Chandra observation of the multiple merger cluster Abell 521
a r X i v :a s t r o -p h /0508585v 2 15 S e p 2005
Astronomy &Astrophysics manuscript no.ferrari˙A521February 5,2008
(DOI:will be inserted by hand later)
Chandra observation of the multiple merger cluster Abell 521
Ferrari,C.1,Arnaud,M.2,Ettori,S.3,Maurogordato,S.4,and Rho,J.5
1Institut f¨u r Astrophysik,Technikerstra?e 25,6020Innsbruck,Austria
2
CEA /DSM /DAPNIA Service d’Astrophysique,CEA Saclay,L’Orme des Merisiers,B?a t.709,91191Gif-sur-Yvette,France
3Osservatorio Astronomico di Bologna,Istituto Nazionale di Astro?sica,via Ranzani 1,I-40127Bologna,Italy 4
Laboratoire Cassiop′e e,CNRS/UMR 6202,Observatoire de la C?o te d’Azur,BP 4229,06304Nice Cedex 4,France
5
Spitzer Science Center,California Institute of Technology,Pasadena,CA 911251
Received 1August 2005;accepted 8September 2005
Abstract.We present the Chandra analysis of the rich galaxy cluster Abell 521(z =0.247).The high resolution of the Chandra observation has allowed us to re?ne the original merging scenario proposed for A521,and to reveal new features in its X-ray emission.A521has strongly substructured ICM density and temperature maps.Its X-ray di?use emission is elongated along a NW/SE direction (SX2)and shows two major components,a main cluster and a northern group of galaxies.This latter is in turn substructured,showing a clump of cold and very dense gas centred on the Brightest Cluster Galaxy (BCG),and a northern tail aligned in the SX2direction.A compression of the X-ray isophotes is also observed South of the BCG.We conclude that the northern group is infalling onto the main cluster along the NW/SE direction.This hypothesis is corroborated by the presence of a hot bar in the ICM temperature map located between the southern and northern regions,as the gas could be compressionally heated due to the subclusters’collision.The hot region corresponds to the eastern part of an over-dense ridge of galaxies,along which it was originally suggested that a merging of subclusters has recently occurred along the line of sight.An alternative hypothesis about the origin of the hot central bar is that we could observe in projection the shock fronts due to this older cluster-cluster collision.However,the two hypothesis do not exclude each other.Two other structures possibly interacting with the main cluster are detected on the West and North-East sides of the BCG.We also uncover the presence of two northern edges in the ICM density,which could be due to the ongoing merging events observed in the central ?eld of the cluster,or even in its outer regions.A521is a spectacular example of a multiple merger cluster made up by several substructures converging at di?erent epochs towards the centre of the system.The very pertubed dynamical state of this cluster is also con?rmed by our discovery of a radio relic in its South-East region.
Key words.galaxies:clusters:general -galaxies:clusters:individual:Abell 521-X-rays:galaxies:clusters
1.Introduction
In the concordant cosmological model (ΛCDM,?m =0.3and ?Λ=0.7),small structures are the ?rst to form,and then they merge giving rise to more and more massive systems in a hierarchical way.Both numerical and obser-vational results show that galaxy clusters form and evolve through the merging of sub-clusters and groups of galax-ies along ?lamentary structures (e.g.West et al.1995;Bertschinger 1998;Durrett et al.1998;Arnaud et al.2000;Bardelli et al.2000;Borgani et al.2004;Adami et al.2005).
Combined optical and X-ray studies have been par-ticularly successful in revealing the dynamics of merg-ing clusters (Flores et al.2000;Henriksen et al.2000;
2Ferrari,C.et al.:Chandra observations of Abell521 tion,and to analyse the complex physical processes acting
during their evolution.
Abell521(z=0.247)is a relatively rich(R=1)cluster
of Bautz-Morgan Type III(Abell1958;Abell et al.1989).
