orbital evolution  calculations shows that a slightly evolved 0.85 \\Msun\\ secondary orbiting a NS can transfer mass to the NS, and reaches a stage in which its mass is reduced to $\\simeq 0.45$ \\Msun, and its optical location in the HR diagram is then compatible with the recently detected optical counterpart of \\psr;  ii) \\psr\\ might represent a system whose evolution has been envisioned by Burderi et al. (2001): the spin and the magnetic moment of the pulsar may keep the system in a radio--ejection  phase in which accretion is inhibited by the  radiation pressure  exerted by the  pulsar on  the overflowing matter while  the mechanism that  drives the Roche lobe  overflow from the companion is still active, thus causing an intense wind which would be very difficult to explain otherwise. This evolution seems to be the only viable possibility to explain the long lasting eclipses and the strong intensity variation randomly occurring in the radio emission.  An artistic impression of the system, according to this scenario, is shown in Figure~\\ref{artistic}. \\begin{figure} \\centering \\psfig{figure=ngc6397_artistic.ps3,height=3.9in,angle=270} \\caption{An artistic impression of the system \\psr\\ based on the scenario proposed in this paper (courtesy of ESO). \\label{artistic}} \\end{figure}  As a final remark we note that $P_{\\rm orb} \\sim P_{\\rm crit}$ suggests the interesting possibility that this system could swiftly switch from the  present radio  pulsar phase  to an  accretion phase  in  which it should be visible as a $L_{\\rm X} \\sim 10^{36}$ ergs s$^{-1}$ LMXB."
    },
    "0208/astro-ph0208550_arXiv.txt": {
        "abstract": "The ionization and kinematic properties of the emission line regions of three subsamples of 6C and 3CR radio galaxies have been compared. The degeneracy between redshift and radio power is broken, and the relative importance of radio power (P), radio size (D) and redshift (z) for the emission line region properties is determined. ",
        "introduction": "The emission line regions (ELRs) of high redshift radio galaxies are generally more extensive than those of lower redshift sources.  Spectroscopic studies of 3CR sources at $z \\sim 1$ (Best, R\\\"{o}ttgering \\& Longair 2000a, 2000b) reveal that the size of the radio source plays an important role in determining the dominant ionization mechanism.  The emission lines of small sources are best explained by the predictions of shock ionization associated with the passage of the radio cocoon. Larger sources usually appear photoionized by the central AGN, and typically exist in a higher ionization state than their smaller counterparts.  Studies of 3CR radio galaxies reveal considerable changes with redshift, with the lower redshift/radio power sources (degenerate in the flux limited 3CR sample) generally having much less extreme kinematics than sources at $z \\sim 1$ (Baum \\& McCarthy 2000; Best et al 2000b).  This degeneracy can be broken by means of a comparison with other radio galaxy surveys, e.g. the 6C sample. ",
        "conclusions": ""
    },
    "0208/astro-ph0208285_arXiv.txt": {
        "abstract": "We show that the mean percentage of stellar light re-radiated by dust is $\\sim 30\\%$ for the Virgo Cluster late-spirals measured with ISOPHOT by Tuffs et al. (2002). A strong dependence of this ratio with morphological type was found, ranging from typical values of $\\sim 15\\%$ for early spirals to up to $\\sim 50\\%$ for some late spirals. The extreme BCDs can have even higher percentages of their bolometric output re-radiated in the thermal infrared. Luminosity correction factors for the cold dust component are given for general use in converting far-infrared (FIR) luminosities derived from IRAS. ",
        "introduction": "Star-forming galaxies contain dust which absorbs some fraction of the emitted starlight, primarily re-radiating it in the far-infrared (FIR). This not only applies to starburst and ultraluminous systems, which radiate almost all their power in the FIR, but also to so-called ``normal'' galaxies - systems which are not dominated by AGN and not undergoing a starburst. Though less spectacular than starburst galaxies, normal galaxies still account for most of the infrared emissivity of the local universe. Their role in the distant universe is observationally a completely open question.  Recent observations with the ISOPHOT instrument (Lemke et al. 1996) on board the Infrared Space Observatory (ISO; Kessler et al. 1996), covering the spectral peak of the dust emission between 100 and 200$\\,{\\mu}{\\rm m}$, have shown that a significant contribution to the FIR luminosity of normal galaxies is actually radiated by grains too cold to be visible to IRAS. Statistical evidence for the existence of a cold dust component was established in our study (Popescu et al. 2002) of the spatially integrated FIR emissions of a complete volume- and luminosity sample of 63 gas-rich Virgo Cluster galaxies measured with ISOPHOT at 60, 100 and 170\\,${\\mu}$m (Tuffs et al. 2002).\\footnote{A subsample of these galaxies was also observed with the ISO LWS instrument by Leech et al. (1999).} These observations represent the deepest survey (both in luminosity and surface brightness terms) of normal galaxies yet measured in the FIR. In particular, these data showed that the cold dust component is present in all galaxies later than S0, i.e. spiral, irregular and blue compact dwarf (BCD) galaxies.  In this letter, these new results are used to evaluate the fraction of starlight emitted in normal galaxies which is re-radiated in the FIR, the first such measurement for these systems. Previous estimates based on the IRAS Bright Galaxy Sample (BGS; Soifer \\& Neugebauer 1991) have established a canonical value of 30\\% for the fraction of starlight to be re-radiated in the FIR in the local universe. However, this value refers to relatively bright FIR sources in which the bulk of the dust emission is radiated in the IRAS 60 and 100\\,${\\mu}$m bands, and is not representative of quiescent systems like the Virgo galaxies. In addition it takes no account of measurements longwards of 120\\,${\\mu}$m, not available at that time. The percentage of stellar light re-radiated by dust was investigated by Xu \\& Buat (1995), using an indirect estimate for the total FIR luminosity. Here we calculate this percentage by using for the first time measurements of the bulk of the dust emission in quiescent normal galaxies.  \\begin{figure} \\plotfiddle{figure1.eps}{2.2in}{0.}{60.}{60.}{-150}{-225} \\caption{The luminosity correction factor $corr2$ for the cold dust versus Hubble type.} \\end{figure}  Before evaluating the percentage of stellar light re-radiated by dust (Sect.~4) we describe our derivation of the total bolometric output in the ultraviolet (UV)/optical/near-infrared (NIR) in Sect.~3 and determine in Sect~2 luminosity correction factors for the cold dust, for general use in converting IRAS luminosities into total FIR luminosities. ",
        "conclusions": "We have shown that the luminosity correction for cold dust does not correlate with morphological type and exhibits a huge scatter, with values ranging from 0 to 150$\\%$. Only the S0a/Sa galaxies have relatively small corrections, as many of them are lacking the two dust temperature components. The big scatter in the correction factors for cold dust suggests that the FIR SEDs are strongly influenced by opacity effects, and not only by variations in the intensity and colour of the radiated starlight. Opacity depends on the geometrical distributions of the stellar populations and of the dust, both on large and small scales, as well as on metallicity, and the sheer sizes of the disks. All these quantities vary not only along the Hubble sequence, but also within a given morphological class, so that the scatter in Fig.~1 is not surprising, and no simple recipe can be considered to predict the luminosity of the cold dust emission component. Ultimately, radiative transfer calculations have to be performed for the derivation of the energy densities that heat the grains and thus produce FIR emission (e.g. Silva et al. 1998, Popescu et al. 2000a, Misiriotis et al. 2001, Popescu \\& Tuffs 2002). Our results also imply that there is not a one to one conversion of the FIR luminosity into star-formation rates.  A second result of this paper is the correlation of the ratio of the dust to the stellar output with morphological type. This correlation can be also interpreted as a sequence from normal to dwarf gas rich galaxies, with the dwarfs having an increased contribution of the FIR output to the total bolometric output. These findings could be important for our perception of the distant Universe, where, according to the hierarchical galaxy formation scenarios, gas rich dwarf galaxies should prevail at those epochs. We would then expect these galaxies to make a higher contribution to the total FIR output in the early Universe than previously expected. This, together with the cosmic-ray driven winds, in which grains can survive and be inserted in the surrounding intergalactic medium (Popescu et al. 2000b), could potentially change our view of the high redshifted Universe."
    },
    "0208/astro-ph0208416_arXiv.txt": {
        "abstract": "{ We have considered the thermal equilibrium in pre-protostellar cores in the approximation where the dust temperature is independent of interactions with the gas and where the gas is heated both by collisions with dust grains and ionization by cosmic rays. We have then used these results to study the stability of cores in hydrostatic equilibrium in the limit where thermal pressure dominates over magnetic field and turbulence.  We compare the density distribution derived in this manner with results obtained in the isothermal case.  We find that for cores with characteristics similar to those observed in nearby molecular clouds, the gas and dust temperatures are coupled in the core interior with densities above $\\sim 3\\times 10^4$~cm$^{-3}$.  As a consequence, one expects that the  gas temperature like the dust temperature decreases towards the center of these objects. However, the regime where gas and dust temperatures are coupled coincides approximately with that in which CO and many other molecular species deplete onto dust grain surfaces. At larger radii and lower densities, the gas and dust temperatures decouple and the gas temperature tends to the value expected for cosmic ray heating alone.  The density structure which one computes taking into account such deviations from isothermality are not greatly different from that expected for an isothermal Bonnor-Ebert sphere.  It is impossible in the framework of these models to have a stable equilibrium core with mass above $\\sim 5$~\\msun\\ and column density compatible with observed values ($N_{\\rm H} > 2\\times 10^{22}$~\\cmsq\\ or $A_{\\rm V} > 10$ mag.). We conclude from this that observed high mass cores are either supported by magnetic field or turbulence or are already in a state of collapse. Lower mass cores on the other hand have stable states where thermal pressure alone provides support against gravitation and we conclude that the much studied object B68 may be in a state of stable equilibrium if the internal gas temperature is computed in self-consistent fashion. Finally we note that in molecular clouds such as Ophiuchus and Orion with high radiation fields and pressures, gas and dust temperatures are expected to be well coupled and hence in the absence of an internal heat source, one expects temperatures to decrease towards core centers and to be relatively high as compared to low pressure clouds like Taurus. ",
        "introduction": "It has long been presumed that during the formation of a star, there is an intermediate phase in which the ``protostar'' is at least approximately in a state of hydrostatic equilibrium or magneto--hydrostatic equilibrium \\citep[see e.g.][]{sal87}.  While this idea in origin was merely based on plausibility arguments, it has received support from the discovery that in least some nearby molecular clouds, one finds embedded ``cores'' of higher density than the surroundings where the observed linewidths are thermally dominated.  That is to say, while there may be local subsonic turbulence, there is no evidence for collapse onto a point mass and it appears that thermal pressure is capable of balancing gravity.  On the other hand, comparisons of the gravitational, thermal, magnetic and turbulent energies of such cores show that all these quantities are equal to within the (considerable) uncertainties \\citep[][]{mg88a,mg88b,mal91,c99}. These data suggest though they do not prove the existence of equilibrium structures which are an intermediate state in the evolution of a {\\it prestellar core}.  This picture has received more observational support with the advent of high quality maps of the millimeter continuum emission of the dust grains within such cores as well as the possibility to study their extinction in the near and mid infrared \\citep{jwm2000,jfmm2001,bap00,llbc94}. These have allowed a more unbiased view to be obtained of the density distribution in prestellar cores. In particular, they have shown that the early molecular line data was highly biased because above a critical density of $\\sim 5\\times 10^4$ cm$^{-3}$ \\citep[see][]{tmc02,cwz02a,cwz02b,kal99} most molecular tracers including CO condense out onto dust grain surfaces.  This depletion has the consequence that in molecular lines one sees mainly a lower density outer shell, whereas the dust emission (or absorption) offers a more unbiased view of the density distribution.  The results from the millimeter continuum and infrared absorption studies have been compared with a variety of theoretical models of hydrostatic cores \\citep{bap00}. One result of such studies has been evidence for a ``flattening'' in the density distribution for radii below a critical value of $r_{\\rm cr}\\simeq 2000$-8000 AU implying a steeper fall off in radius above $r_{\\rm cr}$. Thus for example, \\citet{bap00} model L1544 with a roughly uniform density inside 1900 AU but a rapid fall off outside this radius.  Structures  whose support against gravitational collapse is mainly due to the magnetic field are plausible both because the observed dust continuum maps show large departures from spherical symmetry and because of time scale arguments.  Collapse on a free fall time scale would produce a larger star formation rate than that observed.  There are however some cores where thermal pressure may dominate magnetic pressure and spherical symmetry may be a good assumption. Particularly worthy of note is B68 where \\citet{al01} have recently demonstrated that the density distribution derived from their NIR measurements is consistent with a purely thermally supported hydrostatic model. In fact, they find results consistent with the equilibrium structures discussed by \\citet{e55} and \\citet{b56}, which we will refer to in the following as Bonnor-Ebert spheres. To what extent B68 is exceptional is presently unclear but we note that a large number of cores in the Ophiuchus and Orion clouds \\citep{jwm2000,jfmm2001} appear to have characteristics compatible with Bonnor-Ebert spheres at temperatures of 15--30~K.  This accord between theoretical expectation and observation suggests that a study of the theoretical assumptions may be worthwhile.  One of these assumptions has been that of isothermality which is often based on the concept of ``low optical depth for the cooling transitions''. In reality, the gas in prestellar cores is thought to be cooled mainly by optically thick CO transitions \\citep{g01} although, as mentioned above, the CO seems to disappear at high densities.  Moreover, the typical density for which dust grains and gas become thermally coupled is roughly of the same order as that observed in prestellar cores \\citep[see, e.g.][]{kw84}.  Thus, we decided that a new look at the gas temperature distribution to be expected in such cores seemed warranted.  In this paper we replace the assumption of an isothermal gas with the more realistic condition of thermal balance in the gas, and we evaluate the consequences of an external heating source (the interstellar radiation field) on the structure of the cloud (density and temperature profiles) and its stability properties.  Recent studies of the dust temperature distribution in such objects \\citep{eal01,zwg01} have shown that the dust temperature typically falls by a factor of 2 from edge to center. It seems reasonable to ask how the gas temperature will react in such circumstances. One might also ask  whether the density distribution in hydrostatic equilibrium will depart appreciably from that expected under the isothermal assumption. Will the expected density contrast differ for example from that expected for a Bonnor-Ebert sphere when one calculates the gas temperature in self--consistent fashion?  This article represents an attempt to answer such questions.  The outline of this paper is as follows. In Sect.~2 we give a brief introduction to the theory of structures in hydrostatic equilibrium including the results for an isothermal equilibrium Bonnor-Ebert sphere. In Sect.~3 we discuss the input to our calculations and the simplifications which we have made. In Sect.~4, we present our results for the gas temperature distribution in two model cores whose density distribution has been assumed similar to that observed in L1544 and B68. In Sect.~5 we present our results for non-isothermal hydrostatic equilibria for a variety of conditions.  Here we show among other things that the equilibria obtained depend relatively sensitively on the external radiation field. In Sect.~6 we discuss the observational consequences of our results, and compare the properties of our model clouds to isentropic polytropes. In Sect.~7 we summarize our conclusions. ",
        "conclusions": "We have examined the gas temperature distribution to be expected in interstellar pre--protostellar cores heated by the external ISRF. We find that when (as in observed cores), the central density exceeds $3 \\times 10^4$ \\percc , there is coupling between the gas and dust temperatures and hence the gas temperature (like the dust) decreases with decreasing radius.  At larger radii and smaller densities, the dust and gas decouple and the gas temperature may (for low external pressures as in the Taurus cloud) decrease towards the values expected for heating by galactic cosmic rays of around 10~K. The region where gas and dust temperatures are coupled is somewhat interior to the region where CO is highly depleted.  However, CO depletion does not seem greatly to affect the temperature distribution mainly because cooling by gas--grain collisions becomes dominant.  The observed values of the temperature of around 10~K in many pre-stellar cores allow limits to be placed on the cosmic ray ionization rate similar to the standard value of order $10^{-17}$ s$^{-1}$ based on the measured cosmic ray flux. The fact that measured gas temperatures in cores show so little spatial variation \\citep{bm89,tmc02} suggests to us that the observed thermometers mainly trace layers  of moderate depletion where gas--grain coupling is not playing an important role.  We have also examined the consequences of such a temperature distribution for the density distribution in hydrostatic equilibrium cores. The changes caused by the ``real temperature distribution'' are minor and the characteristics of a ``marginally stable Bonnor-Ebert sphere'' are similar to those in the isothermal case.  An interesting point  which emerges from these calculations is that the temperature in the core nucleus is sensitive to the external radiation field (because the core nucleus as a rule is heated by small particle MIR emission from the borders of the surrounding cloud). This is in particular the case for cores with high external pressure (and hence high density) such as those studied by ~\\citet{jfmm2001,jwm2000} in the Orion B and $\\rho$Oph clouds where $p_{\\rm ext}$ appears to be above $10^6$ \\percc K. In these cases, we expect the dust temperature to follow the gas temperature and thus the behavior should be roughly like a negative index polytrope. The external radiation fields are also higher in Orion and Ophiuchus than in Taurus and thus the core masses may also rise somewhat. For constant external pressure, the mass of the marginally stable Bonnor-Ebert sphere rises with about the 0.35 power of the external radiation field. It would be useful to have direct temperature estimates to confirm these expectations.  Another result of this study is that high mass thermally supported cores are incompatible with high column density. Thus it is difficult to imagine observed high column density (greater than $10^{22}$ \\cmsq) high mass (greater than 10~\\msun) cores going through a series of quasi-equilibrium states en route to collapse. Hence we suspect that higher mass cores  are either not stable or have other (magnetic) means of support.  These other means of support become apparent observationally both because observed cores show large departures from spherical symmetry and also because the observed line widths in most cores  can only be explained in terms of turbulence and often of supersonic turbulence.  Thus thermally supported cores may be the exception rather than the rule. It is significant nonetheless that cores in regions such as Taurus where predominantly low mass star formation is taking place have in general line profiles showing a large component of thermal broadening. This suggests that where (as in Taurus), the pivotal state may be a core with predominantly thermal support, only low mass stars are likely to form.  In clouds such as Ophiuchus and Orion with higher radiation fields and pressures and with cores having predominantly non--thermal support, higher mass stars may become possible and star formation may take another course.  It is also significant that occasionally, one finds cases like B68 where the data are consistent with hydrostatic equilibrium (marginally stable) and pure thermal support. In fact, we find that when one takes the gas temperature dependence into account, B68 is marginally stable. But irrespective of whether this is true or not, B68 gives every sign of being close to the pivotal state from which protostellar collapse will commence."
    },
    "0208/astro-ph0208146_arXiv.txt": {
        "abstract": "We report on unprecedented short-term variations detected in the optical flux from the black hole binary system, V4641 Sgr.  Amplitudes of the optical fluctuations were larger at longer time scales, and surprisingly reached $\\sim 60\\%$ around a period of $\\sim$10 min.  The power spectra of fluctuations are characterized by a power law ($\\propto f^{-\\alpha},\\; \\alpha \\sim -1.7$).  It is the first case in black hole binaries that the optical emission was revealed to show short-term and large-amplitude variations given by such a power spectrum.  The optical emission from black hole binaries is generally dominated by the emission from the outer portion of an accretion disc. The rapid optical fluctuations however indicate that the emission from an inner accretion region significantly contributes to the optical flux.  In this case, cyclo-synchrotron emission associated with various scales of magnetic flares is the most promising mechanism for the violently variable optical emission. ",
        "introduction": "In many steller-mass black holes, the X-ray flux rapidly oscillates in various time-scales (\\cite{lew95XB}; \\cite{wei97CygX1}; \\cite{rut99BHCQPO}).  The short time-scale variations are generally believed to originate from the inner accretion flow onto a black hole. Hence, X-ray fluctuations have received much attention for investigating the physics of the accretion flow and the black hole itself (\\cite{che94BXCQPO}; \\cite{wei98BHspin}; \\cite{hom01J1550}; \\cite{bai01nature}).  Here we report on short-term variations detected in the optical range, at which black hole binaries had been believed to be relatively calm. The object is called as V4641 Sgr ($V=13.8\\;{\\rm mag}$ at quiescence).  It forms a binary system with an orbital period of 2.8 d, containing a normal star of 5--8 solar-mass ($M_{\\odot}$) and a black hole of $\\sim 10 M_{\\odot}$ which accretes the overflowing gas from the normal star (\\cite{oro01v4641sgr}).  This object first received much attention in 1999 September, when it experienced a quite luminous, but unexpectedly short outburst (\\cite{uem02v4641}; \\cite{smi99v4641}).  Another noteworthy feature of this system is highly relativistic jets which were associated with this outburst (\\cite{hje00v4641}).  Radio observations detected superluminal motion of jets with an apparent velocity of $\\gtrsim 9.5c$ (\\cite{hje00v4641}; \\cite{oro01v4641sgr}).  These features make V4641 Sgr an outstanding black hole X-ray transient.  The mechanisms of its outburst and highly relativistic jet are poorly understood.  The next major outburst occurred in 2002 May, during which we detected rapid optical fluctuations (\\cite{mar02iauc}).  \\begin{figure*} \\begin{center} \\FigureFile(170mm,170mm){figure1.ps} \\end{center} \\caption{Light curves of V4641 Sgr during the outburst in 2002 May.  a: The whole light curve of the outburst.  The abscissa and ordinate denote the date and $R_{\\rm c}$ magnitude, respectively.  The gray triangles denote the visual estimation reported to Variable Star Network (VSNET; http://www.kusastro.kyoto-u.ac.jp/vsnet/).  The black, green, and blue points denote our CCD time-series observations.  b, c, d, and e: Light curves on May 24 (b, c) and 25--26 (d, e) showing short-term fluctuations.  Errors of the points are 0.02--0.05 mag in these time-series observations, and their typical errors are indicated in Figure 1c and 1e.} \\label{fig:lc02} \\end{figure*} ",
        "conclusions": "It is quite unusual that short-term and large-amplitude variations, as shown in Figure 1, were detected in the optical range which is a long wavelength region compared with the X-ray range.  In general, the optical emission from black hole binaries is considered to be the thermal emission from the outer, low temperature portion of the disc (\\cite{hyn98J1655multiwavelength}; \\cite{hyn02j1859}).  To date, ordinary black hole binary binaries sometimes showed the optical-flux variations, called ``superhumps'' (\\cite{odo96BHXNSH}).  They can however be explained by modulations at the tidally-distorted outer accretion disc (\\cite{has01BHXNSH}; \\cite{whi88tidal}).  We can naturally understand that variations originated from relatively low temperature region are detected in the optical range.  On the other hand, the optical emission in V4641 Sgr during the 2002 outburst has completely different features which were never observed in such ordinary systems.  The time-scale of fluctuations is much shorter than the orbital period of V4641 Sgr ($\\sim 2.8$ d) and at the outermost part of the accretion disc.  They hence indicate that the optical emission originates from the inner region of the accretion disc where the matter orbits with shorter time-scales and in high-temperature.  The striking similarity to X-ray modulations in GRS 1915+105 also reminds us of inner, short-time scale events.  Another noteworthy feature during the 2002 outburst of V4641 Sgr is an extremely small X-ray/optical flux ratio.  Observations with Rossi X-ray Timing Explorer (RXTE) revealed that the object was active also in X-ray during the optical outburst and reached a peak of about 60 mCrab on May 24.55 \\footnote{$\\langle$http://lheawww.gsfc.nasa.gov/users/swank/v4641sgr/$\\rangle$} (\\cite{mar02iauc}).  The X-ray/optical flux ratio of V4641 Sgr is calculated to be an order of 1, although those of typical systems are $\\sim 500$ (\\cite{tan96XNreview}).  Such a small X-ray/optical flux ratio means a peculiar energy spectrum in the X-ray--optical range, in which the optical emission is exceptionally high.  The extremely small X-ray/optical flux ratio indicates weak contribution of thermal emission from an optically-thick disc.  Instead of thermal disc emission, the synchrotron emission might be acceptable for a dominant source of the optical emission.  During the outburst, the inverted spectrum was reported by radio observations, which strongly indicates a strong contribution from synchrotron emission in the radio range \\footnote{$\\langle$http://www.atnf.csiro.au/people/rsault/astro/v4641/$\\rangle$} \\footnote{$\\langle$http://vsnet.kusastro.kyoto-u.ac.jp/vsnet/Mail/vsnet-campaign-v4641sgr/msg00051.html$\\rangle$}. It is possible that that the strong synchrotron emission dominates even in the infrared--optical range.  Under magnetic fields $B\\sim 10^6$--$10^7$ G, theoretical calculations indicate that the synchrotron emission from accretion discs can be its maximum at the optical range, and furthermore, simulated energy spectra can yield extremely small X-ray/optical ratio (1--10), as in V4641 Sgr (\\cite{mer00j1118}; \\cite{mar01j1118}).  The observed colours also support a significant contribution of synchrotron emission at the optical range.  Our multi-colour photometry on May 24 yielded de-reddened colours of the highly variable component of $B-V=-0.177\\pm 0.039$, $V-R_c=-0.092\\pm 0.039$, and $R_c-I_c=0.004\\pm 0.026$ ($E(B-V)=0.32$ assumed; \\cite{oro01v4641sgr}).  It should be noticed that the colour of $R-I$ is exceptionally red compared with the $B-V$ and the $V-R_c$.  This means that the $I_c$-band flux is considerably in excess of that expected from the $B-V$ and the $V-R_c$. Such a energy spectrum cannot be explained only with thermal disc emission.  The synchrotron emission can be a strong candidate to reconcile these colours.  The above scenario can moreover provide a consistent picture for short-term fluctuations.  The paradigm of cyclo-synchrotron emission from magnetic flares has been recently discussed for a strong candidate of short-term fluctuations in the black hole binaries (\\cite{mer00j1118}; \\cite{kan01nature}).  Various scales of magnetic flares in accretion discs can produce various scales of flux variations of the cyclo-synchrotron emission.  The fluctuations from such flares are reported to produce power spectra with a power law of $\\alpha\\sim 2$ (\\cite{min95lowstatedisk}; \\cite{kaw00BHADfluctuation}).  This is acceptable for $\\alpha$ observed in X-ray emission from ordinary black hole binaries ($\\alpha=1$--$2$) (\\cite{lew95XB}; \\cite{vanderkli89QPOreview}), and also, in optical emission from V4641 Sgr.  The cyclo-synchrotron emission mentioned above is thus the most promising interpretation of the optical seconds--minutes order fluctuations in V4641 Sgr.  While radio observations showed no evidence of short-term fluctuations \\footnote{$\\langle$http://vsnet.kusastro.kyoto-u.ac.jp/vsnet/Mail/alert7000/msg00351.html$\\rangle$}, it can be naturally understand by considering emission sources different from the optical one, for example, ejected clouds or optically-thick post-shock jet (\\cite{mar01j1118}).  The synchrotron emission from jets is also a possible source of the rapid optical variations although no evidence of jets has been reported during the outburst in 2002 May (\\cite{mir98grs1915}).  V4641 Sgr is proposed to be a ``microblazar'' rather than a ``microquasar'' since the jets observed during the 1999 outburst have been proposed to have a largest bulk Lorentz factor of $Gamma\\gtrsim 9.5$ among known galactic sources (\\cite{oro01v4641sgr}; \\cite{mir99jet}).  If the large $Gamma$ is caused by a low inclination of the system, the rapid, large-amplitude fluctuations in the 2002 May outburst may be explained with the Doppler boosting in the jets.  In this case, however, the lack of radio short-term fluctuations may be a more serious problem since we may need another synchrotron source which should be steady and dominant at the radio range.  A noteworthy advantage of V4641 Sgr is its apparent brightness ($V=13.8$) which we can easily observe even with small telescopes, compared with the other objects, for example, GRS 1915+105 ($V>19$) and XTE J1118+480 ($V=18$--$19$).  In a number of black hole binaries, the optical--UV flux is heavily obscured by intersteller extinction.  On the other hand, it can be the most essential wavelength when the synchrotron emission is dominant because its peak frequency can lie at the optical range (\\cite{mer00j1118}; \\cite{mar01j1118}).  V4641 Sgr is a unique object which shows fluctuations caused by the magnetic activity detected in the optical range.  In future, simultaneous observations from X-ray to radio will been performed and enable us to study the short-term fluctuations in multi-wavelengths.  Our discovery of these new features of V4641 Sgr will thus lead a revolutionary advance in our understanding of the magnetic activity in black hole accretion discs.  \\vskip 3mm  We are pleased to acknowledge comments by T. Kawaguchi. We are grateful to many amateur observers for supplying their vital visual CCD estimates via VSNET.  This work is partly supported by a grant-in aid (13640239) from the Japanese Ministry of Education, Culture, Sports, Science and Technology.  Part of this work is supported by a Research Fellowship of the Japan Society for the Promotion of Science for Young Scientists (MU)."
    },
    "0208/gr-qc0208047_arXiv.txt": {
        "abstract": "We study cosmological braneworld models with a single {\\em timelike\\/} extra dimension. Such models admit the intriguing possibility that a contracting braneworld experiences a natural bounce without ever reaching a singular state. This feature persists in the case of anisotropic braneworlds under some additional  and not very restrictive assumptions. Generalizing our study to braneworld models containing an induced brane curvature term, we find that a FRW-type singularity is once again absent if the bulk extra dimension is timelike. In this case, the universe either has a non-singular origin or commences its expansion from a quasi-singular state during which both the Hubble parameter and the energy density and pressure remain finite while the curvature tensor diverges. The non-singular and quasi-singular behaviour which we have discovered differs both qualitatively and quantitatively from what is usually observed in braneworld models with spacelike extra dimensions and could have interesting cosmological implications. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0208/astro-ph0208236_arXiv.txt": {
        "abstract": "{ We have pursued two different methods to analyze the old stellar population near the Galactic plane, using data from  the 2MASS survey. The first method is based on the isolation of the red clump giant population in the color-magnitude diagrams and the inversion of its star counts to obtain directly the density distribution along the line of sight. The second method fits the parameters of a disc model to the star counts in 820 regions. Results from both independent methods are consistent with each other. The qualitative conclusions are that the disc is well fitted by an exponential distribution in both the galactocentric distance and height. There is not an abrupt cut-off in the stellar disc (at least within $R<15$ kpc). There is a strong flare (i.e. an increase of scale-height towards the outer Galaxy) which begins well inside the solar circle, and hence there is a decrease of the scale-height towards the inner Galaxy. Another notable feature is the existence of a warp in the old stellar population whose amplitude is coincident with the amplitude of the gas warp.\\\\ It is shown for low latitude stars (mean height: $|z|\\sim 300$ pc) in the outer disc (galactocentric radius $R> 6$ kpc) that: the scale-height in the solar circle is $h_z(R_\\odot)=3.6\\times 10^{-2}R_\\odot$, the scale-length of the surface density is $h_R=0.42R_{\\odot}$ and  the scale-length of the space density in the plane (i.e. including the effect of the flare) is $H=0.25R_{\\odot}$. The variation of the scale-height due to the flare follows roughly a law $h_z(R)\\approx  h_z(R_\\odot) \\exp \\left(\\frac{R-R_\\odot}{[12-0.6R({\\rm kpc})]\\ {\\rm kpc}}\\right)$ (for $R<\\sim 15$ kpc; $R_\\odot=7.9$ kpc). The warp moves the mean position of the disc to a height $z_w=1.2\\times 10^{-3} R({\\rm kpc})^{5.25}\\sin (\\phi +5^\\circ )$ pc (for $R<\\sim 13$ kpc; $R_\\odot=7.9$ kpc). ",
        "introduction": "Star counts have been widely used for the study of Galactic structure (see Paul 1993), and are becoming increasingly important with the appearance of wide area surveys in the last decades (see for example Bahcall 1986, Majewski 1993, Reid 1993, Garz\\'on et al. 1993, Price 1988). The use of progressively larger and more sensitive samples  combined with detailed models of stellar galactic distribution (for example Bahcall \\& Soneira, 1980, Wainscoat et al. 1992) has overcome most of the original uncertainties and difficulties in the analysis and interpretation of star count data. The  advent of NIR detector arrays has permitted the detailed exploration of the stellar structure in hitherto hidden areas of the Milky Way, such as the Galactic Plane and Bulge (Eaton et al. 1984, Habing 1988, Garz\\'on et al. 1993, Hammersley et al. 1994, Ruphy et al. 1996, L\\'opez--Corredoira et al. 2000, Epchtein 1997, Skrutskie et al. 1997).  This is because at these wavelengths interstellar extinction is significantly less when compared to the visible, while keeping the individual stellar contribution to the observed flux.  There are, however, still controversial or totally unknown parameters in the description of the detailed stellar structure. Some of these are concerned with the radial and vertical distribution of the Galactic disc, and its specific morphology. Several radial scale-lengths have been determined by different groups, even when  using the same data sets. As a general trend, the shorter scale-lengths are found for the surveys using longer wavelengths (Kent et al. 1991), although it could equally be that different source type or regions of the Galaxy are being examined.  There is far less debate on the vertical scale-height, with the canonical value for the old population in the Solar vicinity being about 300 pc. Still more work is needed, however, to differentiate between the distribution being an exponential  or the $sech^2$ (van der Kruit 1988). Finally, features in the stellar disc like internal (Freudenreich 1998, L\\'opez-Corredoira et al. 2001) or external (Habing 1988, Ruphy et al. 1996) cut--offs radii, flares (Kent et al. 1991, Alard 2000), warps (Djorgovski \\& Sosin 1989, Freudenreich 1998, Porcel \\& Battaner 1995, Alard 2000, Drimmel \\& Spergel 2001) and/or local corrugations of the galactic plane (Hammersley et al. 1995) are less well studied, and are more controversial.  In this paper we will address some of the above topics by analyzing the shape of the stellar Galactic disc in two ways. We will first extract a single stellar type by using the deep NIR color--magnitudes diagrams (CMD) from the 2MASS survey (Skrutskie et al. 1997). The analysis will be based on the red clump giants, whose relatively high density and  bright absolute magnitude  give rise to an identifiable feature in the CMDs. Hence,  the distance and extinction to each star can be directly determined only assuming  the absolute magnitude and color of the sources.  As a second approach, we will use the star counts, also taken from 2MASS, and the results will then be  compared with those obtained from the red clump method. Both methods will be used to examine the scale-length, scale-height, flare, outer disc cut-off and warp.  The data for this work has been taken from the 2nd release of the 2MASS project (Skrutskie et al. 1997, http://www.ipac.caltech.edu/2mass/releases/docs.html). We have made use of the available data in that second release for near plane regions in the outer Galaxy ($45^\\circ <l<315^\\circ$). ",
        "conclusions": "In this paper we examine both quantitatively and qualitatively the large-scale structures in the Galactic disc. Two independent methods are applied to the 2MASS data. The first isolates red clump giants in a number of regions near the plane, the counts are then  inverted to directly obtain the disc density. The second fits the parameters of a disc model  in 820 independent near-plane regions. Results from both  methods are consistent with each other implying that both methods are valid. For instance, magnitude and color of the red clump giants must be close to the correct values otherwise there would be significant difference between the two methods for  $H$ and $h_z(R_\\odot)$.  The distribution of sources within the disc is well described by an exponential distribution in both the galactocentric distance and height above the plane. There is a strong flare, i.e. an increase of scale-height towards the outer Galaxy. This flare starts well inside the solar circle, hence there is a decrease of the scale-height towards the inner Galaxy. This is consistent with the observation that the large majority of galaxies have flares (de Grijs \\& Peletier 1997) and theoretically consistent with some scenarios of self-graviting stellar discs (Narajan \\& Jog 2002). The flare removes the need to invoke a cut-off in the outer Galaxy, and certainly there is no evidence for an abrupt cut-off in the stellar disc, to at least $R<15 $~kpc (=4.5$h_R$). For the analysis it was asssumed that the red clump luminosity function could be described as a delta function and whilst it is shown that this could have some minor effect on the result, it could not mask a cut off.  In other spiral galaxies, the question of the cut-off is still not very clear. Pohlen et al. (2002) find that typical radial profiles are better described by a two-slope exponential profile, characterized by an inner and outer scale-length separated at a break radius, rather than a sharply-truncated exponential model. Florido et al. (2001), however, argue that  there are some cases of abrupt truncation in the near infrared emission of the stellar profile. Kregel et al. (2002) finds that 60\\% of the spiral galaxies are truncated, although he also finds that the truncation is beyond $4.4h_R$ for galaxies with $h_R<4$ kpc; so probably, if there existed a cutoff in our Galaxy, it would not be detected in $R<4.5h_R$.  Another noticeable feature is the warp in the old stellar population whose amplitude is coincident with the amplitude of the gas warp. Figs. \\ref{Fig:figpac1} and \\ref{Fig:figpac2} show the form of the disc. S\\'anchez-Saavedra et al. (1990, 2002) or Reshetnikov \\& Combes (1998) claim that most spiral galaxies are warped at optical wavelengths, but it is not easy to disentangle the stellar and gaseous warp.  \\begin{figure}[!h] \\begin{center} \\vspace{1cm} \\mbox{\\epsfig{file=MS2213f22.ps,height=6cm}} \\end{center} \\caption{Cut in the plane YZ of the Galaxy as described by the final disc model.} \\label{Fig:figpac1} \\end{figure}  \\begin{figure}[!h] \\begin{center} \\vspace{1cm} \\mbox{\\epsfig{file=MS2213f23.ps,height=7cm}} \\end{center} \\caption{Isodensity surface with $\\rho =0.005$ star/pc$^3$ as described by the final disc model. Vertical scale multiplied by a factor 5.} \\label{Fig:figpac2} \\end{figure}  Various parameters of the old stellar disc have been measured based on an analysis over  galactocentric distances between 6 kpc and 15 kpc. The study of the inner disc would be somewhat more complicate, especially in the plane, because the disc component is no longer isolated and it is mixed with other possible components: bulge, bar, ring, etc. As is shown in L\\'opez-Corredoira et al. (2001), the disc might be a Freeman type II disc and  inside about 3.5 kpc the disc would be truncated; this deserves another detailed study which is beyond the scope of this paper. At present we can talk about the distribution at $R>5-6$ kpc and representative of the mean old disc at low latitudes (with mean $|z|$ around 300 pc). If there exists a thick disc, the results for higher latitudes would be substantially different. In the following are presented the results from the star counts, since they are potentially more accurate. \\begin{itemize}  \\item The scale-height in the solar circle is $h_z(R_\\odot )=285^{+8}_{-12}$ pc.  \\item The scale-length of the space density in the plane is $H=1.97^{+0.15}_{-0.12}$ kpc.  \\item the scale-length of the surface density is $h_R=3.3^{+0.5}_{-0.4}$ kpc.  \\end{itemize}  The errors do not include systematic uncertainties, which we know to be lower than a 10\\%. For the flare and the warp:  \\begin{itemize}  \\item The variation of the scale-height due to the flare follows roughly a law $h_z(R)\\approx h_z(R_\\odot) e^{\\frac{R-R_\\odot}{(12-0.6R(kpc))\\ {\\rm kpc}}}$ (for $R<\\sim 15$ kpc). \\item The warp moves the mid plane of the disc to the height $z_w=1.2\\times 10^{-3} R(kpc)^{5.25}\\sin (\\phi +5^\\circ )$ pc (for $R<\\sim 13$ kpc), where $\\phi $ is the galactocentric angle ($\\phi _\\odot=0$). \\end{itemize}  The results presented here constitute an important step in constraining of the disc structure, especially for the stellar populations in outer disc, which were previously not very well known. Theoretical implications are not discussed here but it is expected that these results will constitute a valuable test for many theories about the formation and evolution of the galactic discs, as well as studies about the formation of warps, flares.   Acknowledgments: This publication makes use of data products from 2MASS, which is a joint project of the Univ. of Massachusetts and the Infrared Processing and Analysis Center, funded by the NASA and the NSF. We gratefully acknowledge the anonymous referee whose detail report has helped improve the paper. Thanks are also given to O. E. Gerhard, N. Bissantz and C. Alard for helpful comments."
    },
    "0208/astro-ph0208014_arXiv.txt": {
        "abstract": "Trigonometric parallaxes have been measured by Dahn et al.\\ (2002) for 28 cool dwarfs and brown dwarfs, including 17 L dwarfs and three T dwarfs.  Broadband CCD and near-IR photometry ($VRIz^*JHK$) have been obtained for these objects and for 24 additional late-type dwarfs. These data have been supplemented with astrometry and photometry from the literature, including parallaxes for the brighter companions of ten L and two T dwarfs.  The absolute magnitudes and colors are reviewed here.  The $I-J$ color and the spectral type are both good predictors of absolute magnitude for late-M and L dwarfs. $M_J$ becomes monotonically fainter with $I-J$ color and with spectral type through late-L dwarfs, then brightens for early-T dwarfs. In contrast, the $J-K$ color correlates poorly with absolute magnitude for L dwarfs.  Using several other parameters from the literature (Li detection, H$\\alpha$ emission strength, projected rotation velocity, and tangential velocity), we fail to uncover any measurable parameter that correlates with the anomalous $J-K$ color. ",
        "introduction": "Parallaxes of late-M, L, and T dwarfs have been measured from images taken over the past several years with the 1.55 m Strand astrometric telescope at the U.S. Naval Observatory.  The present results have been published recently by Dahn et al. (2002).  The parallaxes are necessary for many purposes, including comparison with evolutionary models, analyzing the kinematics, determining ages, determining temperatures, and identifying outlying objects (perhaps binaries or young objects). When faint companions to bright stars with Hipparcos parallaxes are included, there are now nearly 30 field L dwarfs with measured and published distances, but only five T dwarfs published to date. In this paper, we discuss two applications of the parallax data to our understanding of L dwarfs. ",
        "conclusions": ""
    },
    "0208/astro-ph0208478_arXiv.txt": {
        "abstract": "We present observations of the UV absorption lines in the luminous Seyfert 1 galaxy Mrk 509, obtained with the medium resolution ($\\lambda$/$\\Delta\\lambda$ $\\approx$ 40,000) echelle gratings of the Space Telescope Imaging Spectrograph on the {\\it Hubble Space Telescope}. The spectra reveal the presence of eight kinematic components of absorption in Ly$\\alpha$, C~IV, and N~V, at radial velocities of $-$422, $-$328, $-$259, $-$62, $-$22, $+$34, $+$124, and $+$ 210 km s$^{-1}$ with respect to an emission-line redshift of z $=$ 0.03440, seven of which were detected in an earlier {\\it Far Ultraviolet Spectrographic Explorer (FUSE)} spectrum. The component at $-$22 km s$^{-1}$ also shows absorption by Si~IV. The covering factor and velocity width of the Si~IV lines were lower than those of the higher ionization lines for this component, which is evidence for two separate absorbers at this velocity. We have calculated photoionization models to match the UV column densities in each of these components. Using the predicted O~VI column densities, we were able to match the O~VI profiles observed in the {\\it FUSE} spectrum. Based on our results, none of the UV absorbers can produce the X-ray absorption seen in simultaneous {\\it Chandra} observations; therefore, there must be more highly ionized gas in the radial velocity ranges covered by the UV kinematic components. ",
        "introduction": "Since the launch of the {\\it International Ultraviolet Explorer (IUE)}, it has been known that the UV spectra of Seyfert 1 galaxies show absorption lines intrinsic to their nuclei (Ulrich 1988). With the advent of the {\\it Hubble Space Telescope (HST)}, it is now understood that intrinsic absorption is a common phenomenon, present in more than half of the well-studied Seyfert 1 galaxies (Crenshaw et al. 1999). Among those Seyferts that show absorption, high ionization lines such as N~V $\\lambda\\lambda$1238.8, 1242.8 and C~IV $\\lambda\\lambda$1548.2, 1550.8 are always present, along with Ly$\\alpha$, while lower ionization lines, such as Si~IV $\\lambda\\lambda$1393.8, 1402.8 and Mg~II $\\lambda\\lambda$2796.3, 2803.5, are less common. Typically, the absorption lines are blueshifted (by up to 2100 km s$^{-1}$) with respect to the systemic velocities of the host galaxies, indicating net radial outflow. Although there are examples of strong UV absorption lines near systemic velocities, some are found in highly inclined host galaxies and, therefore, are most likely formed in gas within the plane of the host galaxy (Crenshaw et al. 2001; Crenshaw et al. 2002). Among the blue-shifted absorbers, the ionic columns are highly variable, which may be the result of changes in response to the ionizing continuum (cf. Krolik \\& Kriss 1997; Shields \\& Hamann 1997) or transverse motion (Crenshaw \\& Kraemer 1999). Variability is suggestive of the proximity of the absorbers to the central active nucleus, since it may result from the high densities, hence short recombination timescales, or high transverse velocities, similar to those inferred for the emission-line gas close to the active nucleus. Based on density constraints, it has been shown that in at least two sources, NGC 4151 and NGC 3516, much of the absorbing gas may lie within a fraction of a parsec from the central source (Kraemer et al. 2001a; Kraemer et al. 2002). Another indication of small radial distances is the low line-of-sight covering factors derived from some UV absorption lines (Kraemer et al. 2002; Gabel et al. 2002). In addition to the UV absorbers, the presence of intrinsic absorption, typically in the form of bound-free edges of O~VII and O~VIII, has been detected in the X-ray spectra of a similar fraction of Seyfert 1 galaxies (Reynolds 1997; George et al. 1998). Most recently, spectra obtained with the {\\it Chandra X-ray Observatory (CXO)} have revealed that X-ray absorption lines associated with this material are also blue-shifted (Kaastra et al. 2000; Kaspi et al. 2000, 2001). Although it has been argued that some fraction of the UV absorption arises in the same gas responsible for the X-ray absorption (Mathur, Elvis \\& Wilkes 1995, 1999; Crenshaw \\& Kraemer 1999; Kriss et al. 2000; Kraemer et al. 2002), the connection between the X-ray and UV absorption is complex. In fact, there is often a wide range in ionization states in gas which may be at the same radial velocities (Kaspi et al. 2002). Further evidence for this is the co-spatial X-ray and optical line emission seen in the narrow-line regions (NLR) of Mrk 3 (Sako et al. 2000), NGC 1068 (Ogle et al. 2002), and NGC 4151 (Ogle et al. 2000).  Mrk 509 is a highly luminous (L$_{h\\nu~>~13.6 eV}$ $\\sim$ 10$^{45}$ erg s$^{-1}$; see Kriss et al. [2000]; Yaqoob et al. [2002a]) Seyfert 1 galaxy. Phillips et al. (1983) determined that the nucleus is surrounded by two distinct components of extended ionized gas: a low ionization component, with a velocities indicative of rotation in the galactic disk, and a high ionization component, with velocities blue-shifted with respect to those in the disk. Phillips et al. suggested that this high ionization component is part of an outflowing shell of gas. If gas is distributed within a bicone centered on the active nucleus, as appears to be the case in other Seyfert galaxies (Crenshaw \\& Kraemer 2000; Crenshaw at al. 2000; Ruiz et al. 2001), the axis of the outflow must be roughly parallel to our line-of-sight (see Fig. 8 in Phillips et al.).  There is extended broad Balmer line emission (FWZI $>$ 15,000 km s$^{-1}$), presumably scattered light from the unresolved broad line region, in an axisymmetrical distribution surrounding the active nucleus (Mediavilla et al. 1998), consistent with the geometry suggested by Phillips et al. (1983). Based on recent X-ray observations of Mrk 509 with {\\it XMM-Newton}, Pounds et al. (2001) argued that the inclination of the putative accretion disk within the active nucleus is $<$ 30\\deg, which also fits with the proposed geometry.  Two kinematic components of intrinsic Ly$\\alpha$ and C~IV absorption, at $\\sim$ $-$420 and $+$40 km s$^{-1}$, were detected by York et al. (1984) in high-dispersion {\\it IUE} spectra of Mrk 509. In their {\\it HST}/Faint Object Spectrograph observations $\\sim$ 12 yr later, Crenshaw, Boggess, \\& Wu (1995) found the same components, as well as N~V absorption at these velocities. York et al. (1984) and Crenshaw et al. (1995) suggested that these lines formed in extended regions of ionized gas in the host galaxy. Kriss et al. (2000) obtained a high resolution ($\\lambda$/$\\Delta\\lambda$ $\\approx$ 15,000) {\\it Far Ultraviolet Spectroscopic Explorer (FUSE)} spectrum over the wavelength range 915 -- 1185 \\AA\\ on 1999 November 9, 11 that shows intrinsic absorption in the lines of O~VI $\\lambda\\lambda$1031.9, 1037.6, C~III $\\lambda$977.0, and the H~I Lyman lines, consisting of seven, relatively narrow (FWHM $\\leq$ 63 km s$^{-1}$), kinematic components which are clustered into two groups, at $-$370 km s$^{-1}$ and near the systemic velocity. They suggested that the former was associated with the extended, blue-shifted ionized gas observed by Phillips et al. (1983). In addition to the UV absorption, {\\it ASCA} spectra showed strong evidence for the presence of an X-ray warm absorber (Reynolds 1997; George et al. 1998).  We have obtained simultaneous {\\it CXO}/High Energy Transmission Grating (HETG) and {\\it HST}/Space Telescope Imaging Spectrograph (STIS) medium resolution spectra of Mrk 509 on 2001 April 13. We note that there are no previous {\\it HST} high-resolution ($\\lambda$/$\\Delta\\lambda$ $\\geq$ 10,000) UV spectra, except for a Goddard High-Resolution Spectrograph (GHRS) observation of the intrinsic L$\\alpha$ absorption (Crenshaw et al. 1999), which is highly saturated (see section 2.2). The analysis of the X-ray spectra are discussed in Yaqoob et al. (2002a) and Yaqoob et al. (2002b; hereafter Paper I). In Paper I, we describe the details of the ionizing continuum and the physical nature of the X-ray warm absorbing gas, including model predictions of ionic column densities. In the present paper, we present our analysis of the STIS spectra and the results of photoionization modeling of the UV absorbers. The paper is organized as follows: in Section 2 we describe the observations, and give the details of the measurement of the intrinsic lines, in Section 3 we detail the photoionization modeling of the absorbers, in Section 4 we discuss the implications of the result, and Section 5 gives our summary. ",
        "conclusions": "In general, intrinsic UV (Crenshaw et al. 1999) and X-ray (Kaspi et al. 2000, 2001; Kaastra et al. 2000) absorption lines detected in Seyfert 1 galaxies are blue-shifted, indicative of radial outflow. However, in the case of Mrk 509, the clustering of component velocities near systemic and the presence of red-shifted gas (although this is somewhat uncertain; see Section 2.1) differ from the more obvious cases of radial outflow discussed by Crenshaw et al. (1999). Kriss et al. (2000) suggested that this may be the result of the absorbers at radial distances of $\\sim$ 300 pc, co-located with the low ionization gas detected by Phillips et al. (1983) and moving primarily transversely to our line-of-sight. Indeed, it has been suggested that UV absorption lines form in disk-driven winds (e.g., Proga, Stone, \\& Kallman 2000). If this is the case, the low velocities we observe in Mrk 509 require a viewing angle roughly orthogonal to the direction of the flow, similar to the geometry suggested by Kriss et al. (2000). This scenario is consistent with the evidence that the accretion disk in Mrk 509 is viewed roughly face-on (Pounds et al. 2001). If the widths of the UV absorption lines detected in Seyfert 1s are the result of superposition of multiple kinematic components, viewing an outflow stream from this vantage point might help explain the narrowness of the lines seen in Mrk 509. Nevertheless, there are some problems with this hypothesis. Since Mrk 509 is a high luminosity Seyfert 1, disk-wind models predict that the outflow will be more confined to the disk than in lower luminosity AGN (Konigl \\& Kartje 1994; Murray et al. 1995). The axisymmetric distribution of the reflected broad Balmer line emission in Mrk 509 can be interpreted as evidence for an obscuring torus or wind that covers only a small solid angle (Mediavilla et al. 1998), which is consistent with a ``flattened'' wind. However, most of the UV components have covering factors near unity, which may be difficult to achieve with a flattened wind viewed against a roughly face-on disk, depending on the location of the continuum source relative to the base of the wind. Specifically, the ``launch-pad'' of the wind must be {\\it interior} to the UV continuum source.  Alternatively, it is possible that the observed velocities of the UV absorbers are close to their true radial velocities. In this case, the absorbers may not be part of a disk-driven flow and are, instead, gas that has been elevated off the disk and, therefore, may be more likely to occult the continuum and BLR when the disk is viewed face-on. Although disk-wind models include such a component, both the X-ray and UV absorbers detected in Mrk 509 are too neutral to be the ``hitch-hiker'' gas proposed by Murray et al. (1995) to shield the disk-driven winds or the highly-ionized gas in the inner part of the wind modeled by Proga et al. (2000), since, in both cases, these components may have ionization parameters $\\geq$ 10. Krolik \\& Kriss (1995) proposed that UV absorbers are high density knots embedded in a thermally expanding, highly ionized wind, which would provide a natural explanation for the kinematic association of the UV and X-ray absorbers. However, this model predicts radial velocities similar to the disk-driven winds ($>$ several hundred km s$^{-1}$), hence would still require a special geometry to explain the low radial velocities and large values of C$_{los}$. A simpler scenario would have the absorbers at large radial distances, e.g. 100's of pcs as suggested by Kriss et al. (2000). As noted in Section 3.3, similar UV absorbers were found at large distances in NGC 4151 (Kraemer et al. 2001a). Furthermore, studies of the narrow-line region kinematics indicate a strong drop in radial velocities in emission-line gas at distances $>$ few hundred pcs (Crenshaw \\& Kraemer 2000; Crenshaw et al. 2000; Ruiz et al. 2001), so it is not necessarily surprising that the absorbers within the NLR would have similar kinematics. However, better determination of the radial distances of these absorbers requires constraints on the gas density, which can only be achieved via variability studies."
    },
    "0208/astro-ph0208152_arXiv.txt": {
        "abstract": "{ Supernova 1998bu in the galaxy M96 was observed by COMPTEL for a total of 88 days starting 17 days after the explosion. We searched for a signal in the 847 keV and 1238 keV lines of radioactive $^{56}$Co from this type Ia supernova. Using several different analysis methods, we did not detect SN1998bu. Our measurements should have been sensitive enough to detect $^{60}$Co gamma-rays as predicted from supernova models. Our $2\\sigma$ flux limit is $2.3 \\cdot 10^{-5}$ photons~cm$^{-2}$~s$^{-1}$; this would correspond to 0.35 \\Msol~of ejected $^{56}$Ni, if SN1998bu were at a distance of 11.3 Mpc and transparent to MeV gamma rays for the period of our measurements. We discuss our measurements in the context of common supernova models, and conclude disfavoring a supernova event with large mixing and major parts of the freshly-generated radioactivity in outer layers. ",
        "introduction": "Despite their widespread use as 'standard candles', the physical nature of supernovae of type Ia is still not understood in terms of physical processes; therefore corrections of evolutionary aspects remain empirical \\citep{Georgii:Branch1998,Georgii:Niemeyer2000}. Type Ia supernovae are believed to be caused by thermonuclear explosions of CO white dwarfs \\citep{Georgii:Livio2000,Georgii:Hoeflich1996,Georgii:Nomoto97}. Radioactive energy of $\\simeq$0.5 \\Msol\\ of $^{56}$Ni synthesized in such explosions is considered the driver of all types of observed light from these objects. The nearby SN1998bu supernova was a unique opportunity to directly measure gamma-rays from the $^{56}$Ni decay chain with the Compton Observatory.  The dynamics of a white dwarf explosion are difficult to model due to the range of scales involved in flame ignition and propagation \\citep{Georgii:Iwamoto1999}. Theories which are most successful in describing observations of type Ia supernovae include empirical components for key aspects. Observationally type Ia supernovae are a fairly homogeneous phenomenon \\citep{Georgii:Branch1998}, suggesting a narror range of synthesized $^{56}$Ni masses. Constraints from  models of the bolometric light curve of type Ia supernovae imply that typically 0.3--0.5 \\Msol\\ of radioactive $^{56}$Ni energy are needed\\citep{Georgii:Hoeflich1996}. Observations of the supernova light curve's peak magnitude or of NIR lines of Fe[II] and Co[II] have mostly been used to derive $^{56}$Ni masses; the rather wide range of  inferred $^{56}$Ni (0.1 -- 1.14 \\Msol)\\citep{Georgii:Contardo2000} is difficult to understand if a single well-tuned process is held responsible for the supernovae of type Ia and in particular their \"standard candle\" characteristics. Among different types of models, a rather wide range of $^{56}$Ni masses of 0.1 up to 0.8 \\Msol\\ is discussed (see `Discussion' section below).  Critical parameters of the explosion models are the ignition density of the white dwarf at its core, and the transition from the early deflagration stage (sub-sonic flame propagation) into a detonation (super-sonic flame propagation)\\citep[e.g.][]{Georgii:Leibundgut2000,Georgii:Livio2000}. The former is estimated from evolutionary models of the (uncertain) progenitor and the effective binary accretion rate, involving uncertain issues such as steady or flash-like nuclear burning of the accreted H and He material, or modulation of the accretion flow from the companion through the wind of the white dwarf (at the higher accretion rates required by the lower ignition densities preferred from otherwise excessive production of neutron-rich Fe group isotopes\\citep{Georgii:Nomoto97}). The acceleration of the flame speed can only be treated as a purely empirical parameter of models at present, but critically determines the final $^{56}$Ni mass and the Fe group to lighter element ratio \\citep{Georgii:Iwamoto1999}. Three-dimensional model treatments of this flame ``micro-physics'' is promising, but still a challenging problem \\citep{Georgii:Reinecke1999}.  The fact that type Ia supernovae are rather homogeneous \\citep{Georgii:Branch1998} suggests a clear evolutionary path towards a well-definied presupernova star and a robust ignition condition. This led to the model of binary accretion of H and He rich matter onto a CO white dwarf at a well-tuned accretion rate such that H and especially He nuclear burning proceeds non-catastrophically and the white dwarf C mantle grows in mass until the Chandrasekhar mass limit is reached; the thermonuclear runaway explosion ensues from fast nuclear burning of Carbon ignited at the core due to heating from gravitational pressure and H and He shell-burning heat conduction \\citep{Georgii:Nomoto1982}. Sub-Chandrasekhar mass white dwarfs could also explode as type Ia supernovae \\citep{Georgii:Livne1990}: The merging of two white dwarfs would disrupt the lighter of the two into a C-O envelope accreting onto the more massive white dwarf (\"double-degenerate\" model). As the accretion proceeds to exceed the Chandrasekhar limit, the white dwarf ignites C centrally as above \\citep{Georgii:Iben1984}. There are some doubts if the merging process will avoid core collapse of the merged object and produce a thermonuclear supernova; e.g. transport of rotational energy is critical \\citep{Georgii:Livio2000}. Alternatively, a single-degenerate sub-Chandrasekhar model has been proposed: A He layer built up from accretion and steady hydrogen burning as above may ignite in a flash and thus send a shock wave into the white dwarf core, adding to the gravitational heat and thus also igniting carbon in the center for a lower-mass white dwarf \\citep{Georgii:Nomoto1982,Georgii:Livne1995}. In this scenario much of the radioactive material would be produced towards the outside, resulting in different evolution of radioactivity-derived supernova light. Recent constraints on early spectra and the absence of intermediate-mass elements in the outer fast ejecta disfavor this scenario somewhat \\citep{Georgii:Livio2000}.  A deflagration model was recently favored for SN1998bu on purely spectroscopic arguments \\citep{Georgii:Vinko2001}. And the occurrence near or in a spiral arm and the observation of a light-echo \\citep{Georgii:Cappellaro2001} may suggest that the progenitor system could be younger than those of the average type Ia supernovae, therefore anomalous and igniting at a particularly low density.  With such diversity of models and the difficulties of detailed physics modeling, observations of a variety of aspects of type Ia supernovae are a key to clarify the true nature of these events.  Observations of a large sample of supernovae in UV, optical and infrared bands have been made and discussed widely. But this radiation originates from driving processes deep inside the object, the bolometric light curve and its evolution reflect the supernova envelope structure, with much less information on the core. Spectral information tells us about material mixing and the total kinetic energy. However optical photons are created long after the initial explosion; most information from the early stage of the supernova event is lost. This makes it difficult to discriminate between different explosion models or model parameters.   \\begin{figure*}[ht] \\centering \\begin{minipage}[t]{8.5cm} \\includegraphics[width=\\textwidth,bb= 5 8 185 190]{H3503F1a.eps} \\end{minipage} \\hfill \\begin{minipage}[t]{8.5cm} \\includegraphics[width=\\textwidth, bb= 5 8 185 190]{H3503F1b.eps} \\end{minipage} \\caption{The total energy spectrum (left) and the background subtracted spectrum (right) for a sample position on the grid around SN1998bu.} \\label{Georgii:F1} \\end{figure*}  Observations of  $\\gamma$-rays promise more direct information from the core of the supernova and the explosion mechanism. The radioactivity produced in the initial event decays and produces gamma-ray lines, which can be observed directly once the supernova is transparent to gamma-rays (after about 30-100 days). Even in the early stages, $\\gamma$-rays from radioactivity will escape from outer layers, their intensity depending on ejecta mixing. Differences in the predicted $\\gamma$-ray spectra have been suggested as the key observation to discriminate between models and the extent of mixing \\citep{Georgii:Hoeflich1998,Georgii:Pinto2001}. This is best observed at early times; at times later than ~30 days, differences in total $^{56}$Ni masses can mimic differences between sub- and Chandrasekhar models \\citep{Georgii:Pinto2001}.  The sensitivity of the  $\\gamma$-ray instruments on-board CGRO (OSSE and COMPTEL) limits the observations of type Ia supernovae to events  within about 15 Mpc. To date, only one event, SN1991T, was marginally detected with COMPTEL \\citep{Georgii:Morris1997}. SN1998bu provides a second opportunity for line searches in { type Ia supernovae}. Some theoretical models predict  $\\gamma$-ray line fluxes well above the sensitivity limits of COMPTEL and OSSE for an assumed distance of 11~Mpc. Although COMPTEL lacks the spectral resolution to provide unique and decisive $\\gamma$-ray line shape diagnostics, an independent proof of the radioactive $^{56}$Ni mass origin through detection of the corresponding $\\gamma$-ray line fluxes was attempted, and is important given the complexity and unknowns of conversion of radioactive energy in a supernova envelope.  \\begin{figure*}[tb] \\sidecaption \\includegraphics[width=12cm,,clip]{H3503F2.eps} \\caption{The result of the spectral analysis. The derived $2\\sigma$ upper limits are $4.1 \\cdot 10^{-5}$ photons~cm$^{-2}$~s$^{-1}$ for the 847 keV line and $2.3 \\cdot 10^{-5}$ photons~cm$^{-2}$~s$^{-1}$ for the 1238 keV line.} \\label{Georgii:Line} \\end{figure*}  On May 9.9 UT in 1998 supernova SN1998bu was discovered in the galaxy M96 (NGC 3368) \\citep{Georgii:Villi98}. From wide-band spectrograms it was classified as type Ia \\citep{Georgii:Ayani98,Georgii:Meikle1998}. From an earlier observation \\citep{Georgii:Faranda1998} and an { estimate of the} maximum blue light { at} $t_{\\rm Bmax} =$ 10952.7 $\\pm$ 0.5 TJD  (i.e.~May 19), \\cite{Georgii:Meikle1998} inferred the date of the explosion to be May 2.0 $\\pm$ 1.0 UT (i.e. TJD 10935 $\\pm$ 1). The distance to M96 had been known from HST Cepheid measurements as 11.3 $\\pm$ 0.9~Mpc \\citep{Georgii:Hjorth1997}, though another value of 9.6 $\\pm$ 0.6~Mpc had been derived from measurements of planetary nebulae \\citep{Georgii:Feldmeier1997}. SN1998bu appears to be a typical type Ia event \\citep{Georgii:Jah1999}, its reddening can be attributed to dust in the host galaxy. From its dereddened brightness it was concluded (using the methods described in \\citet{Georgii:Nomoto97} and \\citet{Georgii:Iwamoto1999}) that the total ejected Ni mass was rather typical. A value of 0.77 \\Msol\\ has been derived from analysis of the bolometric light curve \\citep{Georgii:Leibundgut2000,Georgii:Contardo2000}. ",
        "conclusions": "To compare  $\\gamma$-ray upper limits with theoretical Ni mass predictions, the distance to the SN plays an essential role. The host galaxy M96 had a HST-Cepheid-determined distance of 11.6 $\\pm$ 0.9 Mpc \\citep{Georgii:Tanvir1995}, later revised to 11.3 $\\pm$ 0.9 Mpc by \\cite{Georgii:Hjorth1997}. This makes SN1998bu one of seven SN observed in galaxies with a distance set by Cepheid measurements. We adopt this distance of 11.3 Mpc for our analysis.  A second key factor is the transparency of the supernova to $\\gamma$-rays. This is a key issue determining how much radioactive energy is converted into kinetic energy and supernova light. The maximum of the optical light curve (about 10 days after the explosion) is set by a maximum of the product of (declining) energy deposition and (rising) $\\gamma$-ray energy escape \\citep{Georgii:PintoE2001}. Furthermore, the Compton scattering optical depth to 1 MeV $\\gamma$-rays is below unity beyond about 50 days after explosion \\citep{Georgii:Pinto2001}, but absorption corrections to observed line $\\gamma$-rays are probably significant for typical models up to about 100 days after explosion \\citep[see e.g. Fig. 11 in][calculated for energies down to 10~keV, however]{Georgii:Hoeflich1998}. This illustrates clearly the importance of $\\gamma$-ray measurements with high spectral resolution and at those early times, in order to directly address the explosion mechanism: the line shapes and the ratio of the different line intensities from the $^{56}$Ni decay chain can reveal the ratio between deposited and directly radiated radioactive energy (e.g. \\cite{Georgii:Hoeflich1996}).  For illustrative purposes and simplification, we may simply assume as an extreme case that the supernova was transparent for our observation of the gamma-rays from $^{56}$Co decay over days 17-136 with emphasis on the late part; in this case we directly convert our flux limits into $^{56}$Co (and therefore original $^{56}$Ni) masses. Our lowest upper limit for the 1238 keV line of $2.3 \\cdot 10^{-5}$ photons~cm$^{-2}$~s$^{-1}$ then constrains the visible $^{56}$Ni mass to below 0.35 \\Msol. If we then want to reconcile this with the 0.77 \\Msol\\ of total $^{56}$Ni determined bolometrically \\citep{Georgii:Leibundgut2000}, more than half of the $\\gamma$-ray energy would be deposited in the supernova over this time window. We therefore do have to look in detail at the energy deposition efficiency around peak optical luminosity and/or effectiveness of $\\gamma$-ray escape soon thereafter.  For several model classes (detonation, delayed detonation, and sub-Chandrasekhar), $\\gamma$-ray light curves have been calculated through detailed Monte-Carlo photon transport in the expanding supernova \\citep{Georgii:Hoeflich1998,Georgii:Kumagai1998,Georgii:Isern1997,Georgii:Pinto2001}. Considerable variety in the gamma-ray flux by factors up to 5 arises from the different explosion models, envelope structures, and photon transport treatments employed in such calculations. In Fig.~\\ref{Georgii:F5}, expected  $\\gamma$-ray light curves for a few typical models \\citep{Georgii:Isern1998,Georgii:Kumagai1998} are shown, re-scaled for a distance of 11.3 Mpc. In Table  \\ref{Georgii:T1} we list the $^{56}$Ni mass for each of these models, together with time-averaged fluxes over the observation time for each $\\gamma$-ray line. We see that the predicted $^{56}$Co $\\gamma$-ray flux does not follow the straightforward scaling to the amount of $^{56}$Ni, the explosion mechanism and the envelope photon transport determine the time-dependent $\\gamma$-ray fluxes. For the same type of explosion model, predicted $^{56}$Ni masses vary within a factor of two: For the delayed-detonation class of models, values between 0.55 \\Msol\\ and 0.96 \\Msol\\ have been published \\citep{Georgii:Iwamoto1999,Georgii:Woosley1991,Georgii:Isern1997}, as a result of differences in the point at which the initially-slow nuclear burning (deflagration) is assumed to turn into a detonation. This typical intrinsic variability within an explosion type of a factor of two, which directly translates into the   $\\gamma$-ray flux scaling, indicates the systematics which typically remains, within an explosion type. In fact, any of the SNe~Ia scenarios (sub-Chandrasekhar, deflagration, delayed detonations, and pulsating delayed detonation models) has been shown to be capabable to produce a wide variety of $^{56}$Ni masses ranging from $\\simeq$~0.1 to 1 \\Msol\\ (e.g. \\cite{Georgii:Nomoto1984,Georgii:Hoeflich1996,Georgii:Hoeflich2002}).  On the other hand, Fig.~\\ref{Georgii:F5} illustrates that, for approximately the same amount of total $^{56}$Ni, pure deflagration or detonation models are about a factor of two dimmer in   $\\gamma$-rays, while the sub-Chandrasekhar model with substantial $^{56}$Ni sitting further outside reaches a   $\\gamma$-ray flux about twice as large as the typical delayed-detonation model.   \\begin{table}[h] \\caption{Fluxes in the 847 keV and 1238 keV line for different models, averaged over our observation time.} \\label{Georgii:T1} \\begin{minipage}{\\textwidth} \\begin{tabular}{llll} \\hline Model & $^{56}$Ni mass & I$_{\\rm 847~keV}$$\\cdot$10$^5$ & I$_{\\rm 1238~keV}$$\\cdot$10$^5$ \\\\ &[\\Msol] & [ph~cm$^{-2}$s$^{-1}$]&[ph~cm$^{-2}$s$^{-1}$]\\\\ \\hline W7\\footnote{\\cite{Georgii:Kumagai1998}}   & 0.58 & 4.2&3.0\\\\ W7DT$^a$                                  & 0.77 & 5.8&4.1\\\\ HeCD$^a$                                  & 0.72 & 8.2&5.5\\\\ WDD2$^a$                                  & 0.58 & 3.9&2.8\\\\ Deflag.\\footnote{\\cite{Georgii:Isern1998}}& 0.50 & 1.5&1.0\\\\ Del. Det.$^b$                             & 0.80 & 4.3&2.7\\\\ Det.$^b$                                  & 0.70 & 4.0&2.7\\\\ SubCH.$^b$                                & 0.60 & 2.7&1.5\\\\ \\hline \\end{tabular} \\end{minipage} \\end{table}  In Table  \\ref{Georgii:T1} we also list the time-averaged $^{56}$Co   $\\gamma$-ray fluxes of each model, together with our 2$\\sigma$ upper limits. Our SN1998bu flux limits are well below  the \"HeCD\" and the \"W7DT\" model predictions for both lines. The  \"W7\", \"WDD2\", delayed detonation and detonation model fluxes are marginally consistent with our flux limits, while the fluxes predicted from the sub-Chandrasekhar model and the deflagration model are consistent with our limits for both line energies at the adopted distance.  This comparison illustrates that with a $^{56}$Ni mass in the \"typical\" range derived for SN1998bu, around 0.7-0.8 \\Msol\\ \\citep{Georgii:Leibundgut2000}, we should have seen $^{56}$Co   $\\gamma$-rays at least due to those models which turn more rapidly from deflagration into detonation (W7DT) or partially-produce radioactivity in their outer ejecta (HECD). Yet, within Chandrasekhar-type models of the presently-favored type of a delayed transition from deflagration into detonation, a total $^{56}$Ni mass as high as about 1 \\Msol\\ may still be consistent with our measurement.  At a distance of 9.6 $\\pm$ 0.6 Mpc \\citep{Georgii:Feldmeier1997} based on planetary nebulae (PN) all models would be inconsistent with our 1238 keV and 847 keV flux limits, indicating either this distance is incorrect (see \\citep{Georgii:Maoz1999} for a discussion on a possible correction in the distance ladder scale) or that model treatments generally overestimate the $^{56}$Ni masses.  It will require time-resolved measurements of the   $\\gamma$-ray flux (hence a brighter / more nearby supernova or a more sensitive instrument), or exploitation of spectral-shape details as promised by the spectrometer aboard INTEGRAL \\citep[see discussion in][]{Georgii:Isern1997}, to decide among explosion models from gamma-ray line measurements alone."
    },
    "0208/astro-ph0208402_arXiv.txt": {
        "abstract": "{\\small We briefly review the properties and physical consequences of quasiperiodic oscillations (QPOs) seen on many occasions in the X-ray emission from black-hole binary systems. High frequency QPOs ($\\nu >$ 40 Hz) continue to be scrutinized as effects of general relativity, with new attention to the role of resonances in their formation. Low-frequency QPOs (0.05 to 30 Hz) exhibit complicated behavior, with occasions of high amplitude and particular correlations with some X-ray spectral parameters. QPO mechanisms are a requirement for any physical model seeking to explain either (1) the non-thermal X-ray spectrum that is commonly seen and is usually stronger than the accretion disk at times of highest luminosity, or (2) the hard X-ray spectrum evident when there is a steady type of radio jet. } ",
        "introduction": "Most of the brightest celestial X-ray sources recorded in a given year are transient outbursts from black hole binaries in the Galaxy. The eruptions are understood as a consequence of a low rate of mass accretion from the companion star \\cite{tan96}. The material gradually fills the outer regions of an accretion disk surrounding the black hole. When the disk surface density reaches a critical value, matter spirals into the inner disk where it reaches X-ray emitting temperatures before falling into the black hole event horizon \\cite{mey01}.  Such X-ray novae are the parent population for identifying black holes that are remnants of massive stars. The mass of the compact object (typically 5--15 \\msun) is deduced from radial velocity studies of the companion star. In most cases such measurements are only possible when the system has returned to relative quiescence, and the companion can be seen against the glare of the hot gases in the disk. In establishing the nature of the compact object, the critical argument is whether the mass exceeds the upper limit ($\\sim 3.0$ \\msun) for the mass of a neutron star.  There are now 17 ``dynamical black hole'' binaries in the Milky Way or LMC: 14 were first noticed as bright X-ray novae \\cite{oro02}, while the other 3 cases are persistent X-ray sources with O/B type companions.  Observations with the {\\it Rossi} X-ray Timing Explorer (RXTE) have pioneered efforts to further study black holes and their occasional relativistic jets via broadband X-ray observations during active states of accretion. The X-ray timing and spectral properties convey information about physical processes that occur near the black hole event horizon, and one of the primary research goals is to obtain constraints on the black hole mass and spin using predictions of general relativity (GR) in the strong-field regime. There is also the need to understand accretion physics for each of the four distinct emission states that are displayed by so many accreting black holes systems. In this paper we describe recent advances in these topics, with particular attention to the diverse forms of quasiperiodic oscillations (QPOs) \\cite{vdk00} that have been detected during the extensive monitoring programs for X-ray transients conducted with RXTE. ",
        "conclusions": ""
    },
    "0208/astro-ph0208128_arXiv.txt": {
        "abstract": "We investigate the action of the magnetorotational instability (MRI) in the context of iron--core collapse.  Exponential growth of the field on the time scale $\\Omega^{-1}$ by the MRI will dominate the linear growth process of field line ``wrapping\" with the same characteristic time.  We examine a variety of initial rotation states, with solid body rotation or a gradient in rotational velocity, that correspond to models in the literature.  A relatively modest value of the initial rotation, a period of $\\sim$ 10 s, will give a very rapidly rotating PNS and hence strong differential rotation with respect to the infalling matter. We assume conservation of angular momentum on spherical shells.  Rotational distortion and the dynamic feedback of the magnetic field are neglected in the subsequent calculation of rotational velocities.  In our rotating and collapsing conditions, a seed field is expected to be amplified by the MRI and to grow exponentially to a saturation field.  Results are discussed for two examples of saturation fields, a fiducial field that corresponds to $v_{\\rm{A}} = r\\Omega$ and a field that corresponds to the maximum growing mode of the MRI.  We find, as expected, that the shear is strong at the boundary of the newly formed protoneutron star, and, unexpectedly, that the region within the stalled shock can be subject to strong MHD activity.  Modest initial rotation velocities of the iron core result in sub--Keplerian rotation and a sub--equipartition magnetic field that nevertheless produce substantial MHD luminosity and hoop stresses : saturation fields of order $10^{14}$ -- $10^{16}$ G can develop $\\sim$ 300 msec after bounce with an associated MHD luminosity of $\\sim 10^{52}$ erg s$^{-1}$.  Bi-polar flows driven by this MHD power can affect or even cause the explosions associated with core-collapse supernovae. ",
        "introduction": "There is accumulating evidence that core collapse supernovae are distinctly and significantly asymmetric.  A number of supernova remnants show intrinsic ``bilateral\" structure (Dubner et al. 2002). Jet and counter jet structures have been mapped for Cas A in the optical (Fesen \\& Gunderson 1996; Fesen 2001; and references therein).  Chandra X-ray Observatory (CXO) and XMM Newton observations of Cas A show that the jet and counter jet and associated structure are observable in the X-ray and that the intermediate mass elements are ejected in a roughly toroidal configuration (Hughes et al. 2000;  Hwang et al. 2000; Willingale et al. 2002).  HST observations of the debris of SN~1987A show that the ejecta are asymmetric with an axis that roughly aligns with the small axis of the rings (Pun et al. 2001; Wang et al. 2002). Spectropolarimetry shows that substantial asymmetry is ubiquitous in core-collapse supernovae, and that a significant fraction of core-collapse supernovae have a bi-polar structure (Wang et al. 1996, 2001).  The strength of the asymmetry observed with polarimetry is higher (several \\%)  in supernovae of Type Ib and Ic that represent exploding bare non-degenerate cores (Wang et al. 2001).  The degree of asymmetry also rises as a function of time for Type II supernovae (from $\\lta$ 1\\% to $\\gta$ 1\\%) as the ejecta expand and the photosphere recedes (Wang et al.  2001; Leonard et al. 2000, 2001).  Both of these trends suggest that it is the core collapse mechanism itself that is responsible for the asymmetry.  Two possibilities are being actively explored to account for the observed asymmetries.  One is associated with the rotational effect on convection (Fryer \\& Heger 2000) and another is due to the effect of jets (Khokhlov et al. 1999; Wheeler et al. 2000; Wheeler, Meier \\& Wilson 2002).  The time scale of neutrino emission is short compared to the dynamical time of the overlying stellar mantle and envelope. Neutrino asymmetries can yield short-lived, impulsive effects (Shimizu, et al. 1994; Burrows \\& Hayes 1996; Lai et al. 2001), but there are questions of whether expansion and transverse pressure gradients will eliminate transient asymmetries before homologous expansion is achieved (Chevalier \\& Soker 1989). Rayleigh--Taylor and Richtmyer--Meshkov instabilities might produce ``finger'' asymmetries that are preserved, but it is unclear that such finer scale perturbations can reproduce the common feature of a single symmetry axis that is substantially independent of space and time (Wang et al. 2001). Jet calculations have established that non-relativistic axial jets of energy of order $10^{51}$ erg originating within the collapsed core can initiate a bi-polar asymmetric supernova explosion that is consistent with the spectropolarimetry (Khokhlov et al. 1999; Khokhlov \\& H\\\"oflich 2001; H\\\"oflich et al. 2001). The result is that heavy elements (e.g. O, Ca) are characteristically ejected in tori along the equator. Iron, silicon and other heavy elements in Cas A are distributed in this way (Hwang et al. 2000), and there is some evidence for this distribution in SN 1987A (Wang et al. 2002).  Radioactive matter ejected in the jets can alter the ionization structure and hence the shape of the photosphere of the envelope even if the density structure is spherically symmetric (H\\\"oflich et al. 2001).  This will generate a finite polarization, even though the density distribution is spherical and the jets are stopped deep within the star and may account for the early polarization observed in Type II supernovae (Leonard et al. 2000; Wang et al. 2001).  If one of the pair of axial jets is somewhat stronger than the other, jets can, in principle, also account for pulsar runaway velocities that are parallel to the spin axis (Helfand et al. 2001, and references therein).  While a combination neutrino--induced/jet--induced explosion may prove necessary for complete understanding of core-collapse explosions, jets of the strength computed by Khokhlov et al. (1999) are sufficient for supernova explosions.  In this paper, we will explore the possible conditions that could lead to the formation of buoyant bi-polar MHD outflow.  Immediately after the discovery of pulsars there were suggestions that rotation and magnetic fields could be a significant factor in the explosion mechanism (Ostriker \\& Gunn 1971; Bisnovatyi-Kogan 1971; Bisnovatyi-Kogan \\& Ruzmaikin 1976; Kundt 1976).  Typical dipole fields of $10^{12}$ G and rotation periods of several to several tens of milliseconds yield electrodynamic power of $\\sim 10^{44-45}$ erg s$^{-1}$ that is insufficient to produce a strong explosion.  The evidence for asymmetries and the possibility that bi-polar flows or jets can account for the observations suggest that this issue must be revisited.  The fact that pulsars like those in the Crab and Vela remnants have jet-like protrusions (Weisskopf et al. 2000; Helfand et al. 2001) also encourages this line of thought. The present-day jets in young pulsars may be vestiges of much more powerful MHD jets that occurred when the pulsar was born.  The transient values of the magnetic field and rotation could have greatly exceeded those observed today.  Tapping that energy to power the explosion could be the very mechanism that results in the modest values of rotation and field the pulsars display after the ejecta have dispersed.   In the current context, one would like to know not only whether or not the rotation and magnetic field of a nascent neutron star can power the supernova explosion, but, specifically, whether or not bi-polar flows or jets form and whether or not they are sufficiently energetic to drive the explosion. Possible physical mechanisms for inducing axial flows, asymmetric supernovae, and related phenomena driven by magnetorotational effects were considered by Wheeler et al. (2000).  The means of amplifying magnetic fields by linear wrapping associated with differential rotation in the neutron star and possibly by an $\\alpha$~--~$\\Omega$ dynamo were discussed. Whereas Wheeler et al. (2000) focused on the effect of the resulting net dipole field, Wheeler, Meier, \\& Wilson (2002) recognized that the toroidal field would be the dominant component of the magnetic field and explored the capacity of the toroidal field to directly generate axial jets by analogy with magneto-centrifugal models of jets in active galactic nuclei (Koide et al. 2000; and references therein). Wheeler, Meier, \\& Wilson (2002) found that the production of a strong toroidal field, substantially stronger than the $10^{12}$ G field of a pulsar, and strong axial jets driven by that field are possible.  A caveat to consider is that Wheeler, Meier, \\& Wilson (2002) again considered only amplification of the field by ``wrapping,\" a process that only increases the field linearly, and hence rather slowly in time. While conditions of very rapid rotation might exist that lead to a sufficiently rapid growth of the magnetic field, that is not guaranteed. In addition, reconnection might limit the field before it can be wrapped the thousands of times necessary to be interesting. Here we consider the effects of magnetic shearing, the magnetorotational instability (MRI; Balbus \\& Hawley 1991, 1998), on the strongly shearing environment that must exist in a nascent neutron star.  This instability is expected to lead to the rapid exponential growth of the magnetic field with characteristic time scale of order the rotational period.  While this instability has been widely explored in the context of accretion disks, this is the first time it has been applied to core collapse.  We will argue that this instability must inevitably occur in core collapse, that it is likely to be the dominant mechanism for the production of magnetic flux in the context of core collapse, and that it has the capacity to produce fields that are sufficiently strong to affect, if not cause, the explosion.  In \\S 2 we describe our assumptions by which we treat the angular momentum and field amplification in the context of a numerical collapse calculation.  In \\S 3, we present the results of these calculations. Section 4 presents a discussion and conclusions. ",
        "conclusions": "No one would doubt that the progenitors of core collapse supernovae rotate and possess some magnetic field.  Pulsars as remnants of core collapse are manifestly rotating and magnetized.  The question has always been whether rotation and magnetic fields would be incidental perturbations or a critical factor in understanding the explosion.  We have shown here that with plausible rotation from contemporary stellar evolution calculations and any finite seed field with a component parallel to the rotation axis, the magnetorotational instability can lead to the rapid exponential growth of the magnetic field to substantial values on times of a fraction of a second, comparable to the core collapse time. Even a relatively modest value of initial rotation gives a very rapidly rotating PNS and hence strong differential rotation with respect to the infalling matter.  This result seems reasonably robust.  The reason is that the instability condition for the MRI is basically only that the gradient in angular velocity be negative.  This condition is broadly satisfied in core collapse environments.  Rotation can weaken supernova explosions without magnetic field \\citep{FH00}; on the other hand, rotational energy can be converted to magnetic energy that can power MHD bi-polar flow that may promote supernova explosions.  {\\it The implication is that rotation and magnetic fields cannot be ignored in the core collapse context.}  Even artificially limiting the post-collapse rotation to sub--Keplerian values as done here, we find fields in excess of $10^{15}$ G near the boundary of the neutron star.  While this field strength is sub--equipartition, the implied MHD luminosities we have derived are of order $10^{52}$ erg s$^{-1}$.  This is a substantial luminosity and could, alone, power a supernova explosion if sustained for a sufficiently long time, a fraction of a second.  As pointed out by Wheeler, Meier \\& Wilson (2002), the fields do not have to be comparable to equipartition to be important because they can catalyze the conversion of the large reservoir of rotational energy into buoyant, bi-polar MHD flow.  Higher rates of initial rotation that are within the bounds of the evolutionary calculations could lead to even larger post-collapse rotation and even larger magnetic fields.  If the initial rotation of the iron core proves to be substantially lower than we have explored here, then the MRI would be of little consequence to the explosion.  The MHD luminosities derived here are comparable to the typical neutrino luminosities derived from core collapse, $\\sim 10^{52}$ erg~s$^{-1}$. One important difference is that the matter beyond the PNS is increasingly transparent to this neutrino luminosity, whereas the MHD power is deposited locally in the plasma. Another difference is that the neutrino luminosity is basically radial so it resists the inward fall of the collapse, the very source of the neutrino luminosity itself.  In contrast, hoop stresses associated with the magnetic field (see below) will tend to pull inward and force matter selectively up the rotation axis.  We note that for complete self-consistency, one should apply the MRI to the evolution of rotating stars where even a weak field renders the H{\\o}iland dynamical stability criterion ``all but useless\" in the words of Balbus \\& Hawley (1998).  Some steps to explore magnetic viscosity in stellar evolution have been taken. \\citet{spruit98} have invoked a magnetic viscosity that efficiently couples the inner core to the outer expanding giant envelope until the core contraction is more rapid than its rotation. They concluded that if the decoupling is as late as the phase of carbon depletion, the iron core might be rotating with an extremely low angular velocity, $\\sim 10^{-4}$ rad s$^{-1}$, yielding an initial PNS rotation period of 100 s or an angular velocity of $\\sim$ 0.06 rad s$^{-1}$.  This condition requires that the field be of order the rotation energy of the core, but if this condition holds then field amplification by the MRI in the collapse would be negligible. On the other hand, recent calculations by \\citet{HW02} based on a prescription for magnetic viscosity by \\citet{spruit02} yield much more rapidly rotating iron cores. \\citet{HW02} find PNS rotation rates of 4 to 8 ms, consistent with the values we have explored here.  Clearly, much more must be done to understand the magnetorotational evolution of supernova progenitors.  As expected, the shear and hence the saturation fields are often highest at the boundary of the PNS where strong MHD activity is anticipated. Unexpectedly, with our fiducial saturation field, the field can also be large within the standing shock compared to the local pressure since the shock compression there naturally leads to shear in a rotating environment and since gas pressure is relatively low.  The strength of this secondary peak is about the same for the initial solid body profile with $\\Omega_{\\rm{0\\_c}} = 0.2$ rad s$^{-1}$ and for the MM and FH profiles with $\\Omega_{\\rm{0\\_c}} = 1.0$ rad s$^{-1}$.  We have had to limit $\\Omega_{\\rm{0\\_c}}$ to smaller values in the case of initially constant angular velocity in order to not violate the sub--Keplerian condition at larger radii after collapse.  The maximum magnetic fields achieved are generally about the same strength within a factor of 10 for all three initial angular velocity prescriptions we have explored, although the value of $\\Omega_{\\rm{0\\_c}}$ was chosen to be substantially smaller in the initially solid rotation case compared to those where there is a gradient in $\\Omega$.  The configuration of the magnetic field in a precollapse iron core is not well understood.  In this calculation we have assumed there exists a seed vertical field to calculate the growth of the field due to the MRI; however, the MRI can amplify other components of the magnetic field.  The final configuration of the magnetic field after collapse may be less uncertain since the system has a strongly preferred direction due to rotation.  In the context of accretion disks, \\citet{HGB96} show that their fiducial run with initial random magnetic field configuration results in 9\\% radial, 88 \\% toroidal, and 3 \\% vertical components. Most of the shear is in the radial direction, so the radial component is greatly amplified by the MRI and turned into toroidal field due to differential rotation \\citep{BH98}.  The dominant component is most likely to be the toroidal field.  Another uncertainty is the rotational profile.  Although we have formally considered the MRI in a geometry with arbitrary pressure gradients, in practice our assumption of conservation of angular momentum in spherical shells effectively restricts our analysis of the resulting magnetic fields to the equatorial plane.  It is not clear what profile to use in the PNS, since, we note, even the rotational profile of the Sun is not well understood.  A full understanding of the rotational state of a PNS remains a large challenge.  We have assumed various prescriptions for the saturation field.  All are variations on the theme that, within factors of order 2$\\pi$, the saturation field will be given by the condition $v_{\\rm{A}} \\sim r \\Omega$.  Fig. \\ref{vel} compares the velocities of sound ($c_{\\rm{s}}$), Keplerian rotation ($v_{\\rm{kep}}$) and model rotation ($r\\Omega$), and the Alfv\\'{e}n ($v_{\\rm{A}}$) velocity corresponding to $B_{\\rm{sat}}$ for the solid, MM and FH profiles we adopted.  For the sub--Keplerian rotational velocity profiles we employed, the magnetic field saturates at $v_{\\rm{A}} \\sim r\\Omega$.  For the case of $B_{\\rm{max,en}}$, $v_{\\rm{A}} < r\\Omega$ since $B_{\\rm{max,en}} < B_{\\rm{sat}}$ for the MM and FH profiles, and $B_{\\rm{max,en}}$ is especially small for the initial solid rotation profile.  In all cases, the saturation field is sub--equipartition with Alfv\\'{e}n velocity less than Keplerian.  In the numerical disk simulations, about 20 rotations are required to reach saturation.  The region of maximum shear in these calculations, around 15 km, typically has an angular velocity of 500 rad s$^{-1}$ or a period of about 0.013 s.  That means that by the end of the current calculations at 0.387 s, there have been about 30 rotations. Although the prescriptions for the growth and saturation fields we use here are heuristic, this aspect of our results is certainly commensurate with the numerical simulations of the MRI.  The issues of the saturation field and the nature of astrophysical dynamos are still vigorously explored.  There are concerns that even weak fields may limit turbulent cascade at the smallest scales and hence suppress dynamo processes.  Vainshtein, Parker \\& Rosner (1993) point out that this may be avoided by the formation of Coriolis-twisted loops that develop in shearing strong toroidal fields.  Papaloizou and Szuszkiewicz (1992) presented a global stability analysis for conditions where the angular velocity is constant along magnetic lines of force that generally agreed with the local MRI stability analysis of Balbus \\& Hawley (1998) for weak fields.  For strong fields, however, their analysis suggested that fields corresponding to the saturation condition $v_{\\rm{A}} \\sim \\sqrt{R \\Omega c_{\\rm{s}}}$ would be unstable.  For the sub--Keplerian conditions we consider here, this saturation criterion would yield a ratio of magnetic to total pressure that is larger by a factor of $c_{\\rm{s}}/r \\Omega$ than the criteria we used (see Fig.  \\ref{vel}). \\citet{VC01} conclude that the MRI has the required properties for a dynamo, anisotropic turbulence in a shearing flow, to generate both disordered and ordered fields of large strength.  The saturation limits we have adopted here are consistent with those found in numerical calculations of the MRI saturation, but this topic clearly deserves more study.  We have made one major assumption that must be checked more carefully in future work.  That is that we can apply the MRI linear stability analysis derived for no background radial flow to a dynamic situation where radial inflow and convective motions are rapid.  The character of the MRI has never been investigated in this regime with finite background flow.  One rationale for our assumption is that at a given radius one can transfer to a co-moving frame where the local radial velocity is zero.  This is formally correct, but since we then make assumptions about modes with wavelength comparable to the size of the system, this formality may be inapplicable to the problem we want to solve, the nature of the saturation fields.  An ex-post factor justification for our results is that after collapse when the PNS settles nearly into hydrostatic equilibrium, the radial motions are, in fact, small and the traditional MRI is probably valid as illustrated in Fig. \\ref{vrad}. The action of the MRI in the vicinity of the standing shock and the fields we derive there may, however, be questionable on this basis.  This topic deserves more study as a general point of physics, not merely for the application to core collapse. We also note that the MRI could be altered in the context of large radiation pressure.  The radiation pressure is not a major component of the total pressure in the current calculations, so we do not believe this to be a factor here, but this is worth considering in principle.  We have also neglected several feed-back processes. Among these are the build up of magnetic pressure and hoop stresses, the viscous coupling of shells that will tend to suppress the differential rotation, the effects of neutrino viscosity on turbulence, the effects of the magnetic field on neutrino transport, and the effects of centrifugal forces.  We will describe some of these issues briefly.   Both the magnetic pressure and the magnetic viscosity are small for the sub--Keplerian conditions explored here.  For most cases $\\beta^{-1}$ is less than 0.1 for the conditions we have assumed, (the $B_{\\rm{sat}}$ case for the FH profile with $\\Omega_{0\\_\\rm{c}} = 1.0 s^{-1}$ pushes this limit), so the direct dynamical effect of the magnetic field is expected to be small.  The viscous time scale is $\\tau_{\\rm{vis}} \\sim (\\alpha \\Omega)^{-1}(\\rm{r/h})^2$, where $\\alpha$ is the viscosity parameter and h is the vertical scale height, with h $\\sim$ r for our case.  For a magnetically-dominated viscosity, \\begin{equation} \\alpha \\sim \\frac{B_{\\rm{r}} B_{\\phi}}{4 \\pi P} = \\left(\\frac{B_{\\rm{r}}}{B_{\\phi}}\\right) \\frac{B_{\\phi}^2}{4 \\pi P} \\sim 2 \\left(\\frac{B_{\\rm{r}}}{B_{\\phi}}\\right) \\beta^{-1}. \\end{equation} With this expression for $\\alpha$, the viscous time becomes: \\begin{equation} \\tau_{\\rm{vis}} \\sim \\frac{1}{2}\\left(\\frac{B_{\\phi}}{B_{\\rm{r}}}\\right) \\left(\\frac{1}{\\beta^{-1}\\Omega}\\right) \\gg \\Omega^{-1}, \\end{equation} where the final inequality follows from $B_{\\phi} > B_{\\rm{r}}$ and $\\beta^{-1} < 1$.  This prescription for viscosity is reasonable in the absence of convection.  In the portions of the structure that are convective, the viscosity could be enhanced significantly.  In addition to dynamic effects, we note that the fields generated here are well above the QED limit ($B_{\\rm{Q}} = 4.4 \\times 10^{13}$ G).  In this exotic regime, such a strong field has radiative and thermodynamic effects \\citep{duncan00}, although it is not clear that these have profound effects on the dynamics in core collapse supernovae.  We have not discussed the role of neutrinos here, although the processes of neutrino loss and de-leptonization are included in our calculation of the cooling PNS.  It is possible that the neutrino flux affects the magnetic buoyancy (Thompson \\& Murray 2002) and that the magnetic fields affect the neutrino emissivity \\citep{TD96} and interactions with the plasma (Laming 1999).  The time scale for shear viscosity due to neutrino diffusion is much longer than the times of interest here, although magnetic fields and turbulence can make it shorter \\citep{gou98}.  The MRI provides magnetic field and turbulence, so this issue deserves further study.  In addition to affecting the shear, the neutrino viscosity might also affect the turbulence needed to make the MRI work.  Using the expression for the neutrino viscosity at sub-nuclear densities from \\citet{gou98}, \\begin{equation} \\label{nuvisc} \\eta_{\\nu} \\sim 2\\times10^{23}\\left(\\frac{T}{10~{\\rm MeV}}\\right)^2\\rho_{13}^{-1} {\\rm g~cm^{-1}~s^{-1}}, \\end{equation} gives a Reynolds number \\begin{equation} \\label{reynolds} Re \\sim \\frac{v R \\rho}{\\eta_{\\nu}} \\sim \\frac{\\Omega R^2 \\rho}{\\eta_{\\nu}} \\sim 10^5 \\end{equation} near the PNS boundary at $R \\sim 10^6$ cm with $\\Omega \\sim 1000$ rad s$^{-1}$.  This could decrease to Re $\\sim 100$ near the standing shock where the density is lower.  The former is probably large enough to sustain turbulence in the strong shear, but the latter might not be.  When viscosity dominates the dissipation in a collisional plasma, the growth condition that the field growth time be less than the dissipation time yields a constraint on the magnetic field \\citep{BH98}: \\begin{equation} B \\gg \\left(\\frac{15 \\pi}{8} \\nu \\rho \\Omega\\right)^{1/2} = 5\\times10^{7}~\\rm{G}~ \\Omega_3^{1/2}\\left(\\frac{T}{10~{\\rm MeV}}\\right)^{5/4} \\end{equation} This equation implies that B must exceed $\\sim 10^8$ G near the boundary of the PNS where T $\\sim 10$ MeV and smaller values in cooler portions at larger radii.  Even compression of moderate fields in the iron core should exceed this threshold. We note that if one were to use the expression for the neutrino viscosity from eq. (\\ref{nuvisc}), the corresponding limit on the field would exceed $10^{13}$ G for similar parameters.  This limit may not be relevant, however, since in order to use the fluid equations, rather than a kinetic theory, to describe the instability and its damping, the physical lengthscale associated with the viscosity, the mean free path of the dissipative particles, must be substantially less than the damping scale of the turbulence.  This is not likely to be the case for neutrinos in the vicinity of the region of maximum shear in this problem.  The damping length where Re $\\sim$ 1 would be about $10^{-5}$ of $10^6$ cm or about 10 cm from eq. (\\ref{reynolds}), whereas the mean free path of the neutrinos (neglecting blocking and other complications) is about $10^4$ cm at the density of $\\sim 10^{13}$ {g~cm$^{-3}$} and temperature of about 10 MeV that characterize the region of maximum shear \\citep{arnett96}. There may be regions deep in the neutron star where the neutrino mean free path is short enough that this becomes an issue, but we do not think that the neutrinos can affect the eddy turnover and instability of the MRI in regions that are significant for maximum field growth.  An obvious imperative is to now understand the behavior of the strong magnetic fields we believe are likely to be attendant to any core collapse situation.  The fields will generate strong pressure tensor anisotropies that can lead to dynamic response even when the magnetic pressure is small compared to the isotropic ambient gas pressure. As argued in Wheeler, Meier \\& Wilson (2002), a dominant toroidal component is a natural condition to form a collimated magnetorotational wind, and hence polar flow.  A first example of driving a polar flow with the MRI is given by \\citet{HB02}.  A key ingredient to force flow up the axis and to collimate it is the hoop stress from the resulting field.  We have examined the acceleration implied by the hoop stresses of the fields we have derived here, $a_{\\rm{hoop}} = B_{\\phi}^2/4\\pi \\rho r$.  We find that the hoop stresses corresponding to the peak saturation fields can be competitive with, and even exceed, the net acceleration of the pressure gradient and gravity.  The large scale toroidal field is thus likely to affect the dynamics by accelerating matter inward along cylindrical radii.  The flow, thus compressed, is likely to be channeled up the rotation axes to begin the bi-polar flow that will be further accelerated by hoop and torsional stresses from the field, the ``spring and fling\" outlined in Wheeler et al. (2002). The MRI is expected to yield a combination of large scale and small scale magnetic fields.  Even the small scale fields in the turbulent magnetized medium may act like a viscoelastic fluid that would tend to drive circulation in along the equator and up the rotation axis \\citep{williams02} where the same small scale fields could collimate and stabilize the flow even in the absence of large scale toroidal fields \\citep{li02}. On the other hand, small scale fields can result in dissipation by reconnection, an issue that we have not treated here, relying implicitly on the numerical simulations that show a growth in the field that is not eliminated by such effects.  The dynamics of these jets may depart substantially from pure hydrodynamical jets, since they will tend to preserve the flux in the Poynting flow and reconnection can accelerate the matter (see Spruit, Daigne \\& Drenkhahn 2001 and references therein). The poloidal component of the field can be another contributor to plasma waves (Wheeler et al. 2000).  For a complete understanding of the physics in a core collapse supernova explosion, a combination of neutrino--induced and jet--induced explosion may be required.  Understanding the myriad implications of this statement will be a rich exploration."
