A study of giant radio galaxies at RATAN-600
M.L. Khabibullina, O.V. Verkhodanov, M. Singh, A. Pyria, S. Nandi, N.V. Verkhodanova
aa r X i v : . [ a s t r o - ph . C O ] A ug Bull. Spec. Astrophys. Obs., 2011, 66, 171–182 c (cid:13) Special Astrophysical Observatory of the Russian AS, 2018
A Study of Giant Radio Galaxies at RATAN-600
M.L. Khabibullina a , O.V. Verkhodanov a , M. Singh b , A. Pirya b , S. Nandi b , N.V. Verkhodanova a a Special Astrophysical Observatory of the Russian AS, Nizhnij Arkhyz 369167, Russia; b Aryabhatta Research Institute of Observational Sciences, Manora Park, Nainital 263 129, India
Received July 28, 2010; accepted September 15, 2010.
We report the results of flux density measurements in the extended components of thirteen giantradio galaxies, made with the RATAN-600 in the centimeter range. Supplementing them withthe WENSS, NVSS and GB6 survey data we constructed the spectra of the studied galaxycomponents. We computed the spectral indices in the studied frequency range and demonstratethe need for a detailed account of the integral contribution of such objects into the backgroundradiation.
Key words:
Radio lines: galaxies—techniques: radar astronomy
1. INTRODUCTION
According to the generally accepted definition, gi-ant radio galaxies (GRGs) are the radio sources withlinear sizes greater than 1 Mpc, i.e. the largest ra-dio sources in the Universe. They mostly belong tothe morphological type FR II [1] and are identifiedwith giant elliptical galaxies and quasars. In com-parison with normal galaxies, GRGs are quite rare.This makes their statistical study and a detailed re-search into the causes of their formation as a pop-ulation quite difficult. They are the biggest objectsin the visible universe, and it is possible that theycould play a special role in the formation of the large-scale structure. Radio observations of GRGs allow toclarify what caused such gigantic objects to appear.Large dimensions of GRGs as well suggest that thesesources must be at the last stages of evolution.The study of these objects began with the source3C 236 [2]. The models of radio sources [3],[4] predicttemporal variations in the radio luminosity and lin-ear sizes of giant radio sources. According to thesemodels, GRGs should be very old objects (aged over10 yrs) that are presumably located in the environ-ments with decreased matter density, as compared tothe smaller sources with comparable radio luminosi-ties [5]. Komberg and Paschenko [6] have analyzedthe radio and optical data (SDSS, APM) for the ra-dio galaxies and quasars, and concluded that apartfrom the effect of the environment, the giant sizes ofradio sources can be attributed to the population oflong-lived radio-loud active nuclei, which in turn canevolve into to GRGs. Multi-frequency observations [7]have shown that the spectral age of GRGs is longer than the one, expected from the evolutional models.As noted in [8], such radio galaxies may affect theprocesses of galaxy formation, since the pressure ofgas, outflowing from the radio source, may compressthe cold gas clouds thus initiating the developmentof stars on the one hand, and stop the formation ofgalaxies on the other hand. Several groups [9]-[18]continue to study the characteristics of GRGs, long-ing to explain their gigantic sizes. However, no unam-biguous solution of this problem exists up to date.In our recent work [19],[20], based on the analy-sis of radio spectra of giant radio galaxies we havecome to the conclusion that the change in the spec-tral index of giant radio galaxies, depending on theshift with respect to the galactic center, noted earlierin [10] is linked with the particle energy variationsin the components, caused by the pressure variationof the gas, flowing around, i.e. due to the changes inthe medium depending on the distance from the hostgalaxy. However, general conclusions will be more sig-nificant at the integral approach to the GRG popula-tion as a whole. An essential step in the study of thecauses of large sizes of giant radio galaxies is a com-parative study of similar properties of “normal” radiogalaxies [21]-[24]. Note that Soboleva [25] has earliermade the observations of radio galaxies with minutedimensions in the centimeter wavelength range withthe RATAN-600, and discovered that the morphologi-cal structures have virtually identical spectral indices.Therefore, the study of the GRG sample objects willsupplement the data on the radio spectra of this pop-ulation of galaxies.GRGs are as well interesting to study the contami-72 O.V. Verkhodanov, M. Singh, A. Pyria, S. Nandi, N.V. Verkhodanova
Figure 1:
Radio images of giant radio galaxies in the NVSS survey. The circles mark the object components,observed with the RATAN-600.
