Datasets:
ReadingTimeMachine
/
rtm-sgt-ocr-v1

Dataset card Data Studio Files Community
1
source
stringlengths
1
2.05k
⌀
target
stringlengths
1
11.7k
The orbital. epieyclie and vertical frequencies around a Ixerr black hole satisfy (Ixato 1990)where 1«eLl is the dimensionless angular momentuni parameter of the hole (α«O implving a retrograde disc) and r is the radius in units of GAL/c75.
The orbital, epicyclic and vertical frequencies around a Kerr black hole satisfy (Kato 1990)where $-1<a<1$ is the dimensionless angular momentum parameter of the hole $a<0$ implying a retrograde disc) and $r$ is the radius in units of $GM/c^2$.
For a progracde clisc (e>0) we have the ordering &«OQ.<© at any racius. corresponding to à situation as illustrated in Fig.
For a prograde disc $a>0)$ we have the ordering $\kappa<\Omega_z<\Omega$ at any radius, corresponding to a situation as illustrated in Fig.
1.
1.
Fig.
Fig.
2 shows the radial wavelength. A=2r/k. associated with the steady wavelike shape of a prograde warped disc around a err black hole.
2 shows the radial wavelength, $\lambda=2\pi/k$, associated with the steady wavelike shape of a prograde warped disc around a Kerr black hole.
To a first approximation. i.e. considering onlv the first terms in equations (13)). (15)). we have The wavelength. is shorter for more rapidly rotating black holes. and decreases as the mareinally stable circular orbit is approached. but is always at least a Lew times Lf.
To a first approximation, i.e. considering only the first terms in equations \ref{kerr1}) \ref{kerr3}) ), we have The wavelength is shorter for more rapidly rotating black holes, and decreases as the marginally stable circular orbit is approached, but is always at least a few times $H$.
This suggests that these wavelike solutions are. physically realizable anc may be described: with some confidence by theories of long-wavelength bending waves.
This suggests that these wavelike solutions are physically realizable and may be described with some confidence by theories of long-wavelength bending waves.
A more accurate description. of the steady wavelike shape can be obtained by setting the time-derivatives to zero in equations (5)) and (6)).
A more accurate description of the steady wavelike shape can be obtained by setting the time-derivatives to zero in equations \ref{dwdt}) ) and \ref{dgdt}) ).
This results in a second-order ordinary dilflerential equation for the tilt variable VV (2). Ifwe take the first-orcer approximations from equations (13)) (14)). we have
This results in a second-order ordinary differential equation for the tilt variable $W(R)$ , If we take the first-order approximations from equations \ref{kerr1}) \ref{kerr2}) ), we have
scale is detected as a peak in the A-variance spectra. the peak position at about 0.07 is somewhat lower than the scale of the maximum variation 1/(KV2)=0.088.
scale is detected as a peak in the $\Delta$ -variance spectra, the peak position at about 0.07 is somewhat lower than the scale of the maximum variation $1/(k \sqrt{2})=0.088$.
We found this small shift by about in all spectra.
We found this small shift by about in all spectra.
When using the effective filter length. the peak position is very constant independent of the filter type and its diameter ratio.
When using the effective filter length, the peak position is very constant independent of the filter type and its diameter ratio.
The peak remains at the same position but the ripples in the A-variance spectrum move corresponding to the changes in the side lobes of the Bessel function.
The peak remains at the same position but the ripples in the $\Delta$ -variance spectrum move corresponding to the changes in the side lobes of the Bessel function.
Decreasing v relative to the value of 3.0 results in à steeper decay above the peak and an increased number of ripples at large lags.
Decreasing $v$ relative to the value of 3.0 results in a steeper decay above the peak and an increased number of ripples at large lags.
When reducing the diameter ratio v for the Mexican-hat filter the decay above the peak also steepens and we obtain some flattening at the largest lags in the map.
When reducing the diameter ratio $v$ for the Mexican-hat filter the decay above the peak also steepens and we obtain some flattening at the largest lags in the map.
In general we can either achieve the sharp peak and the artificial structures at large lags by the French-hat filter or the broader peak without ripples by means of the Mexican hat.
In general we can either achieve the sharp peak and the artificial structures at large lags by the French-hat filter or the broader peak without ripples by means of the Mexican hat.