After its?rst detection in X-ray with HEAO1(Johnson
et al.1983;Kowalski et al.1984),the dynamical state
of A521has been investigated in detail through a com-
bined X-ray and optical analysis(Arnaud et al.2000;
Maurogordato et al.2000;Ferrari et al.2003).A severe
segregation between the gas and galaxy distributions was
detected.ROSAT/HRI observations revealed the presence
of two peaks of X-ray emission,associated with a dif-
fuse main cluster and with a compact less massive group
(Arnaud et al.2000).Unlike what is usually observed in
relaxed systems,the Brightest Cluster Galaxy(BCG)is
not located at the barycentre of A521,but in the compact
sub-group,with a surprising o?-set from its X-ray peak.
The galaxy isodensity map in the central20’×20’?eld of
A521has a very irregular and strongly sub-clustered mor-
phology.Its general structure follows a NW/SE direction,
crossed by a perpendicular high density ridge of galaxies
in the core region(Arnaud et al.2000;Ferrari et al.2003).
The analysis of the dynamical and kinematic properties of
more than one hundred cluster members con?rmed that
A521is far from dynamical equilibrium:its radial veloc-
ity distribution signi?cantly di?erent from a Gaussian and
characterised by a very high dispersion(1325+145
?100km/s)
is typical of merging systems(Ferrari et al.2003).A de-tailed dynamical analysis revealed at least two di?erent and not contemporary episodes of merging:a)a dynam-ically bound complex of galaxies,hosting the BCG and corresponding to the compact group detected in X-ray,is currently infalling on the plane of the sky toward the cen-tre of the main cluster,and b)two or more sub-clusters have recently collided along the over-dense central ridge, with a collision axis nearly along the line of sight(Ferrari et al.2003).
The recent analysis by Umeda et al.(2004)of A521Hαluminosity function showed that this cluster contains more currently star-forming galaxies than local clusters,consis-tently with the observed Butcher-Oemler e?ect.The ex-cess of star formation(SF)can be at least partly related to the particular dynamical state of A521,since an increase of SF has been observed in several merging systems(e.g. Gavazzi et al.2003;Poggianti et al.2004;Ferrari et al. 2005).
The complex dynamical state of A521and its unique morphological features motivated our Chandra observa-tions with the aim of better characterising the physics of this exceptional cluster.In this paper the Chandra data are analysed.Sect.2brie?y describes the observa-tions and the data reduction.In Sect.3we study the X-ray morphology and the temperature structure of A521. Results are discussed in Sect.4and summarised in Sect.5. As in Ferrari et al.(2003),all numbers are expressed as a function of h75,the Hubble constant in units of75 km s?1Mpc?1.We have used theΛCDM model with ?m=0.3and?Λ=0.7,thus1arcmin corresponds to ～0.217h75?1Mpc in the following.
2.Observations and data reduction
A521was observed with Chandra ACIS-I and ACIS-S in”VFAINT”mode.The datasets were processed and cleaned using CIAO3.2software and calibration?les in CALDB3.0.0.The?rst exposure was done on Dec23, 1999with ACIS-I and focal plane temperature of?110o for an e?ective exposure time of38.0ksec after standard cleaning(88%of the nominal exposure time).On Oct13, 2000,a second exposure of41ksec was done with ACIS-S and focal plane temperature of?120o.After cleaning the light curve from the several?ares present by requiring a mean count rate of0.085cts/s,an exposure of18.4ksec is obtained.
3.Results
3.1.X-ray morphology
The raw image of A521di?use emission is presented in Fig.1.The cluster shows two highest density regions that we will call clumps A and B in the following.They corre-spond respectively to the Northern group and the central part of the main cluster identi?ed in the ROSAT image (Arnaud et al.2000).The new Chandra observations un-cover the presence of several features inside and around each of the two clumps,as it appears more clearly in Fig.2,which represents a smoothed image of the cluster central?eld(7×7arcmin2)in the0.5-5keV energy band. In order to get a smoothed image of the di?use emission of the cluster,the programme“mrp
Ferrari,C.et al.:Chandra observations of Abell521
3 Fig.1.0.4-5keV raw image of A521di?use emission(ACIS-S3?eld)derived by summing the ACIS-I and ACIS-S observations.In the right panel the image has been smoothed with a Gaussian ofσ=2arcsec,and two possible arc-like discontinuities are indicated in red.The ACIS-I CCD gaps are evident in the East and South sides.The whole cluster region and its two main clumps A and B are also indicated.
and c)the general cluster structure out to～5h?1Mpc (Plionis et al.2003).A slight misalignment is present be-tween the main axis of the X-ray and optical emission (SX2and S2).