STUDY OF GIANT RADIO GALAXIES AT RATAN-600
Main parameters of the observed giant radio galaxies
Source Coordinates Redshift Type Angular Flux densityRA+Dec (J2000.0) size, (1.4 GHz),hhmmss+ddmmss arcmin mJyGRG 0139+3957 013930+395703 0.211 II 5.7 801.1GRG 0452+5204 045253+520447 0.109 I 9.7 2869.1GRG 0751+4231 075109+423124 0.203 II 6.0 162.3GRG 0912+3510 091252+351016 0.249 II 6.2 157.4GRG 0929+4146 092911+414646 0.365 II 6.6 165.5GRG 1032+2759 103214+275600 0.085 II 11.0 284.1GRG 1216+4159 121610+415927 0.243 II 5.2 415.5GRG 1343+3758 134255+375819 0.227 II 11.3 131.0GRG 1400+3017 140040+301700 0.206 II 10.8 451.9GRG 1453+3308 145303+330841 0.249 II 5.7 455.5GRG 1521+5105 152115+510501 (0.37) II 4.3 1197.5GRG 1552+2005 155209+200524 0.089 II 19.6 2385.6GRG 1738+3733 173821+373333 0.156 II 6.5 236.0GRG 1918+516 191923+514334 0.284 II 7.3 920GRG 2103+6456 210314+645655 0.215 II 4.8 119.7Table 2:
The observed GRG regions. Sections: c—central, n—northern, s—southern. N t is the number oftransits. The coordinates (right ascension+declination) of the centers of components are listed for the epochJ2000.0 Source Section Coordinates of center N t observed regionsGRG 0139+3957 c 013927.4+395653 1GRG 0452+5204 c 045343.7+520556 11GRG 0751+4231 c 075153.9+422945 10GRG 0912+3510 n 091252.0+351231 5GRG 0912+3510 s 091250.0+350631 1GRG 0929+4146 c 092951.8+414353 10GRG 1032+2756 n 103212.5+275925 3GRG 1032+2756 c 103214.4+275555 3GRG 1032+2756 s 103215.1+275115 1GRG 1216+4159 c 21641.4+415545 11GRG 1343+3758 c 134255.0+375819 2GRG 1400+3017 n 140045.0+302214 3GRG 1400+3017 s 140038.4+301325 3GRG 1453+3308 n 145302.0+331046 4GRG 1453+3308 c 145303.0+330856 2GRG 1453+3308 s 145301.4+330556 1GRG 1521+5105 c 152132.5+510232 7GRG 1552+2005 c 155209.0+200524 8GRG 1738+3733 n 173820.6+373658 2GRG 1738+3733 c 173821.0+373333 2GRG 1738+3733 s 173821.8+373108 1GRG 2103+6456 c 210322.1+645929 974 O.V. Verkhodanov, M. Singh, A. Pyria, S. Nandi, N.V. Verkhodanova
Fig. 1. (Contd.)nation of the microwave background due to their mor-phological properties (as well as their size, shape andorientation) in the pixel domain, or the phase char-acteristics in the harmonic domain [26]. While theirapparent contribution in the background maps in themillimeter range seems to be weak, the angular sizesof sources (up to 10 minutes of arc) produce problemsin the component separation due to the changes in thespectral index in the locations of extended radio com-ponents of galaxies. Therefore, one of the interestingproblems is to evaluate and account for a possiblecontribution of GRGs to the background anisotropy,both their radiation in the millimeter range, and theeffects, occurring in the component separation on themultipole scales ℓ ≥
500 in different frequency ranges.In this paper, we present the results of measure-ments of the flux densities of giant radio galaxies atthe centimeter and decimeter ranges based on the re- sults of two observational sets at the RATAN-600.
