Filename
stringlengths 22
64
| Paragraph
stringlengths 8
5.57k
|
|---|---|
Processed_Self-similar_structure_of_resistive_ADAFs_with_out.txt | Figure 2. Similar to Fig. 1, but for s = 0.2 and different values of βφ (as labeled). |
Processed_Self-similar_structure_of_resistive_ADAFs_with_out.txt | The influence of βφ is displayed in Fig. 2. In this figure, the power law index s is 0.2 and βφ can vary 2 = from 0.01 (i.e., cφ 2), as labeled. Other input parameters are simi- 4cs lar to Fig 1. In Fig. 2, higher values of βφ are not considered because the ratio H/r exceeds unity (see top left panel of Fig. 2). The dependence of the mag- netic heating on the azimuthal field component implies that the stronger fields enhance the magnetic heating and the gas temperature. With increasing the temper- ature (or the sound speed) due to the toroidal compo- nent of the magnetic field, therefore, the disc becomes thicker. These results are in good agreement with previ- ous works (see Ghoreyshi & Shadmehri, 2020). We find that at Pm = 10, for example, the enhancement of βφ from 0.1 to 2.0 causes the ratio H/r to rise by two orders of magnitude. When the vertical field compo- nent is ignored, our results would be similar to find- ings of Faghei & Mollatayefeh (2012). Our results indi- cate that a stronger toroidal component can enhance the infall and the rotational velocities. A combination of equations (26) and (28) exhibits that the radial ve- 1 + βφC3. Thus, the mate- locity is proportional to rial can move radially inward with faster rate when a p stronger azimuthal field component is considered. In or- der to explore the role of toroidal field component in the rotational velocity, it is helpful to compare the second term with the third one in equation (27). For our input parameters, the value of the second term, i.e. the gradi- ent of the total pressure plus the magnetic stress force, rises with increasing βφ. But the third term reduces as βφ increases. Since the value of third term is smaller than the value of second one, the rotational velocity rises for stronger magnetic field. The previous studies of a resistive ADAF also showed that the toroidal com- ponent of the magnetic field can increase these veloci- ties (Faghei & Mollatayefeh, 2012). Note that the spe- cific magnetic prandtl number in which the disc has a non-rotating motion is almost independent of the input parameter βφ. A comparison between Figures 1 and 2 shows that the role of wind in the changes of velocities is more important. One can expect that an increase in βφ leads to a reduction in the surface density which can be because of the increase of radial velocity (see equation (19)). Even if the radial field component is considered, the surface density decrease with increasing βφ (see Mosallanezhad et al., 2013). |
Processed_Self-similar_structure_of_resistive_ADAFs_with_out.txt | variables depends on the magnetic prandtl number. It results from the dependence of magnetic heating on the vertical field component. A comparison between the two terms of the magnetic heating exhibits that the coeffi- cient related to the vertical field component is greater then that related to the toroidal component of the mag- netic field. Thus, the influence of vertical field compo- nent can be more significant. In Fig. 3, one can see that the behavior of disc variables for different values of βz changes at a specific magnetic prandtl number about −1 < 1.1, for example, the ra- Pm tio H/r drops as the input parameter βz rises. But this −1 > 1.1. Since the ro- trend would be reversed if Pm tational velocity is highly sensitive to the vertical field component (equation (27)), this velocity among the vari- ables examined in this paper does not obey this trend. For our input parameters, the sign of the second term in equation (27) is negative and an increase in βz results in a greater absolute value for this term. Hence, a larger βz, unlike βφ, causes the disc material to rotate with a slower rate (see also Cao & Spruit, 2013; Cao, 2014). Mosallanezhad et al. (2013) also showed that there is a such trend for the rotational velocity in the absence of ohmic dissipation. For adopted range of Pm, the disc reaches a non-rotating limit when the vertical compo- nent of the magnetic field is strong enough. We also examine the effect of wind strength on Ω(Pm). Our re- sults indicate that the power law index s plays no role in vanishing the rotational velocity if the vertical com- ponent of the field is absent. This means that Ω(Pm) only due to the presence of strong vertical field falls to zero (compare all plots of Ω(Pm) in our model). Al- though the outflow is not the agent of vanishing Ω(Pm), its strength can affect the magnitude of the magnetic prandtl number Pm in which the non-rotating limit oc- curs. |
Processed_Self-similar_structure_of_resistive_ADAFs_with_out.txt | Figure 3. Similar to Fig. 1, but for s = 0.2 and different values of βz (as labeled). |
Processed_Self-similar_structure_of_resistive_ADAFs_with_out.txt | Figure 4. Similar to Fig. 1, but for s = 0.2 and different values of α (as labeled). |
Processed_Self-similar_structure_of_resistive_ADAFs_with_out.txt | ity parameter (see equation (26)). Thus, an increase in the viscosity parameter enhances the radial velocity. In- deed, the discs with higher viscosity parameter are able to transport higher angular momentum, and therefore the disc material moves inward with faster rate. Subse- quently, the increase of infall velocity leads to a lower surface density (see equation (19)). For small values of the magnetic prandtl number, the decrease of surface density due to the viscosity parameter is not significant. But if the viscosity parameter increases from 0.1 to 1.0, the surface density falls by ten orders of magnitude at high-magnetic prandtl number limit. In such a limit of Pm, however, the change in α from 0.1 to 1.0 causes the infall velocity to increase by seven orders of magnitude. One can also see that the magnetic prandtl number in which the non-rotating limit occurs is nearly indepen- dent of the viscosity parameter. |
Processed_Self-similar_structure_of_resistive_ADAFs_with_out.txt | The existence of outflows and large-scale magnetic fields in the inner regions of an ADAF have been confirmed by numerical simulation. In such regions of ADAFs, the magnetic diffusivity can play an important role in de- termining the properties of ADAFs. In the present pa- per, therefore, we investigated the dynamics of resis- tive ADAFs with coexistent outflows and large-scale magnetic fields. The large-scale magnetic fields were as- sumed to have z and φ components. Here, the outflows can extract the mass, the angular momentum, and the energy from the disc. In the presence of outflows, nu- merical simulations and observations suggest a power- law function for the accretion rate whose power law in- dex represents the wind strength. We assumed a steady state flow and presented the self-similar solutions that are described by a function of the radial distance. |
Processed_Self-similar_structure_of_resistive_ADAFs_with_out.txt | diffusivity, so that the radial velocity of the disc ma- terial approaches the Keplerian speed. This increase is more significant when both vertical and azimuthal com- ponents of the large-scale magnetic field get stronger. Since the discs with higher viscosity parameter are able to transport higher angular momentum, the disc mate- rial moves inward with faster rate. This can lead to a high mass accretion rate. |
Processed_Self-similar_structure_of_resistive_ADAFs_with_out.txt | We found that the surface density generally tends to zero in the high-magnetic diffusivity limit, unless the vertical field component is weak. Since the infall veloc- ity, and therefore accretion rate increase at this limit, the surface density of a resistivity disc decreases sig- nificantly. Any change in the density profile can affect the emitted spectrum of an accretion flow. Although the surface density decreases with the increase of azimuthal field component, the effect of vertical field component on the surface density depends on the magnetic diffusiv- ity. In large-magnetic diffusivity limit, the role of verti- cal field component in the modification of density pro- file is similar to what we found for the azimuthal field component. In the opposite limit, however, the trend of vertical field component will be different. |