    },
    "0208/astro-ph0208291_arXiv.txt": {
        "abstract": "Understanding post-common-envelope evolution is important in the studies of close-binary systems. The majority of all interacting binaries with a compact object in their system (e.g. cataclysmic variables, X-ray binaries) are thought to have passed through a common-envelope (CE) phase.  Current models of interacting binaries all assume, as a starting point, that there is no significant modification of the secondary star compared with a normal star of similar spectral type.  The extent to which the CE significantly alters the composition of the secondary star has yet to be determined.  We are studying the M-type secondary in the pre-CV system EG UMa, in order to determine its atmospheric parameters by comparison with synthetic spectra generated using {\\sc phoenix}. Absorption lines due to several elements have been used in investigating the stellar parameters of effective temperature, gravity and over elemental abundance. In addition, we are searching for anomalies due to s-process elements (such as Ba, Sr, Rb, Y), which may have been deposited in the atmosphere during the CE stage, and prove that CE evolution has occurred.  EG UMa displays strong YO absorption bands, which are normally associated with giants and S stars.  These bands may have been formed as a result of an Y-abundance enhancement introduced by the CE, which would be consistent with the observed possible Rb and Sr enrichment. ",
        "introduction": "Pre-Cataclysmic Variables (pre-CVs) are objects consisting of a white dwarf and a late, low-mass main-sequence star.  The distinguishing feature of these objects is that they have already undergone a phase of common-envelope (CE) evolution, and are sometimes also called post-common-envelope binaries (PCEBs).  Initially the binary separation is large, however the CE phase allows a large amount of angular momentum to be lost in the system, and the binary separation shrinks.  Pre-CV or PCEB stars have not yet reached the evolutionary stage of significant mass-transfer due to Roche-lobe overflow when they would enter the cataclysmic-variable stage. Pre-CVs are therefore excellent objects for studying the initial evolutionary stages as observations are not confused by discs or hotspots.  EG UMa was initially identified as a DA white dwarf (Stephenson 1960) and later a binary companion with emission lines was discovered (Greenstein 1965). The following system parameters have been determined (see Bleach et al. 2000 for details and references): The primary is a cool white dwarf at around 13,000 K, with a mass of 0.64 M$_{\\odot}$ and a radius of 0.013 R$_{\\odot}$, while the secondary is an M4--5 dwarf with a mass of 0.42 M$_{\\odot}$ and a radius of 0.45 R$_{\\odot}$. The orbital period of the system is around 16 hours (Lanning 1982).  However, previous work has indicated that the M dwarf might have a larger radius and mass than that for a main-sequence star of similar spectral type (Bleach et al. 2000). We can use high-resolution spectroscopy to obtain estimates of the gravity in the M dwarf, which would give some insight into the mass and radius inconsistencies. ",
        "conclusions": "Line-profile fitting has been used to determine the atmospheric parameters of the secondary star in the EG UMa binary system. The best fitting solution gives a surface temperature of the M star as 3300 $\\pm$ 100 K and a surface gravity of $\\log g$ = 5.5. This value of $\\log g$ is slightly higher than the average for an early M dwarf.  EG UMa is unique among the stars in the sample by displaying YO bands.  These may be explained by an abundance enhancement of Y, which would be consistent with s-process enhancement from the common-envelope phase of the binary system's evolution.  Similarly, such enrichment has been inferred for Sr and Rb, but not for Ba. Further work is in progress to examine the anomalies in more detail."
    },
    "0208/astro-ph0208258_arXiv.txt": {
        "abstract": "We report a serendipitous discovery of broad absorption lines that were newly formed in the spectrum of the high-redshift, luminous radio-loud quasar TEX 1726$+$344, in a time interval of only 12 years. This is the first quasar showing a transition from narrow absorption lines to broad absorption lines. It also becomes one of the few radio-loud broad absorption line quasars. The gas cloud responsible for these broad absorption lines is derived to have parameters coinciding with those of the remnant of a tidally disrupted star. ",
        "introduction": "Approximately 10\\% of all quasars are ``radio loud'', in the sense that they are more luminous at radio than at optical wavelengths. Also for some unknown reason,  about 10\\% of radio-quiet quasars show broad absorption lines (BAL) in their spectra. Only recently have a handful of radio-loud BAL quasars been discovered (Becker et al. 1997; Brotherton et al. 1998), thus changing our view that BALs only appear in radio-quiet quasars. These quasars will serve as a link between radio-loud and radio-quiet quasars, and will aid in understanding the structure and kinematics of powerful outflows in these quasars. This may eventually lead to a breakthrough in understanding the origin of the radio loudness and the radio-loud/radio-quiet dichotomy. ",
        "conclusions": "We note that absorption line variations with large magnitudes have been observed in the low-redshift radio-quiet quasar PG 1126$-$041 (Wang et al. 1999). Numerous studies on the absorption line variability of Seyfert galaxies also exist in the literature (Malizia et al. 1997; Risaliti, Elvis \\& Nicastro 2002; Akylas et al. 2002), and have been modeled with a single cloud moving through the line of sight (Akylas et al. 2002). Hamann, Barlow \\& Junkkarinen (1997) have identified variable intrinsic absorption lines (FWHM$\\sim$400 km/s) in a radio-quiet quasar Q2343$+$125. BAL variability at a moderate level ($<$40\\%) has been found in several radio-quiet quasars (Barlow 1993). In contrast to these previous studies, TEX 1726$+$344, a high luminosity {\\it radio-loud} quasar developing BALs in a time interval of 12 years, represents a new phenomenon.  The popular disk-wind model (Murray et al. 1995) for both broad emission and absorption lines has been challenged by Ma \\& Wills (2001) because some radio-loud quasars were discovered to have dramatic CIV emission line variability, speculated to be due to the illumination of the jet pointing to the polar region of the quasar. The tidally disrupted stars model now seems to be the best candidate. The broad absorption lines in TEX 1726$+$344 will disappear in just a few years, if they were indeed caused by a tidal disruption event."
    },
    "0208/astro-ph0208544_arXiv.txt": {
        "abstract": "This paper reviews our current understanding of the process of re-ionization of the Universe, focusing especially on those models where re-ionization is caused by UV radiation from massive stars. After reviewing the expected properties of stars at zero metallicity, I discuss the properties of primordial HII regions and their observability. ",
        "introduction": "\\label{sect:intro}  %  The emergence of the first sources of light in the Universe and the subsequent re-ionization of hydrogen marks the end of the \"Dark Ages\" in cosmic history, a period characterized by the absence of discrete sources of light. Despite its remote timeline, this epoch is currently under intense theoretical investigation and will soon begin to be probed observationally. The first reason to study this epoch is the fact that this - together with the epoch of recombination - is the most accessible of the global phase transitions undergone by the Universe after the Big Bang.  Two facts make the formation of structure in the Dark Ages easier to study theoretically than similar processes occurring at other epochs: {\\it i)} the formation of the first structures is directly linked to the growth of linear perturbations, and {\\it ii)} these objects have a known metallicity set by the end-product of the primordial nucleosynthesis. Therefore, the second reason to study this epoch is because it makes it possible to probe the power spectrum of density fluctuations emerging from recombination at scales smaller than accessible by current cosmic microwave background experiments.  In a Universe where structures grow hierarchically, the first sources of light act as seeds for the subsequent formation of larger objects. Thus, the third reason to study this period is that by doing so we may learn about processes relevant to the formation of the nuclei of present day giant galaxies.  In Section \\ref{sect:basics}, I will review our present view on the basic processes leading to re-ionization. I refer the reader to the recent review by Loeb and Barkana \\cite{loebbarkana} for a more detailed discussion. In Section \\ref{sect:theory}, I will review in more detail a few recent theoretical results relevant to the case of re-ionization by stellar UV radiation. The present observational status is summarized in Section \\ref{sect:observations}. The properties of primordial and low metallicity HII regions are discussed in Sections \\ref{sect:primordial} and \\ref{sect:lowmetal}, respectively. Finally, in Section \\ref{sect:future}, I address the observability of these objects by NGST and future large ground-based telescopes. ",
        "conclusions": ""
    },
    "0208/hep-th0208050_arXiv.txt": {
        "abstract": "Current theoretical investigations seem to indicate the possibility of observing signatures of short distance physics in the Cosmic Microwave Background spectrum. We try to gain a deeper understanding on why all information about this regime is lost in the case of Black Hole radiation but not necessarily so in a cosmological setting by using the {\\em moving mirror\\/} as a toy model for both backgrounds. The different responses of the Hawking and Cosmic Microwave Background spectra to short distance physics are derived in the appropriate limit when the moving mirror mimics a Black Hole background or an expanding universe. The different sensitivities to new physics, displayed by both backgrounds, are clarified through an averaging prescription that accounts for the intrinsic uncertainty in their quantum fluctuations. We then proceed to interpret the physical significance of our findings for time-dependent backgrounds in the light of nonlocal string theory. ",
        "introduction": "\\label{intro} Although short distance physics belongs in the realm of quantum gravity, it is possible that some traces may have survived in the low energy observables. Many efforts have been devoted to the exploration of this issue \\cite{tp} and the claims fall in two categories: Hawking radiation \\cite{hawking} is {\\em robust against\\/} nonlinear modifications of short distance physics \\cite{unruh,jacobson,cr}; but Cosmic Microwave Background (CMB) spectrum may generically be {\\em sensitive\\/} to short distance modes \\cite{b.greene} (see also \\cite{tp} for a partial list of other relevant References), except for special classes of modification that ensure adiabatic time evolution of the modes at late times \\cite{mb,branden.jerome}. \\par These claims should be taken as suggestive rather than conclusive for at least the two following reasons: \\par\\noindent {\\em i\\/}) we do not, as yet, have a fundamental theory to describe physics at energies higher than the Planck mass. This means that even for specific short distance models introduced in literature, we do not have a set of equations by which to study these models. Einstein equations and in particular the equation of energy conservation almost always break down in the presence of nonlinear trans-planckian physics \\cite{lm-mb}. We can hope that quantum field theory and modifications to Einstein equations which satisfy Bianchi identity may remain a good description, as approximate tools, while taking the low energy limit; \\par\\noindent {\\em ii\\/}) the spectrum is very sensitive to the initial conditions. We do not know what the initial conditions are in Cosmology but they can contaminate the result of the investigation of the role of nonlinear physics in the CMB spectrum. \\par Despite the difficulties of carrying out such an investigation, it is very exciting to consider the possibility that we may find low energy signatures from very high energy processes in Cosmology. If trans-planckian physics is described within the framework of String Theory (e.g.~see Ref.~\\cite{amanda}) with our present state of knowledge, then its first evidence would be found in cosmological grounds. \\par Let us take the above claims as true. In these notes we want to gain a deeper understanding about probes of Planck scale physics, by posing the following question: Why do the CMB spectrum and Hawking radiation respond differently to nonlinear short distance physics? In order to achieve a comparison between Black Holes and Cosmology, the setup here is the following: Consider a {\\em moving mirror\\/} which follows a certain trajectory along a cartesian direction (see \\cite{birrell} and References therein). In Section~\\ref{mirror} we discuss the response function of a detector in the space to the right of the moving mirror for two limiting cases: {\\em a\\/}) when the mirror's motion ``mimics'' a Black Hole background (Section~\\ref{bhmirror}); {\\em b\\/}) and when the mirror's motion follows the evolution of the Hubble horizon (Section~\\ref{cosmomirror}). We then compare the spectra and argue that the difference between Black Holes and Cosmology with respect to their sensitivity to short distance physics, arises from the fact that since the {\\em stationary\\/} background of Black Holes is characterized by a {\\em length scale much larger\\/} than the Planck length, there are lesser degrees of freedom in the first instance (surface degrees of freedom ``confined'' to the Black Hole horizon), as compared to the dynamic background of Cosmology which, due to non-locality, results in volume degrees of freedom. This leads us on to introducing a spatial averaging prescription to account for the uncertainty in the origin of Hawking radiation that indicates a ``wash-out'' of trans-planckian modes, while in Cosmology there is a natural time averaging to account for the uncertainty in the initial conditions which (at best) enhances the role played by short scale physics at later times. We conclude in Section~\\ref{race} that any traces of Planck scale physics in the spectra, as measured by their departure from thermality, may be larger and thus observable in the cosmological case. We then speculate that our results are related and point very much in the same direction as recent works on cosmological background solutions in String Theory (see \\cite{eva} and References therein). An observable effect becomes possible when quantum fluctuations are comparable to thermal fluctuations. The competition between quantum and thermal fluctuations is larger for volume degrees of freedom than for surface degrees of freedom, thus the possibility that they may be within reach of observation in Cosmology. \\par Recent work in String Theory \\cite{eva} indicates that non-locality may be required for all non-stationary, cosmologically relevant backgrounds (i.e.~Lorentzian vacua). We relate the findings of \\cite{eva} to our backgrounds in order to obtain a physical interpretation for the sensitivity of the CMB spectrum to Planck scale physics as originating from non-locality of time-dependent solutions in String Theory. \\par We use units with $c=\\hbar=1$ and denote by $m_p$ the Planck mass and by $\\ell_p=m_p^{-1}$ the Planck length. Finally, $k_B$ is the Boltzmann constant. ",
        "conclusions": ""
    },
    "0208/hep-ph0208222_arXiv.txt": {
        "abstract": "The lectures describe several cosmological effects produced by neutrinos. Upper and lower cosmological limits on neutrino mass are derived. The role that neutrinos may play in formation of large scale structure of the universe is described and neutrino mass limits are presented. Effects of neutrinos on cosmological background radiation and on big bang nucleosynthesis are discussed. Limits on the number of neutrino flavors and mass/mixing are given. ",
        "introduction": "}  Of all known particles neutrinos have the weakest interactions and the smallest possibly nonvanishing, mass. Thanks to these properties neutrino is the second most abundant particle in the universe after photons. According to observations the number density of photons in cosmic microwave background radiation (CMBR) is $n_\\gamma = 412$/cm$^3$. In standard cosmology the number density of cosmic neutrinos can be expressed through $n_\\gamma$ as \\be n_\\nu + n_{\\bar \\nu} = 3 n_\\gamma /11 = 112/{\\rm cm}^3 \\label{nnu} \\ee for any neutrino flavor ($\\nue$, $\\num$, and $\\nut$), assuming that there is an equal number of neutrinos and antineutrinos.  Knowing the temperature of CMBR, $T_\\gamma = 2.728\\, {\\rm K} = 2.35 \\cdot 10^{-4}$ eV, one can calculate the temperature of cosmic neutrinos: \\be T_\\nu = (4/11)^{1/3} T_\\gamma = 1.95\\,{\\rm K} = 1.68\\cdot 10^{-4} {\\rm eV} \\label{tnu} \\ee which is true if the neutrino mass is much smaller than their temperature, $m_\\nu \\ll T_\\nu$. Otherwise the parameter $T_\\nu$ does not have the meaning of temperature; up to a constant factor it can be understood as the inverse cosmological scale factor $a(t)$. Theory predicts that the spectrum of cosmic neutrinos, even massive ones, is given by the almost equilibrium form: \\be f_\\nu = \\left[\\, \\exp\\left(p/T - \\xi\\right) + 1 \\right]^{-1} \\label{fnu} \\ee with the dimensionless chemical potential $\\xi = \\mu /T$ usually assumed to be negligibly small. However one should note that in the expression above $p$ is the neutrino momentum, while in the equilibrium distribution there stands energy $E= \\sqrt{m^2_\\nu + p^2}$.  There is a small correction to expression (\\ref{fnu}) of the order of $(m_\\nu /T_d)^2$ where $T_d$ is the neutrino decoupling temperature, $T_d \\sim {\\rm MeV}$ - this is the temperature when neutrinos stopped to interact with primeval plasma. This correction appeared because at $T>T_d$ neutrinos were in equilibrium and their distribution depended on $E/T$. Distribution of noninteracting neutrinos should be a function of $p a(t)$. In the case of instantaneous decoupling it turns into $f(\\sqrt{(p/T)^2 + (m_\\nu/T_d)^2})$, while for non-instantaneous decoupling the dependence on mass could be different.  More details about cosmological neutrinos can be found e.g. in a recent review paper~\\cite{dolgov02}.  Neutrinos are normally assumed to possess only usual weak interactions with $(V-A)$-coupling to $W$ and $Z$ bosons. Correspondingly, if $\\mnu=0$, only left-handed neutrinos, $\\nu_L$, i.e. those with spin anti-parallel to their momentum (and parallel for $\\bar\\nu$), possess this interaction, while right-handed neutrinos, $\\nu_R$, are sterile. If $\\mnu \\neq 0$ right-handed neutrinos would be also coupled to intermediate bosons with the strength suppressed as $(\\mnu/E)^2$ and their cosmological number density would be always negligible since their mass is bounded from above by a few eV see eqs. (\\ref{gz},\\ref{gz-2}). If however neutrinos are mixed and massive (possibly with Majorana and Dirac masses) additional three sterile neutrinos could be abundantly produced in the early universe~\\cite{dolgov81}.  It is known from experiment that there are at least  three neutrino families (or flavors), $\\nue$, $\\num$, and $\\nut$. From LEP data the number of light neutrino flavors with $m_\\nu < m_Z/2$ is indeed three: \\be N_\\nu = 2.993 \\pm 0.011 \\label{nnu-lep} \\ee One can find references to original experimental papers in the Review of Particle Physics~\\cite{pdg}.  Direct experiment limits on neutrino masses are~\\cite{pdg} \\be \\mne < 3\\, {\\rm eV},\\,\\,\\, \\mnm < 190\\, {\\rm keV},\\,\\,\\, \\mnt < 18.2\\, {\\rm MeV} \\label{nu-masses} \\ee As we will see below, cosmology allows to derive an upper limit on masses of all neutrino flavors similar to that presented above for $\\mne$.  There is a strong evidence in favor of neutrino oscillations. The best fit solutions to the observed neutrino anomalies indicates maximum mixing between $\\num$ and $\\nut$ with mass difference about $3\\times 10^{-3}$ eV$^2$ (for explanation of atmospheric anomaly) and also large mixing between $\\nue$ and another active neutrino with mass difference between $10^{-3}-10^{-5}$ eV$^2$ (for explanation of the deficit of solar neutrinos). If mass differences are indeed so small then masses of all active neutrinos should be below 3 eV and right-handed neutrinos would not be practically produced in particle interactions in the standard theory, but as we noted below, they may be produced by oscillations.  Except for the above mentioned anomalies, and possibly LSND, neutrinos are well described by the standard electroweak theory. For a recent review of neutrino anomalies see e.g. ref.~\\cite{nu-osc}.  In what follows we discuss the bounds in neutrino masses that can be derived from the magnitude of cosmic energy density and large scale structure of the universe (sec. \\ref{s-mass}). Relation between cosmological neutrinos and CMBR is considered in sec. \\ref{s-cmbr}. In section \\ref{s-bbn} we describe the role played by neutrinos in big bang nucleosynthesis (BBN) and present the limits on the number of neutrino species and possible neutrino degeneracy. Cosmological impact of neutrino oscillations is considered in sec. \\ref{s-nuosc}. The body of the lectures is preceded by a brief presentation of basic cosmological facts and essential observational data (section~\\ref{s-cosmo}). These lectures present a shorter version of the recent review paper~\\cite{dolgov02} where one can find details and a long list or references, many of which are omitted here because of lack of space and time. ",
        "conclusions": ".}\\protect \\label{s-mass}}  \\subsection{Gerstein-Zeldovich limit \\label{ss-gz}}  Since the number density of neutrinos at the present day is known, see eq.~(\\ref{nnu}), it is easy to calculate their contribution into cosmological energy density, $\\rho_\\nu = \\sum m_{\\nu_a} n_\\nu$, if neutrinos are stable. Demanding that $\\rho_\\nu$ does not exceed the known value of energy density of matter we obtain \\be \\sum_a m_{\\nu_a} < 95\\,{\\rm eV}\\, \\Omega_m h^2 \\approx 14\\,{\\rm eV} \\label{gz} \\ee where the sum is taken over all light neutrino species, $ a=e,\\,\\mu,\\,\\tau$. This limit was originally derived by Gerstein and Zeldovich~\\cite{gz} in 1966. Six years later the result was rediscovered by Cowsik and McClelland~\\cite{cm}. In the later paper, however, the photon heating by $e^+e^-$-annihilation was not taken into account and both helicity states of massive neutrinos were assumed to be equally abundant. Correspondingly the resulting number density of relic neutrinos was overestimated by the factor 11/2.  If all active neutrinos are strongly mixed and their mass differences are very small (see the end of sec.~\\ref{s-intr}) then the limit (\\ref{gz}) for an individual mass would be $\\mnu < 4.7$ eV.  The bound (\\ref{gz}) can be noticeably strengthened because neutrino may make only sub-dominant contribution to $\\Omega_m$. Arguments based on large scale structure formation (see below sec.~\\ref{ss-lss}) lead to the conclusion that $\\Omega_\\nu < \\Omega_m/3$ and correspondingly: \\be \\sum_a m_{\\nu_a} < 5 \\,\\,{\\rm eV} \\label{gz-2} \\ee As noted above, in the case of small mass differences the mass bound for a single neutrino would be $m_\\nu < 1.7 $ eV.  \\subsection{Tremaine-Gunn limit \\label{ss-tg}}  Quantum mechanics allows to obtain a lower limit on neutrino mass if neutrinos make {\\it all} dark matter in galaxies, especially in dwarf ones~\\cite{tremaine79}. The derivation is based on the fact that neutrinos are fermions and hence cannot have an arbitrary large number density if their energy is bounded from above to allow formation of gravitationally bound cluster. So to make dominant contribution into dark matter neutrino mass should be larger than a certain value. Gravitationally bound neutrinos would be most densely packed if they form degenerate gas with Fermi momentum $p_f = m_\\nu V_F$. The Fermi velocity $V_F$ can be determined from the virial theorem: \\be V_F^2 = G_N M_{gal} /R_{gal} \\label{vf} \\ee where $G_N = 1/m_{Pl}^2$ is the Newton gravitational constant and $M_{gal}$ and $R_{gal}$ are respectively the mass and radius of a galaxy.  The number density of degenerate neutrinos and equal number of antineutrinos is $n_\\nu = p^3_F/(3\\pi^2)$ and correspondingly their total mass in a galaxy is \\be M_\\nu = 4\\pi R_{gal}^3 m_\\nu n_\\nu /3 \\label{Mnu} \\ee According to observations galactic masses are dominated by invisible matter, so one should expect that $M_\\nu \\approx M_{gal}$. From the equations above we find: \\be m_\\nu = 80\\,{\\rm eV}\\,\\left({300\\,{\\rm km/sec} \\over V }\\right)^{1/4} \\left({ 1 \\, {\\rm kpc} \\over R_{gal}}\\right)^{1/2} \\label{mnutg} \\ee For dwarfs $R_{gal} \\approx 1$ kpc and $V \\approx 100$ km/sec. Correspondingly neutrinos, if they constitute all dark matter in such galaxies, should be rather heavy, $m_\\nu > 100$ eV in contradiction with Gerstein-Zeldovich limit. Thus we have to conclude that dark matter in galaxies is dominated by some other unknown particles.  \\subsection{Neutrinos and large scale structure of the universe \\label{ss-lss}}  Though, as we saw above, massive neutrinos cannot be dominant dark matter particles, they may play an essential role in large scale structure formation and evolution. According to the accepted point of view cosmological structures have been developed as a result of gravitational instability of initially small primordial density perturbations. The latter presumably were generated at inflationary stage due to rising quantum fluctuations of the inflaton field. For reviews and list of references see e.g.~\\cite{bellido00}. It is usually assumed that the spectrum of initial density perturbation has a simple power law form, i.e. Fourier transform of the density perturbations \\be (\\delta \\rho /\\rho)_{in} =\\int d^3 k \\,\\delta (k) \\label{deltarho} \\ee behaves as  $\\delta^2 \\sim k^n$. Moreover, the value of the exponent, $n$, is usually taken to be 1. It corresponds to flat or Harrison-Zeldovich spectrum~\\cite{hz}, as indicated by inflation and consistent with observations.  With the known initial perturbations and equation of state of cosmological matter one can calculate the shape of the evolved spectrum and to compare it with observations. This permits to determine the properties of the cosmological dark matter. In the case of neutrinos density perturbations at small scales are efficiently erased as can be seen from the following simple arguments. Neutrinos were decoupled from plasma when they relativistic. The decoupling temperature is  $T^d \\sim $ MeV, while $\\mnu \\leq 10$ eV. Thus after decoupling neutrinos free streamed practically with the speed of light. Since the flux of neutrinos from neutrino-rich regions should be larger than that from neutrino-poor regions, the inhomogeneities in neutrino distribution would smoothed down at the scales smaller than neutrino free path, $l_{fs}= 2t_{nr}$. Here $t_{nr}$ is the cosmic time from beginning till the moment when neutrinos became nonrelativistic. As we mentioned above neutrinos propagate with the speed of light, so locally their path is equal just to $t$ and factor 2 came from the expansion of the universe. The mass contained inside $l_{fs}$ is \\be M_{fs} = {4 \\pi (2t_{fs})^3 \\over 3} \\, \\rho = m_{Pl}^2 t_{fs} \\label{mfs} \\ee where we used for the cosmological energy density the critical value (\\ref{rhoc}) with the Hubble parameter $H=1/(2t)$, as given by eq.~(\\ref{HofT}). Assuming that the universe was dominated by relativistic matter (photons and three neutrino flavors) till neutrino temperature dropped down to $T_\\nu = \\mnu /3$ and taking into account that $T_\\nu \\approx 0.7 T_\\gamma$ (\\ref{tnu}) we find that the mass inside the free-streaming length is \\be M_{fs} = 0.1 m_{Pl}^3 /\\mnu^2 \\approx 10^{17} M_\\odot ({\\rm eV} /\\mnu)^2 \\label{mfsnum} \\ee where $M_\\odot = 2\\cdot 10^{33}$ g is the solar mass. This result is derived for the case of one neutrino much heavier than the others. It would be modified in an evident way if neutrinos are mass degenerate.  In such a theory the characteristic mass of the first formed objects, $M_{fs}$, is much larger than the mass of large galaxies, $M_{gal} \\sim 10^{12} M_\\odot$ and dark matter with such property is called hot dark matter (HDM). The dark matter particles for which the characteristic mass is smaller than the galactic mass are called cold dark matter (CDM) and the intermediate case is naturally called warm (WDM). In HDM model of structure formation large clusters of galaxies should be formed first and smaller structures could be created from larger ones later by their fragmentation. However such process demands too much time and, moreover, the observations indicate that smaller structures are older. Together with Tremaine-Gunn limit discussed in sec.~\\ref{ss-tg}, it ``twice'' excludes neutrinos as dominant part of dark matter in the universe. However mixed models with comparable amount of CDM and HDM are not excluded. Though the mystery of cosmic conspiracy - why different particles have comparable contribution to $\\Omega$ - becomes in this case even more pronounced: \\be \\Omega_{vac}\\sim (\\Omega_m =\\Omega_{CDM}+\\Omega_{HDM}) \\sim \\Omega_b \\label{Omega-a} \\ee  From the arguments presented here one can see that the larger is the fraction of neutrinos in the total mass density of the universe the smaller should be power in cosmic structures at small scales. This permits to strengthen the upper limit on neutrino mass. Especially sensitive to neutrino mass are the structures at large red-shift $z$ because in neutrino dominated universe small structures should form late and should not exist at large $z$. The neutrino impact on the structure formation was analyzed in refs.~\\cite{nustr} with the typical limits between 1 and 5 eV. More detailed discussion and more references can be found in the review~\\cite{dolgov02}. According to ref.~\\cite{sdss99} Sloan Digital Sky Survey is potentially sensitive to $\\mnu \\leq 0.1$ eV.  \\subsection{ Cosmological limit on heavy neutrino mass \\label{ss-nuh}}  If there exists fourth lepton generation then the corresponding neutrino should be heavier than $m_Z/2 = 45$ GeV to surpass the LEP result~(\\ref{nnu-lep}). If these heavy neutrinos are stable on cosmological time scale, $\\tau_\\nu \\geq t_U \\sim 10^{10}$ years, then their mass density may be cosmologically noticeable. Since such neutrinos are assumed to be very heavy their number density at decoupling should be Boltzmann suppressed and they may escape Gerstein-Zeldovich limit. First calculations of cosmological number density of massive particles were performed by Zeldovich in 1965~\\cite{zeldovich65}. However his result contained a numerical error later corrected in ref.~\\cite{zop65}. The same approach was applied to the calculations of the number/energy density of relic heavy neutrinos practically simultaneously in the papers~\\cite{vdz} where it was found that the mass of heavy neutrino should be above 2.5 GeV to be cosmologically safe.  The number density of massive particle (neutrinos)s which survived annihilation is inversely proportional to the annihilation cross-section $\\sigma_{ann}$ and is approximately given by the expression \\be n_\\nu / n_\\gamma  = \\left( \\sigma_{ann}\\,v\\,m_\\nu m_{Pl} \\right)^{-1} \\label{nmnu} \\ee where $v$ is the c.m. velocity of the annihilating particles and $n_\\gamma$ is the number density of photons in CMBR. For relatively light neutrinos, $\\mnu \\ll m_Z$ (which is not realistic now), the annihilation cross-section is proportional to $\\sigma \\sim \\mnu^2$ and the energy density of heavy relic neutrinos drops as $1/\\mnu^2$. According to the calculations quoted above $\\rho_\\nu = \\rho_c$ for $\\mnu = 2.5 $ GeV.  For higher masses $\\mnu > m_{W,Z}$, the cross-section started to drop as $\\sigma \\sim \\alpha^2 / \\mnu^2$ and the cosmologically allowed window above 2.5 GeV becomes closed for $\\mnu >$ (3-5) TeV~\\cite{dz80}. However for $\\mnu >m_W$ a new channel of annihilation becomes open, \\be \\nuh+\\bar\\nuh \\rar W^+W^- \\label{nuW} \\ee with the cross-section rising as $m_{\\nuh}$~\\cite{ekm89}. The rise of the cross-section is related to the rise of the Yukawa couplings of Higgs boson which is necessary to ensure a large mass of $\\nuh$. Correspondingly the excluded region above a few TeV becomes open again. However the annihilation (\\ref{nuW}) proceeds only in one lowest partial wave and the cross-section is restricted by the unitarity limit~\\cite{griest90}, \\be \\sigma_J < \\pi (2J+1)/p^2. \\label{sigma0} \\ee If one assumes that this limit is saturated then the large values $m_{\\nuh}$ about 100 TeV would be forbidden. In reality the limit should be somewhat more restrictive because it is natural to expect that the cross-section started to drop with rising mass of neutrino before it reaches the unitarity bound. However it is very difficult, if possible at all, to make any accurate calculations in this strong interaction regime.   To summarize this discussion, the cosmic energy density, $\\rho_{\\nuh}$, of heavy neutrinos with the usual weak interaction is sketched in fig. (\\ref{rholfig}). In the region of very small masses the ratio of number densities $n_{\\nuh}/n_\\gamma$ does not depend upon the neutrino mass and $\\rho_{\\nuh}$ linearly rises with mass. For larger masses $\\sigma_{ann} \\sim \\mnh^2$ and $\\rho_{\\nuh}\\sim 1/\\mnh^2$. This formally opens a window for $\\mnh$ above 2.5 GeV. A very deep minimum in $\\rho_{\\nuh}$ near $\\mnh = m_Z /2$ is related to the resonance enhanced cross-section around $Z$-pole. Above $Z$-pole the cross-section of $\\bar \\nuh \\nuh$-annihilation into light fermions goes down with mass as $\\alpha^2/\\mnh^2$ (as in any normal weakly coupled gauge theory). The corresponding rise in $\\rho_{\\nuh}$ is shown by a dashed line. However for $\\mnh > m_W$ the contribution of the channel $\\bar \\nuh \\nuh \\rar W^+W^-$ leads to the rise of the cross-section with increasing neutrino mass as $\\sigma_{ann} \\sim \\alpha^2 \\mnh^2 /m_W^4$. This would allow keeping $\\rho_{\\nuh}$ well below $\\rho_c$ for all masses above 2.5 GeV. The behavior of $\\rho_{\\nuh}$, with this effect of rising cross-section included, is shown by the solid line up to $\\mnh =1.5 $ TeV. Above that value it continues as a dashed line. This rise with mass would break unitarity limit for partial wave amplitude when $\\mnh$ reaches 1.5 TeV (or 3 TeV for Majorana neutrino). If one takes the maximum value of the S-wave cross-section permitted by unitarity (\\ref{sigma0}), which scales as $1/\\mnh^2$, this would give rise to $\\rho_{\\nuh} \\sim \\mnh^2$ and it crosses $\\rho_c$ at $\\mnh \\approx 200$ TeV. This behavior is continued by the solid line above 1.5 TeV. However for $\\mnh \\geq {\\rm a\\,\\, few}\\,\\, {\\rm TeV}$ the Yukawa coupling of $\\nuh$ to the Higgs field becomes strong and no reliable calculations of the annihilation cross-section has been done in this limit. Presumably the cross-section is much smaller than the perturbative result and the cosmological bound for $\\mnh$ is close to several TeV. This possible, though not certain, behavior is presented by the dashed-dotted line. One should keep in mind, however, that the presented results for the energy density could only be true if the temperature of the universe at an early stage was higher than the heavy lepton mass.  \\begin{figure}[htb] \\begin{center} \\leavevmode \\hbox{ \\epsfysize=3.0in \\epsffile{rhol1.eps}} \\end{center} \\caption{Cosmological energy density of massive neutrinos $\\Omega = \\rho_{\\nuh} /\\rho_c$ as a function of their mass measured in eV. The meaning of different lines is explained in the text. \\label{rholfig}} \\end{figure}"
    },
    "0208/hep-ph0208152_arXiv.txt": {
        "abstract": "Large scale magnetic fields represent a triple point where cosmology, high-energy physics and astrophysics meet for different but related purposes. After reviewing the implications of large scale magnetic fields in these different areas, the r\\^ole of primordial magnetic fields is discussed in various physical processes occurring prior to the decoupling epoch with particular attention to the big bang nucleosynthesis (BBN) epoch and to the electroweak (EW) epoch. The generation of matter--antimatter isocurvature fluctuations, induced by hypermagnetic fields, is analyzed in light of a possible increase of extra-relativistic species at BBN. It is argued that stochastic GW backgrounds can be generated by hypermagnetic fields at the LISA frequency. The problem of the origin of large scale magnetic fields is also scrutinized. ",
        "introduction": "Through the last fifty years, the possible existence and implications  of primordial magnetic fields became a very useful cross-disciplinary area at the interface of cosmology, astrophysics and high-energy physics\\cite{rev1,rev2,rev3,rev4,rev5,rev6,rev7}. From astrophysical observations, we do know that planets, stars, the interstellar medium and the intergalactic medium are all magnetized. The magnetic fields in these environments have values ranging from the $\\mu $ G of the intra-cluster medium, to the G  (in the case of the earth) up to presumably $10^{12}$ G, the typical magnetic fields of neutron stars. In principle, observations of magnetic fields in galaxies (and clusters of galaxies) could discriminate between a direct primordial origin of the large-scale fields and a primordial origin mediated by a dynamo amplification. None of the two options are, at the moment, supported by clear observational evidence. Furthermore, in both approaches there  are various theoretical assumptions which have been (but still need to be) carefully scrutinized.  In high-energy physics, the possible existence of intergalactic magnetic fields is one of the crucial unknowns in the analysis of ultra-high energy cosmic rays above the GZK cut-off. The interplay of high-energy physics and astrophysics is indeed present since the origin of this subject. In 1949 the  scientific argument between Fermi \\cite{fermi}, on one side,and Alfv\\'en \\cite{alv1,alv2}, Richtmyer and  Teller \\cite{alv3}, on the other, concerned exactly the possible existence of galactic magnetic fields. Fermi was convinced that high-energy cosmic rays are in equilibrium with the whole galaxy while Alfv\\'en was supporting the idea that high energy cosmic rays are in equilibrium with stars. In order to make his argument consistent, Fermi postulated (rather than demonstrated) the existence of a $\\mu$ G galactic magnetic field. Fermi thought that the origin of this field was  primordial.  In cosmology the possible existence of magnetic fields prior to decoupling can influence virtually all the moments in the thermodynamical history of the Universe. Big-bang nucleosynthesis (BBN), electroweak phase transition (EWPT), decoupling time are all influenced by the existence of magnetic fields at the corresponding epochs. If magnetic fields were originated in the past history of the Universe, their birth should be related, in some way, to the interplay of gravitational and gauge interactions. Superstring theories and higher-dimensional theories, formulated through the past thirty years, pretend to give us some hints on the possible form of such an interplay and their implications may be useful to consider.  The present paper is organized as follows. In Section II the basic ideas on large scale magnetic field structure and observations will be briefly outlined. Section III collects some considerations on the evolution of magnetic fields. In Section IV the problem of the origin will be illustrated with particular attention to models where there is an effective evolution of the gauge coupling. Section V deals with the possible implications of hypermagnetic fields for the EW physics and for the generation of the BAU. In Section VI it will be shown that if hypermagnetic fields are present at the EW epoch, matter--antimatter fluctuations are likely to be produced at BBN. In Section VII the implications of hypermagnetic fields for the GW backgrounds at the LISA and VIRGO/LIGO frequencies will be discussed. In Section VIII some speculations on the possible Faraday rotation of the CMB polarization will be presented. Section IX contains some concluding remarks.   \\renewcommand{\\theequation}{2.\\arabic{equation}} \\setcounter{equation}{0} ",
        "conclusions": "The large scale magnetic fields observed today in the Universe may or may not be primordial and there could indeed be different possibilities. It could be that in the past history of the Universe very strong magnetic fields have been created. These fields could be  strong enough to affect phase transitions and other phenomena in the life of the Universe but, at the same time, too weak to be responsible for the origin of large scale magnetic fields. It could also be that magnetic field were indeed strong enough to act as seeds of presently observed magnetic fields and, in this case we should be able to find evidence that this was indeed the case. In light of this perspective various ``observables'' , possibly affected by the existence of primordial magnetic fields, could be proposed. They include the stochastic GW backgrounds, the Faraday rotation of CMB and the baryon asymmetry of the Universe. From a more theoretical perspective, primordial magnetic fields can be connected to the exsistence of (small and large) extra-dimensions and to the possible dynamics of gauge couplings in the early stages of the evolution of the Universe."
    },
    "0208/astro-ph0208467_arXiv.txt": {
        "abstract": "We use data from the BIMA Survey of Nearby Galaxies (SONG) to investigate the relationship between ellipticity and central mass concentration in barred spirals. Existing simulations predict that bar ellipticity decreases as inflowing mass driven by the bar accumulates in the central regions, ultimately destroying the bar. Using the ratio of the bulge mass to the mass within the bar radius as an estimate of the central mass concentration, we obtain dynamical mass estimates from SONG CO 1-0 rotation curve data.  We find an inverse correlation between bar ellipticity and central mass concentration, consistent with simulations of bar dissolution. ",
        "introduction": "Bars exert gravitational torques on the gas in the disks of spiral galaxies, resulting in gas inflow towards the center (Quillen et al. 1995; Regan, Vogel, \\& Teuben 1997; Regan, Sheth, \\& Vogel 1999). This results in a significant increase in the gas mass in the center of a bar (Sakamoto et al. 1999; Sheth 2001), often leading to increased star formation and even starburst activity in the nucleus (Ho, Filippenko, \\& Sargent, 1997; Jogee, Kenney, \\& Smith 1999). Simulations predict that the increased mass concentration may affect the bar structure and even dissolve the bar itself (Kormendy 1982; Hasan, \\& Norman 1990; Friedli, \\& Pfenniger 1991; Friedli \\& Benz 1993; Hasan, Pfenniger, \\& Norman 1993; Norman, Sellwood, \\& Hasan 1996).  In this paper we investigate whether there is observational evidence for the change in bar shape with mass concentration in the centers of spiral galaxies.  We have used the photometric data of a sample of 13 barred galaxies from the BIMA Survey of Nearby Galaxies (SONG) to determine the bar structure and the CO(1-0) rotation curves to derive central mass concentrations.  We use the bar ellipticity $(1-b/a)$, where $a$ is the semi-major axis and $b$ the semi-minor axis, to be a measure of the bar structure. We define the central mass concentration `\\fc'~ as the ratio of the dynamical mass within the bulge to that within the bar radius. The bulge is the most physically distinct region in the galaxy center and easier to measure than other length scales such as core radius which is used to define central mass concentration in numerical studies (e.g. Norman et al.  1996). We discuss the justification for using the bulge mass in more detail in \\S 5.  To determine \\fc, we have used the rotation curves derived from the CO J=1-0 emission in the galaxies. CO rotation curves were used because CO traces the kinematics of cold molecular gas, which moves along closed orbits in the plane of a galaxy and hence is a good tracer of the dynamical mass distribution in galaxies.  In \\S 2 we describe the galaxy sample and the observational data used in the analysis. In \\S 3 we discuss how we derived bar ellipticities and in \\S 4 we determine \\fc\\ in our sample of galaxies. The statistical analysis is presented in \\S 5 and we discuss the significance of the results in \\S 6. We list our conclusions in \\S 7. ",
        "conclusions": "We have used the BIMA SONG survey data to determine ellipticities and mass concentrations in the centers of nearby barred galaxies. We have used optical or near-infrared images to determine bar shapes and the CO (1-0) rotation curves to derive dynamical masses in the bulge and bar regions of the galaxies.  \\\\ 1) We find an apparent correlation between the bar ellipticity and the central mass concentration. For our sample of 13 galaxies a conservative analysis yields a correlation coefficient of $\\sim -0.8$. The probability that the parent sample is uncorrelated is 0.012, which indicates that it is a statistically significant correlation.  \\\\ 2) The correlation suggests that bar structure is affected by the dynamical mass concentration in the bulge. This may provide evidence that bars evolve as gas flows inwards and mass accumulates in their centers, indicating that the mass concentration affects the bar structure and may eventually dissolve the bar."
    },
    "0208/astro-ph0208184_arXiv.txt": {
        "abstract": "We present a statistical study of the post-formation migration of giant planets in a range of initial disk conditions. For given initial conditions we model the evolution of giant planet orbits under the influence of disk, stellar, and mass loss torques. We determine the mass and semi-major axis distribution of surviving planets after disk dissipation, for various disk masses, lifetimes, viscosities, and initial planet masses. The majority of planets migrate too fast and are destroyed via mass transfer onto the central star. Most surviving planets have relatively large orbital semi-major axes of several AU or larger. We conclude that the extrasolar planets observed to date, particularly those with small semi-major axes, represent only a small fraction ($\\sim$25\\% to 33\\%) of a larger cohort of giant planets around solar-type stars, and many undetected giant planets must exist at large ($>$1-2~AU) distances from their parent stars. As sensitivity and completion of the observed sample increases with time, this distant majority population of giant planets should be revealed. We find that the current distribution of extrasolar giant planet masses implies that high mass (more than 1-2~Jupiter masses) giant planet formation must be relatively rare. Finally, our simulations imply that the efficiency of giant planet formation must be high: at least 10\\% and perhaps as many as 80\\% of solar-type stars possess giant planets during their pre-main sequence phase. These predictions, including those for pre-main sequence stars, are testable with the next generation of ground- and space-based planet detection techniques. ",
        "introduction": "}  Extrasolar giant planets (EGPs) have been detected by the radial velocity method at orbital distances from several AUs (e.g., 47~UMa~c, Gl614b, $\\epsilon$~Eri~b, 55~Cnc~d) to several hundredths of an AU (51~Peg~b et al.) from their central stars (see Mayor \\& Queloz \\cite{mq}; Butler \\& Marcy \\cite{bm}; Butler et al. \\cite{butler97}; Noyes et al. \\cite{noyes}; Cochran et al. \\cite{cochran}; Santos et al. \\cite{santos}; Vogt et al. \\cite{vogt}; Hatzes et al. \\cite{hatzes}; Butler et al. \\cite{butler2001}; Santos et al. \\cite{santos2001}; Fischer et al. \\cite{fischer2002}; Marcy et al. \\cite{marcy2002}; and many others; see also the review by Marcy et al. \\cite{marcyppiv}). Unless giant planets form in place within 1~AU of low mass stars -- unlikely in the context of published formation models (see, for example, Guillot et al. \\cite{guillot96}) -- the observed range of orbital semi-major axes implies that dramatic orbital changes occur after formation. Previous work showed that in gaseous disks and even subsequent particulate disks, giant planets can move from formation distances of around 5~AU to a wide range of final distances, a process known as orbital migration (Lin \\& Papaloizou \\cite{lp86}; Lin et al. \\cite{linetal}; Takeuchi et al. \\cite{takeuchi}; Ward \\cite{ward97a,ward97b}; Trilling et al. \\cite{trillingetal}; Murray et al. \\cite{murray}; Bryden et al. \\cite{bryden}; Kley \\cite{kley99,kley2000}; Del Popolo et al. \\cite{delp}; Tanaka et al. \\cite{tanaka}). The first direct determination of the radius of an extrasolar giant planet via transit measurement (Charbonneau et al. \\cite{c00}, Henry et al. \\cite{h00}) supports rapid inward migration of giant planets, since the large planetary radius requires close proximity to the parent star when the planet's internal entropy was much larger than at present (Guillot et al. \\cite{guillot96}, Burrows et al. \\cite{burrad}), and formation in place is generally considered implausible (see, e.g., Guillot et al. \\cite{guillot96}). Early inward migration is commensurate with the idea of migration caused by disk-planet interactions, as disk lifetimes are not longer than $10^7$~years (e.g., Zuckerman et al. \\cite{zuck}).  In this work, we address the following questions: (1) How efficient is orbital migration? (2) What population of planets survives the migration process? and (3) How does this produced population compare to the observed EGP population? We answer these questions by allowing a large population of giant planets to evolve and migrate in circumstellar disks with various initial conditions (one planet per disk) and determining the final semi-major axis and mass distributions. We do not employ any stopping mechanisms, but instead allow only those planets to survive whose migration timescales are longer than their disks' lifetimes. We compare the final semi-major axis and mass distributions of the surviving model planets to those of the observed EGPs. We reproduce the fraction of planets in small orbits, and predict the frequency of and orbital distributions for giant planets that are as yet unobservable. Of those stars which do form giant planets, a few percent should have giant planets ultimately residing in small orbits, and around one quarter of stars which form planets retain planets in orbits of semi-major axis several AU and beyond. Given the current discovery statistics and their uncertainties, our results imply that giant planet formation is very efficient around low-mass stars: we find that 10\\% to 80\\% of young stars form planets, with the uncertainty dominated by the initial planetary mass distribution (and with some poorly-known uncertainties associated with discovery statistics). The presently detectable portion of the giant planet population represents only about 25\\% to 33\\% of the total extant population of giant planets.  The model we present here is a simple one with a minimum set of physical assumptions. By neglecting specific stopping mechanisms, we provide results that represent a baseline for planet formation statistics, bereft of specific additional assumptions that any given stopping mechanism would require. It is hoped that the calculations and results presented here are transparent enough that others can use them to explore the particular effects that specific assumed stopping mechanisms would have. In short, we use the simplest possible model in order to explore the consequences of and for planet formation. ",
        "conclusions": "}   We have presented a simple, baseline model of planet migration. Our simple model allows us to study the statistical behavior of planets migrating in disks without complications introduced from relatively unconstrained processes, like stopping mechanisms. Our overall results can be useful as a baseline statistical result of planet formation and survival. We have shown that most giant planets, under nominal initial conditions, migrate rapidly relative to the disk lifetime and are destroyed before the circumstellar disk dissipates. We have found that the observed extrasolar planets must represent only the tip of the iceberg -- perhaps 25\\% to 33\\% of the total extant population has been detected -- based on the required distribution of initial parameters. Most planets which have formed and survived reside outside the current detection limit of the radial velocity searches (a similar conclusion was reached by Armitage et al. \\cite{arm}); this prediction is supported with new detections reported in Vogt et al. (\\cite{vogt2002}). We have shown that the planetary initial mass distribution must be biased toward smaller masses in order to produce the shape of the observed EGP mass distribution, yielding our preferred initial mass distribution (see also Armitage et al. \\cite{arm}); again, recent detections by Vogt et al. (\\cite{vogt2002}) support the conclusion that the lowest mass planets are the most common. Finally, because migration in high mass disks destroys almost all giant planets formed therein, low mass disks -- within a factor of~10 or so of the mass of Jupiter -- must be capable of forming giant planets (essentially in agreement with recent observational results by Carpenter (\\cite{carp})). Hence we find that giant planet formation must be a relatively efficient process in disks. Because high mass planets migrate more slowly than low mass planets in a given disk, overall there should be more massive planets at intermediate and large semi-major axes than found close-in. The population of close-in planets, in contrast, should be dominated by smaller mass planets. Zucker \\& Mazeh (\\cite{zm2002}) have shown that this effect is, in fact, observed and statistically significant.  Migrating giant planets may be detrimental to terrestrial planet survival, if terrestrial planets form coevally with giant planets. Planets interior to a migrating giant planet would be disrupted and lost from the system. This of course assumes that smaller planets do not migrate, although they too likely migrate, potentially on even shorter timescales than giant planets (Ward \\cite{ward97a,ward97b}). If terrestrial planets form after the dissipation of gas in the protoplanetary disk, then disruption by a migrating giant planet may be less of a risk (excepting giant planet migration caused by planet-planetesimal interaction (Murray et al. \\cite{murray})). Kortenkamp \\& Wetherill (\\cite{kort}) have considered the case of terrestrial planet formation when Jupiter has both its current and a larger heliocentric distance of 6.2~AU. They have found that accumulation of rocky bodies may be easier with Jupiter at a larger heliocentric distance (that is, pre-migration). It is possible that formation of the terrestrial planets in our Solar System may reveal clues about giant planet migration in our planetary system; certainly, these studies are also relevant to the formation of small, rocky planets in other planetary systems. In the context of large scale migrations, if terrestrial planet formation requires that giant planets have not migrated through the terrestrial zone around 1~AU, then only around~20\\% (flat initial mass distribution) or~3\\% (preferred) of all planet-forming systems qualify (i.e., a few percent of all late-type stars; see sections~\\ref{stats} and \\ref{eff}). If, however, the constraint is merely that there be no giant planet in the immediate vicinity of the terrestrial planet zone % at the onset of terrestrial planet formation (perhaps $10^7$~years), then the vast majority (93\\% for flat initial mass distribution, 99\\% for preferred) of planet-forming systems qualify. This number corresponds to nearly all planet-forming systems which is around~10\\% to~80\\% of late-type stars.  Although radial velocity detections of more distant giant planets will become possible as the time baseline of observations increases, astrometric techniques are more sensitive to giant planets in large orbits. % SIM, the Space Interferometry Mission (Danner \\& Unwin \\cite{sim}), will do a thorough job of detecting giant planets, Uranus-mass objects, and even smaller bodies from small to large semi-major axes ($\\sim$10~AU), with maximum sensitivity achieved for planets approximately 0.3~${\\rm M_{Earth}}$ around 3-5~AU. SIM's ability to test our predictions of the preferred initial giant planet mass distribution will be limited largely by mission lifetime. However, SIM can also analyze disks around young stars with high precision (target resolution of 1~microarcsecond), perhaps mapping out the signature of gaps created by migrating (or non-migrating) giant planets and giving us a rough snapshot of the time-dependent mass distribution of planets during the migration phase itself. (A 1~AU gap at 100~parsecs is 10~milliarcseconds.) For giant planets in the largest orbits, i.e., 20-30 AU from their parent star, direct imaging techniques may be the only practical method for detection since astrometric techniques would require baselines of decades or more.  Our results suggest that techniques to study planet formation around young stars -- radial velocity; high resolution imaging of young stellar systems; searches for gaps such as with SIM and potentially also the Space InfraRed Telescope Facility (SIRTF); searches for planetary outflows (Quillen \\& Trilling \\cite{quillen}); or searches for other indirect evidence, like cometesimals scattered onto stars (Quillen \\& Holman \\cite{qh}) -- should ultimately have a very high success rate. We anticipate that observational data will show that planet formation is taking place around 10\\%~to~80\\% of low mass pre-main sequence stars, and that planet searches around main sequence stars will have a much lower success rate. Planet formation is an ``easy come, easy go'' business, with many planets created and many planets destroyed, and with an important minority -- including our own Jupiter -- surviving."
    },
    "0208/astro-ph0208521_arXiv.txt": {
        "abstract": "We study the solar-cycle variations of solar p-mode travel time for different wave packets to probe the magnetic fields at the base of the solar convection zone. We select the wave packets which return to the same spatial point after traveling around the Sun with integral number of bounces. The change in one-bounce travel time at solar maximum relative to minimum is approximately the same for all wave packets studied except a wave packet whose lower turning point is located at the base of the convection zone. This particular wave packet has an additional decrease in travel time at solar maximum relative to other wave packets. The magnitude of the additional decrease in travel time for this particular wave packet increases with solar activity. This additional decrease in travel time might be caused by the magnetic field perturbation and sound speed perturbation at the base of the convection zone. With the assumption that this additional decrease is caused only by the magnetic field perturbation at the base of the convection zone, the field strength is estimated to be about $4-7\\times 10^{5}$ gauss at solar maximum if the filling factor is unity. We also discuss the problem of this interpretation. ",
        "introduction": "Observations give evidence that magnetic fields on the Sun emerge from below. How and where magnetic fields are generated is a long standing unanswered question in astronomy \\cite{cow34,par55,bab61}. It has been suggested that the boundary between the radiative zone and convection zone (CZ) is the best location for an oscillatory solar dynamo \\cite{spi80,par93,cha97}. Many attempts have been made to detect the magnetic fields in this region \\cite{gou96,bas97,how99,bas00,bas01,eff01,ant01}. Until now no clear evidence of magnetic field in this region has been found.  Here we use the technique of time-distance analysis \\cite{duv93} to measure solar cycle variations of travel time of acoustic waves with different ray paths to probe the magnetic fields at the base of the CZ.  A resonant solar p-mode is trapped and multiply reflected in a cavity between the surface and a layer in the solar interior. The acoustic signal emanating from a point at the surface propagates downward to the bottom of the cavity and back to the surface at a different horizontal distance from the original point.  Different p-modes have different paths and arrive at the surface with different travel times and different distances from the original point. The modes with the same angular phase velocity have approximately the same ray path and form a wave packet. The relation between the travel time and travel distance of a wave packet can be measured by using the temporal cross-correlation between the time series at two points \\cite{duv93}.  Different wave packets penetrate into different depths: the wave packet with a larger phase velocity penetrates into a greater depth. Time-distance analysis measures the travel times of different wave packets to probe the interior of the Sun at different depths \\cite{duv96,kos96,kos00}. The ray path of wave packet computed from a standard solar model with the ray theory for three different phase velocities is shown in Figure 1. If a magnetic field is present at the base of the CZ, it has different effects on different wave packets. It changes the travel time of wave packets which can penetrate into the base of the CZ, while it has no effect on the wave packets which can not reach the base of the CZ. If the magnetic fields at the base of the CZ vary with the solar cycle like the surface magnetic fields, travel time is expected to vary with the solar cycle as well. The change in travel time due to the magnetic fields at the base of the CZ is small because the ratio of magnetic pressure to gas pressure is small. However, the change in travel time increases linearly with the number of bounces between the boundaries of the cavity. Thus the strategy is to measure the change in multiple-bounce travel time. Here we measure the time for a wave packet to travel around the Sun to come back to the same spatial point. If a wave packet takes $N$ bounces to travel around the Sun, the change in travel time would increase by a factor of $N$ relative to the change in one-bounce travel time. Therefore, the problem becomes measuring solar cycle variations of travel time with the auto-correlation function of the time series at the same spatial point. In this study we use two different approaches, the multiple-bounce travel time analysis (MBTTA) and the power spectrum simulation analysis (PSSA), to measure the travel time of wave packets. ",
        "conclusions": "To test whether the shorter travel time at $N=8$ is caused by the analysis procedure, we did the following test. The frequency difference between solar maximum and minimum is smoothed by a fit in the ($l, \\nu$) domain. The mode frequencies at solar maximum are simulated by adding this smooth function to the mode frequencies at minimum. Applying the same procedure as in PSSA to the measured frequencies at solar minimum and the simulated frequencies at solar maximum, we compute the change in travel time for different $N$'s. The result shows that the change in travel time is approximately the same for all $N$'s and there is no additional decrease at $N=8$. This test indicates that the shorter travel time at $N=8$ is not caused by the analysis procedure.  It is unlikely that the shorter travel time at $N=8$ is caused by the spatial distribution of the near-surface magnetic fields. The most prominent spatial pattern of magnetic activity on the surface is the active latitudinal band in each hemisphere. The separation between the centroids of two active bands is about $42^\\circ$ in 1998 and monotonically decreases to about $29^\\circ$ in 2000 (from Greenwich Sunspot Data). If the latitudinal distribution of active regions can cause an additional decrease in one-bounce travel time, it would occur at $N=9$ in 1998 and shift to $N=12$ in 2000. This contradicts to the PSSA results shown in Figure 2. Thus it is unlikely that the shorter travel time at $N=8$ is caused by the separation of two active latitudinal bands on the surface. Since the active longitudes is less prominent than the active latitudes, it is unlikely the anomaly at $N=8$ is caused by the active longitudes.  The fact that the ray path of the wave packet of $N=8$ has the lower turning point at the base of the CZ as shown in Figure 1 suggests that the additional decrease in travel time at $N=8$ may be caused by the solar-cycle varying wave speed at the base of the CZ. The change in wave speed at the base of the CZ could be caused by magnetic field perturbation or/and sound speed ($[\\gamma p/\\rho]^{1/2}$) perturbation. It has been shown that the global measurements can not distinguish these two effects \\cite{zwe95}. The previous study \\cite{kos97} has also indicated that either magnetic field perturbation or sound speed perturbation alone would cause a change in travel time not only for $N=8$ but also for $N<8$, though it is smaller. Thus either magnetic field perturbation or sound speed perturbation alone can not explain the measurements of travel time variation shown in Figure 2.  However, the combination of these two perturbations might be able to explain the measured travel time variations. Although we do not know the mechanism producing the sound speed perturbation, it probably has the magnetic origin because the presence of a magnetic field could change the thermal structure and leads to a change in sound speed.  With the above caution in mind, we will estimate the field strength based on the assumption that the additional decrease in travel time at $N=8$ is caused only by the magnetic fields at the base of the CZ. If we adopt the value of $\\Delta\\tau_8$ measured with MBTTA and PSSA, the fraction of change in travel time due to the wave speed perturbation at the base of the CZ is about $0.015/\\tau_8-0.053/\\tau_8 \\approx 2.6-9\\times 10^{-6}$. If we use the half width of the tachocline, $0.025 R_{\\odot}$, as the width of the magnetic layer at the base of the CZ \\cite{cor01}, the fraction of change in wave speed at the base of the CZ, $\\delta w /w$, is about $2.6-9\\times 10^{-5}$ because the wave packet of $N=8$ spends about one tenth of time inside the tachocline. If the change in wave speed is entirely due to the presence of magnetic fields, the fraction of change in wave speed is $\\delta w/w = \\sin^2\\theta(v^2_A/c^2)/2$, where $v_A=B/(4\\pi\\rho)^{1/2}$ is the Alfven speed, and $\\theta$ is the angle between wave propagation direction and magnetic field \\cite{kos00}. The density $\\rho\\approx 0.2$ g\\,cm$^{-3}$ in the tachocline. Averaging $\\sin^2\\theta$ over all directions yields about 1/2. Thus the magnetic field strength $B \\sim [16\\pi \\rho c^2 (\\delta w/w)]^{1/2} \\sim 4-7\\times 10^5$ gauss if the filling factor of magnetic field is unity. A smaller filling factor would increase the estimated field strength. The field strength estimated here is greater than most theories predict \\cite{fis00}. Such a strong field needs a large degree of subadiabaticity in the tachocline to stabilize it \\cite{gil00}. The approximately constant $\\delta\\tau_N/N$ at $N\\ge 9$ suggests that there is no strong magnetic field in the middle of the CZ.  The problem of above interpretation is that no additional decrease in travel time is detected for the neighboring wave packets of $N=8$ in our measurements. The neighboring wave packets, $N=7$ and 9, may be also influenced by the magnetic fields at the base of the CZ. The influence depends on the width and location of the magnetic layer at the base of the CZ and the width of the wave packets. If we adopt the parameters used above and the ray approximation to estimate the influence on the neighboring wave packets, the additional decrease at $N=7$ is about $40\\%$ of that at $N=8$, while there is no effect on $N=9$. However, the finite width of the wave packets would increase the effect of magnetic fields on the neighboring wave packets. The previous studies have shown that the travel-time sensitivity kernel is wide \\cite{jen00,bir00}. To estimate the effect due to the finite width, we construct the 3-D wave packet by superposing the eigenfunctions of the modes consistent with our phase-velocity filter \\cite{bog97}. The FWHM of the energy distribution in the radial direction for the wave packets of $N=7$ and 9 at the lower turning point is about $0.1 R_{\\odot}$, which is greater than the separation between the lower turning points of two neighboring wave packets, about $0.02-0.03 R_{\\odot}$. Thus the influence of magnetic fields at the base of the CZ on the wave packets of $N=7$ and 9 is not negligibly small compared with that on $N=8$.  This contradicts to the result of our measurements. At this moment, we do not know how the combination of magnetic field perturbation and sound speed perturbation can help resolve this contradiction."
    },
    "0208/astro-ph0208247_arXiv.txt": {
        "abstract": "Application of deconvolution algorithms to astronomical images is often limited by variations in PSF structure over the domain of the images.  One major difficulty is that Fourier methods can no longer be used for fast convolutions over the entire images. However, if the PSF is modeled as a sum of orthogonal functions that are individually constant in form over the images, but whose relative amplitudes encode the PSF spatial variability, then separation of variables again allows global image operations to be used.  This approach is readily adapted to the Lucy-Richardson deconvolution algorithm. Use of the Karhunen-Lo\\`eve transform allows for a particularly compact orthogonal expansion of the PSF.  These techniques are demonstrated on the deconvolution of Gemini/Hokupa'a adaptive optics images of the galactic center. ",
        "introduction": "\\label{sect:intro}  %  The deconvolution of astronomical images becomes tricky when the structure of the point-spread function (PSF) varies significantly over the domain of interest.  A constant PSF is generally not a formal requirement for most deconvolution algorithms, but it is a profoundly useful simplification.  With a constant PSF, one can quickly perform convolutions over the entire image in the Fourier domain, and the difficult problem of trying to estimate the PSF structure as it varies from point to point is completely avoided.  Unfortunately, real astronomical images often do have spatially-variant PSFs.  This problem is especially important for adaptive-optics (AO) imagery.  Presently, all working AO systems can only correct for atmospheric blurring at the location of the guide star or laser beacon.  As one moves away from this location in angle, the correction will degrade rapidly.\\cite{stein}  Traditionally there are two approaches for the deconvolution or analysis of such images.  \\begin{enumerate} \\item The easiest approach is to sweep the problem under the rug!  While it may be facetious to suggest that this as a real solution, it is true that there may be some cases where the PSF structural variations are significant in general, but may be ignored for specific problems. One might imagine a diffraction-limited system in which the form of the sharp core varied over the image, but the broad wings more or less stayed the same; for some problems, the latter component of the PSF may be more important.  \\item Break the image up into many sub-domains, over which the PSF may be regarded as constant.  This procedure is tedious at best, may produce discontinuities when the sub-images are stitched back together, and still must find a way to represent the PSF satisfactorily at all the locations demanded.  \\end{enumerate}  In this paper, I will outline a different approach that incorporates the continuous variation of the PSF into the deconvolution, but that also uses efficient full-image Fourier convolution; in this regard, it should be superior to, but yet simpler than the sub-domain approach. As it happens, this method simply strings together ideas already in the literature for compactly modeling the PSF spatial variations, as well as encoding the variations in a form that allows for efficient convolution. Folding these methods into standard Lucy\\cite{lucy}-Richardson\\cite{rich} deconvolution allows that algorithm to work quickly with a variable PSF; this method is readily adopted, as it actually only incorporates forward convolutions of the PSF (and its transpose) to estimate the deconvolved image.  An important caveat is that while a spatially-variant PSF deconvolution algorithm attempts to use the most accurate PSF possible for any point in the image, this does not imply that the deconvolved image will have uniform resolution.  Although a forward convolution of a source model to match an image with large PSF variations may be readily done, with real image noise the resolution gains offered by deconvolution of such images will always be tied to the intrinsic resolution of the local PSF. In the single-point AO case, for example, one will not be able to deconvolve the entire image to the same fine resolution available at the correction-point.  Variable-PSF deconvolution cannot undo any real loss of structural information that occurs as the resolution degrades away from the reference point. ",
        "conclusions": ""
    },
    "0208/astro-ph0208071_arXiv.txt": {
        "abstract": "The enhancement factor of the resonant thermonuclear reaction rates is calculated for the extremely dense stellar plasmas in the liquid phase. In order to calculate the enhancement factor we use the screening potential which is deduced from the numerical experiment of the classical one-component plasma.  It is found that the enhancement is tremendous for white dwarf densities if the $^{12}$C + $^{12}$C fusion cross sections show resonant behavior in the astrophysical energy range.  We summarize our numerical results by accurate analytic fitting formulae. ",
        "introduction": "In a recent important paper Cussons, Langanke, \\& Liolios (2002) have pointed out the potential resonant screening effects on stellar $^{12}$C + $^{12}$C reaction rates.  The $^{12}$C + $^{12}$C fusion cross sections show noticeable resonant structures down to the lowest energies measured so far in the laboratory $E \\sim 2.4$ MeV (Kettner, Lorenz-Wirzba, \\& Rolfs 1980).  If the resonant structure continues to even lower energies and the astrophysical reaction rate is due to the contributions of narrow resonances, one then has to consider that the entrance channel width of these resonances will be modified in the plasma.  Cussons, Langanke, \\& Liolios (2002) have specifically pointed out the possible importance of the plasma effects on the resonant $^{12}$C + $^{12}$C reactions for a carbon white dwarf environment with $T = 5 \\times 10^{7}$ K and $\\rho = 2 \\times 10^{9}$ g cm$^{-3}$.  They have considered a resonance energy interval 0.4$-$2 MeV.  They have specifically discussed a rather extreme case of the low resonance energy $E_{r}$ = 400 keV and have estimated the overall enhancement of the resonant $^{12}$C + $^{12}$C reaction rates due to the plasma effects for this case.  Cussons, Langanke, \\& Liolios (2002) adopted the method of Salpeter \\& Van Horn (1969) which is based on the lattice model of the dense plasma to calculate the resonant screening effects.  One of the present authors (N. I.) and his collaborators have calculated the enhancement of non-resonant thermonuclear reaction rates in extremely dense stellar plasmas (Itoh, Kuwashima, \\& Munakata 1990).  This work is a natural extension of the works of Itoh, Totsuji, \\& Ichimaru (1977) and Itoh et al. (1979), and improves upon the accuracy of the results of Salpeter \\& Van Horn (1969).  Itoh, Kuwashima, \\& Munakata (1990) have summarized their numerical results by an accurate analytical fitting formula which will be readily implemented in the stellar evolution computations.  The aim of the present paper is to extend the work of Itoh, Kuwashima, \\& Munakata (1990) to the case of resonant reactions.  The present paper is organized as follows.  Physical conditions relevant to the present calculation are made explicit in \\S~2.  Calculation of the enhancement factor of the resonant thermonuclear reaction rates is summarized in \\S~3. The results are presented in \\S~4.  Extension to the case of ionic mixtures is made in \\S~5.  Concluding remarks are given in \\S~6. ",
        "conclusions": "We have presented a calculation of the enhancement of the resonant thermonuclear reaction rates for extremely dense stellar plasmas. The calculation has been carried out by adopting the screening potential derived from the Monte Carlo computations of the classical one-component plasma.  We have summarized our numerical results by an accurate analytic fitting formula to facilitate applications. The present results will be useful if the $^{12}$C + $^{12}$C fusion reaction contains narrow resonances in the astrophysical energy range."
    },
    "0208/astro-ph0208301_arXiv.txt": {
        "abstract": "We have measured the spectrum of UHE cosmic rays using the Flash ADC (FADC) detector (called HiRes-II) of the High Resolution Fly's Eye experiment running in monocular mode.  We describe in detail the data analysis, development of the Monte Carlo simulation program, and results.  We also describe the results of the HiRes-I detector. We present our measured spectra and compare them with a model incorporating galactic and extragalactic cosmic rays.  Our combined spectra provide strong evidence for the existence of the spectral feature known as the ``ankle.'' ",
        "introduction": "The aim of the High Resolution Fly's Eye (HiRes) experiment is to study the highest energy cosmic rays using the atmospheric fluorescence technique.  In this paper we describe the data collection, analysis, and Monte Carlo calculations used to measure the cosmic ray spectrum with the HiRes experiment's FADC detector, HiRes-II.  We also describe the analysis performed on the data collected by the HiRes-I detector and present the two monocular spectra, covering an energy range from $2 \\times 10^{17}$ eV to over $10^{20}$ eV.  We perform a statistical test of the combined spectra which gives strong evidence for the presence of the spectral feature known as the ``ankle.''  We conclude with a fit of our data to a toy model incorporating galactic and extragalactic cosmic ray sources.  The acceleration of cosmic rays to ultra high energies is thought to occur in large regions of high magnetic fields expanding at relativistic velocities\\cite{kn:acceleration}.  Such structures are rare in the neighborhood of the Milky Way galaxy and many of the cosmic rays that we observe may have traveled cosmological distances to reach us.  Hence they are probes of conditions in some of the most violent and interesting objects in the universe.  The highest energy particles from terrestrial particle accelerators have energy $1 \\times 10^{12}$ eV, so the cosmic rays we observe have energies at least five orders of magnitude higher.  Since we observe showers in the atmosphere initiated by the cosmic ray particles, we are sensitive to their composition and to the details of their interactions with matter.  Interactions of high energy protons, traveling large distances across the universe, with photons of the cosmic microwave background radiation can excite nucleon resonances which decay to a nucleon plus a $\\pi$ meson.  This is an important energy loss mechanism for the cosmic rays, and results in the Greisen-Zatsepin-Kuzmin (GZK) cutoff\\cite{kn:gzk}, which is often stated as: cosmic rays traveling more than 50 Mpc should have a maximum energy of $6 \\times 10^{19}$ eV, if sources are uniformly distributed.  Several events above this energy have been seen by previous experiments\\cite{kn:prevexp,kn:flyseye,kn:agasa}, but statistics are low and it is crucial to search for more events above the GZK cutoff.  The spectrum of cosmic rays has few distinguishing features.  It consists of regions of power law behavior with breaks in the power law index.  There is a steepening from E$^{-2.7}$ to E$^{-3.0}$ at about $3 \\times 10^{15}$ eV (called the knee)\\cite{kn:knee} and a hardening at higher energy (called the ankle).  The Fly's Eye experiment\\cite{kn:flyseye}, observing in stereo mode, saw a second knee (or steepening of the spectrum) at $4 \\times 10^{17}$ eV and the ankle at $3 \\times 10^{18}$ eV.  The second knee has also been observed by the Akeno experiment\\cite{kn:akeno}.  The Haverah Park experiment\\cite{kn:hpark} observed the ankle at about $4 \\times 10^{18}$ eV. The Yakutsk experiment\\cite{kn:yakutsk} has seen both the second knee and the ankle.  The AGASA experiment\\cite{kn:agasa}, which has a large enough aperture to collect events with energies of $10^{20}$ eV, observes a higher flux than Fly's Eye, and the ankle at $1 \\times 10^{19}$ eV.  They observe a dip at the GZK threshold, but their spectrum then recovers at higher energies.  The atmospheric fluorescence technique has its basis in the fact that, on average, approximately five UV fluorescence photons\\cite{kn:kakimoto} will be emitted when a minimum ionizing particle of charge $e$ passes through one meter of air.  In HiRes, we detect these photons and reconstruct the development of cosmic ray air showers.  We collect the fluorescence light with spherical mirrors of area 5.1 m$^2$, and focus it on a $16 \\times 16$ array of photomultiplier tubes, each of which looks at about one degree of the sky.  We record the integrated pulse height and trigger time information from each tube, and can reconstruct the geometry of the air shower and the energy of the primary cosmic ray that initiated it.  HiRes consists of two detector sites located on desert hilltops on the U. S. Army's Dugway Proving Ground in west central Utah.  The first site, called HiRes-I, consists of 22 detectors that look between 3 and 17 degrees in elevation and almost 360 degrees in azimuthal angle\\cite{kn:HiRes-INIM}.  This detector uses an integrating ADC readout system which records the photomultiplier tubes' pulse height and time information.  The second site, called HiRes-II and located 12.6 km away, consists of 42 detectors looking between 3 and 31 degrees in elevation, and has a Flash ADC (FADC) system to save pulse height and time information from its phototubes\\cite{kn:HiRes-IINIM}.  The sampling period of the FADC electronics is 100 ns.  Cosmic ray air showers with energies near $10^{20}$ eV and occurring within a radius of 35 km, can trigger the HiRes detectors and can be reliably reconstructed.  The two detector sites are designed to observe cosmic ray showers steroscopically.  This stereo mode observation gives us the best geometric resolution, about 0.6 degrees in pointing angle and 100 m in distance to the shower.  In this mode we make two measurements of the particle's energy and thus can make an empirical determination of our energy resolution.  The limitation of stereo mode is a geometrically imposed lower energy threshold of $10^{18}$ eV.  At this energy the events lie halfway between the two detectors, about 6 km from each.  In monocular mode, the HiRes-II detector can observe events much closer and dimmer than is possible in stereo mode; the energy threshold for this mode is about $2 \\times 10^{17}$ eV.  The geometric resolution is still good: about 5 degrees in pointing angle and 300 m in distance.  In this paper we describe the operation and data analysis for the HiRes-II detector, briefly describe the differences between HiRes-I and HiRes-II, and present the monocular spectra of the two detectors. ",
        "conclusions": "We have measured the flux of UHE cosmic rays with the FADC detector of the HiRes experiment.  Use of Flash ADC information allowed us to reduce systematic errors in reconstruction of events.  We developed our Monte Carlo simulation programs to very accurately model the experiment, and calculated the exposure of the experiment in a way that takes into account the experimental resolution.  The result reported here is in good agreement with the cosmic ray flux measurement made with the HiRes-I detector.  The latter measurement is based on a largely statistically independent data set, with only a limited number of stereo events in common to both analyses.  The result reported here is also consistent with the flux measured by the Fly's Eye experiment using the stereo reconstruction technique.  Above $10^{20}$eV our data is significantly different from that of the AGASA experiment.  The ankle is not seen in the HiRes-II monocular alone, but is apparent in the combined HiRes-I and HiRes-II data.  We have fit our data to a model incorporating both galactic and extragalactic sources of cosmic rays, which includes the GZK cutoff, and find good agreement."