2. RATAN-600 DATA
The initial GRG sample is selected based on thelists from [9]-[11] within the operational range ofthe RATAN-600. The observations of the GRGs wereperformed at the Northern sector of the RATAN-600in the second decade of December 2008, and at theSouthern sector in early January 2010. The contin-uous spectrum radiometers of the Feed Cabin 1 [27]were used at the 1.38, 2.7, 3.9, 6.25, 13 and 31 cmwavelengths. Note that despite the large range ofwavelengths, the elevated noise conditions during theperiod of observations have restricted the data, suit-able for the analysis to four bands only: 2.7, 3.9, 6.25
STUDY OF GIANT RADIO GALAXIES AT RATAN-600
Integral flux densities (mJy) of the radio sources according to the data from the RATAN-600 and theWENSS, NVSS surveys
Source 2.7 cm 3.9 cm 6.25 cm 13 cm 92 cm 21 cm 6.2 cmcomponent RATAN RATAN RATAN RATAN WENSS NVSS GB60139+3957w – – 470 857 – 2120 656c – – 139 252 – 744 317e – – 82 212 – 133 –0452+5204c 417 827 1141 1984 18760 3003 8440751+4231c 103 227 274 476 1365 202 350912+3510n – – 21 <
120 160 56 <
20s – – 70 144 512 101 200929+4146c – – 215 315 1560 200 911032+2759n – – 46 <
120 – 92 <
20c – – 35 <
120 – 75 59s – – 86 108 – 138 561216+4159c – – 123 207 1604 264 771400+3017n – – 61 178 1258 333 73s – – 40 155 1053 155 371453+3308n – – 19 <
40 420 245 <
20c – – 109 138 593 149 131s – – 76 <
40 488 89 < <
20e – – 139 252 – 744 317ee – – 470 857 – 2120 6561738+3733n – – 36 56 152 64 <
20c – – 46 113 720 117 93s – – 29 <
30 133 58 < < <
180 337 124 32Figure 3:
Radio spectra of radio galaxies: a giantGRG 0929+4146, having a steeper spectrum, and anormal binary J092924+414618, the radio brightnessdistribution of which was integrated by the power beampattern in the RATAN-600 observations. Individualspectra for each source are shown, giving the totalcontribution to the observed radio brightness distri-bution. and 13 cm. The size of beam patterns in the centralsection at the observational elevation amounted to25 ′′ , 36 ′′ , 43 ′′ and 57 ′′ , 90 ′′ and 119 ′′ , respectively. Forthe wavelengths of 6.25 and 13 cm we used the spec-tral analyzer subchannels to effectively deal with theinterference. The list of observed sources is given inTable 1, the log book—in Table 2. Note that for thegalaxies GRG 1343+3758 and GRG 2103+64 we didnot manage to reach a sufficient signal-to-noise ratioto be able to detect the sources.Depending on the position angle of the radiostructure, from one to three sections of the sourcehave been made (Table 2). The number of object tran-sits through the telescope’s beam pattern (BP) waslimited by the observational time, granted by the pro-gram committee. To bind the flux densities with the international scale[28] we observed standard calibration sources fromthe RATAN-600 standard list [29],[30]. The curves ofsource transits were analyzed in the FADPS standard76
O.V. Verkhodanov, M. Singh, A. Pyria, S. Nandi, N.V. Verkhodanova
Figure 2:
Radio spectra of giant radio galaxies, constructed based on the RATAN-600 data and the NVSS,WENSS, GB6 surveys (Table 3) etc. The RATAN-600 data are marked by black ellipses. processing system [31],[32]. The first stage of dataprocessing consists in subtracting the low-frequencytrend with a window smoothing of 8 minutes of arcfrom the source transit records. A transition to fluxdensities was performed by integrating the extendedsignal, approximated by a set of Gaussians, and atransition to the flux density scale through calibra-tion. The noise level in the records of single transitsat the wavelengths of 1.38, 2.7, 3.9, 6.25 and 13 cm in the observations at the northern sector at the el-evation of 76 ◦ amounted to 8.1, 5, 36, 3.3 and 65mK/s / , respectively, while in the southern sector atthe elevation of 82 ◦ it amounted to 17.2, 8.9, 18.1,10.7 and 96.6 mK/s / , respectively. The measure-ments of flux densities at the wavelengths of 2.7, 3.9,6.25 and 13 cm are given in Table 3. The table alsolists the values of integrated flux densities of the stud-ied sources, computed from the NVSS (NRAO VLA STUDY OF GIANT RADIO GALAXIES AT RATAN-600
Fig. 2. (Contd.)78
O.V. Verkhodanov, M. Singh, A. Pyria, S. Nandi, N.V. Verkhodanova
Table 4:
Approximation relations for the continuousradio spectra of giant radio galaxies in the wavelengthrange from 92 cm to 2.7 cm
Source Radio spectrumcomponent0139+3957w 3 . − . x c 6 . − . x e 3 . − . x . − . x . − . x . − . x s 0 . − . x . − . x . − . x c − . − . x s 1 . − . x . − . x − .
914 + 1 . x − . x s 3 . − . x . − . x c − .