The equivalent figure shows the same shape of the peak but more pronounced ripples with deeper minima on large scales for the French hat filter d a somewhat steeper decay except for a small discretisation error from a single pixel row which ts treated different when mirroring or continuing periodically.
The equivalent figure shows the same shape of the peak but more pronounced ripples with deeper minima on large scales for the French hat filter and a somewhat steeper decay except for a small discretisation error from a single pixel row which is treated different when mirroring or continuing periodically.
In general the edge truncation always produces a strong and the mirror- a weak smearing of the French-hat A-variance ripples at large lags relative to the periodic continuation.
In general the edge truncation always produces a strong and the mirror-continuation a weak smearing of the French-hat $\Delta$ -variance ripples at large lags relative to the periodic continuation.
Corresponding results for the chess board map show almost the same curves as the sine wave field at all lags above the peak in the A-variance spectra.
Corresponding results for the chess board map show almost the same curves as the sine wave field at all lags above the peak in the $\Delta$ -variance spectra.
But they show a flatter rising part at smaller lags.
But they show a flatter rising part at smaller lags.
The additional high-frequency contributions from the edges of the chess fields increase the A-variance at small lagsthere.
The additional high-frequency contributions from the edges of the chess fields increase the $\Delta$ -variance at small lags.
The situation is somewhat different for the circle map.
The situation is somewhat different for the circle map.
The strong contributions from the "empty" area at large lags visible in Fig.
The strong contributions from the “empty” area at large lags visible in Fig.
5 suppress differences from the different filter shapes in the spectrum.
\ref{fig_circleexample} suppress differences from the different filter shapes in the spectrum.
The spectra obtained from both filter types look very similar.
The spectra obtained from both filter types look very similar.
To quantify the influence of the selection. of the filter shape and the diameter ratio on the detection of the dominant scale length we measure the sharpness of the A-variance peaks shown in Fig. 7..
To quantify the influence of the selection of the filter shape and the diameter ratio on the detection of the dominant scale length we measure the sharpness of the $\Delta$ -variance peaks shown in Fig. \ref{fig_sinedeltas}.
We use the The behaviour of both parameters is shown in Fig.
We use the The behaviour of both parameters is shown in Fig.
8 as a function of the filter diameter ratio v.
\ref{fig_sinecontrast} as a function of the filter diameter ratio $v$.
The upper plot shows the logarithmie. width of the peak: the lower plot the contrast of the peak relative to values at much larger and much smaller lags.
The upper plot shows the logarithmic width of the peak; the lower plot the contrast of the peak relative to values at much larger and much smaller lags.
From the figure it is obvious that we cannot achieve a minimum width and maximum contrasts simultaneously with the same filter. so that some balance has to be found.
From the figure it is obvious that we cannot achieve a minimum width and maximum contrasts simultaneously with the same filter, so that some balance has to be found.
The contrast with respect to smaller lags hardly varies with filter type and diameter ratio but the contrast relative to large lags is drastically changed.
The contrast with respect to smaller lags hardly varies with filter type and diameter ratio but the contrast relative to large lags is drastically changed.
The French-hat filter always produces a narrower peak but the decrease of the peak width towards lower v ratios is accompanied by a deterioration of the contrast with respect to large lags.
The French-hat filter always produces a narrower peak but the decrease of the peak width towards lower $v$ ratios is accompanied by a deterioration of the contrast with respect to large lags.
The Mexican-hat filter shows a continuous improvement of both quantities towards lower ratios but gives a somewhat broader peak.
The Mexican-hat filter shows a continuous improvement of both quantities towards lower ratios but gives a somewhat broader peak.
The slight increase of the contrast with respect to the lower end of the spectrum towards lower v values for both filter types indicates
The slight increase of the contrast with respect to the lower end of the spectrum towards lower $v$ values for both filter types indicates
The onset of massive star formation episodes in galaxies drives their observational properties in almost any wavelength range.
The onset of massive star formation episodes in galaxies drives their observational properties in almost any wavelength range.
The UV and optical become dominated by the continuum of massive. hot and young stars. as well as by the presence of nebular emission lines.
The UV and optical become dominated by the continuum of massive, hot and young stars, as well as by the presence of nebular emission lines.
After a few Myr of evolution. red supergiant stars contribute to most of the near infrared emission.
After a few Myr of evolution, red supergiant stars contribute to most of the near infrared emission.