    },
    "0208/astro-ph0208137_arXiv.txt": {
        "abstract": "We present \\ha\\ spectropolarimetry observations of a sample of 23 Herbig Ae/Be stars. A change in the linear polarisation across \\ha\\ is detected in a large fraction of the objects, which indicates that the regions around Herbig stars are flattened (disc-like) on small scales. A second outcome of our study is that the spectropolarimetric signatures for the Ae stars differ from those of the Herbig Be stars, with characteristics changing from depolarisation across \\ha\\ in the Herbig Be stars, to line polarisations in the Ae group. The frequency of depolarisations detected in the Herbig Be stars (7/12) is particularly interesting as, by analogy to classical Be stars, it may be the best evidence to date that the higher mass Herbig stars are surrounded by flattened structures. For the Herbig Ae stars, 9 out of 11 show a line polarisation effect that can be understood in terms of a compact \\ha\\ emission that is itself polarised by a rotating disc-like circumstellar medium. The spectropolarimetric difference between the Herbig Be and Ae stars may be the first indication that there is a transition in the Hertzsprung-Russell Diagram from magnetic accretion at spectral type A to disc accretion at spectral type B. Alternatively, the interior polarised line emission apparent in the Ae stars may be masked in the Herbig Be stars due to their higher levels of \\ha\\ emission. ",
        "introduction": "Although we seem to understand reasonably well how low mass stars form by cloud collapse and disc accretion, the question of whether high mass stars form in the same way remains completely open. On the one hand, theoretical efforts are being undertaken to realise the formation of high mass stars through disc accretion (e.g. Behrend \\& Maeder 2001). On the other hand, others perform calculations for the formation of high mass stars by stellar collisions and mergers in a dense cluster environment (e.g. Bonnell, Bate \\& Zinnecker 1998). These collision scenarios are invoked because high mass star formation through disc accretion encounters the problem that radiation pressure forces on gas and dust may be able to reverse the infall, preventing the formation of stars with masses larger than 10 \\msun (York \\& Kruegel 1977).  In addressing the question as to whether accretion discs around high mass stars are present, the intermediate mass (2 -- 15 \\msun) pre-main sequence Herbig Ae/Be stars (Herbig 1960) play a crucial role, as these objects are the only higher mass pre-main sequence stars that are visible at optical and infrared wavelengths. Yet, whether or not Herbig Ae/Be stars are embedded in accretion discs has not been determined. Although there is clear-cut evidence for the presence of circumstellar gas and dust (see e.g. Waters \\& Waelkens 1998), there is no consensus on the {\\it geometry} of this material (see e.g. Pezzuto, Strafella \\& Lorenzetti 1997 for a summary of conflicting views). Ideally one would like to solve this issue by directly imaging the environments around these objects. Although there are indications that some intermediate mass stars present geometries deviating from spherical symmetry (e.g. Mannings \\& Sargent 1997; Grady et al. 1999; Shepherd, Claussen \\& Kurtz 2001), the conclusions reached are sometimes contradictory. For instance in the case of AB~Aur, the brightest and best-studied Herbig Ae/Be star, Mannings \\& Sargent infer an almost edge-on geometry, i~=~76\\degree, while Grady et al. claim i~$\\leq$~45\\degree. Furthermore, most of these imaging studies have been performed in the millimetre and radio regimes, probing the geometry on a relatively large scale, between about 100 -- 1000 AU. To make progress in the area of star formation for the higher mass stars, observations exploring the innermost regions around these objects are badly needed. Until sub-milli-arcsecond imaging becomes a reality, linear spectropolarimetry across emission lines such as H$\\alpha$ is the most powerful technique for doing so, as the scattering of starlight on electrons in an ionised medium occurs within a few stellar radii (Rudy 1978; Cassinelli, Nordsieck, \\& Murison, 1987).  This application of spectropolarimetry was first established in studies of classical Be stars (Clarke \\& McLean 1974; Poeckert 1975).  In its simplest form, it is based on the expectation that H$\\alpha$ photons arise over a much larger volume than the stellar continuum photons. For this reason the line photons undergo fewer scatterings off e.g. a circumstellar disc than the continuum photons do. Consequently the emission line flux will be much less polarised than the continuum. In this situation, a change in polarisation across the line profile is expected -- we refer to this {\\it de}polarisation as the classical line effect (because of its first appearance in observations of classical Be stars). The high incidence of these line depolarisations among members of this object class (26 out of 44 in Poeckert \\& Marlborough 1976) indicated that the envelopes of classical Be stars are not spherically symmetric. These findings are now taken as compelling evidence that classical Be stars are embedded in circumstellar discs (see e.g. Waters \\& Marlborough 1992).  Since then, H$\\alpha$ spectropolarimetry has been performed on a variety of other strong line emitting objects, such as B[e] stars (Zickgraf \\& Schulte-Ladbeck 1989; Magalh$\\tilde{\\rm a}$es et al. 1992), Luminous Blue Variables (e.g. Schulte-Ladbeck et al. 1994), novae (Bjorkman et al. 1994) and supernovae (e.g. Cropper et al. 1988). In the case of pre-main sequence stars the technique has recently been applied to a sample of 9 Herbig Be stars by Oudmaijer \\& Drew (1999; hereafter OD99). The main outcome of this pilot study was that a line effect was detected in about half the Herbig Be stars. It is possible that the OD99 detection rate is consistent with discs around all these stars being viewed at angles ranging from face-on to edge-on.  But to be able to make a proper distinction between viewing angle and intrinsic geometry effects, one needs to obtain a larger sample. We pursue this goal by increasing the sample of spectropolarimetric data to about 25 Herbig stars in total.  In addition to increasing the sample, we extend our data set toward the later spectral type Herbig Ae (HAe) stars, where we may be able to detect spectropolarimetry characteristics differing from the earlier type Herbig Be (HBe) stars. A switch in phenomenology may be expected to occur at some point working down the stellar mass range, as different physical mechanisms are likely to play a role at different spectral types.  For instance, radiation pressure forces are likely to play a role for the higher luminosity stars at the early B types, whereas magnetic fields may become more dynamically prominent at the later A types.  The magnetically-channelled accretion model (Ghosh \\& Lamb 1979) that is commonly applied to the lowest mass pre-main-sequence T~Tauri stars (by e.g. Bouvier, Forestini, \\& Allain, 1997) may also be a suitable model as early as spectral type A (Pontefract et al. 2000).  If it does operate, the inner accretion disc around the star is truncated by the magnetic field, and the depolarisation effect may then be absent because the inner hole will necessarily lead to reduced intrinsic continuum polarisation.  Alternatively, the channelled accretion may produce a relatively bright and compact source of H$\\alpha$ emission that may be scattered within the accretion column itself or within the disrupted disc.  This in turn may yield a polarisation signature at H$\\alpha$ that is more complex than the simple depolarisation effect (see McLean 1979; Wood, Brown \\& Fox 1993).  The second goal of this paper is accordingly to explore how the H$\\alpha$ spectropolarimetric signature behaves as a function of Herbig star spectral type.  The paper is organised as follows. In Sect.~\\ref{s_obs} we discuss the way in which our sample was selected, how the observations were performed, and reduced. In Sect.~\\ref{s_results} we first present the continuum polarisation results for the whole sample and subsequently focus on the spectropolarimetry. We outline spectropolarimetric behaviour that can be expected, and develop tools to characterise the observations. We then discuss the individual targets, where we proceed by separating the discussion on the Herbig Be stars (Sect.~\\ref{s_hbe}) from the Herbig Ae stars (Sect.~\\ref{s_hae}). The two groups are subsequently compared in Sect.~\\ref{s_diff}. In Sect.~\\ref{s_disc} we summarise the main outcomes of our study and discuss the possible interpretations.  \\begin{table*} \\begin{minipage}{\\linewidth} \\renewcommand{\\thefootnote}{\\thempfootnote} \\caption{Herbig Ae/Be targets. {\\it V} magnitudes (column 3) and Spectral types (column 4) are taken from {\\sc simbad}. The integration times denote the total exposures (column 6). The errors in our polarisation data (column 7) are of the order of 0.01\\% based on photon-statistics only. Yet, systematic (external) errors in the polarisation are estimated to be 0.1\\%. The errors in the Position Angle $\\theta$ (column 8) are less than a degree. Column (9) gives estimates of the sky PA derived from line excursions; note that the presumed discs are supposed to lie perpendicular (=90\\degree) to this determined sky PA.} \\label{t_cont} \\begin{tabular}{llrlclcrc} \\hline Name          & HD Number  & {\\it V} &Spec. Tp  & Date & Exposure(s)  & $P_{\\rm cont}^{\\rm R}$ (\\%) & $\\Theta_{\\rm cont}^{\\rm R}$ (\\degree) & $\\Theta_{\\rm intr}^{\\rm R}$ (\\degree)\\\\ \\hline MWC 137       &              & 11.2     & Be     &20-12-99         & 24x100  & 6.07 & 160  & 115 $\\pm$ 10\\\\ MWC 1080      &              & 11.6     & B0     &18-12-99             & 8x500   & 1.73 & 77   & \\\\ MWC 166       & HD 53367     &  7.0     & B0     &18-12-99             & 16x150  & 0.20 & 34   &  \\\\ BD+40 4124    &              & 10.7     & B2     &19-12-99         & 24x100  & 1.21 &  8   & 173 $\\pm$ 8\\\\ MWC 361       & HD 200775    &  7.4     & B2Ve   &18-12-99             & 16x100  & 0.81 & 96   &  \\\\ IL Cep        & HD 216629    &  9.3     & B2IV-Vne&19-12-99            & 16x100  & 4.24 & 102  &   \\\\ MWC 147       & HD 259431    &  8.8     & B6pe   &18-12-99             & 24x50   & 1.05 & 100  &  \\\\ MWC 158$^1$   & HD 50138     &  6.6     & B9     &18-12-99         & 16x20   & 0.65 & 59   & 45 $\\pm$ 5\\\\ & HD 58647     &  6.8     & B9IV   &19-12-99             & 16x75   & 0.14 & 127  & \\\\ AS 477        &              & 10.2     & B9.5Ve &19-12-99             & 16x150  & 0.43 & 56   &  \\\\ MWC 120       & HD 37806     &  7.9     & A0     &19-12-99             & 16x50   & 0.35 & 76   & \\\\ KMS 27        & HD 37357     &  8.9     & A0     &19-12-99             & 16x200  & 0.13 & 52   & \\\\ MWC 789       & HD 250550    &  9.6     & A0     &20-12-99         & 32x100  & 0.92 & 174  & 178 $\\pm$ 10\\\\ AB Aur        & HD 31293     &  7.1     & A0pe   &18-12-99         & 16x150  & 0.11 & 54  & 160 $\\pm$ 5\\\\ SV Cep        &              & 10.1     & A0     &20-12-99             & 8x500   & 0.61 & 67  &  \\\\ XY Per        & HD 275877    &  9.4     & A2II   &20-12-99             & 8x300   & 1.60 & 132 &   \\\\ MWC 480       & HD 31648     &  7.7     & A2     &19-12-99             & 16x75   & 0.38 & 52  &  \\\\ & HD 244604    &  9.4     & A3     &19-12-99             & 16x200  & 0.44 & 119 &   \\\\ MWC 758       & HD 36112     &  8.3     & A3     &19-12-99             & 16x100  & 0.07 & 179 &   \\\\ T Ori         &              &  9.5     & A3V    &20-12-99             & 8x500   & 0.39 & 97  &  \\\\ & HD 245185    & 10.0     & A5     &20-12-99             & 8x500   & 0.21 & 168 &   \\\\ & HD 35929     &  8.1     & A5     &19-12-99             & 16x200  & 0.12 & 51  &  \\\\ CQ Tau$^2$        & HD 36910     & 10.7     & F2IVe  &20-12-99             & 8x500   & 0.27 & 83 &   \\\\ \\hline \\end{tabular} \\\\ \\noindent $^1$ MWC~158 does not appear in Table 1 of Th\\'e et al. (1994). The data have been included for completeness.\\\\ \\noindent $^2$ Note that the spectral type for CQ~Tau as given by Th\\'e et al. (1994) is between late A and early F (A8 -- F2), and\\\\ \\hspace*{0.2cm} is indeed considered to be a member of the Herbig Ae/Be group.\\\\ \\end{minipage} \\end{table*} ",
        "conclusions": "\\label{s_disc}  Altogether we have found that 16 out of 23 Herbig Ae/Be stars show a change in linear polarisation across \\ha. Data on 19 objects have been presented here, and are supplemented by 4 more from OD99. Summarised:  \\begin{itemize} \\item{} For the Herbig Be stars: 7 out of 12 reveal a line effect that is consistent with a depolarisation effect caused by a flattened structure in analogy to the presence of discs around classical Be stars.  \\item{} For the Herbig Ae stars: 9 out of 11 show a line polarisation effect characterised by a ``loop'' in $(Q,U)$ space, which suggests compact \\ha\\ emission that is itself polarised by a rotating disc-like medium.  \\end{itemize}  We have found that the changes detected in the Herbig Be stars are like those observed in classical Be stars. It is worth considering whether this 7/12 (or $\\sim$58 percent) detection rate is also as high as found among classical Be stars.  Before making such a comparison, it is necessary to establish the typical sensitivity limit of our spectropolarimetry. This is most easily derived by assessing the radii of the continuum knots in the $(Q,U)$ plane plots: typically these are large enough that excursions out of them that extend no further than 0.2\\% are too unclear to claim as definite detections.  Using this as the detection threshold for polarisation changes across \\ha, we can turn for a comparison to the large classical Be sample of Poeckert \\& Marlborough (1976).  They tabulated the change in line polarisation percentage across \\ha\\ for 48 classical Be stars (their Table 2). In 26 of these stars, the polarisation change was greater than our sensitivity limit of 0.2\\% . This amounts to a 54 percent detection rate to be compared with $\\sim$58 percent here.  These percentages are strikingly alike.  The similarity in both spectropolarimetric behaviour and detection rate is a strong hint that Herbig Be stars are embedded in electron-scattering circumstellar discs of similar characteristics to those associated with classical Be stars.  This allows the interpretation of the non-detections to be examples of Herbig stars with their discs too face-on to yield measurable line depolarisations.  The predominance of $(Q,U)$ loops in the Herbig Ae star spectropolarimetry suggests that compact \\ha\\ emission that is itself scattered and polarised by a rotating, non-spherical medium is the norm.  Prior examples of this spectropolarimetric behaviour and discussion of its origin may be found in studies of the peculiar Be star $\\gamma$~Cas (Poeckert \\& Marlborough 1977) and the Wolf-Rayet binary EZ~CMa (Schulte-Ladbeck et al. 1990).  That the frequency of effects among the Herbig Ae stars is possibly higher than among the Herbig Be stars might be expected as the PA rotation can remain visible over a wider range of viewing angles. In the case of AB~Aur, Pontefract et al (2000) have suggested that the spectropolarimetric evidence of a rotating, flattened structure surrounding a compact \\ha\\ source might be due to magnetically-channelled accretion.  Catala et al. (1999) also invoked this possibility in the face of a detection of time-variable redshifted absorption in He{\\sc i} lines in AB~Aur.  Our data include two instances of stars showing significant redshifted absorption at \\ha\\ (T~Ori and CQ~Tau), and many more instances of $(Q,U)$ loops. (Note that infall features at \\ha\\ were previously found by e.g. Finkenzeller \\& Mundt 1984; de Winter et al. 1999). Independently, the two lines of evidence would not be such a strong indication of magnetically-channelled accretion -- but together they are rather more compelling.  To pursue this issue further, an interesting next step would be to compare the linear spectropolarimetric characteristics for the Herbig stars with those in the lower mass T~Tauri stars, a class of young objects for which the magnetically-channelled accretion model is better established (Edwards et al. 1994; Hartmann 1999). So far only circular spectropolarimetry has been published (Johns-Krull, Valenti \\& Koresko 1999) for T~Tauri stars.  We now consider why there is the observed difference between the Herbig Be and HAe groups, and what it might mean.  One point of view could be that there is a physical transition region in the Hertzsprung-Russell Diagram where the geometry changes, as possibly the mode of presumed accretion changes. While the Herbig Ae stars might experience magnetically-channelled accretion resembling that associated with T~Tauri stars, the Herbig Be star accretion flow may not be disrupted at a magnetospheric radius. The possibility that this might be so inspired the exploration of a radiation-driven disc wind concept for Herbig Be stars and embedded Becklin-Neugebauer objects presented by Drew, Proga \\& Stone (1998). Before the option of a real transition can be accepted it is important to recognise that the difference may instead simply be one of observational contrast.  Crudely speaking, spectral type maps onto effective temperature and, indeed, luminosity -- if differences in evolutionary age are taken to be a secondary influence in our sample.  Assuming all environmental factors such as the geometry and quantity of circumstellar matter around all Herbig Ae/Be stars to be the same, the mere facts of the higher temperatures and luminosities of the HBe stars imply greater extension of the ionised regions around them, and of course larger dust-free cavities. In accord with this, \\ha\\ equivalent width is indeed greater among the HBe stars than among the HAe stars.  In the presence of reduced \\ha\\ line emission and opacity for the HAe stars it is easier to detect the interior component that we require should exist inside an axisymmetric rotating structure. The failure to detect this at \\ha\\ in the early Be stars could then be nothing more than a masking of it by the sheer size of the total line-emitting volume and its optical depth.  We have counted MWC~361 (see Fig.~\\ref{f_hbe}) as an instance of broad line depolarisation.  This may not be the entire story.  The data for this object show a relatively narrow PA rotation in the red wing of \\ha, on top of what otherwise is a well-defined wide depolarisation.  This may hint at some interior \\ha\\ polarisation reminiscent of that seen among the HAe stars. Significantly, perhaps, the \\ha\\ equivalent width measured for MWC~361 (-63\\AA) is lower than for most of the other early HBe stars showing simple depolarisations. A difficulty may however be presented by the case of MWC~166 (HD~53367). This early HBe star also has a relatively small \\ha\\ equivalent width ($-$2.8\\AA\\ rather and $-$14\\AA\\ in OD99), but no evidence of a compact \\ha\\ component. In other words, spectropolarimetrically it does not look like an HAe star. The counter to this is to argue that a reclassification of this object is appropriate. Evidence to support this can be derived from the relative weakness of MWC 166's infrared continuum excess: Hillenbrand et al. (1992) classify it as a Class III source because it is so weak. These Class III objects are hard to distinguish from more mature classical Be stars. Perhaps MWC~361 is the seemingly peculiar case that ultimately betrays the geometric parallels between the environments of HAe and HBe stars. However, caution must be exercised in view of the potential impact of the known binarity of MWC~361 (Millan-Gabet, Schloerb \\& Traub 2001).  There is a test that might distinguish between the two options discussed above. Spectropolarimetry could be performed across less opaque, higher excitation emission lines that can be expected to arise in the same location as the compact \\ha\\ component, if it exists in the HBes. If the extension and optical depth of \\ha\\ emission indeed masks the evidence of a compact source in the HBes, e.g. He~{\\sc i} or higher Balmer line observations could reveal angle rotations ($Q,U$ loops) in both the HBe and the HAe stars. If however, angle rotations were not observed in the HBe population, while appearing among the HAe stars, then the balance of the argument would shift in favour of HBe environments being physically different from those of HAe stars on the smallest scales.  As we have found that flattened structures around Herbig Ae/Be stars are common in our data, we wonder how this compares with the existing literature on this matter. We have therefore checked if the spectropolarimetric behaviour of our HAe sample correlates at all with the so-called UXOr phenomenon (named after UX~Ori). The UXOr phenomenon is believed, by some authors, to be associated with dusty clouds orbiting the young star in a close to edge-on disc-like structure (see e.g. Natta et al. 1997, but see Herbst \\& Shevchenko 1999 for a conflicting view). Six of the HAe stars from our sample appear in the HAe compilation by Natta et al., but only three of these undergo photometric changes larger than 1 mag in the V band characteristic for UXOrs\\footnote{This is a very loose description of the UXOr behaviour. Strictly speaking, the star should undergo a colour reversal (blueing effect) which is characteristic for UX~Ori itself}. As some of the stars that do not show UXOr behaviour do show loops in $(Q,U)$ space, this indicates that the UXOr and spectropolarimetric behaviour have different physical origins: the UXOr behaviour may well be linked with large scale dusty clouds, while the spectropolarimetry yields insights on the structure of the accretion flow much closer to the star.  Recently, Millan-Gabet et al. (2001) conducted an interferometric study at near-infrared wavelengths and they concluded that accretion disc models can be ruled out in most cases. Instead it was claimed that spherical models reproduce the visibility data much better. Note the seeming contradiction with the findings of the spectropolarimetry. But what are the spatial scales involved? The Herbig Ae stars in Sect.~\\ref{s_hae} of this study show loops in $(Q,U)$ space, associated with PA changes measured to be as narrow as $\\sim$ 10 \\ang. This wavelength width translates into a velocity in the rotating disc of about 225 \\kms. Assuming a stellar mass of $\\sim$ 2 \\msun, we find that this rotation speed corresponds to spatial scales of only $\\sim$ 7 stellar radii, or equivalently 0.07 AU. This is to be compared to the scales of 0.5 -- 5 AU that the interferometry of Millan-Gabet et al. can resolve.  We conclude that the spectropolarimetry indeed probes the smallest-scale structures of the circumstellar medium, which are found to be flattened.  \\paragraph*{\\it Acknowledgements}  We thank Matthew Pontefract for his participation in obtaining the observations, and the referee Rens Waters for constructive comments. The allocation of time on the William Herschel Telescope was awarded by PATT, the United Kingdom allocation panel.  JSV is funded by the Particle Physics and Astronomy Research Council of the United Kingdom. The data analysis facilities are provided by the Starlink Project, which is run by CCLRC on behalf of PPARC. This research has made use of the {\\sc simbad} database, operated at CDS, Strasbourg, France."
    },
    "0208/astro-ph0208123_arXiv.txt": {
        "abstract": "Recently, Maiolino et al.~(2001a, \\aap, 365, 28) constructed a sample of active galactic nuclei for which both the reddening $E(\\bv)$ and the column density $\\NH$ to the nucleus could be determined.  For most of the galaxies in their sample, they found that $E(\\bv)/\\NH$ is substantially smaller than for the diffuse interstellar medium of our Galaxy.  They asserted that either the dust-to-gas ratio is lower than in the Galaxy or that the grains are so large that they do not extinct or redden efficiently in the optical. We show that there is no systematic increase in $E(\\bv)$ with $\\NH$ for the Maiolino et al.~(2001a) galaxies, which suggests that the X-ray absorption and optical extinction occur in distinct media.  Maiolino et al.~(2001b, \\aap, 365, 37) suggested that the observed lines of sight for the Maiolino et al.~(2001a) galaxies pass through the ``torus'' that obscures the broad line region and nuclear continuum in Seyfert 2 galaxies and argued that the torus grains are larger than Galactic grains. There is no reason to believe that the lines of sight for these galaxies pass through the torus, since the observed column densities are lower than those typically observed in Seyfert 2 galaxies.  We suggest instead that the X-ray absorption occurs in material located off the torus and/or accretion disk while the optical extinction occurs in material located beyond the torus. The X-ray absorbing material could either be dust-free or could contain large grains that do not extinct efficently in the optical. There is no conclusive evidence that the grains in active galactic nuclei are systematically larger than those in the diffuse interstellar medium of our Galaxy.  We discuss an alternative way to probe the properties of dust in Seyfert tori, but find that observations of Seyfert 2 nuclei with higher resolution than currently available will be needed in order to place stringent limits on the dust. ",
        "introduction": "}  The unified model of active galactic nuclei (AGNs) posits that Seyfert 1 (Sy 1) and Seyfert 2 (Sy 2) galaxies are intrinsically the same, but that obscuring matter along our line of sight to the nucleus extincts the broad lines and nuclear continuum in Sy 2s (see Antonucci 1993 for a review).  The obscuring matter is usually referred to as the ``torus'', although its geometry has not yet been established.  The torus is located at a distance $r \\ltsim 1 \\pc$ from the center, on the basis of the following observational evidence.  First, the infrared spectral energy distribution of AGNs is dominated by thermal emission from grains with a range of temperatures extending up to the sublimation temperature, $\\approx 1500$--$2000 \\K$.  This temperature is reached at a distance $\\ltsim 1 \\pc$ from the center.  Variations in the IR continuum follow variations in the UV/optical continuum with a delay of $\\sim$ hundreds of days, confirming the presence of dust with $r \\ltsim 1 \\pc$.  See \\S 4.3 of Peterson (1997) for further details and references.  Second, VLBI observations of ``megamasers'' in AGNs demonstrate that molecules are located at $r \\ltsim 1 \\pc$ (see Maloney 2002 for a review).\\footnote{The megamaser observations also suggest a warped disk geometry for the torus; see Maloney (2002).} Third, Risaliti, Maiolino, \\& Salvati (1999) have shown that the column density $\\NH \\gtsim 10^{24} \\cm^{-2}$ for at least 50\\% of Sy 2 galaxies, and have pointed out that if the obscuring matter were located at $r \\gtsim 10$--$100 \\pc$, then its mass would exceed the dynamical mass for that region.  In addition to Sy 1s and Sy 2s, there are also intermediate-type galaxies, with weaker broad lines than observed in Sy 1s.  For example, Sy 1.8 galaxies have very weak broad H$\\, \\alpha$ and H$\\, \\beta$ lines and Sy 1.9s have no detectable broad H$\\, \\beta$.  Risaliti et al.~(1999) found that the column density (determined from the photoelectric cutoff in the X-ray spectrum) along the line of sight to the nucleus is typically much higher in Sy 2s than in 1.8s and 1.9s (see Figure \\ref{fig:Sy_NH_dist}).  The lower column densities in intermediate-type Seyferts could possibly result from our line of sight traversing relatively low-density outer regions of the torus.  Alternatively, the X-ray-absorbing material in Sy 1.8s and 1.9s might be unrelated to the torus. Maiolino \\& Rieke (1995) found that Sy 1.8s and 1.9s are mostly found in edge-on host galaxies, implying that the optical extinction is due to dust located beyond the torus.  If the optical extinction is unrelated to the torus, then it seems plausible that the same could be true for the X-ray absorption as well.  Maiolino et al.~(2001a; hereafter M01a) presented a sample of AGNs for which at least some broad lines are observed, but with significant reddening.  For each of the objects in the M01a sample, both the reddening $E(\\bv)$ and the total H column density $\\NH$ along the line of sight to the nucleus have been estimated.\\footnote{The reddening is the extinction at $B$ minus the extinction at $V$; i.e., $E(\\bv) = A_B - A_V$.} The column density is inferred from observations of the absorption of X-rays from the central source. M01a found that, with the exception of a few low-luminosity AGNs, $E(\\bv)/\\NH$ is generally lower than it is in the diffuse interstellar medium (ISM) of our Galaxy, by factors of a few up to $\\sim 100$. M01a assumed that the X-ray-absorbing material and optical-absorbing material are identical, which would imply that either (1) the dust-to-gas ratio in this material is lower than in the Galaxy or (2) the grains in this material are so large that, unlike Galactic grains, they do not extinct and redden efficiently in the optical. In a companion paper, Maiolino, Marconi, \\& Oliva (2001b; hereafter M01b) argued in favor of the latter interpretation. Their argument assumes that the medium probed by M01a is in fact the torus that is responsible for the optical obscuration of the broad line region (BLR) and nuclear continuum in Sy 2s.  Here, we propose an alternative interpretation, namely, that the lines of sight to the nuclei in the M01a sample pass not through the torus, but through lightly or moderately ionized material located off the torus and/or accretion disk, perhaps in a wind.  This material, which is responsible for the X-ray absorption, could be either dust-free or dominated by large enough grains that very little extinction and reddening result.  The observed reddening of the BLR occurs in a dusty medium that is physically distinct from the X-ray-absorbing material.  In \\S \\ref{sec:sample} we describe the M01a sample and present evidence that (1) the torus is not being probed and (2) the X-ray- and optical-absorbing gas are distinct. In \\S \\ref{sec:big_grains} we review the evidence that led M01b to conclude that the grain size distribution of the obscuring material must be dominated by very large grains, and show that this conclusion does not hold in our alternative scenario.  In \\S \\ref{sec:extinction_Sy2}, we discuss a method for determining the extinction per column in Seyfert tori. Currently available observations yield limits that are consistent with the torus dust being similar to Galactic dust, but observations at higher resolution are needed in order to establish more stringent limits. We briefly summarize in \\S \\ref{sec:conclusions}. ",
        "conclusions": "}  M01a constructed a sample of intermediate-type Seyfert galaxies that have been observed in the X-ray and in the optical or infrared and found that $E(\\bv)/\\NH$ is substantially lower than its value in the Galaxy for most of the objects in the sample.  M01b concluded that the only viable explanation for this result is that the grains in Seyfert tori are typically substantially larger than the grains in the ISM of our Galaxy.  Here, we have shown that the material that absorbs the X-rays is probably unrelated to the material that absorbs the optical/infrared radiation (\\S \\ref{sec:sample}, Figure \\ref{fig:EvsNH}) and that the torus probably is not probed by the observations of the M01a sample (\\S \\ref{sec:sample}, Figure \\ref{fig:Sy_NH_dist}). We suggest that, alternatively, the line of sight towards an M01a galaxy passes through ionized material located just off the torus and/or the accretion disk.  This material is responsible for the X-ray absorption, while the optical/infrared extinction occurs in material farther from the nucleus, where the dust may be quite similar to Galactic dust.  The X-ray-absorbing material may be dust-free or may contain large grains that have very small extinction efficiencies in the optical/infrared (\\S \\ref{sec:big_grains}). This material may be associated with a disk wind, which would originate within the dust sublimation radius.  In this case, the material would naturally be dust-free.  The disk wind scenario can be tested by searching for blue-shifted absorption by ions characteristic of lightly or moderately ionized gas, e.g., \\ion{C}{4}.  However, such ultraviolet absorption lines could be difficult to detect due to the moderate levels of extinction.  Although the M01a sample does not present evidence in favor of large grains in AGN tori, nor does it present evidence against large grains. The apparent lack of strong $9.7 \\micron$ silicate absorption in Sy 2s could indicate large grains in Seyfert tori, as pointed out by M01a.  More detailed studies of radiative transfer are needed in order to clarify this point.  We have attempted to obtain limits on the $1.6 \\micron$ extinction per column in Sy 2s (\\S \\ref{sec:extinction_Sy2}), but the HST does not have adequate resolution to exclude nuclear light that is scattered into our line of sight from locations beyond the torus.  Current observations are consistent with the torus dust being similar to Galactic dust; observations at higher resolutions will yield more stringent limits."
    },
    "0208/astro-ph0208409_arXiv.txt": {
        "abstract": "We describe an extensive FUSE survey of highly ionized oxygen in the vicinity of the Milky Way that serves as an example of the type of study that would be desirable for other galactic systems. Understanding the origin of hot gas in the vicinity of galaxies and its relationship to the intergalactic medium presents a major observational challenge.  \\hbox{Ultraviolet} absorption-line spectroscopy is currently the most direct means for comprehensive investigations of the gas in galactic environments, but even with present (and near-term) facilities the number of background objects available to probe nearby galaxy halos and low-redshift cosmological structures is limited. Studying these structures over a range of impact parameters and angular separations would provide fundamental information about the baryonic content of the hot gas, its physical conditions, and its origins.  A large space telescope optimized for high resolution spectroscopy in the 900--3200\\,\\AA\\ wavelength region at a sensitivity sufficient to observe faint AGNs/QSOs at angular separations of $< 1\\deg$  would be ideal for such studies. ",
        "introduction": "Understanding galaxy formation and evolution requires observational information about the hot, highly-ionized gas found in and near galaxies. A complete picture of the relationship between hot gas in galaxies and the intergalactic medium (IGM) does not yet exist because a variety of processes affect the heating and distribution of the interstellar gas in and around galaxies.  Galaxy formation, accretion of satellite galaxies, tidal interactions, star-formation, galactic winds, and galaxy-IGM interactions may all contribute to the production of the hot gas observed in the IGM and extended galactic halos.  In this article, we outline a program we have begun with the Far Ultraviolet Spectroscopic Explorer (FUSE) to study the hot gas in the vicinity of the Milky Way.  The study is described in detail in a series of three articles devoted to probing the highly ionized oxygen (O\\,VI) absorption along complete paths through the Galactic halo and Local Group.  The articles include a catalog of the spectra and basic observational information (Wakker et al. 2002), a study of the hot gas in the Milky Way halo (Savage et al. 2002b), and an investigation of the highly ionized high velocity gas in the vicinity of the Galaxy (Sembach et al. 2002).  Here, we summarize the high velocity gas results and comment on their implications for understanding other types of highly ionized gas. This comprehensive study serves as a prime example of what can be learned if serious efforts are made to conduct large spectroscopic observing programs with existing space-based instrumentation. Similar types of studies conducted with future generations of instruments on the Hubble Space Telescope (HST) and other large space telescopes will enable us to explore the properties of gas in other groups of galaxies in much greater detail than is presently possible.  In Table~1 we list basic information for some of the best ultraviolet (UV) absorption-line diagnostics of hot gas.  The table entries include the atom or ion, rest wavelength of the transition, cosmological redshift required to shift the line(s) to an observed wavelength of 1150\\,\\AA, the temperature at which the atom or ion peaks in abundance under conditions of collisional ionization equilibrium, the thermal width of the line at that temperature, and a logarithmic relative line strength that depends on the cosmic abundance of the element ($A$), the line strength ($f\\lambda$), and the peak ionization fraction ($\\phi$) of the ion in collisionally ionized gas. Larger values of $Af\\lambda\\phi$ indicate easier detectability.  \\begin{table} \\caption{Ultraviolet Diagnostics of Hot Gas at Low Redshift} \\begin{tabular}{lccccc} \\tableline\\tableline Species & $\\lambda$(\\AA) & $z$(1150\\,\\AA) & $T_{CIE}$ (K)\\tablenotemark{a} & b$_{th}$(km~s$^{-1}$)\\tablenotemark{b} & log [$Af\\lambda\\phi$]\\tablenotemark{c} \\\\ \\tableline H\\,I Ly$\\alpha$ & 1215.7 & --- & $<10^5$ & 40.8\\tablenotemark{d} & $-2.06$\\tablenotemark{d}\\\\ H\\,I Ly$\\beta$  & 1025.7 & 0.12 & $<10^5$ & 40.8\\tablenotemark{d} & $-2.86$\\tablenotemark{d} \\\\ H\\,I Ly$\\gamma$ & 972.5 & 0.18 & $<10^5$ & 40.8\\tablenotemark{d} & $-3.32$\\tablenotemark{d} \\\\ C\\,IV & 1548.2, 1550.8 & --- & $1.0\\times10^5$ & 11.8 & $-1.51, -1.81$\\\\ N\\,V  & 1238.8, 1242.8 & --- & $1.8\\times10^5$ & 14.6 & $-2.35,-2.65$\\\\ O\\,IV & 787.7 & 0.46 & $1.6\\times10^5$ & 12.9 & $-1.36$\\ \\\\ O\\,V  & 629.7 & 0.83 & $2.5\\times10^5$ & 16.1 & $-0.97$\\ \\\\ O\\,VI & 1031.9, 1037.6 & 0.11 & $2.8\\times10^5$ & 17.1 & $-1.65, -1.95$\\ \\\\ Ne\\,VIII & 770.4, 780.3 & 0.48 & $5.6\\times10^5$ & 21.5 & $-2.36,-2.66$\\\\ Mg\\,X & 609.8, 624.9 & 0.87 & $1.1\\times10^6$ & 27.4 & $-3.31,-3.61$ \\\\ \\tableline \\end{tabular} \\tablenotetext{a}{Temperature of maximum ionization fraction in collisional ionization equilibrium (Sutherland \\& Dopita 1993).} \\tablenotetext{b}{Thermal line width, $b = (2kT/m)^{1/2}$, at $T=T_{CIE}$ unless indicated otherwise.} \\tablenotetext{c}{Values of $f\\lambda$ are from Morton (1991) and Verner et al. (1994).  Values of $A$ (abundance relative to hydrogen on a logarithmic scale where H\\,=\\,12.00, C\\,=\\,8.55, N\\,=7.97, O\\,=\\,8.87, Ne\\,=\\,8.07, and Mg\\,=\\,7.59) are from Grevesse \\& Noels (1993) and Anders \\& Grevesse (1989). Values of $\\phi$ are from Sutherland \\& Dopita (1993).} \\tablenotetext{d}{Value at $T = 10^5$ K.} \\end{table}   The O\\,VI $\\lambda\\lambda1031.926, 1037.617$ doublet lines are the best UV resonance lines to use for kinematical investigations of hot ($T \\sim 10^5-10^6$\\,K) gas in the low-redshift universe.  Oxygen is the most abundant element heavier than helium, and the O\\,VI lines have large oscillator strengths. Lower ionization UV lines observable at high spectral resolution are generally either much weaker than the O\\,VI lines (e.g., N\\,V $\\lambda1238.821, 1242.804$) or are better tracers of collisionally ionized gas at temperatures $T < 10^5$\\,K (e.g., C\\,IV $\\lambda1548.195, 1550.770$, C\\,III $\\lambda977.020$, Si\\,IV $\\lambda1393.755, 1402.770$, Si\\,III $\\lambda1206.500$). This latter set of ions is also considerably more susceptible to photoionization than O\\,VI.  X-ray spectroscopy of the interstellar or intergalactic gas in higher ionization lines (e.g., O\\,VII, O\\,VIII) is possible with XMM-Newton and the Chandra X-ray Observatory for a small number of sight lines toward AGNs and QSOs, but the spectral resolution (R~$\\equiv \\lambda/\\Delta\\lambda < 1000$) is modest compared to that afforded by FUSE (R~$\\sim 15,000$).  While the X-ray lines provide extremely useful information about the amount of gas at temperatures greater than $10^6$\\,K, the interpretation of where that gas is located, or how it is related to the $10^5-10^6$\\,K gas traced by O\\,VI, is hampered at low redshift by the complexity of the hot ISM and IGM along the sight lines observed. ",
        "conclusions": ""
    },
    "0208/astro-ph0208315_arXiv.txt": {
        "abstract": "Genetic algorithms are a class of heuristic search techniques that apply basic evolutionary operators in a computational setting. We have designed a fully parallel and distributed hardware/software implementation of the generalized optimization subroutine {\\tt PIKAIA}, which utilizes a genetic algorithm to provide an objective determination of the globally optimal parameters for a given model against an observational data set. We have used this modeling tool in the context of white dwarf asteroseismology, i.e., the art and science of extracting physical and structural information about these stars from observations of their oscillation frequencies. The efficient, parallel exploration of parameter-space made possible by genetic-algorithm-based numerical optimization led us to a number of interesting physical results: (1) resolution of a hitherto puzzling discrepancy between stellar evolution models and prior asteroseismic inferences of the surface helium layer mass for a DBV white dwarf; (2) precise determination of the central oxygen mass fraction in a white dwarf star; and (3) a preliminary estimate of the astrophysically important but experimentally uncertain rate for the $^{12}{\\rm C} (\\alpha,\\gamma)^{16}{\\rm O}$ nuclear reaction. These successes suggest that a broad class of computationally-intensive modeling applications could also benefit from this approach. ",
        "introduction": "}  About 5 billion years from now, the hydrogen fuel in the center of the Sun will begin to run out and the helium that has collected there will begin to gravitationally contract, increasing the rate of hydrogen burning in a shell surrounding the core. Our star will slowly bloat into a red giant---eventually engulfing the inner planets, perhaps even the Earth. As the helium core continues to contract under the influence of gravity, it will eventually reach the temperatures and densities needed to fuse three helium nuclei into a carbon nucleus (the $3\\alpha$ reaction). Another nuclear reaction will compete for the available helium nuclei at the same temperature: the carbon can fuse with an additional helium nucleus to form oxygen. The amount of oxygen produced during this process is largely determined by the relative rates of these two competing reactions \\cite{ww93}. Since the Sun is not very massive by stellar standards, it will never get hot enough in the center to produce nuclei much heavier than carbon and oxygen. These elements will collect in the center of the star, which will then shed most of its red giant envelope---creating a planetary nebula---and emerge as a hot white dwarf star \\cite{kd00}.  Once a white dwarf star forms and the nuclear reactions have ceased, its structural and thermal evolution is dominated by cooling, and regulated by the opacity of its thin atmospheric outer layers. It will slowly fade as it radiates its residual thermal energy into space---eventually cooling through a narrow range of temperatures that will cause it to vibrate in a periodic manner, sending gravity-driven seismic waves deep through the interior and bringing information to the surface in the form of brightness variations. This is fortunate, because a detailed record of the nuclear history of the star is locked inside, and pulsations provide the only known key to revealing it.  We can determine the internal composition and structure of pulsating white dwarfs using the techniques of high speed photometry to observe their variations in brightness over time, and then matching these observations with a computer model which behaves the same way. The observational aspects of this procedure have been addressed by the development of the Whole Earth Telescope (WET) network \\cite{nat90}, a group of astronomers at telescopes around the globe who cooperate to produce nearly continuous time-series photometry of a single target for 1-2 weeks at a time. The Fourier spectra of such observations reveal dozens of excited modes with periods in the range 100--1000 seconds, supporting our interpretation of them as non-radial oscillations with gravity as the restoring force ($g$-modes). The WET has now provided a wealth of seismological data on the different varieties of pulsating white dwarf stars.  The physical property of white dwarf models that most directly determines the pulsation frequencies is the radial profile of the Brunt-V\\\"ais\\\"al\\\"a (buoyancy) frequency, which is given by \\begin{equation} N^2 = -g \\left( {d \\ln \\rho \\over dr} - {1 \\over \\Gamma_1} {d \\ln P \\over dr} \\right) , \\end{equation} where $g$ is the local gravity, $\\rho(r)$ the density, $P(r)$ the pressure, and $\\Gamma_1$ is $(\\partial\\ln P / \\partial\\ln\\rho)$ at constant entropy. The magnitude of $N^2$ reflects the difference between the actual and the adiabatic density gradients, and sets the local propagation speed of internal gravity waves. The observed frequencies, in turn, are a measure of the average (inverse) wave speed in the portion of the interior where the waves propagate. Inferring the $N^2$ internal profile from the observed pulsation frequencies is thus a classical {\\it inverse problem}, on par in scope and complexity with similar problems encountered in helio- and geo-seismology.  Consider first the complementary {\\it forward problem}, which consists in computing the oscillation frequencies of a {\\it given} white dwarf structural model. The forward modeling procedure begins with a static, non-rotating, unmagnetized, spherically symmetric model of a pre-white dwarf, which we allow to evolve quasi-statically until it reaches the desired surface temperature. The models must initially satisfy two of the basic equations of stellar structure: the condition of hydrostatic equilibrium, which balances the outward pressure gradient against the inward pull of gravity \\begin{equation} {dP \\over dr} = {G M_r \\over r^2}\\rho\\ , \\end{equation} and the continuity equation ensuring mass conservation \\begin{equation} {dM_r \\over dr} = 4 \\pi r^2 \\rho\\ , \\end{equation} where $M_r$ is the mass contained within a spherical shell of radius $r$. White dwarf stars are compact objects supported mainly by electron degeneracy pressure ($P_e$), and we can describe the core with a simple polytropic equation of state of the form \\begin{eqnarray} P_e \\propto \\left( {\\rho \\over \\mu_e} \\right)^{5/3} , \\end{eqnarray} where $\\mu_e$ is the mean molecular weight per free electron. Cooling is achieved by leaking the internal thermal energy through the opacity of the thin atmospheric layers at a rate consistent with the star's luminosity, and adjusting the interior structure accordingly. Although we initially ignore a third equation of stellar structure (which ensures thermal balance), we do use it to evolve the models in a self-consistent manner. The cooling tracks of our polytropic models quickly forget the unphysical initial conditions and converge with the evolutionary tracks of self-consistent pre-white dwarf models well above the temperatures at which the hydrogen- and helium-atmosphere white dwarfs are observed to be pulsationally unstable \\cite{woo90}.  Next, the $g$-mode pulsation frequencies ($\\sigma_g$) of the evolved models must be calculated for comparison with the observations. Working in the usual spherical polar coordinates $(r,\\theta,\\phi)$, the first step is to express the radial displacement ($\\Xi_r$) experienced by an oscillating fluid element as \\begin{equation} \\Xi_r(r,\\theta,\\phi,t) =\\xi_r(r) Y_\\ell^m(\\theta,\\phi) \\exp(i\\sigma_g t)~, \\end{equation} where the $Y_\\ell^m$ are the usual spherical harmonic functions \\cite{as72}. For a given set of angular and azimuthal quantum numbers $(\\ell,m)$, the linearized adiabatic non-radial oscillation equations reduce to a one-dimensional linear eigenvalue problem for $\\sigma_g$ and $\\xi_r$, described by the following set of equations: \\begin{equation} \\frac{1}{r^2} \\frac{d}{dr}(r^2 \\xi_r)-\\frac{g}{c_s^2} \\xi_r + \\left(1-\\frac{L_{\\ell}^2}{\\sigma_g^2}\\right) \\frac{P'}{\\rho c_s^2}= \\frac{\\ell (\\ell+1)}{\\sigma_g^2 r^2} \\Phi' , \\label{eq-osc1} \\end{equation} \\begin{equation} \\frac{1}{\\rho} \\frac{d P'}{dr}+\\frac{g}{\\rho c_s^2} P' + (N^2-\\sigma_g^2)\\xi_r = -\\frac{d\\Phi'}{dr} , \\label{eq-osc2} \\end{equation} \\begin{equation} \\frac{1}{r^2}\\frac{d}{dr}\\left(r^2\\frac{d\\Phi'}{dr}\\right)- \\frac{\\ell(\\ell+1)}{r^2}\\Phi' = 4 \\pi G \\rho \\left(\\frac{P'}{\\rho c_s^2}+\\frac{N^2}{g}\\xi_r \\right) , \\label{eq-osc3} \\end{equation} where $\\Phi'$ is the perturbation of the gravitational potential, $c_s$ is the sound speed, $L_{\\ell}^2 \\equiv \\ell(\\ell+1) c_s^2/r^2$ is the Lamb or acoustic frequency, and $\\xi_r$ is the (small) radial displacement associated with a given mode of frequency $\\sigma_g$ (see \\cite{unn89} for a detailed derivation). The eigenmodes associated with a given set of $(\\ell,m)$ values possess radial harmonics which can be labeled with a third quantum number ($k$) related to the number of nodes in the corresponding radial eigenfunction, so that the frequencies of individual eigenmodes are best labeled as $\\sigma_{k\\ell m}$.  Inverting a continuous function, in our case $N^2(r)$, from a discrete set of data (the pulsation periods) is well known to be a mathematically ill-posed problem \\cite{cb86,par94}. However, the situation is not as critical as one might imagine because strong physical constraints can be placed on the variations with depth of the Brunt-V\\\"ais\\\"al\\\"a frequency. In white dwarf interiors, the $N^2(r)$ profile is determined by the structural stratification (e.g., variations of density and pressure with depth), which in turn depends on the star's evolutionary history as well as a number of physical parameters such as stellar mass, core chemical composition, surface temperature, and the mass of its surface helium layer, to name but a few. The ill-posed inverse problem for $N^2$ can be then recast in the form of an optimization problem that consists in finding the numerical values for the set of these parameters that yields the optimal fit between the oscillation periods of the corresponding white dwarf structural model, as computed via the forward procedure outlined above, and the observed periods. From the point of view of numerical optimization, this is now a well-posed problem.  With detailed observations and a theoretical model in hand, the next step is to select a suitable numerical optimization method. Models of all but the simplest physical systems are typically non-linear, so finding the optimal match to the observations requires an initial guess for each parameter. Some iterative method is generally used to improve on this first guess until successive iterations do not produce significantly different answers. There are at least two potential problems with this standard approach to model-fitting. The first guess is often derived from the past experience of the person who is fitting the model. This {\\it subjective} method is even worse when combined with a {\\it local} approach to iterative improvement.  Many optimization schemes, such as differential corrections \\cite{pl72} or the simplex method \\cite{kl87}, yield final results that depend to some extent on the choice of initial model parameters. This does not have serious consequences if the parameter-space contains a single, well-defined minimum. But if there are many local minima, then it can be more difficult for a traditional approach to find the globally optimal solution (e.g., see Fig.~1 of \\cite{cha95}).  The multi-modal nature of the optimization problem is not the only modeling pitfall to be reckoned with. A good fit between model periods and data certainly suggests that the model adequately reflects the actual physical structure of the stars themselves. However, the possibility can never be ruled out that other physical characteristics of the white dwarf models, considered known and held fixed in the present modeling work, could also be varied to yield comparably good fits to the observed frequencies. As with any inverse problem, asteroseismic inferences are plagued by the potential for non-uniqueness of the solutions. With this caveat firmly in mind, we proceed. ",
        "conclusions": "The application of genetic-algorithm-based optimization to white dwarf pulsation models turned out to be very fruitful. We are now confident that we can rely on this approach to perform global searches and to provide objectively determined optimal models for the observed pulsation frequencies of white dwarfs, along with fairly detailed maps of the parameter-space as a natural byproduct. The method finally allowed us to measure the central oxygen mass fraction in a pulsating white dwarf star, with an internal precision of a few percent \\cite{mnw00}. We used this value to derive a preliminary measurement of the $^{12}{\\rm C}(\\alpha,\\gamma)^{16}{\\rm O}$ reaction rate \\cite{mwc01,msw02}, which turned out higher than most published values \\cite{kun01}. More work on additional white dwarf stars and possible sources of systematic uncertainty should help to resolve the discrepancy.  Our success with the parallel genetic algorithm leads us to believe that many other problems of interest in astronomy and physics could benefit from this approach. For models that can run in less than a few minutes on currently available processors, and where automated execution is possible, the parallel version of {\\tt PIKAIA} can provide an objective and efficient alternative to large grid searches without sacrificing the global nature of the solution.  Although the number of model evaluations required is still large compared to what can be accomplished in reasonable wallclock time on a single desktop computer, Linux clusters are fast, inexpensive, and are quickly becoming ubiquitous. When combined with software like {\\tt PIKAIA} that can exploit the full potential of such distributed architectures, a new realm of modeling possibilities opens up.   \\begin{acknowledge} We are grateful to the High Altitude Observatory Visiting Scientist Program for fostering this collaboration in a very productive environment for two months during the summer of 2000. This work was supported by grant NAG5-9321 from the Applied Information Systems Research Program of the National Aeronautics \\& Space Administration, and in part by the Danish National Research Foundation through its establishment of the Theoretical Astrophysics Center. The National Center for Atmospheric Research is sponsored by the National Science Foundation. \\end{acknowledge}"
    },
    "0208/astro-ph0208586_arXiv.txt": {
        "abstract": "In this paper, we present preliminary results on the stability of massless particles in two and three-planet systems.  The results of our study may be used to address questions concerning the stability of terrestrial planets in these systems and also the trapping of particles in resonances with the planets.  The possibility of the existence of islands of stability and/or instability at different regions in multi-body systems and their probable correspondence to certain resonances are also discussed. ",
        "introduction": "The discovery of multi-body extrasolar planetary systems during the past few years has once again confronted astrodynamicists with the old question of the stability of such systems.  The discovery of GJ 876 where two planets are locked in a 2:1 mean-motion resonance (Marcy et al.\\ 2001), the confirmation of three planets in orbit about Upsilon Andromedae (Butler et al.\\ 1999), and the discovery of planetary systems around 47 Uma and 55 Cancri with planets in orbits more closely resembling those in the Solar System (Fischer et al.\\ 2002; Marcy et al.\\ 2002), have set the grounds for a deeper look at the problem of the stability of multi-body systems.  Here we present a preliminary investigation of the dynamical stability of the 47 Uma and 55 Cancri systems.  A more detailed analysis, including similar work on GJ 876 and $\\upsilon$ And, will be addressed in future work. ",
        "conclusions": "We have presented results on the stability of the 47 Uma and 55 Cancri systems with and without massless particles.  Dynamical fits were used to determine the initial conditions for the simulations.  A large area of parameter space is consistent with the observations.  The area around 1 AU in the 47 Uma system may not harbor a terrestrial planet, while the opposite is true in the 55 Cancri system.  These results are in rough agreement with previous studies (Fischer et al.\\ 2002; Marcy et al.\\ 2002)."
    },
    "0208/astro-ph0208065_arXiv.txt": {
        "abstract": "The gamma-ray blazar PKS 1510$-$089 has a highly superluminal milli-arcsecond jet at a position angle (PA) of $-28^\\circ$ and an arcsecond jet with an initial PA of $155^\\circ$.  With a $\\Delta$PA of $177^\\circ$ between the arcsecond and milli-arcsecond jets, PKS 1510$-$089 is perhaps the most highly misaligned radio jet ever observed and serves as a graphic example of projection effects in a highly beamed relativistic jet.  Here we present the results of observations designed to bridge the gap between the milli-arcsecond and arcsecond scales.  We find that a previously detected ``counter-feature'' to the arcsecond jet is directly fed by the milli-arcsecond jet. This feature is located $0.3\\arcsec$ from the core, corresponding to a de-projected distance of 30 kiloparsecs.  The feature appears to be dominated by shocked emission and has an almost perfectly ordered magnetic field along its outside edge.  We conclude that it is most likely a shocked bend, viewed end-on, where the jet crosses our line of sight to form the southern arcsecond jet.  While the bend appears to be nearly $180^\\circ$ when viewed in projection, we estimate the intrinsic bending angle to be between $12^\\circ$ and $24^\\circ$.  The cause of the bend is uncertain; however, we favor a scenario where the jet is bent after it departs the galaxy, either by ram pressure due to winds in the intracluster medium or simply by the density gradient in the transition to the intergalactic medium. ",
        "introduction": "\\label{s:intro} Extragalactic radio jets can change direction for a variety of reasons, such as deflections by massive clouds in the interstellar or intracluster medium, growth of hydrodynamical instabilities, or precession at the base of the jet.  Core dominated radio sources are oriented with their jets pointed nearly along our line of sight, greatly exaggerating in projection any intrinsic trajectory changes.  Small intrinsic changes can therefore appear as large angle misalignments between the jet axes observed on parsec and kiloparsec scales.  Indeed, large angle misalignments have been observed in some core dominated radio sources \\citep[e.g. 3C\\,309.1,][]{WKP86}, and the distribution of jet misalignment angles has been extensively studied \\citep[e.g.][]{PR88,WCU92,CM93,ASV96}.  While the distribution is known to have a bimodal shape, with a main peak of misalignment angles near $0^\\circ$ and a secondary peak near $90^\\circ$ \\citep{PR88,CM93}, misalignment angles larger than $120^\\circ$ are quite rare.  Between two recent studies by \\citet*{TME98} of southern EGRET sources and by \\citet*{LTP01} of the Pearson-Readhead sample, the misalignment angles of fifty sources are compiled; just two of those sources have misalignment angles greater than $120^\\circ$, and not one is misaligned more than $150^\\circ$.  Here we discuss the highly misaligned radio jet of PKS 1510$-$089 ($z=0.360$), a radio selected, high polarization quasar \\citep{HB93,SMA84}.  PKS 1510$-$089 has been detected in $\\gamma$-rays by EGRET \\citep{HBB99} and exhibits apparent jet motions in excess of 20 times the speed of light \\citep{H01,W02}. Superluminal motion is a natural consequence of a highly relativistic jet pointed nearly right at us \\citep{R66,BK79}, leading to a compression of the observed time scale and magnifying intrinsic pattern speeds: $\\beta_{app} = \\beta\\sin\\theta/(1-\\beta\\cos\\theta)$ where $\\beta_{app}c$ is the observed speed, $\\beta c$ is the intrinsic pattern speed, and $\\theta$ is the angle between the jet axis and the line of sight.  If traveling at the optimum angle for superluminal motion, $\\beta=\\cos\\theta$, the parsec-scale jet of PKS 1510$-$089, at an apparent speed of $20c$, lies within just three degrees of our line of sight.  \\citet*{OBC88} mapped the arcsecond scale structure of PKS 1510$-$089 at 5, 15 and 22 GHz using the Very Large Array (VLA).  They observed a jet extending nine arcseconds to the south-southeast at 5 GHz and an oppositely directed counter-feature at just 0.3 arcseconds at the higher frequencies. Three epochs of early Very Long Baseline Interferometry (VLBI) observations at 1.7 GHz \\citep[summarized in][]{BPF96} appeared to show the milli-arcsecond scale jet directed toward the southern VLA jet. On the basis of these results, PKS 1510$-$089 has been considered a well aligned source in studies of mis-alignment angles in gamma-ray blazars \\citep{TME98,HJS98,C00}.  More recent Very Long Baseline Array (VLBA) observations ranging from 2 GHz up to 43 GHz \\citep{FC97,KVZC98,H01,J01,W02} show the jet extending up to 40 milli-arcseconds (at 2 GHz) to the north-northwest, directly toward the ``counter-feature'' at $0.3\\arcsec$ observed by \\citet{OBC88}.  These more recent and more sensitive observations (as well as the deep 1.7 GHz observations presented here) show no sign of a southern milli-arcsecond jet, suggesting that the early VLBI results may have simply mis-identified the core.  With a highly superluminal milli-arcsecond jet extending to the north-northeast,  a bright VLA feature directly in its path at $0.3\\arcsec$, and an arcsecond VLA jet oppositely directed by $\\simeq 180^\\circ$, the connection between the jets on these scales is an intriguing puzzle. Here we report the results of observations designed to fill the gap in resolution between the previous VLBI and VLA observations of this source.  The observations are described in \\S{\\ref{s:obs}}, and in \\S{\\ref{s:dis}} we suggest and analyze two general models to explain the jet trajectory of this highly superluminal blazar.  Throughout this paper we assume a cosmology with $H_0 = 70$ km s$^{-1}$ Mpc$^{-1}$ , $\\Omega_m = 0.3$, and $\\Omega_\\Lambda = 0.7$, and we choose a spectral index convention: $S_\\nu\\propto\\nu^{+\\alpha}$. ",
        "conclusions": "We find that the apparent $\\simeq 180^\\circ$ misalignment between the milli-arcsecond and arcsecond radio jets in PKS 1510$-$089 is most likely due to a small angle intrinsic bend of $\\simeq12^\\circ-24^\\circ$, viewed in projection.  Our new images show the highly superluminal VLBI jet to point directly at and nearly connect with a previously detected ``counter-feature'' to the arcsecond jet, leaving little doubt that this feature at $0.3\\arcsec$ is actually part of the approaching jet.  We imaged the feature in detail and find it to be dominated by shock enhanced emission and to have very high levels of polarization confined along its outside edge. The projected magnetic field must be almost perfectly ordered at this point, ruling out a simple bow-shock model where the $0.3\\arcsec$ feature marks the terminal spot in the VLBI jet.  Instead, we prefer a shocked-bend model where the jet bends across our line of sight at this point, creating highly ordered magnetic field either as a direct consequence of the shock itself or through shear.  The bend occurs at a de-projected distance of $\\simeq 30$ kiloparsecs from the nucleus, near the probable boundary of the host galaxy. The cause of the bend is uncertain; however, we favor a scenario where the jet is bent after it departs the galaxy, either by ram pressure from winds in the intracluster medium or by the density gradient in the transition to the intergalactic medium."
    },
    "0208/astro-ph0208253_arXiv.txt": {
        "abstract": "A `complete' sample of 174 M giants classified by Blanco (1986) and later than subtype M0 in the NGC\\,6522 Baade's Window clear field has been investigated to establish some general properties of cool Bulge stars. Photometric information has been obtained from the MACHO database to search for variablility and, where possible, to determine periods. Near- and mid-IR magnitudes have been extracted from DENIS and ISOGAL.  Forty-six semi-regular (SR) variables and two irregular variables were found amongst the 174. Many M5 and all stars M6 or later show variation, whereas earlier subtypes (M1--M4) do not. The DENIS $I-J$ and $J-K_S$ colours and the luminosities of the M stars increase with M sub-class. $K$ tends to increase with log\\,$P$ among the M-type SR variables.  Almost all the variables were detected at 7$\\mu$m during the ISOGAL programme. Excess radiation at 15$\\mu$m, indicative of heavy mass-loss, is associated with high luminosity and late spectral type. The limit of sensitivity of the ISOGAL survey was such that the non-variables were not detected.  Four probable M stars not listed by Blanco (1986), two of which are semi-regular variables, were detected by ISOGAL.  In the case of doubly-periodic SR variables, the longer periods have $K$-mags which place them close to the `D' line of Wood (2000) in a $K$, log $P$ diagram. The unusual MACHO light curve of one particular star, Blanco 26, shows the commencement of a long-period variation with an anomalously short and sharp event and appears to rule out a pulsational model for this phenomenon. ",
        "introduction": "The Baade's Window fields in the inner Bulge of the Milky Way galaxy offer the opportunity of studying a sample of galactic stars at a well-defined distance from the sun and with relatively low interstellar absorption.  The mira variable content of the NGC\\,6522 and Sgr I Baade's Windows has been surveyed by Lloyd Evans (1976). A recent census of miras in Sgr\\,I, which includes the results of blue and near-infrared photographic surveys as well as long-period stars found during the IRAS mid-infrared survey, has been given by Glass et al (1995). In the present work, the general population of M giants, including variables of lower amplitude than the miras, is studied in the NGC\\,6522 Window.  A `complete' sample of M stars from M1 to M8 in an annular field of 42 arcmin$^2$ area surrounding the NGC\\,6522 cluster was obtained by Blanco (1986) using a grism attached to the prime focus camera of the CTIO 4m telescope. It should be noted that the classification into M sub-types by Blanco (1986) is not quite on the Morgan-Keenan (MK) system, in that stars of M2 to M6 were stated to have been classified 1-2 subtypes too late by Terndrup, Frogel and Whitford (1990; see also below). Spectroscopy of all Blanco's stars of subtype M6 or later was obtained by Sharples, Walker \\& Cropper (1990), who tabulated their radial velocities and their TiO (8415) and CaII (8662) line strengths.  Unlike the solar neighbourhood and the Magellanic Clouds, the galactic centre fields do not contain late-type C-rich giants. The relatively small Blanco (1986) field also does not include any miras. Lloyd Evans' (1976) figures for the surface density of miras suggest that only about one could have been expected.  The ISOGAL survey (Omont et al., 2002) at 7 and 15$\\mu$m of sample fields along the Galactic Plane and within the Galactic Bulge included a `fiducial' field around the globular cluster NGC\\,6522. Glass et al.\\ (1999) showed that all stars later than M6 were detected, whereas M1 types were not seen at all. It was found that many AGB stars besides the mira variables possess excesses at 15$\\mu$m, indicating that they are losing mass.  Glass, Alves et al.\\ (2000; see also Alard et al., 2001) showed that nearly all the stars that were detected by ISOGAL in the NGC\\,6522 and Sgr\\,I Baade's windows could be identified with late-type variables in the MACHO database. They studied approximately 300 stars that were seen in both the MACHO $v$ and $r$ colours as well as in both the ISOGAL bands at 7 and 15$\\mu$m. Almost all the non-miras were found to be semi-regular variables, with mass-loss rates that ranged from undetectable to $ \\sim5 \\times$ 10$^{-7}$ $M_{\\odot}$yr$^{-1}$. Semiregular variables were seen to outnumber miras by at least 20:1 in these fields. They had periods ranging from 10-200 days and amplitudes which reached $\\sim$0.3 mag at 100d period. A period of 70 days or more was found to be a necessary, though not a sufficient, condition for detectable mass-loss.  Schultheis and Glass (2001) examined the same sample of stars in the near infrared period-magnitude, period-colour and colour-magnitude diagrams derivable from the DENIS $IJK_S$ dedicated survey of the galactic bulge (Simon et al., 2002, in preparation) and the 2MASS $JHK_S$ survey.  The Semi-regular variables inhabit the upper end of the $J-K_S, K_S$ colour-magnitude diagram and their colours and magnitudes are seen to increase in a general way with period. The well-known separation between miras and late-type giants in the $J-H, H-K$ diagram was shown to be primarily caused by the effect of water vapour absorption on the $J-H$ colour index.  In this paper, we examine the complete Blanco (1986) sample of stars in the NGC\\,6522 field classified M1 or later for variability in the MACHO database to find the dependence of variability on spectral type. Because many of these stars fall within the area surveyed by ISOGAL they could also be examined for mass-loss. The previous work (Alard et al, 2001; Schultheis and Glass, 2001) dealt exclusively with stars detected in both MACHO ($r$ and $i$) and both ISOGAL (7$\\mu$m and 15$\\mu$m) bands and was heavily weighted towards stars with mass-loss. There are only 23 stars in common. ",
        "conclusions": "We have examined a `complete' sample of 174 Blanco M giants in the Inner Bulge of the Galaxy for variability and infrared characteristics.  About 46 of the Blanco (1986) stars are semi-regular variables with amplitudes exceeding about 0.03 mag. Stars with spectral types of M6 or later are variable, together with many of type M5, especially the most luminous. More than one period is fairly common.  There is a continuous increase of average luminosity and of $I-J$ and $J-K_S$ colours with M sub-class. A $K_S$, log $P$ trend is shown by SRVs if in each case the most important shorter period is used when a long period is present.  Thirty-seven Blanco stars out of a possible $\\sim$101 were detected by ISO during the ISOCAM programme. It is clear that there is a high correlation between variability and detection but not whether this is due to luminosity or to temperature effects (late spectral type). NGC\\,6522 therefore contains very few semi-regular variable M-type giants that escaped detection in the work of Alard et al (2001).  Four additional probable M stars detected by ISO, two of them semi-regular variables with late M-subtype colours, have been found in the Blanco (1986) field, which shows that his survey is not as complete as formerly believed.  Among those stars with both short and long periods, the long periods appear to fall close to the `D' line in the $K_S$, log $P$ diagram,  found by Wood (2000) for similar objects in the LMC. The unusual light curve for Blanco 26 suggests that pulsation cannot be responsible for the long period variability."
    },
    "0208/astro-ph0208535_arXiv.txt": {
        "abstract": "There is  now strong  evidence suggesting that  the $  ^{12}$CO J=1--0 transition,   widely   used   to   trace  H$_2$   gas,   significantly underestimates its  mass in metal-poor  regions. In spiral  disks such regions are found in large galactocentric distances where we show that any unaccounted  H$_2$ gas phase is  likely to be  diffuse ($\\rm n\\sim 5-20$~cm$ ^{-3}$) and warmer ($\\rm T_{\\rm kin}\\sim 50-100$ K) than the cool  ($\\rm T_{\\rm  kin}\\sim~15-20$~K) CO-luminous  one.   Moreover we find  that  a  high value  of  the  H$_2$  formation rate  on  grains, suggested  by  recent  observational  work,  can  compensate  for  the reduction of  the available  grain surface in  the metal-poor  part of typical galactic  disks and thus enhance this  CO-poor H$_2$ component which may  be contributing significantly  to the mass and  pressure of spiral disks beyond their optical radius. ",
        "introduction": "The use of $ ^{12}$CO,  the most abundant molecule after H$_2$ itself, to trace  H$_2$ gas is a  widely used practice since  the discovery of bright  $ ^{12}$CO  J=1--0  emission in  Orion  (Wilson, Jefferts,  \\& Penzias  1970).   The  most   easily  excited  H$_2$  transition,  the quadrupole  S(0): $\\rm J_{u}-J_{l}=2-0$,  corresponds to  $\\Delta \\rm E_{20}/k\\sim 510$  K, much too  high to be significantly  populated in the  cool  ISM ($\\rm  T_{\\rm  kin}\\sim 10-50$  K)  revealed  by the  $ ^{12}$CO  J=1--0   transition  ($\\Delta  \\rm   E_{10}/k\\sim  5.5$  K). Furthermore the  large optical depths  of the latter  ($\\tau _{10}\\sim 5-10$)  reduce its  critical density  of $\\rm  n^{(10)}  _{\\rm cr}\\sim 380$~cm$ ^{-3}$  (for $\\rm T_{\\rm kin}=50$~K; Kamp  \\& Zadelhoff 2001) to  $\\sim  \\rm  n^{(10)}   _{\\rm  cr}(1-e^{-\\tau  _{10}})  \\tau  ^{-1} _{10}\\sim  40-80\\ cm^{-3}$,  similar to  the lowest  average densities observed  in Giant  Molecular  Clouds (GMCs).   Hence  the $  ^{12}$CO J=1--0 transition  is expected to be  excited even in  the coldest and most diffuse regions of the molecular~ISM.   However  its large  optical depths  (e.g.  Martin,  Sanders,  \\& Hills 1984;  Sage  \\&   Isbell  1991;  Falgarone  1998),  do   not  allow  a straightforward  use  of  its   luminosity  together  with  a  $\\rm  [ ^{12}CO/H_2]$ abundance  ratio for  H$_2$ mass estimates.   The widely used $\\rm X_{\\rm CO}=M(H_2)/L_{\\rm CO}$ factor (where $\\rm L_{\\rm CO}$ is the  velocity/area-integrated brightness temperature  of $ ^{12}$CO J=1--0)  is based  on  the  notion of  an  effectively optically  thin medium, where the $ ^{12}$CO  line emission arises from an ensemble of small,  radiatively  de-coupled  cells  not overlapping  in  space  or velocity.   Then the  high  optical depths  arise  locally within  the individual cells and the observed wide line profiles result from their macroscopic  motions  rather   than  their  intrinsic,  much  narrower linewidths  (e.g.  Martin  et al.   1984; Young  \\& Scoville  1991 and references   therein).   This   picture,  along   with   the  observed virialization of molecular clouds (see  e.g.  Larson 1981) over a wide range of scales ($\\sim 0.1-100$ pc) allows a statistical derivation of $\\rm X_{\\rm CO}$ as an ensemble-average that is relatively insensitive to local molecular cloud conditions (e.g.  Dickman, Snell, \\& Schloerb 1986; Young \\& Scoville 1991; Bryant \\& Scoville 1996) and, for scales $\\ga 10$ pc,  is expected to yield the correct H$_2$  mass to within a factor of two.   The existence of such cells  is supported by high angular and velocity resolution studies  of galactic clouds (e.g.  Falgarone  et al.  1998; Tauber 1996) while the statistical  robustness of $\\rm X_{\\rm CO}$ for a  variety of conditions  has been  verified using  both observational (e.g.  Young \\& Scoville 1991 and references therein), and theoretical methods (e.g. Kutner  \\& Leung 1985; Dickman, Snell  \\& Schloerb 1986; Maloney \\& Black 1988;  Wolfire, Hollenbach, \\& Tielens 1993; Sakamoto 1996; Bryant \\& Scoville 1996).   However these studies also suggest a likely failure  of this method in two  particular environments, namely a) galactic and/or starburst nuclei where non-virial motions cause the luminous  $ ^{12}$CO  emission to  overestimate the  H$_2$  mass (e.g. Dahmen et  al.  1998; Downes  \\& Solomon 1998), b)  metal-poor regions where  it seriously underestimates  the H$_2$  mass (e.g.   Maloney \\& Black  1988; Arimoto,  Sofue, \\&  Tsujimoto 1996;  Israel  1988, 1997, 1999).  The  last case is the  most serious one since  then $ ^{12}$CO emission is usually faint and  thus precludes the observations of more optically thin isotopes  like $ ^{13}$CO, C$ ^{18}$O  that could yield the H$_2$ mass without using the $\\rm X_{\\rm CO}$ factor.   In  this   paper  we  show   that  the  decreasing   metallicity  with galactocentric radius in disks and  a larger H$_2$ formation rate open up the possibility of large  amounts of H$_2$ gas being there hitherto undetected  by the  conventional  method.  This  phase  will be  warm, diffuse  and devoid  of CO,  and may  contribute significantly  to the pressure and total mass of the disk. ",
        "conclusions": "In the  light of mounting evidence  that the standard  method of using rotational line  emission of  the $ ^{12}$CO  molecule to  trace H$_2$ underestimates its mass at low  metallicities we examined the state of such  gas in  the disks  of  spiral galaxies  at large  galactocentric distances.  Our results can be summarized as follows,  \\noindent A high H$_2$ formation rate along with the presence of dust throughout a typical  HI disk, both  suggested by recent observations,  raise the possibility of an extended H$_2$ gas phase, well past the one inferred by the $ ^{12}$CO brightness distribution.  \\noindent This phase is likely to be  warm and diffuse and it may be responsible for  the  observed  high   total  pressures  at  large  galactocentric distances  in the  Galaxy.  In  the  case of  NGC 891  such a  massive gaseous reservoir may have already been detected through its S(0) line emission."
    },
    "0208/astro-ph0208229_arXiv.txt": {
        "abstract": "The ionising stellar populations  of eleven H{\\sevensize II} regions in the spiral galaxies: NGC~628, NGC~925, NGC~1232 and NGC~1637,  all of them reported to have solar or oversolar abundances according to empirical calibrations, have been analysed using stellar population synthesis models.  Four of the observed regions in the sample, show features which indicate the presence of a population of Wolf-Rayet (WR) stars with ages between 2.3 and 4.1 Myr. This population is sufficient to explain the emission line spectrum of the low metallicity region H13 in NGC~628, taking into account the uncertainties involved in both observations and model computations. This is not the case for the rest of the regions for which a second ionising population is required to simultaneously reproduce both the WR features and the emission line spectrum. Composite populations are also found for half of the regions without WR features, although in this case, the result is based only on the emission line spectrum analysis.  For two of the regions showing WR features, no consistent solution is found, since the population containing WR stars produces a spectral energy distribution which is too hard to explain the emission of the gas. Several solutions are proposed to solve this problem. ",
        "introduction": "Metallicity affects stellar evolution in at least two different ways: by increasing the opacity of the stellar material and through the strengthening of the wind driven mass loss in high mass stars. The former implies a lower effective temperature for the atmospheres of ionising stars of higher metal content even at zero age, while the latter can severely affect the evolution of the most massive stars leading, in the most extreme cases, to the almost complete loss of their outer envelopes.  The first of these effects should be readily observable as an inverse correlation between metallicity and temperature of the ionising radiation for HII regions of different metallicities. In fact, there is a general agreement about the hardening of the ionising radiation in regions of low metal content (e.g. Campbell et al. 1986; Cervi{\\~n}o and Mas-Hesse 1994). The inverse situation, however, i.e. the softening of the ionising radiation in regions of high metal content, is more difficult to establish since, in these cases, the determination of both metal abundance and ionising temperature is rather difficult. For the first one we have to rely , in general, on uncertain semi-empirical calibrations when the involved abundances are higher than about 1/2 solar. For the second one, there are no suitable nebular emission line diagnostics. Recently, Bressolin, Kennicut \\& Garnett (1998) have investigated a possible relation between ionising stellar temperature and metallicity in giant HII regions in spiral galaxies through the use of the softness parameter $\\eta$' (V{\\'\\i}lchez \\& Pagel 1988) concluding that their results are consistent with a significant decrease in mean stellar temperatures of the ionising stars with increasing metallicity. This decrease is however difficult to quantify since in this work the metallicity is derived from the abundance parameter R$_{23}$ (Pagel et al. 1979) and both  $\\eta$' and R$_{23}$ depend to some extent on the degree of ionisation of the nebula (D{\\'\\i}az et al. 1991). On the other hand, the method can be applied to a large number of HII regions and therefore can provide trends statistically significant.  A different approach consists in deriving the metal abundance from the direct determination of the electron temperature in the nebula and the ionising temperature from the fitting of single star photoionisation models. The method requires both high quality spectroscopy and detailed modelling. In a previous paper (D{\\'\\i}az et al. 2000a; DCTG00) we have applied this procedure to HII regions in the spiral galaxy NGC~4258. The combination of these data with similar ones in the literature also seems to point to a decrease of mean ionising temperature with increasing metallicity. Surprisingly, this is so even in the presence of WR stars.  The presence of these stars provides an additional constraint to characterise the ionising stellar population. In the last years, stellar evolution models for massive stars taking into account mass loss at different metallicities have been calculated in order to reproduce the observed features of WR stars. The model predictions are however quite different depending on the assumed mass loss. In fact, models that assume a standard mass loss rate underpredict the number of galactic WR stars in comparison with observations (Maeder \\& Meynet 1994) while an enhanced mass loss rate fits most WR population properties except for the mass loss rate itself (Leitherer, Chapman \\& Koribalski 1997). Therefore, the detailed observation and modelling of giant HII regions showing WR features can help to shed some light on this very important issue.  It would also be desirable to check the hardening of the ionising radiation due to the presence of Wolf-Rayet stars predicted on theoretical grounds (P{\\'e}rez 1997). These stars are supposed to change drastically the spectral energy distribution of an ionising star cluster at energies higher than {$\\sim$} 40 eV. Models predict that this change depends both on the age and the metallicity of the burst (Schaerer \\& Vacca 1998; SV98) and, as a result of this, ionic elemental ratios are expected to change as well as ion-weighted temperatures.  A puzzling result in DCTG00 was the relatively low effective temperatures found in the two ionising clusters with WR features, suggesting either that the rate of high energy photons is lower than expected or that the opacity in WR envelopes is highly efficient. This matter can be investigated through the observation of WR features in high metallicity HII regions where the effect of the mass loss rate on the effective temperature of the ionising stars is supposed to be most enhanced.  In order to analyse the effects of metallicity on both  stellar evolution and ionising radiation, we have analysed a sample of reported high metallicity H{\\sevensize II} regions by combining evolutionary synthesis models for ionising populations and  photoionisation models in a selfconsistent way.  In this paper a detailed analysis of the ionising populations of the observed regions is presented. ",
        "conclusions": "We have analysed the stellar populations of eleven  H{\\sevensize II} regions in the galaxies NGC 628, NGC 1232, NGC 925 and NGC 1637 using spectrophotometric observations between 3500 and 9700 {\\AA}.  We have  derived the physical properties of the regions and their corresponding ionising clusters: filling factor, mass of ionised gas and mass of ionising stars. Most of the regions have small ionising clusters with masses in the range 1500 to 30000 solar masses. The exceptions are the three supergiant HII regions in NGC 1232, CDT1, CDT3 and CDT4 with masses greater than 100,000 solar masses. These values constitute in fact lower limits since the regions are assumed to be ionisation bounded and the presence of dust has not been taken into account.  WR features have been observed in the four supergiant HII regions,  H13 in NGC~628 and CDT1, CDT3, CDT4 in NGC~1232. A very detailed modelling has been carried out for these regions by using two different sets of models: those of SV98 for WR star populations and those of Leitherer et al. (1999) for ionising populations, to try to reproduce simultaneously both the WR features and the emission line spectrum of each region. In all cases, the agreement between both predicted and observed values for the WR luminosities and equivalent widths is excellent. This fact seems to indicate that the stellar evolution assumed by the models is predicting the right WR/O ratios at the different metallicities involved.  The ages of the populations containing WR stars are found to be between 2.3 and 4.1 Myr. This population is sufficient to explain the emission line spectrum of the low metallicity region H13 in NGC~628, taking into account the uncertainties involved in both observations and model computations.  This is not the case for the rest of the regions. For CDT1 in NGC~1232, a second ionising population is required to simultaneously reproduce both the WR features and the emission line spectrum. This second population is older than the previous one by about 5 Myr and contributes about 10\\% of the ionising photons and 5 times the continuum luminosity at 9000 {\\AA}. Composite populations are also found for half of the regions without WR features, although in this case, the result is based only on the emission line spectrum analysis.  For the other two regions containing WR stars, CDT3 and CDT4 in NGC~1232, no consistent solution is found, since the population containing WR stars produces a spectral energy distribution which is too hard to explain the emission of the gas. Several solutions are proposed to solve this problem. Both a reduction in the number of high energy photons of the ionising clusters, which could arise naturally with the inclusion of blanketing by metallic lines in the atmospheres of the WR stars, and/or the assumption that the HII regions be matter bounded can effectively solve the problem. The latter hypothesis has been tested successfully (Castellanos et al. 2002) while the former one still remains to be explored."
    },
    "0208/astro-ph0208159_arXiv.txt": {
        "abstract": "The Capodimonte Deep Field (OACDF) is a multi-colour imaging survey on two 0.5x0.5 square degree fields performed in the BVRI bands and in six medium-band filters (700 - 900 nm) with the Wide Field Imager (WFI) at the ESO 2.2 m telescope at La Silla, Chile. In this contribution the adopted strategies for the  OACDF data reduction are discussed. Preliminary scientific results of the survey are also presented. ",
        "introduction": "In view of the arrival of the VLT Survey Telescope (VST) (a cooperation of ESO, the Capodimonte Astronomical Observatory-OAC-Naples, Italy, and the European Consortium Omegacam, for the design, realization, installation, and operation at ESO Paranal Observatory in Chile of a 2.6 m aperture, 1 degree x 1 degree wide field imaging facility in the spectral range from UV to I bands), the OAC started a pilot project, called the Capodimonte Deep Field (OACDF), consisting in a multi-colour imaging survey using the Wide Field Imager (WFI, Baade et al. 1998 ) at the ESO 2.2~m telescope at La Silla, Chile. The VST will be equipped with OmegaCam, a 16k X 16k array of 32 CCDs, which will cover 1 square degree. The main goal of the OACDF is to provide a large photometric database, mainly oriented to extragalactic studies (quasars, high-redshift galaxies, galaxy counts, lensing, etc.), that can be used also for stellar and planetary research (galactic halo population, peculiar objects like brown dwarfs and cool white dwarfs, Kuiper-Belt objects). Another goal is to gain insight into the handling and processing of data coming from a wide field imager, similar to the one which will be installed at VST.  In this contribution we discuss technical aspects of the OACDF data reduction and present some preliminary scientific results. We focus on the following topics: super flat-fielding and defringing of the mosaics, astrometry and photometric calibration. Such topics will be of primary importance for the processing of the VLT Survey Telescope (VST) data, in the framework of the European consortium ASTRO-WISE (see http://www.ASTRO-WISE.org). Some of the problems encountered are intrinsic to the ESO WFI. Hence, our tests might contribute to a better characterisation of this instrument. In addition, we report on some preliminary scientific results of the OACDF project. The depth of the OACDF (25.1 mag in the R band) allows to achieve the foreseen scientific goals: i) the search for rare/peculiar objects, AGNs and high-redshift QSO's (z$>$3); ii) the search for intermediate-redshift early-type galaxies to be used as tracers of galaxy evolution and iii) the search for galaxy clusters to be used as targets for spectroscopic follow-up's at larger telescopes. ",
        "conclusions": ""
    },
    "0208/astro-ph0208190_arXiv.txt": {
        "abstract": "We present observations of the radio emission from the unusual supernova SN 1988Z in MCG +03-28-022 made with the Very Large Array at 20, 6, 3.6, and 2 cm, including new observations from 1989 December 21, 385 days after the optically estimated explosion date, through 2001 January 25, 4,438 days after explosion. At a redshift $z = 0.022$ for the parent galaxy ($\\sim$100 Mpc for $\\rm{H}_0 = 65 ~\\rm{km} ~\\rm{s}^{-1} ~\\rm{Mpc}^{-1})$, SN 1988Z is the most distant radio supernova ever detected. With a 6 cm maximum flux density of 1.8 mJy, SN 1988Z is $\\sim$20\\% more luminous than the unusually powerful radio supernova SN 1986J in NGC 891 and only $\\sim$3 times less radio luminous at 6 cm peak than the extraordinary SN 1998bw, the presumed counterpart to GRB 980425. Our analysis and model fitting of the radio light curves for SN 1988Z indicate that it can be well-described by a model involving the supernova blastwave interacting with a high-density circumstellar cocoon, which consists almost entirely of clumps or filaments.  SN 1988Z is unusual, however, in that around age 1750 days the flux density begins to decline much more rapidly than expected from the model fit to the early data, without a change in the absorption parameters.  We interpret this steepening of the radio flux density decline rate as due to a change in the number density of the clumps in the circumstellar material (CSM) without a change in the average properties of a clump.  If one assumes that the blastwave is traveling through the CSM at $\\sim2,000$ times faster than the CSM was established (20,000 \\kms ~vs. 10 \\kms), then this steepening of the emission decline rate represents a change in the presupernova stellar wind properties $\\sim10,000$~yrs before explosion, a characteristic time scale also seen in other radio supernovae.  Further analysis of the radio light curves for SN 1988Z implies that the SN progenitor star likely had a ZAMS mass of $\\sim20$--30 $M_\\odot$.  We propose that SNe, such as SN 1986J, SN 1988Z, and SN 1998bw, with very massive star progenitors and associated massive wind ($\\dot M\\gtrsim 10^{-4}\\ M_\\odot$ yr$^{-1}$) have very highly-clumped, wind-established CSM and unusually high blastwave velocities ($>20,000$~\\kms). ",
        "introduction": "Supernova (SN) 1988Z was independently discovered in MCG +03-28-022 (Zw 095-049) near $\\rm{m_B} \\sim 16.4$ by both G.~Candeo on 1988 December 12 (Cappellaro, Turatto, \\& Candeo 1988) and C.~Pollas on 1988 December 14 (Pollas 1988). The SN was classified as a Type II, based on the detection of hydrogen emission lines in the optical spectra by Cowley \\& Hartwich (Heathcote et al.~1988), which also indicated that the SN was quite distant, with redshift $z \\sim 0.022$.  Filippenko (1989) confirmed the Type II identification, with possible resemblance to SN 1987F. The spectra revealed at early times that SN 1988Z was peculiar, with a remarkably blue color, a lack of absorption lines and P-Cygni profiles, very narrow ($\\lesssim$100 km s$^{-1}$ FWHM) [O III] circumstellar emission lines, and a steadily growing, relatively narrow ($\\simeq$2000 km s$^{-1}$ FWHM) component to the H I and He I lines (Stathakis \\& Sadler 1991; Filippenko 1991a,b). Stathakis \\& Sadler (1991), and Turatto et al.~(1993) for later times, analyzed in detail the spectroscopic and photometric observations of SN 1988Z. Schlegel (1990) proposed that SN 1988Z, along with other similar SNe, constituted a new class of SNe, the Type II-``narrow,'' or SNe IIn.  The properties of the optical spectra and light curves indicated strong SN shock-circumstellar shell interaction (Filippenko 1991a,b; Turatto et al.~1993) which, together with its resemblance optically to SN 1986J (Filippenko 1991a,b; see also Rupen et al.~1987, Leibundgut et al.~1991), strongly suggested that SN 1988Z should be a very luminous radio emitter, as SN 1986J is a strong radio source (see Weiler, Panagia, \\& Sramek 1990). Sramek et al.~(1990) reported the radio detection at 6 cm wavelength of SN 1988Z with the Very Large Array (VLA)\\footnote{The VLA telescope of the National Radio Astronomy Observatory is operated by Associated Universities, Inc. under a cooperative agreement with the National Science Foundation.} on 1989 December 21. The supernova is located at RA(J2000) = $10^h 51^m 50\\fs138$, Dec(J2000) = $+16\\arcdeg 00\\arcmin 00\\farcs16$, with an uncertainty of $\\pm$ 0\\farcs2 in each coordinate, which is coincident, to within the uncertainties, with its optical position (Kirshner, Leibundgut, \\& Smith 1989). This initial announcement of the detection of radio emission was followed by a more thorough study and analysis of the first three years of multifrequency radio observations by Van Dyk et al.~(1993a).  They compared SN 1988Z in detail with SN 1986J and concluded that the two SNe are very similar in their radio properties. They suggested that the progenitor to SN 1988Z was a massive [$20 \\leq~ M(M_\\odot) \\leq 30$] star which underwent a high mass-loss phase ($\\dot M \\ga 10^4\\ M_\\odot$ yr$^{-1}$) before explosion (see also Stathakis \\& Sadler 1991).  Luminous radio emission has also been an indicator of X-ray emission from SNe, with synchrotron radio emission being produced as the forward SN shock interacts with the CSM and X-rays being emitted from the corresponding reverse shock region interacting with the SN ejecta (Chevalier \\& Fransson 1994).  For instance, SN 1986J was detected as a luminous X-ray source by Bregman \\& Pildis (1992) and Houck et al.~(1998). Fabian \\& Terlevich (1996) reported the detection of X-rays from SN 1988Z with ROSAT. SN 1988Z is a very luminous X-ray emitter with a bolometric X-ray luminosity of $\\sim 10^{41}~\\rm{erg}~\\rm{s}^{-1}$.  Chugai \\& Danziger (1994) offered two models to explain the unusual characteristics of SN 1988Z: 1) shock interaction with a two component wind consisting of a tenuous, homogeneous medium with embedded higher density clumps, or 2) shock interaction with a similar tenuous, homogeneous medium and a higher-density, equatorial, wind-established, disk-like component, favoring the former over the latter. However, they unexpectedly conclude that the SN ejecta is of low mass ($M < 1\\ M_\\odot$), and that SN 1988Z may have originated from a relatively low-mass star of $M_{\\rm ZAMS} \\sim 8$--10 $M_{\\odot}$, in sharp contrast to the high-mass progenitor suggested by Van Dyk et al.~(1993a) and Stathakis \\& Sadler (1991).  Aretxaga et al.~(1999) collect the observations from X-ray to radio and attempt to estimate the integrated electromagnetic energy radiated by SN 1988Z in the first 8.5 years after discovery. They obtain a value of $\\ge 2 \\times 10^{51}$ erg, perhaps as high as $10^{52}$ erg, which they consider is sufficiently high to suggest that SN 1988Z could be classified as a ``hypernova,'' approaching the 2--$5 \\times 10^{52}$ ergs estimated to have been released in SN 1998bw, the possible counterpart of GRB 980425 (Iwamoto et al.~1998), and perhaps indicative of the collapse of the stellar progenitor core into a black hole. (It is interesting to note here that two SNe IIn, 1997cy and 1999E, may also have been associated with $\\gamma$-ray bursts; see Pastorello et al.~2002 and references therein). Aretxaga et al.~(1999) suggest an ejecta mass of $\\sim15\\ M_\\odot$, and, therefore, a very high-mass progenitor.  Obviously, SN 1988Z is an extremely interesting and, in many ways, unusual object.  Fortunately, due to its intrinsic brightness it has been relatively well-studied in many wavelength bands.  Here we report radio observations of SN 1988Z at multiple wavelengths, including new observations which add another six years of monitoring, more than doubling the coverage reported by Van Dyk et al.~(1993a).  We further interpret its radio emission using a clumpy wind model.  We conclude that SN 1988Z, like SNe 1986J and, possibly, other SNe IIn, as well as the unusual SN 1998bw, arise from the explosions of very massive stars surrounded by highly filamentary CSM. ",
        "conclusions": "The radio emission from SN 1988Z followed an evolution well described by standard models up to an age of $\\sim1750$ days ($\\sim4.8$ yrs), after which its behavior evolved into a much more rapid decline in radio emission without a corresponding change in radio absorption parameters.  This is the first time we have witnessed this in any well-studied RSN, and it implies that highly radio-luminous RSNe, such as SN 1988Z and, possibly, SN 1986J, may follow a different evolutionary path in their presupernova mass-loss than that for more ``normal'' RSNe, such as SNe 1979C and 1980K. However, it is interesting to note that, if one assumes a blastwave velocity $\\sim2,000$ times faster than the RSG wind-established CSM ($\\sim20,000$ \\kms\\ vs.~$\\sim10$ \\kms), the timescale for such a variation in SN 1988Z of $\\sim10^4$ yr, is similar to that seen for SNe 1979C and 1980K (see, e.g., for SN 1979C, Montes et al.~1998; for SN 1980K, Montes et al.~2000), and also for SN 1998bw (Weiler et al.~2001).  In addition, further analysis of the results from Chugai \\& Danziger (1994), which implied an extremely high mass-loss rate of $7 \\times 10^{-4} M_\\odot$ yr$^{-1}$ and a relatively low-mass progenitor with $M_{\\rm ZAMS} \\sim 8$--10 $M_\\odot$ for SN 1988Z, indicates that their estimates are probably unrealistic. Using their modeling, our lower mass-loss rate, $1.2 \\times 10^{-4}\\ M_\\odot$ yr$^{-1}$, implies a significantly higher ejecta mass and, therefore, a higher ZAMS mass of $\\sim20$--30 $M_\\odot$ for the progenitor star.  We have noted that the H${\\alpha}$ data from Aretxaga et al.~(1999), which are also indicative of the SN shock-CSM interaction (Chevalier \\& Fransson 1994), also begin to deviate significantly from their model by day $\\ga$1250.  Although a detailed comparison is beyond the scope of this paper, both the slope before the break and the magnitude of the break are roughly consistent with the radio results.  From our analysis we propose that SNe, such as SNe 1988Z, 1986J, and 1998bw (possibly the counterpart to GRB 980425), with possible very massive star progenitors ($M_{\\rm ZAMS} > 20\\ M_\\odot$) and associated massive winds ($\\dot {M} \\ga 10^{-4}\\ M_\\odot$~yr$^{-1}$), have very highly-clumped, wind-established CSM and unusually high blastwave velocities ($>20,000$~\\kms)."