275 + 12 . e − x s 1 . − . x . − . x e 34 . − . x + 2 . x ee 3 . − . x . − . x . − . x c 1 . − . x s 0 . − . x . − . x Sky Survey) maps [33] at 21 cm, constructed on theVLA radio interferometer (USA), and WENSS (West-erbork Northern Sky Survey) maps [34], constructedat the Dutch radio interferometer in Westerbork at92 cm. The table as well gives the data from the GB6(Green Bank) catalog [35]. Integration of the radiobrightness distribution in the maps was carried outusing an interactive image analysis package SkyView with a preliminary subtraction of the trend. For iden-tifying the objects and making the estimates of theirparameters we also used the CATS database [36],[37].Among the CATS catalogs, we found the identifica-tions in the GB6 [35], VLSS [38], 6C [39], 7C [40], 8C[41], Texas [42], and B3 [43] surveys.One of the problems in observing the GRGs withthe RATAN-600 is the evaluation of flux densities ofmulticomponent sources, where the components haveclose right ascensions, but different declination an-gles. In this case, the radio source is oriented coax- Table 5:
Spectral indices of giant radio galaxy compo-nents at 6.25 and 13 cm
Source Spectral index Spectral indexcomponent 6.25 cm 13 cm0139+3957w –1.29 –1.29c –2.24 –2.24e –1.18 –1.180452+5204c –0.83 –0.830751+4231c –0.70 –0.700912+3510n –0.77 –0.77s –0.56 –0.560929+4146c –0.58 –0.581032+2759n –1.40 –1.40c –0.24 –0.24s –0.62 –0.621216+4159c –0.88 –0.881400+3017n –1.54 –1.27s –1.20 –1.201453+3308n –1.13 –1.13c –0.31 –0.43s –0.69 –0.691521+5105c –0.73 –0.731552+2005w –0.46 –0.46e 0.03 –1.63ee –1.00 –1.001738+3733n –0.57 –0.57c –0.80 –0.80s –0.56 –0.562103+6456c –0.92 –0.92ially (vertically, if projected on the plane) with thebeam pattern in the observations at the meridian.Then, in each section there appears a contribution ofdifferent radio components of the source object dueto an extended form of the beam pattern vertically.One way to restore the obtained signal is to simu-late an extended source with more accurate data fromthe NVSS catalog, its convolution with the calculatedbeam pattern of the RATAN-600 and the calculationof the signal correction, caused by the confusion in thecontributions of different components. This approach,applied in [19] gave a somewhat inflated estimate ofthe fluxes after the signal reconstruction, which maybe due to inaccurate plotting of the source shape.Therefore, we used other estimates based on the ad-ditivity of the convolution operator. In this case, whenthe same beam pattern is used to observe two com-ponents of the source, separated by an angular dis-tance ∆ h , the observed flux density S of one com-ponent, apart from the incoming flux B , gets a partof the flux from the second component kS , where k is the coefficient (the value from the antenna beam STUDY OF GIANT RADIO GALAXIES AT RATAN-600
Top: radio spectra of the GRG 1521+5105components. The RATAN-600 data area markedby black ellipses. Bottom: a radio image ofGRG 1521+5105 (in the center) according to theNVSS data, superimposed on the optical DSS surveyimage. pattern), corresponding to a vertical shift from thecenter of the beam by the angular distance ∆ h .Hence, we find that a real signal from one compo-nent can be estimated as B = ( S − kS ) / (1 − k ).This algorithm was used in the analysis of the GRGobservations in several sections. The results are listedin Table 3.The error in determining the flux densities in theRATAN-600 observations of the sources with fluxdensities greater than 50 mJy was about 10%, andfor the flux densities below 50 mJy it amounted to13% at the wavelength of 6.25 cm. Similarly, at 13 cmthe error of 10% was found for the values greaterthan 180 mJy, and 15 %—for the flux densities be-low 180 mJy.For the sources GRG 0452+5204 andGRG 0751+4231 at the wavelengths of 2.7 and 3.9 cm the observations were carried out in the“beam switching” mode. To take into account apossible flux density drop of an extended objectwhen observed in this mode, we modelled thepassage of two sources through two horns. Themodelling procedure included: the calculation of atwo-dimensional beam pattern of the RATAN-600applying Korzhavin’s method [44] within the FADPSsystem [31] at the observed wavelength, convolutionwith the components of the observed source, andmodelling the object transit through the RATAN-600beam pattern. In addition, we also used in the dataanalysis the BP calculations made by Majorova [45].The conversion factor estimate of the integral fluxdensity of a given extended radio source in the beamswitching mode model was accounted for in theanalysis of real observations. Based on the measurement data we have constructedthe spectra of the radio source components. De-scribing the spectra, we fitted them by the formulalg S ( ν ) = A + Bx + Cf ( x ), where S is the flux den-sity in Jy, x is the frequency logarithm ν in MHz,and f ( x ) is one of the following functions: exp( − x ),exp( x ) or x . The spg system was used for the spec-tral analysis [46]. The spectra of the components aredemonstrated in Fig. 2. The analytical description ofthe continuous spectrum curves for the componentsof the studied GRGs is shown in Table 4.
