MRC B0319-454: Probing the large-scale structure with a giant radio galaxy
aa r X i v : . [ a s t r o - ph ] D ec Mon. Not. R. Astron. Soc. , 1–14 (2002) Printed 26 October 2018 (MN L A TEX style file v2.2)
MRC B0319 − V. Safouris , ⋆ , R. Subrahmanyan , , G. V. Bicknell and L. Saripalli , Research School of Astronomy and Astrophysics, Mount Stromlo Observatory, Australian National University,Cotter Road, Weston ACT 2611, Australia Australia Telescope National Facility, CSIRO, Locked Bag 194, Narrabri, NSW 2390, Australia Raman Research Institute, C V Raman Avenue, Sadashivanagar, Bangalore 560080, India
26 October 2018
ABSTRACT
We present an investigation of the relationships between the radio properties ofa giant radio galaxy MRC B0319-454 and the surrounding galaxy distributionwith the aim of examining the influence of intergalactic gas and gravity associatedwith the large-scale structure on the evolution in the radio morphology. Our newradio continuum observations of the radio source, with high surface brightnesssensitivity, images the asymmetries in the megaparsec-scale radio structure intotal intensity and polarization. We compare these with the 3-D galaxy distribu-tion derived from galaxy redshift surveys. Galaxy density gradients are observedalong and perpendicular to the radio axis: the large-scale structure is consis-tent with a model wherein the galaxies trace the ambient intergalactic gas andthe evolution of the radio structures are ram-pressure limited by this associatedgas. Additionally, we have modeled the off-axis evolution of the south-west radiolobe as deflection of a buoyant jet backflow by a transverse gravitational field:the model is plausible if entrainment is small. The case study presented here isa demonstration that giant radio galaxies may be useful probes of the warm-hot intergalactic medium believed to be associated with moderately over densegalaxy distributions.
Key words: galaxies: active – galaxies: individual (MRC B0319 − In a powerful radio galaxy, reaccelerated jet materialinflates the synchrotron lobes after passage throughthe termination shocks at the jet ends (Scheuer 1974;Begelman et al. 1984). The synchrotron lobes interactwith the surrounding thermal gas and their morpholo-gies are a result of this interaction. In cluster environ-ments, where XMM-Newton and Chandra have beenable to detect and image the relatively dense intra-cluster medium (ICM) in X-rays, the total intensityradio and X-ray contours follow each other and X-ray holes are observed at the locations of the radiolobes (e.g. Boehringer et al. 1993; McNamara et al. 2000;Nulsen et al. 2002; Bˆırzan et al. 2004). These holes areevidence that the expanding radio lobes interact with,and displace the X-ray emitting gas. Lobe interactionwith intra-cluster gas is also apparent in the bending of ⋆ E-mail: [email protected] (VS) radio plumes in wide-angle and narrow-angle tailed radiosources (e.g. Bliton et al. 1998; Hardcastle et al. 2005;Douglass et al. 2008). The morphologies of these sourcessuggest that the radio tails are deflected behind the mov-ing host galaxy, due to the ram-pressure exerted by theICM.Giant radio galaxies, which have linear sizes of > c (cid:13) Safouris, Subrahmanyan, Bicknell & Saripalli ample, if the parent optical galaxy resides at a boundarybetween an over-density and a void, then there may be agradient in the gas density surrounding the radio source.If this gradient is aligned with the radio jet axis, then wewould expect an asymmetry to arise in the lobe lengths,due to differing ram-pressure limitations in the ambientgas on the two sides. If, on the other hand, the gradientin the gas density is transverse to the jet axis, then wewould expect the light synchrotron lobes to evolve trans-verse to the jet axis and in the direction of decreasingambient density and pressure due to buoyancy.The diffuse gas in the environs of giant radio galaxies,outside of rich clusters, most likely pertains to the warm-hot phase of the IGM, whose existence at low-redshiftshas been predicted by large-scale cosmological hydrody-namical simulations of galaxy formation (Cen & Ostriker1999, 2006; Dav´e et al. 2001). In the simulations, thewarm-hot gas follows the filamentary galaxy distributionon large scales and represents unvirialized over-densitiesin the range 10-30. Little is known about this warm-hot gas since its thermal emission is not detectable bypresent day X-ray telescopes. However, the shape takenby the lobes of a giant radio galaxy is a visible manifesta-tion of their interaction with the surrounding gas. Thesegalaxies, therefore, make ideal probes of the unseen am-bient medium, which may be the warm-hot intergalacticmedium (WHIM).Recently, the 1.8 Mpc lobes of the giant radio galaxyMSH 05 − were used to constrain the properties of itsambient thermal gas (Subrahmanyan et al. 2008). Thelobes in this galaxy appear to be relicts and there is anobserved asymmetry, which appears to be related to ananisotropy in the local galaxy distribution. A comparisonof the properties of the radio lobes and those of the ambi-ent gas, which were derived from the surrounding galaxydistribution, indicated that the lobes were highly over-pressured despite their relict appearance. Alternatively,the density-temperature product for the ambient IGMmight be an order of magnitude larger than that pre-dicted by structure formation models indicating signifi-cant feedback in the IGM.In this paper, we present a similar study of anothergiant radio galaxy MRC B0319 −
454 (also referred topreviously as MSH 03 − and PMN J0321 − z ≈ .
6. Its 26 ′ size on the sky, make the measurement ofthe surrounding galaxy distribution possible with multi-object fibre instruments such as AAOmega on the Anglo-Australian Telescope. MRC B0319 − −
454 was pre-viously observed at 843-MHz with the Molonglo Obser-vatory Synthesis Telescope (MOST), and with the Aus-tralia Telescope Compact Array (ATCA) at 20, 13 and 6cm (Jones 1989; Saripalli et al. 1994). In the latter work,Saripalli et al. presented a detailed study of the mor- phology of the giant radio galaxy pointing out a num-ber of features that make the radio source unusual: aunique configuration of five compact hot spots in one ofthe lobes, a prominent jet and counter-jet detected outto exceptional distances, and lobes that are asymmetri-cally shaped and positioned about the core. Consideringthe likely expansion velocity of the source, it was arguedthat light travel-time effects were not responsible for theasymmetry in the lobe separations from the core. Basedon an examination of the projected galaxy distributionaround the host, Saripalli et al. attributed this feature tocorresponding asymmetries in the ambient intergalacticmedium.Given the Mpc size and pronounced asymmetriesin the radio morphology, MRC B0319 −
454 is an excel-lent example of a radio galaxy where the morphology isshaped by the IGM. In this paper, we examine the inter-action between the lobes and the ambient IGM using ournew radio observations made with the specific purposeof imaging the radio lobes fully. Our new improved ra-dio images of MRC B0319 −
454 have been made using anobserving mode designed to have high surface brightnesssensitivity and in full polarization. We use these in con-junction with redshift measurements of the surroundinggalaxy distribution, in an attempt to model the inter-action between the giant lobes with their IGM environ-ment. Our radio observations include polarization mea-surements for the first time.The paper is organized as follows: In Sect. 2we present our new radio continuum images of MRCB0319 − = 0 .
3, Ω Λ = 0 . H = 71 km s − Mpc − . The host elliptical galaxy,ESO 248-G10 (also cataloged as AM 0319 − M R = − . z = 0 . z = 0 . ′′ = 1.2 kpc. New radio continuum observations of MRC B0319 − ′ source in the 20 and 13 cmbands. Earth-rotation synthesis with full UV coveragewere made in 5 separate array configurations that em-phasized low spatial frequencies: 1.5D and 1.5A 1.5-km arrays, a 750B 750-m array, a EW352 352-m ar-ray, and a compact EW214 214-m array. The relativelycompact configurations were for improving sensitivity tothe low surface brightness structures, while the relativelylonger 1.5-km arrays were to enable imaging with sub-arcminute resolution. Visibilities were measured simulta-neously in two bands, 128-MHz wide, centered at 1378 c (cid:13) , 1–14 RC B0319 − and 2368 MHz. The continuum bands were covered in13 independent channels. A journal of the observations isin Table 1. The ATCA antennas have primary beam fullwidth half maximum (FWHM) of about 35 ′ and 21 ′ re-spectively in the 20 and 13 cm bands. Therefore, to accu-rately image the extended emission in this 26 ′ source weadopted the approach of mosaic observing and coveredthe double radio source with 8 separate pointings in a4 × ′ , which is approxi-mately equal to the sky plane Nyquist sampling require-ment for observing in the 13 cm band, was adopted forthe pointings.Data at both frequencies were reduced using stan-dard procedures in the MIRIAD package. The flux-density scale was set using observations of the primarycalibrator, PKS B1934 − − −
454 since the source has abright hot spot. We have not adopted a true joint mosaic-ing approach for this reason. Nevertheless, we note thatthe largest angular scales expected to be reliably repro-duced by the cut-and-paste mosaic are adequate becausethe angular scales of the largest emission structures areexpected to be less than the primary beam FWHM.