    },
    "0208/astro-ph0208473_arXiv.txt": {
        "abstract": "{An improved microscopic diffusion in stars is presented considering in detail the partly ionized stages of metals. Besides, the influence of degenerate electron-gas and of the contribution of radiation to the total pressure has been accounted for. The solution of the diffusion equations is then performed following the scheme of \\cite{Diffc}. By defining one mean charged ion per element very few modifications are necessary to solve the improved diffusion scheme. (A portable FORTRAN routine is provided.) The change in the sound-speed profile of a solar model obtained with the new diffusion description is at most about 25\\% at $r=0.6 \\,R_\\odot$. The biggest effect on low-mass stars is expected near the turn-off, where the convective envelope is shallowest. However, only a difference of at most 40~K in the effective temperature could be observed when assuming either fully or partly ionized metals in the diffusion equation. Nevertheless, the surface metal distribution is strongly altered. ",
        "introduction": "In the last few years the precise measurement of solar oscillations has challenged theorists to compute solar models of gradually higher accuracy. This demanded an improvement of the existing input physics like equation of state \\citep{OPEOS}, opacities \\citep{OP96} and nuclear reaction rates \\citep{Adel98}. In addition, the formerly neglected process of microscopic diffusion has been found to improve the agreement of solar models with helioseismic data considerably \\citep{BP92}. Nevertheless, there exists a significant discrepancy in the sound speed just below the convective between the theoretical predicted and the helioseismic determined value, the reason of which is still unknown.  \\mybf{Unfortunately, the expression ``diffusion'' is not always used in the literature to cover the same physical process(es). In order to avoid confusion, ``diffusion'' is defined in this work like, e.g., in \\cite{Bah98} or \\cite{SGW00}, to describe abundance changes due to pressure, temperature, and concentration gradients neglecting the effects of radiative forces.}  \\mybf{Beside solar models} diffusion is now also implemented in many stellar calculations, e.g., for the computation of globular cluster isochrones \\citep{CCIF97,SGW00}. With the improved models globular clusters have been found to be about \\mbox{1\\,Gyr} younger than determined previously by models without diffusion.  However, recent observations of the surface iron abundance of near turn-off stars \\citep{RCBB01,RC02} suggests that diffusion is much less efficient in metal-poor stars with thin convective envelopes than theoretically predicted. Even worse, the models predict an almost total depletion of heavy elements in the surface of such stars at the turn off (see Sect.~\\ref{lowmass}). Including the \\mynbf{here neglected} effect of radiative forces, the opposite effect can be obtained for some elements, e.g., \\element[][]{Fe} may then be enhanced considerably at the surface \\citep[]{RMR02}. \\mynbf{Thus, radiative levitation is important in certain evolutionary phases, and should be included in future models.}  \\mynbf{Nevertheless,} by assuming an additional mixing process below the convective envelope it would be possible to inhibit any diffusion process. Such a process is also favoured to reduce the discrepancy in the sound speed just below the solar convective zone \\citep{Richard96}.  Various mechanisms have been proposed to cause additional mixing. Before, however, being able to determine the extent of additional mixing processes or other non-standard physics, diffusion should be followed accurately. Therefore, the validity of physical assumptions used to compute the diffusion efficiency is investigated in this work. A common assumption in the calculation of diffusion constants is the complete ionization of all elements. Basically in all stellar model computations including diffusion \\citep[e.g.][]{CCIF97,SGW00,WS00,CFN01} this approximation is made. \\mybf{Exception are the models of \\cite{RMT00} or \\cite{RMR02} which account for partial ionization and radiative levitation in the diffusion treatment.}  Since the cross section for the main scattering process of ions in the stellar plasma, the Coulomb scattering, is proportional to the square of the ion charge, deviations from complete ionization may have an important influence on the diffusivity of the elements. But the diffusion constant of a specific element is not simply direct proportional to the cross section, because diffusion has to obey the laws of mass and charge conservation. Thus, the cross section of each element has to be considered in relation to the ones of all other elements. An exact knowledge of the ionization stage of each element is therefore necessary.  In order to obtain more accurate microscopic diffusion constants the assumption of fully ionized metals is dropped in this work. Instead, the ionization stage of each metal is considered in detail, where the ionization degrees of each element are determined by using an up-to-date EOS of \\cite{Irwin}.  In the next section the implementation of partly ionized elements into the solution of Burgers' equations~\\citep{Burgers} is described. The changes in the solar sound-speed profile and in the evolution of metal-poor stars using this improved description are discussed afterwards. ",
        "conclusions": "Starting from Burgers' equation for a multicomponent fluid the effect of partly ionized metals has been consistently included in the diffusion equation following the procedure of TBL. As an additional feature, electron degeneracy and the contribution of radiation to the total pressure has been taken into account. It has been shown, that for most stars it is sufficient to use one ion per element carrying the mean charge instead of computing diffusion for each ion separately. By this means, the elaborate treatment of partly ionized metals remains feasible. With increasing stellar mass, and thus decreasing stellar convective envelope,  it may become necessary to consider, at least, all \\He-ions separately. However, since these stars are usually hotter, radiative levitation becomes important, which then should be included, too.  The effect of the improved diffusion treatment of partly ionized metals has been investigated in case of the Sun and the TO properties of low-mass metal-poor stars. In order to obtain the accurate ionization degrees of each metal species the equation of state of \\cite{Irwin} has been employed. The sound-speed profile in solar models obtained with this new EOS is in as good agreement with the helioseismic determined one as models with the updated OPAL01-EOS.  An improvement of up to 25\\% between $r=0.25 R_\\odot$ and $r=0.65 R_\\odot$ could be achieved by including the effect of partly ionization into the diffusion equation. However, this value is presently still exceeded by the uncertainties in other input physics like the EOS. Nevertheless, the changes between models treating metals fully and partly ionized have been found to be smaller than claimed by \\cite{TRM98}. \\mybf{The origin of this discrepancy is not clear, but might be due to the use of different equations of state.}  In low-mass metal-poor stars the improved diffusion description causes strong deviations in the surface metal distribution from the initial one. In addition, the depletion of metals is much stronger than in the case when full ionization is assumed. The consequent change in the effective temperature is at most about 40~K for a 1.1~$M_\\odot$ star diminishing with decreasing mass.  The influence of radiative levitation on the diffusive behaviour in particular of the metals has been neglected in this work. This may reduce the amount of depletion, and may further alter the metal distribution. Besides, since the effect of using partly instead of fully ionized metals in the diffusion equation is biggest for small convective envelopes, a small amount of additional mixing below the convective boundary may reduce the effect of diffusion strongly.  With the mass of the convective envelopes of metal-poor stars with $M>1~M_\\odot$ being very small ($\\la 10^{-8}~M_\\odot$), also the amount of mass loss may strongly modify the surface abundances. However, mass-loss mechanisms are only very poorly understood. Therefore, first of all, precise stellar models are needed to disentangle the extent of mass loss and to determine the influence of rotation.  \\acknowledgement{This work has been supported by a Marie Curie Fellowship of the European Community programme ``Human Potential'' under contract number HPMF-CT-2000-00951. I would like to thank M.\\  Salaris for useful comments and inspiring discussions, and A.\\  Weiss, who kindly provided splined versions of the opacity tables. In addition, I am grateful to A.W.\\  Irwin for providing me his equation of state.}  \\appendix"
    },
    "0208/astro-ph0208428_arXiv.txt": {
        "abstract": "{ We present the results of moderate resolution spectroscopy for a globular cluster in the M81 group dwarf spheroidal galaxy DDO~78. The DDO~78 globular cluster, 4 Milky Way globular clusters, spectroscopic and radial velocity standards were observed with the Long-slit spectrograph of the 6-m telescope (SAO RAS, Russia). Lick spectrophotometric indices were determined in the bandpasses adopted by Burstein et al. (1984). We have derived the mean metallicity of the globular cluster in DDO78 to be [Fe/H] $=-1.6 \\pm 0.1$ dex by taking the weighted mean of metallicities obtained from the strength of several absorption features. We have estimated an age for the globular cluster of 9-12 Gyr similar to that found for the Galactic globular cluster NGC~362, which resembles our cluster by chemical abundance and integrated spectrophotometric properties. ",
        "introduction": "The dominant type of dwarf galaxies in the local Universe is the dwarf spheroidal galaxies (DSphs). These systems are characterized by low luminosities ($M_\\mathrm{V} > -14$), low stellar density ($\\mu_\\mathrm{v}(0) \\ga 22 ^m/\\sq\\arcsec$), small spatial extent (core radii $<$ a few hundred parsecs), HI masses $M_\\mathrm{HI} \\la 10^5 M_{\\sun}$, lack of current star formation (Grebel 2000, Stetson et al. 1998). New deep C-M diagrams and spectroscopy of individual red giants in several Local Group dSphs have shown us that, in spite of their current quiescent appearance, most of the systems have had surprisingly complicated evolutionary histories (Harbeck et al. 2002). It would be interesting to compare the chemical and star formation histories for DSphs in the Local group and in the other nearby groups.  It is a difficult task, because the observational magnitude limit does not allow us to observe the HB stars in the galaxies at a distance of $>$ 3 Mpc. Globular clusters (GCs) to be found in some DSphs are ideal probes of chemical histories and evolution of their host galaxies. They are bright enough to be observed at large distances. If the correlations between key spectral indices, metallicities and ages are common for GCs in all types of galaxies, we can determine the age for the given particular GC by using the evolutionary population synthesis models (e. g. Worthey 1994) and compare our results with the data available in literature on the Milky Way globular clusters.  The group of galaxies around the bright spiral M81 is one of the nearest prominent groups in the vicinity of the Local Group. It is situated at a distance of $\\sim 3.7$ Mpc and consists of at least 12 DSphs and 12 dwarf irregular galaxies (Karachentsev et al. 2002). DDO~78 is one of the faintest DSphs of the group ($M_\\mathrm{V}=-12.83$). According to Karachentseva et al. (1987), DDO~78 has a very flat surface brightness profile with $\\mu_\\mathrm{B(0)} = 25.1^m/\\sq\\arcsec$, $r_\\mathrm{ef}=32\\arcsec$, and $B(t)=15.8$. It was not detected in H{\\sc I} by Fisher \\& Tully (1981) and van Driel et al. (1998). An accurate distance modulus of DDO~78 is $(m-M)_0=27.85 \\pm 0.15$ according to Karachentsev et al. (2000). The latter authors found globular cluster candidates in five dwarf spheroidal galaxies of the M81 group and measured their basic photometric parameters. The brightest globular cluster candidate in DSph galaxy DDO~78 has the integrated apparent magnitude $V_\\mathrm{t}=19.45$, the integrated color after correction for Galactic reddening $(V-I)_{0}=1.07$, the angular half-light radius $R(0.5L)=0.3 \\arcsec$, the central surface brightness $\\mu_v(0)=18.0 ^m/\\sq\\arcsec$, the linear projected separation of the globular cluster from the galaxy center 0.26 kpc. Sharina et al. (2001) measured a heliocentric radial velocity of the globular cluster candidate in DDO~78 to be  $55\\ \\pm 10\\  \\mathrm {km}\\ \\mathrm{s}^{-1}$ by cross-correlation with template stars and established the object as a {\\it bona fide} member of the galaxy. The present work continues our study of the dwarf spheroidal galaxies in the M81 group. The spectra of the globular cluster in DDO~78 (Fig.1) are of rather high signal-to-noise ratio  and are suitable for the quantitative spectrophotometric study. ",
        "conclusions": "According to Burgarella et al. (2001) the mean metallicity of metal-poor globular cluster systems is weakly dependent on the host galaxy's properties and is almost \"universal\" at [Fe/H]~$\\sim-1.4 \\pm 0.3$~dex. The metallicity of the GC in DDO78 [Fe/H]~$=-1.6 \\pm 0.1$~dex agrees well with this value. There are two DSph companions of the Milky Way which contain globular clusters: Fornax and Sagittarius DSph galaxies. Fornax DSph has the absolute integral magnitude $M_\\mathrm{V} = -13.7$, and the distance from the Milky Way 140 kpc (van den Bergh 2000). These properties are comparable to those of DDO~78, which is located at the distance of 223 kpc from M81 (Karachentsev et al. 2002). Five Fornax DSph globular clusters have the mean metallicity [Fe/H]~$=-2.04$ dex, essentially the same ages ( $\\mid \\delta t \\mid < 1$ Gyr) and are coeval with the old, metal-poor clusters of our Galaxy (Buonanno et al. 1998) Unfortunately, there are no measured integrated absorption indices for all Fornax GCs in the literature. Fornax DSph GCs 3 and 5 show absorption feature strengths (Huchra et al. 1996) unlike those for the GC in DDO78. Our cluster seems to be more metal-rich than the Fornax DSph GCs. We found only one Galactic globular cluster with measured integral absorption indices, NGC~362, which resembles the GC in DDO78 by all properties within the errors. NGC~362 has the galactocentric radius $R_{gc}=9.2$ kpc, the integral absolute visual magnitude $M_\\mathrm{V} =-8.4$ and resembles by its properties some intermediate-metallicity clusters with $R_{gc} > 8$ kpc, Pal~12, NGC~1851, NGC~1261, NGC~2808 (Rosenberg et al. 1999). These clusters are younger by $ \\sim 2$ Gyr than the metal-poor halo GCs and are associated with so-called `streams' that may be relics of ancient Milky Way satellites which had masses typical of a dwarf galaxy.  The M81 group of galaxies has a similar to the LG structure (Karachentsev et al. 2000). The subgroups around the Sb type galaxies, the Milky Way and M31, have a spatial separation of $\\sim 1$ Mpc and approach to each other at a velocity of $\\sim 130$ $ \\mathrm {km}\\ \\mathrm{s}^{-1}$. Such a situation resembles the M81/NGC~2403 complex. But there are some differences between the two groups. The core members of the M81 group: M81, M82, NGC~3077 and NGC~2976 are known to be closely interacting from the aperture synthesis maps of the 21 cm H{\\sc I} emission (Yun et al. 1994, Boyce et al. 2001). The dwarf spheroidal galaxies are distributed around M81 asymmetrically (Karachentsev et al. 2001). With respect to the group centroid (located between M81 and M82) all DSphs are concentrated in one quadrant. Jonson et al. (1997) reported about the discovery of the high-excitation H{\\sc II} region in  K61, the brightest DSph galaxy of the M81 group and the closest companion to M81. They found the metallicity of H{\\sc II} region, Z, between 0.001 and 0.008, and the age between 2 and 5.2 Myr. The properties of DSph galaxies appear to be correlated with the galaxy mass and with environment. Does the tidal interaction between the brightest M81 group galaxies influence the distribution and star formation histories of DSphs? Accurate numerical simulations are needed."
    },
    "0208/astro-ph0208102_arXiv.txt": {
        "abstract": "Galaxy cluster surveys provide a powerful means of studying the density and nature of the dark energy.  The redshift distribution of detected clusters in a deep, large solid angle SZE or X--ray survey is highly sensitive to the dark energy equation of state.  Accurate constraints at the 5\\% level on the dark energy equation of state require that systematic biases in the mass estimators must be controlled at  better than the $\\sim$10\\% level.    Observed regularity in the cluster population and the availability of multiple, independent mass estimators suggests these precise measurements are possible.  Using hydrodynamical simulations that include preheating, we show that the level of preheating required to explain local galaxy cluster structure has a dramatic effect on X--ray cluster surveys, but only a mild effect on SZE surveys.  This suggests that SZE surveys may be optimal for cosmology while X--ray surveys are well suited for studies of the thermal history of the intracluster medium. \\vspace{1pc} ",
        "introduction": "Galaxy clusters have long been used to study dark matter and cosmology in general.  Cluster surveys in the local universe are particularly useful for constraining a combination of the matter density parameter $\\Omega_M$ and the normalization of the power spectrum of density fluctuations \\citep[we describe the normalization using $\\sigma_8$, the {\\it rms} fluctuations of overdensity within spheres of 8$h^{-1}$~Mpc radius; i.e.][]{henry97,viana99, reiprich02}; surveys that probe the cluster population at higher redshift are sensitive to the growth of density fluctuations, allowing one to break the $\\Omega_M$-$\\sigma_8$ degeneracy that arises from local cluster abundance constraints \\cite{eke96,bahcall98}.  Wang \\& Steinhardt \\cite{wang98} argued that a measurement of the changes of cluster abundance with redshift would provide constraints on the dark energy equation of state parameter $w\\equiv p/\\rho$.  Describing the problem in terms of cluster abundance only makes sense in the local universe , because, of course, one cannot measure the cluster abundance without knowing the survey volume; the survey volume beyond $z\\sim0.1$ is sensitive to cosmological parameters that affect the expansion history of the universe-- namely, the matter density $\\Omega_M$, the dark energy density $\\Omega_E$ and the dark energy equation of state $w$.  A cluster survey of a particular piece of the sky with appropriate followup actually delivers a list of clusters with mass estimates and redshifts-- that is, the redshift distribution of galaxy clusters above some detection limit. Haiman, Mohr \\& Holder \\cite{haiman01} showed how the redshift distribution from large X--ray or Sunyaev-Zel'dovich effect (SZE) cluster surveys allows for precise measurements of the dark energy equation of state that are competitive with constraints possible from studies of supernovae at high redshift.  A series of papers details recent work  to explore the theoretical and observational obstacles to precise cosmological measurements with cluster surveys.  Holder, Haiman \\& Mohr \\cite{holder01b} introduced the Fisher matrix formalism and showed that high yield SZE cluster surveys can provide precise constraints on the geometry of the universe through simultaneous measurements of $\\Omega_E$ and $\\Omega_M$.  Weller, Battye \\& Kniessl \\cite{weller01} demonstrated that future SZE surveys might constrain the variation of the dark energy equation of state $w(z)$.  Hu \\& Kravtsov \\cite{hu02} examined the effects of cosmic variance on cluster surveys as well as including the effects of imprecise knowledge of a larger number of cosmological parameters. Levine, Schulz \\& White  \\cite{levine02} examined an X--ray cluster survey, showing that a sufficiently large survey allows one to measure cosmological parameters and constrain the all--important cluster mass--observable relation simultaneously.  This raises the exciting possibility that cluster surveys are self--calibrating-- that the redshift distribution contains enough energy to solve for cluster structure and cosmology simultaneously!  An important caveat to this work is that the authors assumed that the evolution of cluster structure was perfectly known.  In a more recent work, Majumdar \\& Mohr \\cite{majumdar02} show that if one allows for our imprecise knowledge cluster structural evolution with redshift, the constraint on the dark energy equation of state $w$ evaporates.  In addition, we show that if one incorporates followup measurements-- perhaps from X--ray, SZE or weak lensing-- into the survey one can recover the precise constraints on $w$.  These calculations underscore the importance of incorporating information from multiple observables into future cluster surveys.  Ongoing calculations by several groups will undoubtedly provide additional insights into how to optimize cluster surveys, but today it appears that high yield cluster surveys are as viable a means of studying dark energy as first suggested. ",
        "conclusions": ""
    },
    "0208/astro-ph0208272_arXiv.txt": {
        "abstract": "{ We extend the Kelvin-Helmholtz instability to an expanding background. We study the evolution of a non-viscous irrotational fluid and find that for wavelengths much smaller than the Hubble scale small perturbations of the fluid are unstable for wavenumbers larger than a critical value. We then apply this result in the early universe, treating cold dark matter as a classical fluid with vanishing background pressure. ",
        "introduction": "In the standard cosmological model quantum fluctuations are magnified from microscopic to cosmological scales during a period of exponential expansion (inflation) and source curvature perturbations with a scale-invariant or nearly scale-invariant spectrum (\\cite{LLBook1,LLBook2}). After inflation these perturbations act as seeds for the large scale structure of the universe: first, beginning during radiation domination, non-baryonic dark matter ``falls'' into the potential wells formed during inflation and later, during matter domination but after decoupling, radiation and  baryonic matter follow.  In this picture radiation and baryonic matter can ``fall'' into the potential wells only after decoupling since before that their Jeans length is larger than the Hubble radius. After decoupling perturbations whose wavelength is larger than the Jeans-length grow, but on scales smaller than the Jeans-length they oscillate, leaving imprints in the CMB that can be observed today. On very small scales structure is wiped out through the tight coupling of the baryons to the photons until decoupling (\\cite{Silk}). Unlike in baryonic matter and radiation, perturbations in the non-baryonic dark matter can start growing as soon as their wavelength is smaller than the horizon, since the Jeans scale is negligible for non-interacting dark matter. The  non-baryonic dark matter can therefore cluster and ``fall'' into the potential wells much earlier than the other matter.   In fluid dynamics the study of instabilities (\\cite{chandra}) is well established and has also been extended to the astrophysical context (see \\cite{ino}, and references therein). The Kelvin-Helmholtz instability has been known for nearly 150 years (\\cite{chandra}). In its simplest form two large fluid volumes, separated by an interface, move in opposite directions. Neglecting the stabilizing effects of density and gravity gradients this setup is unstable for all non-zero velocities and wavenumbers.   We study the initial stages in the evolution of small perturbations in an irrotational, non-viscous fluid due to a Kelvin-Helmholtz instability in an expanding universe. We show that even in such a background these instabilities occur. We then apply these results in the early universe. Two large colliding regions of cold dark matter might serve as an example: the Kelvin-Helmholtz instability can arise at the contact surfaces of the two regions. We model the dark matter as a fluid with vanishing background pressure in concordance with the standard cold dark matter models (\\cite{LLBook1,LLBook2}). Depending on when this happens in the evolution of the background the perturbations arising from the instability grow either exponentially or linearly on scales smaller than a critical wavenumber which we determine. On the basis of these results we argue that it will be interesting to test these perturbations in simulations because they could have relevance to the dark matter cusp and halo problems and to anisotropies in the CMB.    In the next section we give the governing equations for the fluid. In Sec.~\\ref{modelsec} we specify the model geometry and the general setup. After giving the background solution in Sec.~\\ref{backsec}, we derive the perturbed equations in Sec.~\\ref{pertsec} and analyse the stability of the system. In Sec.~\\ref{critwavesec} we then calculate the critical wavenumbers for which instability can occur. We apply our results to standard cold dark matter in the final section and discuss possible implications. ",
        "conclusions": "\\label{disksec}  In standard fluid dynamics the Kelvin-Helmholtz instability in the incompressible case develops for all wavenumbers $k>0$. However in an expanding universe we have shown that this is no longer the case and that there are critical wave numbers which separate stable and unstable domains. These wave numbers depend on the expansion of the background and the  background velocity.  For wave numbers larger than the critical wavenumber, $k>k_{\\rm{crit}}$, small perturbations will grow.  More specifically, for a scale factor exponent $\\gamma > 1/2$ perturbations will grow for a wavenumber-dependent time. In the case $\\gamma=1/2$, i.e.~during radiation domination, perturbations grow linearly for all times and for $\\gamma <1/2$ they grow exponentially for all times.  As an illustration we look at the case of cold dark matter at the beginning of matter domination, $\\gamma=2/3$. The physical setting could be two large volumes of cold dark matter colliding or some non-homogeneous velocity flows in the early universe.  In the first case the contact surface would be prone to the Kelvin-Helmholtz instability and in the second the instability would occur inside the flow. From Eq.~(\\ref{defA}) and (\\ref{k1estim}) the physical critical wavenumber at the time of matter and radiation equality is given by \\be \\frac{k_{\\rm {crit}}}{a_{\\rm{eq}}} =\\frac{2\\gamma}{\\bar U t_0}\\frac{a_0}{a_{\\rm{eq}}} \\,, \\ee where $a_{\\rm{eq}}$ is the scale factor at equality. We choose $t_0$ to be the present age of the universe, $t_0=H_0^{-1}$, where $H_0=100kms^{-1}Mpc^{-1} h$, ${a_0}/{a_{\\rm{eq}}}=24000\\Omega_0h^2$ (see (\\cite{LLBook1,LLBook2})), with $\\Omega_0\\sim 1$ and the characteristic velocity is $\\bar U\\sim 100kms^{-1}$  (see \\cite{anne,klypin}). With these numbers the critical scale below which perturbations are amplified is \\be \\lambda_{\\rm{crit}} \\sim 1 kpc \\,. \\ee  Although detailed numerical analysis will be necessary in order to make definite quantitative predictions of the effect of the instability on the density distribution, we can nevertheless speculate that it will lead to a redistribution of power from scales $\\sim \\lambda_{crit}$ to much smaller scales.  A significant feature of this example is that the instability occurs at such a small scale. This is important because it is well known that the cold dark matter model has problems at small scales. It seems to predict higher densities and more small scale structure than is observed. Modifications of the CDM have been tried, e.g. Warm Dark Matter and Self-Interacting Dark Matter but neither of these solves all the problems, see e.g.~(\\cite{primack,tasitsiomi}) for an outline of the ideas and detailed references. While these discrepancies may be, at least partially, due to numerical and observational resolution and selection effects or the consequences of baryonic physics, there still appears to be a residual problem.   The surprising element of the Kelvin-Helmholtz effect is that it occurs even on expanding backgrounds and depending on the value of $\\gamma$ the instabilities can grow exponentially.  Our results indicate that further analysis using numerical methods is needed to explore the effect into the nonlinear regime."
    },
    "0208/astro-ph0208044_arXiv.txt": {
        "abstract": "We present results from a short series of blue, moderate resolution spectra of the microquasar binary, SS~433. The observations were made at a time optimized to find the spectrum of the donor star, i.e., when the donor was in the foreground and well above the plane of the obscuring disk. In addition to the well-known stationary and jet emission lines, we find evidence of a weak absorption spectrum that resembles that of an A-type evolved star. These lines display radial velocity shifts opposite to those associated with the disk surrounding the compact star, and they appear strongest when the disk is maximally eclipsed. All these properties suggest that these absorption lines form in the atmosphere of the hitherto unseen mass donor star in SS~433.   The radial velocity shifts observed are consistent with a mass ratio $M_X / M_O = 0.57 \\pm 0.11$ and masses of $M_O = (19\\pm 7)~M_\\odot$ and $M_X = (11 \\pm 5)~M_\\odot$.  These results indicate that the system consists of an evolved, massive donor and a black hole mass gainer. ",
        "introduction": "%  SS~433 is still one of the most mysterious of the X-ray binaries even after some 25 years of observation \\citep{mar84,zwi89,gie02}. We know that the mass donor feeds an enlarged accretion disk surrounding a neutron star or black hole companion, and a small portion of this inflow is ejected into relativistic jets that are observed in optical and X-ray emission lines and in high resolution radio maps.  There are two basic timescales that control the spectral appearance and system dynamics, a 162~d disk and jet precessional cycle and a 13~d orbital period. The mass function derived from the \\ion{He}{2} $\\lambda 4686$ emission line indicates that the donor star mass is in excess of $8~M_\\odot$ \\citep{fab90}, and the donor is probably a Roche-filling, evolved star \\citep{kin00}. However, the spectral signature of this star has eluded detection. This is probably because the binary is embedded in an expanding thick disk that is fed by the wind from the super-Eddington accretion disk \\citep{zwi91}.  The outer regions of this equatorial thick disk have been detected in high resolution radio measurements by \\citet{par99} and \\citet{blu01}.  We recently showed how many of the properties of the ``stationary'' emission lines can be explained in terms of a disk wind \\citep{gie02}.  The task of finding the spectrum of the donor is crucial because without a measurement of its orbital motion, the mass of the relativistic star is unknown.   The best opportunity to observe the flux from the donor occurs at the precessional phase when the disk normal is closest to our line of sight and the donor star appears well above the disk plane near the donor inferior conjunction orbital phase \\citep{gie02}. This configuration occurs only a few nights each year for ground-based observers.   The choice of spectral region is also important.  \\citet{gor98b} found that the regular eclipse and precessional variations seen clearly in the blue are lost in the red due to an erratically variable flux component. Thus, we need to search for the donor spectrum blueward of the $R$-band in order to avoid this variable component. On the other hand, the color variations observed during eclipses suggest that the donor is cooler than the central portions of the disk \\citep{ant87,gor97}. Thus, the disk will tend to contribute a greater fraction of the total flux at lower wavelengths.   The best compromise is in the blue where there are a number of strong absorption lines in B- and later-type stars.  Here we present the results of a blue spectral search for the donor's spectrum made during an optimal disk and orbital configuration in 2002 June (\\S2).  We first discuss the dominant emission features formed in the jets and disk wind (\\S3). We then focus on a much weaker set of absorption lines (\\S4), and we present arguments linking these to the photosphere of the donor star. ",
        "conclusions": ""
    },
    "0208/hep-ph0208092_arXiv.txt": {
        "abstract": "s{ \\vskip -2.5in \\rightline{hep-ph/0208092} \\rightline{UMN--TH--2108/02} \\rightline{TPI--MINN--02/25} \\rightline{August 2002} \\vskip 1.9in The supersymmetric extension to the Standard Model offers a promising cold dark matter candidate, the lightest neutralino. I will review the prospects for the detection of this candidate in both accelerator and direct detection searches.} ",
        "introduction": "Although there are many reasons for considering supersymmetry as a candidate extension to the standard model of strong, weak and electromagnetic interactions\\cite{reviews}, one of the most compelling is its role in understanding the hierarchy problem\\cite{hierarchy} namely, why/how is $m_W \\ll M_P$.  One might think naively that it would be sufficient to set $m_W \\ll M_P$ by hand. However, radiative corrections tend to destroy this hierarchy. For example, one-loop diagrams generate \\beq \\delta m^2_W = \\mathcal {O}\\left({\\alpha\\over\\pi}\\right)~\\Lambda^2 \\gg m^2_W \\label{four} \\eeq where $\\Lambda$ is a cut-off representing the appearance of new physics, and the inequality in (\\ref{four}) applies if $\\Lambda\\sim 10^3$ TeV, and even more so if  $\\Lambda \\sim m_{GUT} \\sim 10^{16}$ GeV or $ \\sim M_P \\sim 10^{19}$ GeV. If the radiative corrections to a physical quantity are much larger than its measured values, obtaining the latter requires strong cancellations, which in general require fine tuning of the bare input parameters. However, the necessary cancellations are natural in supersymmetry, where one has equal numbers of bosons $B$ and fermions $F$ with equal couplings, so that (\\ref{four}) is replaced by \\beq \\delta m^2_W = \\mathcal {O}\\left({\\alpha\\over\\pi}\\right)~\\vert m^2_B - m^2_F\\vert~. \\label{five} \\eeq The residual radiative correction is naturally small if $ \\vert m^2_B - m^2_F\\vert \\la 1~{\\rm TeV}^2 $.  In order to justify the absence of interactions which can be responsible for extremely rapid proton decay, it is common in the minimal supersymmetric standard model (MSSM) to assume the conservation of R-parity.  If R-parity, which distinguishes between ``normal\" matter and the  supersymmetric partners and can be defined in terms of baryon, lepton and spin as $R = (-1)^{3B + L + 2S}$, is unbroken, there is at least one supersymmetric particle (the lightest supersymmetric particle or LSP) which must be stable.  Thus, the minimal model contains the fewest number of new particles and interactions necessary to make a consistent theory.  There are very strong constraints, however, forbidding the existence of stable or long lived particles which are not color and electrically neutral~\\cite{EHNOS}. Strong and electromagnetically interacting LSPs would become bound with normal matter forming anomalously heavy isotopes. Indeed, there are very strong upper limits on the abundances, relative to hydrogen, of nuclear isotopes\\cite{isotopes}, $n/n_H \\la 10^{-15}~~{\\rm to}~~10^{-29} $ for 1 GeV $\\la m \\la$ 1 TeV. A strongly interacting stable relic is expected to have an abundance $n/n_H \\la 10^{-10}$ with a higher abundance for charged particles.   There are relatively few supersymmetric candidates which are not colored and are electrically neutral.  The sneutrino\\cite{snu} is one possibility, but in the MSSM, it has been excluded as a dark matter candidate by direct\\cite{dir} and indirect\\cite{indir} searches.  In fact, one can set an accelerator based limit on the sneutrino mass from neutrino counting, $m_{\\tilde\\nu}\\ga$ 44.7 GeV \\cite{EFOS}. In this case, the direct relic searches in underground low-background experiments require $m_{\\tilde\\nu}\\ga$ 20 TeV~\\cite{dir}. Another possibility is the gravitino which is probably the most difficult to exclude. I will concentrate on the remaining possibility in the MSSM, namely the neutralinos. ",
        "conclusions": ""
    },
    "0208/astro-ph0208208_arXiv.txt": {
        "abstract": "\\chandra\\ has significantly advanced our knowledge of the processes in the intracluster gas.  The discovery of remarkably regular ``cold fronts'', or contact discontinuities, in merging clusters showed that gas dynamic instabilities at the boundaries between the moving gases are often suppressed, most likely by specially structured magnetic fields.  Cold fronts are not limited to mergers --- \\chandra\\ observed them in the cores of more than 2/3 of the cooling flow clusters, where they often divide the cool central and the hotter ambient gas phases.  The natural state of the low-entropy central gas in clusters thus appears to be to slosh subsonically in the central potential well, with several interesting implications.  Hydrostatic equilibrium is not reached and the hydrostatic total mass estimates cannot be valid at small radii ($r\\lax 100$ kpc).  The kinetic energy dissipated by the sloshing gas may be sufficient to compensate for radiative cooling.  In the absence of a recent merger, the cause of such sloshing is unclear.  It might be related to the mechanical energy generated by the central AGN, evidence of which (the bubbles) is also observed by \\chandra\\ in a large fraction of the cooling flow clusters.  Heat conduction across cold fronts is suppressed completely.  In the bulk of the gas, it is reduced by a factor of $\\sim 10$ or more relative to the classic value, as shown by merger temperature maps.  A spatial correlation between the gas temperature maps and the radio halo maps is observed in several mergers, which may provide clues for the relativistic particle acceleration mechanisms. ",
        "introduction": "\\chandra\\ X-ray Observatory's salient feature is its 1\\as\\ resolution.  The difference it makes for the galaxy cluster studies is illustrated in Fig.\\ \\ref{fig:2a0335}, which shows images of one of the most relaxed nearby clusters, 2A0335+096, made by \\rosat\\ PSPC and \\chandra.  On large linear scales accessible to \\rosat, the cluster is symmetric and, apparently, in hydrostatic equilibrium.  However, in the central 100 kpc, \\chandra\\ reveals a very dynamic gas core.  In this article, we will review the published results and report some new findings on dynamic phenomena in the cluster gas.  Some other interesting \\chandra\\ work, such as the accurate total mass profiles and studies of the AGN interaction with the cluster gas, is presented elsewhere in these proceedings.  We use \\hfifty.  \\begin{figure} \\plotone{taiwproc/taiwproc_2a0335_gray.ps} \\caption{{\\em Left}: \\rosat\\ PSPC image of the relaxed cluster 2A0335+096 ($z$=0.03, $T$=3 keV). {\\em Right}: \\chandra\\ image reveals complex and dynamic structure of its core.} \\label{fig:2a0335} \\end{figure} ",
        "conclusions": "Shock fronts and cold fronts discovered by \\chandra\\ provide unique tools to study the physics of the ICM. In particular,  1. In merging clusters, we can estimate the velocity of the infalling subclusters.  The cold fronts are too sharp and regular for their velocity, indicating that gas dynamic instabilities at the boundaries between the gas phases are often suppressed, most probably by specially structured magnetic fields.  2. Cold fronts are also observed in the cores of most ``relaxed'' clusters, showing that the dense, cool central gas in such clusters is constantly sloshing in the central potential dip. This has several interesting implications: (a) the X-ray hydrostatic method should systematically underestimate the total cluster mass inside $r\\lax 100$ kpc; (b) the dissipated kinetic energy of the sloshing gas is sufficient to compensate for radiative cooling and thus disrupt cooling flows.  Typically, such sloshing involves only a small fraction of the total gas mass, so these findings cannot be generalized to the whole cluster.  3. Thermal conduction appears to be suppressed, compared to the Spitzer value: (a) across cold fronts, completely, which follows from the sharpness of the gas density jumps; (b) in the bulk of the gas, by a factor $\\sim 10$, from the existence of small-scale temperature variations in mergers.  Also, we find that relativistic electrons appear to concentrate in the recently shock-heated gas, even though in most clusters, the merger Mach numbers are apparently insufficient to accelerate electrons to the radio halo energies."
    },
    "0208/astro-ph0208514_arXiv.txt": {
        "abstract": "{We present the first result of a comprehensive spectroscopic study of quasar host galaxies.  On-axis, spatially resolved spectra of low redshift  quasars have  been obtained with  FORS1, mounted  on the 8.2m ESO  Very Large Telescope,  Antu. The spectra are  {\\it spatially deconvolved} using a spectroscopic  version of the ``MCS deconvolution algorithm''. The algorithm decomposes two dimensional spectra into the individual spectra of  the central point-like nucleus and  of its host galaxy.  Applied  to \\obj\\,  at  $z=0.135$ (M$_B$=$-$23.0), it  provides us  with  the spectrum  of   the  host   galaxy  between  3600\\AA\\,   and  8500\\AA\\, (rest-frame), at a  mean resolving power of 700. The  data allow us to measure several of the important Lick indices. The stellar populations and gas ionization state of the host galaxy of \\obj\\, are very similar to  the   ones  measured  for  normal   non-AGN  galaxies.   Dynamical information is  also available for  the gas and stellar  components of the  galaxy.   Using   deconvolution  and  a  deprojection  algorithm, velocity curves are derived for  emission lines, from the center up to 4\\arcsec\\,  away   from  the nucleus of the galaxy. Fitting  a  simple three-components mass model (point  mass, spherical halo of dark matter, disk) to the position-velocity  diagram, we infer  a mass  of M(r$<$1kpc) = $(2.0 \\pm 0.3) 10^{10}$ M$_{\\odot}$ within the central kiloparsec of the galaxy, and a mass  integrated  over  10  kpc  of  M(r$<$10kpc) = $ (1.9  \\pm 0.3)  10^{11}$ M$_{\\odot}$, with an additional 10\\% error due to the uncertainty on the inclination of the galaxy.  This, in combination  with the  analysis of the stellar populations  indicates that  the host  galaxy  of \\obj\\,  is a  normal spiral galaxy. ",
        "introduction": "Luminous quasars are now known to be generally located at the cores of large  galaxies.  With  the recent  discovery that  supermassive black holes appear  to reside in most  if not all  galaxies with substantial bulges (e.g., Magorrian et al.  \\cite{Mago98}), it is now quite likely that  quasar-like  activity  is  an  extremely  common  but  transient phenomenon,  linking the  growth of  now  extinct black  holes to  the violent processes during phases  of active accretion (cf.  McLeod et al. \\cite{leod99}).   However, very  little of the  physical processes operating  during  such  episodes  of  nuclear  activity  is  actually understood: neither do  we know the actual conditions  for fueling (or refueling)  massive  black holes,  nor  the  time-scales involved.   A particularly  interesting problem  is the  importance of  the feedback from  a high-luminosity  Active Galactic  Nucleus (AGN)  onto  its host galaxy, in the  form of huge quantities of  ionizing radiation as well as possible mechanical outflows (jets).  Quasar host  galaxies have been studied almost  exclusively by imaging (e.g.,   Bahcall   et   al.    \\cite{bahcall97},   Stockton   et   al. \\cite{Stockton98}, M\\'arquez  et al \\cite{Marquez01}).   Recent Hubble Space Telescope optical  studies have established that high-luminosity quasars  generally reside  in big  ellipticals, irrespective  of radio properties   (McLeod   \\&   Rieke   \\cite{leod95},   Disney   et   al. \\cite{disney95}, Hughes et al.  \\cite{hugh00}).  There also seems to be a trend that more luminous QSOs are hosted by more massive galaxies (McLure et al.  \\cite{Lure99}, McLeod \\& Rieke \\cite{leod95}).  A  more detailed  understanding of  the physical  conditions  in host galaxies     can    only     be     obtained    from     spectroscopic observations. Available  evidence in  this field is  extremely scarce, and although extensive quasar  host spectroscopy was conducted already in the early 1980s (e.g.,  Boroson et al. \\cite{boroson85}), they have never  really  been  followed  up with  improved  instrumentation  and analysis  techniques, except for  a few  isolated objects  (among them 3C~48  being the  probably best-studied  case in  this field  -- e.g., Chatzichristou   et   al.     \\cite{cha99},   Canalizo   \\&   Stockton \\cite{canal2000}). Particularly important is  the fact that nearly all these  observations  up  to   now  were  designed  as  ``off-nuclear'' spectroscopy, avoiding  the strong contamination from  the AGN itself, but  yielding only  information  on  the host  for  distances of  $\\ga 5-10$\\,kpc from the nucleus.  \\begin{figure*} \\centering \\includegraphics[width=8.7cm]{fig03b.ps} \\includegraphics[width=8.7cm]{fig03c.ps} \\includegraphics[width=8.7cm]{fig03d.ps} \\caption{One dimensional  deconvolved spectrum of \\obj\\,  and its host galaxy. Each  panel corresponds  to one FORS1  grism and  displays the individual flux calibrated spectrum of  the quasar and its host alone. Note that the  host spectrum shows no trace of  the AGN broad emission lines, that  are seen narrow  in the host  galaxy. Note also  the very good agreement in the deconvolution of the 3 grisms in the overlapping regions.} \\label{onedspec} \\end{figure*}  With the  aim of studying the  stellar and gas content  of quasar host galaxies, as well as their dynamics, we have initiated a spectroscopic campaign with the ESO-VLT to  observe a sample of luminous radio-quiet quasars  at low  redshift. The  sample  is based  on the  Hamburg/ESO Survey (HES; Wisotzki et  al.\\ \\cite{wisotzki00}), a wide-angle ($\\sim 7500$~deg$^2$) search  for optically bright ($B \\la  17.5$) quasars in the  southern sky.   This  survey is  particularly  suited to  provide samples for  host galaxy studies,  mainly because of two  reasons: (1) because  of   its  bright   limiting  magnitude,  the   survey  yields low-redshift  quasars in  large numbers,  (2) the  selection  does not encompass  morphological criteria,  i.e.,  there is  no limitation  to point sources as  in most other optical quasar  surveys. This fact and the wide range of selection criteria employed in the HES (cf. Wisotzki et  al.  2000)  ensure  that  quasars  are  selected  irrespective  of morphological properties of their  host galaxies (see also K\\\"ohler et al.\\ \\cite{koehler97}). Our objects all  have $M_B < -23$ and  $z<0.33$.    The present paper  is aimed at exposing the  techniques used to obtain high  quality  spectra  of   quasar  hosts,  decontaminated  from  the contamination  by the  quasar, and  to show  their application  to one object, taken as a test case: \\obj, at $z=0.135$. ",
        "conclusions": "We have undertaken a VLT  program aimed at unveiling the spectroscopic properties  of quasar  host galaxies  for a  sample  of radio quiet, bright quasars at low  redshift. Our long term scientific goal is double: (1) to compare the  stellar populations of quasar hosts  with those of other (``normal'') galaxies, and (2) to study their dynamics, as close as possible to the central AGN.  The  present paper considers \\obj\\, as a test case. An  important  part  of our  work  is  dedicated to  the correction  of  geometrical effects  and  removal  of the  atmospheric blurring. This  is crucial in the  central parts of  the galaxy, with sizes comparable to the seeing disk.  We  find that  the stellar  population of  the host  galaxy  of \\obj\\, compares  well with  that of  the  normal non-AGN  spiral galaxies  of Kennicutt (1992a,b) and Trager  et al. (\\cite{trager98}). They show no trace of  enhanced star formation.   The interstellar medium  does not show significant ionization by the central AGN. Its mass M(r $<$ 10 kpc) = $(1.9 \\pm 0.3) 10^{11}$ M$_{\\odot}$ also compares very well with that of  other spiral galaxies.  To summarize, the host galaxy of \\obj\\, is a normal spiral galaxy."
    },
    "0208/astro-ph0208387_arXiv.txt": {
        "abstract": "Recently \\citet{giz02} suggested that supergranulation has a wave-like component. In this paper I show that the same phenomenon can be observed using surface Doppler shift data, thereby confirming their observations. I am also able to measure the dispersion relation to lower wavenumbers and to extend the results for rotation and meridional flows beyond $\\pm 70^\\circ$ latitude. ",
        "introduction": "Solar supergranulation, which was first described more than 40 years ago by \\citet{lei}, has remained difficult to explain. While it appears to be similar to the granulation, several observations, such as the anomalous flows obtained using correlation tracking or similar methods (e.g. \\citet{beck00}, in the following BS), challenge the interpretation as a simple convective phenomenon. An alternative explanation is that the supergranulation has a wave-like component and observational evidence for this was recently reported by \\citet{giz02} (GDS in the following).  It has been the cause of some concern that the wave like character has not been observed previously, given the many measurements of the rotation rate of the supergranulation and attempts to determine why it appears to be anomalous. A likely explanation is that it was obscured by the limitations of the surface Doppler shift data.  Supergranular motions are predominantly horizontal near the photosphere. Let $V_\\phi$ be the velocity in the longitudinal ($\\phi$) direction and $V_\\theta$ the velocity in the latitude ($\\theta$) direction. The observed Doppler velocity $V$ is then given by \\begin{equation} V=V_\\phi\\sin\\phi + V_\\theta\\cos\\phi\\sin\\theta \\end{equation} under the assumption that the observations are made in the equatorial plane (i.e. that B0 = 0) at a large distance from the Sun. For the $V_\\phi$ component, in particular, the projection leads to a large modulation of the signal as the Sun rotates and to a significant spreading of the power in the frequency domain. ",
        "conclusions": "Possibly the main result is to confirm the results of GDS. Given the differing analysis techniques this represents a significant confirmation and lays to rest the concerns based on the fact that the phenomenon had not been observed previously in surface Doppler shift data.  While not physically motivated it is interesting that the frequencies appear to follow a square root quite well. As can be seen from Figure \\ref{disp} the dispersion relation at $40^\\circ$ is quite similar to that at the equator in both directions.  The small variation of the rotation rate with $l$ is consistent with a simple depth independent flow. This is in stark contrast to the numbers derived from the same data in BS, which varied by 10nHz between $l$=50 and $l$=200, and which were difficult to explain. The rotation rate derived here is remarkably close to the magnetic tracer rate of \\citet{komm93}. The similarity may be related to the fact that the advection of the small magnetic elements appear to be influenced by the supergranulation. This similarity is quite unlike the results in BS, in which the rotation rate, at least at low $l$, exceeded that of the plasma at any depth. This lends further credence to the idea that the flows inferred here represent physical advection rates.  The meridional flow results appear to be reliable to at least $\\pm70^\\circ$ latitude with a clear turnover between $20^\\circ$ and $30^\\circ$ and no sign of a second cell. Note that the meridional flow is not perfectly antisymmetric across the equator. As shown this is largely explained by a P angle error caused by the misalignment of the MDI instrument on the SOHO spacecraft \\citep{toner} and by the difference between the Carrington elements and the true rotation axis of the Sun \\citep{giles}. An error in the rotation axis causes part of the rotation velocity to be misidentified as meridional flow.  Figure \\ref{power} shows that the power peaks around $l=100$, as previously seen. The linewidth appears to be an increasing function of $l$, possibly indicating that smaller scale features damp more quickly. The increase for $l \\le 50$ is likely an artifact of the difficulty in fitting low degree features.  As in GDS Figure \\ref{anis} shows that the power is highly anisotropic at all latitudes and that the anisotropy is in the prograde direction and increasingly towards the equator at higher latitudes.  The physical nature of these waves, if that is indeed what they are, is still unknown. However, with the results presented here, it may be possible to further constrain the models.  Further progress may be possible by extending these results over time or by probing the depth structure using time-distance helioseismology."
    },
    "0208/astro-ph0208452_arXiv.txt": {
        "abstract": "We investigated the orbital evolution of satellite galaxies using numerical simulations.  It has been long believed that the orbit suffers circularization due to the dynamical friction from the galactic halo during orbital decay. This circularization was confirmed by numerous simulations where dynamical friction is added as external force. However, some of the resent $N$-body simulations demonstrated that circularization is much slower than expected from approximate calculations. We found that the dominant reason for this discrepancy is the assumption that Coulomb logarithm $\\log \\Lambda$ is constant, which has been used in practically all recent calculations. Since the size of the satellite is relatively large, accurate determination of the outer cutoff radius is crucial to obtain good estimate for the dynamical friction. An excellent agreement between $N$-body simulations and approximate calculations was observed when the outer cutoff radius is taken to be the distance of the satellite to the center of the galaxy. When satellite is at the perigalacticon, the distance to the center is smaller and therefore $\\log \\Lambda$ becomes smaller. As a result, the dynamical friction becomes less effective.  We apply our result to the Large Magellanic Cloud. We found that the expected lifetime of the LMC is twice as long as that would be predicted with previous calculations.  Previous study predicts that the LMC will merge into the Milky Way after 7 G years, while we found that the merging will take place after 14 G years from now.  Our result suggests that generally satellites formed around a galaxy have longer lifetime than previous estimates. ",
        "introduction": "Recent observations have revealed that there are many satellite galaxies around the Milky Way. In the hierarchical clustering scenario, it is expected many of such dwarf satellites are formed. In fact, one of the most serious problems with the present hierarchical clustering scenario is that it predict too many satellite galaxies, about a factor of 10 more than the number observed in the Local group ({\\it e.g.}, Moore \\etal, 1999). A number of explanations, including exotic theories which relies on hot or self-interacting dark matter, have been proposed.  In this paper, we go back to the basic problem: how long are the satellites lives ?  In other words, how do the orbits of satellites evolve through interaction with the gravitational field of its parent galaxy? The dominant driving force of the evolution is the dynamical friction. For satellites like the LMC-SMC pair and the Sagittarius dwarf, there are many detailed studies of their orbital evolution, in which the dynamical friction is included as the external force operating on the center-of-mass motion of the satellite.  Well known works include Murai and Fujimoto (MCs, 1980) and Ibata and Lewis (Sagittarius, 1998). In both of these studies, and in all other studies where the dynamical friction formula is used, significant circularization of the orbit of the satellite was observed. This circularization is the natural result of the fact that the dynamical friction is proportional to the local density of the background stars, and therefore the strongest at the perigalacticon.  However, recent $N$-body simulations of the orbital evolution of satellites resulted in rather counter-intuitive result. Van den Bosch et al (1999, hereafter BLLS) performed the $N$-body simulation of the satellite, where the parent galaxy was modeled directly as self-consistent $N$-body system. The satellite is modeled as one massive particle with spline potential softening used in PKDGRAV (Dikaiakos \\& Stadel, 1996).  They investigated the evolution of the orbit for wide variety of model parameters such as the mass of the satellite and initial orbital eccentricity. They observed practically no circularization in any of their simulations.  Jiang and Binney (2000, hereafter JB) performed fully self-consistent simulation of the satellite, where both the parent galaxy and the satellite are modeled as self-consistent $N$-body systems. They compared their result with the result of approximate model in which the usual dynamical friction formula is used. Though they argued that the agreement is good, from their figure 3 it is clear that approximate models suffer stronger circularization and evolve faster than their $N$-body counterpart.  Neither of above two papers discussed the reason of this rather serious discrepancy between the result of $N$-body simulations and previous analytic prediction.  The purpose of this paper is to understand its cause. In section 2, we describe our model experiment designed to reproduce the discrepancy observed by BLLS and JB. In section 3 we show our result. Our result is consistent with both of the previous works. $N$-body simulation showed only marginal circularization but approximate calculation using dynamical friction formula showed strong circularization. In section 4, we investigate the reason. There are several possible candidates for the reason. We consider a few of them, and found that a simple modification of the conventional form of the dynamical friction formula results in a quite remarkable improvement of the agreement between $N$-body and approximate calculations. In section 5 we apply our formalism to the LMC. We found that the orbital evolution becomes significantly slower than prediction by previous calculations using conventional formula. For example, the lifetime of LMC was 7 Gyr with conventional formula, but is 14 Gyr with our formalism. We also discuss the implication of our result to the so-called ``dwarf problem''. ",
        "conclusions": "We performed $N$-body simulations of satellite orbits.  We found that the circularization of the orbit due to the dynamical friction is much slower than commonly believed.  This discrepancy was also reported by BLLS, and we can see the same tendency from the numerical result reported by JB.  Previous studies of satellite orbits used the outer cutoff radius of the dark halo as $b_{max}$.   We found that the effective $b_{max}$ should be of the order of $R_s$, the distance of the satellite from the center of the galaxy, which varies as the satellite orbits around the galaxy. Our formula results in a greatly improved agreement with the $N$-body result.  \\subsection{the Large Magellanic Cloud}  \\begin{figure} \\plotone{f3.eps} \\caption{Radial Evolution of LMC. From -10 G years to 10 G years.} \\label{fig:3} \\end{figure}  The Large Magellanic Cloud is the most famous satellite of Milky Way. Its orbit has been investigated from both observation and numerical simulations ({\\it e.g.}, Toomre, 1970; Tremaine, 1976; Lin and Lynden-Bell, 1977; Murai and Fujimoto, 1980).  The importance of the effect of dynamical friction from the galactic halo on the orbit evolution LMC is first emphasized by Tremaine (1976).  By using numerical simulation, \\citet{MF} (hereafter MF) determined the orbital elements and the present phase of the LMC.  They performed a number of backward numerical integrations of the orbits of the LMC and SMC from various initial conditions, and integrated orbits of test particles in the LMC and SMC for each condition.  Comparing the result of distribution of test particles and the observed Magellanic stream, they chose the initial condition which gives the best fit.  In their numerical integration, they assumed a halo expressed by a singular isothermal sphere, which is a simple flat-rotation halo.  In their paper, it is not clear either what assumption or what exact value is adopted for $\\ln \\Lambda$, since there is no discussion on how they determined $\\ln \\Lambda$ though it appeared in their equation (13).  In order to see the effect of changing $\\ln \\Lambda$, we integrated the orbit of LMC both forward and backward in time, using both the constant $\\Lambda$ and variable $\\Lambda$ ($b_{max} = R_s$). In this study, we express LMC as a single Plummer-softened particle with mass $2 \\times 10^{10}\\smass$ and softening length $5$ kpc.  The rotation velocity of the halo is $250$ km/s, same as what is used by MF. We simulated the orbit of the LMC only, since our purpose here is to demonstrate the effect of $\\Lambda$ and not the accurate determination of the orbits of the Clouds.  The solid curve in Figure \\ref{fig:3} corresponds to the orbit obtained when the dynamical friction is calculated using equation (\\ref{eqn:vlambda}). The dashed curve Figure \\ref{fig:3} correspond to the orbit obtained using the formula (\\ref{eqn:dff1}) and (\\ref{eqn:clambda}).  Note that the backward part of this dashed curve is in very good agreement with the result of MF. This agreement strongly suggests that what MF used is indeed a constant $\\Lambda$.   Figure \\ref{fig:3} shows that real evolution of the orbit of LMC (with variable $\\Lambda$) is significantly smaller than what is obtained by MF. 10 Gyrs ago, the ``true'' apogalacticon was only $160$ kpc, while the solution by MF was $180$ kpc.  A more remarkable difference is in the future of the LMC. With the constant $\\Lambda$. The LMC  will fall to the galactic center in only 7 Gyrs with constant $\\Lambda$, while our result suggests that it will take more than 14 G years for the  LMC to fall to the galactic center.   \\subsection{Statistical Evolution of Faint Galaxies}  In semi-anaritic studies of galaxy formation, it has been assumed that the orbits of satellite galaxies evolve through dynamical friction following Chandrasekhar's formula with constant $\\Lambda$.  In this section, we discuss how our result might change our understanding of the statistical evolution of the satellite galaxies.  Our study shows that the time evolution of the eccentricity of satellites is rather small. Thus, we may assume that the distribution of eccentricities of satellite galaxies at present directly reflects that at the formation epoch of the Galaxy. Therefore the distribution of eccentricities of satellites galaxies can be an important clue to the formation of the Galaxy.  The lifetime of the satellite is estimated using the dynamical friction timescale with $\\ln \\Lambda$ taken to be $M_H/M_s$ \\citep{Lac93,Kau94}.  This would cause a quite serious overestimate in the dynamical friction timescale, since the factor one should use is the ratio between the size of the halo and the size of the satellite. If we assume $M\\propto \\sigma^4$, we have $R \\propto M^{1/2}$. Thus, there is at least a factor of two difference in the value of $\\ln \\Lambda$.  Since there are too many other uncertainties in the semi-analytic modeling of the galaxy number evolution, how serious this difference is not clear. However, it certainly affects the estimate of presently observed satellites rather strongly. A more detailed study on this aspect is clearly necessary.  We thank Toshi Fukushige and Sadanori Okamura on stimulating discussions. We also thank Rainer Spurzem, T, Tsuchiya, and Andrea Just for helpful discussions. This work is supported by Grant-in-Aid for Scientific Research B (13440058) of the Ministry of Eduaction, Culture, Culture, Science and Technology, Japan."
    },
    "0208/astro-ph0208178_arXiv.txt": {
        "abstract": "The local expansion field ($v_{220}<1200\\kms$) {\\em and\\/} the cosmic expansion field out to $30\\,000\\kms$ are characterized by $H_{0}=58\\;$[km\\,s$^{-1}$\\,Mpc$^{-1}$]. While the random error of this determination is small ($\\pm2$ units), it may still be affected by systematic errors as large as $\\pm10\\%$. The local expansion is outlined by Cepheids and by Cepheid-calibrated TF distances of a complete sample of field galaxies and by nearby groups and clusters; the cosmic expansion is defined by Cepheid-calibrated SNe\\,Ia. The main source of systematic errors are therefore the shape and the zero point of the P-L relation of Cepheids and its possible dependence on metallicity. GAIA will essentially eliminate these systematic error sources. Another source of systematic error is due to the homogenization of SNe\\,Ia as to decline rate $\\Delta m_{15}$ and color $(B-V)$. GAIA will discover about half of the 2200 SNe\\,Ia which will occur during a four-year lifetime within $10\\,000\\kms$. For many of them ground-based follow-up will provide useful photometric parameters (and spectra), which will allow to fix the dependence of the SNe\\,Ia luminosity on $\\Delta m_{15}$ and $(B-V)$ with high accuracy. At the same time they will yield exquisite distances to a corresponding number of field galaxies. --- GAIA will also revolutionize the very local distance scale by determining fundamental distances of the companion galaxies of the Milky Way and even of some spirals in- and possibly outside the Local Group from their rotation curves seen in radial velocities and proper motions.  Moreover, GAIA will obtain trigonometric parallaxes of RR Lyrae stars, of red giants defining the TRGB, of stars on the ZAMS, of White Dwarf defining their cooling sequence, and of globular clusters, and determine the metallicity dependence of these distance indicators. It will thus establish a self-controlling network of distance indicators within the Local Group and beyond. ",
        "introduction": "\\label{sec:1} An evaluation of GAIA's contribution to the extragalactic distance scale requires as a first step a brief description of the present situation (Section~\\ref{sec:2}). Inadequacies of the present situation of the distance scale are discussed in Section~\\ref{sec:3}. Important improvements expected from GAIA are outlined in Section~\\ref{sec:4}. Some conclusions are given in Section~\\ref{sec:5}. ",
        "conclusions": "\\label{sec:5} Reliable distance indicators (Cepheids, TF distances of complete samples, and nearby cluster distances) require a {\\em local\\/} value of $H_0=59.2\\pm1.4$. Cepheid-calibrated SNe\\,Ia give a {\\em large-scale\\/} value of $H_0=57.4\\pm2.3$. The two values are statistically indistinguishable and one may assume $H_0=58\\pm2$ everywhere.  The quoted error is only the statistical error. But the solution for $H_0$ is dominated by {\\em systematic\\/} errors which could amount to as much as 10\\%. The largest systematic errors are introduced by Cepheids, which form the basis of the local as well as of the large-scale expansion rate. Equation~(\\ref{eq:12}) suggests, in case of a non-linear form of the P-L relation of Cepheids, that $H_0$ is increased by 6\\% for Cepheids with a median period of 30 days, i.e.\\ a typical period for extra\\,-\\,Local Group galaxies. A metallicity dependence of the Cepheid P-L relation could affect $H_0$ by $\\sim\\!5\\%$; if the metallicity correction of Kennicutt et~al.\\ (1998) is taken at face value the effect would go in the direction of decreasing $H_0$. The zero point of the P-L relation seems to be well determined (Table~\\ref{tab:1}), but a systematic error of 3-4\\% cannot be excluded.  The three systematic error sources of Cepheid distances, which propagate with almost full weight into the entire extragalactic distance scale, will essentially be eliminated by GAIA.  The next important source of a systematic error --- at least for the large-scale value of $H_0$ --- comes from the homogenization of SNe\\,Ia as to decline rate $\\Delta m_{15}$ and color $(B-V)$. In fact errors of the slope of the $M-\\Delta m_{15}$ and $M-(B-V)$ relations could introduce systematic errors of the SNe\\,Ia distances of $\\sim 3\\%$ (Parodi et~al.\\ 2000). This problem can entirely be solved if good photometry will be obtained for the many hundreds of SNe\\,Ia to be discovered within $10\\,000\\kms$ by GAIA. This at the same time will yield irreplaceable distances to an equal number of field galaxies (to within random errors of $\\pm5\\%$), which will outline any concerted deviations from pure Hubble flow out to $10\\,000\\kms$.  The one remaining systematic error of $\\la 4\\%$ comes from the difficult photometry of extra-Local Group Cepheids with the wide-field camera (WFPC-2) of HST. GAIA cannot offer a handle on this problem, and it has to await future photometry from space.  GAIA will not only much improve the value of $H_0$, but also the age of the oldest objects. Definitive distances to globular clusters will reduce the error of their ages, and the improvement of the distances of Local Group galaxies {\\em and\\/} the measurement of their proper motions are very important for the determination of the dynamical age of the Local Group (Lynden-Bell 1999). Thus $H_0$ and a minimum age of the Universe can be used as strong priors for the CMB fluctuation spectrum (cf. Netterfield et~al.\\ 2002; Pryke 2001) which will narrow down the possible range of other cosmological parameters like the baryon density $\\Omega_{\\rm b}$, the matter density $\\Omega_{\\rm Matter}$, and the cosmological constant $\\Lambda$."
    },
    "0208/nucl-th0208020_arXiv.txt": {
        "abstract": "We investigate the liquid-gas phase transition of dense matter in supernova explosion by the relativistic mean field approach and fragment based statistical model. The boiling temperature is found to be high ($T_{boil} \\geq 0.7\\ \\MeV$ for $\\rhoB \\geq 10^{-7}\\ \\fmcube$), and adiabatic paths are shown to go across the boundary of coexisting region even with high entropy. This suggests that materials experienced phase transition can be ejected to outside. We calculated fragment mass and isotope distribution around the boiling point. We found that heavy elements at the iron, the first, second, and third peaks of r-process are abundantly formed at $\\rhoB = 10^{-7}, 10^{-5}, 10^{-3}$ and $10^{-2}\\ \\fmcube$, respectively. ",
        "introduction": "It is generally believed that there exist several phases in nuclear matter. Among the phase transitions between these phases, the nuclear liquid-gas phase transition has been extensively studied in these three decades~\\cite{HI-Exp}. It takes place in relatively cold ($T_{boil}$ = (5-8) MeV) and less dense ($\\rhoB\\sim\\rho_0/3$) nuclear matter, and it causes multifragmentation in heavy-ion collisions. When the expanding nuclear matter cools down and goes across the boundary of coexisting region, it becomes unstable against small fluctuations of density or $np$ asymmetry, then various fragments are abundantly formed almost simultaneously. Especially at around the critical point, fragment distribution is expected to follow the power law~\\cite{Fisher}, $Y_f \\propto A^{-\\tau}$, which is one of the characteristic features of critical phenomena. Recent theoretical model studies~\\cite{Dyn-HI,NSE-HI,Hirata2002,Fai-Randrup,Bauer} have shown that it is very difficult to describe this fragment distribution in a picture of sequential binary decays of one big compound nucleus, which has been successfully applied to the decay of nuclei at low excitation. This finding suggests that it is necessary to consider {\\em statistical ensemble of various fragment configurations} rather than one dominant configuration in describing fragment formation at around the boundary of coexisting region.   In the universe, the temperature and density of this liquid-gas phase transition would be probed during supernova explosion. In the collapse and bounce stages of supernova explosion, the density and temperature are high enough to keep statistical equilibrium~\\cite{Bet90,Suz90}. At baryon densities of $\\rhoB \\ge 10^{-5}\\ \\fmcube $, since the density is too high for neutrinos to escape, neutrinos are trapped in dense matter. This leads to an approximate conservation of lepton fraction $Y_L = L/B$ and entropy per baryon $S/B$, where $L$ and $B$ denote the lepton and baryon number, respectively. After the core bounce, supernova matter, which is composed of nucleons and leptons, expands and cools down. As the baryon density and the temperature decrease, charged particle reactions become insufficient and the chemical equilibrium ceases to hold, namely the system freezes out at this point. If the supernova matter goes across the boundary of coexisting region and the boiling point $T_{boil}$ of the liquid-gas phase transition is higher than the freeze-out temperature $T_{fo}$, this matter will dissolve into fragments and form various nuclei in a critical manner. It further keeps equilibrium and expands to the freeze-out point. The statistical distribution of fragments at freeze-out would provide the initial condition for following nucleosynthesis such as the r-process. (See following references on r-process~\\cite{Burbidge,Meyer} and references therein.)  The importance of the nuclear liquid-gas phase transition in supernova explosion was already noticed and extensively studied before~\\cite{Lattimer}. However, there are two more points which we should consider further. First, the main interest in the previous works was limited to the modification of the equation of state (EoS). The nuclear distribution as an initial condition for the r-process was not studied well. Secondly, in constructing the EoS of supernova matter, the mean field treatment was applied in which one assumed one kind of large nucleus surrounded by nucleon and alpha gas~\\cite{TM1-table,LS91}. At temperature much above or below the boiling point, fragment mass distribution is narrow and the one species approximation works well. However, since fluctuation dominates at around the boiling point, it is  necessary to take account of fragment mass and isotope distribution. This distribution of fragments can modify the following r-process nucleosynthesis provided that the freeze-out point is not far from the boiling point.  In this work, we study nuclear fragment formation through the nuclear liquid-gas phase transition during supernova explosion. This process may lead to the production of medium mass nuclei as seed elements and serve as a pre-process of the r-process. We call this process as LG process~\\cite{IOS2001-YKIS01b}. In order to pursue this possibility quantitatively, it is necessary to determine the liquid-gas coexisting region. We find that the liquid-gas coexisting region extends down to very low density keeping the boiling point around $T_{boil} \\sim 1$ MeV in a two-phase coexistence treatment of EoS with the Relativistic Mean Field (RMF) model~\\cite{TM1-table,Sero-Walecka,TM1,Muller,RMF-Other}. In supernova explosion, it can happen that material with $S/B \\geq 10$ is ejected to outside~\\cite{SumiyoshiNext}. Adiabatic paths of ejecta are found to go through the calculated liquid-gas coexisting region even with high entropy. Having this finding of the passage through the coexisting region, we investigate the fragment distribution at around the boiling point in a statistical models of fragments~\\cite{NSE-HI,Hirata2002,Fai-Randrup,Bauer}, referred to as the Nuclear Statistical Equilibrium (NSE) in astrophysics. We show that heavy elements around the first, second, and third peaks of r-process are abundantly formed at $\\rhoB = 10^{-5}, 10^{-3}$ and $10^{-2}\\ \\fmcube$, respectively, with temperatures around and just below $T_{boil}$ in NSE. Furthermore, the isotope distribution of these elements are also well described in this model. We find that it is important to take account of the Coulomb energy reduction from the screening by electrons in supernova matter. Although nuclei formed at high densities $\\rhoB \\sim 10^{-2}\\ \\fmcube $ having very small entropy at around $T_{boil}$ is not likely ejected to outside, heavy elements up to the r-process third peak are already formed statistically at these densities.  This paper is organized as follows. We describe the treatment of two-phase coexistence with RMF model in Sec. 2. The liquid-gas coexistence is shown in the ($\\rhoB,T$) diagram. We demonstrate that it would be possible for a part of ejecta in supernova explosion to experience the liquid-gas coexisting region. The effects of liquid-gas coexistence on EoS, proton fraction, and adiabatic path are also studied. In Sec. 3, we describe the nuclear statistical model of fragments at equilibrium (NSE) to study the production of elements. We take into account the Coulomb energy modification from electron screening. We evaluate the fragment distribution at around the boiling point and in the coexisting region within this statistical model. We found that fragments are formed abundantly even at very low densities if the temperature is around the boiling point. We compare the calculated mass and isotope distributions of fragments with the solar abundance~\\cite{Abundance}. In Sec. 4, we discuss the possibility of the ejection of nuclei synthesized in the coexisting region referring to a hydrodynamical calculation of supernova explosion~\\cite{SumiyoshiNext}. We summarize our work in Sec. 5.   \\vfill\\break ",
        "conclusions": "In this paper, we have investigated the liquid-gas phase transition of supernova matter, and its effects on the fragment formation. We have used two models --- the Relativistic Mean Field (RMF) model and the Nuclear Statistical Equilibrium (NSE) model.   In RMF, we have used the interaction TM1, which has been successfully applied to finite nuclei including neutron rich unstable nuclei, neutron stars, and supernova explosion~\\cite{TM1-table,Sum95b,Sum00,Sum95c}. Leptons are shown to play non-trivial roles such as the symmetrization of nuclear part of supernova matter. As a result, nuclear liquid gains symmetry energy, and the calculated boiling points in supernova matter ($T_{boil} > 1\\ \\MeV$ for $\\rhoB \\geq 10^{-10}\\ \\fmcube$) are comparable to those in symmetric nuclear matter at low densities. Adiabatic paths are shown to go across the boundary of coexisting region even at high entropy such as $S/B \\geq 10$, which is expected to be enough for supernova matter to be ejected to outside. Clear concentration of adiabatic paths to the boundary of coexisting region have been found. All of these findings suggest that at least a part of ejecta in supernova explosion would experience the liquid-gas phase transition before freeze-out.   In NSE, we have used nuclear binding energies of Myers-Swiatecki model~\\cite{MS1994} with Coulomb correction due to electron screening as a medium effect~\\cite{Lattimer}. Since larger species of nuclei become stable with this Coulomb energy correction, we have adopted the mass table of around 9000 nuclei constructed by Myers and Swiatecki~\\cite{MS1994}. Because of the finiteness of nuclei, they lose surface and Coulomb energy compared to the case of coexistence treatment of two infinite matter phases in RMF. The boiling points become slightly lower, but they are still high; $T_{boil} \\geq 0.7$ MeV for $\\rhoB \\geq 10^{-7}\\ \\fmcube$. Calculated fragment mass distributions around $T_{boil}(\\rhoB)$ show enhancement of the iron peak elements, the first, second, and third peak r-process elements at $\\rhoB = 10^{-7}, 10^{-5}, 10^{-3}$ and $10^{-2}\\ \\fmcube$, respectively. In addition, calculated isotope distribution shows that very neutron rich nuclei around and beyond the neutron dripline may exist under thermal and chemical equilibrium in supernova matter with degenerate neutrinos. These unstable nuclei against neutron emission would provide a lot of neutrons after freeze-out, which may help the r-process to proceed.   From the present investigations, we can draw a new scenario for making seed nuclei before the r-process; fragments are abundantly formed through the liquid-gas phase transition of supernova matter before the freeze-out, and this formation of fragments serve to produce the bulk structure of the seed elements. We call this process as the LG process as a pre-process of r-process~\\cite{IOS2001-YKIS01b}.   It is interesting to note that our model based on the liquid-gas coexisting state of supernova matter can even provide the r-process nuclei or their seed in a simple manner based on the condition determined by the dynamics of supernova explosion such as $\\rhoB$, $Y_L$, and $T$. One of the most promising conditions is $\\rhoB = 10^{-5}\\ \\fmcube$. This density roughly corresponds to the neutrino sphere. The entropy at $T_{boil}$ is a little smaller than the ejection criteria, $S/B \\geq 10$ in one-dimensional hydrodynamical calculation of supernova explosion~\\cite{SumiyoshiNext}. However, it would be possible that matter with small entropy can be ejected by convection and/or jet in asymmetric supernova explosion~\\cite{Janka,MacFadyen}. The most conservative freeze-out density for ejection would be $\\rhoB = 10^{-7}\\ \\fmcube$. The entropy at $T_{boil}$ is large enough, and the seed nuclei will be nucleons, $\\alpha$, iron peak nuclei and a small amount of the first peak nuclei of r-process. Higher densities may not be relevant to ejection, but it may be closely related to the nuclear distribution on hot neutron star surface.    In this work, we have assumed equilibrium throughout this paper. One of the key questions is the freeze-out conditions of supernova matter, at which nuclear reactions become less frequent and supernova matter goes off equilibrium in the expansion time-scale. The seed nuclear distribution of the r-process will be given as the nuclear distribution on the freeze-out line in the $(\\rhoB,T)$ diagram. It is important to determine the freeze-out condition in supernova dynamics. Another important direction is to construct a model which includes both of the mean field nature such as in RMF and the statistical nature in NSE. In a present NSE treatment, only the Coulomb correction is included as the medium effects, and medium effects from strong interactions are neglected. This neglection may lead to the overestimate of neutron rich nuclei, as discussed in recent statistical fragmentation models~\\cite{Stat-Isospin}. On the other hand, in the Thomas-Fermi treatment of heavy-nuclei with EoS derived using RMF, since statistical nature or fragment distribution is not taken care of, the treatment is not sufficient especially at around $T_{boil}$. Works in these directions are in progress."
    },
    "0208/astro-ph0208446_arXiv.txt": {
        "abstract": "We report low- and high-resolution spectra of comet C/2002\\,C1 (Ikeya-Zhang) from McDonald Observatory.  The comet had a well-developed ion tail including CO$^+$, CO$^+_2$, CH$^+$, and H$_{2}$O$^+$. We used our high-resolution spectra to search for N$_2^+$. None was detected and we placed upper limits on N$_2^+$/CO$^+$ of $5.4\\times10^{-4}$.  N$_2^+$ was detected in the low-resolution spectra but we show that this emission was probably telluric in origin (if cometary, we derive N$_2^+$/CO$^+ = 5.5\\times10^{-3}$, still very low). We discuss the implications for the conditions in the early solar nebula of the non-detection of N$_2^+$.  These depend on whether the H$_{2}$O ice was deposited in the amorphous or crystalline form.  If H$_{2}$O was deposited in its crystalline form, the detection of CO$^+$ but not N$_2^+$ has implications for H$_{2}$O/H$_2$ in the early solar nebula. ",
        "introduction": "Knowledge of the nitrogen content of comets is important for an understanding of conditions in the early solar nebula. Nitrogen exists in cometary ices in many forms including N$_2$, NH$_{3}$, HCN, etc. Most of these species are chemically reactive in the coma gases making it difficult to use coma observations to unravel the nitrogen chemistry of the solar nebula. However, conditions in the early solar nebula were such that  the dominant equilibrium species of carbon, oxygen and nitrogen should be N$_2$, CO, and H$_{2}$O \\citep{lepr80}. Observations of N$_2$H$^+$ in dense molecular clouds, coupled with chemical models, led \\citet{wowyzi92} to infer that in potential star-forming regions nitrogen is preferentially in N$_2$, rather than NH$_{3}$.  N$_2$ is the least reactive of the nitrogen-bearing species and is thus the most appropriate to study to understand the nitrogen chemistry of comets.  However, observations of cometary N$_2$ are exceedingly difficult. Ground-based spectra are hampered by the telluric N$_2$ atmosphere; spacecraft flyby mass spectrometer observations are compromised by the fact that N$_2$ shares the mass 28 bin with CO, which is known to be quite common in cometary comae.  A suitable method for measuring the N$_2$ content of cometary ices is by studying its ion, N$_2^+$.  Generally, observations are obtained of the N$_2^+$ 1N $B^2\\Sigma_u^+-X^1\\Sigma_g^+$ (0,0) band at 3914\\AA. Successful observations of this band require that the comet have a well-developed ion tail and that the data be obtained with sufficiently high spectral resolution to isolate the band from other cometary emissions and from any telluric N$_2^+$ emission.  Comet C/2002\\,C1 (Ikeya-Zhang) was discovered in early February 2002 and reached perihelion on 2002 March 18 at a heliocentric distance of 0.507\\,{\\sc au}.  Our spectral observations showed that it had a strong ion tail containing ions of CO$^+$, CO$_2^+$, CH$^+$, and H$_{2}$O$^+$.  The comet was relatively bright, making it possible to observe at high spectral-resolution.  Thus, we obtained spectra of Ikeya-Zhang at both high- and low-resolution in order to search for the signature of N$_2^+$ in the tail. In this paper, we report on our non-detection of any N$_2^+$ attributable to the comet and discuss the implications of our derived upper limits. ",
        "conclusions": ""
    },
    "0208/astro-ph0208393_arXiv.txt": {
        "abstract": "{We report the discovery of a small H$\\alpha$ nebula positionally coincident with the candidate neutron star \\ss located at the center of the supernova remnant \\gg . The nebula has a roughly circular shape with a diameter of $\\sim$6$''$ and a flux of $\\sim$$10^{-2}$ photons cm$^{-2}$ s$^{-1}$ in the H$\\alpha$ line. Considering the uncertainties in the distance and energy output from the putative neutron star, we find that such a flux can be explained either in a bow-shock model or assuming that the nebular emission is due to photo-ionization and heating of the ambient gas. ",
        "introduction": "\\label{sect:intro}   The X--ray source \\ss is a strong neutron star candidate, very likely associated with the shell-like supernova remnant \\gg\\ (Aschenbach 1998). \\ss was first seen with the \\textit{ROSAT} satellite (Aschenbach 1998, Aschenbach et al. 1999) and subsequently studied with \\textit{ASCA} (Slane et al. 2001) and \\textit{BeppoSAX} (Mereghetti 2001). Its location,  very close to the geometrical center of \\gg\\  , suggested an association with this young remnant, but the situation was complicated by the presence  of two early type stars (HD 76060 and Wray 16-30) that might have been responsible for the observed X--rays from \\ss, as well as by the presence of other X--ray sources in the vicinity (Mereghetti 2001).  The picture was finally clarified thanks to the accurate localization obtained with the \\textit{Chandra} satellite (Pavlov et al. 2001). The absence of any optical counterparts at the \\textit{Chandra} position, down to magnitudes B$\\sim$22.5 and R$\\sim$21, implies a very high X--ray to optical flux ratio, consistent with an isolated neutron star. The soft  spectrum of \\ss, well described by a blackbody with $kT$$_{\\rm{BB}}\\sim$0.4 keV, is similar to that of other compact X--ray sources found in supernova remnants, such as, e.g., CasA (Mereghetti, Tiengo \\& Israel 2002) and G 296.5+10.0 (Pavlov et al. 2002). For an assumed distance of 1 kpc the X--ray luminosity of \\ss is $\\sim$10$^{32}$$-$10$^{33}$ erg s$^{-1}$. No  pulsations have   been detected so far.  Here we present optical images of the central region of \\gg\\ showing the presence of a small H$\\alpha$ nebula at the position of \\ss.   \\begin{figure*}[!ht] \\vskip 0.0truecm \\centerline{\\psfig{figure=Eg171_f1_d.ps,height=130mm,angle=-90}} \\vskip 0.0truecm \\caption{H$\\alpha$ (left) and R band (right) images of the  central region of \\gg\\, North is to the top, East to the left. The cross indicates the position of \\ss (RA(J2000) = 8$^{h}$ 52$^{m}$ 01$^{s}$.38, Dec(J2000)= $-$46$^{\\circ}$ 17$'$ 53$''$.34,  Pavlov et al. 2001). To locate it, we performed an astrometry  based on the coordinates of a number of stars from the GSC II catalog. The rms of our astrometry fit is 0.07$''$; taking into account possible systematic errors in the GSC catalogue, we estimate an overall error $<$1$''$.} \\label{fig:cfr} \\end{figure*} ",
        "conclusions": "We have discovered a faint ($\\sim$10$^{-2}$ ph cm$^{-2}$ s$^{-1}$) H$\\alpha$ nebula positionally coincident with the point-like X--ray source \\ss , which is thought to be the compact remnant associated to SNR G~266.1$-$1.2. Although this region contains several diffuse H$\\alpha$ features, the positional coincidence and the possible evidence for a peculiar color, compatible with a pure Balmer emission line spectrum, suggest a relation between the H$\\alpha$ nebula A and the putative neutron star. In fact, although the lack of a  period and spin-down measurement for \\ss make the energetics somewhat uncertain, we have shown that the luminosity of the nebula is compatible with the predictions of the two models which have been invoked to explain a few H$\\alpha$ nebulae associated to different kinds of neutron stars.  Different distances are favored by the two scenarios. The model based on photo-ionization suggests  a distance greater than $\\sim$1 kpc, consistent with that inferred from X--ray absorption in the \\gg\\ SNR (Mereghetti \\& Pellizzoni 2001). In the case of a bow-shock nebula, the  most likely distance is smaller than $\\sim$0.5 kpc, due to the presumably small value of $\\dot{E}_{\\rm{rot}}$  and the lack of evidence for a high transverse velocity for the neutron star. Of course we cannot exclude a greater distance in the bow-shock scenario if the neutron star  has a high velocity close to the direction of the line of sight. Although this might be consistent with the circular symmetry of the  nebula, we note that the chance probability for, e.g., $i<10^{\\circ}$ is only $\\sim$7\\%.   More detailed investigations are required to confirm the proposed association between nebula A and \\ss\\, and eventually to discriminate between the two possible mechanisms. In particular, deeper imaging with high resolution can provide information on the shape of the nebula, while high-resolution spectroscopy is needed to measure the width of the H$\\alpha$ line. In the bow-shock model one should see a major fraction of the emission with velocity widths comparable to the shock velocity, larger than that of the narrow lines expected in the photo-ionisation model with thermal velocities $\\approxlt40$ km s$^{-1}$ (Raymond 1991)."
    },
    "0208/astro-ph0208500_arXiv.txt": {
        "abstract": "The standard treatment of gravitational lensing by a point mass lens $M$ is based on a weak-field deflection angle $\\hat{\\alpha} = 2/x_0$, where $x_0 = r_0 c^2/2 G M$ with $r_0$ the distance of closest approach to the mass of a lensed light ray. It was shown that for a point mass lens, the total magnification and image centroid shift of a point source remain unchanged by relativistic corrections of second order in $1/x_0$. This paper considers these issues analytically taking into account the relativistic images, under three assumptions {\\bf A1}--{\\bf A3}, for a Schwarzschild black hole lens with background point and extended sources having arbitrary surface brightness profiles. The assumptions are {\\bf A1:} The source is close to the line of sight and lies in the asymptotically flat region outside the black hole lens; {\\bf A2:}  The observer-lens and lens-source distances are significantly greater than the impact parameters of the lensed light rays; and {\\bf A3:} The distance of closest approach of any light ray that does not wind around the black hole on its travel from the source to the observer, lies in the weak-field regime outside the black hole. We apply our results to the Galactic black hole for lensing scenarios where {\\bf A1}--{\\bf A3} hold. We show that a single factor characterizes the full relativistic correction to the weak-field image centroid and magnification. As  the  lens-source distance increases, the relativistic correction factor strictly decreases.  In particular, we find that for  point and extended sources about $10 \\ {\\rm pc}$ behind the black hole, which is a distance significantly outside the tidal disruption radius of a sun-like source, the relativistic correction factor is minuscule, of order $10^{-14}$. Therefore, for standard lensing configurations, any detectable relativistic corrections to microlensing by the Galactic black hole will most likely have to come from sources significantly closer to the black hole. ",
        "introduction": "Microlensing describes gravitational lensing of a source whose multiple images are not resolved. Two fundamental microlensing observables are the total magnification (photometry) and image centroid shift (astrometry) of images of a lensed source.  These observables have important astrophysical  applications  such as determining the mass and distance to the lens, angular radius of the source, etc. (see, e.g., \\citealt{pac96}, \\citealt{pac98}, \\citealt{bsvb98}, \\citealt{jhp99}, \\citealt{gp02}, and references therein).   A natural issue to explore is how are the photometry  and astrometry of a source being lensed by a point mass changed when the point mass is replaced by a black hole lens.  This could have important implications for the testability of general relativity's predictions about how the gravitational field of a black hole affects light rays. Indeed, the standard theoretical framework for point mass microlensing is based on  relativistic calculations to first-order in $1/x_0$ about a Schwarzschild black hole, where $$ x_0 = \\frac{r_0}{2 \\APrg}, \\qquad \\APrg = \\frac{G \\APmbh}{c^2}, $$ where $\\APmbh$ is the black hole's mass and $\\APrg$ the gravitational radius. \\citet{e00} found that to second-order in $1/x_0$, the relativistic corrections appear in the position and magnification of images due to a point mass lens, while no such correction appears in the total magnification.   \\citet{lw01} also showed that no relativistic correction to second-order in $1/x_0$ occurs for the associated image centroid shift. In this paper, we extend the work of the previous authors by determining an analytical expression under assumptions {\\bf A1}--{\\bf A3} for the full Schwarzschild black hole relativistic correction of the image centroid shift, which includes the total magnification, for point and extended sources with arbitrary surface brightness.  The full relativistic corrections will then be applied to the case of the massive black hole at the center of our Galaxy. Microlensing by the Galactic black hole has been studied by several authors in the weak-field limit of the black hole (e.g, \\citealt{wy92}, \\citealt{as99}, \\citealt{al01}, \\citealt{a01}). In addition, the precession of star orbits in the strong-field regime of the black hole was considered by, e.g., \\citet{j98a,j98b,j99}, \\citet{fm00}. \\citet{ve00} gave a  numerical treatment of the magnifications of several relativistic images for a point source being lensed by a Schwarzschild black hole lens at the Galactic center. They considered sources within our Galaxy that are far away from the black hole (about $8.5 \\ {\\rm k pc}$), while we shall consider sources as close as $10 \\ {\\rm pc}$ to the black hole. Analytical work on magnification due to a Schwarzschild lens was also done by several authors  for point and/or extended sources with uniform   brightness profiles (e.g., \\citealt{o87}, \\citealt{fkn00}, \\citealt{bcis01}, \\citealt{ert02}). As noted above, our microlensing treatment will apply not only to the magnification, but the image centroid of extended sources with arbitrary surface brightness.   We shall also show that a single factor approximates the relativistic corrections to the weak-field total magnification and image centroid due to a Schwarzchild black hole lens at the Galactic center.  The same factor applies to  either a point or extended source. Estimates of this factor will be given for lens-sources distances ranging from $10 \\ {\\rm pc}$ to $100 \\ {\\rm pc}$. In principle, the magnification and image centroid of sources closer to  the Galactic black hole should provide stronger relativistic microlensing signatures. We shall show that even for sources about $10 \\ {\\rm pc}$ behind the black hole, the full relativistic correction is still negligible.       Section~\\ref{sec-image-mag} reviews some basic results about lensing by a Schwarzschild black hole. In Section~\\ref{sec-image-centroid}, we compute explicitly the image centroid due to a Schwarschild black hole acting on point and extended sources with arbitrary brightness profiles.  Our image centroid formula expresses the relativistic image centroid in terms of the weak-field image centroid due to a point mass lens. Section~\\ref{sec-applications} estimates the relativistic corrections to microlensing by the Galactic black hole. ",
        "conclusions": "Previous work on gravitational lensing by the black hole at the Galactic center investigated the relativistic corrections to the weak-field total magnification and image centroid to second order in $1/x_0 = 2 G M/(r_0 c^2)$, where $r_0$ is the distance of closest approach of the light ray to the black hole. It was shown recently that for a point mass lens the total magnification and image centroid shift of a point source remain unchanged by relativistic corrections of second order in $1/x_0$. We computed the relativistic corrections for a Schwarzschild black hole lens under assumptions {\\bf A1}--{\\bf A3}.   These corrections were applied to the case of the massive black hole at the Galactic center. We found that the weak-field magnification and image centroid have approximately the same relativistic correction. This correction is a strictly decreasing function of the lens-source distance $D_{LS}$. For $D_{LS} \\ge 10 \\ {\\rm pc}$, the relativistic correction is of order at most $10^{-14},$ a minuscule correction. Hence, for standard lensing configurations, a nontrivial relativistic correction to microlensing by the Galactic black hole would likely have to come from sources deep inside the black hole's potential well."