3. DISCUSSION OF RESULTS
The spectra constructed (Fig. 2) demonstrate a va-riety of GRG properties. The fact that the spectralindices in the components of the observed radio galax-ies vary significantly is obvious, even the shapes ofthe spectra are different: from a very steep spectrumof the GRG 0139+3957 source, to the spectra withflattening, like in the GRG 1453+3308 source compo-nent.For the sources, observed only in the central sec-tion, the values of flux densities, measured at thewavelengths of 2.7, 3.9, 6.25 and 13 cm, are listed inTable 3, and spectral indices amount to the x argu-ment coefficients from Table 4. For the rest of sourcesthe results are presented in Table 5. Note that inthe case of the GRG 0751+4231 source spectrum theRATAN-600 data points are located higher than thedata from the GB6 and NVSS. In the GB6 catalogthis object is nearly a point object, which explainsthe low level of the cataloged flux value. It is mostlikely that the level of the corresponding value fromthe NVSS is caused by the same factor as in the case80 O.V. Verkhodanov, M. Singh, A. Pyria, S. Nandi, N.V. Verkhodanova when the integration by peak values leads to an in-complete account of weak diffuse emission. The ob-ject GRG 1738+3733 stands out among the observedsources, as both of its extended components have sim-ilar radio spectra, and identical spectral indices.Note that the change in the spectral index of giantradio galaxies, depending on the shift from the galac-tic center was already noted [10]. It is associated withparticle energy variations in the components, causedby the pressure variation in the surrounding gas, i.e.it is due to the changes in the medium, depending onthe distance from the host galaxy.The radio source, observed on the RATAN-600in the region of GRG 0929+4146 turned out to becomposed of two radio galaxies: GRG 0929+4146as such in the shape of a multidimensional ob-ject, stretching along one line, and a double type-FR II radio source with the coordinates ( α =09 h m s , δ = +41 ◦ ′ ′′ ), which merge into oneextended object that can be seen to the left of theGRG 0929+4146 in Fig. 1. The RATAN-600 does notresolve radio galaxies in the meridian transits, andhence Fig. 2 presents the total spectrum. We built aseparate spectrum for each radio galaxy according tothe NVSS, WENSS and 7C survey data (see Fig. 3).The integral radio spectrum of GRG 0929+4146 alonewas approximated by the dependence y = 3 . − . x , and the spectrum of the neighboring radiogalaxy was approximated by the dependence y =1 . − . x , which has a smaller slope and thusdemonstrates that the short-wave observations on theRATAN-600 are dominated by the radio emissionfrom the source J092924+414618.The radio galaxy GRG 1521+5105 was resolvedinto two components. Their spectra are demonstratedin Fig. 4. The integrated radio emission flux densi-ties for the two components are: 368 mJy at thewavelength of 13 cm and 167 mJy at 6.25 cm forJ152103+510600, and 181 mJy and 150 mJy, respec-tively for J152125+510401.The approximations for the radio spectra ofthe components are described by the functions: y = 2 . − . x and y = 1 . − . x . The radiogalaxy GRG 1521+5105, identified with the galaxySDSS J152114.55+510500.9, and having the photo-metric redshift of z = 0 .
37 (according to the NEDdatabase ), is located on the outskirts of the projec-tion of the cluster NSCS J152018+505306 with theredshift z = 0 .
52 (NED) at the angular distance of15 arc minutes from the center. Nevertheless, within10 minutes of arc from the radio galaxy there aremore than 1700 galaxies (according to the NED), and a large number of radio sources (Fig. 4). As we donot have any redshift data for the vast majority ofthem it is difficult to judge the physical belongingof GRG 1521+5105 to any group of galaxies. Still, arich neighborhood of this radio galaxy brings addi-tional interest to search for the reasons of its giganticsize.The RATAN-600 observations allowed to specifythe GRG component spectra and estimate their fluxesin the millimeter wavelength range at the extrapola-tion of the integrated radio spectrum. The flux den-sities of the studied GRG components lie in this partof the spectrum at the level above 0.6 mJy. As theexpected number of GRG-type objects amounts toseveral hundred on the full sphere [26], their contri-bution to the background radiation can, in principle,result in a bias in computing the background fluc-tuation level, not to mention the problem of signalisolation.We plan to further accumulate the data, compilethe lists of new GRGs and perform their observationswith the RATAN-600.