Processed_Self-similar_structure_of_resistive_ADAFs_with_out.txt | Our results showed that the disc becomes thicker with increasing the magnetic diffusivity. The disc thickness decreases as the wind strength or the viscosity parame- ter increases, while the disc becomes thicker for stronger azimuthal field component. We also displayed that an in- crease in the wind strength (or the viscosity parameter) reduces the disc temperature, while stronger azimuthal field component enhances the temperature. The tem- perature of an ADAF decreases due to stronger vertical field component only if small-magnetic diffusivity limit is considered. Indeed, the role of vertical field compo- nent in decreasing the disc temperature depends on the magnetic diffusivity. A comparison between our results and pervious works shows that the effective agent of temperature reduction is outflows and the decreases of temperature due to the vertical field component is seen only in ideal MHD and/or non-ideal MHD with low re- sistivity. |
Processed_Self-similar_structure_of_resistive_ADAFs_with_out.txt | netic field lines to start to reconnect. Hence, the pres- ence of the resistivity can lead to the heating and the ac- celeration of electrons. Furthermore, Ding et al. (2010) suggested that the resistive heating may be helpful to heat the electrons to higher temperatures, as we found in this work. |
Processed_Self-similar_structure_of_resistive_ADAFs_with_out.txt | The author would like to thank an unknown referee for a constructive report. Thanks to Dr. Mohsen Shadmehri for a helpful comment. |
Processed_Cooling_for_instantons_and_the_Wrath_of_Nahm.txt | The dynamics of instantons and anti-instantons in lattice QCD can be studied by analysing the action and topological charge of configurations as they approach a self-dual or anti-self-dual state, i.e. a state in which S/S0 = |Q|. We use cooling to reveal the semi-classical structure of the configurations we study. Improved actions which eliminate discretization errors up to and including O(a4) are used to stabilise instantons as we cool for several thousand sweeps. An analogously improved lattice version of the continuum field-strength tensor is used to construct a topological charge free from O(a4) discretization errors. Values of the action and topological charge obtained with these improved operators approach mutually-consistent integer values to within a few parts in 104 after several hundred cooling sweeps. Analysis of configurations with |Q| ≈ 1 and |Q| ≈ 2 supports the hypothesis that a self-dual |Q| = 1 configuration cannot exist on the 4-torus. |
Processed_Cooling_for_instantons_and_the_Wrath_of_Nahm.txt | states with different winding numbers, and their topological charge is given by the difference be- tween the winding numbers of the states that they connect. Further important properties of (anti-)instantons are spherical symmetry in the four dimensions of Euclidean spacetime, topolog- ical charge of ±1, and an associated action of S0 = 8π2/g2 which is independent of the radius of the instanton. |
Processed_Cooling_for_instantons_and_the_Wrath_of_Nahm.txt | The lattice is a 4-dimensional approximation to Euclidean spacetime. This means that physics on the lattice should ideally approach physics in the continuum if we can take the limit of infinite lattice volume and infinitessimal lattice spacing (V −→ ∞ and a −→ 0). Unfortunately reducing the lattice spacing indefinitely is not practical, nor is increasing the total lattice volume, since these approaches to the continuum limit rapidly increase the time taken to perform computations. We have chosen to remove the non-physical ‘edge’ of the lattice (simulating an infinite volume) by defining periodic untwisted boundary conditions on our lattices. We are therefore performing numerical QCD on a 4-toroidal mesh of points. |
Processed_Cooling_for_instantons_and_the_Wrath_of_Nahm.txt | The discretization errors introduced by the fi- nite lattice spacing are reduced algebraically by Symanzic improvement [2] and non-classical er- rors arising from the self-couplings of the gluon fields are dealt with by the use of mean-field im- provement [3]. Since the lattice we use is topologically equivalent to a 4-torus, we may consider the consequences of the Nahm transform [4] for the stability of the configurations we study. The Nahm trans- form is a duality mapping which interchanges be- tween an SU (N ) configuration with topological charge Q on the torus and an SU (Q) configura- tion with topological charge N on the dual torus. Since there are no instanton solutions in U (1), the Nahm transform implies [5][6] that there can be no self-dual |Q| = 1 configurations on the torus. |
Processed_Cooling_for_instantons_and_the_Wrath_of_Nahm.txt | and c5 is a free parameter which we can use to “tune” the action. By setting c5 = 0 we cre- ate a 4-loop improved action, denoted 4LIS for short, and by setting c5 = 1/10 we eliminate the contribution of S(L(1,2)) and S(L(1,3)), thereby creating a 3-loop improved action, denoted 3LIS for short. For the purposes of investigating a 5LIS we choose to follow DeForcrand et al. and set c5 = 1/20, midway between the 3LIS and 4LIS values. It is important to note that the 3LIS and 4LIS are just special cases of the general 5LIS, and all choices of c5 are free from discretization errors up to and including O(a4), however their O(a6) errors will be different, leading to slightly different results. |
Processed_Cooling_for_instantons_and_the_Wrath_of_Nahm.txt | We hence desire the topological charge calculated on the lattice to approach integer values as the configuration being studied is cooled towards a dilute instanton gas condition. Unfortunately dis- cretization errors tend to lead to non-integer val- ues of Q. Fortunately, however, we can improve the topological charge. |
Processed_Cooling_for_instantons_and_the_Wrath_of_Nahm.txt | Notice that in order to enforce the tracelessness of the Gell-Mann matrices we have subtracted one- third of the trace of A. |
Processed_Cooling_for_instantons_and_the_Wrath_of_Nahm.txt | and k5 is a free parameter, which enables us to create 3-loop, 4-loop, and 5-loop improved defini- tions of the field-strength, all of which are O(a4)- improved. In addition to constructing Q, the field strength may be used in (5) to create a “reconstructed ac- tion”. Since this action is improved in a different manner to the cooling action, comparison of both actions may serve as a mechanism with which to test the improvement procedures. In this brief report we will simply “test the wa- ter” with the values c5 = k5 = 1/20 (5-loop), c5 = k5 = 0 (4-loop), and c5 = 1/10, k5 = 1/90 (3- loop), and show that we obtain excellent results even though many other 5-loop operators, corre- sponding to other choices of c5 and k5, are possi- ble. |
Processed_Cooling_for_instantons_and_the_Wrath_of_Nahm.txt | In practice, because of the large discretization errors present, a cooling scheme based upon the plaquette action will destroy topological struc- ture over the whole lattice if it proceeds for long enough. However, a cooling scheme using an improved action will stabilise at a state with non-trivial structure. We would anticipate cool- ing schemes with small discretization errors will reproduce expected continuum behaviour, espe- cially integer values of Q and S/S0, and remain stable for hundreds of cooling sweeps. When com- paring several cooling schemes, the one which fulfills these criteria best will be the preferred scheme for studying gauge field dynamics. |
Processed_Cooling_for_instantons_and_the_Wrath_of_Nahm.txt | Figure 2. Action obtained with 2-loop (N), 3LIS (△), 4LIS (marked), and 5LIS (•) cooling. |