    },
    "0208/astro-ph0208099_arXiv.txt": {
        "abstract": "We announce the first public release of the SDSS Moving Object Catalog, with SDSS observations for 58,117 asteroids. The catalog lists astrometric and photometric data for moving objects observed prior to Dec 15, 2001, and also includes orbital elements for 10,592 previously known objects. We analyze the correlation between the orbital parameters and optical colors for the known objects, and  confirm that asteroid dynamical families, defined as clusters in orbital parameter space, also strongly segregate in color space. Their distinctive optical colors indicate that the variations in chemical composition within a family are much smaller than the compositional differences between families, and strongly support earlier suggestions that asteroids belonging to a particular family have a common origin. ",
        "introduction": "\\label{sect:intro}  %   SDSS is a digital photometric and spectroscopic survey which will cover 10,000 deg$^2$ of the Celestial Sphere in the North Galactic cap and produce a smaller ($\\sim$ 225 deg$^2$) but much deeper survey in the Southern Galactic hemisphere\\cite{York00}. The survey sky coverage will result in photometric measurements for about 50 million stars and a similar number of galaxies. About 30\\% of the Survey is currently finished. The flux densities of detected objects are measured almost simultaneously in five bands\\cite{F96} ($u$, $g$, $r$, $i$, and $z$) with effective wavelengths of 3551 \\AA, 4686 \\AA, 6166 \\AA, 7480 \\AA, and 8932 \\AA, 95\\% complete for point sources to limiting magnitudes of 22.0, 22.2, 22.2, 21.3, and 20.5 in the North Galactic cap. Astrometric positions are accurate\\cite{Pier02} to about 0.1 arcsec per coordinate (rms) for sources brighter than 20.5$^m$, and the morphological information from the images allows robust star-galaxy separation\\cite{Lupton01} to $\\sim$ 21.5$^m$.   \\subsection{        SDSS Observations of Moving Objects }  SDSS, although primarily designed for observations of extragalactic objects, is significantly contributing to studies of the solar system objects, because asteroids in the imaging survey must be explicitly detected to avoid contamination of the samples of extragalactic objects selected for spectroscopy. Preliminary analysis of SDSS commissioning data\\cite{Ivezic01} showed that SDSS will increase the number of asteroids with accurate five-color photometry by more than two  orders of magnitude (to about 100,000), and to a limit about five magnitudes fainter (seven magnitudes when the completeness limits are compared) than previous multi-color surveys (e.g. The Eight Color Asteroid Survey\\cite{ZTT85}). The main results derived from these early SDSS observations are  \\begin{enumerate} \\item A measurement of the main-belt asteroid size distribution to a significantly smaller size limit ($<1$ km) than possible before. The size distribution resembles a broken power-law, independent of the heliocentric distance: $D^{-2.3}$ for 0.4 km $< D <$ 5 km, and $D^{-4}$ for 5 km $< D <$ 40 km. \\item A smaller number of asteroids compared to previous work. In particular, the number of asteroids with diameters larger than 1 km is about $7\\times10^5$. \\item The distribution of main-belt asteroids in 4-dimensional SDSS color space is strongly bimodal, and the two groups can be associated with S (rocky) and C (carbonaceous) type asteroids, in agreement with previous studies based on smaller samples\\cite{CMZ75}. A strong bimodality is also seen in the heliocentric distribution of asteroids: the inner belt is dominated by S type asteroids centered at $R$ \\about 2.8 AU, while C type asteroids, centered at $R$ \\about 3.2 AU, dominate the outer belt. \\end{enumerate}  The preliminary analysis of SDSS commissioning data was based on a sample of about 10,000 objects. Here we describe the first public catalog of SDSS asteroid observations that includes about 60,000 objects, and show an example of analysis made possible by such a large, accurate and homogeneous database. ",
        "conclusions": ""
    },
    "0208/astro-ph0208176.txt": {
        "abstract": "s{ Attempts to measure extragalactic distances over the last 90 years are briefly described. It follows a short history of the discovery of the expansion of space. Reasons are discussed for the decrease of the Hubble constant from $H_{0}\\approx500$ originally to $H_{0}\\la60$ at present. Remaining problems with Cepheids as local distance calibrators are outlined. } % **************************************************************  % ************************************************************** % 1. Introduction % ************************************************************** ",
        "introduction": "% The conquest of the third dimension beyond the Galaxy was one of the great challenges of the 20th century. It is not possible to do justice to its fascinating and complex history in a few pages. Much of the progress during the first half of the century was made at Mount Wilson. Insight into this part of the history is provided by \\citet{Sandage:95}, and his forthcoming monograph ``An informal history of the Mount Wilson Observatory (1904$-$1950)'' will be the prime reference for the subject with all its ramifications into most fields of astronomy.  % ************************************************************** % 2. Early Galaxy Distances % ************************************************************** ",
        "conclusions": ""
    },
    "0208/astro-ph0208266_arXiv.txt": {
        "abstract": "{We present a spectroscopic catalog of the neighboring massive clusters \\object{Abell~222} and \\object{Abell~223}. The catalog contains the positions, redshifts, $R$ magnitudes, $V-R$ color, as well as the equivalent widths for a number of lines for 183 galaxies, 153 of them belonging to the A~222 and A~223 system. We determine the heliocentric redshifts to be $z=0.2126\\pm0.0008$ for A~222 and $z=0.2079\\pm0.0008$ for A~223. The velocity dispersions of both clusters in the cluster restframe are about the same: $\\sigma = \\tol{1014}{+90}{-71}$ ~km~s$^{-1}$ and $\\sigma = \\tol{1032}{+99}{-76}$ ~km~s$^{-1}$ for A~222 and A~223, respectively.  While we find evidence for substructure in the spatial distribution of A~223, no kinematic substructure can be detected. From the red cluster sequence identified in a color--magnitude--diagram we determine the luminosity of both clusters and derive mass--to--light ratios in the $R$--band of $(M/L)_{\\mathrm{A222}} = (202 \\pm 43)~h_{70}~M_{\\sun}/L_{\\sun}$ and $(M/L)_{\\mathrm{A223}} = (149 \\pm 33)~h_{70}~M_{\\sun}/L_{\\sun}$. Additionally we identify a group of background galaxies at $z \\sim 0.242$. ",
        "introduction": "\\label{sec:introduction} A~222/223 are two Abell clusters at $z \\approx 0.21$ separated by $\\sim14\\arcmin$ on the sky, or $\\sim2600h_{70}^{-1}$ kpc, belonging to the \\citet{1983ApJS...52..183B} photometric sample. Both clusters are rich having Abell richness class 3 \\citep{1958ApJS....3..211A}. While these are optically selected clusters, they have been observed by ROSAT \\citep{1997MNRAS.292..920W,1999ApJ...519..533D} and are confirmed to be massive clusters.  9 spectra of galaxies in the cluster region, most of them being cluster members, were known \\citep{1976ApJ...205..688S,1988ApJ...335..629N} before \\citet[][hereafter PEL]{2000A&A...355..443P} published a list of 53 spectra and did a first kinematical study of this system. PEL also found 4 galaxies at the cluster redshift in the region between the clusters (hereafter ``intercluster region''), indicating a possible connection between the clusters.  We report 184 independent redshifts for 183 galaxies in the field of Abell~222 and Abell~223, more than three times the number of redshifts previously known, as well as equivalent widths for a number of lines.  The paper is organized as follows. In Sect.~\\ref{sec:data-data-reduction} we describe the reduction of the spectroscopic and photometric data and discuss deviations from previous values in the literature. The spatial distribution and the kinematics of the double cluster system are examined in Sect.~\\ref{sec:spat-distr-kinem} with an emphasis on finding possible substructure. We determine the luminosity and mass--to-light ratio of the clusters by selecting the red cluster sequence in Sect.~\\ref{sec:mass-light-ratio}. Our results are summarized in Sect.~\\ref{sec:conclusions}. Throughout this paper we assume an $\\Omega_\\Lambda = 0.7,\\; \\Omega_\\mathrm{m} = 0.3,\\; H_0=70~h_{70}$~km~s$^{-1}$~Mpc$^{-1}$ cosmology. ",
        "conclusions": "\\label{sec:conclusions} We have reported 184 independent redshifts measurements for 183 galaxies in the field of Abell~222 and Abell~223, as well as equivalent widths for [\\ion{O}{ii}], [\\ion{O}{iii}], H$\\beta$, and H$\\alpha$, $R$ magnitudes, and $V-R$ color.  From a sample of 153 galaxies which we identified as cluster members, we derived a mean redshift and restframe velocity dispersion of $z =0.2126 \\pm 0.0008$, $\\sigma_\\mathrm{cor} = \\tol{1014}{+90}{-71}$~km~s$^{-1}$ for A~222 and $z = 0.2078 \\pm 0.0008$, $\\sigma_\\mathrm{cor} = \\tol{1032}{+99}{-76}$~km~s$^{-1}$ for A~223. The values of the redshifts are clearly outside the error margins of the values previously reported by PEL.  By comparing our wavelength calibration to the sky spectrum, which provides an independent wavelength standard, we were able to confirm the accuracy of our data and rule out the possibility of a zero point shift of more than 30~km~s$^{-1}$ for each mask. $R$ and $V$ band photometry was taken from WFI data.  Although the projected density maps of all spectroscopically identified galaxies and of a color selected sample with 702 members clearly show spatial substructure in A~223, neither the DS test nor the DIP statistics were able to find any kinematic substructure. Also no indications of a non--Gaussian parent population could be found.  We fitted a Schechter luminosity function to objects in the red cluster sequence identified in a color--magnitude diagram. Assuming an isothermal sphere model for the cluster we derived $(M/L)$ ratios in $R$--band, which are comparable for both clusters. The computed values are $(M/L)_R = (202\\pm43)~h_{70}~M_{\\sun}/L_{\\sun}\\;$ and $(M/L)_R = (149\\pm33)~h_{70}~M_{\\sun}/L_{\\sun}$ for A~222 and A~223, respectively.  This is within the range of values reported by other groups for other cluster.  \\citet{1978ApJ...226...55D} gave a range of $140-420~h_{70}~M_{\\sun}/L_{\\sun}$ in a study of 12 rich clusters. Typical values for virial mass--to--light ratio are at values of $M/L \\sim 210~h_{70}~M_{\\sun}/L_{\\sun}$ \\citep{1996ApJ...462...32C}. Typical values derived from X--ray masses tend to be somewhat lower than those from virial masses. \\citet{2000ApJ...543..521H} find a median value of $(M/L)_V \\sim 140~h_{70}~M_{\\sun}/L_{\\sun}$ in a study of eight nearby clusters and groups.  We cannot exclude the possibility that the values we report here are biased towards higher values by using an isothermal sphere model. Both cluster geometries clearly deviate from circular symmetric profiles. More robust mass estimates may thus lead to lower $(M/L)$ ratios.  A detailled discussion whether the galaxies between both clusters indeed belong to a structure connecting the cluster pair will be part of a forthcoming weak lensing study of this system."
    },
    "0208/astro-ph0208116_arXiv.txt": {
        "abstract": "We present {\\em ASCA} observations of supernova remnant (SNR) \\snr. The remnant has an irregular shell morphology and is interacting with a molecular cloud, evident from the presence of \\oh\\ masers and shocked molecular gas. The X-ray morphology is consistent with that at radio wavelengths, with a distinct enhancement in the south. The X-ray emission from the SNR is well described by a model of a thermal plasma which has yet to reach ionization equilibrium. The hydrogen column of $\\sim 6.0 \\times 10^{22}$ cm$^{-2}$ is consistent with the large distance to the remnant of $\\sim$ 22 kpc estimated from the maser velocities. We derive an X-ray luminosity of $L_{x}(0.5$-$10.0\\keV)= 1.8\\E{37}\\du^{2}\\ergs\\ps$, which makes G349.7+0.2 one of the most \\xray\\ luminous shell-type SNRs known in the Galaxy. The age of the remnant is estimated to be $\\sim$ 2800 yrs. The ambient density and pressure conditions appear similar to those inferred for luminous compact SNRs found in starburst regions of other galaxies, and provides support for the notion that these may be the result of SNR evolution in the vicinity of dense molecular clouds. ",
        "introduction": "\\label{sec:intro}  Because massive stars evolve quickly, they are often not far from their birth sites when they expire. The result is that many of the supernova remnants (SNRs) produced in the explosive events that mark the endpoint of stellar evolution for these stars are located near the molecular cloud complexes from which the progenitors emerged. The initial expansion of such an SNR is likely to proceed rather effortlessly as the progenitor star has generally sculpted a cavity in the ambient medium by virtue of a strong wind (Chevalier 1999). Eventually the blast wave must contend with the cavity walls, however, and when the cavity resides in a dense molecular cloud the resulting interaction reveals itself spectacularly in X-rays (Chevalier \\& Liang 1989). The remnant sweeps up massive amounts of material and heats it to X-ray emitting temperatures while seeding the cloud with metals synthesized in the supernova explosion. Such young SNRs encountering dense material can transform a large amount of their kinetic energy into radiation, appearing as bright (radio and \\xray) emission sources, often with irregular morphologies. They may be representative of a larger class of compact SNRs identified as bright radio sources in starburst regions of other galaxies (e.g. Kronberg, Biermann, \\& Schwab 1985; Antonucci \\& Ulvestad 1988; and Smith et al. 1998). Chevalier \\& Fransson (2001) have proposed that these sources represent SNRs that have evolved in the high density interclump medium of molecular clouds, and that a similar population that has escaped such high density regions is responsible for driving galactic winds in the host galaxies.  A good example of this type of SNR is N132D in the Large Magellanic Cloud (LMC). It is a luminous ($\\sim 5 \\E{37}\\ergs\\s^{-1}$ in X-ray band) small diameter SNR ($\\sim$ 44\\arcsec = 11.7 pc) that is evolving into a cavity wall on the edge of a molecular cloud (Hughes 1987, Banas et al. 1997). The X-ray emitting material comprises several hundred solar masses and the overall abundances are characteristic of the LMC interstellar material,  implying that the bulk of the emission is from swept--up material. However, optical spectra (Danziger \\& Dennefeld 1976) show that N132D is an oxygen rich SNR whose abundances are consistent with a $\\sim 20 M_\\odot$ progenitor (Blair, Raymond, \\& Long 1994), and high resolution X-ray spectral studies show the presence of an ejecta component as well (Hwang et al. 1993, Behar et al. 2001) indicating that N132D is a relatively young SNR; the large amount of swept-up material, as well as the high X-ray luminosity, are the result of an explosion in the high density surroundings of a molecular cloud.  G349.7+0.2 appears in some ways to be a Galactic counterpart to N132D. It is an SNR with a small angular size ($r \\sim 1$~arcmin) and the third highest radio surface brightness next to Cas~A and the Crab Nebula. Its nonthermal radio emission ($\\alpha_{r}\\simeq-0.5$) and roughly circular morphology, classifies it as a shell--type SNR (Shaver et al. 1985). However, it has an emission peak near the southeastern edge rather than a prominent limb--brightened structure and central cavity typical of shell--type remnants. The early H\\,{\\sc i} absorption measurements showed that G349.7+0.2 lies beyond the tangent point, with the kinematic distance in the range $13.7<d<23\\kpc$  (Caswell et al. 1975). \\oh\\ masers have been recently found interior to G349.7+0.2 with radial velocities $\\sim +16 {\\rm\\ km\\s}^{-1}$ indicating a kinematic distance $d \\approx 22.4$~kpc (Frail et al. 1996), for which the SNR radius is 6.4~kpc. This maser emission, detected in about 10\\% of the Galactic SNRs, has been recognized as a signpost of SNR interactions with molecular clouds (see Koralesky et al. 1998 and references therein; also Lockett, Gauthier \\& Elitzur 1999). The masers are located along the radio ridge extending from the southern emission peak to the north. CO observations toward G349.7+0.2 revealed a molecular cloud associated with the SNR, and with the masers (Reynoso \\& Mangum 2000, Reynoso \\& Mangum 2001) which have been produced by shocks from the expanding SNR blast wave (Lazendic et al. 2002).  \\begin{inlinefigure} \\includegraphics[width=1.2\\linewidth]{f1.ps} \\caption{ The broad band (1.0-10.0 keV) {\\em ASCA} SIS image of \\snr. Contours are from the ATCA 18--cm continuum image, with levels at 10, 17, 33, 66, 132, 198 and 264 mJ\\,beam$^{-1}$. Crosses mark \\oh\\ maser positions. Comparison of \\emph{Chandra} and \\emph{ASCA} } \\vspace{0.1in} \\label{fig-asca} \\end{inlinefigure}  G349.7+0.2 went undetected in the \\rosat\\ all--sky survey. As we show below, this is the result of the high interstellar absorption of soft \\xray\\ photons. The remnant was detected in the \\asca\\ Galactic plane survey (Yamauchi et al. 1998), and we have followed up this detection with a 60 ks observation that was obtained in two pointings. ",
        "conclusions": "The \\asca\\ SIS and GIS spectra of the SNR G349.7+0.2 are well described by a model for a somewhat underionized, optically thin, thermal plasma. The spectral analysis confirms that it is a distant, young SNR that has swept up considerable interstellar material, consistent with the picture that \\snr\\ is interacting with an adjacent molecular cloud that is observed in CO. Velocity measures based on OH masers produced by this cloud interaction yield a distance to the remnant, based upon which we conclude that \\snr\\ is one of the most X-ray luminous shell-type SNRs known in the Galaxy. Combined with its high radio luminosity, this \\snr\\ appears to be a young analog of the bright SNRs observed in starburst galaxies, many of which have evolved in such high density/pressure regions that they have already become radiative and X-ray faint. We note that N132D, a similar SNR in the LMC, shows evidence for ejecta in its X-ray emission. Given the relatively young age, it may be that such a component persists in G349.7+0.2 as well, but is dominated by the emission from swept up material. The high column density to the SNR rules out any opportunity to detect oxygen ejecta, but evidence for Fe-K ejecta is also seen in N132D, and could be observable in G349.7+0.2 with higher sensitivity measurements. Future high resolution X-ray studies, already planned, offer the possibility to search for such a component."
    },
    "0208/astro-ph0208320_arXiv.txt": {
        "abstract": "After estimating and removing all ejection-related Doppler components from the redshifts of the QSOs near NGC 1068, the remaining redshift is assumed to be intrinsic. This well-studied case is the first example in which it has finally been possible to separate the intrinsic redshift component from the cosmological and other Doppler components. It is shown that this leads to intrinsic redshift components that occur at exact multiples of z = 0.062 and are defined by the relation z$_{\\rm i}$ = 0.062[10$N - M]$ where $N$ and $M$ can have only certain discrete values. It is also shown that the intrinsic redshifts given by this relation agree closely with the redshifts found to be preferred in analyses of quasar emission-line redshifts. It is now immediately apparent that the peaks in the distribution of quasar emission line redshifts also occur at pure harmonics of z = 0.062. Although these peaks have previously been fitted by a log(1+z) relation, the above explanation now seems more appropriate. These results indicate that the $\\Delta$z = 0.062 redshift interval, long known to be present in QSO emission-line redshifts below z = 0.6, is also present in an intrinsic component of QSO redshifts up to z $\\sim$ 2. This appears to imply that the $\\Delta$z = 0.062 redshift interval represents a fundamental redshift spacing that arises from the very nature of the intrinsic redshifts. ",
        "introduction": "It has been shown \\citep{bel02a,bel02b} from the redshifts and positions of 12 of the compact, high-redshift objects near NGC 1068 that they appear to have been ejected outwards along the rotation axis of NGC 1068 in four similarly structured triplets that differ only in their size, rotation angle and rotation axis tip angle. In the previous studies \\citep{bel02a,bel02b} no attempt was made to remove the redshift components due to the triplet ejection velocities. Assuming here that the model is correct, it is shown that it is possible to determine the rotation axis tip angle for each triplet at the time of its ejection. All ejection-related Doppler components can then be removed allowing the residual intrinsic redshift components to be obtained. This is the first instance where an attempt has been made to derive accurate values for the intrinsic components. Previously, they have always been contaminated with cosmological and other Doppler-related components. It is shown further that the intrinsic redshift values derived for the sources near NGC 1068 can be used to define a general relation, and that this relation, in turn, predicts the values found to be preferred in emission line redshift distributions covering redshifts from z = 0.06 to z = 2.   \\begin{figure} \\hspace{-1.0cm} \\vspace{-2.0cm} \\epsscale{1.0} \\plotone{f1.eps} \\caption{\\scriptsize{Source triplets near NGC 1068 are defined by the four ellipses labelled A - D. Objects designated as pairs are joined by a solid line. Associated singles are connected by dashed lines. The location of NGC 1068 is shown by the open square. The rotation axis of the central torus is indicated by the vertical dashed line through the galaxy. Single and pair midpoints are joined by heavy solid lines. \\label{fig1}}} \\end{figure}   \\begin{figure} \\hspace{-1.0cm} \\vspace{-2.0cm} \\epsscale{1.0} \\plotone{f2.eps}\\caption{\\scriptsize{Age versus rotation angle. \\label{fig2}}} \\end{figure} ",
        "conclusions": "By removing all ejection-related Doppler components from the redshifts of the compact QSOs near NGC 1068 it has been shown that the remaining intrinsic components can be defined by the relation z$_{\\rm i}$ = $0.062[10N - M]$ where $N$ = 2, $M = [n(n+1)/2]$ and $n$ = 0,1,2,3,4,5. This is the first instance in which it has been possible to separate out the intrinsic component. It is shown that this same relation will fit the results obtained by \\citet{bur90} for redshifts between z = 0.06 and z = 0.6, if $N$ = 1 and $M = n$, where $n$ = 0,1,2,... 9. The $M$-values for the $N$ = 3 state follow directly from these two states and it is shown that together these three states predict intrinsic redshifts that agree closely with those values found to be preferred in emission-line redshift analyses from z = 0.06 to z = 2. These results indicate that the $\\Delta$z = 0.062 redshift interval may thus represent a fundamental redshift spacing that arises from the very nature of the intrinsic redshifts.  Because this relation includes a constant (z$_{\\rm gmax}$ = 0.62) which can be related to gravitational redshifts, the intrinsic components may have a gravitational origin. If so, this increment may be evidence for some form of gravitational quantization.  It is concluded here that the log(1+z) relation, previously fitted to the peaks in the emission-line redshift distribution, may not be the appropriate one to describe the intrinsic redshifts.  By fine-tuning the tip angles it has also been possible to predict accurate redshift values for sources 10 and 12. It is important to note that values of the Doppler components calculated here could change if the redshifts of sources 10 and 12 can be measured and turn out to be significantly different from those estimated above. Small differences might still be accommodated by adjusting the relevant tip angle, provided the age versus rotation angle relation in Fig. 2 is not destroyed. However, the excellent agreement between the intrinsic redshifts given by eqn 1, and the redshift peaks found in quasar emission-line analyses, strongly suggests that the above relation is a significant one."
    },
    "0208/astro-ph0208006_arXiv.txt": {
        "abstract": "This paper describes the search for astronomical sources of high-energy neutrinos using the AMANDA-B10 detector, an array of 302 photomultiplier tubes, used for the detection of Cherenkov light from upward traveling neutrino-induced muons, buried deep in ice at the South Pole. The absolute pointing accuracy and angular resolution were studied by using coincident events between the AMANDA detector and two independent telescopes on the surface, the GASP air Cherenkov telescope and the SPASE extensive air shower array. Using data collected from April to October of 1997 (130.1 days of livetime), a general survey of the northern hemisphere revealed no statistically significant excess of events from any direction. The sensitivity for a flux of muon neutrinos is based on the effective detection area for through-going muons.  Averaged over the Northern sky, the effective detection area exceeds 10,000~m$^2$ for $E_{\\mu}\\approx10$~TeV. Neutrinos generated in the atmosphere by cosmic ray interactions were used to verify the predicted performance of the detector. For a source with a differential energy spectrum proportional to $E_{\\nu}^{-2}$ and declination larger than +40$^{\\circ}$, we obtain $E^2(dN_{\\nu}/dE)\\leq 10^{-6}$~GeVcm$^{-2}$s$^{-1}$ for an energy threshold of 10 GeV. ",
        "introduction": "\\label{intro} \\setcounter{footnote}{0} Nature provides precious few information carriers from the deep recesses of space, and it is imperative to develop techniques to exploit each one. Throughout history, the photon messenger has made vital contributions to the understanding of the observable Universe. In this paper, we present results from a new generation of telescope designed to detect a very different kind of information carrier, high energy neutrinos (where $E_{\\nu}>$ 1 TeV).  The search for astronomical sources of high energy neutrinos is one of the central missions of the Antarctic Muon and Neutrino Detector Array (AMANDA) \\citep{Aman99}. In this paper, we describe a general search for continuous\\footnote{Although the flux limits reported in this paper are computed assuming continuous emission, upper bounds could be generated for periodic or episodic emission as well.} emission from a spatially localized direction in the northern sky, restricted to declinations greater than +5$^{\\circ}$. The search technique is conceptually simple: a point source would be identified by a statistically significant enhancement over expected background fluctuations from a particular direction.  Expected background is readily obtained experimentally from off-source sky bins within the same band of declination. In contrast, unresolved, or diffuse signals, are characterized by an isotropic distribution and backgrounds are estimated by detector simulation programs. The most favorable flux predictions for point sources are several orders of magnitude lower than the most optimistic predictions for diffusely distributed sources. However, atmospheric neutrino background is diffusely distributed as well, so the level of intrinsic background in the diffuse search is also several orders of magnitude higher. While signal-to-noise considerations favor the search for diffuse emission over point source searches, the interpretation of a diffusely distributed signal is more ambiguous. Thus, the search for point sources complements the search for diffuse sources. The latter search is described in \\citet{Hill}. The more specific searches for point emission from Gamma Ray Bursters \\citep{AmanGRB} and quasi-pointlike emission from galactic dark matter trapped in the core of the earth \\citep{AmanWIMP} are presented in separate papers since those analyses were optimized for different flux spectra and different background characteristics. ",
        "conclusions": "\\label{discussion}  The previous sections have shown that AMANDA-B10 has unprecedented sensitivity to high energy neutrinos and possesses the necessary angular response and background rejection to search for point emission of these particles from astronomical objects; \\textit{i.e.,} it is a novel telescope that detects the neutrino messenger. The sensitivity and angular response were determined by simulation.  The reliability of these programs was established by utilizing the known signals generated by (downgoing) atmospheric muons and (upgoing) atmospheric neutrinos.  The angular response was confirmed by the study of air shower events that triggered both AMANDA-B10 and SPASE. Systematic uncertainty in the analysis procedure was also addressed.  The search for point sources of high-energy neutrinos revealed no candidates. A set of event selection criteria was determined by optimizing the signal to noise ratio for a signal with a hard energy spectrum, yet this analysis retains reasonable sensitivity for softer spectra. The upper limits on muon flux for all search bins in the northern hemisphere are presented in Table~\\ref{tb:allskyflux}.  The neutrino flux limits in Fig.~\\ref{fig:nulimit} are inferred from the assumption of a power-law energy spectrum. This procedure is reliable if the mean energy of the neutrino-induced muon is compatible with the energy response of the detector. For example, Fig.~\\ref{fig:mu_energy} shows that $E_{\\mu}$ at the detector brackets the interval between 0.1~TeV and $10^3$~TeV for source spectra proportional to $E^{-2}$. Two lines of evidence show that the simulated energy response of the detector is valid over this interval. First, the agreement between the detected and expected rates of atmospheric neutrinos shows that the response of AMANDA is being correctly modeled in the sub-TeV region. Second, the tails of the \\nch\\ distribution are sensitive to brighter events within AMANDA, which are roughly equivalent to single muons with energy above 1~TeV. We know of no reason to doubt the predicted energy response for $E_{\\mu}<10^{3}$ TeV. Evaluation and calibration of the energy response beyond $10^{3}$ TeV remain an ongoing activity.  Not all model predictions are well characterized by power-law energy spectra. Therefore, Table \\ref{table:limit} shows the results for a selection of models in the literature. The inferred limits on neutrino flux apply to point sources with continuous emission (or episodic emission averaged over the time interval of data collection) and power-law energy spectra with a fixed spectral index. The limits presented here for sources at large positive declination complement existing data, so that comparable limits now exist for the entire sky.  During 1997, the TeV gamma-ray emission of two nearby AGN blazars (Markarian 421 and 501) were observed to exhibit episodic flaring. If neutrino emission follows the same time variability, then it may be possible to improve the signal-to-noise ratio by eliminating the periods of relatively low output. Multiple detection of Mkn~501 from several air Cherenkov instruments allowed nearly continuous monitoring, including periods when the moon was shining. However, monitoring by multiple instruments only extended from March to late August. Due to uncertainties in the details of the time dependence of the gamma emission, \\textit{neutrino} flux limits are not greatly improved by restricting the analysis to high-flux periods of gamma-ray emission.  While this paper describes an analysis dedicated to the search for point sources, another strategy was developed based on the event selection of the atmospheric neutrino analysis \\citep{Biron02}. The results of this complementary analysis are consistent with the results presented here. The absolute efficiency was extracted by comparing to the known flux from atmospheric neutrinos. Moreover, the second analysis was subject to different systematic uncertainties.  The method based on the atmospheric neutrino analysis retained a smaller event sample of 369 events, of which $\\sim270$ are expected from atmospheric neutrinos. The cut selections produce an implicit optimization on more vertical events and/or softer energy spectra. Figure~\\ref{fig:biron_comp} compares the average effective area of the two analysis for an assumed differential spectra proportional to $E^{-2}$. The best flux limits for soft spectra are obtained by atmospheric neutrino analysis, but the neutrino and muon flux limits for either analysis are much larger than obtained for an assumed power law of index of -2.0.  While the flux limits for any particular source or direction in the northern hemisphere can be extracted from this analysis (see Table~\\ref{tb:allskyflux}), flux limits -both integral and pseudo-differential- for a pre-selected list of 62 sources have been reported \\citep{Biron02}.  These include all known TeV gamma ray blazars, nearby QSOs, and galactic TeV gamma ray sources in the northern hemisphere. The list also includes microquasars, the five most luminous AGNs in wavelength bands that span across MeV, X-ray, infrared, and radio bands. Of particular interest are radio galaxies with strong emission at GHz frequencies. We have also investigated BL Lacs that are close to the arrival directions of the very highest energy cosmic rays \\citep{Tinyakov} and the 10 reported cosmic ray doublets at extreme energies \\citep{Uchihori}."
    },
    "0208/astro-ph0208189_arXiv.txt": {
        "abstract": "We investigate the possibility that present-day galactic haloes contain a population of massive black holes (MBHs) that form by hierarchical merging of the black hole remnants of the first stars. Some of the MBHs may be large enough or close enough to the centre of the galactic host that they merge within a Hubble time. We estimate to what extent this process could contribute to the mass of the super-massive black holes (SMBHs) observed in galactic centres today. Many MBHs will not reach the centre of the main halo, however, but continue to orbit within satellite subhaloes. Using a semi-analytical approach that explicitly accounts for dynamical friction, tidal disruption and encounters with the galactic disk, we follow the dynamics of the satellites and their MBHs and determine the abundance and distribution of MBHs in present-day haloes of various masses. Considering two different accretion scenarios we also compute the bolometric luminosity function for the MBHs. ",
        "introduction": "The presence of super-massive black holes (SMBHs) at the centres of most galaxies appears by now firmly established. SMBHs have estimated masses in the range $10^6 - 10^9$ ~\\Msun and various correlations have been observed between the mass of SMBHs and properties of the galactic bulge hosting them. The first of these to be established were correlations between the mass of the SMBH, $M_{smbh}$ and the mass or luminosity of the galactic bulge,  $M_{bulge}$ and $L_{bulge}$ respectively \\cite{magorrian98,kormendy00,laor01}. More recently, a tighter correlation was found between $M_{smbh}$ and the bulge velocity dispersion, $\\sigma_{bulge}$ at some fiducial distance from the centre \\cite{gebhardt00,merritt01}. An equally tight correlation has also been determined between $M_{smbh}$ and the bulge's light profile, as parameterised by a shape index, $n$ \\cite{graham01}.  Since these correlations extend well beyond the direct dynamical influence  of the SMBH it seems likely that there is a close link between the formation of SMBHs and the formation of their host galaxy. A recent analysis finds that the masses of SMBHs appear to be correlated with the host circular velocity even beyond the optical radius \\cite{ferrarese02}. If this is confirmed, it indicates that the SMBHs are linked to properties of the host dark matter halo. This would be the strongest hint yet that there must be a hierarchical merging component to the growth of SMBHs, since the properties of halos are primarily determined in the context of their hierarchical build up.  Most models put forward to account for the correlations assume a close link between galaxy and SMBH formation as a starting point, although they subsequently proceed along either or both of two routes to explain how the SMBHs grow in mass. One is to consider that the SMBH mass increases mainly by the merging of smaller precursors. This requires SMBH precursors to have been present in galaxies from very early on \\cite{madau01,menou01,schneider02} It might allow the observed correlations to be set up over a long period of time with a potentially large number of mergers through the dynamical interactions between the merging galaxies and SMBH precursors. However, BH merging by itself might ultimately be highly inefficient especially for low mass BH binary systems, for which it would be extremely difficult to progress from a mutually bound configuration to the stage where emission of gravitational radiation draws the binary constituents to final coalescence.  Another mechanism considered is growth mainly by gas accretion within the host bulge. In this case a strong non-gravitational interaction between the growing SMBH and the bulge has to be invoked. An example of this is the radiative feedback of an accreting SMBH that changes the gas dynamics in the bulge so as to effectively control its own gas supply and establish a relation between $M_{smbh}$ and $\\sigma_{bulge}$ \\cite{silk98}. A similar route is followed by  models that tie $M_{smbh}$ to the amount and properties of gas in the bulge \\cite{adams01}. The latter itself may depend on the previous merging of the galaxy with others and so provides a way of combining SMBH mass growth through both mergers and accretion \\cite{haehnelt00}.  As an example of the merger-only scenario it has been shown that the merging of the massive black hole (MBH) remnants of the first stars in the Universe could account for the inferred overall abundance of SMBHs today \\cite{schneider02}.  However, gas accretion during the optically bright QSO phase may be able to account for most of the present day SMBH mass density \\cite{yu02b}, although this process alone would probably not allow ordinary stellar mass BHs to become as large as the most massive SMBHs observed today ($M_{smbh} \\grtsim 10^9$ \\Msun) (see e.g. Richstone ~\\ea 1998). Even if stellar mass BHs were accreting at the Eddington limit, there would not be enough time for the required mass increase to occur. The presence of massive BH seeds at prior to the QSO phase and/or subsequent merging of MBHs therefore appears to be necessary.  In this paper we explore this idea further to determine an upper limit on the mass to which SMBHs can grow through mergers of lower mass precursors and more importantly what the implications are for the presence of a remnant population of lower mass MBHs in the galactic halo. In doing so we assume efficient merging between MBHs, but we also consider the effect of relaxing this assumption. As the `seeds' in the merging hierarchy, we consider massive black holes (MBHs) of some mass $M_{seed}$ that are remnants of the first stars in the Universe, forming within high-$\\sigma$ density peaks at redshifts of $z \\sim 24$. We use Monte Carlo merger trees to describe the merging of haloes and then follow the dynamical evolution of merged/accreted satellite haloes and their central MBHs within larger hosts, explicitly accounting for dynamical friction, tidal stripping and disk encounters. A key prediction is that $\\sim 10^3$ MBHs in the mass range $1 - 1000 \\times M_{seed}$  should be present within the galactic halo today as a result of this process.  We start by describing the origin of seed MBHs in section \\ref{sec:mbhorigin}. In section \\ref{sec:mbhmerge}, we explain how the subsequent merging of their haloes could lead to a build-up of a population of MBHs in present-day galactic haloes, as well as contribute to the mass of a central SMBH. Ways of detecting the population of halo MBHs, particularly via their X-ray emission, are described in section \\ref{sec:mbhdetect}. We conclude with a summary of our findings in section \\ref{sec:summary}. ",
        "conclusions": "\\label{sec:summary} We have used a semi-analytical approach to track the merger history of massive black holes and their associated dark matter haloes, as well as the subsequent dynamical evolution of the MBHs within the new merged halo. In particular we have looked at the possibility that MBHs that are the remnants of massive population III stars, forming in low mass haloes at redshifts $z \\sim 24$, could hierarchically build up to contribute to the present-day abundance of central galactic SMBHs. If this is the case then a number of remnant MBHs is expected to orbit inside galactic haloes. Although our analysis has been carried out for one of the currently favoured $\\Lambda$CDM cosmological models, we expect our findings to hold for any model that provides for hierarchical structure formation such as CDM models in general, but notably excluding Warm Dark Matter and other models with a cut-off or discontinuity at some specific scale in their corresponding cosmological  matter power spectrum. \\\\ The main findings of our analysis are: \\begin{enumerate} \\item For Milky-Way sized galaxies, of the order $10^3$ MBHs that have not reached the host centre are expected to orbit within the halo. Around 1/3 of these will be seed mass MBHs, 85 per cent of these MBHs with masses up to $10 \\times M_{\\bh,seed}$.  \\item For a seed MBH mass of $260$ $(1300)$ \\Msun ~some 5 to 8 (2 to 3) MBHs with masses above $10^5$ $(10^6)$ \\Msun ~are expected in the halo of Milky-Way sized galaxies.  \\item Hierarchical merging of seed MBHs with masses of $M_{\\bh} \\sim 10^3$ \\Msun forming in haloes collapsing from $3 \\sigma$ peaks in the matter density field at $z \\sim 24$ can contribute up to 10 per cent to the present-day mass density contained in SMBH. Another mechanism for the SMBH to gain mass, such as gas accretion, appears inevitable.  \\item Depending on the size of a baryonic core remnant around the MBHs, they could be significant sources of X-rays and possibly account for the ultra-luminous off-centre X-ray sources that have been found in a number of galaxies. Accretion from the host ISM is probably not important. \\end{enumerate} We find that the mass functions for all seed MBH masses considered are essentially the same and only shifted along the mass axis proportional to the mass of the seed MBHs. This is because it is the mass of the satellite haloes and not that of the MBHs that dominates their dynamical evolution in a host halo.  Our findings are consistent with the results of another recent investigation by Volonteri, Haardt \\& Madau (2002). We find the total mass density in a Milky-Way sized galactic halo is about a factor 10 higher than their value inferred from the density function of `wandering' BHs in galactic haloes. This is what we would expect on the basis of the difference in seed BH masses and height of the peaks where initial collapse occurred (their $3.5 \\sigma$ vs our $3 \\sigma$).  Furthermore, we find the mass of a central SMBH in a Milky-Way sized halo to be $1.7 \\times 10^6$ \\Msun. Accounting for the difference in seed MBH masses used this agrees to within a factor 2 with the central SMBH mass of $\\sim 5\\times 10^5$ \\Msun for a galaxy sized halo ($\\sigma \\sim 100 {\\rm ~kms}^{-1}$) as implied by their $M_{\\bh} -\\sigma$ relation (with no gas accretion). This also happens to coincide with the mass determined for the SMBH in the Milky Way, although the Milky Way SMBH is known to lie significantly below the observed $M_{\\BH} - \\sigma $ relation.  However, the slightly non-linear $M_{\\BH} - M_{bulge}$ correlation corresponds to a $M_{\\BH} - \\sigma $ relation whose logarithmic slope ($\\sim 4.0$) does not match the much flatter one they determined ($\\sim 2.9$) for $3 \\sigma$ collapse and no gas accretion. We believe this to be primarily a result of the different assumptions made about the MBH merger process. In particular the inclusion of triple BH interactions and sling-shot ejections, that they find, would probably lead to even lower central SMBH masses in our analysis.  While the fiducial model of Volonteri ~\\ea is based on the collapse of $3.5 \\sigma$ peaks and a seed MBH mass of 150 \\Msun, we consider $3 \\sigma$ peaks and a higher mass for the seed MBHs, both of which imply a higher mass density in MBHs at high redshift. This in turn means that less gas accretion onto MBHs is needed to match the $M_{\\BH} - M_{bulge}$ that is actually observed in nearby galaxies. In any case even for our lightest seed MBH mass considered, any resulting central galactic SMBH in our analysis would need to accrete at least 50-100 times its own initial mass to match the observed relation. This is in accord with a number of studies (see e.g. Yu \\& Tremaine 2002 and references therein) that find the present day SMBH mass density to be consistent with the amount of gas accreted during the optically bright QSO phase. On the other hand gas accretion during the QSO phase alone cannot explain growth from a stellar mass BHs to the most massive SMBHs ($\\ge 10^9$ \\Msun). Even if stellar mass BHs are accreting at he Eddington limit, the QSO phase would not last long enough to accommodate the required number of e-folding times for the BHs to grow to SMBH size. The need for intermediate mass seed BHs and/or some merging of MBHs/SMBHs is therefore necessary to explain the presence of the most massive SMBHs.  Our numerical results depend on a number of parameters that are not yet well constrained, notably the exact height of the fluctuations in the matter density field that are supposed to collapse to form the first baryonic objects and the initial mass function of metal poor stars forming inside these. While the former could possibly be determined better by improved numerical simulations, we have shown that, particularly for the abundance of MBHs in the halo, our results hold qualitatively for a wide range of different  IMFs.  If the halo MBHs could be uniquely identified by their X-ray emission or otherwise, then within the context of our model they could also be used to tag (remnants of) substructure orbiting in a galactic halo. In this way they would complement counts and location  of dwarfs and star clusters as measures of substructure in the galaxy and the halo.  Our results for the growth and present-day mass of the central SMBHs do depend sensitively on how efficiently MBHs merge at the host centre. Here we have taken the view that during major mergers any MBHs orbiting within the core region of the host will be dragged towards the central SMBH quickly, aided by the massive inflow of gas. Due to the increased non-homogeneity, violent dynamical evolution and departure from spherical symmetry during this phase, analytical estimates of dynamical time scales presumably overestimate the time required for MBHs to travel to the centre.  However, a more detailed analysis of this process will be required for the calculation of event rates of mergers between central and inspiralling MBHs and the accompanying gravitational wave emission. \\vspace{0.5cm} \\\\ The authors wish to thank R. Bandyopadhyay and G. Bryan for useful discussions and are grateful to the referee for a number of helpful comments. RRI acknowledges support from Oxford University and St Cross College, Oxford. JET acknowledges support from the Leverhulme Trust."
    },
    "0208/astro-ph0208376_arXiv.txt": {
        "abstract": "{A systematic search of the 2MASS point source catalog, covering 47\\% of the sky, was carried out aiming to reveal any hidden globular clusters in our Galaxy. Eight new star clusters were discovered by a search algorithm based on finding peaks in the apparent stellar surface density, and a visual inspection of their vicinities yielded additional two. They all are concentrated toward the Galactic plane and are hidden behind up to ${\\rm A}_V$=20 mag which accounts for their late discovery. The majority of new clusters are associated with H{\\sc ii} regions or unidentified {\\it IRAS} sources suggesting that they are young, probably similar to Arches or open clusters. Only one candidate has morphology similar to a globular cluster and the verification of its nature will require deeper observations with higher angular resolution than the 2MASS data. } ",
        "introduction": "There are about 150 known Galactic globular clusters (GC hereafter; Harris \\cite{har96}). The majority of them were discovered through optical searches, biased against highly obscured objects. Since the Galaxy is estimated to have 160$\\pm20$ GCs (Harris \\cite{har91}), a certain number of GCs may still be hidden behind the Galactic disk.  The Two Micron All Sky Survey (2MASS) offers an opportunity to carry out a systematic and unbiased search for missing GCs because it covers in an uniform way the Galactic plane in near infrared wavelengths $(J, H$ and $K_S$ bands) where the extinction is almost ten times smaller in comparison with the optical part of the spectrum (Bessell \\& Brett \\cite{bes88}; used throughout this letter). Using the 2MASS data base Hurt et al. (\\cite{hur00}) found two new GCs: 2MASS GC01 and 2MASS GC02 (see also Ivanov et al. \\cite{iva00}). Later Dutra \\& Bica (\\cite{dut00}, \\cite{dut01}) presented a sample of about 90 new infrared star clusters, stellar groups and candidates, mostly discovered from visual inspection of the 2MASS images. Recently, Reyl\\'{e} \\& Robin (\\cite{rey02}) reported two new clusters discovered with DENIS. They applied a combined surface density--integrated flux--color criterion, that detected in addition 22 known clusters.  We report the first results of a systematic and objective search of new clusters in the currently released part of the 2MASS point--source catalog, covering 47\\% of the sky. We also give a short description of the technique, used to locate cluster candidates. The list of objects presented here is not aimed to be complete in any sense. ",
        "conclusions": "\\subsection{Cluster Parameters}  The search yielded 247 candidates that satisfied the $3\\sigma$ and 50 stars excess criteria described in the previous section. Of those, 105 were known clusters, present in SIMBAD. Incidentally, 2MASS GC01 was rejected based on insignificant peak $(2.5\\sigma),$ while 2MASS GC02 was not present in the released point source catalog. We inspected visually the 2MASS images of the remaining candidates, and found two more objects. No obvious objects were present in 134 cases. The basic data for the new clusters is given in Table~\\ref{TblCandidates}. A mosaic of true color images of nine clusters is shown in Figure~\\ref{Fig_CC01}, constructed from the 2MASS $JHK_S$ images. It is not surprising that all candidates are situated close to the Galactic plane. This region suffers from the highest extinction which makes it easy to hide unknown clusters. All candidates are at least partially resolved, and many of them show extended emission, that might indicate ionized gas or faint population, unreachable with the 2MASS data.  \\begin{table}[t] \\begin{center} \\caption{Parameters of the cluster candidates. The first eight objects were identified by the automatic algorithm, and the last two were found after a visual inspection. See Sec.~\\ref{IndivObj} for comments on individual objects.} \\label{TblCandidates} \\begin{tabular}{l@{}c@{ }c@{}c@{}c@{}c} \\hline \\multicolumn{1}{c}{ID} & \\multicolumn{1}{c}{R.A.  Dec.} & \\multicolumn{1}{c}{{\\it l   b}} & \\multicolumn{1}{c}{D} & \\multicolumn{1}{c}{$K_S$,$J$-$K_S$,} & \\multicolumn{1}{c}{$A_V$} \\\\ \\multicolumn{1}{c}{CC} & \\multicolumn{1}{c}{(J2000.0)} & \\multicolumn{1}{c}{} & \\multicolumn{1}{c}{$\\arcmin$} & \\multicolumn{1}{c}{$H$-$K_S$} & \\multicolumn{1}{c}{mag} \\\\ \\multicolumn{1}{c}{} & \\multicolumn{1}{c}{} & \\multicolumn{1}{c}{} & \\multicolumn{1}{c}{} & \\multicolumn{1}{c}{mag} & \\multicolumn{1}{c}{} \\\\ \\hline 01 & 05:13:26 $+$37:27.0 & 169.19 $-$0.90 & 3.0 & 7.5 0.5 0.6 & 6-13 \\\\ 02 & 06:15:53 $+$14:16.0 & 196.21 $-$1.20 & 2.0 & 6.3 1.2 1.0 & 6-13 \\\\ 03 & 06:59:14 $-$03:55.0 & 217.30 $-$0.05 & 2.8 & 6.2 0.8 0.1 & 6-13 \\\\ 04 & 07:00:32 $-$08:52.0 & 221.85 $-$2.03 & 4.0 & 7.5 1.8 0.9 & 9-17 \\\\ 05 & 07:00:51 $-$08:56.5 & 221.96 $-$1.99 & 2.4 & 6.8 0.3 0.0 & 9-17 \\\\ 06 & 07:24:14 $-$24:38.0 & 238.48 $-$4.28 & 4.5 & 6.5 0.8 0.7 & 9-17 \\\\ 07 & 07:30:40 $-$15:18.0 & 230.98 $+$1.49 & 2.8 & 6.1 0.7 0.6 & 9-17 \\\\ 08 & 08:19:10 $-$35:39.0 & 254.01 $+$0.25 & 2.8 & 6.3 0.5 0.7 & 4-12 \\\\ \\hline 09 & 06:59:43 $-$04:04.0 & 217.49 $-$0.02 & 1.0 & ...         & 4-12 \\\\ 10 & 08:18:28 $-$35:47.5 & 254.05 $+$0.05 & 0.5 & 9.7 2.0 0.8 & 12-20 \\\\ & & & & 10.3 2.4 1.0 & \\\\ \\hline \\end{tabular} \\end{center} \\end{table}  \\begin{figure} \\caption{Mosaic of true color images for nine of the new clusters listed in Table~\\ref{TblCandidates}. The top left is CC\\,01, the numbers increase from left to right, and toward the lower rows. CC\\,09 is skipped. Blue is $J$, green is $H$, and red is $K_S$. North is up and East is to the left. The individual image are 4.8$\\times$4.8 arcmin.} \\label{Fig_CC01} \\end{figure}  The cluster coordinates are impossible to estimate by fitting of radial profiles because of the relatively low number of members. This forced us to apply an alternative method for finding the centers. We adopted a trial center, and minimized the sum of the distances to the point sources within 1.5 arcmin. Whenever possible, obvious non-members were excluded based on the color-magnitude diagrams. Then we moved the trial center in a rectangular grid pattern across the face of the cluster. The coordinates, presented here are accurate within 30 arcsec, and agree within this uncertainty with visual estimates. In both cases we used 2MASS world coordinates, from the point source catalog, or from the image header, respectively. The diameters were determined after visual inspection of the K-band images. They should be considered lower limits because some fainter stars may well be bellow the detection limit of the 2MASS atlas. The total magnitudes are obtained with aperture measurements, with diameters 100-180 arcsec. To remove the contribution from the foreground stars we subtracted the flux measured through the same aperture near the clusters from the flux measured at the cluster position. The large variations of the extinction and the apparent stellar density lead to errors of about 0.5 mag.  We carried out a thorough search for known objects near the new clusters. The majority of them were found to be associated with H{\\sc ii} ionized regions indicating that they may be young clusters, perhaps similar to the recently discovered Arches cluster. The SIMBAD identifications are discussed in Sec.~\\ref{IndivObj}. The most promising candidate for an unknown globular cluster is CC01, based on its appearance. CC08 and CC10 possess morphologies typical of open clusters.  To verify further the nature of our candidates we constructed luminosity functions (LF hereafter) of the areas near the stellar surface density peaks, and compared them with LFs of circular regions with the same areas, well away from the objects. An example is shown in Figure~\\ref{FigHist}. The excesses of stars at the alleged cluster positions are obvious despite of the low statistics of individual bins. The color-magnitude (CMD hereafter) and color-color diagrams provide an additional test of our candidates. Figure~\\ref{FigCMD} shows the CMDs of CC01 and CC04. An excess of stars is evident in both cases. However, these diagrams are of little help for unobscured or lightly obscured clusters.  \\begin{figure} \\resizebox{\\hsize}{!}{\\includegraphics{h02.eps}} \\caption{Example of luminosity functions of cluster stars. The top panel is CC01. The histogram, shaded top left to bottom right refers to the stars within 3 arcmin from the cluster center. The histogram, shaded from top right to lower left includes stars from a circular annulus at about 10 arcmin from the cluster center. Both regions have the same area. The bottom panel is for CC08, the central radius is 1.5 arcmin, and the annulus is at about 5 arcmin.} \\label{FigHist} \\end{figure}  \\begin{figure} \\resizebox{\\hsize}{!}{\\includegraphics{cmd04.eps}} \\caption{Color-magnitude diagrams for CC01 (left) and CC08 (right). The top panels show all stars within 3 and 1.5 arcmin from the cluster centers, respectively. The lower panels show stars in annuli with the same areas as the central regions, at about 10 and 5 arcmin from the cluster centers, respectively. The excess of red cluster members is evident in both cases. A reddening vector, corresponding to visual extinction $\\rm A_V$ = 10 mag is also shown.} \\label{FigCMD} \\end{figure}  The estimate of the extinction toward the clusters is a problem on its own. It is impossible to determine whether we see the tip of the main sequence or the the tip of the red giant or red supergiant branch, given the relatively shallow depth of the 2MASS point source photometry. We adopted a simple foreground screen, although it is clear that some of the clusters suffer from differential reddening. Next, we made two extinction estimates, assuming that the reddest sequence is either at $J-K_S\\sim0$ mag, typical for the blue stars, or at 0.9-1 mag, which is the usual color of the red supergiant and red giant branches. We neglected completely any metallicity effects because they are much smaller than the uncertainties in the $E(J-K_S)$ which are of order of 0.5 mag, leading to errors of $\\sim4$ mag in $A_V$. The results are included in Table~\\ref{TblCandidates}. Using the stellar colors to estimate the extinction toward clusters inside our Galaxy is more reliable than far-infrared techniques (i.e. Dutra \\& Bica \\cite{dut00}), because the background dust emission can easily be confused for emission from foreground obscuring material.  \\subsection{Comments on Individual Objects\\label{IndivObj}}  {\\bf - CC01} A rich cluster, well resolved by the 2MASS. It resembles morphologically a globular or a compact young cluster, similar to Arches. Further observations are needed to solve this ambiguity. The location of the candidate coincides with an emission nebula Min\\,2-58 (Minkowski \\cite{min48}), clearly seen on both blue and red DSS images (while the stellar cluster is not). A radio-selected H{\\sc ii} region within 30 arcsec was reported by Lockman (\\cite{loc89}).\\\\ {\\bf - CC02} Resolved by the 2MASS. Extended nebulocity is present on both blue and red DSS images. There is an indication for faint stellar population on the red image. An H{\\sc ii} region SH\\,2-269\\,B (Sharpless \\cite{sha59}) lies within the borders of our candidate. Water maser emission was detected by Comoretto et al. (\\cite{com90}) indicating that this is a young cluster embedded in a dense molecular core.\\\\ {\\bf - CC03} Partially resolved by the 2MASS. Compact (D$\\sim 1$ arcmin) nebulocity is visible on the red DSS image, with no trace of the embedded stars. An H{\\sc ii} region BFS\\,56, associated with a molecular cloud was found at the same position by Blitz, Fich, \\& Stark (\\cite{bli82}). A nearby IRAS source (06567-0350) was identified by Magnier et al. (\\cite{mag99}) as possible young stellar object. Therefore, the found cluster is likely young.\\\\ {\\bf - CC04} Well resolved on the 2MASS images. Extended nebulocity is visible on the red DSS plate, with a few bright stars that might be associated with it. Blitz et al. (\\cite{bli82}) reported an H{\\sc ii} region BFS\\,64 nearby.\\\\ {\\bf - CC05} Partially resolved by the 2MASS. A compact nebulocity (D$\\sim 1$ arcmin) is visible on the red DSS image, with no associated stars. Fich \\& Terebey (\\cite{fic96}) reported a cloud (221.9-2.0B) that might hide on-going star formation. Indeed, a young stellar object CPM\\,33 was discovered by Campbell, Persson, \\& Matthews (\\cite{cam89}), which leads as to believe that this may be a young cluster.\\\\ {\\bf - CC06} Partially resolved by the 2MASS. A patchy nebula with a few stars near the center is visible on the red DSS image. However, their association is not obvious due to heavy foreground contamination. The nebula is included in the H{\\sc ii} region catalog of Brand, Blitz, \\& Wouterloot (\\cite{bra86}) as BRAN\\,22C. CO emission, associated with this object was detected by Brand et al. (\\cite{bra87}), suggesting the presence of a dense core.\\\\ {\\bf - CC07} Well resolved by the 2MASS. A patchy faint nebula with a few stars near the center is visible on the red DSS image. Some of the stars coincide with the brighter patches, indicating a physical association. The nebula is cataloged by Sharpless (\\cite{sha59}) as H{\\sc ii} region SH\\,2-299\\,B.\\\\ {\\bf - CC08} Well resolved on the 2MASS images. It is visible on the red DSS image as a loose concentration of faint stars, indicating even lower extinction than the one estimated from the near-infrared color-magnitude diagram (Figure~\\ref{FigCMD}, left). The morphology suggests that this might be an open cluster.\\\\ {\\bf - CC09} Well resolved by the 2MASS. Visible on the red DSS, suggesting low extinction. We refrained from measuring the cluster brightness because of the faint apparent magnitude and the foreground contamination. Resembles an open cluster or a distant OB association.\\\\ {\\bf - CC10} Partially resolved by the 2MASS. Compact nebulocity (D$\\sim$1 arcmin) visible on the red DSS image. It is associated with IRAS\\,08165$-$3538, and appears to be a compact young cluster that does not match our search criteria for richness. It was observed in the near infrared as an unresolved source by Liseau et al. (\\cite{lis92}) and Lorenzetti et al. (\\cite{lor93}). Their measurements are also listed in Table~\\ref{TblCandidates} (last row), and agree with ours, within the uncertainties."
    },
    "0208/hep-th0208094_arXiv.txt": {
        "abstract": "We investigate light-cone structure on the world-volume of an unstable D-brane with a tachyon decaying inhomogeneously by using a field theoretical description. It is shown that (i) light-cones governing open strings are narrower than those governing closed strings and will eventually collapse inward in all directions except at kinks, where the tachyon remains at the top of its potential; and that (ii) light-cones governing open strings at a kink will be narrowed only in the direction perpendicular to the kink surface. It is also shown that (iii) future-directed light-cones governing open strings near a kink are tilted towards the kink, compared with those governing closed strings. The result (i) implies that open strings except at kinks are redshifted, compared with closed strings, and will eventually cease to be dynamical. On the other hand, the result (ii) shows that open strings on a kink surface can move freely along the kink surface and are dynamical but do not feel the existence of the spatial dimension perpendicular to the kink surface. The result (iii) indicates that open strings near a kink have tendency to move towards the kink. Hence, the light-cone structure vividly illustrates how open strings behave during the dynamical formation of a kink. We also discuss about a possibility that the early universe has a network of various dimensional D-branes, black-branes and tachyon matter. A problem associated with the network and a possible solution to the problem are discussed. ",
        "introduction": "Time-dependent backgrounds are recently getting more important than ever in the study of string theory. Actually, spacelike branes~\\cite{Gutperle-Strominger} or a rolling tachyon~\\cite{Rolling-tachyon} have been attracting a great deal of interest. Some of non-BPS objects such as a D-brane-anti-D-brane pair and a non-BPS D-brane are by their nature unstable and should become time-dependent under perturbations. Since non-BPS objects are expected to play important roles in exploration of string dualities beyond the BPS level, one would like to study their properties including their dynamics. Hence, it is important to study the dynamics of tachyons living on those objects, as their unstable nature is characterized by the existence of tachyons.   It is not only in exploration of the string duality web but also in exploration of the early universe where tachyons can play important roles. The so called tachyon cosmology was recently initiated and investigated by many authors~\\cite{tachyon-cosmology}. In this context a tachyon can dominate our universe at an early epoch and, thus, the dynamics of tachyons is essential.   The dynamics of a tachyon can actually be described in many different ways. For example, it was shown by boundary states of an unstable D-brane that the pressure of tachyon fluid approaches zero as the tachyon rolls down~\\cite{tachyon-matter}, it is a field theoretical description~\\cite{Garousi,BRWEP,Kluson,tachyon-matter} that has been extensively used in the tachyon cosmology, and it was by using the boundary string field theory (BSFT) that some exact properties of the tachyon condensation were derived in refs.~\\cite{Garasimov-Shatashvili,Kutasov-Marino-Moore}.   One of well-known properties of the tachyon effective action for an unstable D-brane is that the pressure at late time falls off exponentially as the tachyon field evolves from any spatially homogeneous initial configuration towards the minimum of the potential. This was shown by the boundary states~\\cite{tachyon-matter}, in the field theoretical description with a specific form of the tachyon potential~\\cite{field-theory}, and in the BSFT~\\cite{Sugimoto-Terashima,Minahan} although there are some differences in details.   Another well-known property is the absence of plane-wave solutions around the minimum of the tachyon effective potential.  This is easily inferred from evidences showing that the minimum describes a configuration without D-branes~\\cite{no-branes}. The absence of plane-wave solutions was shown in the field theoretical description~\\cite{field-theory} and the BSFT~\\cite{Ishida-Uehara}.   Knowing the absence of plane-wave solutions around the minimum, it is interesting to ask how this is achieved from the open string points of view. Before the tachyon decay, open strings are dynamical on the world-volume of the unstable non-BPS brane. On the other hand, after the tachyon decay, there are no open string states. Hence, it seems natural to ask ``what really happens to open strings in the process of the tachyon decay?'' In this respect, a non-perturbative confinement mechanism on the brane~\\cite{Bergman-Hori-Yi} and the classical equivalence between a gauge system and a string fluid~\\cite{Gibbons-Hori-Yi} are very suggestive.   In ref.~\\cite{Sen-kink} Sen conjectured that a tachyonic kink on the world-volume of an unstable D$(p+1)$-brane is a BPS D$p$-brane. In this context, the tachyon field for the kink configuration is usually supposed to be static. On the other hand, in the context of the rolling tachyon or the tachyon cosmology, the tachyon is usually spatially homogeneous or nearly homogeneous. Hence, it is interesting to investigate a time-dependent kink, or a highly inhomogeneous rolling tachyon. It may describe the dynamical formation of the kink as a BPS D-brane and, thus, may provide us with more knowledge about the dynamics of the tachyon. Far from the kink, the tachyon condensation is expected to proceed without being affected by the existence of the kink. Hence, to this region the previous works on the homogeneous rolling tachyon can perhaps be applied. However, it is not clear what dynamically happens to open strings near a kink.    In this paper we give yet another view on the fate of open strings during tachyon decay. This view can be applied to the formation process of a tachyonic kink, or inhomogeneous tachyon decay. Our strategy in this paper is to analyze the light-cone structure on the world-volume of an unstable D-brane by using a field theoretical description. In particular, we compare light-cones governing open strings with those governing closed strings. In this view, it is shown that open strings except at tachyonic kinks are redshifted away. On the other hand, open strings on a kink surface remain dynamical but do not feel the existence of the spatial dimension perpendicular to the kink surface. Moreover, open strings near a kink have tendency to move towards the kink during the decay process.   We finally discuss in the context of the tachyon cosmology about a possibility that the early universe has a network of various dimensional D-branes, black-branes and tachyon matter. A problem associated with the network and a possible solution to the problem are discussed.    This paper is organized as follows. In Sec.~\\ref{sec:action} we review an effective field theoretical description of an unstable D-brane and introduce a metric governing open strings. In Sec.~\\ref{sec:tachyon-decay} we analyze an inhomogeneous tachyon decay, or a time-dependent kink, and the light-cone structure on the world-volume. In Sec.~\\ref{sec:cosmology} we discuss some cosmological implications including a problem associated with a D-brane network in the early universe and a possible solution of it. Sec.~\\ref{eqn:summary} is devoted to a summary of this paper. ",
        "conclusions": "\\label{eqn:summary}  We have investigated light-cone structure on the world-volume of an unstable D-brane with a tachyon decaying inhomogeneously by using a field theoretical description. It has been shown that (i) light-cones governing open strings are narrower than those governing closed strings and will eventually collapse inward in all directions except at kinks, where the tachyon remains at the top of its potential; that (ii) light-cones governing open strings at a kink will be narrowed only in the direction perpendicular to the kink surface; and that (iii) future-directed light-cones governing open strings near a kink are tilted towards the kink, compared with those governing closed strings. The result (i) implies that open strings except at kinks are redshifted, compared with closed strings, and will eventually cease to be dynamical. On the other hand, the result (ii) shows that open strings on a kink surface can move freely along the kink surface and are dynamical but do not feel the existence of the spatial dimension perpendicular to the kink surface. The result (iii) indicates that open strings near a kink have tendency to move towards the kink. In this way, the light-cone structure vividly illustrates how open strings behave during the dynamical formation of a kink.    Finally, we have also discussed about a possibility that the early universe has a network of various dimensional D-branes, black-branes and tachyon matter. An associated problem was pointed out and a possible solution of the problem was proposed."
    },
    "0208/astro-ph0208410_arXiv.txt": {
        "abstract": "s{ We derive cosmological parameters from the CBI measurements of the Cosmic Microwave Background (CMB) angular power spectrum. Our results provide an independent confirmation of the standard $\\Omega_{\\rm tot}=1$ $\\Lambda$CDM model within the adiabatic, inflationary paradigm. Above $\\ell=2000$ the observations show evidence of power in excess of that expected in the standard models. We use hydrodynamical simulations to show how Sunyaev-Zeldovich Effect (SZE) may account for the excess power for models with fluctuation amplitude $\\sigma_8\\sim 1$ which is in the high end of the range allowed by the primary CMB observations.} ",
        "introduction": "Increasingly accurate measurements of the angular power spectrum of the Cosmic Microwave Background (CMB) have begun to constrain cosmological models of structure formation. Previous experiments such as \\Boomerang\\cite{Netterfield01}, \\DASI\\cite{DASI01} and \\Maxima\\cite{Maxima01} have now measured precisely the shape of a first peak at $\\ell \\sim 220$, consistent with an $\\Omega_{tot}=1$ $\\Lambda$CDM universe with adiabatic, inflationary seeded perturbations. A significant detection of a second peak has also been established \\cite{deBernardis01} with evidence for a third. The \\cbi\\ observations have now confirmed another important element of the adiabatic, inflationary paradigm, the damping at high multipoles due to the viscous drag over the finite width of the last scattering surface \\cite{Sievers02}. B. Mason\\cite{Mason02a}, in these proceedings, gives a description of the \\cbi\\ instrumental setup and observing strategy and discusses the power spectra. Here we present the cosmological parameter fits obtained from the observations and discuss the nature of the possible excess observed on the smallest angular scales at $2000<\\ell<4000$.  During the year 2000 observing season, the \\cbi\\ covered three deep fields of diameter roughly $1^\\circ$ \\cite{Mason02a,Mason02b}, and three mosaic regions, each of size roughly 13 square degrees \\cite{Pearson02}. The instrument observes in 10 frequency channels spanning the band $26-36$GHz and measures 78 baselines simultaneously. Our power spectrum estimation pipeline is described in \\cite{Myers02} and involves an optimal compression of the ${\\cal O}(10^5)$ visibility measurements of each fields into a coarse grained lattice of visibility estimators. Known point sources are projected out of the data sets when estimating the primary anisotropy spectrum by using a number of constraint matrices. The positions are obtained from the (1.4GHz) NVSS catalog \\cite{Condon98}. When projecting out the source we use large amplitudes which effectively marginalize over all affected modes. This insures robustness with respect to errors in the assumed fluxes of the sources. The residual contribution of sources below our $S_{1.4}=3.4$ mJy cutoff is treated as a white noise background with an estimated amplitude of $0.08\\pm 0.04$ Jy/sr${^-1}$ \\cite{Mason02a,Mason02b}. ",
        "conclusions": "Our analysis of the \\cbi\\ observations has yielded parameters consistent with the standard $\\Omega_{tot}=1$, $\\Lambda$CDM model. These results, based on measurements extending to much higher $\\ell$ than previous experiments, provide a unique confirmation of the model. The dominant feature in the data is the decline in the power with increasing $\\ell$, a necessary consequence of the paradigm which has now been checked. In summary under weak prior assumptions the combination of \\cbi\\ and \\DMR\\ data gives $\\Omega_{tot} = 1.01_{-0.06}^{+0.09}$, and $n_s = 1.02_{-0.07}^{+0.11}$, consistent with inflationary models; $\\Omega_{\\rm cdm}h^2=0.12\\pm0.03$, and $\\Omega_{\\Lambda} = 0.64_{-0.14}^{+0.11}$. With more restrictive priors, flat+weak-$h$+LSS, are used, we find $\\Omega_{\\rm cdm}h^2 = 0.13_{-0.01}^{+0.02}$, consistent with large scale structure studies; $\\Omega_b h^2 = 0.025_{-0.008}^{+0.010}$, consistent with Big Bang Nucleosynthesis; $\\Omega_m = 0.37\\pm 0.11$, and $\\Omega_b = 0.060\\pm 0.020$, indicating a low matter density universe; $h = 0.65_{-0.12}^{+0.12}$, consistent with the recent determinations of the Hubble Constant based on the recently revised Cepheid period-luminosity law; and $t_0=14.0_{-1.2}^{+1.2}$ Gyr, consistent with cosmological age estimates based on the oldest stars in globular clusters. The combination of CMB measurements and LSS priors also enables us to constrain the normalization $\\sigma_8$. We find that for flat+weak-$h$+LSS priors we obtain $\\sigma_8=0.89_{-0.10}^{+0.14}$. Thus it appears that the normalization required to explain the excess with the SZE is in the upper range of the independent result based on the primary CMB signal and LSS data.   The 2001 observing season data is now being analyzed. Although the data will double the overall integration time it is not expected to increase the confidence of the high-$\\ell$ measurements as the observations were aimed at doubling the mosaic area and not at further integration of the deep fields. Follow-up surveys of the deep fields in the optical range and correlation with existing X-ray catalogs may establish whether the measurement is indeed a serendipitous detection of the SZE and will be part of future work. However, the observations have highlighted the potential for SZE measurements to constrain $\\sigma_8$ via the highly sensitive dependence of the angular power spectrum to the amplitude of the fluctuations ${\\cal C}^{SZ}\\sim\\sigma_8^7$, although precise calibration of the theories from either numerical or analytical methods are required to make such conclusions feasible\\cite{Bond02}. The \\cbi\\ is currently being upgraded with polarization sensitive antennas for the 2002/2003 observing season.  This work was supported by the National Science Foundation under grants AST 94-13935, AST 98-02989, and AST 00-98734. Research in Canada is supported by NSERC and the Canadian Institute for Advanced Research. The computational facilities at Toronto are funded by the Canadian Fund for Innovation.  We are grateful to CONICYT for granting permission to operate the CBI at the Chajnantor Scientific Preserve in Chile.   \\begin{figure} \\begin{center} \\psfig{figure=f1.eps,height=3.1in} \\end{center} \\caption{The SZE powers spectrum for $\\sigma_8=1$. The triangle (blue) points are an optimal combination of all data. The hatched (green) area represents the 2-sigma confidence region for the high-$\\ell$ CBI band. The area bounded by dash-dotted curves (red) cover the results from 20 MMH SZ maps while the area bounded by long-dashed (cyan) line cover the results from the 10 SPH maps. The square (yellow) points are the results of the $\\sigma_8=0.9$ 400 Mpc result scaled to $\\sigma_8=1.0$ using our empirical relation. \\label{fig:szspectrum}} \\end{figure}"
    },
    "0208/astro-ph0208283_arXiv.txt": {
        "abstract": "We have mapped the water maser emission associated with the infrared centers IRS1 and IRS3 of the NGC\\,2071IR star forming region at four epochs over $\\sim 4$ months with the Very Long Baseline Array (VLBA).  We detected 269 maser features with $\\sim 1$\\kms~linewidths and measured 30 proper motions. In each infrared center, the water maser emission appears to trace parts of a protostellar disk and collimated outflow. The disk components are $\\sim 9$ and $\\sim 17$ AU long, in IRS\\,3 and IRS\\,1 respectively, and $\\sim 2$ AU wide. They are identified as disks by their compact size, elongation parallel to the direction of known IR polarization, central location in the maser maps, small internal proper motions, and proximity to $\\lambda1.3$\\,cm continuum emission. The outflows have axes perpendicular to the disks and exhibit proper motions of up to $\\sim 42$\\kms. They are outlined by maser emission up to $\\sim 260$ AU from the protostars. The IRS\\,3 outflow appears to be conical on one side, while the IRS\\,1 outflow comprises a  narrowly collimated bipolar flow surrounded by outward-facing, funnel-shaped cavities. The detection of water maser emission tracing such compact disk components and specifically conical or funnel-shaped structures is unusual.  The fact that the distributions are similar in IRS\\,3 and IRS\\,1 may indicate the two infrared centers are roughly coeval.  NGC\\,2071IR provides a rare opportunity to resolve the structures and dynamics of disks and outflows together, and to do so for two protostars that are only $\\sim 2000$ AU apart (in projection) in a deeply embedded star forming region of intermediate luminosity. ",
        "introduction": "The infrared cluster NGC\\,2071IR lies in the Lynds 1630 dark cloud in Orion at a distance of 390 pc \\citep{anthony-twarog82}.  The $\\sim 30''$ diameter cluster has been resolved into 8 distinct near infrared sources \\citep{walther93} and has a total luminosity of 520\\,L$_{\\odot}$ \\citep{butner90}, which is suggestive of intermediate mass star formation. \\citet{snell86} first detected radio continuum counterparts to the IR sources, specifically IRS1 and IRS3 separated by 6''. While IRS\\,1 dominates the luminosity at near infrared  wavelengths, IRS\\,3 is a significant contributor at longer wavelengths \\citep{snell86,kawabe89}.  NGC\\,2071IR also hosts a well-studied molecular outflow that has been mapped in the emission of CO \\citep{scoville86, moriarty-schieven89, chernin92}, H$_2$ \\citep{garden90,aspin92}, CS \\citep{zhou91,kitamura92}, SO and SiO \\citep{chernin93}, HCO+ \\citep{girart99} and  NH$_{3}$ \\citep{zhou90}. The arcsecond-resolution H$_2$ observations of \\citet{aspin92} indicate IRS\\,1 is the likely source for the large scale outflow while those of \\citet{garden90} show elongated molecular emission associated with IRS\\,3 as well. Images of radio continuum emission, with resolutions as high as $0\\rlap{.}''1$, show elongated emission from thermal jets coincident with both infrared sources \\citep{torrelles98,smith94,snell86}.  Water maser emission ultimately associated with NGC\\,2071IR was detected even before the IR cluster and molecular outflows \\citep{schwartz75,pankonin77,campbell78}.  Both IRS\\,1 and IRS\\,3 host the water maser emission \\citep{tofani95,torrelles98}, which indicates the presence of substantial columns of dust laden, warm (300 - 1000 K), dense ($10^8$ - $10^{10}$ cm$^{-3}$) gas that is probably shock excited \\citep{elitzur92}.  \\citet{torrelles98} observed the masers  using the VLA with resolution of $0\\rlap{.}''1$.  In IRS\\,1, they observed a distribution of masers elongated parallel to the roughly east-west radio jet.  In IRS\\,3, they observed a distribution more or less perpendicular to the jet and suggestive of a disk.  The difference in the structure for the maser sources, supplemented by the higher extinction toward IRS\\,3, was inferred to indicate IRS\\,3 is less evolved.  In general, the study of intermediate and high-mass protostars is difficult because examples are hundreds of parsecs away and are contained within crowded fields.  As a result, even observations with $0\\rlap{.}''1$ resolution can be confusion limited. However, long baseline interferometric observations, with milliarcsecond (mas) resolution,  can be used to map regions unambiguously (and estimate proper motions) when high brightness temperature emission, such as water maser emission, is present \\citep[e.g.,][and references therein]{claussen98,moscadelli00,patel00,furuya00,torrelles01,gwinn92}.  The proper motion study of NGC\\,2071IR presented here concentrates on one of the closest regions of intermediate mass star formation whose structure and dynamics we show have not been fully resolved in previous studies.  In Section 2, we present the observations and calibration, followed by a  description of how we estimated proper motions in Section 3.  In Section 4 we discuss the compact protostellar disk and outflows found in both IRS\\,1 and IRS\\,3, and consider how these structures bear upon other lower angular resolution observations of the whole infrared cluster. Conclusions are presented in Section 5. ",
        "conclusions": "We have detected 269 water maser features and 30 proper motions around IRS\\,3 and IRS\\,1 in NGC\\,2071IR over four epochs with the VLBA.  In both IR sources the maser emission appears to trace a protostellar disk and associated outflow.  The disk components comprise centrally located, elongated clumps of masers, with position angles similar to those of local IR polarization and small internal proper motions.  As well, the disk components lie in close proximity to peaks in the continuum emission from compact radio jets.  In contrast, the outflow components are characterized by larger proper motions directed away from the putative disks.  The structures of the outflows are suggestive of a one-sided conical flow (in IRS\\,3) and a bipolar high-speed flow surrounded by outward-facing, wide-angle cavities (in IRS\\,1).  The limbs of the outflows may mark the interface between outflowing and rotating infalling material.  The gross similarity of structure between IRS\\,1 and IRS\\,3 indicates that they are at similar stages of development.  In both IR sources, intact protostellar disks may act as reservoirs of warm dense gas that supports maser action.  Consequently the inferred disks must be relatively quiescent.  In both IR sources, the outflows are richly populated with water masers, which indicates there is substantial ambient or infalling high-density material broadly distributed within $\\sim 100$ AU of the protostars.   Although the extinction toward IRS\\,3 is larger \\citep{walther93}, the evolution of the embedded protostar therein does not seem markedly less far along.  The detection of water maser emission tracing  compact disk components (10 to 20 AU long and $\\sim 2$ AU wide), conical outflow, and wide-angle cavities is unusual. These detections have required observations with the angular resolutions characteristic of VLBI.  Disks traced by water masers on larger scales have been inferred from   lower angular resolution data \\citep{shepherd99,torrelles97,torrelles98}, but confusion from  blending of maser features is a risk in these cases, and confirmation requires higher angular resolution observations.  The water maser features in IRS\\,3 and IRS\\,1 sample the underlying dense gas incompletely, which limits our inferences. In order to solidify the interpretation renewed long baseline interferometric observations are necessary. Water maser emission in regions of low and intermediate mass star formation is highly time variable \\citep[e.g.,][]{wilking94,torrelles98a}, and the accumulation of maps for many epochs, registered and superposed, will better trace the underlying dense gas structures around the protostars."
    },
    "0208/hep-ph0208125_arXiv.txt": {
        "abstract": "{We reconsider the Standard Model interactions of ultra-high energy neutrinos with matter. The next to leading order QCD corrections are presented for charged-current and neutral-current processes. Contrary to popular expectations, these corrections are found to be quite substantial, especially for very large (anti-) neutrino energies. Hence, they need to be taken into account in any search for new physics effects in high-energy neutrino interactions.  In our extrapolation of the parton densities to kinematical regions as yet unexplored directly in terrestrial accelerators, we are guided by double asymptotic scaling in the large $Q^2$ and small Bjorken $x$ region and to models of saturation in the low $Q^2$ and low $x$ regime. The sizes of the consequent uncertainties are commented upon. We also briefly discuss some variables which are insensitive to higher order QCD corrections and are hence suitable in any search for new physics.}   \\begin{document} ",
        "introduction": "Ultra-high energy neutrinos and their interactions continue to attract much attention. This is in spite of the fact that no such neutrino has been seen so far (and hence bounds been placed on their fluxes~\\cite{flux_limits}). Much of the continuing interest has been occasioned by the observation, in more than one detector, of  ultra-high energy cosmic rays (UHECR). An interaction of such cosmic rays with either the microwave background radiation or even the atmosphere would presumably lead to the generation of charged pions and through their decay, to extremely energetic neutrinos~\\cite{PJ}. Alternatively, {\\em primary} ultra-high energy neutrinos themselves could lead to UHECR, thereby avoiding the GZK bound. A possible source for such primary neutrinos is the decay of an extremely massive primordial relic or even a cosmic string~\\cite{BS}. Whatever their origin, it can safely be asserted that cosmic high energy neutrinos are inextricably linked to the very high energy cosmic rays. The experimental detection of such neutrino fluxes is thus expected to provide rare insight into the origin of such cosmic rays and probably to physics beyond the SM as well. Such observations have the additional promise of probing stellar structures \\cite{stellar_structure}, for unlike charged particles, cosmic neutrinos do not suffer any bending due to inter-galactic magnetic fields and hence arrive on earth in a direct line from their source.  Consequently, various experimental efforts are being planned. Pilot experiments, based on the optical detection of C\\v{e}renkov light emitted by the muons created in charged current reactions of neutrinos with nucleons either in water or in ice, include the Antarctic Muon And Neutrino Detector Array (AMANDA)~\\cite{amanda} in the South Pole ice and the one at Lake Baikal \\cite{baikal}. The next generation experiments using similar techniques comprise the Neutrino Telescope and Abyss environmental RESearch (ANTARES)~\\cite{antares}, the Neutrino Experiment SouthwesT Of GReece (NESTOR) project in the Mediterranean~\\cite{nestor}, as well as ICECUBE~\\cite{icecube}, the proposed kilometer scale version of the AMANDA detector. Recently, arguments have been forwarded in favour of facilities based on the detection of radio pulses emanating from the electromagnetic showers created by neutrino interactions in ice and other materials. The primary advantage of such a technique would be the scalability up to an effective area of $10^4\\,{\\rm km}^2$ and the Radio Ice C\\v{e}renkov Experiment (RICE) experiment at the South Pole~\\cite{rice} is a functioning prototype. It has also been realized that neutrinos can initiate horizontal Extensive Air Showers (EAS) which could be detected by giant ground arrays and fluorescence detectors such as the cosmic ray Pierre Auger Project~\\cite{auger-neut}. Deeply penetrating EAS could also be detected by observing their fluorescence light from space based  instruments such as the Orbiting Wide-angle Light-collector (OWL)~\\cite{owl} and the Extreme Universe Space Observatory (EUSO)~\\cite{euso}. Finally, there is the newly approved balloon experiment ANtartic Impulsive Transient Antenna (ANITA) which will look for radio C\\v{e}renkov pulses created by ultra-high energy neutrino interactions and emanating from very large chunks of the Antarctic ice cap.  Its energy threshold is about $10^{18}$ eV, so it will primarily be looking for GZK neutrinos \\cite{anita}.  These experiments, taken together, are sensitive to neutrino energies of upto $10^{11}$ GeV or so. The actual event rates are somewhat uncertain though, as they depend crucially on both the predicted neutrino fluxes, as well as on the ultra-high energy neutrino cross sections which, for want of a better method, we may only estimate by a reasonable extrapolation beyond the measured regime. The interaction of UHE neutrinos with matter is through deep inelastic scattering of the neutrinos with protons and neutrons. Over the last few years, numerous issues with regard to the nature of the cross section of $\\nu-N$ scattering (where N is a proton or a neutron) have gained importance and some of these are discussed in Refs.\\cite{GQRS, RSSSV, DKRS}.  Most of the discussion on $\\nu-N$ scattering has been based on the leading order (LO) expressions for neutrino nucleon scattering (i.e. $\\as$ independent). The usual procedure followed has been to use the lowest order parton level cross section and convolute it with the LO or sometimes even next to leading order (NLO) parton distributions. QCD corrections to the partonic cross sections have typically been neglected, in view of the high energies involved and the consequent small value of the strong coupling constant $\\as$. While, at first sight, such an approximation may seem appropriate, it must be borne in mind that the consequent uncertainties may limit the sensitivity of neutrino telescopes~\\cite{amanda,antares,nestor,icecube,rice} (and to a smaller extent, the cosmic-ray detectors~\\cite{auger-neut,owl,euso}) to physics beyond the Standard Model. Amongst possible such scenarios, of particular interest are theories with supersymmetry~\\cite{susy}, extended gauge or higgs sector~\\cite{exotic} or, more recently, those with a low energy gravity sector~\\cite{extradim}. Perhaps, of even more importance, are the effects on the determination of neutrino mixing parameters~\\cite{oscill}, and neutrino-tomography of the earth's interior~\\cite{tomography}. The importance of a more accurate estimation of the neutrino interaction rates as well as their kinematical distributions, thus, cannot be overstated.  In this paper, we explicitly calculate the ${\\cal O}(\\as)$ QCD corrections to the parton model result and show that while it is not very large, it is by no means negligible. Moreover, we study carefully the behavior of this correction as a function of neutrino energy and find behavior which is not necessarily very intuitive. For example, there is a delicate interplay between the magnitude of $\\as$, the structure of the higher order integrals and the size of the parton distributions (particularly the gluon) in LO and NLO. This gives a non-trivial energy dependence to the ratio of the LO and NLO cross section (which we will call the $K$-factor). We work throughout in the $\\overline{MS}$ scheme.  Another issue which is of relevance in these energy ranges (and which, again, has been addressed in Refs.\\cite{GQRS, RSSSV, DKRS}) is the question of carrying out perturbative calculations at ultra low Bjorken $x$ (down to $10^{-8}$). No data exists to help in parametrisations of parton distribution functions at such values of $x$ and one can only be guided in these regions by a somewhat improperly understood physical picture of a highly dense nucleon of partons. Data from HERA stops around $x\\simeq 10^{-5}$ or so and below that, some physical picture of shadowing and saturation effects (particularly at low $Q^2$) needs to be incorporated to understand the physics of a nucleon with a high density of partons.  We have tried to address both these issues in this paper. We have explicitly calculated the ${\\cal O}(\\as)$ corrections to the partonic cross sections and convoluted them with appropriate parton distributions. We have also addressed the issue of extrapolation of the partonic distributions to regions where simple DGLAP evolution is not expected to hold.  The paper is organised as follows. In Section 2, we present a discussion and justification for the various partonic distributions that we have used in various parts of the $(Q^2,x)$ plane. In Section 3, we present detailed expressions for the ${\\cal O}(\\as)$ corrections to the lowest order partonic cross section for neutrino and antineutrinos scattering against an isoscalar target. These expression are, of course available elsewhere but for the sake of completeness and clarity, we feel it would be useful to present them in a form that is amenable to discussions later in this paper. In Section 4, we present our results for LO and NLO cross sections, both for the differential distributions $d \\sigma/d \\log x$ and $d\\sigma/d \\log Q^2$, as well as the total cross section. This has been done for neutral as well as charged current cross sections. Section 5 has a short discussion on saturation and the final section makes a few concluding remarks. ",
        "conclusions": ""
    },
    "0208/hep-ph0208131_arXiv.txt": {
        "abstract": "Applying the microcanonical definition of entropy to a weakly interacting and self--gravitating neutralino gas, we evaluate the change in the local entropy per particle of this gas between the freeze out era and present day virialized halo structures. An ``entropy consistency'' criterion emerges by comparing theoretical and empirical estimates of this entropy.  We apply this criterion to the cases when neutralinos are mostly B-inos and mostly Higgsinos, in conjunction with the usual ``abundance'' criterion requiring that present neutralino  relic density complies with $0.2<\\Omega_{{\\tilde\\chi^1_0}} < 0.4$ for $h\\simeq 0.65$.  The joint application of both criteria reveals that a much better fitting occurs for the B-ino than for the Higgsino channels, so that the former seems to be a favored channel along the mass range  of $155\\,\\hbox{GeV} <  m_{{\\tilde\\chi^1_0}} < 230 \\,\\hbox{GeV}$. These results are consistent with neutralino annihilation patterns that emerge from recent theoretical analysis on cosmic ray positron excess data reported by the HEAT collaboration. The suggested methodology can be applied to test other annihilation channels of the neutralino, as well as other particle candidates of thermal WIMP gas relics. ",
        "introduction": "There are strong theoretical arguments favoring lightest supersymmetric particles (LSP) as making up the relic gas that forms the halos of actual galactic structures. Assuming that  {\\it R} parity is conserved and that the LSP is stable, it might be an ideal candidate for cold dark matter (CDM), provided it is neutral and has no strong interactions. The most favored scenario \\cite{Ellis,Report,Torrente,Roszkowski,Fornengo,Ellis2} considers the LSP to be the lightest neutralino ($\\tilde\\chi_1^0$), a mixture of supersymmetric partners of the photon, $Z$ boson and neutral Higgs boson \\cite{Report}. Since neutralinos must have decoupled once they were non-relativistic, it is reasonable to assume that they constituted originally a Maxwell-Boltzmann (MB) gas in thermal equilibrium with other components of the primordial cosmic plasma. In the present cosmic era, such a gas is practically collision--less and is either virialized in galactic and galactic cluster halos, in the process of virialization or still in the linear regime for superclusters and structures near the scale of homogeneity\\cite{KoTu, Padma1,Peac}.  Besides the constraint due to their present abundance as main constituents of cosmic dark matter ($\\Omega_{{\\tilde\\chi^1_0}} \\sim 0.3$), it is still uncertain which type of annihilation cross section characterizes these neutralinos. In this paper we present a method that discriminates between different cross sections, based on demanding that (besides yielding the correct abundance) a theoretically  estimated entropy per particle matches an empiric  estimate of the same entropy, both constructed for actual galactic dark halo structures. The application of this ``entropy consistency'' criterion is straightforward because entropy is a state variable that can be evaluated at equilibrium states, irrespective of how enormously complicated the evolution between each such state might have been. In this context, the two fiducial equilibrium states of the neutralino gas are (to a good approximation) the decoupling (or ``freeze out'') epoque and their present state as a virialized relic gas. Considering simplified forms of annihilation cross sections, the joint application of the abundance and entropy--consistency criteria favors the neutralinos as mainly ``B--inos'' over neutralinos as mainly ``higgsinos''. These results are consistent with the theoretical analysis of the HEAT experiment~\\cite{HEAT-TH,HEAT1,HEAT2} which aims at relating the observed positron excess in cosmic rays with a possible weak interaction between neutralinos and nucleons in galactic halos.  The paper is organized as follows. In section 2 we describe the thermodynamics of the neutralino gas as it decouples. The microcanonical ensemble entropy is applied in section 3 to the post--decoupling neutralino gas to estimate the change in entropy between freeze out and present day conditions, leading to a suitable theoretical estimate of the entropy per particle. In section 4 we obtain an empirical estimate of this entropy based on actual halo variables, while in section 5 we examine the consequences of demanding that these two entropies coincide. We summarize our results in section 6. ",
        "conclusions": "We have presented a robust consistency criterion that can be verified for any annihilation channel of a given dark matter candidate proposed as the constituent particle of the present galactic dark matter halos. Since we require that the empirical estimate $\\shacem$ of present dark matter haloes must match the theoretical value $\\shacth$, derived from the microcanonical definition and from freeze out conditions for  the candidate particle, the criterion is of a very general applicability, as it is largely insensitive to the details of the extremely complicated evolution of the neutralino gas from its freeze out era (hence, it is also insensitive to the structure formation scenario that might be assumed). Further, the details of the present day halo structure enter only through an integral feature of the dark halos, the central escape velocity, thus our results are also insensitive to the fine details concerning the central density and the various models describing the structure of dark matter halos. A crucial feature of this criterion is its direct dependence on the physical details ({\\it{i.e.}} annihilation channels and mass) of any particle candidate.  Recent theoretical work by  E. A. Baltz {\\it{et al.}} \\cite{HEAT-TH} confirmed that neutralino annihilation in the galactic halo can produce enough positrons to make up for the excess of cosmic ray positrons experimentally detected by the HEAT collaboration \\cite{HEAT1,HEAT2}. Baltz {\\it{et al.}} concluded that for a boost factor $B_s \\sim 30$ the neutralinos must be primarily B-inos with mass around 160 GeV. For a boost factor $30 < B_s < 100$, the gaugino--dominated SUSY models complying with all constraints yield neutralino masses in the range of $150\\,\\hbox{GeV} <  m_{{\\tilde\\chi^1_0}} < 400 \\,\\hbox{GeV}$. On the other hand, Higgsino dominated neutralinos are possible but only for $B_s \\sim 1000$ with masses larger than 2 TeV. However, our results show this second option to be unlikely and are in agreement with the predictions that follow from \\cite{HEAT-TH}, as we obtain roughly the same mass range for the B-ino dominated case (see figure 1b) and the Higgsino channel is shown to be less favored in the mass range lower than TeV's.  We have examined the specific case of the lightest neutralino for the mostly B-ino and mostly Higgsino channels. The joint application of the ``entropy consistency'' and the usual abundance criteria clearly shows that the B-ino channel is favored over the Higgsino. This result can be helpful in enhancing the study of the parameter space of annihilation channels of LSP's in  MSSM models, as the latter only use equations (\\ref{eqxf}) and (\\ref{eqOmega0})--(\\ref{eqYinf}) in order to find out which parameters yield relic gas abundances that are compatible with observational constraints \\cite{Ellis,Report,Torrente,Roszkowski,Fornengo,Ellis2}. However, equations (\\ref{eqxf}) and (\\ref{eqOmega0})--(\\ref{eqYinf}) by themselves are insufficient to discriminate between annihilation channels. A more efficient study of the parameter space of MSSM can be achieved by the joint usage of the two criteria, for example, by considering more general cross section terms (see for example \\cite{Report}) than the simplified approximated forms (\\ref{sleptons}) and (\\ref{Wboson}). This work is currently in progress.\\\\"
    },
    "0208/astro-ph0208297_arXiv.txt": {
        "abstract": "s{The XXXVIIth Rencontres de Moriond on \"The Cosmological Model\" is briefly summarized.  Almost none of the current observations argues against the popular Cold Dark Matter + $\\Lambda$ concordance model.  However, it remains to be tested how astrophysical uncertainties involved in the interpretation of the different data sets affect the derived cosmological parameters. Independent tests are still required to establish if the Cold Dark Matter and Dark Energy components are `real', or just `epicycles' that happen to fit the current data sets well.} ",
        "introduction": "It is a challenging task in `data compression' to summarize briefly a conference so rich in ideas and observational results, covered in over 100 oral presentations. On the observational side, we were fortunate to hear at this meeting for the first time the results from two CMB experiments: CBI and VSA, which largely confirmed earlier results for the CMB acoustic peaks. We also heard updates on redshift and cluster surveys at different wavelengths.  On the theoretical side, we learnt about the most advanced numerical simulations, and on ideas which relate fundamental physics (e.g.`Brane World') to cosmological models.  The exponential growth of data has changed the character of the subject, in the sense that models can now be assessed quantitatively in great detail.  Below is a modest attempt to summarize the approaches of estimating the `best fit' cosmological model.  It is interesting that the title of meeting is ``The Cosmological Model'', perhaps implying the good agreement within the community on the  concordance $\\Lambda$-Cold Dark Matter model.  As this model has been so successful and popular, it is timely to ask `what can go wrong', and if other models are still possible. ",
        "conclusions": ""
    },
    "0208/astro-ph0208542_arXiv.txt": {
        "abstract": "A kinetic equation for Compton scattering is given that differs from the Kompaneets equation in several significant ways.  By using an inverse differential operator this equation allows treatment of problems for which the radiation field varies rapidly on the scale of the width of the Compton kernel.  This inverse operator method describes, among other effects, the thermal Doppler broadening of spectral lines and continuum edges, and automatically incorporates the process of Compton heating/cooling.  It is well adapted for inclusion into a numerical iterative solution of radiative transfer problems. The equivalent kernel of the new method is shown to be a positive function and with reasonable accuracy near the intitial frequency, unlike the Kompaneets kernel, which is singular and not wholly positive.  It is shown that iterates of the inverse operator kernel can be easily calculated numerically, and a simple summation formula over these iterates is derived that can be efficiently used to compute Comptonized spectra.  It is shown that the new method can be used for initial value and other problems with no more numerical effort than the Kompaneets equation, and that it more correctly describes the solution over times comparable to the mean scattering time. ",
        "introduction": "The most fundamental approach to treating Compton scattering of photons from thermal electrons is to use a Boltzmann-like kinetic equation for the photon distribution function.  The basic physics of the Compton process is incorporated into certain scattering functions, or kernel functions, that specify the probability of an initial photon scattering into various frequency and angular ranges.  This Boltzmann equation gives highly accurate results, but can also involve a heavy computational burden, especially when the number of scatterings is large.  The Boltzmann equation can be substantially simplified when certain conditions are met.  Consider the case where the photons and electrons are both non-relativistic, so that $h\\nu/mc^2 \\ll 1$ and $kT/mc^2 \\ll 1$.  Here $h$ is Planck's constant, $\\nu$ is the photon frequency, $m$ is the electron mass, $c$ is the speed of light, $k$ is Boltzmann's constant, and $T$ is the electron temperature.  In this case the scattering kernel is relatively narrow in frequency.  In fact the spreading is due to the thermal Doppler width is of order $\\Delta\\nu \\sim \\nu \\alpha^{1/2}$, where, \\be \\alpha = {{kT} \\over {mc^2}}.  \\e{1.1} \\ee  When the width of the scattering kernel is small compared to the scale over which the radiation field changes substantially, it is possible to approximate the scattering terms in the Boltzmann equation by a second-order differential operator acting on frequency.  Such approximate equations are generally called Fokker-Planck equations, but for Compton scattering the term {\\em Kompaneets equation} is used, in honor of its originator \\citep{K57}.  In its original form the Kompaneets equation applies strictly only to homogeneous, isotropic radiation fields in which there is time dependence but no spatial transport.  However, since the Compton scattering process itself is not very anisotropic, the Kompaneets scattering term is often used for non-isotropic radiation fields, introducing some additional degree of approximation.  In this paper, for simplicity, the transport equation is also written without the spatial transport terms, but it should be understood that the scattering term, within the isotropic approximation, is directly applicable to cases involving spatial transport as well.  The Kompaneets equation has been a very useful tool for describing the process of Compton scattering from thermal electrons in astrophysics, when the conditions mentioned above are met.  Second-order partial differential equations are much easier to handle numerically than full Boltzmann equations.  Within its limitations the simple Kompaneets equation manages to incorporate a number of important physical properties and effects, namely,  \\begin{enumerate} \\item Conservation of photon number. \\item Detailed balance (correct equilibrium solution). \\item The frequency spreading due to the thermal Doppler effect. \\item The frequency shift due to thermal Doppler effect (inverse Compton effect). \\item The frequency shift due to electron recoil (Compton effect). \\item Stimulated scattering. \\end{enumerate} The Kompaneets equation incorporates the first two properties exactly. The remaining effects are accurately represented only to lowest order in the parameters $h\\nu/mc^2$ and $\\alpha=kT/mc^2$.  For all its many positive features, however, the Kompaneets equation suffers from one major shortcoming: it is unable to treat cases where the radiation fields varies significantly on the scale of Compton frequency shifts, such as can occur in the neighborhood of spectral lines and continuum jumps.  The usual approach in such cases has been to revert to the full Boltzmann equation, which contains the detailed scattering kernels, or to use a Monte Carlo method.  The simplicity of the second order differential operator is thereby lost.  An entirely different approach to the treatment of Compton scattering was presented by \\citet[ hereafter RH]{RH}, designed to apply to conditions typical in stellar atmospheres.  Under these conditions the radiation fields change rapidly over the scale of the electron thermal frequency shift $\\Delta\\nu$ in the neighborhood of spectral lines and continuum edges. The method of RH handles such rapidly varying radiation fields in a numerically efficient way.  An essential feature of the RH method is that it expresses the radiation field as a differential operator acting on the emissivity, not vice versa.  This departure from the traditional Fokker-Planck type of operator leads us to call this an {\\em inverse operator method}.  As originally formulated, the RH method incorporated only numbers 1 and 3 of the above physical effects, namely, photon conservation and the spreading due to the thermal Doppler effect.  This was not a serious limitation when applied to normal stellar atmospheres, where these are the predominant effects.  However, it seemed desirable for have a method that combined the advantages of the Kompaneets and RH equations, which then would be applicable to a much wider class of problems.  The purpose of this paper is to present such a new kinetic equation for Compton scattering, which (like the Kompaneets equation) incorporates all the physical effects 1--6 above, but which also (like the RH method) has the ability to treat rapidly varying radiation fields.  While the idea for this new inverse operator method was motivated by the RH method, the derivation presented here is based on a simple alteration of the Kompaneets equation, which is considerably simpler than the approach used in RH.  The new method is only accurate to first order in $h\\nu/ mc^2$ and $\\alpha=kT/mc^2$, and it assumes that the scattering process can be well approximated by isotropic emission.  However, since the Kompaneets equation itself has these same limitations, the new method cannot be judged inferior because of them.  From the standpoint of applications to stellar atmospheres, there are several advantages of the new kinetic equation.  Like the Kompaneets equation, it now incorporates the processes of Comptonization of the radiation field and the Compton heating/cooling of the gas, which may be of importance in certain cases involving X-ray irradiation, for example.  Another feature is that it satisfies detailed balance, and thus correctly describes the approach to thermal equilibrium.  At the same time, like the RH method, it also can handle the distortion of the line profiles due to Doppler broadening due to scattering on the electrons.  The numerical implementation of the method involves only a minor modification of the RH method, which essentially maintains the latter's favorable timing requirements, namely proportional to the first power of the number of frequency points.  For problems other than stellar atmostpheres, there are also advantages of the new method.  For example, in studying Comptonization of X-rays, \\citet{LR79a,LR79b,LR80} wrote the solution to the transfer problem as a sum of products of probability of scattering $k$ times and the $k$-th iterated kernel function for Comptonization.  As we shall see, the equivalent kernel function for the Kompaneets equation is not wholly positive and is highly singular, being a linear combination derivatives of delta functions up to order two.  The ``iterated'' kernels would involve even higher derivatives, and any solution involving sums over such functions would be entirely unworkable. On the contrary, the kernel functions of the inverse operator method are quite normal functions, and their iterates can be found stably by numerical means, making the method of Lightman \\& Rybicki more practical.  In \\S \\ref{basic} the basic properties of the Kompaneets equation will be reviewed and the equations of the new inverse operator method will be derived.  It is demonstrated that the inverse operator mehtod, like the Kompaneets equation, satisfies all six of the properties listed above.  A comparison with the RH method will be given.  In \\S \\ref{kernels} we discuss properties of a rather general Boltzmann equation and show how one may define the equivalent kernels of any approximate version of it.  This is then done for the Kompaneets and inverse operator methods.  Numerical results are presented for the inverse operator kernel, and its properties, including its accuracy, are discussed.  In \\S \\ref{iterated} iterates of the kernel function are defined and discussed, and numerical examples are given.  In \\S \\ref{summation} a useful formula is derived that reduces certain summations occuring in the formalism of \\citet{LR79a,LR79b,LR80} to numerically tractable forms.  In \\S \\ref{iv} it is shown how initial value problems can be treated using the inverse operator method. Numerical results show the advantages of the inverse operator method over the Kompaneets method for short times, of the order of the mean scattering time.  In \\S \\ref{summary} a short review is given, and some possibilities for future work are discussed. ",
        "conclusions": "\\label{summary}  We have demonstrated a simple modification of the Kompaneets method that maintains most of its desirable properties, but which has much better behavior at very short times, of order of the mean free scattering time.  This inverse operator method has an equivalent kernel function that is positive and nonsingular, unlike that of the Kompaneets equation.  The kernel function is roughly of the right shape in the central regions with an accuracy of about 20\\%.  We have shown that for many applications the inverse operator method requires no more numerical effort than the Kompaneets equation.  We have shown how the iterated kernels of the inverse operator method can be efficiently computed numerically.  A summation formula involving iterated kernels is given that reduces the effort in computing Comptonized spectra using the method of \\citet{LR79a,LR79b,LR80}. We have shown that initial value problems can be solved using the inverse operator method with no more effort than the Kompaneets equation, but with noticeable improvement in the solution at times comparable to the mean scattering time.   In this paper there has been no attempt to advance beyond the physics of the Kompaneets equation.  The main goal has been simply to point out the special advantages that occur when one reverses the role of the second order differential operator that connects the radiation field with the emission coefficient.  However, in future work it would be desirable to overcome some of the limitations of the inverse operator method, as others have done for the Kompaneets equation. There does not seem to be any reason to expect any difficulties in extending the present theory to anisotropic scattering or polarization.  It would be desirable to find ways to increase the accuracy of the inverse operator method.  Obvious improvements would come with using higher order expansions of the coefficients in the parameters $\\alpha$ and $h\\nu/mc^2$, but this will not solve all of the accuracy issues.  More improvement might be gained by expressing the emission coefficient as a sum of $N>1$ terms, as in RH.  Alternatively, and more in keeping with the present approach, one could include higher order moments and derivatives in a Kramers-Moyal expansion \\citep[see, e.g., ][]{Riskin} of the Boltzmann equation, and then use this to form new, higher order, inverse operators.  The present work has concentrated on one particular Fokker-Planck equation, the Kompaneets equation.  However, the idea of using inverse operators clearly could be generalized and applied to other Fokker-Planck equations.  Since each physical situation has its own special properties, it is not possible to predict whether an inverse operator approach would provide significant advantages in any particular case, but it might be worth investigating.  An intriguing possibility is that higher order Kramers-Moyal expansions might have better properties when modified by the use of inverse operators, which perhaps might avoid some of the difficulties known to exist for the traditional expansions beyond the Fokker-Planck approximation \\citep{Pawula}."