The mosaic image of MRC B0319 −
454 at 1378 MHz isshown in Fig. 1. The image was made with a beam ofFWHM 52 ′′ × ′′ at a position angle of 0 ◦ . The rms noisein the image in the vicinity of the radio galaxy is 0.25 mJybeam − . The image dynamic range, defined as the ratioof the peak brightness over the entire image to the rmsnoise, exceeds a 1000:1. The total intensity image of thegiant radio source shows an edge-brightened double radiosource with two large radio lobes that are located to thenorth-east (NE) and south-west (SW) of the radio core(whose location is marked in Fig. 1). Extended emissionis also detected associated with a partial jet that extendsfrom the core in the direction of the bright SW hot spot.The jet, as in previous MOST and ATCA images is tracedonly over a distance of 7 ′ from the core, which is aboutone-third of the distance between the core and SW hotspot. A weak radio source, which is associated with a b J =17 . z = 0 .
07 and its locationon the sky along the path of the SW jet and towards theend of its visible length is likely a chance coincidence: thegalaxy is probably a few tens of Mpc beyond the radiosource.Two broad peaks of enhanced radio emission are lo-cated adjacent one another and on either side of the ra-dio axis at the end of the NE lobe. The ridge-like peakon the eastern boundary is elongated parallel to the ra-dio axis, extends further from the core relative to theneighbouring peak located on the western boundary, andis also slightly brighter than the second peak. It maybe noted here that higher resolution images of this lobe(Saripalli et al. 1994) resolve the ridge along the east-ern boundary into a chain of hot spots, and suggest thatthe NE jet currently terminates at a recessed hot spotlocated in-between the two broad peaks seen in Fig. 1.The post-hot spot lobe material in this NE lobe appearsto be distributed along the radio axis. The radio corecomponent, coincident with the host galaxy, appears tobe enveloped by this lobe plasma, at least in projection.Our new ATCA mosaic image at 1378 MHz, with highersensitivity to extended emission compared to previousimages, reveals low surface-brightness and extended lobematerial in the vicinity of the radio core. The boundaryof lobe plasma in parts of the NE lobe close to the core isnot sharp: the wider spacing in the logarithmic radio con-tours along the NW and SE boundaries indicate a relaxedstate for this lobe material. Additionally, the distributionof this cocoon plasma close to the core is not symmetricabout the radio axis: the lobe is more extended on theNW side.The SW radio lobe is markedly different in its radiostructure compared to the NE lobe. The SW lobe is dom-inated by a bright hot spot, which protrudes from the endof a low surface brightness lobe. There is a large emissiongap between the SW lobe and the core. A secondary peakof enhanced emission, also noted in previous MOST andATCA images, is observed on the eastern boundary ofthis SW lobe and recessed from the bright hot spot; thismay be the site of a past hot spot or the site where the c (cid:13) , 1–14 Safouris, Subrahmanyan, Bicknell & Saripalli current jet bends before terminating at the bright hotspot.The expansion and movement of the post-hot spotplasma in this SW lobe is extremely asymmetric aboutthe source axis: the flow appears directed to the NWand perpendicular to the radio axis. Our new 1378-MHzATCA mosaic image clearly reveals an extension to thelow surface brightness cocoon material, perpendicular tothe radio axis and in the NW direction. The boundaryof the SW lobe appears to be relatively sharply boundedalong the SE. Towards the NW, although the boundariesof the extended emission appear to be defined, the lobesurface brightness fades away more gradually into theplume-like extension.The 26 ′ angular size corresponds to a projected lin-ear size of 1.9 Mpc. We measure the total flux den-sity at 1378 MHz to be 3.86 Jy for the giant radiogalaxy, which is in close agreement with the 1472-MHzflux density measurement in Saripalli et al. (1994). Theimplied radio power is 4 × W Hz − . Consider-ing the absolute R-band magnitude of the host galaxy, M R = − . − −
454 is a high excitation radio galaxy — an opticalspectrum of the host galaxy shows strong narrow emis-sion lines (Bryant & Hunstead 2000). The host galaxiesof FR-II radio sources tend to have high excitation linesin their optical spectra (Hardcastle et al. 2007); in thisrespect the emission line properties of MRC B0319 − Our new 2368 MHz ATCA mosaic image of MRCB0319 −
454 is shown in Fig. 2. The image was made witha beam FWHM of 32 ′′ × ′′ at a position angle of 0 ◦ .The rms noise on the image, in the vicinity of the giantradio galaxy, is 0.15 mJy beam − and the image dynamicrange exceeds 1000:1. We measure the total flux densityat 2368 MHz to be 2.46 Jy. This is notably higher thanthe corresponding flux density measurement of 2.04 Jy inSaripalli et al. (1994) and implying that the earlier 13-cmband ATCA image may have missed a significant fractionof the radio emission.We find good agreement between the previous andnew ATCA images made in the 13-cm band. Notably,both show the same multiple hot spot complex at theend of the NE lobe. Our lower resolution image reveals4 peaks at the lobe end (see the inset in Fig. 2). Thebrightest and most compact of these is located at the verytip of the lobe, and in projection, is seen to protrude fromthe lobe end. Additionally, as would be expected from theincreased total flux density in the new images, our 13-cmband image presented here recovers some of the extendedlow-surface-brightness lobe emission in the vicinity of the radio core, which was not detected in the previous 13-cmATCA image.The extended structure of the giant radio galaxy asobserved in the 13-cm band image compares well withthat in the 20-cm band image; however, the lowest sur-face brightness features seen in the 20-cm band imageare barely imaged at 13 cm. For example, a slight flare inthe lowest contour along the NW edge of the NE lobe isobserved in our 13-cm band image: the asymmetry in theNE lobe in the vicinity of the core is not as prominent inthe 13-cm band compared to the 20-cm band. The 13-cmband image also hints at the extension of low-surface-brightness material in the SW lobe towards NW: onceagain, although this faint material is reliably detected inthe 20-cm band image, it is relatively fainter at 13-cm. Polarization images at 1378 and 2368 MHz were con-structed from CLEANed Stokes I, Q, and U images madewith a beam of FWHM 52 ′′ × ′′ at a P.A of 0 ◦ . Thepolarization position angle images at the two frequencieswere used to compute the distribution of rotation mea-sure (RM) over the giant radio source. Since the differ-ence in the orientations of the polarization vectors at 1378and 2368 MHz is observed to be small, we have assumedthat our RM estimates do not suffer from nπ ambigui-ties. The computed magnitudes of RM are small over theentire source: mean RM is 0 rad m − and the 1- σ scatteris less than 7 rad m − . It may also be noted here thatour low RM values are consistent with those expectedfor this line of sight through the Galaxy in the analysisof Simard-Normandin & Kronberg (1980). Additionally,within the errors, we do not observe any significant gra-dients in the RM distribution across the source.The distribution in the 1378 MHz polarized intensityis shown in Fig. 3. Overlaid are vectors showing the ob-served orientations of the projected electric field ( E -field)with bar lengths proportional to the fractional polariza-tion. The small scatter in RM over the source implies thatthe intrinsic position angles of the electric field are withinabout 10 ◦ of the observed angles. The polarized intensityimage shows that there are two regions of relatively in-tense polarized emission at the end of the NE lobe, whichcoincide with the broad peaks in total intensity. The frac-tional polarization in these regions is about 20% at ourresolution. A narrow channel of low polarized intensityand low fractional polarization runs between the two re-gions. Along this channel the polarization vectors sharplychange in position angle; therefore, the channel is likelya result of beam depolarization (Haverkorn et al. 2000).Another channel of low polarization is observed along theradio axis in the region mid-way between the radio coreand the end of the NE lobe. To each side of this channelthere are two rails of polarized emission where the frac-tional polarization is about 20%. The projected magneticfield lines, which are perpendicular to the displayed E -vectors, follow the total intensity contours over most ofthe NE lobe excepting the central parts. The magneticfield is circumferentially oriented along all of the bound-aries at the far end of the NE lobe, as well as along the c (cid:13) , 1–14 RC B0319 − edges of the lobe close to the core component, where en-hancement in the fractional polarization is also observed.The distribution of polarized emission over the SWlobe is very similar to the total intensity radio structure.There is a bright peak in polarized emission at the loca-tion of the bright hot spot and there are weaker peaksin polarized intensity recessed from the hot spot. Overthe extended low surface brightness regions of the SWlobe, the distribution of polarized emission is fairly uni-form and the fractional polarization is about 15%. Thefractional polarization is enhanced at the total intensityhot spot as well as at the lobe boundaries. The fractionalpolarization values are particularly high, and in the range30-50%, along the eastern and SW edges. We have col-lapsed our image in fractional polarization of the low-surface-brightness SW lobe—excluding the hot spot—along a direction perpendicular to the source axis. Theresulting profile (Fig. 4), shows that the fractional polar-ization is enhanced at the SE end, has a shallow minimumin the central regions of the lobe, then increases towardsthe NW. However, there is a second dip in the fractionalpolarization at the base of the extension observed in thislobe towards NW and within the extension the fractionalpolarization values increase again to about 30% at theend.We have computed the average depolarization ratio(DR; a ratio of the fractional polarization at 1378 MHzto that at 2368 MHz) over different regions of the giantsource. The ratio is about 1.0 at the bright ends of theNE and SW lobes. In the NE lobe, the depolarizationratio is about 0.9 in the central regions of the lobe awayfrom the hot spots and decreases to 0.75 in the vicinity ofthe radio core. In the SW lobe, the depolarization ratiois ≈ We have collated radio flux density measurements for thegiant radio source from the literature and these are tab-ulated in Table 2. The value at 408 MHz was computedfrom the 408-MHz all sky survey of Haslam et al. (1982);the quoted error reflects the uncertainty in the foregroundsubtraction. The total spectrum for the entire source is apower law with spectral index α ≈ − .