The heating of interstellar dust particles by the powerful UV photons induces the thermal re-emission of large amounts of energy in the domain.
The heating of interstellar dust particles by the powerful UV photons induces the thermal re-emission of large amounts of energy in the domain.
The injection of tonizing photons into the surrounding gas generates thermal radio emission. which is replaced by non-thermal emission as the ionizing power of the burst declines and the more massive stars begin to explode as supernovae.
The injection of ionizing photons into the surrounding gas generates thermal radio emission, which is replaced by non-thermal emission as the ionizing power of the burst declines and the more massive stars begin to explode as supernovae.
The direct relation between the strength of the star formation episode and the intensity of the different observable parameters has allowed a number of star formation rate calibrators to be defined. such as UV continuum. emission lines intensity. far infrared or radio luminosities (Kennicutt1998;Rosa-Gonzálezetal.2003:Bell 2003).
The direct relation between the strength of the star formation episode and the intensity of the different observable parameters has allowed a number of star formation rate calibrators to be defined, such as UV continuum, emission lines intensity, far infrared or radio luminosities \citep{Kennicutt98,Rosa02,Bell2003}.
These calibrators have proven to be invaluable for statistical studies of the star formation history of the Universe.
These calibrators have proven to be invaluable for statistical studies of the star formation history of the Universe.
Star-forming regions are also the source of conspicuous X- emission. generated by individual stars. by the injection of laree amounts of mechanical energy heating the interstellar medium. by supernova remnants. and by binary systems transferring mass to a compact primary (Cervifio.Mas-Hesse.&Kunth2002:PersicRephaeli 2002).
Star-forming regions are also the source of conspicuous X-ray emission, generated by individual stars, by the injection of large amounts of mechanical energy heating the interstellar medium, by supernova remnants, and by binary systems transferring mass to a compact primary \citep{Cervino02,Persic02}.
. All of these individual components are in principle directly linked to the strength. of the star formation episode. so that the X-ray luminosity could also be used as an estimator of star formation rates (SFR).
All of these individual components are in principle directly linked to the strength of the star formation episode, so that the X-ray luminosity could also be used as an estimator of star formation rates ).
Several authors have in recent. years discussed. the feasibility of using the X-ray luminosity as an eestimator.
Several authors have in recent years discussed the feasibility of using the X-ray luminosity as an estimator.
Ranalli.Comastri.&Setti(2003) proposed an empirical calibration of the soft (0.5-2.0 keV) and hard (2-10 keV) X-ray luminosities. based on their correlation with the far infrared (FIR) and radio luminosities. and using the known calibrations of these parameters as proxies.
\citet{Ranalli03} proposed an empirical calibration of the soft $0.5$ $2.0$ keV) and hard $2$ $10$ keV) X-ray luminosities, based on their correlation with the far infrared (FIR) and radio luminosities, and using the known calibrations of these parameters as proxies.
Grimmetal.(2003) studied the correlation between the number of high- X-ray binaries (HMXB) and theSFR.. deriving different calibrations of the hard X-ray luminosity for low and high star formation rates.
\citet{Grimm03} studied the correlation between the number of high-mass X-ray binaries (HMXB) and the, deriving different calibrations of the hard X-ray luminosity for low and high star formation rates.
Persieetal.(2004) obtained a calibration of the hard X-ray luminosity as a eestimator by assuming that most of the emission in this range is associated with HMXB. and scaling from the number of HMXB to the oof our Galaxy.
\citet{Persic04} obtained a calibration of the hard X-ray luminosity as a estimator by assuming that most of the emission in this range is associated with HMXB, and scaling from the number of HMXB to the of our Galaxy.
Gilfanovetal.(2004)— confirmed the of the hard X-ray luminosity associated with HMXB. using slightly different slopes at high and low star formation rates.
\citet{Gilfanov2004} confirmed the of the hard X-ray luminosity associated with HMXB, using slightly different slopes at high and low star formation rates.
In a recent paper. Persic&Rephaeli(2007) found indeed that the collective hard X-ray emission of young point sources correlates linearly with the star formation rate derived from the far infrared luminosity.
In a recent paper, \citet{Persic07} found indeed that the collective hard X-ray emission of young point sources correlates linearly with the star formation rate derived from the far infrared luminosity.