The X-ray peak of the main cluster(labelled X in Fig.3)is close to the second brightest cluster galaxy and its position corresponds to the barycentre of the optical emission of A521.South of SX1,the axis perpendicular to SX2(see Fig.3),the cluster appears rather relaxed with quite regular isophotes(Figs.2and3).
The Northern part(North of SX1)is much more com-plex,as it shows evidence of several sub-structures and elongations in the ICM distribution.The most prominent structure is the clump A(see Figs.1and3).It is brighter than the equivalent region centred on the main cluster and roughly elliptical in shape,with a main axis along the SX2direction.Two internal sub-structures are detected,a northern clump and a southern one,labelled respectively A BCG in Fig.3.The brightest peak of clump A, located inside the A
BCG shows an elongation more toward North-West with respect to the SX2direction,being in fact more aligned with the main axis of the optical distribution and of the BCG(S2in Ferrari et al.2003,green line in Fig.3).A compression of the X-ray isophotes is also observed in the South of the BCG along the S2direction.The clump A
g in Fig.3).The observations of Ferrari et al.(2003)have shown that all of these blobs are at the same redshift of the cluster,and that they could be galaxies falling onto the brightest cluster object.A
N structure is centred on a bright object(a star,based on the spectral analysis by Ferrari et al.2003),but its X-ray emission is extended.The northern part of the whole substructure A (A
BCG.
Two other less prominent substructures are present in the North.First,a North-East clump which appears as an excess of emission east of clump A(NE ellipse in Fig.3). Second,we observe an elongation of the X-ray isocontours on the West side of the main cluster X-ray peak towards a sub-structure(labelled W in Fig.3)at～2.5arcmin from it in a North-West direction.Fig.3shows that several galax-ies are concentrated in the W region,i.e.8quite bright ob-jects(I AB=18.5-19)and several faint ones.Spectroscopic observations reveal that3of them are con?rmed cluster members,but note that in this region the spectroscopic
4Ferrari,C.et al.:Chandra observations of Abell
521
Fig.2.0.5-5keV image of A521di?use emission (7×7arcmin 2central ?eld).The image,exposure corrected and background subtracted,is the sum of ACIS-I and ACIS-S observations,and it has been ?ltered using a wavelet transform with a detection level of 3.7σ.The overlaid isocontours are logarithmically spaced by a factor of 0.1.observations are complete at no more than 50-60%level (Ferrari et al.2003).
The X-ray morphology of A521is therefore regular South of the SX1direction,strongly substructured in the northern part.This is even clearer in Fig.4,which shows the residuals obtained after subtracting in each region the emission from the symmetric region with respect to the X-ray peak of the main cluster (labelled X in Fig.3).This way we subtract in the North the corresponding ’unper-turbed’part of the main cluster,as measured in the South.Clumps A ,W and NE emerge clearly North of SX1(with a possible tail of X-ray emission towards South-East for the substructure W ).
Finally,Chandra observations revealed two sharp edges in the X-ray surface brightness in the region to the North of the BCG (labelled A1and A2in Fig.1),with changes by a factor of 2over scales shorter than 10arcsec (see Fig.5).To emphasise the departure from a smoothed spherically symmetric emission,we modelled the surface brightness of the region with two β?models,one obtained as best-?t results of the radial pro?le ex-tracted from the Northern (P.A.=[?90o ,90o ])semicircle centred on the BCG (RA,Dec)=(04:54:06.9,-10:13:20),the second one extracted from the Southern semicircle centred at (RA,Dec)=(04:54:07.9,-10:14:27).Then,we
compute the surface brightness pro?le in two strips cross-ing the two edges both in the (0.5?5)keV exposure-corrected image and in the faked two-dimensional emis-sion obtained by the sum of the two β?models.The faked pro?le is indicated by dashed lines in Fig.5.These edges are located orthogonal to the main X-ray axis SX2and trace the Northern boundary of the clump A (see Fig.1).