84 over the range408 MHz – 4850 MHz (the spectral index α is defined us-ing the relation S ν ∝ ν α ). The total spectra for the NEand SW lobes appear straight over the frequency range843–4850 MHz (where separate flux density estimates areavailable for the two lobes) with α = − .
76 for the SWand α = − .
86 for the NE lobe. We note that the 843-MHz flux density for the SW lobe (Jones 1989) is similarto the 1378 MHz value derived from our image: it is likelythat the MOST measurement is missing flux density. Atlower frequencies (80 - 160 MHz), we only have measure-ments of the flux density in the NE lobe. These are belowthe extrapolation of the high frequency power-law spec-trum. The measurements indicate that the NE lobe hasa flattening of the spectral index below 408 MHz; how-ever, it is possible that the low frequency MSH values are erroneous: Mills et al. (1960) note that the 80-MHzmeasurement is affected by sidelobes.We have computed the spectral index distributionover the giant radio source between 1378 and 2368 MHzusing images with beams of FWHM 52 ′′ × ′′ at P.Aof 0 ◦ . The resulting image is shown in Fig. 5; the imagehas been blanked where the pixel values in the individualimages are less than 4 times the rms noise.The distribution in spectral index shows a steepeningof the spectral index along the axis and towards the corein the case of the NE lobe, and in a direction transverseto the source axis and towards NW in the case of the SWlobe. In both cases the steepening is along the axes ofthe lobes. If we assume that older parts of the lobes havesteeper spectra, as a consequence of spectral aging, thenthe observed spectral gradients suggest a backflow in theNE lobe from the hotspots towards the core, and a trans-verse flow off axis in the case of the SW lobe. We havemade profiles of these observed spectral index gradientsby collapsing the spectral index distribution image of theNE lobe perpendicular to the source axis, and that ofthe SW lobe along the source axis. The profile along theNE lobe (Fig. 6) shows that the spectral index smoothlysteepens along the length of the lobe from the bright endtowards the core. The profile in the SW lobe (Fig. 7)shows an abrupt change in spectral index between thebulk of the SW lobe and the extension towards NW: astep in the spectral index is observed between these twofeatures. The mean spectral index preceding the step is − .
7, in the extension the mean value is − . The host galaxy, ESO 248-G10, is a luminous gi-ant elliptical galaxy with a warped central dust lane(Saripalli et al. 1994). Images of the parent galaxy at op-tical and near-infrared wavelengths are consistent with amodel in which the host elliptical is triaxial with the ra-dio axis along the minor axis (Bryant & Hunstead 2000).Based on the orientation of the dust-lane, Bryant et al.argue that the radio jet axis makes an angle of 65 ◦ to theline of sight: the jet to the NE is pointed away from uswhere as the SW jet is towards us.In the following sections, we present our 2dF spec-troscopy of objects in the neighbourhood of the host anduse these, along with archival 6dF data, to infer the dis-tribution of galaxies and large-scale structure in the en-vironment of the giant radio source. ◦ field multi-fibre spectroscopy We obtained optical spectra of objects in the vicinity ofMRC B0319 −
454 using the AAOmega instrument on theAnglo-Australian Telescope (AAT), and its predecessor,the Two degree Field (2dF) instrument. The 2dF facilityis a multi-object spectrograph, located at the prime fo-cus of the AAT, capable of measuring ≈
400 simultaneousspectra within a 2 ◦ -diameter field in a single observation(Lewis et al. 2002). The AAOmega instrument, which isthe successor to 2dF, is a dual-beam (blue and red arm) c (cid:13) , 1–14 Safouris, Subrahmanyan, Bicknell & Saripalli bench-mounted spectrograph that uses the 2dF fibre po-sitioner at the prime focus. It provides greater through-put, stability and resolution than the decomissioned 2dFspectrographs.Our targets for the multi-fibre spectroscopy weregalaxies with b J magnitudes in the range of 15.0–19.5selected from the SuperCOSMOS catalogue within a 2-degree field around the host galaxy. The selection was re-stricted to those objects identified to be galaxies (and notstars) in the catalogue. A total of 1033 candidate galaxieswere identified in the chosen magnitude range. However,approximately 200 objects in the resulting target list werepreviously observed in a 2dF survey of the Horologium-Reticulum supercluster, which fortuitously covered aspace-volume including MRC B0319 −
454 (Klamer et al.2004). We retained about 20 of these objects in our targetcatalogue as a consistency check, and omitted reobserv-ing the others.Two separate field configurations were prepared forthe robotic fibre positioner using the CONFIGURE rou-tine that is part of the 2dF user software. For this alloca-tion exercise, we gave priority to objects with b J magni-tudes close to that of the host galaxy. Specifically, objectswith 15 . < b J . . < b J . . < b J . .
05 - 0.10 is shown inFig. 8. The host galaxy, which has a measured redshiftof z = 0 . z = 0 .
066 -0.084. The sky distribution of galaxies in the two con-centrations is shown in Fig. 9. The contours in the plotshow the location of the radio galaxy and the open circlesdenote the positions of known galaxy clusters within thefield: these are S0345, A3111, A3112 and APMCC 369and their redshifts are z = 0 . ± .
004 of the hostredshift (all the galaxies in the redshift space concen-tration containing the host); small crosses mark the skypositions of galaxies in the larger concentration in thehigher redshift range. In both redshift ranges there ap-pears to be a deficit of galaxies to the NE side of theradio source.In order to identify galaxy concentrations, includ-ing any that might be associated with the host, ESO248-G10, we applied the Huchra & Geller (1982) friends-of-friends group finding algorithm to our galaxy redshiftdataset. In this technique, association is determined bythe projected spatial and velocity separations betweenindividual galaxies. In particular, a galaxy is a friend ofanother galaxy if it is within a limiting projected dis-tance D L of that galaxy, and if the magnitude of the ve-locity difference between the two galaxies is less than V L .Large values of D L and V L result in most of the galax-ies included into large associations comprising multiplegroups strung out in velocity space. Very small values re-sult in the identification of tight groups and sub-groupswith most galaxies excluded from the associations. Weexplored a range of parameters, systematically reducingthe parameter space from D L = 4 Mpc and V L = 1000km s − until the identified groups appeared as compactseparate features in plots of R.A. versus declination, R.A.versus velocity, and declination versus velocity. The cho-sen values, D L = 1 Mpc and V L = 400 km s − , resultedin the identification of the galaxy concentrations shownin Fig. 10. For clarity, isolated galaxies and groups withless than 5 members are not plotted in this figure. The c (cid:13) , 1–14 RC B0319 − host of the radio source is in a concentration with sevenidentified members. Two of the members appear close to-gether in SuperCOSMOS optical images of the field, andthere is a third galaxy, of unknown redshift, also locatedwithin 10 arcseconds of one of these on the R-band image.A second concentration, also consisting of seven identi-fied members in our redshift database, is located about40 ′ to the SW of the concentration around the host andat a similar redshift. All other concentrations identifiedby our friends-of-friends algorithm are offset from thesetwo in velocity space by > − . It is interestingto note that all the identified galaxy concentrations arelocated towards the SW side of the 2 ◦ field, only isolatedgalaxies occupy the sky region to the NE.The host galaxy, with apparent magnitude b J =16 .
10, appears to lie close to the centre of the associatedconcentration on the sky. The host galaxy is the bright-est member of this concentration, the second brightestgalaxy has a magnitude b J = 16 .
25 and the other mem-bers are fainter than b J = 17 .