Processed_Cooling_for_instantons_and_the_Wrath_of_Nahm.txt | rithm stabilises a configuration at a state with S/S0 ≈ 6, indicating the presence of six in- stantons and/or anti-instantons, while plaquette- based cooling gradually destroys the structure of the configuration. In Figure 2 we compare the values of the action obtained with 2-loop, 3LIS, 4LIS, and 5LIS-based cooling on the same configuration used in Fig- ure 1. We can see that the 4LIS curve drops well below the desired integer value of six, indicat- ing the presence of large negative discretization errors. This comparison and similar work with other configurations have lead us to conclude that 3LIS-based cooling gives the most integer-like re- sults in general, and furthermore it is faster than 4LIS or 5LIS-based cooling, since it requires fewer computational steps. |
Processed_Cooling_for_instantons_and_the_Wrath_of_Nahm.txt | In Figure 3 we compare the various improved Q values obtained as we cool a configuration. The 3-loop improved operator gives the most continuum-like results, coming to within around 1 part in 6000 of an integer, and has the small- est computational cost of all the O(a4)-improved operators. Similar comparisons with other config- urations have lead us to select the 3-loop operator as our preferred choice for studying the topologi- cal charge of cooled configurations. |
Processed_Cooling_for_instantons_and_the_Wrath_of_Nahm.txt | are highly reliable, and hence the destabilization of |Q| ≈ 1 configurations is due to the conse- quences of the Nahm transform, indicating that on the lattice, just as in the continuum, a con- figuration may not simultaneously fulfill the two criteria S/S0 = 1 and |Q| = 1 on the 4-torus. |
Processed_Accurate_OH_maser_positions_from_the_SPLASH_pilot_.txt | state OH masers are also detected towards circumstel- lar envelopes of evolved giant and supergaint stars, such as Miras and OH/IR stars (Nguyen-Q-Rieu et al. 1979; Sevenster et al. 1997a, 1997b, 2001a), and are typically dominated by emission from the 1612 MHz satellite line (i.e. F = 1 → 2, 1612 MHz; Reid & Moran 1981). The maser emission in the 1720 MHz satellite-line transition (F = 2 → 1), is usually associated with supernova rem- nants (SNRs) (Goss & Robinson 1968) and traces the in- teraction between SNRs and surrounding dense molec- ular clouds. Ground-state OH masers are also found associated with planetary nebulae (PN; Uscanga et al. 2012), comets (G´erard et al. 1998) and the centres of ac- tive galaxies (Baan et al. 1982). |
Processed_Accurate_OH_maser_positions_from_the_SPLASH_pilot_.txt | the Galactic Plane, between Galactic longitudes of 315◦ and 45◦ and Galactic latitudes of −3◦ and +3◦. The survey of Sevenster et al. observed only the 1612 MHz OH line and was thus heavily favoured towards detect- ing evolved star OH masers. Therefore, these previous surveys suffer from biases (e.g. towards HMSF regions or evolved star sites) and thus the full population of each type of astrophysical object mentioned above may not be comprehensively studied. Although it is clear that ground-state OH masers are detected towards different types of astrophysical objects (HMSF regions, the en- velopes of evolved stars, SNRs), the proportion of masers associated with each kind of object remains unknown. Studying the proportion will yield a better understand- ing of the ground-state OH maser populations and char- acteristics of their associated astrophysical objects. |
Processed_Accurate_OH_maser_positions_from_the_SPLASH_pilot_.txt | In order to reduce biases caused by targeted surveys or surveys only observing specific ground-state OH tran- sitions, the Southern Parkes Large-Area Survey in Hy- droxyl (SPLASH) was carried out with simultaneous ob- servations of all four ground-state transitions. SPLASH surveyed 176 square degrees of the southern Galactic plane and Galactic Centre (Dawson et al. 2014), i.e. be- tween Galactic longitudes of 332◦ and 10◦ and Galactic latitudes of −2◦ and +2◦ (152 square degrees), plus an extra region around the Galactic Centre, where coverage was extended up to Galactic latitudes +6◦ for Galactic longitudes 358◦ to 4◦ (24 square degrees). Unsurpris- ingly, the survey has detected about 600 sites exhibit- ing OH maser emission. The observations were con- ducted with the Australia Telescope National Facility (ATNF) Parkes 64-m telescope and achieved a mean rms (root-mean-square) point-source sensitivity of ∼65 mJy in each 0.18 km s−1 channel in maser-optimised cubes, but were limited by the spatial resolution (about 13′). This spatial resolution is insufficient to reliably identify the astrophysical objects associated with the maser emis- sion. Therefore, higher spatial resolution observations are needed, which is the motivation for this work. |
Processed_Accurate_OH_maser_positions_from_the_SPLASH_pilot_.txt | The majority of OH masers detected in SPLASH are in the 1612 MHz satellite-line transition and show a double-horned spectral profile. These masers are evolved star OH masers, which also occasionally exhibit detec- tions in the 1665 and/or 1667 MHz main-line transi- tions. Star formation OH masers usually have many spectral features in the main-line transitions. Each spectral feature corresponds to one maser spot on the sky, which is usually considered to arise in a single, well-defined position (Walsh et al. 2014). Star forma- tion OH masers preferentially trace a later stage of the HMSF process, compared to methanol masers and wa- ter masers (e.g. Caswell & Haynes 1987; Caswell 1998; Breen et al. 2010a). As mentioned above, OH masers in the 1720 MHz transition are quite rare and are commonly associated with shocked gas, which can be detected in SNRs (Goss & Robinson 1968), HMSF regions (Caswell 2004; Edris et al. 2007), occasionally asymptotic giant branch (AGB) stars (Sevenster & Chapman 2001b) and PNe (Qiao et al. 2016). |
Processed_Accurate_OH_maser_positions_from_the_SPLASH_pilot_.txt | km s−1) to maximise the sensitivity to maser spots that have a typical OH maser line-width (based on the Parkes SPLASH spectra; Dawson et al. 2014). |
Processed_Accurate_OH_maser_positions_from_the_SPLASH_pilot_.txt | The absolute uncertainty in maser positions depends on the phase noise during the observations, which is re- lated to the distance between the phase calibrator and the target region and the accuracy of the known loca- tions of the phase calibrators, the locations of the an- tennas, as well as the atmospheric conditions. A typi- cal value of 1′′ for the absolute positional uncertainty is adopted. This number is based on previous OH maser surveys with the ATCA by Caswell (1998), who found absolute positional uncertainties of OH masers were bet- ter than 1′′. The relative positional uncertainty between maser spots detected in the same field of view can be more accurate than the absolute uncertainty. The 1- θ σ relative positional uncertainty is approximately 2S/N (Reid et al. 1988), where θ is the convolved size of the source (i.e., the beam size in the case of point sources, such as masers). This allows us to compare the distribu- tions of maser spots within a site on smaller scales than the absolute uncertainty, but does not allow us to com- pare the relative positions of maser spots with other data sets (e.g. IR survey: GLIMPSE, WISE; Benjamin et al. 2003, Churchwell et al. 2009, Wright et al. 2010; as de- scribed below) to better than 1′′. |
Processed_Accurate_OH_maser_positions_from_the_SPLASH_pilot_.txt | The above manual method of searching for and char- acterising features is at least as accurate as an auto- mated method (Walsh et al. 2012, 2014). After identi- fying maser spots, their positions are determined in the following steps. An integrated intensity map (in Miriad a moment 0 map) is created with the unbinned channels over which the maser spot is detected. This integrated in- tensity map has the highest signal-to-noise ratio for that maser spot and therefore exhibits the most accurate po- sition. If two maser spots are overlapping in velocity, we identify their positions by only integrating over velocities where the maser spot in question dominates the flux den- sity. The Miriad task imfit is used to fit the integrated intensity map and obtain both the position and relative uncertainty in the position of the maser spot. The peak velocity and peak flux density of a maser spot are deter- mined from the spectrum, extracted at the position of the maser spot using Miriad task uvspec. The integrated flux density is also calculated using the spectrum of the maser spot (see Section 3.1 for more details). We used a systematic approach to determine the velocity range for each maser spot: in order to improve the S/N of the spectra, we averaged the data by binning five channels to give a velocity resolution of about 0.45 km s−1. We calcu- lated the rms noise (1-σ) of the binned data and identify the velocity range as those channels with flux density in excess of 3-σ. In addition, we excluded any part of the spectrum with noise. Where spectra appear to show two overlapping velocity components with a trough between them, we classify each peak as a separate spot if the dif- ference in flux density between the peak of the weaker line and the lowest point of the trough is greater than the 1-σ noise on the data. This method includes most emission from maser spots, but at the same time does not include weak emission from channels such that the noise may dominate in that velocity range. However, this method is not perfect and so the derived velocity ranges and integrated intensities are intended to be used as a guide. This is particularly the case when two maser spots are blended in both velocity and position. |