Stricklandetal.(2004a) demonstrated that the luminosity of diffuse X- emission in star-forming galaxies is directly proportional to the rate of mechanical energy injection from the young. massive stars into the interstellar medium of the host galaxies.
\citet{Strickland04b} demonstrated that the luminosity of diffuse X-ray emission in star-forming galaxies is directly proportional to the rate of mechanical energy injection from the young, massive stars into the interstellar medium of the host galaxies.
A similar result was found by Grimesetal.(2005) from the analysis of a sample of starburst galaxies of different types (from dwarf starbursts to ultraluminous infrared galaxies). which concluded that the mechanism producing the diffuse X-ray emission in the different types of starbursts was powered by the mechanical energy injected by stellar winds and supernovae into the surrounding medium.
A similar result was found by \citet{Grimes05} from the analysis of a sample of starburst galaxies of different types (from dwarf starbursts to ultraluminous infrared galaxies), which concluded that the mechanism producing the diffuse X-ray emission in the different types of starbursts was powered by the mechanical energy injected by stellar winds and supernovae into the surrounding medium.
Recently Rosa-Gonzálezetal. confirmed the reliability of the soft X-ray luminosity as an eestimator from the analysis of a sample of star-forming galaxies in the at redshifts
Recently \citet{Rosa07} confirmed the reliability of the soft X-ray luminosity as an estimator from the analysis of a sample of star-forming galaxies in the at redshifts
an analytical Milky-Way-like potential consisting of a disk-. a bulge-. and a halo component.
an analytical Milky-Way-like potential consisting of a disk-, a bulge-, and a halo component.
The coordinate system is chosen such that the disk of the host galaxy Hes in the xy-plane.
The coordinate system is chosen such that the disk of the host galaxy lies in the xy-plane.
The disk 1s modeled by a Miyamoto-Nagai potential (Miyamoto&Nagat 1975). with My=1.0xI0!! Me. a=6.5 kpe. and by=0.26 kpe.
The disk is modeled by a Miyamoto-Nagai potential \citep{miya1975}, , with $M_{\rm d}=1.0 \times 10^{11}$ $_{\sun}$, $a_{\rm d}=6.5$ kpc, and $b_{\rm d}=0.26$ kpc.
The bulge is represented by aHernquist potential (Hernquist 1990). with M,=3.4x10! Mo. and a,=0.7 kpe.
The bulge is represented by aHernquist potential \citep{hern1990}, with $M_{\rm b}=3.4 \times 10^{10}$ $_{\sun}$, and $a_{\rm b}=0.7$ kpc.
The dark matter halo is a logarithmic potential. with vo=186.0 km s7!. and nis12.0 kpe.
The dark matter halo is a logarithmic potential, with $v_{\rm 0}=186.0$ km $^{-1}$, and $r_{\rm halo}=12.0$ kpc.
This set of parameters gives à reasonable Milky-Way-like rotation curve.
This set of parameters gives a reasonable Milky-Way-like rotation curve.
In this paper. we focus on ECs and UCDs located far from the galactic disk in the halo of the respective galaxies.
In this paper, we focus on ECs and UCDs located far from the galactic disk in the halo of the respective galaxies.
As orbital parameters for such objects are unknown. we chose a polar orbit between galactic radii 20 and 60 kpe for our simulations.
As orbital parameters for such objects are unknown, we chose a polar orbit between galactic radii 20 and 60 kpc for our simulations.
These values are motivated by the projected distances to the M31 ECs of 13 to 60 kpe (Mackeyetal.2006) and the projected distances of Fornax UCDs between 8 and 74 kpe (Mieskeetal.2008).
These values are motivated by the projected distances to the M31 ECs of 13 to 60 kpc \citep{mackey06} and the projected distances of Fornax UCDs between 8 and 74 kpc \citep{mieske08}.
. Figure 4+ illustrates the chosen orbit. which has an orbital period of about 860 Myr.
Figure \ref{fig_orbits} illustrates the chosen orbit, which has an orbital period of about 860 Myr.
In our formation scenario. the CCs are most likely formed at the peri-galactic passage of the parent galaxy where the impact of the interaction is strongest.
In our formation scenario, the CCs are most likely formed at the peri-galactic passage of the parent galaxy where the impact of the interaction is strongest.
Therefore. we start our caleulations at the peri-galactic distance and integrate all models up to 5 Gyr.