2. The concentration is dis-tributed over a projected linear extent of about 1 Mpcon the sky, indicating that this might be a loose group.We measure a velocity dispersion of 196 km s − for thegalaxies identified with this concentration and also notethat the host lies close to the centre of the distributionin velocity space. The concentration to the SW is dis-tributed over a larger sky area and has a larger velocitydispersion, σ = 272 km s − , and has a mean velocitythat is 290 km s − higher than that of the concentrationaround the host. We have examined the ROSAT All SkySurvey archival data at the location of the 2 ◦ field anddo not find any X-ray emission associated with the hostgrouping or with the neighbouring grouping to the SW.However, there is low surface brightness extended emis-sion coincident with the position of the distant z = 0 . × ×
25 Mpc, which in-cluded the two concentrations of galaxies in the redshiftrange 0.060–0.066. Galaxies within ± . ◦ in R.A. anddeclination and with redshift offsets within − . < ∆ z < . n/ ¯ n ) in asky plane at the redshift of the host. The fractional over-density at the location of the host is 12.4. The peak frac-tional over-density within the cube is 15.8 and is at anoffset position, with respect to the host galaxy, of − − n/ ¯ n exceeding 6 within the cube are either associated withthe host group or with the SW group. The jets in MRC B0319 −
454 make an angle of 65 ◦ to the line of sight with the NE jet pointed away fromus. We have examined the fractional over-density valuesin the box, with smoothing radius 1.25 Mpc, along theinferred axis of the giant radio source. ∆ n/ ¯ n is fairlyconstant over the region of the NE lobe, and the value of∆ n/ ¯ n somewhat exceeds 10. The SW lobe is located out-side the extent of the concentration with which the hostgalaxy is associated, and obviously lies in a region of sig-nificantly lower galaxy number density: the galaxy over-density within the cube decreases along the path lengthof the SW jet and has values as low as ∆ n/ ¯ n = 2 in thevicinity of the SW hot spot. If the galaxies are a tracer ofgas, the local galaxy distribution constitutes evidence foran asymmetry in the gas density on the two sides of thehost; additionally, the distribution is evidence for a den-sity gradient in the ambient medium of the radio galaxy,between the radio core and the SW hot spot.We have also computed the gravitational field inthe vicinity of the radio galaxy following the methoddescribed by Subrahmanyan et al. (2008). Again, pecu-liar galaxy velocities have been ignored in this compu-tation and distances to galaxies along the line of sightare based on redshift values only. First, we estimated thetotal mass in the cube as the product of the box vol-ume, V , and mean matter density in the Universe, ρ m .Assuming that the galaxies trace this matter, an averagemass, M g = ( V × ρ m ) /N g was assigned to each galaxy,where N g = 37 is the total number of galaxies in thecube. This resulted in a mass of M g = 2 . × M ⊙ as-sociated with each galaxy. The acceleration vector at anylocation within the cube was then computed by summingthe contribution from each M g point mass. Nearby galax-ies were excluded from the sum, so that the computedgravitational field is that due to surrounding large-scalegalaxy structures, and not individual objects. Galaxiesoutside of the cube were also ignored. The resultant ac-celeration vector at the location of the host has compo-nents: g RA = − . × − m s − , g DEC = 4 . × − m s − , and g z = 1 . × − m s − . The magnitude ofthis acceleration vector is g s = 1 . × − m s − andthe component on the sky plane is directed NW at aP.A of − ◦ . Assuming the age of the Universe to be τ = 13 . g s τ = 82 km s − . Within the extent of the group,the magnitude of the gravitational acceleration increaseswith distance from the host and takes on values about afactor of two greater than that at the location of the hostgalaxy; outside the group the acceleration declines withdistance from the group. Within a radius of about 1 Mpcof the host galaxy, the acceleration appears to be dom-inated by the mass associated with the group of whichthe host is a member. At the location of the SW lobe thevector components are much smaller: g RA = 2 . × − m s − , g DEC = 6 . × − m s − , and g z = 5 . × − m s − ; this corresponds to a velocity of g s τ = 25 km s − and is roughly directed towards the concentration asso-ciated with the host galaxy. At the location of the SWhot spot and lobe, the gravity is not dominated by thesecond galaxy concentration to the SW. c (cid:13) , 1–14 Safouris, Subrahmanyan, Bicknell & Saripalli
We have used archival data from the 6dFGS to in-fer the distribution of galaxies in the vicinity of MRCB0319 −
454 on larger scales. We computed the 3-D spatialdistribution of galaxy number density using 6dF galax-ies within a cube of side 85 Mpc centred at the host. Atotal of 860 galaxies within ±
10 deg in R.A. and declina-tion and within ± .
01 of the redshift of the host galaxywere included in this analysis. In order to compute thegalaxy over-density, we have smoothed the data using atop-hat function with radius 6 Mpc. At this smoothingscale, the mean number density of galaxies in the cubeis 1.2 and the rms is 1.5. Figure 12 shows the fractionalover-densities (∆ n/ ¯ n ) in the sky plane at the redshift ofthe host galaxy. At the location of the host, the fractionalover-density is 3.9 with this smoothing scale. This plotshows that the host appears to be embedded within alarge-scale galaxy filament that is oriented in a NNE -SSW direction. The filament extends more than 60 Mpcin our plot and has a projected width of approximately15 Mpc, which is the width between points with frac-tional overdensity of two at this smoothing scale. Frac-tional over-density values in front of and behind the hostgalaxy reveal that the filament has an apparent depthof only 8 Mpc along the line of sight. However, the realphysical size, in redshift space, is likely to be larger thanthis because of the Kaiser effect in this filament that hasa low fractional overdensity: there may be an apparentcompression in the distribution of measured velocities dueto the infall, in comoving space, of galaxies towards thefilament centre along the line of sight (Kaiser 1987). Inaddition, the filament is observed to bend away from usin the northern parts of the cube and, conversely, bendand extend towards us in the southern parts. Within thegalaxy filament, the fractional over-density takes on val-ues up to about 9 with this smoothing scale. The hostgalaxy is offset from the centre of the filament, in red-shift space, by approximately 1.5 Mpc towards us. Thetotal extent of the radio source—about 1.9/sin(65 ◦ ) Mpc= 2.1 Mpc—is much smaller than the width of this large-scale filament of galaxies, and the entire radio source, in-cluding the galaxy concentration associated with the hostgalaxy, are embedded within and located in the centralparts of the filament. As projected on the sky, the radiojets are roughly in a direction parallel to the large-scalefilament; however, the inclination of the radio axis to theline of sight makes the SW lobe somewhat more distantfrom the axis of the large-scale filament.Using an identical technique to that described inthe previous section, we have computed the gravitationalfield resulting from the large-scale structure using the6dF survey galaxies in the 85 Mpc side cube. The av-erage mass assigned to each galaxy in this cube was M g = 2 . × M ⊙ . In this analysis galaxies within4.0 Mpc (almost 1 degree) of the location at which any ac-celeration vector was computed were ignored. Therefore,the resulting acceleration vector field is that due to thelarge-scale structure and not influenced by local galaxiesor concentrations. The resulting gravitational field, com-puted on the sky plane at the redshift of the host galaxy, is displayed in Fig. 13; the vector lengths represent thecomponent on the sky plane. The vector components atthe location of the host are g RA = − . × − m s − , g DEC = 2 . × − m s − , and g z = 5 . × − m s − .The magnitude of the acceleration is g s = 5 . × − ms − , most of which is directed along the line of sight awayfrom us, as would be expected from the inference madeabove that the host galaxy lies in front of the axis of thefilament. The corresponding velocity is g s τ = 253 km s − and on the sky plane the acceleration is directed towardsP.A. of − ◦ . The relatively small acceleration vector atthe location of the radio source is consistent with thefinding that the host galaxy is located embedded withinthe filament and not in the peripheral regions. − The NE lobe in MRC B0319 −
454 is aligned along theradio axis, whereas the SW lobe appears to be extendedin a direction perpendicular to the source axis. The con-trasting shapes and orientations taken by the two lobesare indicative of two very different flow histories. The to-tal intensity structure in the NE lobe is suggestive of amodel in which the jet axis remains fairly steady overmost of its length, but has significant jitter at the end.The broad peak at the northern end of the lobe was prob-ably the site of past hot spots, and the chain of hot spotsalong the other eastern rim, which protrude past the endof the lobe, the sites of current hot spots. Strong anddirected backflow is indicated along the source axis andtowards the core, which continues beyond the core. Thebackflowing cocoon material appears to bend away fromthe source axis and towards NW, filling and inflating arelatively relaxed and asymmetric bridge in the vicinityof the core.The gradual steepening of the spectral index distri-bution along the length of the NE lobe suggests thereare significant age gradients along the lobe axis: the ra-dio spectrum is steeper in the vicinity of the radio corewhere the aged electron population has had more timeto lose energy via synchrotron emission. If the depolar-ization (see Sect. 2.3) is a result of entrainment, which ishigher in aged cocoon material, then the DR distributionsupports this view.The radio structure in the SW lobe is suggestive ofa different flow history. There is a large emission gap be-tween the SW lobe and the radio core, and it appears thatthe movement of post-shocked plasma is to the NW andnot along the source axis. Alternative models in whichthe plume-like feature to the NW trace the path of pastjet termination points due to, for example, jet precession,or models invoking movement of the host galaxy with re-spect to the ambient IGM in which the jet material isdeposited are unlikely because they would be expected tomanifest as symmetric or inversion symmetric lobe dis-tortions on the two sides.Enhancements in fractional polarization as well asan orientation of the B-field parallel to boundaries maybe interpretated as arising from a compression of a mage-tized plasma, with a tangled field, at locations where the c (cid:13) , 1–14 RC B0319 − flow terminates on ram-pressure interaction with ambi-ent thermal plasma. The B-field orientation is transverseto the source axis in the bright SW hot spot, as might beexpected from the interaction between the jets and am-bient gas at the termination shock. The magnetic fieldlines generally follow the total intensity contours alongthe boundaries of the SW lobe indicating that the lobesare not relaxed but are compressed at the boundarieswhere the expansion is ram pressure limited. Away fromthe hot spot and in the central regions of the SW lobethe B-field is oriented along NW-SE indicative of a flowof the post hot spot material towards NW. The enhancedfractional polarization and circumferential B-field in theregions of the SW lobe just before the extension towardsNW, manifesting in the intermediate peak in the sliceprofile in Fig. 4, suggest that the flow is discontinuousacross the SW lobe and that the extension towards NWmight have a separate origin.A model that might be considered is one in whichthe post hot spot material inflates a lobe at the loca-tion of the hot spot, which subsequently bends by ≈ ◦ to be directed along the line of sight, and then bendsonce again into the plane of the sky and towards NWto form the extension. In this picture, we would be ob-serving new plasma that has freshly been accelerated atthe hot spot, and old plasma, which resides behind or infront of the new plasma, that in projection on the sky,appear to form one continuous structure oriented NW.An argument against such a model is that the total in-tensity image does not show an enhancement prior tothe extension towards NW. Also, it is difficult to explainwhy the lobe should suddenly bend in this way. One pos-sibility, is that the backflow is deflected by the thermalgaseous halo associated with the galaxy group of whichthe host is a member, in the same way that backflowsare deflected by large angles when they encounter galaxyhalos (Kraft et al. 2005).The total radio spectrum for the SW lobe is straightover the frequency range 1378-4850 MHz. It flattens at843 MHz, but as noted earlier, this is probably becausethis observation missed some of the flux density. Valuesof the two point spectral index, α , over the SW lobetake on values α ≈ − . α ≈ − . − . > B me = 0 . B MB = 0 .