    },
    "0208/astro-ph0208068_arXiv.txt": {
        "abstract": "{\\small We report the discovery of fast ($\\sim 100$ sec) 2.2 $\\mu${m} flares from the microquasar GRS 1915+105, which are superimposed on longer ($\\sim 30$ min), brighter flares corresponding to episodes of jet formation.  The number and strength of the sub-flares in each bright flare varies, and does not seem to correlate in any obvious way with the underlying light curve or with changes in the X-ray emission. However, the fact that these sub-flares only occur in tandem with the larger flares indicates that they might be related to the jet ejection process.} ",
        "introduction": "Simultaneous X-ray and infrared light curves of the microquasar GRS 1915+105 were obtained on August 14, 1997 \\cite{eiken}, one of the first observations showing the disk-jet interaction in this source (Figure \\ref{fig:lcurve}).  GRS 1915+105 exhibited quasi-regular flaring behavior in the infrared during the course of these observations (interpreted as synchrotron emission from ejected plasma), and the flares were seen to correlate with changes in the X-ray emission (interpreted as emptying and refilling of the system's inner accretion disk).\\\\ \\indent Smaller amplitude variability, on faster timescales, was also observed in some of the infrared flares. In this contribution, we have reanalyzed the original data (obtaining higher signal-to-noise photometry than previously presented) and identified all such ``sub-flares'' above a given detection threshold, in an attempt to characterize this previously unknown phenomenon.  \\begin{figure}[htb] \\centering \\epsfig{file=fig1.eps,width=11cm} \\caption{Light curves of GRS 1915+105 on August 14, 1997 UT (Eikenberry et al. 1998). The data in the top panel were taken in K-band (2.2 microns) using the Palomar 5-meter telescope and have been dereddened by 3.3 magnitudes.  The data in the bottom panel were taken using the Proportional Counter Array (PCA) on the Rossi X-ray Timing Explorer (RXTE).  Both light curves have time resolution of one second.  The numbered infrared flares correspond to those shown in Figure 2.} \\label{fig:lcurve} \\vspace{-0.2cm} \\end{figure}  \\begin{figure}[htb] \\centering \\psfig{file=fig2.ps,width=17cm} \\caption{The reanalyzed data from Figure 1, showing the six infrared flares in which sub-flares were detected.  In each case, the numbered panel shows the reanalyzed light curve from Figure 1, with arrows marking each statistically significant sub-flare.  The next panel down shows the light curve after application of a high pass filter to remove all frequencies lower than $\\sim$ 1/(500 sec).  The bottom panel shows the light curve of a field star near GRS 1915+105, which we used to calibrate variation due to atmospheric effects.  In (4) and (6), the top panel shows the X-ray light curve from Figure 1 at the time corresponding to the sub-flaring behavior.} \\label{fig:sub} \\vspace{-0.25cm} \\end{figure} ",
        "conclusions": "The sub-flares we detected are shown in Figure \\ref{fig:sub}.  They have typical rise times of 40 to 100 seconds (with a few longer) and typical amplitudes of 25 to 70 mJy.  The most important point about the sub-flares is that they are all superimposed on larger flares, indicating a possible relationship to the jet.  Some of the large flares are multiple-peaked (2,3), with several strong sub-flares superimposed on them, while other large flares only have one peak (1,6), but it is sharp enough to be detected as a sub-flare by our algorithm.\\\\ \\indent There is no obvious correlation between the properties of the main flare and the number and strength of sub-flares superimposed on it.  The X-ray coverage of the sub-flare detections is poor, but in one case for which they overlap (4), a faint sub-flare is detected at the beginning of the main flare, indicating a possible connection to the X-ray ``spike'' which marks the beginning of the jet ejection. Two other flares (2,3) also show evidence for a faint sub-flare near the beginning of the main flare.  Taken together, this suggests that these sub-flares originate near the accretion disk and might be associated with the jet formation process."
    },
    "0208/astro-ph0208318_arXiv.txt": {
        "abstract": "High resolution (R$\\sim$20,000), high signal-to-noise (S/N $\\geq$ 100) spectra were collected for $\\sim$40 symbiotic stars with the Asiago echelle spectrograph~ over the same 8480-8740 \\AA~ wavelength range covered by the ESA Cornerstone mission GAIA, centered on the near-IR CaII triplet and the head of the Paschen series. A large number ($\\sim$ 140) of cool MKK giant and supergiant templates were observed with the same instrumentation to serve as a reference and classification grid.\\\\ The spectra offer bright prospects in classifying and addressing the nature of the cool component of symbiotic stars (deriving T$_{{\\sl eff}}$, log{\\sl g}, [Fe/H], [{\\sl $\\alpha$}/Fe], V$_{{\\sl rot}}$sin{\\sl i} both via MDM-like methods and syntetic atmosphere modeling) and mapping the physical condition and kinematics of the gas regions responsible for the emission lines. ",
        "introduction": "The spectral region around the near-IR CaII triplet and the head of the Paschen series is among the most promising to investigate the cool component of symbiotic stars. First of all the region is free from telluric absorptions and thus well accessible from the ground (Munari 1999). Secondly, given the dependance of the interstellar extinction on wavelength, the red region is more suitable for the study of the symbiotic sample, which has a spatial distribution flattened toward the Galactic plane.  However, the dominant reasons to move toward these red wavelengths lie in their high diagnostic potential of the nature of the cool giant and the reduced disturbing presence of the nebular continuum. Furthermore, this region (precisely the 8480-8740 \\AA\\ interval) has been selected for the spectral survey ($3\\times10^8$ objects, complete to $V\\sim 17.5$ mag) that the ESA astrometric Cornerstone mission GAIA should perform starting 2010.  The region is dominated by some of the strongest features visible in cool stars ( CaII triplet, FeI, TiI, MgI and SiI lines, as well as lines from CN and TiO bands ) . The presence and activity of the hot companion is well traced by \\clearpage \\begin{figure} \\plotfiddle{Marrese_fig1.ps}{19 truecm}{0}{80}{80}{-285}{-40} \\caption{Normalized spectrum of the symbiotic star EG~And over the GAIA wavelength region ($\\lambda\\lambda$ 8480-8740 \\AA). No line appears in emission over the well developped absorption continuum of the M giant (strongest lines belonging to CaII, FeI, TiI and MgI, with many weaker lines contributed by CN and TiO molecules).} \\end{figure} \\begin{figure} \\plotfiddle{Marrese_fig2.ps}{19 truecm}{0}{80}{80}{-285}{-40} \\caption{Normalized spectrum of the symbiotic star Z~And over the GAIA wavelength region ($\\lambda\\lambda$ 8480-8740 \\AA). Note the presence of double peaked emission core in the CaII-triplet absorption lines and the P13, P14, P15 and P16 Paschen lines in emission. The absorption spectrum of the M giant is undisturbed and rich in FeI, TiI and MgI absorptions (apart from the forest of weak lines due to CN and TiO molecules).} \\end{figure} \\clearpage \\noindent Paschen, CaII, HeI and NI emission lines which are strong and abundant in this wavelength interval. The latter is away from the veiling effect of the nebular blue continuum, that by filling-in aborption features tendes to mimic hotter spectral types, reduced metallicities and higher gravities for the cool component. ",
        "conclusions": ""
    },
    "0208/astro-ph0208404_arXiv.txt": {
        "abstract": "A robust analysis of galaxy structural parameters, based on the modeling of bulge and disk brightnesses in the BVRH bandpasses, is presented for 121 face-on and moderately inclined late-type spirals. Each surface brightness (SB) profile is decomposed into a sum of a generalized \\sersic\\ bulge and an exponential disk. The reliability and limitations of our bulge-to-disk (B/D) decompositions are tested with extensive simulations of galaxy brightness profiles (1D) and images (2D)\\@. We have used repeat observations to test the consistency of our decompositions.  The average systematic model errors are \\lapprx20\\% and \\lapprx5\\% for the bulge and disk components, respectively. The final set of galaxy parameters is studied for variations and correlations in the context of profile type differences and wavelength dependences.  Galaxy types are divided into three classes according to their SB profile shapes; Freeman Type-I and Type-II, and a third ``Transition'' class for galaxies whose profiles change from Type-II in the optical to Type-I in the infrared.  Roughly 43\\%, 44\\%, and 13\\% of Type I, II, and Transition galaxies respectively comprise our sample.  Only Type-I galaxies, with their fully exponential disks, are adequately modeled by our 2-component decompositions and our main results focus on these profiles.  We discuss possible interpretations of Freeman Type-II profiles.  The \\sersic\\ bulge shape parameter for nearby Type-I late-type spirals shows a range between $n=$0.1--2 but, on average, the underlying surface density profile for the bulge and disk of these galaxies is adequately described by a double-exponential distribution. The distribution of disk scale lengths shows a decreasing trend with increasing wavelength, consistent with a higher concentration of old stars or dust (or both) in the central regions relative to the outer disk. We confirm a coupling between the bulge and disk with a scale length ratio $\\langle r_e/h \\rangle = 0.22 \\pm 0.09$, or $\\langle h_{bulge}/h_{disk} \\rangle = 0.13 \\pm 0.06$ for late-type spirals, in agreement with recent N-body simulations of disk formation.  This ratio increases from $\\sim0.2$ for late-type spirals to $\\sim0.24$ for earlier types. These observations are consistent with bulges of late-type spiral galaxies being more deeply embedded in their host disk than earlier-type bulges, as discussed by Graham (2001). Bulges and disks can thus preserve a nearly constant $r_e/h$ but show a great range of surface brightness for any given effective radius. The similar scaling relation for early and late-type spirals suggests comparable formation and/or evolution scenarios for disk galaxies of all Hubble types. In the spirit of Courteau, de~Jong, \\& Broeils (1996) but using our new, more extensive data base, we interpret this result as further evidence for regulated bulge formation by redistribution of disk material to the galaxy center, in agreement with models of secular evolution of the disk. ",
        "introduction": "\\label{sec:intro}  Stellar density distributions provide important constraints for bulge and disk formation models. Historically, astronomers have embraced the $r^{1/4}$ brightness ``law'' \\citep{deVauc48} and exponential brightness profile\\footnote{The exponential nature of galaxy disk profiles emerges naturally in analytical models of disk formation (e.g.\\@ Lin \\& Pringle 1987; Dalcanton, Spergel, \\& Summers 1997; Ferguson \\& Clarke 2001).} (de Vaucouleurs 1959a; Freeman 1970) to model the light distribution of the galaxy bulge and disk, respectively\\footnote{It is important to remind ourselves from the onset that bulge-to-disk decompositions, and inward extrapolations of the disk into the central bulge and/or bar, may have no physical (or dynamical) basis. They provide a convenient description of the light distribution of a galaxy's components that are otherwise dynamically coupled.  The effective integrals of motions are likely similar for all the co-spatial components, though kinematically distinct bulges (counter-rotating nuclei) are known to exist.}. Departures from the standard de~Vaucouleurs profile in the {\\it inner} light distribution of early- and late-type spirals have however been demonstrated in a number of early studies \\citep{deVauc59, vanHout61, Burstein79}, including the Milky Way \\citep{Kent91}.  \\citet{AndSan94}, \\citet{deJong96a}, Courteau, de~Jong, \\& Broeils (1996), and \\cite{Carollo99} later used small samples of high-quality surface brightness (SB) profiles to establish the exponential profile as a better match to {\\it late}-type disk bulges; thus SB profiles of most late-type spirals are best modeled by a double-exponential fit to the bulge and disk.  A broader analysis suggests a range of bulge shapes from early- to late-type spirals \\citep{AndPelBal95, deJong96a, CourdeJBro96, Graham01}. Most of these analyses rely on the modeling of a generalized surface density function such as that proposed by \\citet{Sersic68}; \\begin{equation} \\label{eq:sersic} I(r)=I_{0}\\,{\\exp\\left\\{-{\\left({r\\over{r_{0}}}\\right)^{\\rm{1/n}}}\\right\\}} \\end{equation} or, in magnitudes, \\begin{equation} \\label{eq:sersicmag} \\mu(r)=\\mu_{0}+ 2.5\\log(e){\\left\\{\\left({r\\over{r_{0}}}\\right)^{\\rm{1/n}}\\right\\}}. \\end{equation} where $\\mu_0$ ($I_{0}$) is the central surface brightness (intensity), $r_{0}$ is a scaling radius, and the exponent $1/n$ is a shape parameter that describes the amount of curvature in the profile.  For $n=1$ or 4 one recovers a pure exponential or the de~Vaucouleurs $r^{1/4}$ profile respectively.  Collectively, the works above suggest that the bulge shape parameter $n$ correlates with absolute luminosity and half-light radius, such that bigger, brighter systems have larger values of $n$.  This result was extended to brightest cluster galaxies by \\citet{Graham96}. \\citet{CourdeJBro96} also demonstrated a tight correlation between the bulge and disk exponential scale lengths, for all spiral types, with $h_b/h_d = 0.1\\pm0.05$ (where $h=r_{0}$ and $n=1$ in Eq.~\\ref{eq:sersic}). The exponential nature of late-type galaxy bulges and the correlation between bulge and disk scale lengths was interpreted by \\citet{CourdeJBro96} as evidence for regulated bulge formation by redistribution of disk material to the galaxy center by a bar-like perturbation. We will return to this important constraint for secular evolution models in \\S~\\ref{subsec:sec_ev}.  This study focuses on the development of a reliable set of observables and constraints for structure formation models.  An important goal is to measure the range of the \\sersic\\ $n$ parameter for virialized disk systems. The analyses described above are reproduced and expanded upon with the largest multi-band survey of its kind to date and a clearer understanding of model limitations than previously attained. We aim to characterize and quantify the intrinsic structural properties of the bulge and disk and the extent of their variation with wavelength. These characterizations are made through reliable modeling of bulge and disk parameters from SB profile decompositions. Multi-wavelength information also provides insight about structural variations within and among galaxies due to dust and stellar population effects.  While some of these issues have been addressed before, there remains a number of significant measurement uncertainties and technical limitations which we now investigate thoroughly.  This paper is organized as follows: a brief description of the database is given in \\S~\\ref{sec:data} and in \\S~\\ref{sec:simulations} we discuss our B/D decomposition algorithms (1D and 2D) and the simulations to test the reliability of our technique. For the readers interested mostly in final profile decompositions and results, a summary of the simulation results and guidelines is given in \\S~\\ref{sec:sim_summ}. Actual B/D decompositions of galaxy SB profiles are presented in \\S~\\ref{sec:decomps}, followed by a discussion and interpretation of the results in terms of secular evolution models in \\S~\\ref{sec:results}. A discussion on the nature of Freeman Type-II profiles is also presented in \\S~\\ref{sec:results}. We conclude with future directions in \\S~\\ref{sec:summary}.  Two appendices present (A) a discussion of the functional form for the \\sersic\\ coefficient $b_n$, and (B) decomposition results for our Type-I profiles. ",
        "conclusions": "\\label{sec:summary}  This study has focused on the development of rigorous B/D decomposition techniques using a new, comprehensive, multi-band survey of late-type spiral galaxies.  We examine three types of SB profiles, Freeman Type-I and Type-II, and a third ``Transition'' class for galaxies whose profiles change from Type-II in the optical to Type-I in the infrared.  This distinction is important since Type-II and Transition profiles cannot be adequately modeled by a simple two-component model of the bulge and disk.  Thus, our main results are based on Type-I profiles.  Based on extensive simulations, careful treatment of sky and seeing measurement errors, and repeat observations we are confident that systematic errors are $\\la20$\\% for the bulge components, including the \\sersic\\ shape parameter, and $\\la5$\\% for disk components.  The main conclusions from our simulations and final profile decompositions are as follows:  \\begin{itemize}  \\item Simulations to determine the range of acceptable solutions for any B/D decomposition program are crucial.  The reliability of bulge model parameters is limited by the relative size of the bulge and seeing disk, seeing errors, the intrinsic bulge shape, sky brightness and errors.  Disk parameters are fairly robust to systematic errors, with the exception of improper bulge shapes and sky errors which can have dramatic effects on both modeled disk and bulge components. \\item The \\sersic\\ bulge shape parameter for nearby late-type galaxies shows a range between $n=0.1-2$, but, on average, their underlying surface brightness distribution is best described by a double-exponential model of bulge and disk. \\item Disk scale lengths decrease at longer wavelengths, indicative of a higher concentration of older stars and/or dust in the central regions relative to the outer disk. \\item We confirm and reinforce the result of \\citet{CourdeJBro96} of a structural coupling between the bulge and disk of late-type spirals.  We find $\\langle r_e/h \\rangle = 0.22 \\pm 0.09$, or $\\langle h_{\\rm bulge}/h_{\\rm disk} \\rangle = 0.13 \\pm 0.06$, independent of wavelength.   A mild trend with Hubble type is observed with $\\langle r_e/h \\rangle = 0.20 - 0.013(T-5)$ ($1\\sigma = 0.09$), ranging from $\\langle r_e/h \\rangle \\sim 0.20$ for late-type spirals to $\\langle r_e/h \\rangle \\sim 0.24$ for earlier types. These results are consistent with scenarios of bulge formation in which bulges of late-type spiral galaxies are more deeply embedded in their host disk than earlier-type bulges.  Under this ``iceberg'' scenario, bulges and disks can thus preserve a nearly constant $r_e/h$ but show a great range of $\\mu_e$ for any given $r_e$. The observed scale ratio is consistent with numerical simulations of self-gravitating disks and probably related to the stellar dynamics of an actual or pre-existing barred system.  \\item The inner brightness profile signatures of Type-II galaxies are likely explained by a combination of dust extinction and stellar population effects and perhaps linked to the occurence of a bar, but no decisive conclusion can be derived at present. \\end{itemize}   \\newpage"
    },
    "0208/hep-th0208080_arXiv.txt": {
        "abstract": "We study the system of Schwarzschild anti de Sitter (S-AdS) bulk and FRW brane for localization of gravity; i.e. zero mass gravitons having ground state on the brane, and thereby recovering the Einstein gravity with high energy correction. It has been known that gravity is not localized on AdS brane with AdS bulk. We prove the general result that gravity is not localized for dynamic branes whenever $\\Lambda_4 < 0$, and is localized for the curvature index $k = 1$ only when $\\Lambda_4 > 0$ and black hole mass $M \\neq 0$, else it is localized for all other FRW models. If the localization is taken as the brane world compatibility criterion for cosmological models, then it would predict that negative cosmological constant on the brane is not sustainable. ",
        "introduction": "The idea that our universe has dimensions more than four has been around since the first attempts to unify fundamental forces. The Kaluza-Klein theories were one of the first attempts towards this direction in which the size of extra dimension was taken to be of the order of Planck length. In its new avatar in the form of brane world scenarios, it has attracted much attention in recent times. In these models our physical universe is envisioned as a four dimensional hypersurface in a five dimensional bulk spacetime. The standard model matter is confined to the brane but gravity, by its universal character, can propagate in the extra dimension.  Early attempts in this direction were based on extra dimensions being of the order of millimeter \\cite{arkani} which is less than the current observational limits on low scale gravity. The warped extra dimension models of Randall and Sundrum (RS) \\cite{rs1,rs2} have seen intense activity. Their two brane model had a brane with negative tension which led to anti gravity \\cite{shiromizu} and was hence ruled out. In their single brane model, the flat brane sits in an AdS bulk which is $Z_2$ symmetric. The $Z_2$ symmetry of the extra dimension arises from the reduction of M theory to $E_8 \\times E_8$ heterotic string theory \\cite{witten}. RS were able to  recover Newton's inverse square law with correction terms. These correction terms arise from the massive Kaluza-Klein(KK) modes on the brane and their effect can be tested at sub-millimeter level \\cite{hoyle}. There have also come about quite a few generalizations,  for eg.  in the form of thick branes (see for eg. \\cite{csaki}), AdS branes \\cite{karch} and brane models without $Z_2$ symmetry \\cite{colyuri}.  The brane and bulk spacetimes are joined through the Israel junction conditions \\cite{israel}. The Einstein equation on the brane is modified and there are various interesting cosmological consequences of the modifications \\cite{cosmo1}, including CFT effects \\cite{cft}. Some of the important solutions of bulk spacetime are branes in AdS bulk and FRW branes in Schwarzschild-AdS (S-AdS) bulk \\cite{karch, bdel, kraus, s-ads, sd2}.  Extra dimensions, though introduced to deal with problems in standard model of particle physics and also motivated by unification schemes, give a new color to the cosmological aspects of our physical universe. The S-AdS bulk modifies the Friedmann equation on the brane which now contains two extra terms. First is a matter density squared term and the other is a dark radiation term propotional to the mass parameter of the bulk black hole. The latter is contribution of the bulk Weyl curvature. The density squared term dies out at low energies and the dark radiation term is severely constrained by the big bang nucleosynthesis. Ultimately, at low energies there survives no observational signature which can distinguish this theory from the standard cosmology. However, there have been some attempts by incorporating string theory motivated correction terms in the action which result in modifications available  at the low energy sector \\cite{gb}. They can be tested against current observations, eg. supernova Ia observations \\cite{sn}. Such observations have provided a valuable tool to test standard cosmological model and other novel ideas like those with variable cosmological constant \\cite{vishwa}. In one of the studies based on AdS-CFT correction to S-AdS bulk it was found that the theory provides a definite low energy signature which fits very well with the supernova Ia observations for a wide range of parameters and also brings relief to the age of the universe problem in the standard cosmological model \\cite{svd}.  However, before one addresses the cosmological problems it is important to answer the question of localization of gravity on the brane harboring such models. Localization is not guaranteed a priori as we shall see that it does not happen for all  FRW models, and in particular for negative cosmological constant on the brane. Apart from $\\Lambda$, the curvature index $k=0, \\pm1$ and presence of black hole in the bulk which will make Weyl curvature in the bulk non-zero will also have bearing on this question. In particular, for the conformally non flat Nariai bulk, it has been shown that there can not exist any normalizable bound massless graviton \\cite{sd1}. Another example of non localization is provided by AdS brane in AdS bulk \\cite{karch} and we shall in this investigation establish that this feature holds good more generally even when brane is non empty. We shall establish the general result: \\emph{gravity is never localized whenever $\\Laf < 0$ on a dynamic brane, is localized for $k = 1$ only when $\\Laf > 0, M \\neq 0$ and is else always localized on the brane for all other FRW models}.  The cosmologically interesting FRW brane resides in S-AdS bulk for which the most pertinent question of localization was addressed by authors for a special case \\cite{sd2}. Here we would expand upon our previous work and deal with all possible cosmological scenarios. We would consider the perturbations of the bulk metric to study allowed ground state of massless graviton on the brane together with motion of brane in the bulk spacetime. It is well known that a small off tuning from the critical brane tension makes the brane dynamic \\cite{kraus}. For an FRW expanding universe on the brane, brane will have to be moving in bulk spacetime. Hence the brane equation of motion would also play a critical role alongside the perturbation equation in determining localization of gravity on the brane. In particular, localization would require brane to be ever expanding which would rule out negative cosmological constant on the brane. This is a definite prediction of the brane world cosmology. The most critical and attractive feature of brane world gravity is that it is localized on the brane. Localization could hence serve as compatibility criterion for cosmological models. Further, it is interesting that brane dynamics in the bulk, apart from other physical considerations, also determines that mass of the bulk black hole must be positive.  In the previous studies, the Gaussian normal coordinates for AdS bulk were employed, we shall however carry out the perturbation analysis in the natural curvature coordinates for S-AdS bulk spacetime. This should not matter in the end for localization must be coordinate independent physical property. In Sec. II, we carry out the perturbation analysis and recover RS model results in Sec. III. In the following three sections are discussed the cases of the curvature parameter $k = 0, 1 $ and $ -1$ for the vanishing and non vanishing mass of the bulk black hole. We would show that a number of interesting cosmological models are indeed compatible and allowed, in particular inflationary universes with positive cosmological constant on the brane which is favored by the current observations \\cite{sn}. For completeness we shall also consider static brane which will not be of practical use for it can not harbor expansion. This will be followed by conclusion.  In the next section we study the graviton fluctuations of the bulk metric and derive the wave equation governing the effect of extra dimension on its modes. ",
        "conclusions": "Localization of gravity on the brane is certainly a critical requirement for brane world cosmology. We have established the general result that {\\em for a dynamic FRW brane in S-AdS bulk, gravity is never localized whenever $\\Lambda_4 < 0$, is localized for $k = 1$ only when $\\Lambda_4 > 0$ and $M\\neq 0$ and else is always localized on the brane for all other FRW models.}  Localization hinges on two features: one, potential for zero mass graviton to be negative and second, brane must always be expanding or static at a location. Except for the case $k = 1$, the former condition holds good without any constraint on the parameters. For $k = 1$, presence of black hole in the bulk is necessary to make $V_f < 0$. That is, black hole in the bulk helps confinement of gravitons on the brane. In all other cases, the critical role is played by eq. (19), which determines  whether brane would have bounce or not. For ever expansion, this equation must not have a real positive root. It is clear that only negative $\\Laf$ and positive $k$ favor bounce and consequently oscillatory universe, while $M$ helps expansion. Asymptotically, $\\Laf$ dominates over all others and hence when it is negative, there would always occur a bounce and thereby no localization. Presence of black hole critically matters only in the case of $k = 1$, where it turns potential negative. In all other cases, its role is facilitative rather than critical. Non negative $\\Laf$ and non positive curvature index augurs well with expansion and from the bulk they are joined constructively by mass of the black hole. The opposing effect comes from negative $\\Laf$ and positive $k$. It is expansion which is critical for localization of gravity for zero mass graviton should asymptotically see vanishing potential. It would be interesting to investigate the localization of other spin fields and new interactions which might result in this setting, as has been done for the case of RS brane \\cite{bajc}.  Though static branes are not of cosmological interest, they exhibit a direct relation between mass of the bulk black hole and energy distribution on the brane. A static brane cannot have negative curvature. When it is flat, both black hole mass and effective cosmological constant $\\eta$ must vanish and this gives the non emepty brane generalization of RS brane. It could be located freely anywhere except at $y = 0$. When $k = 1$, both black hole mass and $\\eta$ must be non zero and its location is fixed at $y^2 = 2M$ which would always lie outside the horizon. The brane is kept static by the balance between black hole mass and $\\eta$, which should always be positive. On the other hand $\\Laf$ could have any sign.  Based on localization of gravity on the brane it clearly follows that brane world cosmology makes the definite prediction that in a dynamic universe \\emph{the cosmological constant on the brane cannot be negative and the Universe must be ever expanding}. This is an observationally testable prediction. The present observations indicate accelerating expansion which augurs well with this prediction. The main message of our study is that FRW cosmologies on the brane with S-AdS bulk spacetime are well founded gravitationally.  We thank Luis Anchordoqui, Roy Maartens, Carlos Nunez, S. Shankaranarayanan and R. G. Vishwakarma for various useful discussions and comments. PS is supported by a research grant from Council for Scientific \\& Industrial Research."
    },
    "0208/astro-ph0208362_arXiv.txt": {
        "abstract": "Nuclear medium effects in the neutrino cooling of neutron stars through the  reaction channel $\\gamma\\gamma \\rightarrow \\pi^0 \\rightarrow \\nu_R\\bar{\\nu}_L (\\nu_L \\bar{\\nu}_R)$ are incorporated. Throughout the paper we discuss different possibilities of right-handed neutrinos,  massive left-handed neutrinos and standard massless  left-handed neutrinos (reaction is then allowed only with medium modified vertices). It is demonstrated that multi-particle effects suppress the rate of this reaction channel in the dense hadron matter by $6 - 7$ orders of magnitude that does not allow to decrease existing experimental upper limit on the corresponding $\\pi^0 \\nu\\bar{\\nu}$ coupling. Other possibilities of the manifestation of the given reaction channel in different physical situations, e.g. in the quark color superconducting cores of the most massive neutron stars, are also discussed. We demonstrate that in the color-flavor-locked superconducting phase for temperatures $T\\lsim (0.1\\div 10)$~MeV (depending on the effective pion mass and the decay width)  the process is feasibly the most efficient neutrino cooling process, although the absolute value of the reaction rate is rather small. ",
        "introduction": "\\par Many years ago Pontecorvo and Chiu and Morrison\\cite{ponte} suggested that the process $\\gamma \\gamma \\rightarrow \\nu \\bar{\\nu}$ might play an important role as a mechanism for stellar cooling. Gell-Mann\\cite{gm} subsequently showed that this process is forbidden in a local (V-A) theory. However, it can occur at the one-loop level which has been computed by Levine\\cite{levine} for an intermediate-boson (V-A) theory, and the stellar energy loss rate through $\\gamma \\gamma \\rightarrow \\nu \\bar{\\nu}$ was found to be smaller than the rates for competing processes (pair annihilation $e^+ e^- \\rightarrow \\nu \\bar{\\nu}$ and photo-neutrino production $\\gamma e \\rightarrow e \\nu \\bar{\\nu}$). This result is not modified when the cross section of the above process is computed in the standard model, as was shown by Dicus\\cite{dicus}. Only for very peculiar neutrino coupling to photons or unnaturally large neutrino masses this reaction overwhelms the result of the standard model\\cite{natale}.            There is still another possibility proposed by Fischbach {\\it et al.} \\cite{fisch}, where the reaction $\\gamma \\gamma \\rightarrow \\nu \\bar{\\nu}$ could be significant. This is the case when the process is mediated by a pseudoscalar resonance and the latter decays into $\\nu \\bar{\\nu}$ due to the existence of right-handed neutrinos or due to new interactions beyond the standard model. It was assumed that in astrophysical conditions only the pion resonance could be important (next in mass not strange $\\eta$-resonance is too heavy in standard conditions) and the process was termed the pion-pole mechanism. Thus the process which we will continue to  discuss in this paper  is $\\gamma \\gamma \\rightarrow \\pi^0 \\rightarrow \\nu \\bar{\\nu}$.    Of course, if the temperature is high enough, and on the other hand, the pion dispersion relation in matter allows for the quasiparticle spectrum branch, there appears a significant number of  thermally equilibrated pion quasiparticles. Then the process $\\pi^0\\rightarrow \\nu \\bar{\\nu}$ may also be important. In this process the initial thermally equilibrated pion is on its mass-shell modified in the matter. In the process $\\gamma \\gamma \\rightarrow \\pi^0 \\rightarrow \\nu \\bar{\\nu}$ the initial reaction states contain no pion, the virtual pion only transfers the interaction from thermally equilibrated photons in the $\\gamma\\gamma$ annihilation process to produced $\\nu\\bar\\nu$. As we will see below, the process $\\gamma \\gamma \\rightarrow \\pi^0 \\rightarrow \\nu \\bar{\\nu}$ has an output of the energy $Q$ varying with the temperature as $Q\\propto T^n$ where the power $n$ changes with the temperature typically from  $n=3$ for rather high temperature ($T$ is still much smaller than the pion mass $m_{\\pi}=140~$MeV) to $n=11$ for low temperatures. The process $\\pi^0\\rightarrow \\nu \\bar{\\nu}$ having essentially larger phase space volume (one particle in the initial state) yields however the exponentially suppressed output of the energy, $Q\\propto T^{3/2} e^{-m_{\\pi}/T}$ at $T<m_{\\pi}$  since the initial particle is massive, in difference with photons. Concentrating on the discussion of the rate of the $\\gamma \\gamma \\rightarrow \\pi^0 \\rightarrow \\nu \\bar{\\nu}$ reaction channel we shall also compare it with the rate of the competing $\\pi^0\\rightarrow \\nu \\bar{\\nu}$ process and other relevant processes.          Temperatures of the order of $10 - 60$ MeV are expected in interiors  of  proto-neutron stars formed in supernova (SN) explosions during the early cooling phase of the proto-neutron star evolution, according to existing numerical simulations\\cite{burrows}. It is exactly in such a situation the pion-pole mechanism is the most effective ($T\\sim m_{\\pi}/a $, $a\\sim 2\\div 50$). Indeed, in Ref.\\cite{sn} the mechanism was applied to the case of supernova SN1987A and it was found to be a quite important process even if the pion partial production rate of neutrinos ($\\Gamma (\\pi^0 \\rightarrow \\nu_R \\bar{\\nu}_L (\\nu_L \\bar{\\nu}_R $ ))) is many orders of magnitude smaller than the presently accepted value of the experimental upper limit. In case of $\\nu_L \\bar{\\nu}_R $ with the vacuum vertex  the process rate is proportional to the squared neutrino mass. The result \\cite{sn} was criticized by Raffelt and Seckel \\cite{raffelt} basing on the fact that in the calculation of Ref.\\cite{sn} it was used the vacuum value for the total pion width, $\\Gamma_{\\pi}^{vac}$, which is a tiny quantity, $\\Gamma_{\\pi}^{vac}\\simeq 0.58\\cdot 10^{-7}m_{\\pi}$. They argued that in supernova cores pion states are damped mostly by nucleon absorption rather than by the free decay so that the width in the medium ($\\Gamma_{\\pi}^{med}$) is much larger than in vacuum and the pion-pole mechanism should be strongly suppressed. Besides, in the medium the process may go on massless left-handed neutrinos  through the intermediate nucleon particle $-$ nucleon hole states, i.e. as  $\\gamma\\gamma \\rightarrow \\pi^0\\rightarrow nn^{-1}\\rightarrow \\nu_L \\bar{\\nu}_R$ reaction channel. The processes  $\\pi^0_{virt} \\rightarrow nn^{-1}\\rightarrow \\nu_L \\bar{\\nu}_R$  on virtual pions have been discussed in \\cite{VS86}.            Without any doubt the neutrino emission from the dense hadronic component in neutron stars is the subject of strong modifications due to collective effects in the nuclear matter \\cite{dv}, and it is interesting to quantitatively know how much the pion-pole mechanism is influenced by the properties of the dense medium and even, if the process is strongly suppressed, what is the main effect causing this suppression. In this work we compute the cooling rate due to the neutrino emission through the pion-pole mechanism in the reaction $\\gamma\\gamma\\rightarrow\\pi^0 \\rightarrow \\nu_R \\bar{\\nu}_L $ (or $\\nu_L \\bar{\\nu}_R )$ for the conditions of proto-neutron stars including the effects of dense medium in the pion polarization operator (Section 2). Then  we discuss other relevant reaction channels. In order to clarify their connection to the process $\\gamma\\gamma\\rightarrow\\pi^0 \\rightarrow \\nu_R \\bar{\\nu}_L $ (or $\\nu_L \\bar{\\nu}_R )$ we review in the Appendix the optical theorem formalism, see the discussion in \\cite{dv}, that allows to calculate consistently the reaction rates including complicated medium effects. Continuing to study possible physical situations where the pion pole cooling mechanism could be important we consider in Section 3 a possible consequence of the pion pole mechanism in the case of neutron stars with the color superconducting quark cores. We argue that in the color-flavor-locked superconducting phase  for temperatures $T\\lsim (0.1\\div 10)$~MeV the process is feasibly the most efficient neutrino cooling process, although the absolute value of the reaction rate is rather small. Then we draw our conclusion. ",
        "conclusions": "In conclusion, (i) we evaluated the cooling rate of the proto-neutron stars  via the pion pole mechanism taking into account the nuclear medium effects and we showed that with the width effects included this mechanism gives no constraints on the corresponding $\\pi^0 \\rightarrow \\nu_R\\bar{\\nu}_L (\\nu_L \\bar{\\nu}_R)$ decay width. (ii) We also discussed possible contribution of this mechanism to the cooling of different astrophysical systems, as is the case of neutron stars with dense quark cores. We found that  the process $\\gamma\\gamma\\rightarrow \\pi^0\\rightarrow \\nu\\bar\\nu$ is proved to be the most efficient process in the color-flavor-locked superconducting core at temperatures $T\\lsim (0.1\\div 10)$~MeV depending on the effective pion mass and the pion decay width. Depending on what the future tells us about $\\Gamma (\\pi^0 \\rightarrow \\nu \\bar{\\nu})$ and on the possibility of the color superconductivity in neutron stars we believe that these questions deserve a further more detailed study (perhaps even further studying of a dependence on medium effects of all the terms appearing in the cross section of $\\gamma\\gamma \\rightarrow \\pi^0 \\rightarrow \\nu_R\\bar{\\nu}_L (\\nu_L \\bar{\\nu}_R)$)."
    },
    "0208/astro-ph0208154_arXiv.txt": {
        "abstract": "We are using the Isaac Newton Telescope Wide Field Camera to survey open cluster fields for transiting hot Jupiter planets.  Clusters were selected on the basis of visibility, richness of stars, age and metallicity.  Observations of NGC 6819, 6940 and 7789 began in 1999 and continued in 2000.  We have developed an effective matched-filter transit-detection algorithm which has proved its ability to identify very low amplitude eclipse events in real data. Here we present our results for NGC 6819.  We have identified 7 candidates showing transit-like events.  Colour information suggests that most of the companion bodies are likely to be very-low-mass stars or brown dwarfs, intrinsically interesting objects in their own right. ",
        "introduction": "The technique of discovering planets from their photometric transits enables us to probe fainter stars for planetary companions than are accessible to radial velocity surveys.  This allows us to probe a wider variety of stellar environments.  The technique also offers us the possibility of discovering statistically significant numbers of hot Jupiter planets from the ground within a relatively short timescale.  Such a sample could be used to establish the relationships between planetary formation and key properties such as metallicity, age, etc. Janes (1996) suggested that clusters would be ideal targets for a transit search.  They provide large numbers of stars within a relatively small area, the ages and metallicities of which can be determined. Radial velocity (RV) surveys have indicated that stellar metallicity in particular seems to be correlated with the presence of planets.  Brown et al. (2001) reported surveying the globular cluster 47 Tuc with the Hubble Space Telescope, but interestingly found no transits.  In the light of the RV results, this raises the question of whether and how this deficit is related to the low metallicity of the cluster and/or star/radiation density.  To address this question, we are surveying a number of open clusters with a range of ages/metallicities.  Open clusters do not suffer from the dense crowding of stars present in globulars and so can be studied with ground-based telescopes. They are also generally younger and higher in metallicity. ",
        "conclusions": ""
    },
    "0208/astro-ph0208224_arXiv.txt": {
        "abstract": "We present a combined photometric and spectroscopic analysis of the star formation history and space density of a sample of 485 luminous, $M_V - 5 \\log h < -20.5$, field early--type galaxies at redshifts $0.3 \\la z \\la0.6$. The observed $b_{\\rm J}ori$ colours as a function of redshift, mean absorption line strengths and [OII] 3727 emission properties are used to constrain the star formation history of the galaxies.  The mean star formation history of the early--type galaxy sample is consistent with an old ($z_f>1$), passively evolving luminosity--weighted stellar population. Twenty--one percent of the sample possess detectable [OII] 3727 emission consistent with a low level ($\\la 1$ M$_{\\odot}$ yr$^{-1}$) of on--going star formation. Parametric and non--parametric estimates of the space density of the sample are derived. The integrated luminosity density at $z\\simeq 0.4$, allowing only for passive luminosity evolution, is in excellent agreement with the local, ($\\langle z \\rangle = 0.1$), luminosity density of early--type galaxies.  Overall, the sample properties are consistent with a galaxy formation scenario in which the majority of luminous field early--type galaxies formed at redshifts $z>1$ and have largely evolved passively since the formation epoch. ",
        "introduction": "The star--formation history and space density of luminous early--type galaxies in field\\footnote{Throughout this paper, ``field'' is taken to indicate regions outside rich galaxy clusters.} and cluster environments at significant cosmological look--back times place strong constraints upon theories of galaxy formation and evolution. Within the class of hierarchical formation models, early--type galaxies are predicted to result from a combination of major galaxy mergers (i.e. those involving roughly equal mass progenitors) together with the accretion of lower mass galaxies (Baugh, Cole {\\&} Frenk 1996; Kauffmann 1996).  The merger rate and thus the production rate of early--type galaxies is predicted to be a strong function of local galaxy density; early-type galaxies in clusters arise from an early cosmic epoch of rapid merging and intense star formation whereas field early--type galaxies arise from a more protracted period of merging and star formation extending to redshifts $z \\la 1$. Thus, in hierarchical models, field early--type galaxies are predicted to display lower mean ages and a greater dispersion in star formation history with respect to cluster early--type galaxies. For example, Kauffmann (1996) reports that cluster elliptical galaxies (i.e. those located within parent halos of mass $10^{15}$M$_{\\odot}$) display luminosity weighted stellar ages between 8 and 12.5 Gyr compared to ``field'' elliptical galaxies (i.e. those located within parent halos of mass $10^{13}$M$_{\\odot}$) displaying stellar ages between 6 and 11 Gyr. These predictions contrast with those resulting from models in which early--type galaxies observed in both field and cluster environments formed from the monolithic collapse of galaxy--mass (i.e. $10^{11}-10^{12}$M$_{\\odot}$) overdensities at early cosmic times (Larson 1974).  The stellar populations in such galaxies are predicted to have formed in an early, largely coeval, burst of star formation and to have evolved passively since that time.  A important link between theoretical predictions and the observed properties of early--type galaxies is achieved via studies of the Fundamental Plane (hereafter FP) formed by such objects. Studies of the FP for luminous early--type galaxies in clusters now extend to redshifts $z \\sim 0.5$ (e.g. Pahre, Djorgovski {\\&} de Carvalho 1998; J{\\o}rgensen {\\etal} 1999).  The observed evolution of the FP parameters versus redshift in accordance with the predictions of a passively evolving stellar population, combined with a low intrinsic dispersion among the galaxies, imply that the majority of early--type galaxies in rich clusters formed at some early ($z>2$), coeval epoch. Further evidence that luminous early--type galaxies in rich clusters are composed of uniformly old stellar populations is derived from the observation that these galaxies display impressively uniform colours over the redshift interval $0 < z < 1$ (Bower {\\etal} 1992; Ellis {\\etal} 1997; Stanford, Eisenhardt \\& Dickinson 1998). The exceptionally narrow range of colours displayed both internally, and between clusters, at significant cosmological look--back times ($\\sim 8\\,$Gyr) strongly supports an early, coeval formation epoch.  Given the relative abundance of data available for early--type galaxies in rich clusters at redshift $z\\sim0.5$, the generation of samples of field early--type galaxies with which to test the central predictions of galaxy formation theories is of considerable interest. Initial attempts to form the FP relation for distant field early--type galaxies (van Dokkum {\\etal} 2001; Treu {\\etal} 2001) indicate that field early--type galaxies exhibit an FP relation similar in the mean to that observed in rich clusters but displaying an increased scatter about the mean relation. These studies are limited by small sample sizes ($\\sim 20$ galaxies) and include galaxies over an extended redshift interval, $0.1<z<0.5$, leading to a degree of uncertainty due to the need for evolutionary corrections required to bring the galaxies to a common epoch.  The formation of early--type field galaxies via hierarchical merging also implies that their number density should be a strong function of look--back time (Kauffman, Charlot \\& White 1996). However, the failure to demonstrate any deficit in the number density of luminous field spheroidal galaxies at redshifts $z < 1$ compared to local studies places strong constraints upon the proposed epoch of hierarchical merging (Schade {\\etal} 1999).  It is important to note however, that linking the predictions of numerical models directly to observations is hindered by uncertainties associated with galaxy classification (e.g. Abraham 1999).  Differences between the star--formation history of  galaxies in field and cluster environments are likely to be augmented by the fact that the environment of rich clusters appears to actively suppress on--going/recent star--formation. In a study of 10 distant ($z \\sim 0.5$), rich clusters Dressler {\\etal} (1999) and Poggianti {\\etal} \\ (1999) report that 20{\\%} of galaxies possess so--called ``post--starburst'', (a+k), spectra, indicative of a recent cut--off in star--formation activity combined with an old stellar population. The frequency of the a+k class in a comparable field sample is considerably lower ($\\la 4 {\\%}$), although the significance of this result is low given the small size of the field sample. However, the idea that star--formation activity in rich cluster galaxies is either suppressed (via ram--pressure stripping of neutral gas) or rapidly increased at early times (via galaxy--galaxy interactions and mergers) leading to exhaustion of the available gas supply, remains compelling.  While the advent of the Sloan Digital Sky Survey has recently allowed the definition of very large samples of early--type galaxies (Eisenstein {\\etal} 2001, Bernardi {\\etal} 2002) most existing studies of luminous early--type field galaxies over the redshift interval $0 < z < 1$ have been confined to small numbers of objects present in galaxy redshift surveys designed to constrain the global properties of the galaxy population, e.g. the Canadian Network for Observational Cosmology field galaxy redshift survey (CNOC; Yee, Ellingson \\& Carlberg 1996), the Canada France Redshift Survey (Hammer {\\etal} 1997), and the AUTOFIB galaxy redshift survey (Heyl {\\etal} 1997). Given the under representation of luminous early--type field galaxies in current galaxy surveys we have obtained multi--colour photometry and medium resolution optical spectroscopy for a large, well--defined sample of distant, luminous early--type field galaxies (Willis, Hewett {\\&} Warren 2001, hereafter Paper I). The observational data is of sufficient quality to place strong constraints on the star formation history and space density evolution of the population.  By obtaining data for field galaxies with redshifts and luminosities comparable to those studied in rich cluster environments we can perform an important test of the central predictions of hierarchical models of galaxy formation and evolution.  In Section 2 we review the compilation of a spectroscopic sample of 581 luminous field early--type galaxies (Paper I).  The star--formation history of the sample, employing both colour--redshift information and spectroscopic diagnostics is investigated in Section 3.  In Section 4 we calculate the space density of the galaxies and the paper concludes by considering the implications for models of galaxy formation and evolution in Section 5. Unless otherwise indicated, values of $\\Omega_M = 0.3$, $\\Omega_{\\Lambda} = 0.7$ and $h= {\\rm H_0} / 100$ km s$^{-1}$ Mpc$^{-1}$ are adopted for the cosmological parameters describing the evolution of a model Friedmann--Robertson--Walker universe.  \\section[]{The galaxy sample}  A full description of the galaxy catalogue employed in this paper is given in Paper I but a summary of the main points is included here for completeness. A total of $9599$ objects were selected in an area of 220 deg$^2$ spread over 7 United Kingdom Schmidt Telescope (UKST) fields at high Galactic latitude (Paper I; Section 2). The effective area of the survey is 198.0 deg$^{2}$ and the surface density of candidate early--type galaxies at the faint magnitude limit ($i<18.95$) is $\\simeq 50 \\, {\\rm{deg}}^{-2}$.  Magnitudes, in the natural UKST $b_{\\rm J}ori$ photometric system, calibrated using $8\\arcsec$ diameter aperture magnitudes from CCD observations, are available for all objects and satisfy the magnitude and colour selection criteria summarised in Table 1.  \\begin{table} \\caption{Photometric selection criteria.} \\label{tab_photo_select} \\begin{tabular}{ccccc} 16.5 & $<$ & $m_i$ & $<$ & 18.95 \\\\ 2.15 & $<$ & $ b_{\\rm J} - or $ & $<$ & 3.00 \\\\ & & $ or - i $ & $<$ & 1.05 \\\\ 2.95 & $<$ & $ b_{\\rm J} - i $ & & \\\\ \\end{tabular} \\end{table}  The early--type galaxy sample was constructed employing a combination of photometric and morphological criteria. The photometric selection criteria shown in Table \\ref{tab_photo_select} were chosen to include the predicted colours of a passively evolving early--type galaxy over the redshift interval $0.3<z<0.55$ while reducing contamination from main sequence K-- and M--type stars. Images on each photographic plate were classified as stellar or non--stellar on the basis of a classification parameter (CP) computed from the surface brightness profile (Paper I; Section 2.1). Application of a suitable rescaling value determined for each plate expresses the CP value as the number of standard deviations separating each object from the centre of the stellar locus. Averaging the individual CP values over all plates on which a particular object is detected then permits robust separation of the stellar locus from the distribution of galaxian objects down to the magnitude limit $i = 18.95$. The candidate early--type galaxy sample was constructed by selecting the set of objects that satisfy the photometric criteria described in Table \\ref{tab_photo_select} and display CP values $> 1.5$. In order to constrain the level of incompleteness in the early--type galaxy sample introduced by these criteria, a sample of candidate stellar objects satsifying the above photometric criteria and CP $<1.5$ was also constructed.  Spectroscopy of a sub--sample of $\\sim 600$ galaxies was obtained during 1997--1998 using the Two Degree Field (2dF) facility at the 3.9--m Anglo--Australian Telescope (Paper I; Section 3). The contamination of the sample by stars is extremely small, $\\sim 2\\%$, while the spectroscopic redshift completeness rate, excluding the few stars, is very high $\\ga 95\\%$. Indeed, the majority of the incompleteness in the spectroscopic identifications is due to instrumental or astrometric problems unrelated to the intrinsic nature of the target objects, i.e. close to $100\\%$ of the galaxies in the photometric sample are early--type galaxies with redshifts $0.25 < z < 0.63$. In addition to observations of early--type galaxy candidates, a sample of 33 stellar candidates were included in the 1998 September 16--17 observations in order to determine directly the level of incompleteness in the early--type galaxy sample. All of the 33 stellar candidates display spectra consistent with that of a K-- or M--type star, thus constraining the level of incompleteness in the early--type galaxy sample to $<3${\\%}.  The total of 581 galaxy redshifts was compiled from a number of observing runs, including some commissioning observations where the data quality was not ideal. In order to ensure a high degree of uniformity the analysis in this paper is confined to 485 galaxies observed in a single two--night run with 2dF on 1998/09/16--17. This sample of 485 galaxies with redshifts is in turn divided into two sub--samples: Sample A, consisting of 371 objects, contains all spectra with reliably determined continuum shapes, and Sample B, consisting of all 485 objects. The necessity for the two sub--samples arose from two factors: the contamination of some of the galaxy spectra by light from an uncovered LED (59 objects) and the identification of a ``tail'' in the distribution of galaxy continuum spectral properties arising from instrumental/reduction uncertainties (55 objects). The contamination and reduction uncertainties did not impair the assignment of redshifts but the continuum properties of the galaxies so affected are not reliable. The investigation of continuum--dependent properties, such as the generation of the composite spectrum, is confined to the 371 galaxies in Sample A.  A comparison of the effective volume sampled by the spectroscopic early--type galaxy sample to the typical space density of rich galaxy clusters indicates that the early--type galaxy sample described in Paper I is overwhelming drawn from the field. A direct estimate of the effective survey volume is presented in Section \\ref{sec_vacc} and is computed as the mean accessible volume for galaxies drawn from spectroscopic sample B occupying the redshift interval $0.28<z<0.6$. Comparing the effective survey volume to the observed space density of rich galaxy clusters drawn from a similar redshift interval indicates that the anticipated fraction of cluster early--type galaxies within the sample is $< 5-10${\\%}.  One of the principal conclusions of Paper I was that virtually all the galaxy spectra display properties that are entirely consistent with observations of local galaxies (Paper I; Section 4). The composite early--type galaxy continuum is strikingly similar to a local composite early--type galaxy spectral energy distribution (SED) drawn from the atlas of Kinney {\\etal} (1996). However, there are some identifiable differences. Eighty--one objects in Sample A (104 objects in Sample B) possess detectable [OII] 3727 emission (Paper I; Section 4.3) and this sub--sample of galaxies exhibiting detectable [OII] 3727 emission is henceforth referred to as ``[OII]--emitting''. The larger sub--sample of galaxies (290 in Sample A; 381 in Sample B) with no detectable [OII] 3727 emission are henceforth referred to as ``Normal''. Possible biases against the detection of passively evolving early--type galaxies that have experienced a recent star formation episode are considered in detail in Section \\ref{sec_col_effect}. However, the lack of any significant colour differences between the Normal and [OII]--emitting early--type galaxy samples combined with the low star formation rates implied by the observed [OII] 3727 emission fluxes suggests that no significant bias exists. Indeed, the [OII] 3727 emission line luminosity is function is sufficiently well constrained to reject the existence of a population of intermediate star forming early--type galaxies at the limit of the photometric selection criteria.  The observed number--redshift, N($z$), distribution of all early--type galaxies with reliable redshifts (Sample B) is consistent with the distribution predicted by applying the survey selection criteria to a model early--type galaxy population described by a representative spectral evolution and luminosity function model of Pozzetti {\\etal} (1996). The extreme range of rest--frame, $V$--band, absolute magnitudes present in the sample is $-22.5 \\la M_V - 5 \\log h \\la -20.5$ with an inter--quartile range of 0.55 magnitudes (Section \\ref{sec_mag_est}).  The early--type galaxy photometry presented in Paper I does not include a correction for Galactic extinction. To facilitate a comparison with recent results in the literature (Section 4.4) extinction corrections from Schlegel, Finkbeiner {\\&} Davis (1998) have been made to the galaxy magnitudes. Total to selective extinction values of $A/E(B-V)$ = 4.035, 2.649 and 1.893 have been adopted to correct galaxy magnitudes in the $b_{\\rm J}$--, $or$-- and $i$--bands respectively.  \\section[]{Star formation history}  \\subsection[]{Colour--redshift evolution} \\label{sec_colour}  Determining the possible range of age and metallicity consistent with the observed evolution of broad--band colours places important constraints upon the star formation history of the luminosity--weighted stellar population. Most simply, to what extent does the colour--redshift distribution for the galaxy sample conform to the predictions of a single, evolving stellar population model? Additional information is also present in the distribution of galaxy properties, specifically, what are the physical implications of the existence of the galaxy sub--samples, Normal and [OII]--emitting, defined in Section 2 and how homogeneous are the galaxies within each sub--sample?  The broad--band colours of the galaxy sample are compared to a range of model stellar populations based upon the GISSEL96 spectral synthesis code (Leitherer {\\etal} 1996). Two fiducial stellar population models are considered; an instantaneous burst and an exponentially decreasing burst of star formation of e--folding time scale $\\tau=1\\,$Gyr. Each model is described by a Scalo (1986) Initial Mass Function (IMF) over the mass interval 0.1 to 125 M$_{\\odot}$. The colour--redshift evolution predicted for each model as a function of formation redshift and metallicity is investigated within a spatially flat ($\\Omega_M=0.3$, $\\Omega_{\\Lambda}=0.7$, $h=0.7$) and an open ($\\Omega_M=0.3$, $\\Omega_{\\Lambda}=0.0$, $h=0.7$) model universe.  At the median redshift of the galaxy sample ($\\langle z \\rangle \\simeq 0.4$) the age of the universe in the spatially flat and open models is approximately $9\\,$Gyr and $7\\,$Gyr old respectively.  Colour--redshift loci on the $b_{\\rm J}-or$ versus $or-i$ plane are compared to observed colour--redshift values via the two--colour probability density function $G (b_{\\rm J}-or,or-i \\, | \\, z)$ (Appendix \\ref{app_mod_col}). The function $G$ describes the two--colour probability density of a simulated population of galaxies at a redshift $z$, characterised by a spectro--photometric model computed for a given formation redshift and metallicity, combined with the photometric uncertainties in the $b_{\\rm J}ori$ passbands.  Hence, observed $b_{\\rm J}ori$ colour distributions are compared to a simulated distribution of galaxy colours described by the formation redshift ($z_f$) and metallicity ($\\log [Z/Z_{\\odot}]$) of the spectro--photometric evolution model. Following the maximum likelihood technique of Sandage, Tammann \\& Yahil (1979), one searches for individual values of $z_f$ and $\\log [Z/Z_{\\odot}]$ that maximise the combined probability density for realising the observed galaxy colour--redshift values.  Let $p(u,v \\, | \\, z)$ be the differential probability density that a galaxy of observed redshift $z$ will display colours in the interval $u=b_{\\rm J}-or$ to $u+{\\rm d}u$ and $v=or-i$ to $v+{\\rm d}v$, i.e.  \\begin{equation} { p(u,v \\, | \\, z) \\propto \\frac{G(u,v \\, | \\, z)}{\\int_{-\\infty}^{\\infty} \\int_{-\\infty}^{\\infty} G(u^{\\prime},v ^{\\prime}\\, | \\, z) H(u^{\\prime},v^{\\prime}) \\, {\\rm d}{u^{\\prime}} \\, {\\rm d}{v^{\\prime}}}. } \\end{equation}  {\\noindent}Where $H(u,v)$ represents a two--dimensional window function equal to unity when the galaxy colour lies within the colour--selection boundaries (Section 2) and equal to zero when the colour lies outside this region, i.e.  \\begin{eqnarray} \\label{eqn_col_window} H &=& 1 \\;\\; \\mbox{  if    } \\; 2.15 \\le u \\le 3.00 \\\\\\nonumber && \\,\\,\\, \\;\\; \\mbox{  and  } \\; 2.95 - u \\le v \\le 1.05 \\\\ \\nonumber H &=& 0 \\;\\; \\mbox{  otherwise.} \\nonumber \\end{eqnarray}  {\\noindent}Equation \\ref{eqn_col_window} describes the case $E(B-V)=0$. For the case $E(B-V)>0$, early--type galaxy $b_{\\rm J}ori$ photometry and photometric selection limits are adjusted accordingly. The likelihood--term is then constructed from the combination of individual probability density values estimated for the $k=1, N$ galaxies in the sample for the given combination of parameters ($z_f$, $\\log [Z/Z_{\\odot}]$), i.e.  \\begin{equation} { \\ln {\\mathcal L} = \\sum_{k=1}^{N} \\ln p \\, (u_k,v_k \\, | \\, z_k)  } \\end{equation}  {\\noindent}Maximising the likelihood--term over the defined two--dimensional parameter space equates to determining the most probable combination of input parameters to generate the observed colour--redshift values. Confidence intervals on the returned parameter values may be estimated by determining the surfaces of likelihood defined by the expression  \\begin{equation} { \\ln {\\mathcal L} = \\ln {\\mathcal L}_{max} - {\\frac{1}{2}} {\\chi}_{\\beta}^{2} (M), } \\label{eqn_max_err} \\end{equation}  {\\noindent}where ${\\chi}_{\\beta}^{2} (M)$ is the $\\beta$--point of a $\\chi^2$ distribution with $M$ degrees of freedom (Efstathiou, Ellis \\& Peterson 1988). Contours describing a $4\\sigma$ likelihood difference relative to the maximum likelihood value are evaluated. This criterion is adopted to encompass known uncertainties in the optical/infrared colours of evolving galaxies predicted by current population synthesis models (Charlot, Worthey {\\&} Bressan 1996). Such likelihood surfaces provide constraints only on the relative likelihood of various parameter combinations.  Figure \\ref{age_metal_red} displays relative likelihood contours in the $z_f$ versus $\\log [Z/Z_{\\odot}]$ plane generated for the Normal galaxy sub--sample. The likelihood contours for each model identify a well--defined locus of star formation scenarios ranging from young, above--solar metallicity stellar populations ($z_f=1$, $\\log [Z/Z_{\\odot}]=0.2$) to old, solar metallicity populations ($z_f>3$, $\\log [Z/Z_{\\odot}]=0.0$). The properties of [OII]--emitting galaxies are statistically identical to the Normal sub--sample; the relative likelihood contours (not shown) are centered on the same region in the projected $z_f$ versus $\\log [Z/Z_{\\odot}]$ plane but extend over a larger area due to the smaller sample size.  Although no colour--magnitude (CM) relation for early--type galaxies has been incorporated, the effect on the analysis is anticipated to be negligible. In their analysis of early--type galaxies in cluster CL 1358+68 ($z\\simeq0.33$), van Dokkum {\\etal} (1998) report a CM--relation $(B-V)_z = 0.866-0.018(V_z-20.7)$.  At $z=0.4$ their redshifted $B$ and $V$ bands (i.e. subscript $z$) approximately sample observed frame $or$ and $i$. The absolute $V$--magnitude range in our sample is relatively small; the inter--quartile range is 0.54 magnitudes (Section \\ref{sec_mag_est}). Therefore, one anticipates the introduction of an additional dispersion term in the early--type galaxy colours of $\\sim 0.01$ magnitudes, i.e. small compared to the colour distribution resulting from both the observed uncertainties and estimated intrinsic dispersion (see below).  \\begin{figure} \\psfig{figure=./fig1.ps,height=7.5cm,angle=270.0} \\caption{Contours of equal likelihood in the log metallicity, $\\log [Z/Z_{\\odot}]$, versus formation redshift, $z_f$, plane generated from the colour--redshift analysis of the Normal sub--sample.  Each contour delineates the $4\\sigma$ likelihood region associated with each model; exponentially declining burst (solid line), instantaneous burst (dashed line), flat universe (heavy line), open universe (light line).} \\label{age_metal_red} \\end{figure}  The likelihood contours presented for each model in Figure \\ref{age_metal_red} indicate that the luminosity--weighted stellar populations describing the early--type galaxy sample are consistent with a range of evolving stellar populations where formation redshift, metallicity and/or star formation history vary to some degree.  Should the galaxies form over a range of redshifts, then the luminosity--weighted metallicity varies systematically as a function of formation epoch, with more metal rich galaxies formed at later epochs.  If star formation history is characterised by a sequence of bursts then shorter bursts are necessary to match the observed colour--redshift distribution at lower redshifts. The constraints on the duration of bursts at high redshift are poor because the colours of stellar populations more than $\\sim 4\\,$Gyr old are extremely similar.  More complex behaviour involving variation of all three model parameters are of course possible.  Ferreras, Charlot {\\&} Silk (1999) present a markedly similar result, from a sample of 599 morphologically classified early--type (E/S0) cluster galaxies observed at comparable redshift and luminosity. Their Figure 3(a) may be most directly compared to the instantaneous burst model realised within an open universe presented in Figure \\ref{age_metal_red}. The result of Ferreras {\\etal} (1999), and that presented here, indicate that the luminosity--weighted stellar populations of field and rich cluster galaxies were formed over a comparable range of redshift (i.e. $z>1$) and metallicity ($\\log [ Z/ Z_{\\odot}]=0.0 - 0.2$).   The observed dispersion of early--type galaxy colours about a given model locus results from the observed  photometric uncertainties and from an intrinsic component including some combination of age, metallicity and star formation history variations. The amplitude of the intrinsic dispersion component for a particular colour was estimated by first calculating the expectation value of the distribution of galaxy colour deviations from a given model colour locus, normalised by the photometric uncertainty, i.e. $\\langle F_{colour} \\rangle$. For example, considering the $b_{\\rm J}-or$ colour term one writes \\begin{multline} F^2_{(b_{\\rm J}-or), k} = {[(b_{\\rm J}-or)_k - (b_{\\rm J}-or)_{model}(z_k)]}^2 \\\\ / \\, \\sigma^2_{(b_{\\rm J}-or),k}, \\label{frac_expect} \\end{multline} for a particular galaxy $k$ of observed colour $(b_{\\rm J}-or)_k$, where $(b_{\\rm J}-or)_{model}(z_k)$ is the colour predicted by the selected model locus at the redshift $z_k$ of the galaxy and $\\sigma_{(b_{\\rm J}-or),k}$ is the observed photometric uncertainty. To simplify the calculation of the colour--deviation statistic, only galaxies that deviate from a particular colour locus in the sense that they fall on the side of the locus opposite to the colour--cut that defines the sample selection are included. This limits the calculation to 270, 261 and 241 galaxies in the $b_{\\rm J}-or$, $or-i$ and $b_{\\rm J}-i$ colours respectively, but ensures that the measured dispersions are largely unaffected by the photometric selection criteria (Table 1).  Given the normalised deviation of objects from a particular colour locus averaged across the sample, the ``typical'' intrinsic contribution per galaxy may be calculated by comparing the observed deviation to the photometric uncertainty for any colour, i.e. \\begin{equation} \\sigma_{int}^2 = (F \\sigma_{obs})^2 - \\sigma_{obs}^2, \\label{sigma_int} \\end{equation} where $\\sigma_{int}$ is the intrinsic colour dispersion and $\\sigma_{obs}$ is the median photometric uncertainty associated with the selected colour.  Figure \\ref{spread_points} compares the intrinsic spread calculated for the $b_{\\rm J}-or$ versus $or-i$ plane to colour variations generated by a given stellar population model when varied about central values of formation redshift and metallicity. In addition, the dispersion resulting from varying the assumed star--formation history between an instantaneous burst model and a model employing an exponentially decaying burst of star formation, described by an e--folding time scale of $\\tau = 1$ Gyr, is indicated as a function of formation redshift at fixed metallicity. Figure \\ref{spread_points} indicates that the intrinsic colour dispersion observed in the early--type galaxy sample is consistent with only a very modest spread in formation redshift, metallicity or star formation history. For example, the intrinsic colour variations are consistent with an early--type galaxy formation redshift varying over the interval $2<z<3$ at constant metallicity.  Conversely, at fixed formation redshift, the intrinsic colour variations are consistent with a metallicity interval $0.0 < \\log [ Z/Z_{\\odot}] < 0.2$. Though more complex scenarios involving the effect of combined variations of formation redshift, metallicity and star formation timescale are also allowed by the data, the low root mean square amplitude of colour variations about a given model locus implies that the early--type galaxy sample describes a predominantly uniform population drawn from a narrow range of instrinsic properties.  \\begin{figure} \\psfig{figure=./fig2.ps,height=7.5cm,angle=0.0} \\caption{Amplitude of intrinsic variation of early--type galaxy colours due to varying model stellar population parameters.  The solid square with error bars indicates the intrinsic dispersion values, calculated employing Equations \\ref{frac_expect} and \\ref{sigma_int}, for the galaxy sample.  The extent of the error bars represents the variation in the intrinsic dispersion when the colour errors enclosing 68{\\%} of the galaxy distribution about the median value are employed instead of the median value.  The fiducial model consists of an exponentially declining burst of star formation of timescale 1 Gyr formed at redshift $z=3$ with solar metallicity.  Open triangles indicate the colour sequence generated by varying the metallicity of the fiducial model by $+0.05,\\,0.1,\\,0.15$ and $0.2$ in log metallicity from bottom left to top right. Open squares indicate the colour variation produced by changing the formation redshift of the fiducial model to $z=2.7, 2.4, 2.1$ and $1.8$, again, ordered from bottom left to top right. All simulations are considered for a ``flat'' universe model (see text). All colours are computed for model galaxies at $z=0.4$, the median redshift of the early--type galaxy sample.} \\label{spread_points} \\end{figure}     \\subsection[]{[OII] 3727 emission} \\label{sec_opt_col}  \\subsubsection[]{Star formation or active galactic nuclei?}  [OII] 3727 emission is detected in 104 galaxies ($\\sim 25${\\%}) of galaxies from Sample B\\footnote{Although the spectra of a small number of galaxies in sample B possess continua that may be affected by contaminating LED emission (Section 2), this does not affect the detectability of [OII] emission and the discussion of the [OII] emission properties is based on galaxies in the larger Sample B.} (Paper I; Section 4.3). Star--formation or the presence of an active galactic nucleus could be responsible for the presence of [OII] 3727 emission.  Several lines of evidence suggest that the origin of the emission in the majority of galaxies is star formation activity. The majority of active galactic nuclei (AGN) exhibit relatively strong [OIII] 5007, with the flux ratio [OIII] 5007/[OII] 3727 $\\ga 3$ (e.g. Baldwin, Phillips \\& Terlevich 1981). However, a composite spectrum of the 104 galaxies with detectable [OII] 3727 emission shows no evidence for the presence of the [OIII] 4959 5007 doublet or of H$\\beta$ in emission. Indeed, in the entire galaxy sample there are no individual detections of [OIII] 5007 or H$\\beta$ in emission. The flux limit for detection of [OIII] 5007 in the galaxy spectra is a factor $\\simeq 1.5$ poorer than for [OII] 3727, due primarily to the higher sky--background at $\\sim 7000$\\AA. The lack of detections of [OIII] 5007 is thus surprising if the line ratios were typical of AGN but is entirely consistent with the identification of the source of [OII] emission as low--excitation star formation where the [OIII] 5007/[OII] 3727 ratio is often unity or significantly smaller.  An estimate of the fraction of the galaxy sample hosting AGN can be made from the number of galaxies that are radio sources. The excellent astrometric properties of the galaxy sample (Paper I; Section 2) and the FIRST radio survey (Becker, White \\& Helfand 1995) allow unambiguous identification of galaxies with radio sources using a matching radius of only $2\\,$arcsec. The FIRST survey includes a significant proportion of the three UKST equatorial fields included in the full photometric catalogue of $9599$ candidate galaxies. Of 3170 galaxy candidates within the FIRST survey region, 95 lie within $2\\,$arcsec of a FIRST source. A further 6 galaxies posses multiple FIRST sources, within a $10\\,$arcsec radius, that strongly suggest the galaxy is the origin of radio--lobe emission.  Thus, only 101 of 3170 ($\\simeq 3\\%$) of the galaxies appear to be radio sources to $1\\,$mJy at $1.4\\,$GHz, compared to the much larger percentage showing evidence for [OII] 3727 emission.  The FIRST survey does not cover the southern fields where the 2dF spectroscopy was obtained but the NVSS (Condon {\\etal} 1998) extends to declination $-40^{\\circ}$, including sources to $2.5\\,$mJy at $1.4\\,$GHz. Unfortunately, the positional accuracy of the NVSS is such that a $15\\,$arcsec pairing radius is necessary when matching the catalogue to the galaxy sample. Comparison of the matches between the galaxy sample and the NVSS and FIRST surveys in the three equatorial regions shows excellent agreement. NVSS includes all but the galaxies corresponding to the faintest FIRST sources with the addition of an extra $\\sim 30\\%$ of spurious ``matches'' due to the much larger pairing radius employed. Applying the same matching radius to the galaxies included in the 2dF spectroscopy produces 14 galaxy--NVSS matches out of 421 galaxies within the NVSS survey area. Ten of the NVSS--galaxy matches are expected to represent physical association, with 4 galaxies paired by chance. The galaxies in the spectroscopic sample thus show a very similar incidence of radio emission to that inferred using the FIRST survey.  Of the 14 NVSS matches none are included among the 104 [OII] 3727 emission line objects and the composite spectrum of the 14 objects shows no evidence for any anomalous features. Thus, based on the observed limits to the flux ratio of the [OII] 3727 and [OIII] 4959 5007 emission lines and the small incidence of radio emission among the galaxy sample we conclude that star--formation activity, rather than AGN, is responsible for the majority of the [OII] 3727 emission detected in the galaxy spectra.  \\subsubsection{The [OII] 3727 luminosity function and evolution of the SFR}  [OII] 3727 emission arises within HII regions ionised by populations of massive, short--lived O-- and B--type stars (Hogg {\\etal} 1998; Osterbrock 1989). Given the short lifetime of such populations ($\\sim 10^6$ yrs), [OII] 3727 emission offers an almost instantaneous snapshot of the current star--formation rate on cosmological time--scales.\\footnote{ While this statement is qualitatively true, the calculation of the true star formation rate requires further information, including the Initial Mass Function. The strength of [OII] 3727 emission resulting from an ionising field is also sensitive to the physical state of the interstellar medium -- particularly the electron density and metallicity.} The luminosity of an [OII] 3727 emission line of flux $f$ in a galaxy observed using 2dF at redshift $z$ is \\begin{equation} { {\\rm{L}}_{{\\rm[OII]}} = 4 \\pi d_L^2(z) ( f/A ), } \\label{eqn_l_OII} \\end{equation} where $d_L(z)$ is the luminosity distance corresponding to a redshift $z$ in the specified cosmological model. $A$ is the fraction of total source flux intercepted by a 2dF fibre. The form of the correction for the fraction of [OII] 3727 emission intercepted by a fibre is based on the assumption that the distribution of continuum and [OII] 3727 emission in the galaxies is co--spatial. The observed emission line fluxes were measured by fitting a Gaussian model to each continuum--subtracted line (Paper I; Section 4.3).  Kennicutt's (1992) relation, derived from a sample of local galaxies, between mean [OII] 3727 luminosity and star--formation rate (SFR);  \\begin{equation} { \\rm{SFR} (\\rm{M}_{\\odot} \\rm{yr}^{-1}) = \\frac{L_{[OII]}}{5 \\times 10^{41}} \\; ({\\rm ergs} \\; {\\rm s}^{-1}). } \\label{eqn_ken} \\end{equation} is used to estimate the SFR.  [OII] 3727 emission luminosity, rather than [OII] 3727 equivalent width, is employed because, while equivalent width provides a more straightforward indicator of emission relative to the ionising continuum, any sky--subtraction errors introduce significant uncertainties into measurements of equivalent width.  \\begin{figure} \\psfig{figure=./fig3.ps,height=7.5cm,angle=270.0} \\caption{SFR, estimated from [OII] 3727 luminosity, versus redshift for 104 galaxies with detectable emission. Open squares represent data obtained on Night 1 (1998/9/16), filled squares represent data obtained on Night 2 (1998/9/17). The solid and dotted lines indicate the detection limit for nights 1 and 2 respectively.} \\label{plot_sfr} \\end{figure}  Figure \\ref{plot_sfr} displays the distribution of SFR versus redshift, demonstrating that SFRs inferred from [OII] 3727 luminosities are small ($\\la  1$ M$_{\\odot}$ yr$^{-1}$). The evolution of [OII] 3727 emission, and by implication the SFR, as a function of redshift may be investigated by calculating the [OII] 3727 emission--line luminosity function for the galaxy sample at different redshifts. The nature of the sample is well--suited to the application of the $1 / \\, V_{acc}$ method (Avni \\& Bahcall 1980). The maximum accessible volume associated with each [OII]--emitting galaxy in the sample is calculated by integrating the co--moving volume element per unit solid angle, ${\\rm{d}}V / \\, {\\rm{d}}z$, multiplied by the photometric selection probability as a function of redshift, $W(z)$ (Equation \\ref{eqn_wz}), over the redshift interval within which the estimated [OII] 3727 luminosity exceeds the detection threshold for the identification of [OII] 3727 emission in the galaxy spectrum, i.e. \\begin{equation} { V_{acc} = c \\; {\\rm{d}}\\Omega \\int_{z_{min}}^{z_{({\\rm{L_{[OII]}}}\\geq {\\rm{L_{lim}}}(z))}} {\\frac{{\\rm{d}}V}{{\\rm{d}}z} \\, W(z) \\, {\\rm{d}z},} } \\label{eqn_acc_OII} \\end{equation} The factor d$\\Omega$ scales the accessible volume by the solid angle covered by Sample B (d$\\Omega \\simeq 10.2$ deg$^2$) and the factor $c$ corrects for redshift incompleteness and is equal to the inverse of the redshift completeness rate as a function of apparent $i$--magnitude (Paper I; Section 3.5). The cumulative space density of [OII]--emitting galaxies displaying $\\rm{L_{[OII]}} > L$, $\\Phi_{\\rm{[OII]}}(>{\\rm{L}})$,  is obtained by summing the inverse accessible volume of each galaxy.  Evolution of the early--type galaxy SFR as a function of redshift was investigated by computing the cumulative space density of [OII]--emitting galaxies as a function of $\\rm{L_{[OII]}}$ in three redshift shells ($0.28<z<0.37$ (36 galaxies), $0.37 \\le z < 0.43$ (32 galaxies), and $0.43 \\le z < 0.6$ (31 galaxies)), chosen to include approximately equal numbers of galaxies. The photometric selection function was computed employing a model described by exponentially decaying burst of star formation of e--folding time scale $\\tau=1\\,$Gyr formed at $z_f=3$ with metallicity $\\log [Z/Z_{\\odot}]=0.0$ in a spatially flat universe ($\\Omega_M=0.3$, $\\Omega_{\\Lambda}=0.7$, $h=0.7$). The enclosed volume associated with each [OII]--emitting galaxy, $V_{enc}$ (Section 4.2), was also calculated. The resulting cumulative distributions, Figure \\ref{vacc_OII}, show a marked increase ($\\Delta \\log \\Phi \\sim 0.5$) in the space density of bright ($\\rm{L_{[OII]}} > 10^{41}$ ergs s$^{-1}$) [OII] 3727--emitters at redshifts $z>0.43$.  Any difference in the luminosity function between the two lower redshift shells depends on the exact placement of the redshift boundary (at $z\\sim0.37$) and a larger sample is required to draw quantitative conclusions. However, the significant increase in the space density of bright ($\\rm{L_{[OII]}} > 10^{41}$ ergs s$^{-1}$) [OII]--emitters at redshifts $z>0.43$ is robust and does not vary significantly as the $z\\sim0.43$ shell boundary is perturbed.  \\begin{figure} \\psfig{figure=./fig4.ps,height=7.5cm,angle=0.0} \\caption{Cumulative space density of [OII]--emitting galaxies versus [OII] 3727 luminosity. Data for three redshift shells $0.28<z<0.37$ (solid line), $0.37 \\le z <0.43$ (dotted line) and $0.43 \\le z < 0.6$ (dashed line) are shown.} \\label{vacc_OII} \\end{figure}  \\subsubsection[]{The effect of star--formation on the colour selection} \\label{sec_col_effect}  The conclusion that the fraction of early--type galaxies possessing bright ($\\rm{L_{[OII]}} >10^{41}$ ergs s$^{-1}$) [OII] 3727 emission increases significantly for redshifts $z \\ga 0.4$ is a potentially important result (Section \\ref{sec_sf_conc}). However, it is necessary to verify that the presence of star formation activity does not lead to a bias in the selection of galaxies as a function of redshift that might produce a spurious evolutionary trend.  Consider a galaxy with a passively evolving stellar population formed in an exponentially--decaying burst of star formation ($\\tau=1\\,$Gyr), with solar metallicity, initiated at redshift $z_f = 3$, that later experiences a burst of star--formation, occurring at $z_b \\sim 1$. For the adopted cosmological parameters, the age of the burst at the epoch of observation ($z \\simeq 0.4$) is $4\\,$Gyr. The star--burst is assumed to involve a population of solar metallicity with a Scalo IMF and an e--folding time--scale $\\tau = 1\\,$Gyr. Model predictions for mass fractions equal to 10, 5, 1 and 0.1{\\%} of the total stellar mass in the galaxy were generated using the GISSEL96 code.  Figure \\ref{model_burst} shows the predicted behaviour of a galaxy experiencing a burst of star--formation, as described above, in the $b_{\\rm J}-or$ versus $or-i$ plane as a function of redshift. Modest star--bursts (0.1{\\%} and 1{\\%} of the total stellar mass) display colours over the redshift interval $0.3 \\la z \\la 0.55$ that are very similar to the underlying model early--type galaxy population ($\\Delta (b_{\\rm J}-or) \\la 0.15$). More massive star--bursts (5{\\%} and 10{\\%}) produce markedly bluer colours at redshifts $z \\ga 0.3$. However, it is a feature of all rapidly fading burst models that optical colours ``synchronise'' to those of an old stellar population at post--burst ages $t_b \\sim 4 - 5$ Gyr ($z \\sim 0.25$ within the current model), explaining the tight locus formed by the models at $z<0.3$.  \\begin{figure} \\psfig{figure=./fig5.ps,height=7.5cm,angle=0.0} \\caption{Predicted $b_{\\rm J}-or$ versus $or-i$ colours as a function of redshift for model early--type galaxies experiencing a star--burst at redshift $z_b = 1$ (see text for details). Symbols along the tracks are at increments $\\Delta z = 0.1$. The tracks terminate, on the right, at $z_{max}=0.6$. Note the colour ``synchronisation'' at $z \\sim 0.3$ for all models.} \\label{model_burst} \\end{figure}  A star--burst at $z_b \\sim 1$ causes the early--type galaxies to become bluer in $b_{\\rm J}-or$, moving the galaxies towards the selection boundaries, with the effect most pronounced at the highest redshifts. Thus, any burst of star--formation corresponding to $\\ga 1\\%$ of the total stellar mass would result in early--type galaxies at redshifts $z>0.4$ becoming lost from the sample.  We conclude that the observed increase in the fraction of galaxies with high [OII] 3727 at redshifts $z \\ga 0.4$ is robust and can only increase in the event that a fraction of early--type galaxies experienced massive bursts.   The similarity of the distributions of the Normal and [OII]--emitting galaxies in the colour--redshift plane, combined with the low--level of the [OII] 3727 emission present, constrains the fraction of the stellar mass involved in a star--burst at $z_b \\sim 1$ for the majority of the sample to be $\\la 0.1\\%$. However, the degree of bluing is a strong function of the mass of the burst and it is not possible to exclude the possibility that some fraction of early--type galaxies experience massive $\\ga 5\\%$ bursts of star formation at $z_b \\sim 1$. Redshift $z \\sim 1$ is the upper limit for which useful constraints may be placed upon star--burst activity using the current sample. Figure 4 illustrates why the signature of bursts occurring at higher redshift are difficult to identify due to the strong degree of colour synchronisation that takes place some $4\\,$Gyr after the bursts. Conversely, the limits on the mass of bursts of star--formation occurring at redshifts $z \\la 1$ are even tighter than the $\\sim 0.1\\%$ limit deduced for a burst redshift of $z_b = 1$.  \\subsection[]{Absorption line indices} \\label{sec_absorb_index}  The measurement of spectral absorption line indices of old stellar populations in early--type galaxies provides a probe of both relative age and metallicity of the luminosity weighted stellar population. Absorption line measures are computed using the Lick/IDS line strength indices (Worthey {\\etal} 1994). The dependence of the Lick/IDS indices, defined by central bandpass and adjacent pseudo--continua, on the age and metallicity of the underlying stellar population have been investigated extensively (Worthey 1994).  Absorption line strengths in the composite spectra of the Normal and [OII]--emitting galaxies were compared. The composite spectrum of each sub--sample was matched to the resolution of Lick/IDS spectra via smoothing with a Gaussian filter of wavelength--dependent FWHM (see Appendix A of Worthey \\& Ottaviani, 1997). This ensures that the indices measured are on the same scale as those from the Lick/IDS data, to within a small additive constant. The dependence of the indices on galaxy velocity dispersion is also a potential concern.  Kuntschner (2000) attempts to correct for galaxy velocity dispersion in the measurement of Lick/IDS indices on early--type galaxies and report corrections $\\la$5{\\%} at velocities $\\la300\\,$kms$^{-1}$ in the value of the H$\\beta$ and C4668 indices employed in this paper (see below). Given that the velocity dispersion correction is small and is difficult to measure accurately without stellar calibration exposures, we do not attempt such a correction here. Errors in the measured absorption indices measured were determined via a bootstrap procedure: absorption indices were measured in composite spectra drawn (with replacement) from the Normal sub--sample. A determination was made for each sub--sample realisation, with the number of galaxies contributing to each realisation equal to the number in each of the galaxy sub--samples.  One--$\\sigma$ errors were calculated from the resulting distributions of absorption indices measured for 1000 such realisations.  The models of Worthey {\\etal} (1994) predict index values for a passively--evolving, instantaneous--burst stellar population of given age and metallicity. Kuntschner \\& Davies (1998) advocate the use of the C4668 and Balmer line indices to form a (relatively) independent estimate of the metallicity and age respectively of early--type galaxies in the Fornax cluster. However, they note that galaxy properties inferred from absorption indices should be regarded solely as {\\em relative} measures.  Figure \\ref{ew_grid1} shows the C4668 and H$\\beta$ indices for Normal and [OII]--emitting early--type galaxies drawn from Sample A. \\begin{figure} \\psfig{figure=./fig6.ps,height=7.5cm,angle=0.0} \\caption{Comparison of C4668 and H$\\beta$ Lick/IDS absorption indices measured for the composite spectrum for Normal (square) and [OII]--emitting galaxies (triangle) to indices measured from synthetic stellar populations of varying age and metallicity. 1$\\sigma$ errors are derived via a bootstrap analysis (see text). Stellar population predictions are derived from the models of Worthey (1994). Labels placed at the bottom of each grid indicate $\\log [ Z / Z_{\\odot} ]$, labels placed to the right of each grid indicate stellar population age.} \\label{ew_grid1} \\end{figure} The observed offset in the strength of the H$\\beta$ index between the two sub--samples is in the opposite sense to that which would be expected if the [OII]--emitting galaxies exhibited low--luminosity nebular HII emission, effectively reducing the strength of H$\\beta$ in absorption and leading to overestimation of the apparent age of the galaxy. One may therefore conclude that any offset caused by nebular HII emission in the early--type galaxy sample as a whole is smaller than the estimated error in the H$\\beta$ index determined for the [OII]--emitting galaxies (or alternatively that it is unrelated to [OII] 3727 emission). The absorption line properties of each sub--sample are statistically identical, indicating that the luminosity weighted stellar populations of Normal and [OII]--emitting galaxies share a common star formation history.  \\subsection[]{Conclusions} \\label{sec_sf_conc}  To summarise, the mean colour--redshift distribution of the sample is consistent with a luminosity--weighted stellar population that formed at redshift $z>1$. No significant star formation has occurred since that time. If the population formed over an extended period then there must have been a well--defined tight relationship between the mean metallicity of the stars and the formation epoch in order for the galaxy population as a whole to exhibit such a small dispersion in observed properties.  The exact relation between formation redshift and metallicity is dependent upon the assumed star formation rate as a function of time and the cosmological model.  The strength of the mean absorption line indices of both the Normal and [OII]--emitting sub--samples indicates that each sub--sample is consistent with an old ($\\sim12$ Gyr), approximately solar metallicity ($\\log [Z/Z_{\\odot}] \\simeq 0.0$ dex) luminosity--weighted stellar population. The results of the colour--redshift and absorption line analyses are consistent and indicate that, modulo the assumed star formation rate as a function of time, the mean luminosity weighted stellar population in luminous field early--type galaxies formed at a redshift $z \\gg 1$ with approximately solar metallicity.  Observation of [OII] 3727 emission in $\\sim25${\\%} of the galaxies indicates that these galaxies are experiencing low SFRs of only $\\la 1 \\, h^2 \\, {\\rm{M}}_{\\odot} \\, {\\rm{yr}}^{-1}$. This low level of star formation is consistent with the results of the colour--redshift and absorption index analyses that indicate that Normal and [OII]--emitting early--type galaxies possess essentially identical luminosity--weighted stellar populations. The space density of the brightest [OII]--emitting galaxies displays an increase of a factor $\\sim 3$ at the largest redshifts. However, the colour selection criteria that define the sample mean that recent, $z_b \\simeq 1$, star formation events must involve $< 0.1${\\%} of the total stellar mass of the galaxies.  