4. ACKNOWLEDGMENTS
The authors thank Yu. Sotnikova for her help with theRATAN-600 observations and S.A. Trushkin for valuablediscussions. The study made use of the NED databaseof extragalactic objects, the NASA/IPAC ExtragalacticDatabase, operated by the Jet Propulsion Laboratory,California Institute of Technology, under the contractwith the National Aeronautics and Space Administration.The authors also used the CATS [36],[37] database ofradio astronomy catalogs, and the FADPS [31],[32] sys-tem for processing the radio astronomy data. This workwas supported by the Leading Scientific Schools of Rus-sia (S. M. Khaikin school) grant, and the RFBR grant(project nos. 09-02-92659-IND and 09-02-00298). O.V.V.is also grateful for the partial support of the Dynasty foun-dation. References
B. L. Fanaroff and J. M. Riley, MNRAS , 31 (1974).R. G. Strom and A. G. Willis, A&A , 36 (1980).C.R. Kaiser, J. Dennett Thorpe, and P. Alexander, MN-RAS , 723 (1997).K. Blundell, S. Rawlings, and C.J. Willott, AJ , 677(1999).C.R. Kaiser and P. Alexander, MNRAS , 515 (1999).B. L. Komberg, I. N. Paschenko, Astronomy Reports ,1086 (2009), arXiv:0901.3721.K. H. Mack, U. Klein, C. P. O’Dea, et al. A&A , 431(1998). http://cats.sao.ru http://sed.sao.ru/ ∼ vo/fadps e.html STUDY OF GIANT RADIO GALAXIES AT RATAN-600
M. Jamrozy, J. Machalski, K. H. Mack, and U. Klein,A&A , 467 (2005).A. P. Schoenmakers, K. H. Mack, A. G. de Bruyn, et al.,A&AS , 293 (2000).A. P. Schoenmakers, A. G. de Bruyn, H. J. A. Roettgering,and H. van der Laan, A&A , 861 (2001).L. Lara, I. Marquez, W. D. Cotton, et al., A&A , 826(2001).L. Lara, G. Giovannini, W. D. Cotton, et al., A&A ,899 (2004).L. Saripalli, R. W. Hunstead, R. Subrahmanyan, andE. Boyce, AJ , 896 (2005).C. Konar, D. J. Saikia, C. H. Ishwara-Chandra, andV. K. Kulkarni, MNRAS , 845 (2004).C. Konar, M. Jamrozy, D. J. Saikia, and J. Machalski,MNRAS , 525 (2008).M. Jamrozy, C. Konar, J. Machalski, and D. J. Saikia,MNRAS , 525 (2008).J. Machalski, M. Jamrozy, S. Zola, and D. Koziel,A&A , 85 (2006).S. Nandi, A. Pirya, S. Pal, et al., MNRAS , 433(2010), arXiv:1001.3998.M. L. Khabibullina, O. V. Verkhodanov, M. Singh et al.,Astron. Zh. , 627 (2010), arXiv:1009.4539.M. L. Khabibullina, O. V. Verkhodanov, M. Singh et al.,Astron. Rep. , 392 (2011), arXiv:1108.3295.M. L. Khabibullina and O. V. Verkhodanov, Astrophys.J. Suppl. , 123 (2009), arXiv:0911.3741.M. L. Khabibullina and O. V. Verkhodanov, Astrophys.J. Suppl. , 276 (2009), arXiv:0911.3747.M. L. Khabibullina and O. V. Verkhodanov, Astrophys.J. Suppl. , 340 (2009), arXiv:0911.3752.O. V. Verkhodanov and M. L. Khabibullina, Astron. Lett. , 7 (2010), arXiv:1003.0577.N. S. Soboleva, Astrofiz. Issl. (Izvestiya SAO) , 50(1981).O. V. Verkhodanov, M. L. Khabibullina, M. Singh, etal., in Proc. Intern. Conf. Problems of Practical Cos-mology , Ed. by Yu.V. Baryshev, I.N.Taganov, and P.Teerikorpi (Russian Geograph. Soc., St. Petersburg,2008), p. 247.N. A. Nizhelskii, A. B. Berlin, A. M. Pilipenko et al., in