45 nT; therefore, particleaging is dominated by inverse Compton losses due to themicrowave background radiation and not synchrotron ra-diation. This is typically so for giant radio sources, whichhave large expanded lobes and low equipartition mag-netic fields (Ishwara-Chandra & Saikia 1999).Breaks in the emission spectrum due to both syn- chrotron and inverse-Compton losses are expected to oc-cur at a frequency ν T = 1 . × B synch ( B + B ) t GHz, (1)where t (in Myr) is the spectral age or time since acceler-ation. We estimate a spectral age t < . × yr for thebulk of the SW lobe. The steeper two-point spectral in-dex, with α ≈ − .
3, in the extension to the SW lobesuggests an aged plasma. A plausible model for this ex-tension is that if the injection spectrum had an emissionspectral index α = − .
7, the population might have agedover a time exceeding > × yr while experiencingcontinuous injection of reaccelerated electrons—resultingin a spectral break at frequencies < × yr. There are a number of physical models that can relate theobserved extended emission structure of the radio sourceand the ambient environment. The environmental influ-ence may be in the form of the gravitational field of thelarge-scale distribution of matter, and the gas associatedwith the large-scale structure. We assume here that thegalaxies trace the mass as well as the gas. In the followingsub-sections, we consider some models that could poten-tially be relevant to the evolution of the radio structure.We examine their relevance to this case study, in an at-tempt to shed light on the physical processes leading tothe formation of the asymmetries in MRC B0319 − The morphology of the SW lobe is plume-like, directedNW and moving away from the radio axis. We model thislobe as a backflow from the hot spot that is initially di-rected towards the core, but is deflected off the source axisand towards the NW because of buoyancy. The buoyantforces are a result of a gravitational field whose directionis transverse to the source axis. We have shown previ-ously that the gravitational field on local scales is largelydirected towards the host galaxy group and not NW-SEas we envisage here. Nevertheless, the direction of thegravitational acceleration could plausibly be transverseto the jet axis if the separation (in line of sight distance)between the host and SW groups is closer than that in-ferred by attributing redshifts to distances. As noted inSect. 3.2, the velocity separation between the two galaxyconcentrations is 290 km s − . However, peculiar motionscould contribute significantly to this velocity difference,placing them closer in real space than one might inferfrom the redshifts alone. In this case, the gravitationalfield in the vicinity of the SW lobe would be dominatedby both the host and SW galaxy groups, and we wouldexpect the gravitational acceleration vector to be directedbetween these two mass concentrations, and roughly SE-NW.In the analysis below we assume that hydrostatic c (cid:13)000
7, the population might have agedover a time exceeding > × yr while experiencingcontinuous injection of reaccelerated electrons—resultingin a spectral break at frequencies < × yr. There are a number of physical models that can relate theobserved extended emission structure of the radio sourceand the ambient environment. The environmental influ-ence may be in the form of the gravitational field of thelarge-scale distribution of matter, and the gas associatedwith the large-scale structure. We assume here that thegalaxies trace the mass as well as the gas. In the followingsub-sections, we consider some models that could poten-tially be relevant to the evolution of the radio structure.We examine their relevance to this case study, in an at-tempt to shed light on the physical processes leading tothe formation of the asymmetries in MRC B0319 − The morphology of the SW lobe is plume-like, directedNW and moving away from the radio axis. We model thislobe as a backflow from the hot spot that is initially di-rected towards the core, but is deflected off the source axisand towards the NW because of buoyancy. The buoyantforces are a result of a gravitational field whose directionis transverse to the source axis. We have shown previ-ously that the gravitational field on local scales is largelydirected towards the host galaxy group and not NW-SEas we envisage here. Nevertheless, the direction of thegravitational acceleration could plausibly be transverseto the jet axis if the separation (in line of sight distance)between the host and SW groups is closer than that in-ferred by attributing redshifts to distances. As noted inSect. 3.2, the velocity separation between the two galaxyconcentrations is 290 km s − . However, peculiar motionscould contribute significantly to this velocity difference,placing them closer in real space than one might inferfrom the redshifts alone. In this case, the gravitationalfield in the vicinity of the SW lobe would be dominatedby both the host and SW galaxy groups, and we wouldexpect the gravitational acceleration vector to be directedbetween these two mass concentrations, and roughly SE-NW.In the analysis below we assume that hydrostatic c (cid:13)000 , 1–14 Safouris, Subrahmanyan, Bicknell & Saripalli equilibrium defines the pressure and density distribu-tions; heating and sweeping up of the ambient gas by theexpanding radio lobe is neglected. A possible in-fall of theradio source and local environment towards the centre ofthe large-scale filament is also ignored since peculiar ve-locities are unknown. We use the following notation: ρ and ρ ext are the mass densities of the jet backflow andexternal medium respectively, p and p ext are the pressuresin the lobe backflow and external medium respectively, v bf is the velocity of the backflow, R c is the radius ofcurvature in the backflow, and g is the gravitational fieldwhich causes an hydrostatic pressure gradient in the am-bient gas. We define a coordinate system such that theflow is in the x - y -plane, with the propagation of the jethot spot along the x -axis, as visualized in Fig. 14. Thevelocity of the hot spot, v hs , and velocity of the backflow, v bf , are both in the frame of the host galaxy.Following Worrall et al. (1995) (and Schatzmann(1978) in more detail), the dynamics of a buoyant jetbackflow in such a medium are descirbed by Euler’s equa-tion ∂ v bf ∂t + ( v bf · ∇ ) v bf = − ∇ pρ + g , (2)where the gravitational field, the external density andpressure of the hydrostatic are related by: g = ∇ p ext ρ ext . (3)We assume that at each location along the backflowinglobe, the lobe plasma is in pressure equilibrium with theexternal gas. Hence, the momentum equations for thebackflow become: ∂ v bf ∂t + ( v bf · ∇ ) v bf = −∇ p ext „ ρ − ρ ext « . (4)It therefore follows that the velocity, v bf , of the backflowand the radius of curvature, R c , are related by v R c = g ρ ext ρ „ − ρρ ext « . (5)We therefore have: v R c ≈ ρ ext ρ g (6)for ρ << ρ ext .Let us now consider the various parameters, whichenter into equation (6), namely the velocity of the back-flow, the density of the external medium, and the densityof the lobe material. We have already estimated g in thevicinity of the radio source (see Sect. 3.2); based on theprojected geometry of the SW lobe, we estimate the back-flowing SW lobe plasma to have a radius of curvature ofapproximately 250 kpc.Recent estimates of the backflow velocity in giantradio galaxies have indicated v bf = 0.03 - 0.04 c (e.g.Lara et al. 2000; Jamrozy et al. 2005). These estimatesare based on spectral aging techniques and rely on theargument, based on source morphologies, that the meanbackflow velocity is equal to the mean head advance ve-locity (i.e. h v bf i = h v hs i ). In Sect. 4, we estimated thatthe time elapsed since the particles in the SW lobe exten-sion were last accelerated is > × yr. These particles are separated from the current location of the jet termi-nation by a projected distance of 630 kpc. Therefore, weestimate a separation velocity (the lobe velocity with re-spect to the hot spot) of < . c . If we also assume equaladvance and backflow velocities, then v bf < . c ; theupper limit is similar to the values derived for other giantradio sources.We estimate the external gas density using ourknowledge of the 3-D galaxy distribution. Our analy-sis suggests that the source MRC B0319 −