The conclusions regarding the star formation history of luminous field early--type galaxies appear to be inconsistent with the hierarchical merging of gas rich, massive galaxies at redshifts $z \\la 1$. Rather, the observations are more consistent with the accretion of gas rich, low mass (presumably dwarf) galaxies leading to a small burst of star formation superimposed upon a dominant, quiescent stellar population. The increase in the space density of the brightest [OII]--emitting galaxies at large redshift may indicate an increase as a function of redshift in the accretion rate of low mass galaxies. However, such a conclusion is critically dependent on the identification of the galaxy population studied here as the progenitors of early--type galaxies today. If in fact the galaxy population studied here represents only a small fraction of the present--day population of early--type galaxies then the conclusions concerning the formation history outlined above could be extremely misleading.  The importance of progenitor--bias in the context of the evolution of early--type galaxies in clusters has been stressed by van Dokkum \\& Franx (2001). In the context of the sample of field early--type galaxies it is essential to demonstrate that the space density of the population at $z \\simeq 0.4$ is comparable to the space density of early--type galaxies today and this question is addressed in Section 4.  \\section[]{The luminosity function}  The evolution of the space density of early--type galaxies as a function of redshift places strong constraints upon the extent to which the early--type galaxy population formed via the hierarchical merging of galactic ``sub--units''. The 485 galaxies of Sample B allow a direct estimate of the luminosity function at $z\\simeq 0.4$ to be made. A parametric estimator allows a comparison with parametric descriptions of luminosity functions from other samples while a non--parametric estimator provides a direct determination of the luminosity function without the imposition of a possibly inappropriate model representation. Both types of estimator are employed here.  \\subsection[]{Absolute Magnitudes} \\label{sec_mag_est}  The analysis of the colour--redshift distributions for each galaxy sub--sample in Section 3.2 demonstrated that the observed colours are consistent with the predictions of a single passively--evolving stellar population model, modulo a small intrinsic dispersion. The effects of passive luminosity evolution over the redshift interval $0.3 \\le z \\le 0.6$ are significant and the derived rest--frame absolute magnitudes need to be corrected for passive stellar luminosity evolution.  In addition, referring luminosity function parameters to a common epoch provides a necessary reference point against which complementary studies of early--type galaxy may be compared. To correct the apparent $b_{\\rm J}ori$ magnitudes of each galaxy to a common epoch it is necessary to compute the apparent magnitude of that galaxy as a function of redshift, i.e., \\begin{multline} m_n(z) = m_n - A_n - 5 \\log [d_L(z_n)/d_L(z)] \\\\ - ([e+k](z_n)-[e+k](z)) \\, , \\label{eqn_magz} \\end{multline} where each galaxy with magnitude $m_n$ and redshift $z_n$ is corrected for Galactic extinction, $A_n$, the effect of differential luminosity distance as a function of redshift, $d_L(z)$, and a differential evolution plus $k$--correction term as a function of redshift $[e+k](z)$ appropriate to each photometric passband. Rest--frame absolute $V$--band magnitudes are then determined from apparent $i$--band magnitudes computed at a common redshift, $z_c=0.4$,  i.e., \\begin{equation} { M_V = m_n(z_c) - 25 - 5 \\log d_L(z_c) - k(z_c) + (V-i)_{z=0.4}, } \\label{eqn_abs} \\end{equation} where $k(z_c)$ is determined from the SED of a $z=0.4$ galaxy generated by the specified spectral evolution model and $(V-i)_{z=0.4}$ is the rest frame $V-i$ colour for an early--type galaxy at $z=0.4$. Figure \\ref{abs_mag_hist} shows the absolute magnitude distribution computed for an exponentially decaying burst of star formation of e--folding time scale $\\tau=1\\,$Gyr formed at a redshift $z_f=3$ with solar metallicity within a spatially flat universe (see Table \\ref{tab_lf_models}:  ``Model 2''). \\begin{figure} \\psfig{figure=./fig7.ps,height=7.5cm,angle=0.0} \\caption{Rest--frame absolute $V$--band magnitudes, corrected for passive luminosity evolution, at a common epoch, $z=0.4$. See text for discussion.} \\label{abs_mag_hist} \\end{figure}  \\subsection[]{The luminosity function: ${1 \\, / \\, V_{acc}}$ estimation} \\label{sec_vacc}  In the $1 / V_{acc}$ estimator (Avni \\& Bahcall 1980) the maximum accessible volume associated with each galaxy in the sample is computed by integrating the co--moving volume element per unit solid angle, ${\\rm{d}}V / \\, {\\rm{d}}z$, multiplied by the probability that a galaxy of given magnitude and colour enters the sample as a function of redshift, $W(z)$ (Equation \\ref{eqn_wz}), over the specified redshift limits, i.e. \\begin{equation} { V_{acc} = c \\; {\\rm{d}}\\Omega \\int_{z_{min}}^{z_{max}} \\frac{{\\rm{d}}V}{{\\rm{d}}z} \\, W(z) \\, {\\rm{d}}z, } \\label{eqn_acc} \\end{equation} The factor d$\\Omega$ scales the accessible volume by the appropriate solid angle and the factor $c$ corrects for redshift incompleteness. The enclosed volume associated with each galaxy, $V_{enc}$, is calculated by replacing the integration limit $z_{max}$ by $z_{gal}$.  Redshift limits of $z_{min}=0.28$ and $z_{max}=0.60$ were adopted to exclude galaxies at redshifts where the probability $W(z)$ becomes very small. The resulting number of galaxies in the Normal and [OII]--emitting sub--samples were 367 and 99 respectively. The cumulative $1/V_{acc}$ luminosity function, $\\Phi(<M)$, is calculated by summing the inverse accessible volumes for each galaxy (Figure \\ref{clf_type}).  \\begin{table} \\caption{Spectral evolution models and cosmological parameters employed in luminosity function analyses. For exponentially decaying star formation histories, the e--folding time scale is set to $\\tau=1\\,$Gyr.  The ``non--evolving'' spectral evolution model employs the Kinney SED to determine the $k$--correction term. Although $h$--dependent values are normalised to $h=1$ the adopted value of $h=0.7$ determines the age--redshift relation.} \\label{tab_lf_models} \\begin{tabular}{clcccc} Model & \\multicolumn{3}{c}{Spectral model} & \\multicolumn{2}{c}{Cosmological model} \\\\ & SF history & $z_f$ & $\\log [Z/Z_{\\odot}]$ & $\\Omega_M$ & $\\Omega_{\\Lambda}$ \\\\ 1 & Instantaneous & 1.4 & 0.0 & 0.3 & 0.7 \\\\ 2 & Exponential   & 3.0 & 0.0 & 0.3 & 0.7 \\\\ 3 & Non--evolving & --  & --  & 0.3 & 0.7 \\\\ 4 & Instantaneous & 2.6 & 0.0 & 0.3 & 0.0 \\\\ 5 & Exponential   & 4.0 & 0.1 & 0.3 & 0.0 \\\\ 6 & Non--evolving & --  & --  & 0.3 & 0.0 \\\\ \\end{tabular} \\end{table}  Internal consistency was verified by comparing cumulative luminosity functions for the two sub--samples and for each 2dF field using a two--sample Kolmogorov--Smirnov (KS) test. Figure \\ref{clf_type} shows cumulative luminosity functions for the Normal and [OII]--emitting sub--samples together with the combined sample. A KS--test applied to the Normal and [OII]--emitting sub--samples gives $D_{max}$ and associated probabilities that the samples are drawn from the same distribution, of: (0.11, 0.34), indicating no significant differences. \\begin{figure} \\psfig{figure=./fig8.ps,height=7.5cm,angle=0.0} \\caption{Cumulative luminosity functions corresponding to each spectral sub--class compared to the combined galaxy sample. Each luminosity function was generated employing ``Model 2'', i.e. an exponentially decaying burst of star formation at $z_f = 3.0$ with an e--folding time scale of $\\tau=1\\,$Gyr within a spatially flat universe.} \\label{clf_type} \\end{figure} $D_{max}$ and associated probabilities for samples from each 2dF field compared to the remaining three fields are (0.10,0.36), (0.13,0.10), (0.06,0.90), (0.12,0.16) for fields 1, 2, 3, and 4 respectively. The sample thus shows no evidence for significant differences in the luminosity function from field to field, although the effects of cosmic variance are likely responsible for the smaller probabilities associated with fields 2 and 4.  The effective volume probed by the spectroscopic early--type galaxy sample is computed as the mean accessible volume of all galaxies drawn from Sample B occupying the redshift interval $0.28<z<0.6$, i.e. \\begin{equation} { V_{eff} = \\frac{1}{N} \\sum_{i=1}^{N} V_{acc,i}, } \\end{equation} where the index $i$ is summed over the 466 early--type galaxies for which $V_{acc}$ is computed.  Effective survey volumes computed for each galaxy formation model presented in Table \\ref{tab_lf_models} are displayed in Table \\ref{tab_eff_vols}. \\begin{table} \\caption{Effective spectroscopic survey volume computed for each galaxy formation model presented in Table \\ref{tab_lf_models}.} \\label{tab_eff_vols} \\begin{tabular}{lc} Model & $V_{eff} (\\times 10^5 \\, h^{-3} \\, {\\rm Mpc}^3)$ \\\\ 1 & 7.1 \\\\ 2 & 7.7 \\\\ 3 & 6.5 \\\\ 4 & 5.5 \\\\ 5 & 5.3 \\\\ 6 & 4.7 \\\\ \\end{tabular} \\end{table} The effective survey volume may be compared to the space density of rich galaxy clusters in order to constrain the contribution of the field early--type galaxy sample by cluster early--type galaxies. The space density of rich ($R \\ge 1$) Abell clusters at redshifts $z < 0.6$ is $5.6 \\times 10^{-6} \\, h^3 \\, {\\rm Mpc}^{-3}$ (Postman et al. 1996). Comparing this space density to the effective survey volumes for a spatially flat universe indicates that the spectroscopic early--type galaxy sample may contain four rich galaxy clusters. At a redshift $z=0.4$, the photometric selection criteria will sample the cluster early-type galaxy magnitude distribution down to $\\sim m_{10}$ and we therefore assume a cluster contribution rate in the full photometric sample of $\\sim 10${\\%}. However, the strict lower limit on the spatial sampling of galaxy targets imposed by the 2dF instrument (greater than 30 arcseconds) implies that it is unlikely that the presence of a rich cluster in the spectroscopic target fields will contribute more than a few galaxies to the sample. This conclusion is supported by the absence in the spectroscopic sample of any groups of 5 to 10 galaxies within 2000 kms$^{-1}$ velocity intervals and within a projected radius of 5 arcminutes -- thus implying a cluster contribution rate for the spectroscopic subsample of $<5-10${\\%}.   \\subsection[]{The luminosity function: the STY maximum likelihood estimator} \\label{sec_sty}  The STY estimator (Sandage, Tammann \\& Yahil 1979) is unbiased by density variations and, although the density normalisation is not constrained, the method provides a robust determination of the shape of the luminosity function.  The galaxy luminosity function may be defined via the number of galaxies with magnitudes in the interval $M$ to $M+{\\rm{d}}M$ within a volume element d{\\bf x}$^3$ to be $\\phi(M,{\\bmath {\\rm{x}}}) \\, {\\rm{d}}M \\, {\\rm{d}}{\\bmath {\\rm{x}}}^3$. Based on the results of the previous section we assume that $\\phi(M,{\\bmath{\\rm{x}}})=\\phi(M)$, where the given functional form of the luminosity function is characterised by a constant space density $\\phi^{\\ast}$ per unit volume and magnitude interval. We further assume that the shape parameters describing $\\phi(M)$ do not vary with redshift. Thus, one may express the probability density that a sample galaxy observed at a redshift $z$ will display an absolute magnitude $M$ as \\begin{equation} { p(M|z) \\propto {\\frac{\\phi(M) \\, f(m)\\,S(z)\\,({\\rm d}V/{\\rm d}z)} {\\int_{z_{min}}^{z_{max}}\\int_{-\\infty}^{\\infty} \\phi(M^{\\prime}) \\, f(m^{\\prime}) \\, S(z^{\\prime}) \\, ({\\rm d}V/{\\rm d}z^{\\prime})\\, {\\rm{d}}M^{\\prime} \\, {\\rm d}z^{\\prime}}}. } \\end{equation} The factor $S(z)$ describes the colour--redshift selection probability (Equation \\ref{eqn_colour_select}) and ${\\rm d}V/{\\rm d}z$ is the co--moving differential volume element per unit solid angle. The factor $f(m)$ indicates the fraction of galaxies entering the early--type galaxy sample as a function of apparent magnitude, $m$ \\begin{eqnarray} f(m) &=& 0 \\;\\;\\;\\;\\;\\;\\;\\;\\; {\\mbox{where}} \\: M>M_{faint}(z) \\\\\\nonumber && \\;\\;\\;\\;\\;\\;\\;\\;\\;\\;\\; {\\mbox{or}} \\: M<M_{bright}(z) \\\\\\nonumber f(m) &=& c(m) \\:\\:\\:\\: {\\mbox{where}} \\: M_{bright}(z)<M<M_{faint}(z).\\nonumber \\end{eqnarray} The factor $c(m)$ is the redshift incompleteness as a function of apparent magnitude $m_{i}$. The relation between $m_{i}$ and rest--frame absolute $V$--band magnitudes is given by equations \\ref{eqn_magz} and \\ref{eqn_abs}.  The luminosity function is parameterised in terms of a Gaussian distribution of characteristic magnitude $M^{\\ast}$ and width $\\sigma_M$. The photometric errors are incorporated by convolving the intrinsic form of the luminosity function, $\\phi_{int}(M)$, with a Gaussian function of standard deviation equal to $\\sigma_{obs}$, the observed $1\\sigma$ $i$--band magnitude error of each galaxy, to generate an observed luminosity function, \\begin{equation} { \\phi_{obs} = {\\frac{1}{\\sqrt{2\\pi}\\sigma_{obs}}} \\int_{-\\infty}^{\\infty} \\phi_{int}(M^{\\prime}) \\, \\exp \\left ( {\\frac{-{(M-M^{\\prime})}^2}{2\\sigma_{obs}^2}}\\right ) {\\rm{d}}M^{\\prime}. } \\end{equation} The maximum likelihood procedure identifies values of $M^{\\ast}$ and $\\sigma_M$ that maximise the log likelihood equation over the $k=1, N$ sample galaxies drawn from the redshift interval $0.28<z<0.6$, i.e. \\begin{equation} { \\ln {\\mathcal L} = \\sum_{k=1}^{N} \\ln p \\, (M_i|z_i) \\, . } \\end{equation} Confidence intervals in the $M^{\\ast}$ versus $\\sigma_M$ may be displayed by computing contours of equal likelihood according to Equation \\ref{eqn_max_err}. The characteristic density is estimated by normalising the integrated number of galaxies predicted for a given luminosity function model to the total number of early--type galaxies observed in the full photometric sample (Section 2, Paper I; Section 2), i.e. \\begin{multline} N_{total} = {\\rm{d}}\\Omega \\int_{z_{min}}^{z_{max}} \\int_{m_{bright}}^{m_{faint}} \\phi_{obs}(m(M,z)) \\\\ \\times \\, S(z) \\, ({\\rm{d}}V/{\\rm{d}}z) \\, {\\rm{d}}m \\, {\\rm{d}}z. \\label{eqn_phi_calc} \\end{multline} The apparent magnitude of an early--type galaxy of absolute $V$--magnitude, $M$ and redshift, $z$, is calculated employing Equations \\ref{eqn_magz} and \\ref{eqn_abs} assuming a mean Galactic extinction of $E(B-V)=0.0294$, calculated using the full photometric sample.  \\subsection[]{Results and discussion}  The results of the maximum likelihood analysis for each spectral evolution and cosmological model combination described in Table \\ref{tab_lf_models} are given in Table \\ref{tab_lf_results}. Figure \\ref{lf_cont} displays error contours associated with the luminosity function parameters determined for Model 1. \\begin{figure} \\psfig{figure=./fig9.ps,height=7.5cm,angle=270.0} \\caption{Contours of relative likelihood corresponding to 1 (solid) and $2\\sigma$ (dashed) joint error distributions for luminosity function parameters $M^{\\ast}_V-5\\log h $ and $\\sigma_M$. Contours are computed for Model 1 (Table \\ref{tab_lf_results}). The open square indicates the formal maximum.} \\label{lf_cont} \\end{figure} Error contours represent the relative uncertainty associated with a given combination of spectral evolution and cosmological model and do not account for uncertainties in the adopted spectral evolution model or cosmological parameters. Errors are thus likely to be underestimated and the luminosity function parameters returned by the maximum likelihood procedure should be considered solely as representations of the data over a limited range of absolute magnitude (Lilly {\\etal} 1995). Given the limited range of absolute magnitude and the strong correlation between the errors of the parameters of the Gaussian fit to the luminosity function, the integrated luminosity density is a far more robust quantity. Consideration of the integrated luminosity density has the further advantage that results based on fitting different functional forms for the luminosity function may also be intercompared.  A significant degree of uncertainty can arise when comparing luminosity functions of galaxies defined using different data sets or selected according to different selection criteria. It is necessary to establish both that the same type of objects are included in the determination of the luminosity functions and that the transformations between measured quantities, particularly absolute magnitudes, are well understood. The situation is compounded, as here, when a comparison between luminosity functions of galaxy samples at very different redshifts is attempted. Fortunately, the recent availability of the Sloan Digital Sky Survey (SDSS) Early Data Release (EDR) (Stoughton et al. 2002) and the determination of the luminosity function for early--type galaxies (Eisenstein et al.  2001, Bernardi et al. 2002), allow a relatively direct comparison between our results at $z\\simeq 0.4$ and those pertaining to the present day to be made. Bernardi et al. present luminosity functions in the SDSS $g^*r^*i^*z^*$ bands for a sample of early--type galaxies with $\\langle z \\rangle \\simeq 0.1$.  Early--type galaxies within the SDSS are defined according to a combination of morphological (concentration and surface brightness) and spectral (no detectable emission lines and projected velocity dispersion greater than the effective instrumental limit) criteria. The resulting galaxy sample possesses a well--defined mean colour--redshift locus consistent with a uniformly old stellar population, combined with a small Gaussian dispersion about the mean relation. Therefore, in the absence of additional star--formation events, the $z \\simeq 0.4$ early--type galaxy population will evolve passively to form early--type galaxies that satisfy the SDSS selection criteria. \\begin{table} \\caption{Gaussian luminosity function parameters estimated via STY maximum likelihood procedure. Parameters are valid over the approximate absolute magnitude interval $-22.5<M_V-5\\log h<-20.5$. The $2\\sigma$ single parameter errors are $-{M_V^{\\ast}}^{+0.11}_{-0.15}$ for characteristic magnitude and ${\\sigma_M}^{+0.08}_{-0.06}$ for the width term. The characteristic space density, $\\phi^{\\ast}$, is computed via Equation \\ref{eqn_phi_calc} for the central values of $M_V^{\\ast}-5\\log h$ and $\\sigma_M$ determined for each model and is expressed units of $10^{-3} \\, h^3 \\, {\\rm{Mpc}}^{-3} {\\rm{mag}}^{-1}$. The $2\\sigma$ uncertainty in the characteristic space density expresses the variation in $\\phi^{\\ast}$ associated with the corresponding $2\\sigma$ variation in the luminosity function shape parameters. This source of uncertainty dominates over the Poisson uncertainty associated with the photometric sample size. } \\label{tab_lf_results} \\begin{tabular}{cccc} Model & $-M_V^{\\ast}+5\\log h$ & $\\sigma_M$ & $\\phi^{\\ast}$ \\\\ 1 & 21.28 & 0.47 & $1.50 \\pm 0.25$ \\\\ 2 & 21.23 & 0.48 & $1.46 \\pm 0.25$ \\\\ 3 & 21.28 & 0.50 & $1.59 \\pm 0.23$ \\\\ 4 & 21.12 & 0.47 & $1.86 \\pm 0.29$  \\\\ 5 & 21.10 & 0.46 & $1.96 \\pm 0.32$  \\\\ 6 & 21.10 & 0.49 & $2.17 \\pm 0.32$  \\\\ \\end{tabular} \\end{table}    The Gaussian luminosity function parameters of Bernardi et al. in the $g^{\\ast}$--band are transformed to rest--frame $V$--band values at a redshift $z=0.4$. The redshift $z=0$ $g^{\\ast}$--band characteristic magnitude is corrected for passive luminosity evolution to a redshift $z=0.4$ employing the $Q$--parameter estimated by Bernardi et al. The zero--point transformation between the $g^*$--band absolute magnitudes on the SDSS AB system of Bernardi et al. to those in the $V$--band on the Vega system is readily accomplished using the appropriate transmission functions and the early--type galaxy template of Coleman, Wu \\& Weedman (1980) employed by Bernardi et al. Application of transformations to account for aperture to total magnitudes and a residual zero--point difference result in a total shift of UKST $i$--band magnitudes of $-0.48\\,$mag relative to the SDSS system (see Appendix \\ref{app_sdss_comp} for a comprehensive discussion). The two samples are then compared in terms of the $V$--band luminosity density integrated over the common absolute magnitude interval sampled by each survey,  $-23 \\la M_V^{\\ast} - 5 \\log h \\la -21$, i.e. \\begin{equation} { \\rho_L = \\int_{M_{bright}}^{M_{faint}} \\phi(M) \\, 10^{-0.4(M-M_{{\\odot},V})} \\, {\\rm d}M, } \\label{eqn_lum_den} \\end{equation} which is expressed in units of $\\times 10^7 \\, h^2 \\, L_{{\\odot},V} \\, {\\rm Mpc}^{-3}$, assuming that the $V$--band absolute magnitude of the Sun is $M_{{\\odot},V}=4.83$. \\begin{table} \\caption{Luminosity density values calculated for SDSS and $z=0.4$ early--type galaxy samples according to Equation \\ref{eqn_lum_den}. Values are expressed in units of $\\times 10^7 \\, h^2 \\, L_{{\\odot},V} \\, {\\rm Mpc}^{-3}$ (see text). The Poisson error for each values is approximately 1{\\%}. Note that the SDSS luminosity function values from which the luminosity density is generated are presented by Bernardi et al. for the cosmological model $\\Omega_M=0.3$, $\\Omega_{\\Lambda}=0.7$ and $H_0=70$ km s$^{-1}$ Mpc$^{-1}$. Hence, for consistency, the SDSS figure is compared only to Models 1--3 described in Table \\ref{tab_lf_models}.} \\label{tab_ld} \\begin{tabular}{ccc} Model & $\\rho_L$ & fraction compared to SDSS \\\\ SDSS & 7.21 & -- \\\\ 1 & 6.87 & 0.95 \\\\ 2 & 6.38 & 0.88 \\\\ 3 & 7.19 & 1.00 \\\\ \\end{tabular} \\end{table} Comparison of the luminosity density determined for the SDSS and $z=0.4$ samples (Table \\ref{tab_ld}) indicates that, within the uncertainties imposed by the spectro--photometric model employed to describe the $z=0.4$ early--type galaxy sample and the transformations applied to place the two samples on a common photometric system, there is no significant shortfall in luminous field early--type galaxies at redshift $z=0.4$ compared to $z=0.1$. ",
        "conclusions": "The properties of luminous, field early--type galaxies at redshifts $z<1$ in principle provide an important test of the hierarchical model of galaxy formation. We have presented an analysis of the mean star formation history and space density of a sample of luminous field early--type galaxies selected over the redshift interval $0.3 \\la z \\la 0.6$.  Overall, the mean star formation history of the sample is characterised by an apparently old ($z_f>1$), solar to slightly above--solar metallicity luminosity--weighted stellar population that has evolved passively since the formation epoch. The colour--redshift evolution, mean absorption line properties and mean spectrum of the sample present a consistent picture of an old, quiescent stellar population. The exact range of formation redshift and metallicity permitted depends upon the assumed time dependence of the major star formation event and the cosmological model via the look--back time versus redshift relation. The mean properties of the sample are markedly similar to the properties of morphologically--selected luminous elliptical galaxies in rich cluster environments at redshifts $z<1$ (Ferreras {\\etal} 1999). Though neither result in isolation constrains the extent to which early--type galaxies in field or cluster environments represent a co--eval or co--metal population, the broad similarity between the star formation history of the dominant stellar mass component in such galaxies is consistent with similar formation conditions for each population.  Approximately one--quarter of the early--type galaxy sample displays detectable [OII]3727 emission consistent with on--going star formation rates $\\la 1.5 h^2$ M$_{\\odot}$ yr$^{-1}$. Consideration of the number redshift distribution of [OII] 3727 emitting galaxies and the effects of a recent burst of star formation upon the colour selection of such galaxies, limits the typical stellar mass content of such star formation events to $<1${\\%} of the galaxy mass for a burst occurring at redshift $z=1$. However, the space density of the strongest [OII] 3727 emitting galaxies in the sample displays a marked increase at redshifts $z \\ga 0.4$. Although further detailed observations are required to determine the true nature of such star formation events, the star formation rate associated with each event combined with the increase in space density at higher redshift is consistent with an increasing accretion rate of low mass, gas rich galaxies with increasing look--back time.  Comparison of the luminosity density of the $z = 0.4$ early--type galaxy sample to the sample of $z = 0.1$ early--type galaxies from the SDSS shows no evidence of significant change in the luminosity density of luminous field early--type galaxies between these two epochs. In common with results obtained for morphologically selected samples of elliptical galaxies obtained over similar redshift intervals (Schade {\\etal} 1999), these results show no evidence that significant numbers of luminous early--type/elliptical galaxies have formed via merging at redshifts $z<1$. The hypothesis that luminous field early--type galaxies formed via the hierarchical merging of gas rich disk galaxies is not rejected by our results. However, it does seem clear that the bulk of star formation associated with the formation of such galaxies occurred at redshifts $z>1$ and the assembly of the most massive early--type galaxies in the field, via any postulated merging process, was largely complete by redshift $z\\simeq 0.5$."
    },
    "0208/astro-ph0208538_arXiv.txt": {
        "abstract": "We study the effects of mergers on the structural properties of disc-like systems by using Smooth Particle Hydrodynamical (SPH) numerical simulations in hierarchical clustering scenarios. In order to assess the effects of mergers on the mass distributions we performed a bulge-disc decomposition of the projected surface density of the systems at different stages of the merger process. We  assumed an exponential law for the disc component and the S\\'ersic law for the bulges. We found that simulated objects at $z=0$ have bulge profiles with shape parameters $n\\approx 1$, consistent with observational results of spiral galaxies. The complete sample of simulated objects at $z=0$ and $z>0$ shows that $n$ takes values in the range $n\\approx 0.4 - 4$. We found that secular evolution tends to produce exponential bulge profiles, while the fusion of baryonic cores tends to increase the $n$ value and helps to generate the correlation between $B/D$ and $n$. We found no dependence on the relative mass of the colliding objects. Our results suggest that  mergers, through secular evolution and fusions, could produce the transformation of galactic objects along the Hubble sequence by driving a morphological loop that might also depend  on the properties of the central galactic potential wells, which are also affected by mergers. ",
        "introduction": "The origin of the Hubble sequence is still a controversial issue. In particular, spiral galaxies seem to evolve along this phenomenological classification although the  physical mechanisms behind these morphological changes are not fully understood. The structural parameters, the gas abundances and the star formation activity vary along galaxies in the Hubble sequence in the sense that disc-dominated systems are also the more gaseous ones, experiencing on-going star formation. The large database of galaxies  gathered in the last years have allowed to get more detailed information on the properties of different morphological type galaxies in the local Universe.  Observational results show that when a double exponential  decomposition is applied to the sufarce luminosity density of late-type spirals, the scalelengths of the bulge and disc components seem to be restricted to a certain value $<r_{\\rm b}/r_{\\rm d}> \\approx 0.10$ (Courteau, de Jong \\& Broeils 1996, hereafter CdJB96). This restriction in the range of possible scalelengths for spirals has been interpreted as a proof for secular evolution to be the responsible mechanism for bulge formation from an already in place disc structure. Recently, MacArthur, Courteau \\& Holtzman (2002, hereafter MCH02) studied a larger sample of late-type spirals and carried out a bulge-disc decomposition assuming a S\\'ersic law for the bulge surface luminosity density. These authors found a continuous distribution of the shape parameter of the bulge ($n\\approx 0.2-2$) with a  maximum at $n \\approx 0.90$. They also found a restricted range of values for the scalelengths of the bulge and disc when the shape parameter is $n \\approx 1$, $<r_{\\rm b}/r_{\\rm d}>=0.13 \\pm 0.06$ in the $R$-band. Previous works have also found that bulges of spirals can be fitted by a S\\'ersic law with different shape parameters (e.g., Andredakis, Peletier \\& Ballcels 1995, hereafter APB95; Khosroshahi, Wadadekar \\& Kembhavi 2000,  hereafter KWK00; Graham \\& de Block 2001,  hereafter GdB01). APB95 and GdB01 also found a correlation between the luminosity bulge-to-disc ratio $B/D$ and the shape parameter of the bulge, which also correlates with morphological type. These observational results suggest a possible connection between the formation of the bulge and disc components. They also support the idea  that $n$ could be used as a good indicator of position in the Hubble sequence.  Several theories have been developed to explain the formation of the disc and bulge components. In the case of the disc systems, the standard model is based on three hypothesis: the angular momentum is acquired through cosmological torques (Peebles 1969), baryons and dark matter have the same specific angular momentum content ($J$), and this specific angular momentum is conserved during the collapse and cooling of baryons (Fall \\& Efstathiou 1980,  hereafter FE80). This model has been successful in reproducing several observational results in analytical and semi-analytical models (e.g., Dalcanton, Spergel \\& Summers 1997;  Mo, Mao \\& White 1998). However, serious problems arose in numerical simulations where an important angular momentum transfer from baryons to the dark matter haloes during mergers was detected, breaking the condition of $J$ conservation. Dom\\'{\\i}nguez-Tenreiro, Tissera \\& S\\'aiz (1998,  hereafter DT98) showed that  a disc-like structure with observational counterpart (S\\'aiz et al. 2001,  hereafter S01) can be built up if a compact stellar bulge is allowed to form without depleting the gas reservoir of the system. These stellar bulges provide stability to the gaseous disc systems which are capable of conserving a non-negligible fraction of its angular momentum during violent events. Supernova energy feedback could also contribute to the formation of the disc component, regulating the star formation rate and preventing early catastrophic depletion of the gas into stars. Probably both mechanisms, the formation of a compact stellar bulge which assures the axisymmetrical character of the potential well and energy feedback, work together in nature to allow the formation of  spiral galaxies (e.g. Weil, Eke \\& Efstathiou 1998).  The formation of bulges is a more complex task  since several mechanisms could be acting together such as monolithic collapse (Gilmore \\& Wyse 1998), mergers (Kauffman, Guiderdoni \\& White 1994) and secular evolution (Pfenniger \\& Norman 1990). In general, analytical models and pre-prepared simulations have focused in one or two of them at the time (e.g., van den Bosch 1999; Aguerri, Balcells \\& Peletier 2001,  hereafter A01). These models have been successful in explaining some properties of spiral galaxies, despite their approximations.  The current paradigm for the formation of the structure favors a hierarchical clustering scenario where the structure forms by aggregation of substructure. Hence, a galactic object experiences the effects of collapse, merger, interaction and probably secular evolution  in a non simple fashion, which can also strongly depend on redshift. In particular, violent events (i.e.  mergers, interactions) can have important effects on the internal properties of the objects, such as their mass distribution (Mihos \\& Hernquist 1996) and star formation activity (Tissera et al. 2002,   hereafter T02).  Minor and major mergers of disc systems with satellites have been extensively studied.  However, most of these works do not consider the presence of the stellar bulges. The relevance of this component in the overall evolution of a disc system during a merger has been firstly pointed out by Mihos \\& Hernquist (1994) in a study of pre-prepared mergers. Recently, A01 studied the effects of mergers on the structural parameters of bulges of disc-like systems by assuming that initially bulges form with an exponential profile. These authors found that collisionless mergers produce a migration to higher $n$ and that this effect is proportional to the relative mass of the colliding systems. The simulations used in this analysis were non-cosmological pre-prepared ones where on going star formation and gas dynamics were not included. Recently, T02 show for the first time in cosmological SPH simulations how the central mass concentration in disc-like systems can grow by collapse, merger and secular evolution.  These numerical results suggest that a key point in the formation of both bulges and discs in hierarchical clustering scenarios seems to be mergers. Mergers have been traditionally suggested as a possible mechanism to drive the Hubble sequence since they can modify the dynamical and astrophysical properties of galactic objects. Hence, the question would be how, in hierarchical scenarios, mergers work to shape disc-like objects and if the properties of these objects resemble those of  current spirals. In this paper we will focus on the analysis of the mass distributions of disc-like systems and how they are modified during mergers, paying special attention to the comparative study of the effects of secular evolution triggered by tidal fields and of the actual collision of the baryonic clumps. In order to assess the effects of mergers on the mass distributions we perform a bulge-disc decomposition of the projected surface density of the systems at different stages of the merger process. We  then study how such different structural parameters, including the shape parameter defined by S\\'ersic (1968), evolve during the orbital decay phase and fusion of the satellite.  Our simulations include the effects of gravitation, hydrodynamics, cooling and star formation in a cosmological framework. Hence, the set of mergers that we analyse are given by the particular evolutionary history of each galactic object. This is a crucial point since the merger parameters and physical characteristics of the colliding objects are not ad-hoc choices but result from the consistent formation of the structure in a hierarchical scenario. The drawback of our approach is a lower numerical resolution compared to those used in studies of pre-prepared mergers. Assessment of possible numerical problems are discussed throughout the paper.  In section 2 we present the analysis of the simulations. In section 3 we discuss the results. Section 4 summarizes the conclusions. ",
        "conclusions": "We have studied the properties of the baryon distributions in galaxy-like objects focusing on the effects of mergers. We resort to observations of spiral galaxies of different morphology to constrain our findings.  We found that on average, galactic objects formed in hierarchical clustering scenarios reproduce the angular momentum and structural parameter distributions of spiral galaxies, if a stellar bulge is allowed to form and early gas depletion is avoided. We have succeeded in these two aspects but at the expense of inhibiting star formation on discs. A consistent treatment of energy feedback may help to remove this caveat.  These simulations have allowed us to study how mergers change the distribution of baryons by analysing the evolution of their structural parameters. We found that, on average, galactic objects tend to  have nearly exponential bulges at all redshift. However, note that these simulations produce bulges with shape parameters in the range $0.5 - 4$.  For those systems with $n\\approx 1 $ bulges we found a correlation among their bulge and disc scalelengths in very good agreement with observations. However, for $n>1$ the scalelength distribution  is disorder and displaced to smaller bulge scalelenghts. Observations show the same behaviour. The opposite distribution is found for systems with $n<1$ bulges  which have larger $r_{\\rm b}/r_{\\rm d}$ values and larger dispersions. We found the disc scalelengths to be approximately independent of $n$ parameters, so that this correlation implies that more concentrated objects have smaller  bulge scalelengths (or large effective radius). We also found a maximum  $r_{\\rm b}/r_{\\rm d}$  of $\\approx 0.40$ for $n\\rightarrow0$. Higher numerical resolution simulations are needed to study the formation of such low mass surface profiles.  In order to test our results for low resolution problems in the determination of the structural parameters of the galaxy-like objects, we followed Steinmetz \\& M\\\"uller (1994) and used the bootstrap technique to estimate the effects of low particle number statistics within the objects. We calculated an accuracy better than 25 per cent for the scalelengths and shape parameters.  We found that gas inflows during the orbital decay phase tend to produce important changes in the mass distributions generating $n\\approx 1$ profiles, while the fusions of the baryonic cores tend to increase the $n$ parameter. As a consequence, a morphological loop can be driven by mergers which might be responsible of triggering secular evolution as well as of the baryonic core fusions. The triggering of secular evolution is found to be linked to the presence of a stellar bulge so that systems with well-formed stellar bulges do not experience early gas inflows. Hence, the pace of this morphological loop could be regulated by the properties of the galactic central potential wells which are also affected by the merger history of the objects (see also Tissera \\& Dom\\'{\\i}nquez-Tenreiro 1998 and Tissera 2000).  We found that the simulated  mass bulge-to-disc ratios are within observed range. It is also noted that during the ODP larger changes are observed in the $B/D$ ratios of those objects that experience gas inflows. The changes during this period are, however, quite disorder. It is at the fusion of the baryonic clumps that changes in the shape parameters are correlated with changes in the bulge-to-disc ratio. In our simulations, the actual fusions are responsible of significantly increasing the mass concentration at the centre. Hence, we found that the fusion of the baryonic cores could be the process that determine the observed correlation between the luminosity $B/D$ ratio and the shape parameter or morphological type. However, we found no dependence on the relative masses of the colliding objects.  Overall, our results indicate that the morphological properties of galactic objects are the result of their merger histories within a hierarchical clustering scenario. Based on the good agreement found so far with observations we support the hypothesis of mergers as the main morphological driver along the Hubble sequence. Consequently, the particular and detailed history of substructure aggregation could be a key point in the determination of the astrophysical properties of galaxies."
    },
    "0208/astro-ph0208049_arXiv.txt": {
        "abstract": "We report on the discovery of a diffuse X-ray source with ASCA, presumably associated with a molecular cloud in the vicinity of the supernova remnant RX~J1713.7$-$3946. The energy spectrum (1--10 keV) of the hard X-ray source shows a flat continuum, which is described by a power-law with a photon index of $\\Gamma = 1.0^{+0.4}_{-0.3}$. We argue that this unusually flat spectrum can be best interpreted  in terms of characteristic bremsstrahlung emission from the loss-flattened distribution of either sub-relativistic protons or mildly relativistic electrons. The strong shock of RX~J1713.7$-$3946, which is likely to interact with the molecular cloud, as evidenced by CO-line observations, seems to be a natural site of acceleration of such nonthermal particles. The observed luminosity of $L_{\\rm X} = 1.7 \\times 10^{35} $ erg~s$^{-1}$ (for a distance of 6 kpc) seems to require a huge kinetic energy of about $10^{50}$ erg in the form of nonthermal particles to illuminate the cloud. The shock-acceleration at RX~J1713.7$-$3946 can barely satisfy this energetic requirement, unless (i) the source is located much closer than the preferred distance of 6 kpc and/or (ii) the mechanical energy of the supernova explosion essentially exceeds $10^{51}$ erg. Another possibility would be that an essential part of the lost energy due to the ionization and heating of gas, is somehow converted to plasma waves, which return this energy to nonthermal particles through their turbulent reacceleration on the plasma waves. ",
        "introduction": "Supernova remnants (SNRs) are commonly believed to be major sites for the production of Galactic cosmic rays. The diffusive shock acceleration mechanism naturally accounts for the hard power-law production spectra of cosmic rays in their sources, but, as long as the particle injection rate remains an unsolved problem, the theory does not tell us conclusively what fraction of the initial kinetic energy of an explosion can be transferred to cosmic rays (see e.g. the recent review by \\cite{Malkov01}). Therefore, it is extremely important to  derive the total energy contained in  subrelativistic and relativistic  particles from  observations of the relevant components of nonthermal electromagnetic radiation.  Radio synchrotron emission provides information about  GeV electrons in most shell-type SNRs.  Synchrotron X-ray emission recently  discovered in SN~1006 \\citep{Koyama95}  and some other shell-type SNRs  indicates that the electron acceleration continues effectively up to multi-TeV energies. It is believed, however, that the accelerated protons constitute the major fraction of nonthermal energy to which the mechanical energy of a supernova explosion is transferred  through shock acceleration. The best way to explore the proton component of accelerated particles in SNRs is the detection of  gamma-rays by the decay of $\\pi^0$ mesons produced in collisions between cosmic-ray protons and ambient matter \\citep{Drury94, Naito94}. The detection of gamma-rays of hadronic origin from SNRs  by current and planned space- and ground-based instruments requires high-density environments  in or close to the particle-acceleration sites. Large molecular clouds interacting with shells of SNRs may act as effective gas targets for the production of $\\pi^0$ mesons and their subsequent decay to high-energy $\\gamma$-rays \\citep{ADV94}. It has recently been claimed \\citep{Butt01} that there exists a compelling positional association of the unidentified $\\gamma$-ray source  3EG J1714$-$3857 with a dense molecular cloud (cloud A) overtaken, most probably, by  the shock front  of  the SNR RX~J1713.7$-$3946 (G 347.3$-$0.5). Moreover, it  has been argued that the TeV gamma-rays from this SNR detected by the CANGAROO collaboration are also the result of  interactions of accelerated protons with nearby dense gas targets \\citep{Enomoto02}, though \\citet{Reimer02} claimed that this interpretation seems to be inconsistent with EGRET observations of this region.  The flux of the subrelativistic component of the accelerating protons is generally inconclusive both on observational and theoretical grounds, but it is quite possible that this component dominates the total energetics of nonthermal particles. Unfortunately, subrelativistic protons arriving at the Earth are not a representative sample of the cosmic-ray flux  in the Milky Way owing to the effect of solar modulation. Ionization losses of these particles  play an important role in heating both diffuse neutral gas and molecular clouds, and in generating free electrons in molecular clouds, which  are crucial for interstellar chemistry. It is therefore of great importance  to measure the energy released in the form of  subrelativistic particles  by cosmic-ray accelerators.  The ideal diagnostic tool to probe subrelativistic cosmic rays is the nuclear prompt $\\gamma$-ray line emission at MeV energies \\citep{Ramaty79}. However,  conservative estimates show that the fluxes of even  most prominent $\\gamma$-ray lines  are only marginally detectable because of limited sensitivities of   gamma-ray instruments in the MeV band. On the other hand, as we discuss below, the search for the sites of concentrated subrelativistic protons using bremsstrahlung X-rays could be very promising with the help of superior sensitivities of current X-ray detectors, even though only a small fraction ($\\sim 10^{-5}$) of the nonthermal energy of subrelativistic protons and electrons is radiated away in the form of bremsstrahlung X-rays. Remarkably, if the subrelativistic cosmic rays reside  in a dense gas environment, the ionization losses would result in a significant flattening of the low-energy spectra, and thus leading to the  characteristic ``$1/\\varepsilon$''  type bremsstrahlung X-ray spectrum \\citep{Uchiyama02}.  Such an unusually hard X-ray spectrum may serve as a distinct signature of  nonthermal bremsstrahlung origin of X-ray emission.  In our previous paper \\citep{Uchiyama02} we reported on the detection of hard X-rays from a localized region in the SNR $\\gamma$~Cygni,  which we attributed to $1/\\varepsilon$ bremsstrahlung  from the loss-flattened electron distribution. In this paper we present a more pronounced  example of such hard X-radiation arriving  from a massive cloud in the vicinity of  SNR RX~J1713.7$-$3946. ",
        "conclusions": "\\label{sec:discuss}  \\subsection{On the Origin of Flat X-Ray Emission}  We have discovered a new X-ray source, AX~J1714.1$-$3912, with an extent of $\\sim 10\\arcmin$, characterized by an extremely flat power-law with $\\Gamma = 1.0^{+0.4}_{-0.3}$. The luminosity of the source is estimated to be $L_X = 1.7 \\times 10^{35} d_6^2$ erg~s$^{-1}$, where $d_6$ is the distance  to the source normalized to 6 kpc. Since the luminosity increases linearly with a high-energy cutoff for the $\\varepsilon^{-1}$ type spectrum, the above estimate should be considered as a lower limit for the luminosity. It cannot be the radiation of thermal gas, unless we assume an extremely hot optically thin plasma with a temperature significantly exceeding 10 keV. Since the formation of such a thermal source seems to be quite problematic,  given  its extended character, below we assume that the X-radiation has a nonthermal origin. AX~J1714.1$-$3912 has a good spatial association with cloud A (see figure~\\ref{fig:ximage}). In the following discussion, therefore, we assume that the X-ray emission actually comes from cloud A. If so, this would be a new striking nonthermal phenomenon in molecular clouds.  Nonthermal particles that are illuminating cloud A in the X-ray band could be supplied by either internal or external accelerators. A potential  candidate for an external accelerator is the strong shock of SNR RX~J1713.7$-$3946, which is likely to interact with cloud~A, as evidenced by a very high ratio of CO($J$=2--1)/CO($J$=1--0) \\citep{Butt01}. The particles can be accelerated in the shell  of   RX~J1713.7$-$3946 and afterwards enter cloud~A, given that the shell is a certain site of particle acceleration, which follows from synchrotron X-ray emissions.  It is possible that particle acceleration also takes place at the secondary shocks inside (or in the vicinity of) cloud~A initiated by the main shock of the SNR. Finally, if there is no \\emph{physical} link between cloud~A and SNR RX~J1713.7$-$3946,  this would imply the existence of unseen internal accelerators inside the molecular cloud. Although we cannot exclude any of these possibilities, the preferred option seems to be the ``cloud interacting with SNR'' scenario.  Formally, there are also several options for the nonthermal X-ray production mechanisms: synchrotron radiation, inverse Compton (IC) scattering, and electron and/or proton bremsstrahlung. However, here more definite conclusions can be drawn.  \\subsection{Difficulties in Synchrotron and IC Processes}  Synchrotron X-ray emission by multi-TeV electrons could not explain the flat  X-ray spectrum. Because of the fast synchrotron cooling on a timescale of $\\tau_{\\rm syn} \\sim 150\\ (\\varepsilon /5~{\\rm keV})^{-1/2}$ yr (for a magnetic field of $25\\,\\mu$G typical in molecular clouds), the photon index of the synchrotron spectrum at X-ray energies becomes $\\Gamma=\\alpha_{\\rm e}/2+1$, where $\\alpha_{\\rm e} \\sim$ 2--2.2 is the electron acceleration index. Thus, the differential spectrum of synchrotron X-radiation cannot be flatter than $\\varepsilon^{-2}$. Note that the power-law fits of the X-ray spectra of regions 2--4 ($\\Gamma =$ 2.1--2.3) agree with the synchrotron origin of X-rays from these regions.  Inverse Compton scattering is precluded by low brightness of synchrotron radio emission. Relatively low-energy (GeV) electrons that produce X-rays via  scattering off the cosmic microwave background photons, also emit synchrotron radiation at radio frequencies. Thus, assuming that the X-ray flux is due to IC scattering, we can calculate the radio flux density of the associated synchrotron emission.  For a magnetic field of $25\\,\\mu$G the latter is expected at a level of $10^4$ Jy at  843 MHz, which is higher, by three orders of magnitudes, than the upper limit on the radio flux from the direction of cloud A  \\citep{Slane99}. Therefore, we can safely exclude the IC origin of the observed X-radiation.  \\subsection{1/$\\varepsilon$ Bremsstrahlung} \\label{sec:brems}  X-rays can be produced through the bremsstrahlung of nonthermal electrons and/or protons interacting with the cloud gas. The energy distribution, for either accelerated protons or electrons within the cloud, is characterized by the position of the break energy,  $E_{\\rm br}$, which is set by equating the lifetime against the ionization losses and the age of the accelerator,  $\\tau_{\\rm ion} = \\tau_{\\rm age}$. At energies below $E_{\\rm br}$, for which $\\tau_{\\rm ion} < \\tau_{\\rm age}$, the particle distribution becomes ``loss-flattened'' due to ionization losses \\citep{Uchiyama02}. As a result, the bremsstrahlung spectrum obey a characteristic energy distribution close to  $I(\\varepsilon) \\propto 1/\\varepsilon$, provided that the energy spectrum of the subrelativistic particles becomes harder than $E^{-1/2}$. Almost independent of the details of the acceleration spectrum, this criterion could  be satisfied by the loss-flattened distribution below $E_{\\rm br}$, which is controlled by the density of the cloud $n$ and the  source age $\\tau_{\\rm age}$. For $n =10^3\\ \\rm cm^{-3}$ and $\\tau_{\\rm age} = 10^3$ yr, the break appears at $E_{\\rm p,br} \\simeq 20$ MeV in the proton distribution, and correspondingly at $\\varepsilon_{\\rm p,br} = (m_{\\rm e}/m_{\\rm p}) E_{\\rm p,br} \\simeq 11$ keV in the proton bremsstrahlung spectrum. On the other hand, the breaks are $E_{\\rm e,br} = \\varepsilon_{\\rm e,br} \\simeq 7$ MeV for electrons. At photon energies anywhere below $\\varepsilon_{\\rm br}$, the characteristic $1/\\varepsilon$ power-law is predicted. Note that the $1/\\varepsilon$ bremsstrahlung photons below $\\varepsilon_{\\rm br}$ are produced predominantly by particles in  a narrow energy interval  around $E_{\\rm br}$. The spectral shape of the flat X-ray spectrum of AX~J1714.1$-$3912 agrees perfectly with the $1/\\varepsilon$ emission of either proton bremsstrahlung (PB) or electron bremsstrahlung (EB) from the loss-flattened distribution.  Whereas the PB and EB mechanisms are expected to give rise to the same $1/\\varepsilon$ X-ray spectrum, the PB luminosity peaks at (hard) X-rays and the EB peaks at gamma-rays for typical parameters. In the case of the $1/\\varepsilon$ EB emission, the photon flux per each logarithmic interval in photon energy anywhere up to $\\varepsilon_{\\rm e,br} \\sim 7$ MeV is constant. Then, the observed X-ray flux of $4\\times 10^{-3}$ photon cm$^{-2}$ s$^{-1}$ in the 1--10 keV band implies that the 1--10 MeV flux should be comparable to the Crab. This is inconsistent with non-detection by the COMPTEL instrument \\citep{Schoenfelder96} onboard the Compton Gamma Ray Observatory. Moreover, unless we introduce a sharp cutoff in the electron distribution at an energy of around 100 MeV, the EB model overshoots the $\\gamma$-ray flux reported by the EGRET instrument. Therefore, we may conclude that  the gamma-ray data favor proton bremsstrahlung, rather than the electron bremsstrahlung scenario.  \\subsection{Energetics Requirement of 1/$\\varepsilon$ Bremsstrahlung}  The X-ray luminosity of the $1/\\varepsilon$ bremsstrahlung emission of either protons or electrons can be estimated as \\begin{equation} \\label{eq:L_x} L_X \\sim \\eta_{\\rm p,e} \\left( \\frac{W_{\\rm p,e}}{\\tau_{\\rm age}} \\right) , \\end{equation} where $W$ is the total amount of kinetic energy contained in X-ray emitting particles, $\\tau_{\\rm age}$ is the age of the source (accelerator), and $\\eta=\\tau_{\\rm ion}/\\tau_{\\rm brem}$. Here, we take into account that the bulk of X-rays are produced by particles from the break region, thus $\\tau_{\\rm age}=\\tau_{\\rm ion}$. For the parameters chosen above, $\\eta_{\\rm p}=3 \\times 10^{-5}$ and $\\eta_{\\rm e}=9 \\times 10^{-5}$. These factors demonstrate the low efficiency of the bremsstrahlung mechanism  relative to the ionization losses, implying a huge,  $\\dot{W}_{\\rm p}=5.6 \\times 10^{39} d_6^2$ erg~s$^{-1}$ and $\\dot{W}_{\\rm e}=1.9 \\times 10^{39} d_6^2$ erg~s$^{-1}$, injection power in protons and electrons, respectively. Correspondingly, the total amounts of energy  released during operation of the accelerator are $W_{\\rm p}^{\\rm sub} = 1.8 \\times 10^{50} d_6^2 (\\tau_{\\rm age}/10^3 {\\rm yr})$ erg and $W_{\\rm e} = 7.2 \\times 10^{49} d_6^2 (\\tau_{\\rm age}/10^3 {\\rm yr})$ erg.  If the particles are accelerated in the shell of SNR RX~J1713.7$-$3946, and only a relatively small (10\\% or so) fraction of these particles enter cloud A, the total kinetic energy, which is transferred to subrelativistic protons, should be at least $10^{51}$ erg; the electron bremsstrahlung model alleviates the total energy by a factor of about 3. Given the limited energy budget of a supernova shock of about $10^{51}$ erg, we face a serious problem to support the required X-ray luminosity, unless the system is closer to the Earth than estimated by the radial velocity of cloud A. For example, a distance of 2 kpc would reduce the energy requirement to a quite comfortable level of $\\sim 2 \\times 10^{50}$ erg. Several other scenarios may be invoked to overcome the energy budget  problem. The simplest assumption would be that the explosion energy significantly exceeds $10^{51}$ erg, and that more than 10\\% of this energy is released in the form of subrelativistic particles. It is also possible that the nonthermal particles are accelerated inside the cloud, e.g.  by the shock initiated at the collision of the blast wave of RX~J1713.7$-$3946 with cloud A \\citep{Bykov00}. Finally, we may  speculate that the essential part of the energy which goes into the ionization and heating of gas, and also the excitation of plasma waves, is returned  to subrelativistic protons through acceleration on these plasma waves. Apparently, all of these assumptions need detailed quantitative studies.  AX~J1714.1$-$3912 and cloud A are located within the error box of the unidentified $\\gamma$-ray source 3EG J1714$-$3857 \\citep{Hartman99}. The GeV $\\gamma$-ray flux may be explained by $\\pi^0$ gamma-rays by collisions between the shock-accelerated protons and cloud~A. \\citet{Butt01} argued that an electron-bremsstrahlung origin for the GeV flux is less likely because of the faint synchrotron radio emission of cloud~A. The total energy liberated in $\\gamma$-rays through the decay of neutral pions is roughly estimated to be $L_{\\gamma} \\sim (1/3) \\sigma_{\\rm pp} n c W_{\\rm p}^{\\rm rel}$ where $\\sigma_{\\rm pp} \\simeq 30$ mb is the proton--proton inelastic cross-section, and $W_{\\rm p}^{\\rm rel}$ is the energy content in GeV protons, within cloud~A. The EGRET luminosity, $L_{\\gamma} = 6.5\\times 10^{35} d_6^2$ erg~s$^{-1}$, corresponds to $W_{\\rm p}^{\\rm rel} = 2.2 \\times 10^{48} d_6^2 (n/10^3 {\\rm cm}^{-3})^{-1}$ erg. Thus, with the proton-bremsstrahlung model for AX~J1714.1$-$3912, the energy content in subrelativistic protons far exceeds that in relativistic protons, $W_{\\rm p}^{\\rm sub} \\simeq 80 \\  W_{\\rm p}^{\\rm rel}$. In this case, it would appear that the acceleration/injection mechanisms allow a small fraction of protons to be accelerated to relativistic energies. Another possibility   could be that  the bulk of accelerated relativistic protons have already diffused away, while the subrelativistic protons are still being captured within the molecular cloud.   \\vspace{2mm}  We thank V. Dogiel for useful discussions. Y.U. is supported by the Research Fellowships of the Japan Society for the Promotion of Science for Young Scientists."
    },
    "0208/astro-ph0208080_arXiv.txt": {
        "abstract": "Systematic surveys are being proposed to discover a significant number of galaxies at $z\\simeq 6$, which is now suggested as the epoch when the reionization era of the Universe ends. To plan such surveys, we need a reasonable expectation of the surface density of high redshift galaxies at different flux limits. Here we present a simple prediction of the surface density of $5.5 \\leq z \\leq 6.5$ galaxies in the optical regime, $extrapolating$ from what is already known about galaxies at $z\\simeq 3$. This prediction is consistent with the results of nearly all known searches for objects at $z\\simeq 6$, giving confidence that we may use it to plan optimal combination of survey depth and sky coverage in searching for such objects. We suggest that the most efficient strategy with existing ground-based facilities is to do medium-depth ($m_{AB}\\simeq 24.0$ -- $24.5$ mag), wide-field (a couple of square degrees) survey using a wide-field camera at a 4m-class telescope. As the predicted surface density at this brightness level is very sensitive to the value of $L^{*}$, the result of such a survey can be easily used to constrain the luminosity evolution from $z\\simeq 3$ to 6. ",
        "introduction": "In the last several years, our knowledge about the Universe at high redshift has been gradually extended to $z\\simeq 6$. As of today, five galaxies at $z > 5.5$ (Weymann \\etal 1998; Hu \\etal 1999, 2002; Dawson \\etal 2001) and four quasars at $z > 5.5$ (Fan \\etal 2000, 2001) have been spectroscopically confirmed. The complete Gunn-Peterson trough detected in the $z=6.28$ SDSS quasar by Becker \\etal (2001) and further investigation (Fan \\etal 2002) led these authors to tentatively identify the end of reionization epoch of the Universe at $z\\simeq 6$. Thus the assessment of $z\\simeq 6$ galaxy number counts at different brightness levels will have a very direct cosmological impact, since it will quantify the number density of UV-emitting objects that are the physical cause of the reionization. Several systematic surveys aimed at discovering a significant number of $z\\simeq 6$ galaxies have now been proposed and are being carried out. To make such surveys efficient, we need to have a rough idea of the surface density of galaxies at $z\\simeq 6$. There are a few theoretical predictions on the surface density of galaxies at high redshifts, either based on N-body simulations (cf. Weinberg \\etal 1999, 2002) or based on semi-analytic formalisms (cf. Robinson \\& Silk 2000). However, those predictions are more qualitative than quantitative at this point, because they either are limited by finite volume and finite resolution, or have to rely on several parameters which remain very uncertain in the absence of significant amounts of actual data. Thus one may prefer not to base survey plans directly on such predictions at the moment.  In this paper we present a simple, observational approach. We may $assume$ a reasonable luminosity function for galaxies at $z\\simeq 6$ by $extrapolating$ the known results at $z\\simeq 3$, which is the highest redshift where the luminosity function of galaxies has been quantified over a wide enough brightness range (Steidel \\etal 1999). We will use existing data at $z$\\cge 5 to constrain the normalization of this $extrapolated$ luminosity function. Once the $z\\simeq 6$ luminosity function is estimated in this way, the surface density can be calculated in a straightforward manner, $i.e.$, by numerically integrating the luminosity function over the volume occupied by unit sky-coverage in the redshift bin of interest. In \\S 2, we present the details of such a surface density prediction over the range $5.5 \\leq z \\leq 6.5$. We compare our prediction to all the available observations in \\S 3. Comparison to two theoretical models is made in \\S 4. A summary is given in \\S 5. ",
        "conclusions": "\\subsection{The Effect of Luminosity Evolution}  The most crucial assumption made in our prediction is that the luminosity evolution from $z\\simeq 3$ to 6 is not significant, so that the value of $L^{*}$ at around restframe 1400\\AA\\ is still the same at $z\\simeq 6$ as at $z\\simeq 3$. However, there are several possibilities where this condition could break down. There are at least two major competing effects which are relevant. One possibility is that the merger/star-forming rate could be lower at $z\\simeq 6$, and so would bring down the value of $L^{*}$. On the other hand, both dust extinction and metallicities could be lower at earlier epochs as well, which would make $L^{*}$ brighter. Since these effects tend to cancel each other out, we believe that to first order our assumption is reasonable. Needless to say, the reality could be more complicated that what we assumed.  For example, it has been suggested that $L^{*}$ evolves mildly in the form $L^{*}\\propto (1+z)^{\\beta}$, where $\\beta$ varies from $-1$ to $-1.5$ ($e.g.$ Lanzetta \\etal 1999). As described below, the surface density at the bright end is extremely sensitive to the $L^{*}$ value. Therefore, a direct comparison between this simple prediction, which serves as a first approximation, and any future observations can give quantitative estimates on how the luminosity evolves from $z\\simeq 3$ to 6. For the sake of completeness, here we discuss how the possibilities mentioned above would affect our prediction.  The time interval between $z=3$ and $z=6$ is about 1.28 Gyr in a $\\Omega _M=0.3$, $\\Omega _\\Lambda =0.7$, and $H_0=65$ universe, and is likely only sufficient to allow one major merger ($cf.$ Makino \\& Hut 1997). Hence a higher merger rate alone would at most make $L^{*}$ at $z\\simeq 3$ twice as bright as at $z\\simeq 6$. On the other hand, a higher merger rate would certainly make star formation rate higher, and this would further contribute to the luminosity. The later effect, however, has not yet been well quantified. As a very rough estimation, the higher star formation rate could contribute another factor of few in increasing $L^{*}$ at $z\\simeq 3$. Thus the overall effect of merger/star formation rate difference would make $L^{*}$ at $z\\simeq 3$ four to six times as bright as at $z\\simeq 6$, or a difference of 1.5--2 mag.  The dust content and metallicities are likely to increase from $z\\simeq 6$ to 3, and these would affect $L^*$ in the opposite way. For example, one of the more reddened object in Steidel et al.'s sample, MS 1512-cB58, is quoted having $E(B-V)\\simeq 0.3$ mag (Pettini \\etal 2001). Assuming the extinction law of Calzetti \\etal (2000), this number translates to $A_{1400\\AA}\\simeq 2.6$ mag. This means the galaxies at $z\\simeq 6$ could get brighter by 2.6 mag at most, if dust extinction is largely absent at this redshift. The metallicity of galaxies at large redshift is again very hard to quantify and seems to have a wide spread ($c.f.$ Nagamine \\etal 2001); but in any case, it is not likely that this effect would contribute more than 0.5 mag increase in brightness at $z\\simeq 6$ at around rest-frame 1400 \\AA.  To investigate how the surface density could be affected by the luminosity evolution, we also calculated the surface densities for different $M^{*}$ values. Specifically, we looked at the limits of the $M^{*}$ value where the high normalization case starts to conflict with the known constraints, and where the low normalization case begins to be consistent with those constraints. Since $\\Omega _M=0.3$ and $\\Omega _\\Lambda = 0.7$ are currently the most widely accepted values, for the sake of simplicity, we will only discuss our (0.3, 0.7) model here. In the high normalization case, if $M^{*}$ is brighter by more than 0.7 mag at $z\\simeq 6$, the predicted counts will conflict with our CTIO upper limit. If $M^{*}$ is fainter by only 0.3 mag at $z\\simeq 6$, on the other hand, the counts will conflict with the lower limit derived from the SDSS QSO hosts by more than a factor of two. In the low normalization case, $M^{*}$ needs to be brighter by at least 2.0 mag to make the counts consistent with the Keck lower limit. In the mean time, however, such $M^{*}$ value makes the counts inconsistent with our CTIO upper limit, overpredicting the counts by a factor of two. This situation further confirms that the low normalization case can be rejected.   \\subsection {Summary}  We present a simple empirical approach to predict the galaxy surface density at $z\\simeq 6$, which $extrapolates$ the known luminosity function of $z\\simeq 3$ galaxies to $z\\simeq 6$. Our approach is based on only two observational results, namely, the observed luminosity function of $z\\simeq 3$ Lyman-break galaxies and the number of $5.5\\leq z \\leq 6.5$ galaxies in the HDF-N, and the assumption that there is no $strong$ luminosity evolution for galaxies from $z\\simeq 3$ to $z\\simeq 6$. The biggest uncertainty in our estimates comes from the normalization, $i.e.$, the actual number density of $z\\simeq 6$ galaxies in the HDF-N down to the limit of $m_{AB}$ = 27.0 mag, for which we used one per WFPC-2 field and one per NIC-3 field as our low and high normalization, respectively. We checked our results against the constraints derived from all known observations. It seems that the low normalization case can be rejected. The high normalization case, on the other hand, is consistent with most constraints if the prediction is made in the (0.3, 0.7) or (1, 0) models. The only observation with which our predictions do not agree is the narrow-band $z=5.7$ \\Lya emitter result reported by Rhoads \\& Malhotra (2001), whose number density is at least 5 times higher than our prediction. If their result were used as the normalization, it would suggest that the number of $z\\geq 5.5$ objects in the HDF be at least 6 times as many as actually found. Since these emitters still need future spectroscopic identification to judge their real nature, we conclude that this is only a potential conflict. On the other hand, a direct comparison between our prediction and any future observations can give quantitative estimation on how the luminosity evolves at different epochs, as indicated in the previous section.  To summarize, we believe that our simple approach can be used to plan future surveys, where there will always be compromise between depth and sky coverage. As our prediction indicates, currently the most realistic way to find a significant number of $z\\simeq 6$ galaxies with the available ground-based facilities is to do multi-color, medium-depth and wide-field surveys reaching continuum $m_{AB}\\sim 24.0$ -- $24.5$ mag from 8400\\AA\\ to the CCD Q.E. cut-off at around 1$\\mu m$, and covering a couple of square degrees. There are now several wide-field CCD cameras available at telescopes of sufficient light-gathering power, e.g., the MOSAIC-I/II at the KPNO/CTIO 4m's, the CFH12K at the CFHT and the Suprime-Cam at the Subaru. Carefully designed surveys at a 4m class telescope could possibly discover a few dozen $L>L^{*}$ $z\\simeq 6$ galaxies within a few nights of observation (Yan \\etal 2002, in preparation). On the other hand, deep, pencil-beam surveys from the ground are not likely to be very successful even with 8-10m class telescopes. As Table 1 indicated, pencil-beam surveys with a few square arcmin field of view would have to reach at least $m_{AB}=27$ mag in the difficult spectrum regime redder than 8400\\AA\\ to discover a significant number of such objects. Since at least two bands of observation at similar depth are needed to select drop-out candidates, the telescope time required is very costly if not unrealistic. In the immediate future, it should be possible to use the the Advance Camera for Surveys, which was installed on board HST in Macrh 2002, for drop-out searches to better constrain the luminosity function at $z\\simeq 6$. Accessing the faint end of the luminosity function at $z\\ge 6$ will be one of the major goals that we will pursue with the Next Generation Space Telescope (NGST) out to redshifts as high $z\\simeq 9-10$, and possibly beyond."
    },
    "0208/astro-ph0208425_arXiv.txt": {
        "abstract": "Several statistics are applied to groups and galaxies in groups in the Two degree Field Galaxy Redshift Survey. Firstly we estimate the luminosity functions for different subsets of galaxies in groups. The results are well fitted by a Schechter function with parameters $M^{\\ast}-5\\log(h)=-19.90 \\pm 0.03$ and $\\alpha=-1.13\\pm0.02$ for all galaxies in groups, which is quite consistent with the results by Norberg et al. for field galaxies. When considering the four different spectral types defined by Madgwick et al. we find that the characteristic magnitude is typically brighter than in the field. We also observe a steeper value, $\\alpha=-0.76\\pm 0.03$, of the faint end slope for low star-forming galaxies when compared with the corresponding field value. This steepening is more conspicuous, $\\alpha=-1.10\\pm 0.06$, for those galaxies in more massive groups (${\\mathcal M} \\gsim 10^{14} h^{-1} M_{\\odot}$) than the obtained in the lower mass subset, $\\alpha=-0.71\\pm 0.04$ (${\\mathcal M}<10^{14} h^{-1} M_{\\odot}$).  Secondly, we compute group total luminosities using Moore, Frenk \\& White prescriptions. We define a flux-limited group sample using a new statistical tool developed by Rauzy. The resulting group sample is used to determine the group luminosity function finding a good agreement with previous determinations and semianalytical models.  Finally, the group mass function for the flux-limited sample is derived. An excellent agreement is obtained when comparing our determination with analytical predictions over two orders of magnitude in mass. ",
        "introduction": "Groups of galaxies constitute one of the most suitable laboratories for the study of properties of intermediate galaxy density environments and their consequences on the process of galaxy formation and evolution. Furthermore, several hints about the large scale structure of the universe and how structures evolve in the universe can be drawn from the statistical studies of groups and their properties. Some of them, for instance the luminosities and morphological types of their member galaxies, are sensitive to the processes of mergers and interactions between individual galaxies inside a potential well. Meanwhile, other properties as group abundance as a function of total luminosity or mass, can provide constraints to hierarchical clustering scenarios and cosmological models. In this paper, we provide a robust statistical study about the luminosities of group galaxy members, and also on global group properties such as total luminosities and masses, using one of the largest group catalogues at the present. The accuracy of these determinations allow us a fair comparison with analytical and semianalytical predictions.  Most studies on the environmental dependence of the galaxy luminosity function have been carried out either in the field or in rich clusters using the largest two dimensional catalogues and small redshift surveys. Regarding field galaxy luminosity functions, several works have been devoted to its determination in the last decade (Loveday et al. 1992, Marzke, Huchra \\& Geller 1994, Lin et al. 1996, Zucca et al. 1997, Ratcliffe et al. 1998). With the advent of large redshift surveys such as the Two degree Field Galaxy Redshift Survey (2dFGRS) and the Sloan Digital Sky Survey (SDSS), more reliable statistical results have been obtained. The field luminosity functions determined by Blanton et al. (2001) (SDSS) and Norberg et al. (2002) (2dFGRS) show an excellent agreement, finding a luminosity function accurately described by a Schechter function with parameters $M_{b_J}^{\\ast}-5\\log(h) \\simeq-19.66$ and $\\alpha\\simeq-1.21$. In particular, the 2dFGRS allowed the determination of the field luminosity function for galaxies of different spectral types (Madgwick et al 2002) finding a systematic steepening of the faint end slope moving from passive $(\\alpha=-0.54)$ to active $(\\alpha=-1.5)$ star forming galaxies, and also a corresponding faintening of $M^{\\ast}$.  A controversial issue about the luminosity function of galaxies in clusters is the steepening at the faint end, compared with field galaxy luminosity function (Valotto et al 1997, Trentham 1997, L\\'opez-Cruz et al 1997). Recently Goto et al. (2002) computed a composed luminosity function using 204 clusters taken from SDSS (York et al. 2000). They found that the slopes of the LF's become flatter toward redder color band and that have brighter characteristic magnitude and flatter slopes than the field LF.  The lack of statistical results on smaller overdensities such as groups of galaxies is mainly due to the fact that two dimensional identification privileges the largest overdensities. Analyzing a sample of 66 groups of galaxies identified in redshift space, Muriel, Valotto \\& Lambas (1998) find a flat faint end for the galaxy luminosity function in groups ($\\alpha \\simeq -1.0$) compared with the luminosity function in clusters where a large relative number of faint galaxies is present. It is important to remark that most of the luminosity function estimations in groups and clusters of galaxies are computed subtracting background and foreground contamination due to the lack of spectroscopic information for galaxy members.  One step further, in order to understand the transition between galaxy and galaxy systems luminosities, is the computation of the luminosity function of galaxy groups. Moore, Frenk \\& White (1993), reanalysing the groups in the Center for Astrophysics (CfA) redshift survey, developed a method for the estimation of the total luminosity of groups identified in magnitude-limited galaxy surveys. This method allowed them the computation of the luminosity function of galaxy systems for a sample of 163 groups with at least tree members. Another attempt to determine the luminosity function of virialized systems was made by Marinoni, Hudson \\& Giuricin (2002). They used the Nearby Optical Galaxy (NOG) sample, which comprise $\\sim 7000$ galaxies with $cz \\leq 6000 \\kms $ and $B \\leq 14$, finding a very good agreement with Moore, Frenk \\& White (1993) previous determination.  On the other hand, the abundance of haloes as a function of mass constitutes a key point in both, the determination of a cosmological model and the understanding of the structure collapse. At the present, the more popular models for halo abundance are the analytical model of Press \\& Schechter (1974) for spherical collapse, the Sheth \\& Tormen (1999) model for ellipsoidal collapse and Jenkins et al. (2001) fit obtained from numerical simulations. Many efforts have been made to determine the mass function of galaxy systems from observations (Bahcall \\& Cen 1993, Biviano et al. 1993, Girardi et al. 1998). Recently, Girardi \\& Giuricin (2000) have computed the mass function for a sample of nearby loose groups by Garcia (1993). They found the group mass function to be a smooth extrapolation of the cluster mass function and a reasonable agreement with the Press \\& Schechter (1974) predictions.  Currently, one of the largest group catalogue was constructed by Merch\\'an \\& Zandivarez (2002). They have identified groups on the 2dF public 100K data release using a modified Huchra \\& Geller (1982) group finding algorithm that takes into account 2dF magnitude limit and redshift completeness masks. This catalogue constitutes a large and suitable sample for both, the study of processes in group environment and the properties of the group population itself. The global effects of group environment on star formation was analysed by Mart\\'{\\i}nez et al. (2002) using this catalogue. They have found a strong correlation between the relative fraction of different galaxy types and the parent group virial mass. For groups with $M\\gsim 10^{13} M_{\\odot}$ the relative fraction of star forming galaxies is significantly suppressed, indicating that even intermediate mass environments affect star formation. Dom\\'{\\i}nguez et al. (2002) presented hints toward understanding local environment effects affecting the spectral types of galaxies in groups by studying the relative fractions of different spectral types as a function of the projected local galaxy density and the group-centric distance. A similar analysis were performed in known galaxy clusters and their environments in the 2dFGRS by Lewis et al (2002).  The aim of this work is to use the Merch\\'an \\& Zandivarez (2002) group catalogue to obtain reliable determinations of internal and global properties of groups: luminosity functions of galaxies in groups, group luminosity and mass functions. The outline of this paper is as follows. In section 2, we present a revised version of the 2dF Galaxy Group Catalogue (2dFGGC) used throughout this work. Section 3 describes the methods and results of the luminosity function of galaxies in groups while in section 4 description corresponds to the luminosity function of galaxy groups. The computation of group mass function and a comparison with analytical models are presented in section 5. Finally, in section 6 we summarize our conclusions. ",
        "conclusions": "Here, we have applied several statistical analysis to groups and galaxies in groups taken from  an updated version of the Merch\\'an \\& Zandivarez (2002) group catalogue (2dFGGC). We have focused on an accurate determination of the luminosity functions of galaxies in groups, and the luminosity and mass functions of groups taking advantage of the statistical power of the 2dFGGC.  In the LF computations, we have used a version of the Choloniewski (1987) approach to the $C^-$ method by Lynden-Bell (1971), adapted to the particular sky coverage of the 2dFGRS. The choice of this particular estimator of the luminosity function is inspired in the conclusions of the comparative analysis of LF estimators by Willmer (1997), who states that the $C^-$ method is less affected by inhomogeneities in the sample. The resulting luminosity function for galaxies in groups (Figure \\ref{fig1}) is well fitted by a Schechter function with shape parameters $M^{\\ast}-5\\log(h)=-19.90\\pm 0.03$ and $\\alpha=-1.13\\pm 0.02$ as determined by a STY fitting procedure. These values are quite consistent with those obtained by Norberg et al. (2002) for field galaxies in the 2dFGRS.  We have performed a similar analysis in subsamples of galaxies (Figure \\ref{fig2} and Table 1) defined by the spectral types of Madgwick et al. (2002). In general, the characteristic magnitudes $M^{\\ast}$ are shifted to higher luminosities with respect to the field values found by Madgwick et al (2002), irrespectively of spectral type. This shift may be due to galaxy dynamical interactions such as mergers, which are expected to be much more frequent in systems with low velocity dispersions as the majority of our group sample.  The faint end slopes of the luminosity functions, $\\alpha$, are consistent with those corresponding to field galaxies except for low star forming, Type 1, galaxies, that show a steeper value in groups (Figure \\ref{comp}). We deepen our analysis for Type 1 galaxies exploring the behaviour of $\\alpha$ for two subsets of groups: low (${\\cal M}\\leq 10^{14} h^{-1} M_{\\odot}$) and high (${\\cal M}>10^{14} h^{-1} M_{\\odot}$), virial masses. We observe an increase in the faint end slope of the Type 1 LF ($\\alpha=-1.10\\pm 0.06$) for galaxies in high mass systems meanwhile for galaxies in low mass systems it remains closer to the global value ($\\alpha=-0.71\\pm 0.04$) (Figure \\ref{type1}). This effect could be the result of internal processes in higher mass system environments such as ram-pressure and galaxy harassment, that are not expected to be significantly important in smaller overdensities.   We have defined group luminosities following Moore, Frenk \\& White (1993), as the sum of its observed luminosity plus a normalised integral of the galaxy luminosity function below the flux limit of the survey. This definition has proved to be reliable in the computation of group total luminosity using observations and artificial flux and volume-limited catalogues. The main aim of this computation is the determination of the luminosity function for our group catalogue. Since the $C^{-}$ method for luminosity function estimation requires a fair selection criterion, we have used the Rauzy (2001) $T_C$ statistics to determine an apparent magnitude cut-off for the 2dFGGC that ensures the highest level of completeness. Our final flux-limited group sample comprises 922 groups with apparent magnitudes brighter than $b_J=15.6$ (Figure \\ref{rauzy}). The resulting group luminosity function for this sample (Figure \\ref{groupLF}) is consistent with Moore, Frenk \\& White (1993) determination for groups in the CfA redshift survey and with the semianalytical model prediction of a $\\Lambda$ cold dark matter cosmology performed by Benson et al. (2000). This agreement is acceptable in the luminosity range $M_{b_J}\\lsim -21.5$. The differences observed at fainter luminosities are mainly due to the lack of groups with less than 4 members in the 2dFGGC. The intrinsic characteristics of the group finding algorithm used in the construction of the 2dFGGC by Merch\\'an \\& Zandivarez (2002), determine that below this limit any resulting sample has an unacceptable level of contamination by spurious detections.  The flux-limited group sample adopted in the luminosity function computation is also fair enough for the determination of the group mass function. This determination was achieved by using an adapted version of the $1/V_{\\rm max}$ that considers the sky coverage of the group sample. Finally, a comparison with analytical predictions of halo abundances is made using a simple scheme to relate group virial masses and that corresponding to the adequate overdensity in our cosmological model. The results are displayed in Figure \\ref{fig6}, where it can be seen a notorious agreement with the Press \\& Schechter (1974), Sheth \\& Tormen (1999) and Jenkins et al. (2001) mass functions for masses ${\\mathcal {M}} \\gsim 10^{13} h^{-1}M_{\\odot}$. Again, the disagreement for low masses can be attributed to absence of poor groups.  The statistical significance of the group catalogue used in this work allowed us to obtain very important clues about internal properties of intermediate mass systems and also to observe the level of agreement obtained from both analytical and semianalytical predictions in the framework of a $\\Lambda$ cold dark matter model."
    },
    "0208/astro-ph0208339_arXiv.txt": {
        "abstract": "Using even-order frequency splitting coefficients of global p-modes it is possible to infer the magnetic field in the solar interior as a function of radial distance and latitude. Results obtained using GONG and MDI data are discussed. While there is some signal of a possible magnetic field in the convection zone, there is little evidence for any temporal variation of the magnetic field in the solar interior. Limits on possible magnetic field in the solar core are also discussed. It is generally believed that the solar dynamo is located in the tachocline region. Seismic studies do not show any significant temporal variation in the tachocline region, though a significant latitudinal variation in the properties of the tachocline are found. There is some evidence to suggest that the latitudinal variation is not continuous and the tachocline may consist of two parts. ",
        "introduction": "Helioseismology has been successful in probing the spherically symmetric structure (Gough et al.~1996) of the Sun as well as the rotation rate in its interior (Thompson et al.~1996; Schou et al.~1998). To the first order, rotation affects only the frequency splitting coefficients which represent odd terms in the azimuthal order $m$ of the oscillation modes. The even terms in these splitting coefficients, can arise from second order effects of rotation, magnetic field or any latitudinal dependence in the structure. It is not possible to distinguish between the effects of a magnetic field and aspherical perturbations to the solar structure (Zweibel \\& Gough 1995).  The even order splitting coefficients are fairly small, and no definitive results have so far been obtained regarding the magnetic field strength in the solar interior. Dziembowski \\& Goode (1989) using data from the Big Bear Solar Observatory claimed to find evidence for a mega Gauss field near the base of the convection zone. Improved data from the Global Oscillation Network Group (GONG) project (Hill et al.~1996) and the Michelson Doppler Imager (MDI) instrument (Rhodes et al.~1997) on board the SOHO satellite has not confirmed these results (Antia et al.~2000).  Instead of a magnetic field one can invoke aspherical structure to explain the even coefficients of frequency splittings. In this case, it is possible to apply an inversion technique to determine the latitudinal dependence in solar structure variables like the sound speed and density (Antia et al.~2001a). The advantage of this approach is that it can give the location of perturbation giving rise to the observed even splitting coefficients.  The GONG and MDI instruments have been observing the Sun for the last 7 years and it is also possible to study possible temporal variation in the internal magnetic field. It is well known that the frequencies of solar oscillations vary with time and this variation is correlated with solar activity (Elsworth et al.~1990; Libbrecht \\& Woodard 1990). Similarly, the even splitting coefficients are also known to vary with time and their variation is correlated to the corresponding component of observed magnetic flux at the solar surface (Libbrecht \\& Woodard 1990; Woodard \\& Libbrecht 1993; Howe et al.~1999; Antia et al.~2001a). However, most of these temporal variations are found to arise from perturbation near the solar surface (Basu \\& Antia 2000; Antia et al.~2001a). There is little evidence for any significant temporal variations in the solar structure below the thin surface layers.  \\begin{figure*} \\hbox to \\hsize{\\resizebox{\\figwidth}{!}{\\includegraphics{bcza2.ps}} \\hfil\\resizebox{\\figwidth}{!}{\\includegraphics{bcza2obs.ps}}} \\caption{ The left panel shows the splitting coefficients $a_2$ from a toroidal magnetic field concentrated near the base of the convection zone, plotted as a function of the lower turning point of the modes. The magnetic field is given by Eqs.~(2,3) with $k=2$, $\\beta_0=10^{-4}$, $r_0=0.713R_\\odot$ and $d=0.02R_\\odot$. In the right panel these coefficients are compared with observed values. Each point represents an average over 25 neighbouring modes. The estimated contribution from rotation has been subtracted from the observed splittings plotted in the figure.} \\end{figure*}  Inversions for rotation rate (Thompson et al.~1996; Schou et al.~1998) have shown that the observed differential rotation at the solar surface continues through the convection zone, while in the radiative interior the rotation rate is more or less independent of latitude. The transition takes place close to the base of the convection zone in a region which has been named as the tachocline (Spiegel \\& Zahn 1992). It is generally believed that the solar dynamo operates in the tachocline region. Hence it would be interesting to look at the temporal variations in the solar structure and rotation rate in the tachocline region. Howe et al.~(2000) found a 1.3 year periodicity in the equatorial rotation rate at $r=0.72R_\\odot$. But other investigations (Antia \\& Basu 2000; Corbard et al.~2001) did not find any systematic variation in the same region. Helioseismic inversions are unreliable in the tachocline region and the properties of the tachocline have been studied using forward modelling approach (Kosovichev 1996; Antia et al.~1998; Charbonneau et al.~1999). These results have also not shown any significant temporal variations in the tachocline properties (Basu \\& Antia 2001).  Although no temporal variations have been seen in the tachocline properties, there is a definite latitudinal variation in the position and possibly also in the thickness of the tachocline (Charbonneau et al.~1999; Basu \\& Antia 2001). On the other hand, there is no latitudinal variation in the depth of the convection zone or the solar structure in the tachocline region. These results appear to be contradictory as it is believed that the tachocline region is mixed by some rotationally induced instability (Richard et al.~1996; Brun et al.~1999). Hence, it would be interesting to study the latitudinal variation in the tachocline with accumulated data over the last seven years.  The global modes of oscillations used in these studies can only give information about the large scale structure and magnetic field in the solar interior. To study smaller features like active regions we need to use local helioseismic techniques, like the time-distance analysis (Duvall et al.~1993) or the ring diagram technique (Hill 1988). In this work we present some results obtained using the ring diagram technique, while the time-distance analysis of active regions is described by Kosovichev (2002). ",
        "conclusions": "Using the even order splitting coefficients it is possible to study the magnetic field and departures from spherical symmetry in the solar interior. Unfortunately, it is not possible to distinguish between these two possibilities. The seismic data from GONG and MDI covering the last 7 years does not show any signal from possible toroidal magnetic field concentrated near the base of the convection zone. An upper limit on such a concentrated field is about 150 kG. The seismic data shows a broad feature around $r=0.9R_\\odot$ and a latitude of $60^\\circ$ which may be due to a magnetic field or aspherical perturbation to the solar structure. The aspherical perturbations to the sound speed are at the level of $10^{-4}$ in this region and if these are due to a magnetic field we may expect a field strength of 70 kG. We do not expect large scale ordered magnetic field inside the convection zone but it is possible that some randomised magnetic field is present. This feature extends to the base of the convection zone where it has a magnitude of $5\\times10^{-5}$, which will correspond to a field strength of 250 kG. The observed splitting coefficients do not yield a tight constraint on the magnetic field in the solar core as very few modes penetrate to the core. However, any magnetic field in this region will yield significant distortion at the solar surface. From the observed distortion we can put an upper limit of $B^2/(8\\pi p_0)<10^{-5}$ in the solar core ($r<0.4R_\\odot$). This corresponds to a field strength of 7 MG at the centre, or 3MG at $r\\approx0.2R_\\odot$ or 0.8 MG at $r\\approx 0.4R_\\odot$.  There is no significant temporal variation in aspherical component of sound speed or density in the solar interior. This also applies to possible temporal variations in magnetic field. Thus any temporal variation in solar interior is less than about $5\\times10^{-5}$, which is the error estimate inside the convection zone at low latitude. By taking temporal average over the high activity and low activity data sets it appears that there may be a small increase in sound speed asphericity with activity at the level of $10^{-5}$ in a broad region centred at latitude of $60^\\circ$. This is comparable to the error estimate and its significance is not clear. The MDI data do show some temporal variation, but a closer look shows that most of the temporal variation has occurred during the time when SOHO had lost contact. Thus this is likely to be an artifact of systematic error introduced during recovery of SOHO. This systematic error will also affect other inferences about temporal variation obtained using the MDI data. This systematic error appears to be predominantly in high degree ($\\ell>110$) modes.  It is generally believed that the solar dynamo is operating in the tachocline region. But no significant temporal variation is seen in properties of the tachocline. Nevertheless, the tachocline is known to be prolate and there is indeed a significant latitudinal variation in the position and thickness of the tachocline. However, this latitudinal variation may not be continuous and it appears that the tachocline may consist of two parts one at low latitude where the rotation rate increases with radius and second one at high latitude where the rotation rate decreases with radius. These two parts may be located at different depths and have different thicknesses. But there may be no significant variations within each part. The thickness difference between the two parts should match the depth variation to ensure that the lower limit of mixing due to the tachocline is essentially independent of latitude."