Processed_Accurate_OH_maser_positions_from_the_SPLASH_pilot_.txt | Given that there are very few detections of 1720 MHz masers, we have used that band to investigate the contin- uum emission associated with star formation OH masers. We concatenated 3 zoom bands at this frequency which resulted in a bandwidth of 2 MHz. The data were re- duced using standard techniques for ATCA continuum data, resulting in a typical rms noise of 5 mJy. The re- sults are detailed in Section 4.5. |
Processed_Accurate_OH_maser_positions_from_the_SPLASH_pilot_.txt | cates whether the maser site is a new detection. There are 44 maser sites which only have one maser spot, while the remaining 171 sites have more than one. The richest maser site (G336.075−1.084) has 26 maser spots in to- tal at 1612, 1665 and 1667 MHz and the second richest maser site (G337.404−0.402 from MAGMO) exhibits 21 maser spots at 1665 and 1667 MHz. |
Processed_Accurate_OH_maser_positions_from_the_SPLASH_pilot_.txt | are minor axis uncertainty, major axis uncertainty and position angle. The twelfth column is the identification of each maser site and the reason for this identification. |
Processed_Promoting_and_countering_misinformation_during_Aus.txt | this misinformation has received scrutiny in the media and academic research. Here we study Twitter communities spreading this misinforma- tion during the population-level event, and investigate the role of online communities and bots. Our in-depth investigation of the dynamics of the discussion uses a phased approach – before and after reporting of bots promoting the hashtag was broadcast by the mainstream media. Though we bot-like bots, which present as genuine humans. accounts were differ- we Further, the between ences following Twitter insights. First, Supporters promoted misinfor- mation by engaging others directly with replies and mentions response, using information. Opposers inter- Second, their centrality action more the appeared last to audience. Finally, unaffiliated accounts shared the same URLs as Opposers over Supporters by a ratio of 9:1 in the last phase, hav- in the first phase. This ing shared mostly Supporter URLs exposing foiled Supporters’ communi- misinformation campaigns. We cation strategies observed here in other misinformation-related discussions and could inform counter-strategies. |
Processed_Promoting_and_countering_misinformation_during_Aus.txt | up to the January 6 Capitol Riots in America (Scott, 2021; Ng et al, 2021). |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | A search for H I absorption has been made in 23 radio galaxies selected from a complete sample. The observations were made with the Australia Telescope Compact Array (ATCA), with the Very Large Array (VLA) and for one galaxy with the Westerbork Synthesis Radio Telescope (WSRT). In five galaxies H I absorption was detected. We investigate how the de- tection rate is distributed among galaxies with different radio and optical properties. Among the Fanaroff-Riley (FR) type I radio galaxies, only one object (10% of total) was detected in H I absorption. The H I absorption in these objects is likely to come from a nuclear disk, as found for other galaxies of this type (e.g. NGC 4261 and Hydra A). The low detection rate is consistent with the hypothesis (as suggested by optical and X-ray data) that the “standard” pc scale, geometrically thick torus is not required in low-luminosity radio galaxies. This is consistent with earlier optical work. In the case of FR type-II powerful radio galaxies, no H I absorption has been detected in broad line radio galaxies, while three out of four narrow-line radio galaxies have been detected (the one non-detection having quite a high upper limit). All these are compact or small radio galaxies. To first order this is consistent with the predic- tions of the unified schemes, assuming that the H I absorption is due to an obscuring torus. However, the indications of this being the only cause of the absorption are not very strong. In particular, we find that in two of the three detected objects that the H I is blueshifted compared to the systemic velocity. In the third galaxy (PKS 1549–79) two redshift systems (from the optical lines) are found. The uncertainty in the systemic velocity derived from optical lines is discussed. Finally, by also considering data available in the literature, we find a tendency for radio galaxies with a strong component of young stellar population and far-IR emission to show H I absorption. The overall richer ISM that is likely to be present in these galaxies may be a factor in producing the absorption. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | observations of the H I absorption against radio continuum sources open the possibility of detecting relatively small quantities of H I even in high redshift objects. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | H I absorption has been found in a large number of different types of AGN. By looking at the properties of the absorption, it appears that this absorption can be caused by a variety of gaseous components. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | of ≤ 100 pc can be the cause for such absorption. In the case of Cygnus A VLBA observations (Conway 1998) show that the ab- sorption is occurring against the counter-jet, and that it is elongated spatially in a direction perpendicular to the radio jet. This strongly suggests that the H I absorption is due to a circumnuclear torus that blocks the view of the quasar nucleus of Cygnus A. In the case of NGC 4621, the H I absorption on the VLBI scale (van Langevelde et al. 2000) appears against the counter-jet at a projected distance from the nucleus of only ∼5 pc. The cause of the absorption is be- lieved to be a thin disk (∼1.3 pc thick). Finally, in Hydra A (Taylor 1996) the absorption is likely to arise from a disk with thickness of ∼30 pc. All of these structures could be part of a continuous struc- ture which spans all the way from a many-kpc scale disk down to the classical accretion disk of a few tens of parsec in size. This has also been found to be the case for Seyfert galaxies (Gallimore et al. 1999). |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | A class of radio sources that appears to be particularly suit- able for detecting and studying the H I absorption are Compact Symmetric Objects (CSO) and Compact Steep Spectrum (CSS) ra- dio sources, as noted by Conway (1996). These are strong radio sources, a few kpc (i.e. sub-galactic) in size and are thought to be very young ( <∼ 104 yr). It is claimed that in these objects the most likely cause of the H I absorption is also a nuclear disk/torus. How- ever, the interpretation of some of these absorption profiles relies on the value of the systemic velocity that often is very uncertain. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | Evidence of very broad absorption (∼400-500 km s−1, com- pared to the 50-200 km s−1 found in general in radio galaxies) has been found in luminous infrared galaxies (Mirabel 1989) like 4C 12.50 and 3C 433. In particular, Mirabel (1989) noted that for 4C 12.50 the shape of the H I absorption profile bares a striking similarity with that of starburst galaxies like Arp 220 (Mirabel 1982), suggesting a link between radio-IR spirals and powerful ra- dio galaxies in the early stages of their evolution (like 4C 12.50). |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | The interaction between the radio plasma and the interstellar medium can also play a role. In the Seyfert galaxy IC 5063, very broad (∼700 km s−1) H I absorption, blueshifted with respect to the systemic velocity (Morganti et al. 1998, Oosterloo et al. 2000) has been observed coincident with a radio lobe about 1 kpc from the nucleus. This is clear evidence for a strong interaction between the radio plasma and the ISM in the galaxy, causing the cold gas to be ’blown out’ from the centre of the galaxy at large velocities. Evidence for interaction between the radio plasma and the ISM has been found in at least two radio galaxies: the superluminal object 3C 216 (Pihlstr¨om et al. 1999) and 3C 326 (Conway et al. in prepa- ration). On even larger scale (> 10 kpc), H I absorption against the radio lobes has been found in the radio galaxy Coma A (Morganti et al. 2000a). |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | In some galaxies (e.g. Centaurus A, van der Hulst et al. 1983; NGC 1052 van Gorkom et al. 1986), additional narrow profiles are clearly seen, redshifted with respect to the systemic velocity by a few hundred km s−1. These profiles could correspond to gas clouds falling towards the AGN. Van Gorkom et al. estimated that these infalling clouds could provide the right amount of fuel for the AGN. It is clear that radio galaxies show a diversity of H I absorption properties. However, most previous studies have been based on het- erogeneous samples of objects, often including the most spectacu- lar objects in a given redshift range. In order to establish the H I absorption properties of the general population of radio galaxies it is important to study a well-defined sample. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | other characteristics of the galaxies (e.g. orientation, as one would expect if the nuclear disk/tori are the cause of absorption) we have searched for H I absorption in a sample of radio galaxies for which a wealth of information (from radio, optical and X-ray observations) is available. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | Here we present the results obtained for 23 radio galaxies ob- served so far. In §2 we describe the observations that were made ei- ther with the Australia Telescope Compact Array (ATCA) or with the Very Large Array (VLA) depending on the declination of the objects and for one galaxy with the Westerbork Synthesis Radio Telescope (WSRT). In §3 we describe the results obtained and we discuss in more detail the galaxies in which H I absorption is found. In §4 we discuss the occurrence of the H I absorption in the context of the overall characteristics of the sources, in §5 we discuss the difficulties in the interpretation of the data due to uncertainties in the systemic velocities derived from optical emission lines. Finally, in §6 we discuss the likely origin of the absorption in the galaxies of our sample. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | The observed radio galaxies have been selected from the 2-Jy sam- ple of radio sources (Wall & Peacock 1985, Tadhunter et al. 1993, 1998; Morganti et al. 1993, 1999 and references therein). This sam- ple includes strong (S2.7 GHz > 2 Jy) radio sources with decli- nation south of 10◦ and redshift less than 0.7. Although the aim was to have a “complete” sample of radio galaxies observed in H I, the actual observed sample has been constrained by three techni- cal limitations: 1) the observed objects have a radio core typically stronger than ∼60 mJy (one exception is the galaxy PKS 1949+02 (3C 403)); 2) the redshift limit reachable with the available re- ceivers is between 0.15 and 0.22 (depending on the telescope used); 3) even inside this range a few objects could not be observed be- cause their redshifted H I would fall in spectral regions with strong known radio interference. While the first two constraints have in- troduced some bias on the selection of the observed objects, the latter has just randomly affected the sample. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | The radio and optical characteristics of the observed galax- ies are summarised in Table 1, mainly taken from Morganti et al. (1997). In Table 1 we also included the values of R (defined as the ratio of core to extended radio fluxes and considered to be a orienta- tion indicator). We have observed nine Fanaroff-Riley (FR) type-I galaxies objects, eight FR type-II, two intermediate morphology galaxies, one BL Lac, and three compact radio galaxies (but one of them is the very low radio power galaxy NGC 4696). As de- rived from our optical spectra (Tadhunter et al. 1993, 1998), four galaxies are narrow line radio galaxies (NLRG), four are broad line radio galaxies (BLRG) and the remainder have weak optical emis- sion lines (WLRG, EW[OIII] < 10 ˚A). Note that some of the FR type-II selected have weak emission lines. Because of the limita- tions mentioned earlier, most of the objects in our sample are FR type-I radio galaxies or broad-line radio galaxies (that are known to have, on average, stronger radio cores compared to the narrow-line galaxies, Morganti et al. 1997). Also the large number of WLRG FR-II galaxies is due to the fact that they have relatively strong cores (Morganti et al. 1997). |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | used the 6 km configuration of the ATCA and the B-array config- uration of the VLA. The spatial resolution of the ATCA and VLA data is typically between 6 and 10 arcsec. The log of the obser- vations is given in Table 2. PKS 1333–33 was observed with both instruments. The ATCA observations of PKS 2152–69 are part of a larger study of the radio continuum properties of this object (Mor- ganti et al. in prep). One galaxy (3C 459) was also observed with the WSRT. The system temperature of the new 21-cm frontends used at the WSRT is low (∼ 27 K) even at the relatively low fre- quency of the redshifted H I in this object (1165 MHz), therefore allowing us to better estimate the characteristics of the absorption in this galaxy. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | With the ATCA we were able to observe galaxies up to a red- shift z ∼ 0.15, corresponding to 1235 MHz. In the ATCA obser- vations we used a 16 MHz bandwidth and 512 channels in order to be sure to have both a large velocity range and good velocity resolution. The velocity resolution obtained is ∼ 6.6 km s−1 be- fore Hanning smoothing. Each object was observed a number of times (for approximately 20 minutes each time) throughout a 12 h observing period in order to obtain as much as possible uniform uv coverage. As bandpass and flux-density calibrator we have used PKS 1934–638, assumed to have (according to the latest result by Reynolds, 1996) 15.1 and 14.9 Jy at 1235 and 1405 MHz, the low- est and highest frequencies used. We spent between 10 min and 1 hour on the bandpass calibrator, depending on the flux of the core of the observed sources. The log of the ATCA observations is given in Table 2a. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | The VLA observations were made with a 6.25 MHz band- width and 64 channels. The spectral resolution obtained is ∼21 km s−1. Most of the observed radio galaxies have z < 0.1, but we also included two objects with z ∼ 0.22. The observing time and frequency for each galaxy are given in Table 2b. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | backend, 10 MHz bandwidth 256 channels given about 12 km s−1 velocity resolution. This galaxy was observed for 3.6 h. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | The spectral data were calibrated with the MIRIAD package (Sault et al. 1995). The continuum subtraction was done by using a linear fit through the line-free channels of each visibility record and subtracting this fit from all the frequency channels (“UVLIN”). The estimated rms noises given in Tables 2a, b and c are taken from Hanning smoothed images obtained with natural weighting. We have also used the line-free channels for producing an image of the continuum emission. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | A few more objects from the 2-Jy sample have data available from the literature: NGC 4261 (PKS 1216+06, 3C 270) observed by Jaffe & McNamara (1994) and recently by van Langevelde et al. 2000; Hydra A observed by Dwarakanath et al. (1995) and Taylor (1996); 3C 317 (PKS 1514+07) observed by Jaffe (1991), Centau- rus A (van der Hulst et al. 1983), PKS 1934–63 (V´eron-Cetty et al. 2000), PKS 0023–26 and PKS 2135–20 (Pihlstr¨om et al. in prep). We shall consider also these objects in our discussion. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | Of the 23 radio galaxies we observed, five galaxies show H I ab- sorptions: PKS 1318–43 (NGC 5090), PKS 1549–79, PKS 1814– 63, PKS 1717–00 (3C 353) and PKS 2314+03 (3C 459). No H I in emission has been detected. The results obtained for our sample are summarised in Table 3. ∆S represents the maximum depth of the absorption, or its 3σ limit. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | PKS 1549–79 and PKS 1814–63. For these objects the velocities are derived from new or re-analysed spectra and more details are given below in the notes. For the extended objects an image (ob- tained using the line-free data and using uniform weighting) of the radio continuum is also shown. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | the flux (core or total depending on the size of the source) derived from the continuum image obtained from the line-free channels in our observations (see column 4 of Table 3 for the flux density of the nucleus). For our calculations the covering factor has been as- sumed to be cf = 1 and we will not be able to put constrains on this value until we can image the absorption using VLBI. Thus, the derived optical depths (and the column densities) can be considered as lower limits. The values of the peak optical depth go from 0.5% for 3C 459 and ∼3% for PKS 1318–43 and PKS 1549–79, up to ∼11% for 3C 353 and ∼21% for PKS 1814–63. The full width of the H I profiles measured at 20% of the peak profile intensity (W20) are given in column 7 of Table 3. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | The integrated column density (column 8 in Table 3) has been estimated as: NH ≃ 1.83 × 1018Tspin R τ dv, where Tspin is the spin temperature in Kelvin. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | The spin temperature is a very uncertain parameter in deriving the column densities. The value of the spin temperature is essen- tially unknown for extragalactic sources, so typical Galactic values (between 100 and 1000 K, Heiles & Kulkarni 1988) are often as- sumed. However, the presence of a strong continuum source near the H I gas may significantly increase the spin temperature because the radiative excitation of the H I hyperfine state can dominate the usually more important collisional excitation (see e.g. Bahcall & Ekers 1969). If the gas producing the H I absorption is in a cir- cumnuclear torus (i.e. at a distance ≤ 100 pc from the nucleus), it will be irradiated by the hard X-ray emission from the nucleus. According to Maloney et al. (1996, and references therein), purely atomic gas in these conditions has a stable equilibrium temperature of ∼8000 K. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | H I absorption is detected against the core (see Fig. 1). The H I profile appears double peaked, although the relatively high noise level does not allows us to say much about its shape. The full width at 20% of the peak is ∼150 km/s. The H I velocity derived as arith- metic mean of the velocities at 20% of the peak intensity of the H I profile is ∼3425 km s−1. This is very close to the systemic velocity of the galaxy (3421 km s−1, Tadhunter et al. 1993). The peak optical depth is about 3% and the integrated column density 2.7 × 1018Tspin cm−2. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | For the undetected galaxies, a 3σ limit to the peak optical depth was estimated and is given in Table 3. For these galaxies, the upper limits to the column density (also given in Table 3) were calculated by assuming a profile width of 30 km s−1. The upper limits to the optical depth are typically between 0.5 and a few %, with the extreme values of ∼9% for 3C 278 and 3C 403 – objects with particularly weak cores. Some of the upper limits are due to observations not being deep enough, but many are significant upper limits indicating that in those objects no H I absorption is observed at the level at which it is typically observed in the other radio galax- ies. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | PKS 1549–79 is a core-jet radio source of ∼400 pc (about 120 mas) in size and has been studied in detail at radio frequencies by King (1994) and King et al. (1996) using the southern VLBI network. Overall it has quite a flat spectral index because the flux is domi- nated by the core component (about 3 Jy at 2.3 GHz and spectral index α ∼ 0.0). |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | The H I absorption detected (see Fig. 2) appears to be formed by a narrow component (W20 ∼ 100 km s−1) peaked at ∼45580 km s−1, although a broader component (W20 = 250 km s−1), mainly on the blue side of the narrow component, is possibly de- tected. The peak optical depth of the narrow component is about 2.5%. The integrated column density (including the broad absorp- tion) is ∼3.6 × 1018Tspin cm−2. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | A new optical spectrum was obtained for this galaxy with the 3.6-m ESO telescope. A detailed discussion of the optical char- acteristics of PKS 1549-79 will be presented in Tadhunter et al. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | Figure 1. Radio continuum image (Top) of PKS 1318–43 (NGC 5090) at 21 cm obtained from the line-free channels. The contour levels are: 3.5 mJy beam−1 to 274 mJy beam−1 in steps of factor 1.5. H I absorption spec- tra (Bottom) observed against the core of the radio galaxy PKS 1318–43 (NGC 5090). The systemic velocity (from Tadhunter et al. 1993) is indi- cated. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | Figure 2. H I absorption spectra (after Hanning smoothing) in the source PKS 1549–79. Two different velocities have been derived from the [O III]λ5007 and [O II]λ3727 emission lines. See text for details. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | high redshift it has been detected by IRAS with a IR luminosity of LIR = 1.6 × 1011 L⊙ if we use the 100µm upper limit as a detection. This derived luminosity is high for a radio galaxy. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | PKS 1717–00 is a strong radio galaxy with a FR-II morphology. The radio structure is shown in the continuum image in Fig.3 and a detailed study of the structure of this source, and in particular of the well collimated jet (and counter-jet), has been done by Swain, Bridle & Baum (1999). Despite being a powerful radio galaxy, only weak [O III]λ5007 and Hα emission lines have been detected in the spectrum of this object (Tadhunter et al. 1993). The optical contin- uum of this object is very red (Simpson et al. 1996). |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | The systemic velocity is 9120 ± 59 km s−1 (or z = 0.030421 ± 0.000197) from the RC3 (De Vaucouleurs et al. 1991), very close to the peak of the H I absorption. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | PKS 1814–63 is a Compact Steep Spectrum (CSS) source with a basic double-lobed structure oriented almost north-south (Tzioumis et al. 1996). Because VLBI radio continuum observations are avail- able for this galaxy at only one frequency (2.3 GHz), no informa- tion is available on which of these structures (if any) correspond to the core. The overall extent of the source is 410 mas corresponding to 328 pc. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | Figure 4. H I absorption spectra (after smoothing) in the source PKS 1814- 63. The systemic velocity recently determined is indicated (see text for de- tails). |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | on Vhel = 19220 km s−1, and has an optical depth of a few per- cent, more similar to what usually found in other galaxies. The inte- grated column density of the deep component is ∼1.7 × 1019Tspin cm−2. Including also the broad component the column density is ∼ 2 × 1019Tspin cm−2. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | We have re-analyzed the optical spectrum presented in Tad- hunter et al. (1993). Because of the presence of a bright star near the object, the quality of the spectrum is low. However, by doing a more careful extraction of the spectrum, we have now detected the emis- sion line at higher S/N, and, consequently, have been able to deter- mine a more accurate redshift. We derive a new value for the red- shift of the galaxy of z = 0.0645, corresponding to Vhel = 19350 km s−1. This is different from the previously derived redshift. Al- though this new redshift will require further confirmation with bet- ter data, the H I absorption appears now mostly blueshifted respect to the new systemic velocity. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | In this galaxy the optical continuum spectrum is bluer than that of a typical early-type galaxy, and absorption features suggestive of young stellar populations have been detected (Miller 1981). The emission-line spectrum indicates moderate ionization. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | The radio morphology shows a strong nuclear component and two very asymmetric lobes (Ulvestad 1985 and Morganti et al. 1999). The size of the radio emission is only ∼10 arcsec (corre- sponding to ∼48 kpc). The central component detected in the VLA image (see e.g. Morganti et al. 1999) has been found by higher reso- lution Merlin observations (P. Thomasson private communication) to be resolved in two components, both with a steep spectral index. At the resolution of the WSRT data the continuum emission from 3C 459 is unresolved, while at the resolution of the VLA the continuum emission is slightly resolved (see Fig. 5). |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | The peak optical depth is only about 0.7% in the VLA data and even lower (∼0.4%) in the WSRT data. The difference between the two is most likely due to a dilution effect: the WSRT data include the continuum flux from the entire source while from the VLA data it is clear that the absorption is coming from the core and/or eastern lobe. Only higher resolution observations will tell us whether the absorption is really against the core or more against the lobe. The FWHM of the absorption appears to be quite broad in our VLA data (∼400 km s−1) and is confirmed by the higher quality WSRT pro- file shown in Fig. 5. The absorption is centered on Vhel ∼ 65550 km s−1. The systemic velocity is 65924 km s−1 (corresponding to a redshift of 0.2199, Spinrad et al. 1985). |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | The integrated column density derived from the VLA data is similar to those found for most other radio galaxies: ∼ 2.7 × 1018Tspin cm−2. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | H I absorption is detected in 5 galaxies of the 23 observed. The detection rate is between 20 and 30% depending whether we also consider the data available in the literature. Among the latter, 7 galaxies were observed and 5 detected: these numbers are likely to be biased toward detection (it is less likely that non-detections are published!). A similar detection rate was found in the only other systematic search for H I absorption (van Gorkom et al. 1989). |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | Figure 5. Radio continuum image of 3C459 from VLA at 6cm (Toppannel, from Morganti et al. 1999) and from the line free channels of the VLA 21 cm data presented in this paper (Middlepannel, the contour levels are from 0.195 to 3.9 Jy beam−1 in steps of 390 mJy beam−1. Bottom pannel: H I absorption spectra (from the VLA data (after smoothing, fill line) and from the WSRT data (dashed line). The systemic velocity (from Spinrad et al. 1985) is indicated. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | is worth mentioning that the undetected source (PKS 2135–20) is likely to be a broad line radio galaxy. Broad wings have been de- tected in the Mg II line of this galaxy (Shaw et al. 1995). This would be consistent with the trend for BLRGs described above. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | An important issue for understanding the origin of the H I absorp- tion is how the systemic velocity of the galaxy compares with the velocity of the H I. This can be a complicated (and often under- estimated) problem and therefore it is worthwhile to emphasize the uncertainties connected with the values of the systemic velocities. It was already pointed out by Mirabel (1989) how the systemic veloc- ities derived from emission lines can be both uncertain and biased by motions of the emitting gas. As discussed in the notes above, two of our galaxies detected in H I illustrate these problems. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | In PKS 1549–79 the old redshift available in the literature (Tadhunter et al. 1993) and derived from [O III]λ5007 emission lines (as for the majority of radio galaxies) suggested the H I be- ing strongly redshifted. However, as discussed above, from a newly obtained optical spectrum we find that the [O II]λ3727 lines give a radial velocity about 660 km s−1 higher compared to that derived from the [O III]λ5007 line, and a velocity more consistentwiththe HI absorption (see Fig. 1). What should we consider as systemic velocity of the galaxy? The [O III]λ5007 line appears to be very broad, one of the broadest observed in radio galaxies, suggesting that this line comes from a region where the gas is very disturbed. This region could be closer to the nucleus, in agreement with the suggestion made by Hes et al. (1993), while the [O II]λ3727 lines would originate farther away from the nucleus and from a less dis- turbed region. Thus, the value derived from the [O II]λ3727 lines is likely to be closer to the real systemic velocity of this galaxy. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | in which this ef- fect in 4C12.50 (PKS 1345+12) two redshift systems were found: one containing the high-ionization lines and the other (at +240 km s−1 with respect to the first one) containing the low-ionization lines. Moreover, al- though a large uncertainty is present, the redshift of the H I absorp- tion features appears to be the same as that of the low-ionization emission lines, therefore supporting this as the systemic velocity. We will come back to the similarities between PKS 1549–79 and 4C 12.50 in §6.2.3. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | appears now to be mainly blue-shifted with only part of the shal- low absorption observed at the systemic velocity (unlike what was obtained using the old value for the systemic velocity) as shown in Fig. 4. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | However, better quality spectra will be necessary to confirm this result in particular for 2314+03. Finally, in the case of PKS 1549–79, the H I absorption has a velocity about 100 km s−1 lower than that derived from the [O II]λ3727, therefore, considering the errors, is only marginally blueshifted. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | As mentioned in the Introduction, circumnuclear disks or tori are believed to be the most likely causes of H I absorption in radio galaxies. Thick or warped disks surrounding the central regions are required by unified schemes in the case of FR-II radio galaxies to explain the large column densities seen in X-ray observations, the broad lines seen in reflected light and the far-infrared colours. On the other hand, while the unification of BL Lacs and FR-I re- quires relativistic beaming, the presence of such a thick disk in such objects is still a matter of discussion (Urry & Padovani 1995). In particular, while in BL Lacs broad lines have been detected (Ver- meulen et al. 1995, Corbett et al. 1996) the situation is still unclear for FR-Is. Thus, for these radio galaxies the presence of thick disks surrounding the central black hole is not required to explain the existing observations in term of the unified schemes. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | We have detected H I absorption in only one of the FR-I ra- dio galaxies of the 23 sources observed. The absorption in this case appears to be at the systemic velocity with a width of ∼150 km s−1and a relatively low column density (2.7 ×1018Tspin cm−2). The absorption line could be due to a nuclear disk as is sug- gested for some of the other FR-I galaxies of the 2-Jy available the in literature (e.g. Hydra A, NGC 4261). This is in contrast with the narrow profiles observed in galaxies like Centaurus A which were interpreted in term of single clouds in front of the radio continuum. At a first glance, the low detection rate of H I absorption in FR- I galaxies is somewhat surprising given that, as shown from recent HST images, gas and dust, in particular circumnuclear dust-lanes, are commonly present in low-luminosity radio galaxies (see e.g. de Koff et al. 1996, Chiaberge et al. 1999, Capetti et al. 2000). |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | Figure 6. Distribution of R (ratio of core to extended radio fluxes, at 4.8 GHz from Morganti et al. 1993) for (Top) the FR-I radio galaxies of the 2Jy sample observed in H I (either from this work or from the literature) and (Bottom) for the remaining FR-I galaxies in the 2Jy sample. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | One possibility to explain the low detection rate is that the selection of the objects is biased: because only the objects with a relatively strong core could be observed, we may have biased the sample to objects more beamed toward us. This is very likely to be the case for some objects: for example in the case of 3C 78, the image taken by the Hubble Space Telescope (HST) shows point- like nuclear emission (Chiaberge et al. 1999) and there is evidence for the presence of an un-absorbed, nuclear X-ray source, (Trussoni et al. 1999) both supporting the lack of nuclear obscuration as re- lated to the particular viewing angle. Similar conclusions have been drawn from X-ray observations of PKS 0625–35 (Trussoni et al. 1999). |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | To check how much such a bias may have affected our sample, we have compared the distribution of R for the FR-I radio galaxies of the 2Jy sample observed in H I (either from this work or from the literature) with the values of R for the remaining FR-I galaxies (i.e. not yet observed in H I). |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | The two distributions are presented in Fig. 6. The values of R of the two distributions cover a similar range, with only two cases (Fornax A and Hercules A) of very low values of R among the objects not observed in H I. Indeed, they were not observed because of the very low values of their core flux. Thus, the galaxies observed in H I do not appear to be strongly biased toward high R values. Moreover, we note that Hydra A, one of the FR-I in the sample with a relatively low value of R (log R = −2.17), shows H I absorption at ∼50% level (Dwarakanath et al. 1995) and even higher in the VLBI observations while other objects with similar value of R are undetected at the ∼1% level. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | Moreover, despite the presence of the dusty structures, optical studies show that the nuclei of FR-I galaxies appear basically unob- scured. Chiaberge et al. (1999) find Central Compact Cores (CCC) in 85 % of the FR-I galaxies selected from the 3CR catalogue and observed with HST. Thus, their fraction of unobscured objects is comparabletothefractionofgalaxiesundetectedinHI absorption. Based on the CCC detection rate, they derive a thickness over size ratio < ∼ 0.15. If we take the values derived for NGC 4261 (a circu- lar velocity Vc of 610 km s−1 for a radius of r = 5 pc and velocity dispersion σ = 130 km s−1) and using the thin disk relation (i.e. the thickness of the disk is h = rσ/Vcr) we find that this gives a 10% probability of H I absorption from the disk against the nucleus, thus similar to the detection rate in the FRI sample as a whole. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | In summary, our results for FR-I galaxies support the idea the most of the H I absorption in these objects comes from a thin nu- clear disk as is found for other galaxies of this type (NGC 4261 and Hydra A). However, they do not provide the definitive evidence against a thick torus because, e.g. a thick pc scale torus could be molecular especially given that the quasar component in FRIs, if present, may be weak, so that there may be less photo-dissociated molecular gas. Nevertheless, the low detection rate of H I absorp- tion is consistent with that derived from optical work (Chiaberge et al., 1999), i.e. the “standard” pc scale, geometrically thick torus is not present in low-luminosity radio galaxies. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | For FR-II radio galaxies, the presence of a thick nuclear disk is a vital ingredient in the unified schemes hypothesis. Broad line radio galaxies are supposed to be seen pole-on and therefore obscura- tion from the torus should not occur. Narrow-line radio galaxies (NLRG) are viewed more edge-on, and it might be expected that H I absorption would be more common in such objects. We will discuss this in some more detail. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | H I absorption seems indeed absent in BLRGs, although the statis- tics are still limited. For 3C 445 we only have a relatively high up- per limit to the optical depth: ∼4%. However, the remaining three have a tighter upper limit to the optical depth of ∼1%. An even lower limit of ∼0.5% has been derived in the compact radio galaxy PKS 2135-20 (Pihlstr¨om et al. in prep.): this is in clear contrast with the detections found for the other compact sources in the sam- ple. It appears that our data are consistent with the assumptions that BLRGs are viewed pole-on and the absorption is related to an obscuring torus. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | than the one derived from the X-ray data. A similar discrepancy has been found between the column densities derived from IR/optical data and X-ray data, with the former much lower than the latter. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | In this case, the discrepancy between these values can be ex- plained by assuming that the radio-emitting jets in 3C445 are more spatially extended than the region emitting the X-rays. Therefore they are less likely to be affected by HI absorption associated with dense material close to the central AGN. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | Of the four narrow line radio galaxies observed by us, all but one are detected and the undetected (3C 403) has a very high limit to the optical depth due to the weakness of the radio core. The col- umn densities found in our sample go from ∼ 3 × 1018Tspin cm−2 for PKS 1547–79 and 3C 459 up to a few times 1019 cm−2 for PKS 1814–63. The latter value is similar to the value found in the prototype FR-II radio galaxy Cygnus A (Conway & Blanco 1995). Two more objects (PKS 0023–26 and PKS 1934-63) have been ob- served by other groups (see §2) and both detected. The higher de- tection rate in NLRGs compared to that in BLRGs is what is ex- pected for unified schemes. However, other factors may also play a role. |
Processed_HI_absorption_in_radio_galaxies:_effect_of_orienta.txt | Due to the selection criteria, theNLRGsobservedinoursam- ple are mainly compact or small objects. As pointed out by Heck- man et al. (1983), a radio source of sub- or at most galactic size rep- resents an optimum background against which galactic H I might be seen in absorption, in particular if the radio continuum is confined to a scale comparable to the likely dimensions of a possible nu- clear disk. Indeed, Conway (1996) found a high detection rate of H I absorption in compact objects. Also from this study we confirm that compact radio sources with narrow emission lines have a high chance of being detected. Of the 6 compact/small radio galaxies in the sample (including the two galaxies observed by Pihlstr¨om et al.), the only one undetected is likely to be a broad line radio galaxy (PKS 2135–20, Pihlstr¨om et al.). |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.