454 does notreside in an extremely over-dense environment ( e.g. , agalaxy cluster); instead, we have shown that the hostgalaxy is a member of a loose group that is embed-ded within a large-scale galaxy filament. The large-scalestructure represents only moderate over-densities. There-fore, in terms of the mean baryon density, ρ b , we expectthat 1 < ρ ext /ρ b <
100 in the galaxy filament environ-ment of the SW lobe.In order to estimate the density of the lobe materialin MRC B0319 − a . The lowerand upper Lorentz factor cutoffs to this distribution are γ min and γ max respectively. In this case the internal massdensity of the lobe plasma is related to the energy density, u p , in the synchrotron emitting particles by ρ = m p m e c ( a − a − γ − u p , (7)where we have assumed that γ max ≫ γ min . We computethis expression, in conjunction with minimum energy as-sumptions, using minimum Lorentz factors of γ min = 10 and 10 , since values within this range have been inferredfor jets and lobes previously (see e.g. Blundell et al.2006; Worrall & Birkinshaw 2006). For the particle en-ergy density calculation, we adopt a maximum Lorentzfactor of γ max = 10 . For example, γ min = 10 , gives u p = 4 . × − J m − and ρ = 2 . × − kg m − ,while γ min = 10 , results in u p = 2 . × − J m − and ρ = 1 . × − kg m − . The implied lobe densi-ties, expressed as a fraction of the external gas density, κ = ρ/ρ ext , are in the range 10 − < κ < − .For comparison, an electron-positron jet with thesame γ min values would result in lobe densities that are afactor & lower than those above and a density ratioin the range 10 − < κ < − . For each plasma type, wenote that the lobe densities and corresponding κ -valuesare lower limits since they do not account for possibleentrainment of the relatively dense IGM.Any entrainment of IGM gas into the cocoon of theradio galaxy would pollute the lobe cavity and increase itsdensity. This quantity, however, would remain less thanthat in the IGM so that κ <
1. The actual amount of en-trainment might depend on the source history, the powerof the jets and the environment in which the source is sit-uated. We have examined the polarization properties overthe bulk of the SW lobe, in an attempt to shed light on c (cid:13) , 1–14 RC B0319 − their thermal content, but find that these do not provideany useful limits on ρ . Rather, the measured polarizationindicates that the Faraday depth over the source is < < − for the SW lobe, which has an estimated depth of430 kpc and a minimum energy magnetic field 0.1 nT.If we allow for the presence of protons, the implied massdensity is < × − kg m − . At the upper limit, thisimplies κ > ρ ext values, andtherefore, is a poor constraint.The inner sources in restarting or double-double ra-dio galaxies are a useful probe of the plasma in whichthey are enveloped. Kaiser et al. (2000) examined a sam-ple of double-double restarting radio sources (with sizes & κ ∼ . − − total lobedensities are much greater, and at least a few percent ofthe IGM density, because of contaminating thermal gas.We therefore consider that κ could be as high as 0.01 ifMRC B0319 −
454 has entrained IGM gas during its evo-lution. In such a case, it is interesting to note that if thejets are electron-positron in composition, then the lobedensity would be completely dominated by the entrainedgas. For buoyancy to account for the movement of theSW lobe that has a density ρ = κρ ext , the gravitationalacceleration is required to be g = 7 × − κ „ v bf − « „ R c
250 kpc « − m s − , (8)for our adopted values of R c and v bf . Thus for a lobe with κ = 0 .
01, it is necessary that g ≈ × − m s − . Fora lobe with no entrainment and κ = 10 − , the requiredacceleration is g = 7 × − m s − .In Fig. 15 we have plotted the value of the gravita-tional field that is necessary to deflect the lobe backflowfor a range of κ values (solid lines). Estimates of the mag-nitude of the gravitational field in the vicinity of the SWradio lobe based on the local and large-scale galaxy dis-tributions are displayed with dotted lines. We also plotas a function of external gas density, the field that is re-quired to buoyantly move the lobe if there has been noentrainment and γ min = 10 (dashed line) or 10 (dot-dashed line). In this latter calculation, we have expressedthe external density as a fraction of the mean baryondensity of the Universe, ρ b , which is the baryon densityparameter Ω Baryon times the critical density ρ critical .Figure 15 shows that if the amount of entrained ther-mal gas in the cocoon is negligible, and therefore thecocoon is relatively light (10 − < κ < − ), the grav-itational field required to deflect the lobe via buoyancyencompasses the range of what we estimate at the SWlobe ( g ≈ × − - 6 × − m s − ), due to the sur-rounding mass distribution. From the above analysis, weconclude that the off-axis evolution in the SW lobe mayresult from buoyancy forces in the backflow, if there is minimal entrainment. Alternatively, this may also occurif the entrained gas is confined to a narrow boundarylayer along the edge of the lobe. The external densitiesimplied by the buoyancy model are in good agreementwith those expected in the galaxy filament environmentof the giant radio source. They represent moderate over-densities in the gas distribution and are therefore withinthose expected for the warm-hot phase of the IGM (e.g.Bregman 2007). The external pressures signified by themodel also hint at an ambient gas with warm-hot tem-peratures. Pressure equilibrium between the lowest sur-face brightness regions of the SW lobe and the exter-nal gas implies an IGM pressure of 1 × − N m − inthe vicinity of the flow. For the considered range of ex-ternal densities, this equates to an IGM temperature inthe range 2 × (10 − ) K. At the lower end (corre-sponding to higher external densities), the gas tempera-ture is well matched to that of the WHIM (i.e. 10 − K). Therefore, a buoyant movement of the lobe plasmawould support our initial hypothesis that the lobes ofMRC B0319 −
454 are interacting with the warm-hot gasphase of the IGM.The above interpretation is an appealing one sinceit signifies evidence of an interaction between the radiosource and WHIM. Nevertheless, we consider below fur-ther mechanisms which may drive the off-axis movementof the SW lobe plasma.
We next consider a variant of the above model, where theSW lobe extension is treated as a bubble rather than aback flow. The synchrotron bubble is embedded withina thermal gaseous medium and rises against the gravityvector. In this case, the dynamics are determined by thebalance between the buoyant and drag forces which actupon the bubble in the IGM (e.g. Gull & Northover 1973;Br¨uggen & Kaiser 2001). Equating these forces implies aterminal velocity for the bubble of v T = r gVAC D (9)where V is the volume of the rising bubble, A is thearea of its cross-section, and C D is the drag coefficient.We approximate the displaced SW lobe as a sphere with V /A = 130 kpc and take C D = 0 .