In summary,we distinguish the following main struc-tures in the X-ray observation of A521:
–clumps A and B :the two main clumps in the X-ray emission of A521.Clump B is the central part of the main cluster,which seems to be nearly unperturbed in the southern region.Clump A is a northern sub-cluster centred on the BCG;
–SX1,SX2,S2:SX2is the axis connecting the X-ray peaks of the main cluster and of the northern sub-cluster (with a position angle of 163?1),SX1its per-pendicular direction,S2the optical main elongation axis;
–A
N and A BCG is a centred on the
BCG galaxy and A
1
The position angle being de?ned from North to East.
Ferrari,C.et al.:Chandra observations of Abell5215
Fig.3.Top:image of the di?use X-ray emission of A521(0.5-5keV)-Bottom:deep,I-band optical image of the cluster obtained with CFH12k@CFHT.Contours of the X-ray di?use emission are overlaid(logarithmically spaced by a factor of0.1as in Fig.2).The main sub-structures detected in the ICM distribution are indicated by circles and ellipses:a)the highest density clumps A(Northern sub-cluster)and B(central part of the main cluster)b)the main sub-structures of clump A,centred on three X-ray peaks,A N(centred on a stellar object) and A
6Ferrari,C.et al.:Chandra observations of Abell
521
Fig.4.Image of the residuals of A521X-ray emis-sion:di?erence between the image of Fig.2and the same image rotated by180degrees around the X-ray peak of the main cluster(X in Fig.3).
3.2.Temperature analysis
3.2.1.Temperature map from hardness ratio
Fig.6shows a temperature map of A521obtained using the hardness ratio technique.Images of the region covered by the ACIS-S3?eld of view have been extracted in the energy bands0.5-2keV and2-5keV from the ACIS-S and ACIS-I event?les.Point sources have been detected and removed using the CIAO tools wavdetect and dmfilth. Each image has been background subtracted using blank ?eld data and corrected for vignetting e?ect and exposure variations.The resulting ACIS-S and ACIS-I count rate images in each energy band have then been added and adaptively smoothed using the CIAO tool csmooth.The smoothing scales were de?ned from the raw ACIS-(S+I) image in the2?5keV energy band with a minimum sig-ni?cance of4σand a maximum signi?cance of5σ.
The2-5keV smoothed image has been divided by the corresponding0.5-2keV image to obtain a hardness ratio map,which has been converted to a temperature map.The theoretical conversion factors have been com-puted using an absorbed thermal model(tbabs(mekal) in XSPEC11.3.1)with a column density?xed to the Galactic value of5.79×1020cm?2,a redshift of0.247and an abundance?xed to0.4,convolved with the instrument responses.Since the images have been corrected for vi-gnetting e?ects,we used the on-axis Auxiliary Response and Redistribution Matrix?les,obtained accordingly to the period and con?guration of the observation.An in-verse edge to account for the underestimate of the e?ec-tive area around2keV is also applied(see Vikhlinin et al.
2005).
Fig.5.Counts distribution corresponding to the re-gions of the two fronts detected at the North of clump A(points)and modelledβ-model from relaxed re-gions of the cluster(dashed line).The positions of the two arc-like discontinuities are indicated in grey.
A521is clearly characterised by a highly sub-structured temperature map,which presents:
-a cold region(T≤5keV)in the North-East(labelled NE
BCG).It is surrounded by an annulus of warmer gas(A
Ferrari,C.et al.:Chandra observations of Abell5217
-a gradual decrease of temperature in the South-West sector(Sect1and Sect2).The temperature gradient is less pronounced in the South-East sector,with a possible cold substructure South-East of the cluster barycentre(SE).
3.2.2.Spectral analysis
Spectra in di?erent regions of interest were extracted,to-gether with the corresponding Auxiliary Response and Redistribution Matrix?les.As above,an absorbed ther-mal model is used to?t the accumulated spectra in bins with a minimum number of20counts,with a column den-sity?xed to the Galactic value.An inverse edge to account for the underestimate of the e?ective area around2keV is also applied(see Vikhlinin et al.2005).