Therefore, we start our calculations at the peri-galactic distance and integrate all models up to 5 Gyr.
We stop the integrations at 5 Gyr to save computing time. as the structural parameters change only very slightly afterwards.
We stop the integrations at 5 Gyr to save computing time, as the structural parameters change only very slightly afterwards.
To analyze the impact of a polar orbit relative to an inclined orbit. we recalculated a subset of our models also on an inclined orbit (see Fig. 4...
To analyze the impact of a polar orbit relative to an inclined orbit, we recalculated a subset of our models also on an inclined orbit (see Fig. \ref{fig_orbits},
Orbit 2).
Orbit 2).
The orbit is expected to have its largest impact on the most extended and lowest mass CC models.
The orbit is expected to have its largest impact on the most extended and lowest mass CC models.
The adcitional computations are indicated by open circles in Fig. 3..
The additional computations are indicated by open circles in Fig. \ref{fig_grid}.
In addition. for the most extended and least massive model = 160 pe and M**=10? Me). where the tidal field has the (RClargest impact. and for the most extended and most massive model (Re= 160 pe and MS10° Mo) complementary calculations on a circular orbit at a galactocentric distance of 60 kpe were performed.
In addition, for the most extended and least massive model $R_{\rm pl}^{\rm CC} =$ 160 pc and $M^{\rm CC} = 10^{5.5}$ $_{\sun}$ ), where the tidal field has the largest impact, and for the most extended and most massive model $R_{\rm pl}^{\rm CC} =$ 160 pc and $M^{\rm CC} = 10^{8}$ $_{\sun}$ ) complementary calculations on a circular orbit at a galactocentric distance of 60 kpc were performed.
The numerical modeling was performed with the particle-mesh code ddeveloped by Metz(2008).
The numerical modeling was performed with the particle-mesh code developed by \cite{metz}.
. It is a new C++ implementation of the FORTRAN particle-mesh code citepfellOO using object oriented. programming techniques.
It is a new C++ implementation of the FORTRAN particle-mesh code \\citep{fell00} using object oriented programming techniques.
mmakes particular optimal use of modern multi-core processor technologies.
makes particular optimal use of modern multi-core processor technologies.
The code solves the Poisson equation on a system of Cartesian grids.
The code solves the Poisson equation on a system of Cartesian grids.
In order to get good resolution of the star clusters two erids with high and medium resolution are focused on each star cluster following their trajectories.
In order to get good resolution of the star clusters two grids with high and medium resolution are focused on each star cluster following their trajectories.
The individual high resolution grids have a size of +80 pe and cover an entire star cluster. whereas the medium resolution grid of every star cluster has a size between +800 pe and +1200 pe embedding the whole initial CC.
The individual high resolution grids have a size of $\pm$ 80 pc and cover an entire star cluster, whereas the medium resolution grid of every star cluster has a size between $\pm$ 800 pc and $\pm$ 1200 pc embedding the whole initial CC.
The local universe is covered by a fixed coarse grid with a size of £70 kpe. which contains the orbit of the CC around the center of the galaxy.
The local universe is covered by a fixed coarse grid with a size of $\pm$ 70 kpc, which contains the orbit of the CC around the center of the galaxy.
All grids contain 128° erid cells.
All grids contain $^{3}$ grid cells.
The galaxy is represented by an analytical potential (see Sect. 3.2)).
The galaxy is represented by an analytical potential (see Sect. \ref{potential_mw}) ).
For each particle in the CC the acceleration from the galactic potential is added as an analytical formula to the grid-based acceleration computed by solving the Poisson equation.
For each particle in the CC the acceleration from the galactic potential is added as an analytical formula to the grid-based acceleration computed by solving the Poisson equation.
The formationprocess of the merger object depends on the compactness of the initial. CC.
The formationprocess of the merger object depends on the compactness of the initial CC.
A measure of how densely a CC is filled with star clusters for an equal number NCC of star clusters is given by the parameter « (Fellhaueretal.2002).. where Ry and Rr are the Plummer radius of a single star cluster and the Plummer radius of the CC. respectively.
A measure of how densely a CC is filled with star clusters for an equal number $N_{\rm 0}^{\rm CC}$ of star clusters is given by the parameter $\alpha$ \citep{fell02a}, where $R_{\rm pl}^{\rm SC}$ and $R_{\rm pl}^{\rm CC}$ are the Plummer radius of a single star cluster and the Plummer radius of the CC, respectively.