75 (Churazov et al.2001). Adopting a gravitational field g = 5 . × − ms − , as estimated at the location of the host galaxy in thelarge-scale galaxy filament, we calculate a terminal veloc-ity v T = 80 km s − for the lobe. At this speed, it wouldtake more than 2 × yr for the lobe plasma to movethe observed 200 kpc distance off-axis. This time-scaleis grossly longer than the estimated spectral ages ( ∼ yr) for different components of the source —including theextension at the NW end of the SW lobe —and signifiesthat buoyant forces are unlikely to be responsible for themovement of the lobe if it is a relict bubble. More gen-erally, this analysis suggests that buoyant forces actingon bubbles of synchrotron lobe plasma in the IGM areunimportant unless the bubbles have significantly largerradii and are in gravitational fields that are at least anorder in magnitude or more greater than what we have c (cid:13) , 1–14 Safouris, Subrahmanyan, Bicknell & Saripalli estimated for MRC B0319 − The most striking feature in the morphology of the giantradio source MRC B0319 −
454 is the side-to-side asym-metry in the lobe extents from the core. The NE lobeextends over a deprojected distance of 640 kpc from thecore, while the end of SW lobe is at a much greater dis-tance of 1430 kpc. Light-travel time effects are one pos-sible mechanism for causing such asymmetries; however,as noted in Saripalli et al. (1994), these effects would besmall for the range of velocities that the ends of the lobesmight take. A better explanation relates to correspondingasymmetries in the IGM. In this section, we examine theambient gas distribution for such asymmetries. Again, itis assumed that the galaxies trace the matter.The galaxy distribution within the host galaxy groupsuggests that there might be a gas density gradient aboutthe host. Specifically, the host galaxy appears to lie atthe center of the group. The most luminous members ofthe group are located to the north of the host, with theexception of the second brightest member, which resideswest of the host. Since low luminosity X-ray groups showevidence for irregular distributions in the X-ray emittingintra-group gas, with the emission preferentially biasedto the luminous galaxies in the group (Mulchaey 2000),we might expect there to be less thermal gas south of theradio core in this group.Based on the asymmetry in the lobe extents from thecore, we infer that the ends of the SW and NE lobes haveadvanced with a velocity ratio of 2:1. Assuming that theadvance of the leading lobe edge is ram pressure limitedby the density of the external gas, we expect that the ex-ternal density is 4 times larger on the NE side of the corecompared to that on the SW side. In Sect. 3.2, we notedthat there was evidence in the surrounding galaxy distri-bution for a density gradient in the IGM along the radiojet axis of the source (see also Fig. 11). At a smooth-ing scale of 1.25 Mpc, the fractional over-density in thegalaxy counts is about 10 in the region of the NE lobeand decreases along the axis of the southern jet to ap-proximately 2 in the vicinity of the SW hot spot. As-suming that the galaxies trace the gas, this gradient inthe fractional density contrast, together with the derivedmean galaxy number density of 0.3, is consistent with theside-to-side length asymmetry in this source. Again, thissuggests that the length asymmetry in MRC B0319 − − n/ ¯ n ≈
10) inthe region of the NE lobe. Further examination of thefractional over-density in the vicinity of the radio source,at a smoothing scale of 1.25 Mpc, shows no evidence fordensity gradients within the region of the group occupiedby the NE lobe. Slice profiles along the length of the NElobe, along the line of sight through the NE lobe, andtransverse to the axis of the NE lobe, show that the frac-tional overdensities are a constant and a little over 10 ineach of these directions. The fractional over-density val-ues do decline at the periphery of the group. However,the NE lobe appears to be located close to the centre ofthe galaxy density distribution associated with the group.The uniform density in the region occupied by the NElobe is consistent with the axially symmetric backflow ofthe hot spot plasma along the radio axis in this lobe.The accompanying high fractional over-density values inthe region of the galaxy group, along with the radio mor-phology of the source, suggest that the relatively denserintra-group medium confines the NE lobe and stalls anyexpansion losses. This has resulted in a relatively brightand luminous NE lobe.A similar analysis of the fractional over-density val-ues in the region of the SW lobe suggests a different sce-nario. As mentioned already, there is a decline in thefractional over-density along the jet axis from the radiocore to the lobe end. In addition, we find that there is agradient in the over-density along a line of sight throughthe SW lobe: the fractional over-density is about a factorof 2 larger behind the lobe compared to that in front, andat the location of the lobe. Also, as might be expectedfrom the radio morphology, there exists a gradient in theover-density that is transverse to the radio jet axis andoriented SE to NW. Fractional over-density values alongthis direction show that there is more galaxy density onthe SE side of the source compared to the NW side. Thisgradient may be viewed as arising from the combined ef-fects of the host group and neighbouring galaxy concen-tration. The numbers we derive suggest that the externaldensities are a factor of 2 or more higher on the SE sideof the lobe relative to the NW side.Such a density gradient may have caused the asym-metric expansion of the SW lobe plasma in this direc-tion. The plasma on the NW side of the radio jet axisextends about three times further from the jet axis, thanthe plasma on the SE side. The plasma may be expand-ing on the two sides with a velocity ratio of 3:1. If thisside-ways expansion is ram pressure limited, then we ex-pect the external density on the SE side of the axis to bea factor 9 larger than that on the NW side. Our analysisof the fractional density contrast indicates that the gasdensities on the two sides of the SW lobe are in the ra-tio 2:1 when examined with a smoothing scale 1.25 Mpc;but these could be larger on smaller smoothing scales. Al-though the density gradient is in the direction expectedfor this model, the local galaxy number density ( < . c (cid:13) , 1–14 RC B0319 − In particular, the decreased ambient density to theNW of the SW lobe, may have caused the synchrotronplasma to break confinement along this edge. In this case,the low surface brightness extension and its relativelysteep spectral index distribution, could be attributed toexpansion losses, provided that the electron energy spec-trum is curved, or has a high frequency break that hasshifted to the observing range.Finally, we note that the extent of the NE lobe fromthe core is similar to the extent of the detected SW jetfrom the core. The gas associated with the group may bedistributed to equal extents about the host (although asnoted earlier, we expect more gas to the north based onthe locations of the most luminous galaxies). The gaseousenvelope, which likely confines the NE lobe, might alsobe responsible for the confinement and detectability ofthe SW jet in this region.To summarize, the qualitative correspondance be-tween the density distribution and the radio source mor-phology is suggestive of a physical origin for the structureand asymmetries in the interaction between the radio jetplasma and ambient gas inhomogenieties: a likely mech-anism is ram pressure limitation of the lobe advance andexpansion by the gas density. In this picture, the asym-metries are caused by the locations and orientations ofthe radio lobes with respect to the gaseous media asso-ciated with the host galaxy group and the neighbouringgalaxy concentration south of the host.
We have presented new high sensitivity radio imagesof the powerful radio galaxy MRC B0319 −
454 togetherwith observations of the surrounding large-scale struc-ture. Using these we have sought to understand the in-fluence of the ambient gas and gravity on the evolutionof the jets and post hot spot plasma in this remark-able source. Our observations are qualitatively consis-tent with a model wherein the galaxies trace inhomo-geneities in the intergalactic gas, and these in turn deter-mine the evolution of the giant radio source. In this re-spect, our study has reached similar conclusions to thoseof Subrahmanyan et al. (2008), who also showed thatasymmetries in the morphology of the giant relict source,MSH J0505 − − The Australia Telescope Compact Array is part of theAustralia Telescope, which is funded by the Common-wealth of Australia for operation as a National Facilitymanaged by CSIRO. This research has made use of theNASA/IPAC Extragalactic Database, which is operatedby the Jet Propulsion Laboratory, California Instituteof Technology, under contract with the National Aero-nautics and Space Administration. We thank the Anglo-Australian Observatory for data obtained with the 2dFduring service time and the AAOmega instrument dur-ing the Science Verification program. We acknowledge theuse of the RUNZ code written by Will J. Sutherland. Weare grateful to Professor R. Ekers for useful discussionsand comments on this manuscript. We thank the anony-mous referee for helpful comments and suggestions.
REFERENCES