The temperatures,abundances,?uxes and rest-frame luminosities of the whole cluster(big circle in Fig.12) and of the two main clumps A and B(smaller circles in Fig.1)are listed in Table1.The quoted values have been derived using ACIS-I data,for which a local background can be estimated.They do not show signi?cant variation when blank?eld background are used,validating the use of blank?eld data for the temperature map.
The spectroscopic temperatures(T spec)of the di?er-ent regions identi?ed in the temperature map(see previ-ous section)have also been estimated.The ACIS-I and ACIS-S spectra were?tted simultaneously3.These tem-peratures are compared in Table2with the temperatures obtained using the hardness ratio technique.The latter (T HR)have been estimated using both the mean value from the temperature map and the ratio of the count rates in the smoothed images used to derive the temperature map,these two methods giving results which always di?er by less than0.1keV.
The temperatures derived from the spectral analysis are in good agreement,within1σerror,with the value derived from the hardness-ratio technique(see Table2), with no systematic di?erences.The largest discrepancy(～1.1σ)is observed for the region around the BCG(A
2The CCD gaps have been masked in doing the spectral analysis of the whole cluster.
3We checked that fully consistent results are obtained using the ACIS-I exposure only
Region T spec T HR
(keV)(keV)
T4.10±0.66 4.5
A
NBCG6.51±0.89 6.0
Ridge E5.26±1.75 6.6
Center7.44±1.197.4
Sect15.77±0.68 5.9
Sect24.77±0.66 4.6
SE4.70±0.88 5.3
BCG, Fig.3),centred on the BCG,which is the highest density region of the principal northern substructure(clump A). The very high angular resolution of Chandra has helped to solve one of the open questions of the previous X-ray analysis of A521(Arnaud et al.2000).In the ROSAT ob-servations a compact group in the North of the cluster was also detected,hosting the BCG.Curiously,it was not centred on its brightest galaxy,but on a northern posi-tion.Chandra observations show two main X-ray peaks in the subcluster A corresponding to the ROSAT group, one centred on a stellar object(A
BCG),which were not resolved by the previous X-ray observations.The A
8Ferrari,C.et al.:Chandra observations of Abell
521
Fig.6.Temperature map obtained through the hardness ratio technique(see text for details).The colours range from purple(～3.5keV)to red(≥8keV).Left:the isocontours of the X-ray di?use emission of A521are superposed (logarithmically spaced by a factor of0.1as in Fig.2).Right:the temperature isocontours and the corresponding k T values are overlaid in black.The regions used for the spectroscopic analysis are superposed in white.
Region T Z f X(0.5-2keV)f X(bol)L X(0.5-2keV)L X(bol)
(keV)(Z⊙)(10?12erg/s/cm2)(10?12erg/s/cm2)(1044erg/s)(1044erg/s) Table1.The results of spectral?tting in the whole cluster(circular region centred at(RA,Dec)=(04:54:07.9,-10:14:42.6)and with a radius of4arcmin)and in the two main clumps A and B(circular regions centred respectively at(RA,Dec)=(04:54:06.6,-10:12:56.3)and(RA,Dec)=(04:54:07.9,-10:14:42.6),and with radius of0.8and1.0arcmin). 1σerrors are given.N H has been?xed to the galactic value(5.79×1020cm?2).
ies(～0.5-1.5keV,Forman et al.1985),is associated to
the whole system of galaxies surrounding the BCG.Its
temperature,colder than in the rest of the cluster,is in
agreement with the lower radial velocity dispersion of this
region(256+82
?133kms?1in a circle of240h75?1kpc around
the brightest galaxy,Ferrari et al.2003).What we need to investigate is the origin of the high gas density in A
4The density ratio has been measured as the square root of the surface brightness ratio.The0.5-5keV surface bright-ness maps(ACIS-I and ACIS-S)have been used,considering the A
BCG group could therefore be due to stripped material related to the cannibalism of the BCG.Galaxy cannibalism occurs very early in a cluster lifetime.If it would happen after clus-ter virialization,we should expect randomly oriented cD galaxies,while it has been shown that the shape of cD’s
Ferrari,C.et al.:Chandra observations of Abell5219 aligns with its nearest neighbour,the cluster shape and
the?laments of large scale structure(West1994).The
scenario that the BCG of A521is becoming a cD galaxy
in a dynamically young group is therefore supported by
the observed alignment of the BCG main axis with the
cluster major axis and with the nearest cluster neighbour
(Plionis et al.2003).It is also in agreement with the idea
that cD’s form via the merging of galaxies in the centre of
poor groups,which then fall into richer clusters(Merritt
1984;Zabludo?&Mulchaey1998and references therein).