In general high values of « accelerate the merging process because the star clusters already overlap in the center of the CC. whereas low values hamper the merging process.
In general high values of $\alpha$ accelerate the merging process because the star clusters already overlap in the center of the CC, whereas low values hamper the merging process.
Our models cover a-values of 0.4. 0.2. 0.1. 0.05 and 0.025.
Our models cover $\alpha$ -values of 0.4, 0.2, 0.1, 0.05 and 0.025.
High values of a (x 0.1) correspond to compact CCs with overlapping star clusters in the center (Fig.
High values of $\alpha$ $ \ge 0.1 $ ) correspond to compact CCs with overlapping star clusters in the center (Fig.
2aa and b) where the majority of star clusters merge within a few Myr (Fig. 5)).
\ref{fig_ini_configs}a a and b) where the majority of star clusters merge within a few Myr (Fig. \ref{fig_time_evol}) ).
Low values of a ἐς 0.05) correspond to extended CCs (Fig.
Low values of $\alpha$ $ \le 0.05 $ ) correspond to extended CCs (Fig.
2cc) where the merging process can take up to several hundred Myr.
\ref{fig_ini_configs}c c) where the merging process can take up to several hundred Myr.
The parameter a for the model matrix is shown in Fig.
The parameter $\alpha$ for the model matrix is shown in Fig.
3 increasing from extended to compact CCs.
\ref{fig_grid} increasing from extended to compact CCs.
Also the tidal field has to be taken into account as it counteracts the merging process.
Also the tidal field has to be taken into account as it counteracts the merging process.
An estimate of the influence of the tidal field on the CC ts given by the parameter (Fellhaueretal.2002).. which is the ratio of the cutoff radius R° of the CC and its tidal radius f°.
An estimate of the influence of the tidal field on the CC is given by the parameter \citep{fell02a}, which is the ratio of the cutoff radius $R_{\rm cut}^{\rm CC}$ of the CC and its tidal radius $r_{\rm t}^{\rm CC}$.
An order of magnitude estimate of the tidal radiusis given by King (1962).. where Μος is the mass of the CC. Meaty, 18 the galaxy mass within Ay. A, and A, are the peri- and apo-galactic distances. and e=(Ry—Rp)/(Ra+05 is the eccentricity of the orbit.
An order of magnitude estimate of the tidal radiusis given by \cite{king62}, , where $M_{\rm CC}$ is the mass of the CC, $M_{\rm gal,R_p}$ is the galaxy mass within $R_{\rm p}$ , $R_{\rm p}$ and $R_{\rm a}$ are the peri- and apo-galactic distances, and $e=(R_{\rm a}-R_{\rm p})/(R_{\rm a}+R_{\rm p})=0.5$ is the eccentricity of the orbit.
If the star cluster distribution Hes within the tidal radius of the CC (B« 1) the influence of the tidal field on the merging
If the star cluster distribution lies within the tidal radius of the CC $\beta < 1$ ) the influence of the tidal field on the merging
ihe time scale studied here.
the time scale studied here.
All other details of the code ean be found in Sion(1995). and references therein.
All other details of the code can be found in \citet{sio95} and references therein.
Numerical simulations are carried out bv. switching on accretion for the duration of a dwarf nova outburst interval and then shutting it off to follow the cooling of the white cwarf.
Numerical simulations are carried out by switching on accretion for the duration of a dwarf nova outburst interval and then shutting it off to follow the cooling of the white dwarf.
In this wav the effects of compressional heating and boundary laver irradiation can be assessed quantitatively,
In this way the effects of compressional heating and boundary layer irradiation can be assessed quantitatively.
The matter is assunied (o accrete πο with the same entropy as the white dwarl outer lavers.
The matter is assumed to accrete 'softly' with the same entropy as the white dwarf outer layers.
From theoretical considerations. one expects the accreting malter. as it (ransils through the boundary laver region. to increase ils temperature (since 1 boundary laver temperature is high: Z5,o>1;).
From theoretical considerations, one expects the accreting matter, as it transits through the boundary layer region, to increase its temperature (since the boundary layer temperature is high: $T_{BL} >> T_{*}$ ).