Begelman M. C., Blandford R. D., Rees M. J., 1984,Reviews of Modern Physics, 56, 255Bˆırzan L., Rafferty D. A., McNamara B. R., WiseM. W., Nulsen P. E. J., 2004, ApJ, 607, 800Bliton M., Rizza E., Burns J. O., Owen F. N., LedlowM. J., 1998, MNRAS, 301, 609Blundell K. M., Fabian A. C., Crawford C. S., ErlundM. C., Celotti A., 2006, ApJ, 644, L13Boehringer H., Voges W., Fabian A. C., Edge A. C.,Neumann D. M., 1993, MNRAS, 264, L25Bregman J. N., 2007, ARA&A, 45, 221Br¨uggen M., Kaiser C. R., 2001, MNRAS, 325, 676Bryant J. J., Hunstead R. W., 2000, ApJ, 545, 216Cen R., Ostriker J. P., 1999, ApJ, 514, 1Cen R., Ostriker J. P., 2006, ApJ, 650, 560Churazov E., Br¨uggen M., Kaiser C. R., B¨ohringer H.,Forman W., 2001, ApJ, 554, 261Dav´e R., Cen R., Ostriker J. P., Bryan G. L., HernquistL., Katz N., Weinberg D. H., Norman M. L., O’SheaB., 2001, ApJ, 552, 473Douglass E. M., Blanton E. L., Clarke T. E., SarazinC. L., Wise M., 2008, ApJ, 673, 763 c (cid:13) , 1–14 Safouris, Subrahmanyan, Bicknell & Saripalli
Fleenor M. C., Rose J. A., Christiansen W. A., Hun-stead R. W., Johnston-Hollitt M., Drinkwater M. J.,Saunders W., 2005, AJ, 130, 957Fleenor M. C., Rose J. A., Christiansen W. A.,Johnston-Hollitt M., Hunstead R. W., DrinkwaterM. J., Saunders W., 2006, AJ, 131, 1280Gull S. F., Northover K. J. E., 1973, Nature, 244, 80Hardcastle M. J., Evans D. A., Croston J. H., 2007,MNRAS, 376, 1849Hardcastle M. J., Sakelliou I., Worrall D. M., 2005, MN-RAS, 359, 1007Haslam C. G. T., Salter C. J., Stoffel H., Wilson W. E.,1982, A&AS, 47, 1Haverkorn M., Katgert P., de Bruyn A. G., 2000, A&A,356, L13Huchra J. P., Geller M. J., 1982, ApJ, 257, 423Ishwara-Chandra C. H., Saikia D. J., 1999, MNRAS,309, 100Jamrozy M., Klein U., Machalski J., Mack K.-H., 2004,in M´ujica R., Maiolino R., eds, Multiwavelength AGNSurveys Large-Scale Radio Structure in the Universe:Giant Radio Galaxies. p. 431Jamrozy M., Machalski J., Mack K.-H., Klein U., 2005,A&A, 433, 467Jones D. H., Saunders W., Colless M., Read M. A.,Parker Q. A., Watson F. G., Campbell L. A., BurkeyD., Mauch T., The 6dFGS team 2004, MNRAS, 355,747Jones D. H., Saunders W., Read M., Colless M., 2005,Publications of the Astronomical Society of Australia,22, 277Jones P. A., 1989, Proceedings of the Astronomical So-ciety of Australia, 8, 81Kaiser C. R., Schoenmakers A. P., R¨ottgering H. J. A.,2000, MNRAS, 315, 381Kaiser N., 1987, MNRAS, 227, 1Klamer I., Subrahmanyan R., Hunstead R. W., 2004,MNRAS, 351, 101Kraft R. P., Hardcastle M. J., Worrall D. M., MurrayS. S., 2005, ApJ, 622, 149Lara L., Mack K.-H., Lacy M., Klein U., Cotton W. D.,Feretti L., Giovannini G., Murgia M., 2000, A&A, 356,63Ledlow M. J., Owen F. N., 1996, AJ, 112, 9Lewis I. J., Cannon R. D., Taylor K., Glazebrook K.,Bailey J. A., Baldry I. K., Barton J. R., Bridges T. J.,Dalton G. B. e., 2002, MNRAS, 333, 279McNamara B. R., Wise M., Nulsen P. E. J., David L. P.,Sarazin C. L., Bautz M., Markevitch M., Vikhlinin A.,Forman W. R., Jones C., Harris D. E., 2000, ApJ, 534,L135Mills B. Y., Slee O. B., Hill E. R., 1960, AustralianJournal of Physics, 13, 676Mulchaey J. S., 2000, ARA&A, 38, 289Nulsen P. E. J., David L. P., McNamara B. R., JonesC., Forman W. R., Wise M., 2002, ApJ, 568, 163Safouris V., Subrahmanyan R., Bicknell G. V., SaripalliL., 2008, MNRAS, 385, 2117Saripalli L., Subrahmanyan R., Hunstead R. W., 1994,MNRAS, 269, 37Schatzmann M., 1978, Zeitschrift Angewandte Mathe-matik und Physik, 29, 608 Scheuer P. A. G., 1974, MNRAS, 166, 513Simard-Normandin M., Kronberg P. P., 1980, ApJ, 242,74Slee O. B., 1995, Australian Journal of Physics, 48, 143Subrahmanyan R., Saripalli L., Safouris V., HunsteadR. W., 2008, ApJ, 677, 63Worrall D. M., Birkinshaw M., 2006, in Alloin D., ed.,Physics of Active Galactic Nuclei at all Scales Vol. 693of Lecture Notes in Physics, Berlin Springer Verlag,Multiwavelength Evidence of the Physical Processes inRadio Jets. p. 39Worrall D. M., Birkinshaw M., Cameron R. A., 1995,ApJ, 449, 93Wright A. E., Griffith M. R., Burke B. F., Ekers R. D.,1994, ApJS, 91, 111 c (cid:13) , 1–14 RC B0319 − This paper has been typeset from a TEX/ L A TEX file pre-pared by the author. c (cid:13) , 1–14 Safouris, Subrahmanyan, Bicknell & Saripalli
Table 1.
Journal of ATCA observationsArray Frequencies Date Duration(MHz) (hr)EW367 1378/2368 2003 Sep 02 12750B 1378/2368 2003 Sep 18 12EW352 1378/2368 2003 Oct 02 121.5D 1378/2368 2003 Nov 16 121.5A 1378/2368 2004 Apr 02 12
Radio core / Host galaxy
Figure 1.
New ATCA 1378-MHz mosaic image of MRC B0319 −
454 made with a beam FWHM 52 ′′ × ′′ at a P.A of 0 deg.Contours are at -1, 1, 2, 4, 8, 16, 32, 64, 128 and 256 mJy beam − . Grey scales are shown in the range 1-100 mJy beam − usinga linear scale. The rms noise in the image is 0.25 mJy beam − . The half-power size of the synthesized beam is displayed in thebottom right-hand corner. The radio core, which coincides with the location of the host galaxy, is labeled. This image, as well asall others displayed herein, has been corrected for the attenuation due to the primary beam of each pointing.c (cid:13) , 1–14 RC B0319 − Figure 2.
New ATCA 2368-MHz mosaic image of 0319 −
454 made with a beam FWHM 32 ′′ × ′′ at a P.A of 0 ◦ . Contours are at-1, -0.5, 0.5, 1, 2, 4, 8, 16, 32, 48, 64 and 128 mJy beam − . Grey scales are shown in the range − − using a linearscale. The rms noise in the image is 0.15 mJy beam − . The half-power size of the synthesized beam is displayed in the bottomright-hand corner. The inset shows the hot spot complex at the end of the NE lobe. Contours are at -1, -0.5, 0.5, 1, 2, 4, 8, 16, 32,48, 53, 60, 64 and 128 mJy beam − . Grey scales are shown in the range − − using a linear scale. Table 2.
Radio flux densities of MRC B0319 − . ± . . ± . . ± . . ± . . ± (cid:13) , 1–14 Safouris, Subrahmanyan, Bicknell & Saripalli (a)
Direction of profile(b)
Figure 3.
Distribution of the 1378-MHz polarized intensity over the NE lobe (panel a) and SW lobe (panel b) as observed witha beam of FWHM 52 ′′ × ′′ at a P.A. of 0 ◦ . Electric field vectors are displayed with lengths proportional to the fractionalpolarization at 22-cm wavelength. The length of the vectors shown in the top right-hand corner corresponds to 100% polarization.Contours show the 1378-MHz total intensity at -1, 1, 2, 4, 8, 16, 32, 64, 128 and 256 1 mJy beam − . The half-power size ofthe synthesized beam is displayed in the top right-hand corner. The grey-sacle is in the range 0–30 mJy beam − and 0–7 mJybeam − in the NE and SW lobe panels respectively. The arrow shows the direction of the mean polarization profile displayed inthe following figure. c (cid:13) , 1–14 RC B0319 − Figure 4.
Profile of fractional polarization in the SW lobe. The direction of the profile is indicated in the previous figure.
Direction of profile in Fig. 6Direction of profilein Fig. 7
Figure 5.
Distribution of spectral index over the giant source as computed from images at 2368 and 1378-MHz, made with beamsof FWHM 52 ′′ × ′′ at a P.A of 0 ◦ . Contours of the 1378-MHz total intensity at -1, 1, 2, 4, 8, 16, 32, 64, 128 and 256 mJy beam − are overlaid. The spectral index is shown in grey-scales in the range 0 to − − using a linear scale. The arrows indicatethe directions of the spectral index profiles displayed in the following figures.c (cid:13) , 1–14 Safouris, Subrahmanyan, Bicknell & Saripalli α Figure 6.
Mean profile of spectral index along the NE lobe. α Figure 7.
Mean profile of spectral index across the SW lobe. c (cid:13) , 1–14
RC B0319 − host galaxy Figure 8.
Redshift distribution of galaxies within the 2-degree field in the range z = 0.05 - 0.10. The host redshift is z = 0 . (cid:13) , 1–14 Safouris, Subrahmanyan, Bicknell & Saripalli
A3112APMCC 369S0345 A3111
Figure 9.
Spatial locations of galaxies in the range z = 0 . z = 0 . (cid:13) , 1–14 RC B0319 − host grouphost group Figure 10.
Locations of galaxy groups with 5 or more members in R.A-velocity space (top panel) and declination-velocity space(bottom panel). R.A and declination are plotted relative to the host galaxy. Each group has been assigned a different symboland/or colour.c (cid:13) , 1–14 Safouris, Subrahmanyan, Bicknell & Saripalli (a)(b)
Figure 11.
Fractional over-densities in the vicinity of the radio source. Contour levels are shown at ∆ n/ ¯ n = 2, 4, and 8. Panel(a) is in a sky plane at the location of the host galaxy, whereas panel (b) is off-set in line of sight distance by 3.8 Mpc. Thefractional over-densities displayed in panel (b) are behind the host galaxy. R.A and declinations are off-set relative to the hostgalaxy position. c (cid:13) , 1–14 RC B0319 − Figure 12.
Numerical values of fractional galaxy over-density in a 20 deg field around the host. The data have been smoothedwith a top-hat function of radius 6 Mpc. R.A and declination are off-set relative to the position of the host.c (cid:13) , 1–14 Safouris, Subrahmanyan, Bicknell & Saripalli
Figure 13.
Gravitational acceleration vectors at the redshift of the host galaxy projected onto the sky. Blue vectors (with crosses)are directed out of the page, while red vectors (with circles) point into the page. c (cid:13) , 1–14
RC B0319 − hs xy υυ Figure 14.
Schematic of the buoyant jet backflow in the x - y -plane. Figure 15.
Gravitational acceleration as a function of external density, expressed in terms of the mean Baryon density in the localUniverse. The accelerations required to move thermally contaminated lobes of density κρ ext are shown with solid lines. Also shownare the gravitational accelerations required to buoyantly move a lobe with no thermal contamination and γ min = 10 (dashedline) or γ min = 10 (dot-dashed line). The lower and upper horizontal dotted lines show the the estimated magnitudes of thegravitational field at the location of the SW lobe measured from local and large-scale galaxy distributions.c (cid:13)000