    },
    "0208/astro-ph0208175_arXiv.txt": {
        "abstract": "We present a new set of isochrones in which the effect of the $\\alpha$-element enhancement is fully incorporated. These isochrones are an extension of the already published set of Y$^2$ Isochrones (Yi et al. 2001: Paper~1), constructed for the scaled-solar mixture. As in Paper~1, helium diffusion and convective core overshoot have been taken into account. The range of chemical compositions covered is $ 0.00001 \\le Z \\le 0.08 $. The models were evolved from the pre-main-sequence stellar birthline to the onset of helium burning in the core. The age range of the full isochrone set is 0.1 -- 20\\,Gyr, while younger isochrones of age 1 -- 80\\,Myr are also presented up to the main-sequence turn-off. Combining this set with that of Paper~1 for scaled-solar mixture isochrones, we provide a consistent set of isochrones which can be used to investigate populations of any value of $\\alpha$-enhancement. We confirm the earlier results of Paper~1 that inclusion of $\\alpha$-enhancement effects further reduces the age estimates of globular clusters by approximately 8\\% if [$\\alpha$/Fe]=+0.3. It is important to note the metallicity dependence of the change in age estimates (larger age reductions in lower metallicities). This reduces the age gap between the oldest metal-rich and metal-poor Galactic stellar populations and between the halo and the disk populations. We also investigate whether the effects of $\\alpha$-enhancement can be mimicked by increasing the total metal abundance in the manner proposed by Salaris and collaborators. We find such simple scaling formulae are valid at low metallicities but not at all at high metallicities near and above solar. Thus it is essential to use the isochrones rigorously computed for $\\alpha$-enhancement when modeling metal-rich populations, such as bright galaxies. The isochrone tables, together with interpolation routines have been made available via internet. ",
        "introduction": "This paper presents an extension of the $Y^2$ Isochrones: scaled-solar mixtures (Yi et al. 2001: Paper~1). Its purpose is to study the effect of the $\\alpha$-element enhancement on theoretical isochrones. Already for some time, it has become clear that the abundances of some chemical elements, in particular the elements synthesized by nuclear $\\alpha$ capture reactions (the so-called $\\alpha$-elements; e.g., O, Ne, Mg, Si, S, Ca, and Ti etc.), are enhanced with respect to iron in Population II stars. \\citet{Nor01} is one of the latest such studies which report that metal-poor stars do not have a scaled-solar chemical composition. One popular interpretation for this is that Population II stars formed in a relatively fast collapse of the proto-galaxy, in which a rapid chemical enrichment and $\\alpha$-enhancement through efficient feedback from massive stars took place. Similar $\\alpha$-element enhancements have also been observed in super metal-rich (greater than solar) stellar populations. Spectrophotometric studies of giant elliptical galaxies \\citep{Pel90, Rich92}, and more recent spectroscopic studies of stars in the bulge \\citep{McW99} and thick disk \\citep{Pro00} of the Galaxy, using the Keck~I telescope, indicate that various degrees of $\\alpha$-element enhancement are commonplace. A thorough discussion of the significance of stellar abundances in understanding the evolution of stellar populations can be found in the reviews by \\citet{Whe89} and \\citet{McW97}. Applications to elliptical galaxies, which also appear to be $\\alpha$-enhanced, have been performed by \\citet{Wei95} and \\citet{SW98}, using the spectral indices of \\citet{Wor92}.  Because opacity tables for $\\alpha$-enhanced mixtures covering the complete relevant range of temperature and density became available only recently, several sets of theoretical isochrones have been constructed assuming scaled-solar abundances for the elements heavier than helium. However, the obvious limitation is that they refer to the scaled-solar compositions only. For stars with $\\alpha$-enhanced chemical compositions, one needs then to construct stellar models and isochrones for the particular distribution of metals. Although it is possible in principle to calculate a large number of possible combinations of $\\alpha$-enhanced ratios, it would not be practical considering the degree of the uncertainties in abundance analysis and the size of the computational effort. \\citet{Van00}, and \\citet{Salas00} are among the latest who published sets of the $\\alpha$-enhanced isochrones assuming a particular choice for the $\\alpha$-enhanced chemical compositions, based on the OPAL opacities (Rogers \\& Iglesias 1995, Iglesias \\& Rogers 1996).  In this study, we construct a set of $\\alpha$-enhanced isochrones which is fully consistent with the standard set of $Y^2$ Isochrones previously released for scaled-solar mixtures (Paper~1). The purpose of this research is twofold. One is to provide an extensive and consistent set of the $Y^2$ Isochrones for general use. And, the other is to explore the validity of the common practice of mimicking $\\alpha$-enhanced isochrones by adopting more metal-rich non-$\\alpha$-enhanced isochrones, utilizing  a simple scaling formula in the heavy element abundance \\citep{Chi91, Cha92, Sal93}. ",
        "conclusions": "This paper presents an extension of the $Y^{2}$ isochrones \\citep{paper1} including the effects of $\\alpha$-element enhancement. Two values of $\\alpha$-enhancement, [$\\alpha$/Fe]=$+$0.3 and $+$0.6 have been considered.  As in Paper 1, pre-MS evolutionary phases are included, allowing for the construction of younger isochrones. Isochrones including the complete hydrogen burning phase, suitable for population synthesis, have been constructed for the age range 0.1 -- 20\\,Gyr. In the case of stars with a convective core near the MS, overshoot by 0.2 pressure scale height at the edge of the convective core has been taken into account. An additional set of very young isochrones for ages 1 -- 80\\,Myr is presented. They are unsuitable for population synthesis because their post-MS phases are not included but would be useful for comparing with the CMDs of very young stellar populations.  We confirm that the use of simple scaling formulae, such as the one introduced by Salaris et al.~(1993), can approximate the effect of $\\alpha$-enhancement on the isochrone morphology especially at low metallicities. Thus, such formulae, in the absence of appropriate models, can be useful to the studies of globular clusters and dwarf spheroidal galaxies. However, the range of validity of such scaling laws is limited. In general, because of the complex interaction between opacities, the equation of state and nuclear processes, it is not possible to derive a scaling formula that is applicable for a large range of $Z$.  From the point of view of stellar population chronology, this paper confirms that for a given [Fe/H], the updated isochrones that include the effects of $\\alpha$-element enhancement lead to significant age reductions of up to 23\\% if [$\\alpha$/Fe]=$+$0.3, as shown in Table \\ref{age}. Approximately a third of this reduction is due to a factor of two increase in the $\\alpha$-element abundances (see the difference between the second and fifth columns in Table \\ref{age}). A new result, which is equally important, is the age increase at high $Z$, which removes the age gap that used to exist between the oldest metal-rich and metal-poor stellar populations.  Caveats of this work include the treatment of $\\alpha$-enhancement in the temperature-color transformation. Theoretical stellar properties (temperature, gravity, etc.) are converted to observable quantities (magnitudes and colors) based on a stellar spectral library (in our case, either the Green et al. table or the Lejeune et al. table). Such a spectral library is usually calibrated empirically so that it reproduces the colors of sample stars in the Milky Way for a given set of parameters. In fact, the calibration is based on a limited and incomplete sample. For example, such Milky Way samples may imply [$\\alpha$/Fe]=0.3 for [Fe/H]=$-$2, [$\\alpha$/Fe]=0.15 for [Fe/H]=$-$0.7, and [$\\alpha$/Fe]=0.0 for [Fe/H]=0. Users of these isochrones should be warned that for this reason, interpolated isochrones, even if well within our parameter range ([$\\alpha$/Fe] = 0.0 -- 0.6), are not necessarily based on a proper temperature-color transformation scheme. Some colors are more sensitive to [$\\alpha$/Fe] than others. Thus, a full $\\alpha$-enhancement treatment will be possible only when temperature-color transformation schemes are available for a wide range of $\\alpha$-enhancement as well.  Similarly, the importance of $\\alpha$-enhancement effects on the spectral line analysis for clusters and galaxies has been recognized \\citep{Vaz01} as well. The effects of $\\alpha$-enhancement occur mainly in two steps. The first is in the stellar model construction through opacities and energy generations. The next is through the convolution of the stellar models with a stellar spectral library. For this reason, one must have a properly $\\alpha$-enhanced stellar spectral library for a wide range of $\\alpha$-enhancement in order to generate consistent population synthesis models.  Despite the caveats, we believe that these new isochrones with $\\alpha$-enhancement options will be useful in many studies. The full set of isochrones and a FORTRAN package that work for age, metallicity, and $\\alpha$-enhancement interpolation are available at the following internet sites.  {\\tt http://www-astro.physics.ox.ac.uk/$\\sim$yi/yyiso.html}  {\\tt http://www.astro.yale.edu/demarque/yyiso.html}  {\\tt http://csaweb.yonsei.ac.kr/$\\sim$kim/yyiso.html}"
    },
    "0208/astro-ph0208496_arXiv.txt": {
        "abstract": "{ We present a sample of 17 type 2 Seyfert galaxies which have an X-ray column density lower than 10$^{22}$ cm$^{-2}$. The Compton thin nature of these sources is strongly suggested by isotropic indicators. We estimate the fraction of these sources to be in the range of 10\\% - 30\\% of the population of type 2 Seyfert galaxies. Furthermore, this fraction appears to increase progressively at lower luminosities. The simple formulation of the Unified Model for Seyfert galaxies is not applicable in such sources since the pc-scale molecular torus is not likely to be responsible for the low column density observed, instead the absorption observed is likely to originate at larger scales. According to this hypothesis, in these objects the broad line regions are covered by some dusty obscuring material. In particular, this could occur in objects with dust lanes, patches or HII regions. However, we cannot rule out that in the lowest luminosity sources the BLR is weak, absent or has faded away. This last scenario is consistent with the predictions of some recent theoretical models for low luminosity AGNs. ",
        "introduction": "Seyfert galaxies belong to the class of Active Galactic Nuclei (AGN). According to the standard model, in active galaxies an accretion disk around a massive black hole produces a hard X-ray continuum, which photoionizes the Broad Line Region (BLR, where broad emission lines originate) and the Narrow Line Region (NLR, where narrow emission lines originate) located at $<$ 1 pc and at $<$ 100 pc from the nuclear engine respectively. Seyfert galaxies are classified as type 1 or type 2. Type 1 have both narrow forbidden lines (FWHM $\\leq$ 10$^{3}$ km/s) and broad Balmer lines (FWHM $\\sim$ 10$^{4}$ km/s) in their optical spectrum, while type 2 have only narrow lines. Actually they are the same object: type 2 Seyferts harbour a BLR, but this is obscured from view in some directions by a molecular torus (Unification Model; Antonucci, 1993).  Optical spectropolarimetry measurements of scattered broad permitted lines provide strong evidence in favour of the unified model (Antonucci \\& Miller 1985). At least 35\\% of Seyfert 2 galaxies have broad emission lines seen in polarized light (Tran 2001; Moran et al. 2000), therefore a good fraction of Seyfert 2 galaxies seem to host a hidden Seyfert 1 nucleus.  More evidence in favour of the unified model comes from the X-ray spectra: the column density  of neutral hydrogen in type 2 Seyferts is significantly higher than in type 1 objects as  would be expected if, for the type 2 sources, the nucleus is observed through the torus (Turner et al. 1997; Smith \\& Done 1996). Observed column  densities range from 10$^{22}$ cm$^{-2}$ to higher than 10$^{24}$ cm$^{-2}$ for $\\sim$ 96\\% of the objects (Risaliti et al. 1999; Bassani et al. 1999).  However, not all Seyfert 2 galaxies have a Broad Line Region in polarized light and not all Seyfert 2 galaxies have column densities higher than 10$^{22}$ cm$^{-2}$. Polarimetric surveys of complete samples of Seyfert 2s indicate that a large fraction of these objects (up to 50\\%) do not show a hidden BLR typical of an obscured Seyfert 1 nucleus. Furthermore there have been some recent examples of Seyfert 2 galaxies, such as NGC 3147 (Ptak et al. 1996), NGC 4698 (Pappa et al. 2001) and NGC 7590 (Bassani et al. 1999), which have no or low absorption measured from the X-ray spectrum. It can be argued that these are Compton thick objects i.e. in which the medium is thick to Compton scattering such that the transmitted component is completely suppressed below 10 keV and the 2-10 keV spectrum is dominated by reprocessed components. In this case the hard X-ray spectrum is characterized by a flat Compton reflection component from the inner surface of the torus and/or a steeper component ascribed to an ionized, warm scattering medium. When the absorbing medium has column density N$_{H}$ $>$ 10$^{24}$ cm$^{-2}$, then the transmitted component can be observed above 10 keV. Therefore, in these sources the true column density can only be estimated by higher energy data for N$_{H}$ $>$ 10$^{24}$ cm$^{-2}$ or measured indirectly by comparing the X-ray luminosity with the Far-Infrared or [OIII] luminosities for even higher column densities. In the above mentioned sources these absorption indicators suggest that they are actually Compton thin objects. At the moment at least $\\sim$ 4\\% of Seyfert 2s have N$_{H}$ $<$ 10$^{22}$ cm$^{-2}$ (Risaliti et al. 1999). The exact nature of these peculiar Seyfert 2s is still unclear, as it is not obvious what obscures their Broad Line Region: they may be intrinsically different objects than those explained by the unified theory or in other words they may be the ``true'' Seyfert 2 galaxies which are sometimes discussed in the literature (Tran 2001).  In this paper we have collected a sample of Seyfert 2 galaxies characterized by low X-ray absorption in order to study their properties, estimate their abundance and understand better their nature. ",
        "conclusions": ""
    },
    "0208/astro-ph0208033_arXiv.txt": {
        "abstract": "{Several popular cosmological models predict that most of the baryonic mass in the local universe is located in filamentary and sheet-like structures associated with galaxy overdensities. This gas is expected to be gravitationally heated to $\\sim 10^6 ~K$ and therefore emitting in the soft X-rays. We have detected diffuse soft X-ray structures in a high Galactic latitude ROSAT field after point source subtraction and correction for Galactic absorption. These diffuse structures have an X-ray energy distribution that is much softer than expected from clusters, groups or unresolved emission from AGNs, but are consistent with that expected from a diffuse warm intergalactic medium. To discriminate between a Galactic or extragalactic nature of the diffuse gas we have correlated the soft X--map with multiband optical images in this field. We have found a significant overdensity of galaxies in correspondence with the strongest diffuse X-ray structure. The photometric redshift distribution of the galaxies over the X-ray peak has an excess over field galaxies at $z\\sim0.45$. This result strongly suggests that the diffuse X-ray flux is due to extragalactic emission by warm gas associated with an overdense galaxy region at $z\\sim0.45$. ",
        "introduction": "The mismatch between the density of baryons observed in the local Universe and the baryon density observed and predicted at high redshift is currently one of the most puzzling issues in cosmology. While the density of baryons, $\\Omega _b$, observed in stars and gas ($\\rm{HI}+ \\rm{HII}$) in the local Universe does not exceed 0.01 \\citep{fukugita}, observations of the Ly$\\alpha$ forest at $z=2$ \\citep{rauch} and Big Bang nucleosynthesis constraints \\citep{burles} both give $\\Omega _b \\simeq 0.04$ or larger \\citep{pettini}. One possibility is that at $z<1$ the baryonic gas falls onto the cosmic web pattern and is heated by shock mechanisms forming filamentary and sheet-like structures \\citep{cen,dave}. Such diffuse gas, called Warm--Hot Intergalactic Medium (WHIM), should be detectable in the soft X-rays as a consequence of having a temperature in the range $\\rm 10^5 ~K < \\rm{T} < 10^7 ~K$. A tight connection with the filamentary distribution of galaxies is expected from N-body simulations (see \\citet{bond} for a theoretical discussion). \\\\ Various models of diffuse X-ray emission have been proposed by several authors during the past year \\citep{phillips,kuntz,bryan}; nonetheless, observational evidence for diffuse/filamentary X-ray emission is still sparse. \\citet{wang} found an excess of emission in some ROSAT fields which they ascribed to a diffuse component of the X-ray background due to WHIM. \\citet{soltan} detected a correlation signature between the soft X-ray background and galaxies. Filamentary soft X-ray structures were identified by \\citet{warwick} in various overlapping ROSAT pointings near the Lockman Hole. Structures laid among clusters were found by \\citet{kull} in the core of the Shapley Supercluster and \\citet{tittley} along the line of sight connecting two Abell clusters. \\citet{scharf} presented tentative evidence of a soft X-ray filamentary structure which seems to correlate with the density of galaxies measured in the I-band. Recently, \\citet{bagchi} pointed out the discovery of a presumably filamentary structure both in radio and soft X-ray traced by a galaxy arc.   We have re-analyzed several ROSAT pointings toward a region close to the Lockman Hole with exposure times $> 10 \\rm{~ksec}$ and, after removal of the point sources and correction for Galactic absorption, we have detected diffuse X-ray emission. We have started a program of optical multiband wide field imaging of the regions showing diffuse X-ray emission with the goal of detecting associated galaxy overdensities, which would support the extragalactic nature of the large scale soft X-ray structures.  In this paper we report preliminary results on one of the ROSAT fields for which we have obtained HI radio data and which has been partially mapped in five optical bands. We present the discovery of a galaxy overdensity at a photometric redshift $z\\sim$0.45 spatially coincident with the most prominent diffuse X-ray structure in the ROSAT field.\\\\ The plan of the paper is as follows: in Sect.\\ref{xmaps} we discuss the analysis of the X-ray ROSAT maps, in Sect.\\ref{HI} the reduction of the radio data and the HI absorption correction for X-ray maps are presented, in Sect.\\ref{galaxies} we discuss the optical images analysis. Sect.\\ref{photom z} contains the photometric redshift estimate and, finally, in Sect.\\ref{summary} a summary of the results is presented.  \\begin{figure*}[!] \\begin{center} \\vskip2truecm \\caption{Panel {\\em (a)} Soft (R2, $0.14-0.284 \\rm{~keV}$) map smoothed to 15 arcsec showing the point sources; {\\em (b)} Hard (R6+7, $0.73-2.04 \\rm{~keV}$) map with the same smoothing. The inner circle (1~deg in diameter) encloses the region where the point source removal is reliable. The two crosses indicate the location of the structures discussed in Fig.\\ref{maps}.} \\label{maps_p} \\end{center} \\begin{picture}(100,-100) \\put(70,245){{\\em (a)}} \\put(280,245){{\\em (b)}} \\end{picture} \\end{figure*} \\begin{figure*}[t!] \\begin{center} \\includegraphics[angle=0, width=0.4\\textwidth]{fg2a.ps} \\includegraphics[angle=0, width=0.4\\textwidth]{fg2b.ps} \\end{center} \\begin{center} \\includegraphics[angle=0, width=0.4\\textwidth]{fg2c.ps} \\includegraphics[angle=0, width=0.4\\textwidth]{fg2d.ps} \\end{center} \\begin{center} \\caption{ {\\em (a)} Soft X-ray map smoothed to a resolution of 5.9 arcmin after removing the point sources identified in the R2 and R6+7 maps; {\\em (b)} Wavelet map showing extended structures with high ($4\\sigma$) statistical significance; {\\em (c)} Map of Galactic HI column density (contours from 5.3 to 10.7~10$^{19}$cm$^{-2}$, spaced by 0.6~10$^{19}$), lighter gray areas correspond to lower HI columns; {\\em (d)} Soft map corrected for HI absorption. As in Fig.\\ref{maps_p} the inner circle (1 deg in diameter) encloses the region where the point source removal is reliable. The two structures, marked with crosses, are in a region where the ROSAT PSF is only $30\\%$ worse than that in the center.} \\label{maps} \\end{center} \\begin{picture}(100,-100) \\put(70,525){{\\em (a)}} \\put(280,525){{\\em (b)}} \\put(70,310){{\\em (c)}} \\put(280,310){{\\em (d)}} \\end{picture} \\end{figure*} ",
        "conclusions": "\\label{summary}  We have analyzed a deep ROSAT field in a region of high Galactic latitude and low Galactic HI absorption. After removal of the point sources and after correction for absorption, the softest map show evidence for diffuse/extended structures. These diffuse structures have an X-ray energy distribution that is much softer than expected from clusters, groups or unresolved emission from AGNs, but it is consistent with emission from the diffuse warm intergalactic medium expected by the cosmological models, both in terms of shape (plasma at kT$\\sim$0.3~keV) and of flux ($\\rm \\langle F_{0.2-0.3keV}\\rangle \\approx 10^{-12}~erg~cm^{-2}s^{-1}deg^{-2}$).  To discriminate between a Galactic or extragalactic nature of the diffuse gas we have correlated the soft X--map with multiband optical images in this field. The most prominent diffuse X-ray structure in the ROSAT map appears associated with an overdensity of galaxies at a photometric redshift of $\\sim 0.45$. This association, along with the X-ray properties of the former (sect.\\ref{xmaps}), strongly suggest that we are observing an extragalactic structure most likely tracing the warm intergalactic medium predicted by cosmological theories, in this case at redshift $\\sim 0.45$, which is expected to be the main reservoir of baryonic matter at low redshifts.  The second most prominent diffuse X-ray structure ``2'' is not associated, in our optical maps, with a pronounced galaxy overdensity, although there are some clumps of galaxies surrounding it (Fig. \\ref{clust_map}). Either this structure is not extragalactic (e.g. associated to our Galactic halo) or it is associated with galaxies at redshift higher than 0.8, where most galaxies escape detection in our optical images. Finally, this may be a case of warm baryonic gas at relatively low redshift, not enclosing galaxy overdensities, but possibly bridging those few galaxy clumps detected in its surrounding.  Multi-object spectroscopy should provide a critical test on the nature of these diffuse structures. Indeed the specroscopic information would both confirm the galaxy redshift distribution and give an estimate of the involved virial masses to be compared with the temperature inferred for the barionic gas."
    },
    "0208/astro-ph0208519_arXiv.txt": {
        "abstract": "We have carried out a submillimeter continuum and spectroscopic study of the W43 main complex, a massive star-forming region, which harbors a giant \\hii region.  The maps reveal a filamentary structure containing $\\sim 50$ fragments with masses of $40-4\\,000~\\msun$ and typical diameters of 0.25~pc.  Their large sizes, large non-thermal velocities ($\\Delta v \\sim 5~\\kms$), and high densities ($n\\htwo\\sim 10^6~\\cmc$) suggest that they are protoclusters and excellent sites to form massive stars.  Follow-up observations are necessary, but we have already identified three protoclusters to be very good candidates for containing very young massive protostars.  The starburst cluster, that excites the giant \\hii region has a large impact on the molecular complex.  However, it remains unclear if this first episode of star formation is triggering the formation of new massive stars, through ionization shocks crossing the closeby molecular clouds.  W43 is thus an ideal laboratory to investigate massive star formation from the protostellar phase to that of giant \\hii regions.  Moreover, the very active star-forming complex W43 may be considered a Galactic mini-starburst region that could be used as a miniature model of starburst galaxies. ",
        "introduction": "High-mass (OB; $M_\\star>8~\\msun$) stars are believed to form in clusters within molecular cloud complexes.  Due to their high ultraviolet luminosity, massive young stellar objects (YSOs) first heat, then ionize, and disrupt their surrounding molecular cloud. Owing to their large distances to the Sun and the complex interplay between massive YSOs and the neighboring interstellar medium, the formation of high-mass stars is still poorly understood.  Many studies have been devoted to embedded YSOs that have already developed an \\hii region and thus are easily detectable in far-infrared and centimeter continuum surveys (see a review by \\citealt{chur99}).  In contrast, possible precursors of ultracompact \\hii regions (UCH\\mbox{\\sc ~ii}s), i.e. massive YSOs in their main building phase, have only been discovered recently (\\citealt{hunt00, bran01, srid02} and references therein).  These massive protostars, also called high-mass protostellar objects (HMPOs), are inconspicuous in the IRAS bands and have no, or very little free-free emission.  They can be detected in (sub)millimeter dust continuum and molecular high-density tracers and frequently display H$_2$O/CH$_3$OH maser emission.  Therefore, submillimeter continuum imaging of molecular cloud complexes is ideal to probe clouds surrounding UCH\\mbox{\\sc ~ii}s and HMPOs simultaneously.  Making a census of those deeply embedded phases is the first necessary step to gain insight into the processes leading to the formation of a massive star.  The W43 star-forming complex is located in the inner spiral arm of our Galaxy, at 5.5~kpc from the Sun \\citep{wils70}.  W43 is well-known for its giant \\hii region emitting $10^{51}$ Lyman continuum photons per second and a far-infrared continuum luminosity of $\\sim 3.5\\times 10^6~\\lsun$ \\citep*{smit78, lest85}.  The main ionizing source of W43 was discovered in near-infrared images by \\citet{lest85} and confirmed by \\citet*{blum99} as a cluster of Wolf-Rayet (WR) and OB main sequence stars.  The inner 20~pc of W43 (at $l\\sim 30.75\\degr$, $b\\sim -0.06\\degr$) contain this WR/OB cluster and a $10^6~\\msun$ molecular cloud called G30.8-0.0.  The W43 main complex was mapped in CO lines and parts of it in higher density tracers such as H$_2$CO and \\hcops lines, and 1.3~mm continuum \\citep*{bieg82, lisz95, moon95}.  Several sources were identified by these authors but a complete and comprehensive sample of UC\\hii regions and massive protostars is still lacking.  Identifying such a sample is essential for the present paper, which aims at presenting a global scenario of (massive) star formation in W43.  Therefore, we will carefully investigate the distribution of dense molecular clouds in the W43 main complex, in particular to identify sites of future or on-going massive star formation.  A detailed study of W43 may help constrain the properties of distant starburst galaxies.  Indeed, the large luminosity and ionizing flux of the W43 giant \\hii region are similar to those of NGC~3603 or M17 ($10^5-10^7~\\lsun$ and $10^{50}-10^{51}$~Lyc$\\,$s$^{-1}$) , taken to be representative of clusters and \\hii regions in starburst galaxies (e.g. \\citealt{tapi01}).  The NGC~3603 stellar cluster is qualified as ``starburst'' because it consists of several tens of WR, O, and B-type stars (e.g. \\citealt{bran99}).  The W43 cluster of main sequence stars is likely to be similar but more heavily reddened ($A_{\\rm v}=30$~mag versus 4~mag in NGC~3603).  Since the W43 stellar cluster is closely associated with giant molecular clouds, we may be witnessing another burst of star formation.  This would make the W43 main complex a Galactic mini-starburst region, i.e. a miniature model of the stellar and gas content of starburst regions in distant galaxies.  In the present paper, we report a submillimeter continuum and spectroscopic study of the W43 main complex.  From the imaging and deep spectroscopic measurements presented in Sect.~2, we make a complete census of the compact and dense cloud fragments in W43 (Sect.~3).  In Sect.~4, we investigate the impact of the giant \\hii region on the molecular complex, the nature of these cloud fragments, and the global characteristics of the mini-starburst W43.  Finally, Sect.~5 summarizes our conclusions. ",
        "conclusions": "W43 is a molecular and \\hii complex whose ionized gas has been far better studied than its molecular component.  We have imaged the W43 main cloud at submillimeter wavelengths in the 1.3~mm and $350~\\micron$ continuum and \\hcop(3-2) line emission.  In addition, we have obtained deep \\hcop(3-2) and \\htcop(3-2) spectra at selected locations.  Our main findings can be summarized as follows:  \\begin{enumerate}  \\item A multiresolution analysis on our submillimeter continuum maps identifies $\\sim 50$ compact fragments.  Our $350~\\micron$ continuum observations along with the 3.5~cm image of \\cite{bal01} show that their 1.3~mm emission is largely thermal emission from cool dust. \\item These bona-fide cloud fragments have diameters varying from 0.09~pc to 0.56~pc and masses spanning the range $20~\\msun$ to $3\\,600~\\msun$.  Their large size and turbulent line width ($\\Delta v \\sim 5~\\kms$) suggest they are protoclusters, i.e. clouds that contain many smaller-size and denser structures and will form star clusters.  Those protocluster candidates have large mean densities ($n\\htwo\\sim 10^6~\\cmc$) reminiscent of the direct progenitors of individual low-mass stars.  The W43 protocluster candidates thus constitute an excellent sample for studies of the earliest stages of massive star formation. \\item  The present, unbiased survey is sensitive to protoclusters either being pre-stellar or containing massive protostars or UC\\hii regions.  Five of the W43 protoclusters are confirmed to contain massive YSOs in their HMPO or UC\\hii phase.  Follow-up observations are needed to determine the evolutionary state of the remaining protocluster candidates. \\item In W43, the centimeter and infrared emission from the giant \\hii region dominates, preventing the detection of more compact sources that could be associated with UCH\\mbox{\\sc ~ii}s.  Notably, at least two of the three IRAS point sources of the W43 main cloud are not stellar in nature; they are instead associated with ionization fronts possibly all excited by the closeby WR/OB cluster. \\item While the low-density clouds surrounding the starburst cluster may have been blown away, the densest parts of the W43 molecular clouds seem to remain at the systemic velocity given by Galactic motions. A more precise kinematic study is necessary to determine if the ionization front associated with the giant \\hii region is compressing the molecular clouds and triggering star formation. \\item W43 is the site of at least two remarkably efficient episodes of massive star formation.  Indeed, it is known to harbor a starburst cluster containing several WR and OB stars ($\\sim 10^6~\\lsun$).  We show here that the molecular complex is currently undergoing a second mini-starburst with a star formation efficiency of $\\sim 25\\%/10^6$~yr and possibly a final stellar density of $\\sim 100$ stars/pc$^3$ over (14~pc)$^3$.  Learning about the global characteristics of this Galactic mini-starburst region should help constraining the physical processes at work in the distant starburst galaxies. \\end{enumerate}"
    },
    "0208/astro-ph0208205_arXiv.txt": {
        "abstract": "We used optical spectroscopy of the neutron star X-ray transient XTE~J2123--058 to measure the rotational broadening of the binary companion's stellar absorption lines and determined that the companion's projected rotational velocity is $v\\sin i = 121^{+21}_{-29}$~km~s$^{-1}$.  This value is considerably larger than the measurements of $v\\sin i$ obtained previously for three other neutron star systems where the values are between 34 and 65~km~s$^{-1}$.  The larger value is likely due to the combination of high binary inclination and short (6~hr) orbital period for XTE~J2123--058.  Along with previously measured parameters, the rotational broadening measurement allowed us to determine the binary parameters, including the ratio of the neutron star mass to the companion mass, $q = M_{1}/M_{2} = 2.7^{+1.3}_{-0.9}$, the neutron star mass, $M_{1} = 1.46^{+0.30}_{-0.39}$\\Msun, and the companion mass, $M_{2} = 0.53^{+0.28}_{-0.39}$\\Msun, assuming a Roche lobe filling companion synchronously rotating with the binary orbit.  These values are consistent with a previous measurement where the H$\\alpha$ emission line was used to determine the semiamplitude of the neutron star's radial velocity curve ($K_{1}$).  It is a significant result that the two methods give consistent values.  We also report the first measurement of the XTE~J2123--058 companion's radius.  Assuming synchronous rotation, we obtain $R_{2} = 0.62^{+0.11}_{-0.15}$~\\Rsun, which is in-line with that expected for a late K-type star on or close to the main sequence.  Finally, we report the first precise spectroscopic determination of the binary orbital period ($P_{orb} = 21442.3\\pm 1.8$~seconds). ",
        "introduction": "The X-ray transient XTE~J2123--058, discovered in 1998 \\citep{lss98}, is a neutron star low-mass X-ray binary (LMXB) with a 6~hr orbital period \\citep{tomsick99,zurita00}.  Although many of the properties of this system are rather typical of LMXBs such as type I X-ray bursts and high frequency quasi-periodic oscillations (QPOs), XTE~J2123--058 distinguishes itself from other LMXBs by having high Galactic latitude ($b = -36^{\\circ}$) and a high and relatively well-determined binary inclination ($i = 73^{\\circ}\\pm 4^{\\circ}$; Zurita et al.~2000\\nocite{zurita00}). A main reason for the good inclination constraint is that partial eclipses are present in the outburst optical light curves \\citep{tomsick99,zurita00}.  For optical observations, the high Galactic latitude is advantageous since the extinction along the line-of-sight to the source is low.  The high binary inclination is useful for measuring the rotational broadening because the widths of the companion's absorption lines increase with inclination.  The ultimate goal of this project is to obtain a precise measurement of the neutron star mass.  Such mass measurements are important for constraining neutron star equations of state (EOS) and for understanding the evolution of neutron star systems.  Precise mass measurements have been made for millisecond radio pulsars (MSPs, Thorsett \\& Chakrabarty 1999\\nocite{tc99}), but these measurements are lacking for LMXBs.  Since it is theoretically possible to spin-up neutron stars in LMXBs to millisecond periods via accretion, a link between LMXBs and MSPs has long been suspected \\citep{alpar82}. Although there is substantial evidence to support this picture, the prediction that rapidly rotating neutron stars in LMXBs should be more massive by 0.1 to 0.5\\Msun~(Bhattacharya 1995\\nocite{bhattacharya95}) than those that have not been spun-up has not been tested.  Previous work on optical observations of XTE~J2123--058 in quiescence resulted in measurements of the semiamplitude of the companion's radial velocity curve ($K_{2}$), the companion's spectral type, the distance to the source and the systemic velocity \\citep{tomsick01,casares02}. \\cite{casares02} also obtained a measurement of the semiamplitude of the neutron star's radial velocity curve ($K_{1}$) using the H$\\alpha$ emission line that is present in the spectrum, giving a determination of the mass ratio ($q = K_{2}/K_{1} = M_{1}/M_{2}$). Although they considered their $K_{1}$ measurement to be tentative due to uncertainties concerning using the broad H$\\alpha$ line to infer the radial velocity of the neutron star, they used their results to obtain a neutron star mass of $1.55\\pm 0.31$~\\Msun~(68\\% confidence errors).  Here, we use the observations of \\cite{tomsick01} to determine the projected rotation rate of the companion ($v\\sin i$) from a measurement of the rotational broadening \\citep{mrw94}.  From this, we obtain a measurement of $q$ that relies only on measurements of the companion's absorption lines and is independent of the \\cite{casares02} mass ratio measurement.  Although rotational broadening measurements have been used to obtain mass measurements for several black hole binaries, this has only been previously accomplished for three other neutron star LMXBs:  Cyg~X-2 \\citep{cck98}; Cen~X-4 \\citep{torres02}; and 2S~0921--630 \\citep{shahbaz99}.  A measurement of $v\\sin i$ was also claimed for Aql~X-1 \\citep{scc97}; however, the subsequent discovery of a field star within $0^{\\prime\\prime}.46$ of Aql~X-1 \\citep{wry00} is likely to have some impact on this measurement. Thus, XTE~J2123--058 provides a relatively rare opportunity to obtain a rotational broadening measurement in a neutron star LMXB, which is an important step toward obtaining a precise neutron star mass measurement. This study also provides a valuable test of the method most commonly used to measure compact object masses since we expect to obtain a neutron star mass that is considerably less than the 5-15\\Msun~compact object masses obtained for the black hole systems. ",
        "conclusions": "We used moderate resolution optical spectra from Keck Observatory to carry out a rotational broadening measurement for the XTE~J2123--058 companion.  After a detailed analysis where we account for possible sources of systematic error, we obtained $v\\sin i = 121^{+21}_{-29}$~km~s$^{-1}$ for the companion's projected rotational velocity.  Using this result, our determination of the spectroscopic orbital period and previous measurements of $K_{2}$ \\citep{tomsick01} and $i$ \\citep{zurita00}, we calculated the values of the binary parameters, including $q = 2.7^{+1.3}_{-0.9}$, $M_{1} = 1.46^{+0.30}_{-0.39}$ and $M_{2} = 0.53^{+0.28}_{-0.39}$, assuming a Roche lobe filling companion synchronously rotating with the binary orbit.  Assuming only synchronous rotation, we also obtained the first measurement of the XTE~J2123--058 companion's radius, $R_{2} = 0.62^{+0.11}_{-0.15}$~\\Rsun, which is in-line with that expected for a late K-type star on or close to the main sequence.  One of the most significant results of this work is that our measurements of the binary parameters using the rotational broadening method are consistent with the values found by \\cite{casares02} by using the H$\\alpha$ emission line to determine $K_{1}$.  Although the comparison of the two techniques for obtaining compact object masses is limited to a precision of 20-30\\% by the errors on $M_{1}$, this provides an important test of the methods currently used to measure compact object masses.  However, there is some indication that $q$ is underestimated by the H$\\alpha$ emission line method from the fact that \\cite{zurita00} obtain values of $q$ between 3 and 5 when modeling the XTE~J2123--058 outburst light curves to find the binary inclination. While our measurement of $q$ is consistent with this range, \\cite{casares02} find a lower value of $q = 2.1\\pm 0.4$. This issue could be explored further by measuring the ellipsoidal modulations for XTE~J2123--058 in X-ray quiescence.  Also, the error on $v\\sin i$ in this work is dominated by statistical errors and could be reduced through further spectroscopic observations."
    },
    "0208/astro-ph0208211_arXiv.txt": {
        "abstract": "We present a sample of 150 narrow-line Seyfert 1s (NLS1s) found within the Sloan Digital Sky Survey Early Data Release (EDR), only two of which were previously identified as such.  This substantially increases the known number of NLS1s, and provides a basic method by which to identify many more with subsequent releases of SDSS data.  With its large size and homogeneous, well-defined selection criteria, this sample will help alleviate two major problems which have plauged NLS1 research in the past; namely, their relative rarity and significant differences in selection algorithms between the known samples.  45 of these SDSS-selected NLS1s are detected at energies of 0.1--2~keV in the ROSAT All-Sky Survey (RASS), and are found to have ultrasoft X-ray spectra with photon indices of $\\Gamma \\ga 2$, in agreement with previous results for NLS1s. However, about 10--20 of those NLS1s that were not detected by ROSAT have optical properties very similar to the detected objects, and so should also have been detected by the RASS.  This may be due to either significant intrinsic absorption in many NLS1s, or a significant sub-class of NLS1s that have uncharacteristic, intrinsically flatter (hence harder) X-ray spectral energy distributions. ",
        "introduction": "Since their initial classification by \\citet{ostpog85} narrow-line Seyfert 1s (NLS1s) have gained noteriety as interestingly extreme examples of active galactic nuclei (AGN).  They were initially defined by their relatively narrow permitted emission lines \\citep[FWHM $\\la$ 2000~km~s$^{-1}$;][]{goodrich89}, strong \\ion{Fe}{2} relative to H$\\beta$, and weak [\\ion{O}{3}].  In their analysis of 87 bright AGN, \\citet{bg92} found a strong anticorrelation between the strengths of the [\\ion{O}{3}] and \\ion{Fe}{2} lines (the primary correlation behind their so-called eigenvector 1 or Principal Component 1 [PC1]).  NLS1s lie at the extreme, low--[\\ion{O}{3}] end of PC1.  The authors suggested that this may be due to a high accretion rate relative to the Eddington rate.  A more recent analysis by \\citet{boroson02} reinforces this claim, noting that NLS1s consistently exhibit the lowest estimated central black hole masses for similar luminosities and the highest inferred relative accretion rates among the various AGN types.  Much attention has been devoted in recent years to the X-ray spectra of NLS1s. While AGN have long been known to emit a substantial fraction of their luminosity as X-rays, a soft X-ray ``excess'' was noted among Seyfert galaxies \\citep[e.g.,][and references therein]{puch92}. Boller, Brandt, \\& Fink (1996) found a strong anticorrelation between the X-ray photon index ($\\Gamma$, where $f_E \\propto E^{-\\Gamma}$) and FWHM(H$\\beta$), with NLS1s having $\\Gamma \\ga 2.5$.  Consequently, selection on the basis of ultrasoft X-ray emission has proven effective in the discovery of new NLS1s \\citep{grupe00}. This soft excess is thought to be the high-energy (Wien) tail of thermal emission from the inner accretion disk.  Since extremely high temperatures are required to produce this emission, it again follows that NLS1s may be powered by low-mass black holes at high relative accretion rates \\citep{pounds95, wang96}.  To date, a combination of X-ray and optical selection and serendipity has resulted in the discovery of a large number of NLS1s \\citep[see][for a review]{pogge2000}: for example, the sample compiled by \\citet[][hereafter VVG01]{vvg01} consists of 64 NLS1s with z~$<$~0.1, B~$<$~17.0, and $\\delta > -25\\degr$, while the {\\it Catalog of Quasars and Active Nuclei} \\citep[10th edition:][]{vv01} lists 205 NLS1s among Seyferts and QSOs. While these provide a starting point for studying the role of NLS1s among AGN phenomena, the sample size is relatively small and quite heterogeneous due to the wide variety of selection criteria employed. Since many known NLS1s were first discovered in X-rays, it is difficult to ascertain with confidence whether the extreme X-ray softness exhibited by most catalogued NLS1s is a fundamental property or a subtle selection effect. Clearly, a large and homogeneous optically selected sample would be advantageous in resolving these issues.  Such a sample is now becoming available in the form of the Sloan Digital Sky Survey \\cite[SDSS;][]{york2000}, in particular the Early Data Release \\citep[EDR;][]{stoughton02}, released in mid-2001.  The EDR contains spectra of approxmiately 4000 quasars \\citep{schneider02} as well as a large number of Seyfert galaxies and other AGN.  Photometric data are measured in five bands \\citep[$u^\\prime,g^\\prime,r^\\prime,i^\\prime,$ and $z^\\prime$;][]{fukugita96}, and criteria based on these bands are used to select QSO candidates for spectroscopic follow-up, as described in \\citet{richards02}. Parameters such as redshifts, magnitudes and linewidths are stored in a searchable database, with photometric properties in the ``PhotoObj'' class and spectral properties in the ``SpecObj'' and ``SpecLine'' classes.  This database also has built into it an ``ExternalCatalog'' class, which contains all EDR objects within 60\\arcsec\\ of objects catalogued in the ROSAT All-Sky Survey (RASS), as well as some ROSAT-measured properties of these objects (see section~\\ref{xrayprop}).  By submitting a query restricted to objects with narrow H$\\beta$ emission lines and then analyzing the resulting spectra, we have identified 150 NLS1s in the EDR, only two of which have been previously classified as such.  In the following section we discuss in more detail the selection criteria and subsequent analysis.  The overall selection methods and optical properties are discussed, as well as objects which were previously known and/or misidentified.  Finally, we report on those objects which were also observed in RASS, and give possible reasons why some were not detected when they should have been. ",
        "conclusions": "}  The 150 SDSS-selected NLS1s presented in this paper represent a significant increase in the total known number of these extreme AGN.  They comprise approximately 15\\% of the EDR ``QSO'' database at $z \\la 0.5$ and have very well-defined color and linewidth selection criteria.  45 of these NLS1s were also detected with good confidence in the 0.1--2~keV band of the ROSAT All-Sky Survey and exhibit ultrasoft X-ray spectra.  Of the NLS1s that were \\emph{not} detected, several have similar optical properties to NLS1s seen in the RASS, and thus should have been detected as well.  This may be due to either high intrinsic absorption or harder X-ray spectra (or both) among the undetected objects. More optical and X-ray data will almost certainly be helpful in determining the cause behind this.  It should also be noted that while most of our objects clearly fall within the defining NLS1 criteria, a substantial fraction are near the 2000~km~s$^{-1}$ FWHM cutoff.  Additionally, several of the spectra had low signal--to--noise ratios.  This may hide characteristics (such as a broad component in H$\\beta$) which would reclassify the object as a Seyfert 1.5 or other type.  While these objects all appear to be NLS1s from the data given, it is likely that some may be reclassified when higher resolution, higher signal-to-noise spectra are obtained.  Thus, this sample should be considered a list of very strong NLS1 candidates rather than a definitive list.  Nevertheless, this sample demonstrates that large numbers of NLS1s (and other interesting objects) with well-constrained selection criteria can indeed be found in the SDSS Early Data Release. Since the EDR represents only about 5\\% of the total spectroscopic survey \\citep{stoughton02}, the full SDSS catalog will provide a definitive resource for the discovery and study of NLS1s."
    },
    "0208/gr-qc0208060_arXiv.txt": {
        "abstract": " ",
        "introduction": "A compact object of solar-mass size orbiting a supermassive black hole is one of promising candidates for the source of gravitational waves. Since the internal structure of such a compact object may be neglected in this situation, we may adopt the black hole perturbation approach with the compact object being regarded as a point particle. In the black hole perturbation approach, we consider the metric perturbation induced by a point particle of mass $\\mu$ orbiting a black hole of mass $M$, where $\\mu\\ll M$. At the lowest-order in the mass ratio ($\\mu/M$), the motion of the particle follows a geodesic of the background spacetime. In the next order, however, the particle moves no longer along a geodesic of the background because of its interaction with the self-field.  Although this deviation from a background geodesic is small for $\\mu/M\\ll1$ at each instant of time, after a large lapse of time, it accumulates to become non-negligible. For example, a circular orbit will not remain circular but becomes a spiral-in orbit and the orbit eventually plunges into the black hole.  If the time scale of the orbital evolution due to the self-force is sufficiently long compared to the characteristic orbital time, we may adopt the so-called adiabatic approximation in which the orbit is assumed to be instantaneously geodesic with the constants of motion changing very slowly with time. In the Schwarzschild background case, we may assume the orbit to lie on the equatorial plane and the geodesic motion is determined by the energy and (the $z$-component of) angular momentum of the particle. In this case, the time variation of the energy and angular momentum can be determined from the energy and angular momentum emitted to infinity and absorbed into the black hole horizon by using the conservation law.  However, there are cases when the adiabatic approximation breaks down. For example, in the case of an extremely eccentric orbit or an orbit close to the inner-most stable circular orbit, the orbital evolution will not be adiabatic because the stability of the orbit is strongly affected by an infinitesimally small reaction force. Furthermore, in the Kerr background, there is an additional constant of motion, known as the Carter constant. Intuitively, it describes the total orbital angular momentum, but unlike the case of spherical symmetry, it has nothing to do with the Killing vector field of the Kerr geometry. The lack of its relation to the Killing vector makes us impossible to evaluate the time change of the Carter constant from the gravitational waves emitted to infinity and to event horizon, even in the case when the adiabatic approximation is valid. Thus it is in any case necessary to derive the self-force of a particle explicitly.  The gravitational self-force $F^\\mu$ is formally given as \\[ \\frac{d^2 z^{\\alpha}}{d\\tau^2} +\\Gamma_{\\mu\\nu}^{\\alpha}\\frac{dz^{\\mu}}{d\\tau}\\frac{dz^{\\nu}}{d\\tau} =F^{\\alpha} \\,, \\] where $\\{z^{\\alpha}(\\tau)\\}$ represents the orbit with $\\tau$ being the proper time measured in the background geometry and $\\Gamma_{\\mu\\nu}^{\\alpha}$ is the connection of the background. The self-force arizes from the metric perturbation $h_{\\mu\\nu}$ induced by the particle: \\[ \\tilde g_{\\mu\\nu} = g_{\\mu\\nu} + h_{\\mu\\nu}\\,, \\] and it is expressed as \\[ F^{\\alpha}[h]= -\\mu P_{\\beta}^{\\alpha} (\\bar{h}_{\\beta\\gamma;\\delta} -\\frac{1}{2}g_{\\beta\\gamma} {\\bar{h}^{\\epsilon}}{}_{\\epsilon;\\delta} -\\frac{1}{2}\\bar{h}_{\\gamma\\delta;\\beta} +\\frac{1}{4}g_{\\gamma\\delta} {\\bar{h}^{\\epsilon}}{}_{\\epsilon;\\beta} )u^{\\gamma}u^{\\delta}\\,, \\] where ${P_{\\alpha}}^{\\beta}={\\delta_{\\alpha}}^{\\beta}+u_{\\alpha}u^{\\beta}$, $\\bar{h}_{\\alpha\\beta}=h_{\\alpha\\beta} -\\frac{1}{2}g_{\\alpha\\beta}h$ and $u^{\\alpha}=dz^\\alpha/d\\tau$.  The metric perturbation diverges at the location of the particle and so does the self-force. Thus the above formal expression is in fact meaningless. Fortunately, however, it is known that the metric perturbation in the vicinity of the orbit can be divided into two parts under the harmonic gauge condition; the direct part which has support only on the past light-cone emanating from the field point $x^\\mu$ and the tail part which has support inside the past light-cone, and the physical self-force is given by the tail part of the metric perturbation which is regular as we let the field point coincide with a point on the orbit; $x^\\mu\\to z^\\mu(\\tau)$ \\cite{Mino:1996nk,Quinn:1999kj}. It must be noted that the direct part can be evaluated by local analysis, i.e., only with the knowledge of local geometrical quantities. Therefore the physical self-force can be calculated as \\[ \\lim_{x\\to z(\\tau)}F_{\\alpha}[h^{\\rm tail}(x)] = \\lim_{x\\to z(\\tau)} \\left(F_{\\alpha}[h(x)]-F_{\\alpha}[h^{{\\rm dir}}(x)]\\right). \\] Furthermore, it has been revealed recently by Detweiler and Whiting \\cite{Detweiler:2002mi} that the above devision of the metric can be slightly modified so that the new direct part, called the S part, satisfies the same Einstein equations as the full metric perturbation does, and the new tail part, called the R part, satisfies the source-free Einstein equations, and that the R part gives the identical, regular self-force as the tail part does. The important point is that the S part can be still evaluated locally near the orbit without knowing the global solution.  When we perform this subtraction, we must evaluate the full self-force and the direct part under the same gauge condition. But the direct part is, by definition, defined only in the harmonic gauge. On the other hand, the full metric perturbation is directly obtainable only by the Regge-Wheeler-Zerilli or Teukolsky formalism \\cite{Regge:1957td,Zerilli:wd,Teukolsky:1973ha, Chrzanowski:wv}. Therefore one must find a gauge transformation that brings both the full metric perturbation and the direct part of the metric perturbation to those in the same gauge. This is called the {\\it gauge problem} \\footnote{Furthermore, the gravitational self-force is, because of the equivalence principle, a gauge-variant notion. To give a genuinely physical meaning to it, one must solve the second order metric perturbation completely. This is however beyond the scope of the present paper.}.  For the direct part, methods to obtain it under the harmonic gauge condition were proposed \\cite{Mode-sum,Mino:2001mq,Barack:2001gx,Barack:2002mh}. However, it seems extremely difficult to solve the metric perturbation under the harmonic gauge because the metric components couple to each other in a complicated way. This is one of the reasons why the gauge problem is difficult to solve.  Recently, Barack and Ori \\cite{Barack:2001ph} gave a useful insight into the gauge problem. They proposed an intermediate gauge approach in which only the direct part of the metric in the harmonic gauge is subtracted from the full metric perturbation in the RW gauge. They then argued that the gauge-dependence of the self-force is unimportant when averaged over a sufficiently long lapse of time. Using this approach, the gravitational self-force for an orbit plunging into a Schwarzschild black hole was calculated by Barack and Lousto \\cite{Barack:2002ku}. But they also pointed out that the RW gauge is singular in the sense that the resulting self-force will still have a direction-dependent limit for general orbits. The situation becomes worse in the Kerr background where the only known gauge in which the metric perturbation can be evaluated is the radiation gauge \\cite{Chrzanowski:wv}, but the metric perturbation becomes ill-defined in the neighborhood of the particle, i.e., the Einstein equations are not satisfied there \\cite{Barack:2001ph}.  In this paper, as a direct approach to the gauge problem, we consider a formalism to calculate the metric perturbation in the harmonic gauge. We focus on the Schwarzschild background. Instead of directly solving the metric perturbation in the harmonic gauge, we consider the metric perturbation in the RW gauge first, and then transform it to the one in the harmonic gauge. Namely, we derive a set of equations for gauge functions that transform the metric perturbation in the RW gauge to the one in the harmonic gauge.  The paper is organized as follows. In Sec.~\\ref{sec:form}, we formulate the gauge transformation from the RW gauge to the harmonic gauge. First we decompose the gauge transformation generators into the Fourier-harmonics components. Then the generators are devided into three parts; the odd parity part and the even parity part which is further devided into scalar and divergence free parts. By the above procedure, we find a set of decoupled equations for the gauge functions. In Sec.~\\ref{sec:discussion}, we summarize our formulation and discuss remaining issues. In Appendix \\ref{app:RWZ}, we recapitulate the equations for the Regge-Wheeler-Zerilli formalism by correcting typos in Zerilli's paper \\cite{Zerilli:wd}. ",
        "conclusions": "\\label{sec:discussion}  In this paper, to solve the gauge problem of the gravitational self-force, we have considered the gauge transformation from the Regge-Wheeler gauge to the harmonic gauge and have presented a formalism to obtain the infinitesimal displacement vector of this transformation, $\\xi^{\\mu}$. First, we have performed the Fourier-harmonic expansion of $\\xi^{\\mu}$ and divided it into the odd and even parity parts. The odd part has only one degree of freedom and it turns out that the gauge transformation can be found by solving a single second-order differential equation for the radial function. As for the even parity part, we have further divided it into scalar and vector parts where the scalar part is given by the gradient of a scalar function and the vector part is divergence-free. The scalar part has by definition only one degree of freedom, and we have found that it can be obtained by solving two second-order differential equations consecutively. These two equations are found to be identical to the $s=0$ Teukolsky equation. The vector part has two degrees of freedom, and the gauge transformation equations give equations that are coupled in a complicated way. However, by introducing two auxiliary variables which satisfy the $s=\\pm1$ Teukolsky equations, we have succeeded in deriving a decoupled second-order equation for one of the gauge functions with the source term given by the auxiliary variables. Interestingly, this second-order equation has the same form as the $s=0$ Teukolsky equation. The other gauge function is then simply given by applying a differential operator to the first.  Since all the equations to be solved have the form analogous to or equal to the Regge-Wheeler equation, we can derive analytic expressions for their homogeneous solutions by using the Mano-Suzuki-Takasugi method \\cite{Mano1}, and construct the Green function from these homogeneous solutions. So we conclude that the gauge transformation can be solved by using the Green function method, and we can construct the metric perturbation in the harmonic gauge. In practice, however, it may not be easy to solve for the gauge transformation since it involves products of Green functions with double integrals. Derivation of the gauge transformation functions in a closed, practically tractable form is left for future study.  Another approach to the gauge problem is to consider the self-force in a gauge different from the harmonic gauge, similar to (but very different in principle from) the intermediate gauge approach proposed by Barack and Ori \\cite{Barack:2001ph}. Here the recent result by Detweiler and Whiting \\cite{Detweiler:2002mi} becomes crucial. Their observation that the S part and the R part play the identical roles as the direct part and the tail part, respectively, and that the S part satisfies the same inhomogeneous Einstein equations as the full metric perturbation enables us to define the S part and the R part of the metric perturbation unambiguously in an arbitrary gauge as long as the gauge condition is consistent with the Einstein equations. For example, given the S part of the metric perturbation in the harmonic gauge, one can perform the gauge transformation of it to the RW gauge and the resulting metric perturbation which satisfies the Einstein equations can be identified as the S part of the metric perturbation in the RW gauge. Then, after solving the Regge-Wheeler-Zerilli equations to obtain the full metric perturbation, it is straightforward to derive the R part of the metric perturbation in the RW gauge \\cite{Mino}. The calculation of the self-force in the RW gauge in this manner is in progress \\cite{SNS2}.  Finally, we comment on the self-force in the case of the Kerr background. In the Schwarzschild case, it was possible to use the Regge-Wheeler-Zerilli formalism to obtain the metric perturbation in the RW gauge. However, in the Kerr case, there is no known gauge in which the full metric perturbation can be calculated. The Chrzanowski method \\cite{Chrzanowski:wv} based on the Teukolsky formalism can give the metric perturbation in the (ingoing or outgoing) radiation gauge, but only outside the range of radial coordinates the orbit resides in. One possible way to circumvent this difficulty is to consider first the regularization of the Weyl scalar $\\Psi_4$. Given an orbit, $\\Psi_4$ can be calculated by the Teukolsky formalism, and the S part of it, $\\Psi_4^{\\rm S}$, can be calculated from the S part of the metric perturbation in the harmonic gauge, $h_{\\mu\\nu}^{\\rm S,H}$, \\begin{eqnarray} \\Psi_4^{\\rm S} = \\hat{\\Psi}_4[h_{\\mu\\nu}^{\\rm S,H}] \\,, \\end{eqnarray} where $\\hat{\\Psi}_4$ is the operator to derive the Wely scalar from a given metric perturbation. Then the R part of $\\Psi_4$ can be derived by subtracting the S part from the Weyl scalar, \\begin{eqnarray} \\Psi_4^{\\rm R} = \\Psi_4 -\\Psi_4^{\\rm S} \\,. \\end{eqnarray} Now $\\Psi_4^{\\rm R}$ satisfies the homogeneous Teukolsky equation. Hence using the Chrzanowski method, we may construct the R part of the metric perturbation in the radiation gauge and derive the self-force. Since this procedure involves many derivative operations, the metric perturbation $h_{\\mu\\nu}^{\\rm S,H}$ has to be evaluated to with a sufficiently high accuracy which may be practically a difficult task, if not impossible. Feasibility of this method should surely be investigated."
    },
    "0208/astro-ph0208357_arXiv.txt": {
        "abstract": "We have continued our long term study of the double-neutron-star binary pulsar PSR~B1534+12, using new instrumentation to make very high precision measurements at the Arecibo Observatory.  We have significantly improved our solution for the astrometric, spin, and orbital parameters of the system, as well as for the five ``post-Keplerian'' orbital parameters that can be used to test gravitation theory.  The results are in good agreement with the predictions of general relativity.  With the assumption that general relativity is the correct theory of gravity in the classical regime, our measurements allow us to determine the masses of the pulsar and its companion neutron star with high accuracy: $1.3332\\pm0.0010M_\\odot$ and $1.3452\\pm0.0010M_\\odot$, respectively. The small but significant mass difference is difficult to understand in most evolutionary models, as the pulsar is thought to have been born first from a more massive progenitor star and then undergone a period of mass accretion before the formation of the second neutron star.  PSR~B1534+12 has also become a valuable probe of the local interstellar medium.  We have now measured the pulsar distance to be $1.02\\pm0.05$~kpc, giving a mean electron density along this line of sight of 0.011\\,cm$^{-3}$.  We continue to measure a gradient in the dispersion measure, though the rate of change is now slower than in the first years after the pulsar's discovery. ",
        "introduction": "\\label{sec:intro}  The discovery \\citep{ht75a} of the first pulsar in a binary system, PSR~B1913+16, introduced the possibility of powerful new experimental tests of gravity in the strong field and radiative regimes.  Over a quarter century, measurements of this pulsar were shown to be in excellent agreement with the predictions of general relativity \\citep{tw89,dt91,tay94}.  A dozen years ago the discovery \\citep{wol91a} of PSR~B1534+12---a bright, relatively nearby pulsar, also with a neutron-star companion---promised the independent opportunity to confirm the tests of gravity done earlier with PSR~B1913+16. Furthermore, B1534+12's favorable, nearly edge-on orbital geometry allowed new experimental tests that were complementary to those done with PSR~B1913+16 \\citep{twdw92}.  In earlier work \\citep[hereafter Paper~I]{sac+98} we analyzed data taken over a six year period with the Arecibo 305~m telescope, the 43-m telescope at Green Bank, West Virginia, and the 76-m Lovell Telescope at Jodrell Bank Observatory, U.K.  We found that those data were in agreement with the predictions of general relativity at the limit of the measurement uncertainties: about 1\\%.  We showed that the orbital period was decreasing, as expected if the system is losing energy in the form of gravitational radiation.  Because the observed orbital decay rate is contaminated by kinematic effects which depend on the unknown pulsar distance \\citep{dt91,ctk94}, the radiation predictions of general relativity cannot be tested with this pulsar to better than $\\sim30\\%$ without an independent distance measurement. However, with the assumption that general relativity is the correct theory of gravity, we were able to invert the kinematic model to measure the pulsar distance to 20\\%.  Over the past four years, we have continued our studies of PSR~B1534+12, using the recently upgraded Arecibo telescope together with a new data acquisition system \\citep{sst+00} that fully removes the dispersive effects of the interstellar medium from the signal.  In addition to extending the length of the data set by more than half, these new data are of substantially higher quality than those available for Paper~I.  In this paper, we report on timing studies based on the complete 11.5-year Arecibo data set.  We present the improved tests of general relativity that are now possible, give a pulsar distance accurate to 5\\%, and report on dispersion measure variations.  In subsequent papers, we will describe the analysis of secular variations in the pulsar radiation pattern that have been interpreted as evidence for geodetic precession of the pulsar, and we will discuss single-pulse studies. ",
        "conclusions": "\\label{sec:disc}  \\subsection{Observed Change in Orbital Period:  Distance to PSR\\,B1534+12}  The observed $\\dot P_b$ is measured in the reference frame of the solar system barycenter, and must be corrected to the center-of-mass frame of the binary pulsar system before it can be compared to the value predicted by GR.  The largest correction is kinematic, involving the relative acceleration of the two reference frames.  It can be broken down into the vertical acceleration in the Galactic potential, acceleration in the plane of the Galaxy, and an apparent centripetal acceleration due to the transverse velocity of the pulsar binary \\citep{dt91}: \\begin{equation}\\label{eq:gal} \\left(\\frac{\\dot{P_b}}{P_b}\\right)^{\\rm gal} = -\\,\\frac{a_z\\sin b}{c} \\,-\\,\\frac{v_0^2}{cR_0} \\left[\\cos l + \\frac{\\beta}{\\sin^2 l + \\beta^2}\\right] +\\mu^2\\frac{d}{c}. \\end{equation} Here $a_z$ is the vertical component of galactic acceleration, $l$ and $b$ the pulsar's galactic coordinates, $R_0$ and $v_0$ the Sun's galactocentric distance and galactic circular velocity, $\\mu$ and $d$ the pulsar's proper motion and distance, and $\\beta=d/R_0 - \\cos l$. As we do not have a precise distance to the pulsar through either a timing or an interferometric parallax measurement, the best independent distance estimate still comes from the \\citet{tc93} model of the free electron content of the Galaxy; this model puts the pulsar at a distance of $0.7\\pm0.2$\\,kpc.  At this distance the \\citet{kg89} model of the Galactic potential yields an estimate of $a_z/c=(1.60\\pm0.13)\\times10^{-19}\\,\\mbox{s}^{-1}$.  We assume $v_0=222\\pm20\\,$km\\,s$^{-1}$ and $R_0 = 7.7\\pm0.7$\\,kpc, as in \\citet{dt91}.  Then, summing the terms in equation~(\\ref{eq:gal}) and multiplying by $P_b$, we find the total kinematic bias to be \\begin{equation} \\left(\\dot P_b\\right)^{\\rm gal} = (0.037\\pm0.011)\\times10^{-12}\\,. \\end{equation} The uncertainty in this correction is dominated by the uncertainty in distance, which is only roughly estimated by the Taylor and Cordes model.  The slight decrease in the uncertainty from that given in Paper~I results from an increase by a factor of ten in the precision of the proper motion measurement.  Our measurement of the intrinsic rate of orbital period decay is therefore \\begin{equation} \\left(\\dot P_b\\right)^{\\rm obs} - \\left(\\dot P_b\\right)^{\\rm gal} = (-0.174\\pm0.011)\\times 10^{-12}\\,. \\end{equation} The uncertainty is completely dominated by the uncertainty on the kinematic correction, which is nearly a factor of 4 larger than the measurement uncertainty on $\\dot P_b^{\\rm obs}$.  In GR, the orbital period decay due to gravitational radiation damping, $(\\dot{P_b})^{\\rm GR}$, can be predicted from the masses $m_1$ and $m_2$ (eq.~\\ref{eq:pbdot}), which in turn can be deduced from the high precision measurements of $\\dot{\\omega}$ and $\\gamma$.  As listed in Table~\\ref{tab:orbparms}, the expected value is \\begin{equation} \\left(\\dot P_b\\right)^{\\rm GR} = -0.192\\times 10^{-12}\\,.\\label{eq:pbdotgr} \\end{equation} As in Paper~I, our measured value differs from this prediction, now by some 1.7 standard deviations, and again, assuming that GR is the correct theory of gravity, we can derive the true distance to the pulsar from the ``excess $\\dot P_b$'' parameter and equation~\\ref{eq:gal} above \\citep{bb96}.  Our improved distance is $d=1.02\\pm0.05$~kpc (68\\% confidence limit).  The uncertainty is still dominated by the measurement uncertainty of $(\\dot P_b)^{\\rm obs}$, rather than uncertainties in the galactic rotation parameters or the acceleration $a_z$, though the Galactic model uncertainties will ultimately limit the distance measurement to a few percent accuracy. The timing parallax for this system is still not significantly measured, but is constrained to be less than 1.5\\,mas, in good agreement with the GR-derived result.  Our new distance result is consistent with the $1.1\\pm0.2$\\,kpc determined in Paper~I, but we have improved the uncertainty by a factor of 4.  Our previous distance led to a downward revision of the estimated double-neutron-star inspiral rate visible to gravitational-wave observatories such as LIGO.  The new measurement leads to a small (15\\%) increase in the number density of similar systems in the local universe, with the uncertainty now dominated by the uncertain scale height of such binaries \\citep[e.g.,][]{knst01}.  \\subsection{Dispersion Measurements and the Local Interstellar Medium}\\label{sec:dmvar}  As noted in \\S\\ref{sec:timing}, we find a significantly different DM and DM derivative in the post-upgrade era than before the upgrade (Figure~\\ref{fig:dmvar}).  We argue that while the change in reference DM value is largely due to the use of new standard profiles with a slightly different frequency alignment, the difference in the rate of change (more than a factor of 4 slower since the upgrade) is physical. It is true that the profile at both frequencies is undergoing a secular evolution due to geodetic precession of the pulsar's spin axis \\citep[][Stairs {\\it et al.}, in prep.]{arz95} and thus one might suspect a contribution to the DM derivative from different temporal evolution at the two observing frequencies.  We explore this possibility by calculating the DM difference between our 1999 May and 2001 June observing campaigns, for the best-fit post-upgrade DM derivative versus the best-fit {\\it pre-}upgrade DM derivative, finding that if the pre-upgrade DM trend had continued, the 2001 June DM would have been smaller by about 0.0005 cm$^{-3}$\\,pc.  This large a difference would imply a difference in offset between the 430\\,MHz and 1400\\,MHz profiles of 10\\,$\\mu$s, or 0.27 bins in a 1024-bin profile.  By comparing the actual cumulative profiles at these epochs using the same cross-correlation algorithm used to calculate TOAs, we estimate that the 430--1400\\,MHz offset has shifted by no more than about 0.05 bins.  As the TOAs are determined by the strong main peak of the pulse and the most noticeable evolution is taking place in the low-level emission near the base of the pulse \\citep{stta00} this is perhaps not surprising.  We therefore conclude that the observed changes in DM slope do represent changes in the structure of the interstellar medium between the Earth and the pulsar.  With our current data set, it is not possible to construct a satisfactory structure function with which to characterize the length scales of turbulence in the interstellar medium \\citep[e.g.,][]{pw91}. Because of strong refractive scintillation at 1400\\,MHz, the high-frequency data set remains quite sparse (Figure~\\ref{fig:dayres}). Even though we may use a wider-bandwidth instrument such as the recently commissioned 100-MHz ``WAPP'' spectrometer for future 1400\\,MHz observations, the broadband nature of the refractive interstellar scintillation may make it difficult to accurately sample the dispersion measure variations with resolution finer than roughly 6 months.  \\subsection{The Neutron Star Masses: Testing Evolutionary Theory}  The best mass estimates for the two neutron stars in the PSR~B1534+12 system come from the DDGR timing model.  The measured values are $m_1 = 1.3332\\pm0.0010\\,M_{\\odot}$ and $m_2 = 1.3452\\pm0.0010\\,M_{\\odot}$. The masses derived from the DD model and equations~\\ref{eq:r} and \\ref{eq:s} are in good agreement with these values, though the measurement uncertainties are much larger (Figure~\\ref{fig:chisq}).  Although the masses of the two neutron stars are very similar, it is now clear that the pulsar is significantly less massive than its companion.  This is contrary to initial expectations from binary evolution.  The pulsar is rapidly spinning, and has the low magnetic field characteristic of a ``recycled pulsar'' that has been spun-up by mass transfer from a companion star.  Assuming a monotonic relation between progenitor mass and neutron star mass, we therefore might predict that the pulsar would be more massive at birth, and that the mass difference would only increase during the later evolution of the system.  This perhaps naive prediction has already been challenged by the binary pulsar system PSR~B2303+46. Recently, the companion in this close, eccentric binary was shown to be a white dwarf rather than a second neutron star \\citep{vk99}.  The eccentric orbit implies that the white dwarf formed first, from the star that was originally more massive.  Presumably, mass transfer to the initially less massive star pushed it above the minimum mass needed to form a supernova and a neutron star.  A similar process may have been important in the formation of the PSR~B1534+12 system.  Although the pulsar was formed from the initially more massive star, mass transfer to its companion star probably resulted in a mass inversion before either neutron star was formed.  To move beyond the simplest (or most naive) predictions about the relative masses of the pulsar and companion in the B1534+12 system requires significant improvements in our theoretical understanding of several areas of stellar and binary evolution.  First, we must understand the relationship between progenitor and neutron star masses, which may not be purely monotonic.  Then we must understand how the mass lost by the pulsar's progenitor or gained by the companion's progenitor affected the evolution of the stellar cores and the amount of fallback during the supernova events.  Finally, we must understand in some detail the mass transfer that spun up the pulsar. Although these are all challenging problems, there is some hope for future guidance from the growing number of precisely measured neutron star masses.  \\subsection{Test of Relativity}  PSR~B1534+12 permits the second test of general relativity based on the $\\dot\\omega$, $\\gamma$, and $\\dot P_b$ parameters of a binary pulsar system, and adds significant measurements of the Shapiro-delay parameters $r$ and $s$.  The left-hand sides of equations~(\\ref{eq:omdot}--\\ref{eq:s}) represent quantities measured in a theory-independent fashion and listed in the DD column of Table~\\ref{tab:orbparms}.  If GR is consistent with these measurements and there are no significant unmodeled effects, the five curves corresponding to equations~(\\ref{eq:omdot}--\\ref{eq:s}) should intersect at a single point in the $m_1$-$m_2$ plane.  These curves are presented in Figure~\\ref{fig:massmass}, in which a pair of lines delimit the 68\\% confidence limit for each PK parameter (a single line is drawn for $\\dot\\omega$, whose uncertainty is too small to show). It is clear that the $\\dot\\omega$, $\\gamma$, $r$, and $s$ curves intersect, though $r$ is still poorly measured.  The curve obtained from the observed DD value of $\\dot P_b$ can be made to intersect the others, as discussed above, by setting the pulsar distance to $1.02\\pm0.05$\\,kpc rather than the $0.7\\pm0.2$\\,kpc estimated from the dispersion measure.  A filled circle at $m_1=1.3332~M_\\odot, m_2=1.3452~M_\\odot$ marks the DDGR solution of Table~\\ref{tab:orbparms}, and its location on the $\\dot\\omega$ line agrees to within 0.05\\% with the measured DD values of $\\gamma$ and $s$.  This provides a highly precise test of the validity of general relativity using only non-radiative timing parameters, an important complement to the mixed $\\dot\\omega$--$\\gamma$--$\\dot P_b$ test provided by PSR~B1913+16."
    },
    "0208/astro-ph0208482_arXiv.txt": {
        "abstract": "New HI observations are presented for a complete sample of 109 low luminosity star-forming galaxies taken from the KPNO International Spectroscopic Survey (KISS), the first CCD-based wide-field objective-prism survey for emission-line galaxies.  This sample consists of all star-forming galaxies with $M_B > -18.0$ and $cz <$ 11,000 km s$^{-1}$ from the first H$\\alpha$-selected survey list.  The galaxies in this list lie within a 1.3 deg wide strip centered on $\\delta($B$1950) = 29\\arcdeg30\\arcmin$ that spans the range $\\alpha($B$1950)=12^{\\mbox{\\scriptsize h}}15^{\\mbox{\\scriptsize m}}$ to $\\alpha($B$1950)=17^{\\mbox{\\scriptsize h}}0^{\\mbox{\\scriptsize m}}$.  Overall, 97 out of 109 galaxies have been detected in HI.  We confirm the weak trend of increasing gas richness with decreasing luminosity found by previous authors.  Gas richness is also shown to be weakly anti-correlated with metallicity.  The dependence of star formation rates (SFRs) and HI gas depletion timescales on metallicity is examined.  The median solar metallicity based SFR and gas depletion timescale are 0.1639 M$_{\\odot}$ yr$^{-1}$ and 5 Gyrs, respectively.  Corrections for variations in metallicity decreases SFRs by $\\sim0.5$ dex and increases gas depletion timescales by an average of $\\sim$8 Gyrs.  The majority of galaxies in this sample still have large reservoirs of HI gas, and despite their large current star formation rates, could have formed stars in a quasi-continuous manner for a Hubble time.  Finally, we present the first HI mass function for low luminosity star-forming galaxies and show that this subpopulation contributes 10-15\\% of the overall HI density in the local universe.  We conclude that if the HI mass function of the Universe does indeed have a steeply rising low-mass slope as suggested by previous authors, it is not due to the population of low luminosity star-forming galaxies.  Comparison of the number densities from the HIMFs in the range $10^8<M_{HI}/M_{\\odot}<10^9$ implies that  25-50\\% of galaxies in this regime are currently undergoing a strong episode of star formation. ",
        "introduction": "In an attempt to develop a unified understanding of the different morphological classes of dwarf galaxies, efforts have been made to relate the various observed types through some evolutionary scheme.  A number of scenarios have been proposed (see Skillman \\& Bender 1995, Ferrara \\& Tolstoy 2000 for excellent reviews).  One possibility is that the gas-rich dwarf irregulars (dIs) go through an extremely strong burst of star formation, lose the majority of their gas after the starburst event, and eventually fade and evolve into the gas-poor dwarf ellipticals (dEs).  During the bursting phase, the dIs would resemble massive HII regions.  Another possibility is that the dIs and dEs have common progenitors but have evolved differently because of environmental effects.  Although these schemes are highly debated, they form a good context for investigating the comparative structural properties of the different classes of dwarfs.  Several questions arise when considering these evolutionary scenarios.  Can {\\it any} dI initiate a large star-formation episode and appear as a BCD?  In other words, is the BCD phenomenon a stage of galaxy evolution common to all gas-rich dwarfs or only to a select sub-sample with special characteristics?  Further, do the progenitor galaxies go through repeated intense bursts of activity, or is the bursting phase unique in the object's lifetime?  If the bursts repeat, then what fraction of the time is the galaxy bursting (i.e., what is the duty cycle)?  Answers to these questions can also help us to determine if starbursting dwarfs can be used as a fair tracer of the more general population of low-mass galaxies.  Since starbursting dwarfs are far more easily detected than their quiescent counterparts, this issue is of great importance to investigations of the faint-end/low-mass tails of the luminosity/HI mass functions, and to studies of large-scale structure, particularly since low-mass galaxies may be less biased and better tracers of the overall mass distribution (e.g. Santiago \\& da Costa 1990, Loveday et al. 1995, Lee et al. 1999).  Within this context, it is interesting to study the HI properties of low luminosity star-forming galaxies, since the quantity of HI indicates the amount of fuel available for driving a starburst.  Additional constraints may be placed on the star-formation history  and chemical evolution of bursting dwarfs when HI data are used in conjunction with the star-formation rates and abundances inferred from optical spectroscopy.  Also, dynamical masses of these systems may be estimated from the velocity widths of HI profiles.  With this combination of optical and radio information, it is possible to assess the plausibility of gas-rich dIs as progenitors of bursting dwarfs, and to attempt to answer the questions posed above.   One can further study the differences in the distribution of HI in various morphological types by obtaining HI interferometric maps.  This is generally only done after initial single-dish HI observations have been made and viable candidates for interferometric mapping have been identified.  Hence, observations of the neutral gas content of bursting dwarfs is an important step toward a more complete understanding of this class of galaxy.  Previous 21-cm observations of star-forming dwarf galaxies (Thuan \\& Martin 1981; Gordon \\& Gottesman 1981; Staveley-Smith, Davies \\& Kinman 1992; Thuan et al. 1999; Smoker et al. 2000; Salzer et al. 2002) have been fruitful in establishing the properties of their neutral gas.  However, the samples used in these previous studies are comprised of objects from a number of different catalogs.  For example, Thuan and Martin (1981) used a combination of dwarfs from the Markarian, Haro and Zwicky catalogs, while Salzer et al. (2002) selected dwarfs from the Michigan, Wasilewski and Case surveys, in addition to the catalogs used by Thuan and Martin.  HI observations are needed for a complete, uniformly selected sample of dwarf galaxies since the studies which are based on such heterogeneous collections of objects may be biased in unrecognized ways.  To remedy this situation, we have obtained HI observations for a new, ultra-deep, complete sample of actively star-forming, low luminosity galaxies using the Arecibo Observatory 305-m radio telescope.  Our sample consists of low luminosity (M$_{B} \\geq -18.0$), nearby ($cz \\leq $11,000 km s$^{-1}$) , H$\\alpha$ selected galaxies from the first survey strip of the KPNO International Spectroscopic Survey (KISS) (Salzer et al. 2000, hereafter Paper I; Salzer et al. 2001, hereafter KISS Red 1 [KR1]), a new objective prism survey for extragalactic emission-line objects.  In this paper, we present the 21-cm HI line observations of the overall sample.  The remaining sections are organized as follows.  An overview of the sample and observational procedures are given in $\\S$ 2.  The HI data are presented $\\S$ 3, and the HI properties are described in $\\S$ 4.  Finally, our conclusions are summarized in $\\S$ 5.  Suggestions for further work involving the dwarf candidates in this sample are given throughout the paper.  All distance dependent quantities assume H$_{\\mbox{\\scriptsize o}} = $75 km s$^{-1}$ Mpc$^{-1}$. ",
        "conclusions": "A complete sample of 109 low luminosity ($M_B>-18.0$), nearby ($cz<11,000$ km s$^{-1}$), H$\\alpha$-selected star-forming galaxies from the KISS catalog have been observed at 21-cm.  Our detection rate is 89\\% (97/109).  By examining the KISS composite $B$ and $V$-band survey images, we have found that 9\\% (10/109) have companions of comparable or greater optical brightness within the $\\sim3\\farcm5$ Arecibo beam.  We find that our non-detections and upper-limits for confused sources do not bias the sample in terms of $L_B$ or $M_{HI}$.  The HI properties of this sample are as follows:  \\begin{enumerate} \\item Our sample includes true dwarf galaxies as well as larger but heavily-extincted edge-on spiral galaxies.  This is reflected in our broad distribution of HI line width (39 km s$^{-1}$ to 311 km s$^{-1}$, median=132 km s$^{-1}$, mean=139 km s$^{-1}$, uncorrected for inclination). \\item The range of HI gas richness for this sample (as defined by $M_{HI}/L_B$) is the same as in previous HI surveys of late-type galaxies.  We report weak anti-correlations between the gas-richness and metallicity, and the gas-richness and blue luminosity.  This is consistent with previous results for different samples of dwarf galaxies. \\item Using the models of Ferrara \\& Tolstoy (2000), our galaxies are shown to have a large range of dark-to-visible mass fractions (0 $<M_{dark}/M_{stars+gas}< $300). \\item The median HI gas depletion timescale for this sample increases from $\\sim$5 Gyr to $\\sim$13 Gyr when a metallicity-dependent SFR is used to calculate timescales instead of a solar-metallicity SFR.  Accounting for metallicity is important since the galaxies in this sample have low luminosities, and tend to be metal-poor.  One interpretation of these statistics is that these galaxies will not deplete their gas supplies for another Hubble time, if they continue to form stars at their current rate.  These results also lift the requirement that BCDs have star formation histories dominated by short bursts of activity, and show that a more nearly continuous mode of star formation is possible.  \\item By computing an HI mass function for this sample, it is shown that the low luminosity star-forming galaxies in KISS contain 10-15\\% of the overall neutral hydrogen in the universe.  We find that $\\rho_{HI}(\\mbox{KISS}\\; M_B >-18.0)=7.0\\times 10^6 \\; M_{\\odot}$ Mpc$^{-3}$, or $\\Omega_{HI} = 4.5 \\times 10^{-5}$ with a $\\sim20\\%$ statistical error.  The HIMF of this sub-population does not exhibit a steeply rising slope at low-masses.  This is consistent with the result that gas richness ($M_{HI}/L_B$) does not increase at a significant level with decreasing galaxy luminosity.  In the range $10^8<M_{HI}/M_{\\odot}<10^9$, we find that 25-50\\% of all galaxies are currently undergoing a strong episode of star formation.  \\end{enumerate}"
    },
    "0208/astro-ph0208577_arXiv.txt": {
        "abstract": "A 3D simulation of the upper part of the solar convective zone is used to derive constraints about the averaged and dynamic properties of solar turbulent convection. Theses  constraints are then used to compute the acoustic energy supply rate $P$ injected into the solar radial oscillations  according to the theoretical  expression in  Samadi \\& Goupil (2001). The result is compared with solar seismic data. \\\\Assuming, as it is usually  done,    a gaussian model for the frequency component $\\chi_k(\\nu)$ of the model of turbulence, it is  found that the computed  $P(\\nu)$ is underestimated compared with the solar seismic data by a factor $\\sim 2.5$. \\\\A frequency analysis of the solar simulation shows that the gaussian model indeed does not correctly model  $\\chi_k(\\nu)$ in the frequency range where the acoustic energy injected into the solar $p$-modes is important  ($\\nu \\simeq 2 - 4$~mHz). One must consider an additional non-gaussian component for $\\chi_k(\\nu)$ to reproduce its behavior. Computed values  of $P$ obtained with this non-gaussian component reproduce better the solar seismic observations. This non-gaussian component leads to a Reynolds stress contribution of the same order than the one arising from  the advection of the turbulent fluctuations of entropy by the turbulent motions. ",
        "introduction": "Turbulent motions in the upper convective zone of solar like stars,  generate acoustic energy, which in turn is injected at the rate $P(\\nu)$  into $p$-modes oscillations ($\\nu$ is the frequency of a given mode). Solar-type oscillations are therefore meant as stochastically excited by turbulent convection. This type of excitation process concerns low massive stars with an outer convective zone.  Measurement of the  rate of acoustic energy, $P(\\nu)$, injected into solar-like oscillations is one of the goals of future space seismic missions COROT (Baglin et al 1998) and Eddington (Favata et al 2000). These seismic data will  then make it possible to constrain the theory of excitation and damping of solar-type oscillations, which in turn will provide valuable information about the properties of stellar convection zones.  Theoretical formulations for $P(\\nu)$ offer the advantage of testing {\\it separately} several properties entering in excitation mechanism. Here we consider the formulation by Samadi \\& Goupil (2001, see section~\\ref{formulation} and also Samadi (2001)  for a detailed summary).  Computation of  $P(\\nu)$  requires an accurate knowledge of the dynamic properties of turbulence in stars. Unfortunately, the current observations of the solar granulation  cannot provide a precise enough description of the turbulent spectrum properties.  In the present work we use a 3D simulation of the upper part of the solar convective zone to determine the dynamic properties of solar turbulence (see section~\\ref{the3Dconstraints}) which are necessary to compute $P(\\nu)$. Our calculations are then compared with  the helioseismic  constraints from the GOLF/SOHO instrument (see section~\\ref{consequences}). ",
        "conclusions": "Our investigation demonstrates the non-gaussian character of the stochastic excitation of solar $p$-modes. Indeed,  the gaussian function (GF) as a model for $\\chi_k(\\nu)$ leads to an important over-estimation of $P(\\nu)$ whereas  the Gaussian Exponential Function (GEF), which decreases more slowly with  $\\nu$ than the GF does,  results in a better agreement between  $P(\\nu)$ and the seismic solar observations. The maximum value of $P(\\nu)$ is now  reproduced without any adjustement of free parameters  in contrast with previous approaches.  The non-gaussian character of stochastic excitation causes the Reynolds stress contribution to be of the same order as the contribution arising from  the advection of the turbulent fluctuations of entropy by the turbulent motions (not shown here, see Samadi et al 2002b). This last result is  in better agreement with recent results by Stein \\& Norlund (2001)  and contrasts with  previous results by Samadi et al (2001) based on the gaussian assumption for $\\chi_k(\\nu)$ and  also with  the estimation carried out by Goldreich et al (1994)."
    },
    "0208/hep-th0208140_arXiv.txt": {
        "abstract": "Vacuum structure and global cosmic strings are analyzed in the effective theory of  self-interacting O(2) scalar fields on (3+1)-manifolds with conical singularities. In the context of one-loop effective action computed by heat-kernel methods with $\\zeta$-function regularization, we find an inhomogeneous vacuum of minimum energy and suggest some reason why low-energy global strings are likely to be generated at the conical singularities. ",
        "introduction": "Cosmic strings remain somewhere in our universe as relics of cosmological phase transition at early times. These linear topological defects have been extensively studied in connection with cosmological problems~\\cite{Vil}. Global strings were proposed as a candidate of producing primordial density perturbations leading to galaxy formation~\\cite{VE}. However, the original purpose seems unlikely to be achieved due to radiation of Goldstone bosons around oscillating closed loops of strings~\\cite{Dav} and other stringent astrophysical bounds~\\cite{LH}. Recently other physical applications are also investigated, {\\it e.g.}, time evolution of global string network~\\cite{YY} and possible generation of several types of global strings in high density QCD~\\cite{FZ}.  When infinite straight U(1) global strings are coupled to Einstein gravity, static solutions with cylindrical symmetry encounter unavoidable singularity~\\cite{Gre}. With a negative cosmological constant, global vortices become seeds for charged BTZ black holes in (2+1) dimensions without physical singularity~\\cite{KKK}. Such spacetime structure with horizons is extended to a viable model of 6-dimensional Randall-Sundrum type brane worlds, which is composed of our world on the 3-brane and 2 extra-dimensions identified with spatial plane of global vortices~\\cite{CK}.  The questions we are interested in in this paper are the vacuum structure and the global strings in the effective theory with conical singularities. Suppose that an infinite straight local cosmic string was generated throughout a cosmological phase transition in the very early universe, {\\it e.g.}, a Nielsen-Olesen vortex string at grand unification (GUT) scale. If there has been no other phase transition up to the late-time transition of a very low energy, {\\it e.g.}, electroweak (EW) scale or lower scale, the supermassive local cosmic string may be observed through some gravitational effects, i.e., background spacetime has a deficit angle. Even though a phase transition occurs at the low energy, released latent heat or other fluctuations may be not enough to {\\it melt} this high-energy cosmic string. So the survived string may affect the new vacuum structure of spontaneous symmetry breaking. In section 2, we assume an O(2) scalar theory with conical singularity in the background spacetime for tractability, and compute one-loop effective action including a $\\delta$-function singularity. In section 3, we analyze vacuum of the effective theory and show that an inhomogeneous vacuum, jumping from symmetric local maximum at the points of conical singularities to broken global minimum at the other space, is minimum energy configuration instead of homogeneous broken vacuum. In section 4, we address our main issue that the production of a global string along the supermassive local cosmic string seems favorable and natural.  The obtained low-energy global string containing the supermassive local cosmic string core looks like an infinite straight candle with a heavy wick. This kind of possibilities was firstly proposed in the case of superconducting strings, {\\it e.g.}, the O(10) string with an inner core of $\\tilde{\\rm U} (1)$ magnetic flux~\\cite{Wit}. It may also be understood as a simplified field-theoretic calculation of daily life experience that the structure of extend objects can easily formed at the site of a defect or dust when a vapor (or a liquid) is cooled down to the liquid (or the solid). ",
        "conclusions": ""
    }
}