Notice that a similar case of a cold and very dense clump
of gas,detected around a BCG with multiple nuclei and
aligned along the merging axis of the cluster,has been ob-
served with Chandra in A3266(Henriksen&Tittley2002).
In that case,the BCG is probably cannibalising galaxies
from a merging subcluster.
In agreement with the previous optical analysis,we
therefore conclude that the main northern clump A is a
group of galaxies in interaction with the main cluster.The
subcluster shows a higher density component centred on
the BCG,a northern tail of gas,and compression of the
X-ray isophotes South of the BCG.All these results are
in agreement with the merging scenario suggested by our
optical analysis of the cluster(i.e.the group A is infalling
toward the main cluster along a North/South direction).
By comparing the ICM temperature and density maps of
A521with the numerical simulations of Ricker&Sarazin
(2001),we suggest that the group A and the main cluster
are in a pre-merger phase(～–0.5Gyr from the closest
cores encounter),with a quite low impact parameter(～<
1?2r s)and a merger axis nearly perpendicular to the line
of sight.
A higher ICM temperature has been observed in the
region surrounding the clump A
10Ferrari,C.et al.:Chandra observations of Abell
521
Fig.8.Iso-density contours of the galaxies with B <27and I <20superimposed in black on the temper-ature map of A521.The over-dense ridge of galaxies is indicated in red.
SX2
E2
E1
S2
A_BCG
SX1
A1
A2
Fig.9.Projected galaxy density map of the red sequence galaxies in A521central ?eld (15×20arcmin 2).The map has been built on the ba-sis of a multi-scale approach (see Ferrari et al.2005for more details).The positions of the A
BCG group is
falling from North onto the main cluster,the edges might be interpreted as residuals of the sloshing activity of the ICM during the merger that is taking place along the SX2axis.However,due to the very complex optical and X-ray properties of A521,nothing can exclude that A1and A2could be alternatively related to other merging events in A521,either in its central ?eld (e.g.the possible collisions of the clumps
W
or NE with the main cluster),or in its outer regions not covered by our Chandra observations.
Ferrari,C.et al.:Chandra observations of Abell52111
In this respect,several other substructures appear at larger scales in the North of the iso-density map of the projected distribution of the red sequence galaxies(see Fig.9).New optical observations(EFOSC2@3.6m ESO) have recently revealed the presence of several galaxies at the cluster redshift in the two Northern clumps E1and E2 (Fig.9),con?rming that they are very likely other merging subclusters at1.5-2h75?1Mpc from the cluster centre.A radio relic in the South-East region of the cluster has also been discovered through new VLA observations of A521 (Ferrari2003,see Appendix A for more details).
5.Summary and conclusions
Through our Chandra observations of A521we have con-?rmed that this cluster is in a disturbed dynamical state, as shown by previous X-ray and optical analysis(Arnaud et al.2000;Maurogordato et al.2000;Ferrari et al.2003).
A sketch of the possible merging scenario in the central ?eld of the cluster covered by our Chandra observations is shown in Fig.10,in which the following features emerge: -a main cluster centred on the X-ray/optical barycentre of the system;
-a group of galaxies(clump A)with its ICM density peak centred on the BCG,which is infalling on the main cluster along a NW/SE direction(～-0.5Gyr from the closest cores encounter);
-two other structures possibly interacting with the main cluster(W and NE),the former in the central phases of an o?-axis collision coming somewhere from S-SE,the latter at the beginning of the interaction and coming from North-East.The nature of these two substruc-tures,and in particular of NE,is however very uncer-tain;
-two edges in the ICM density(A1and A2),probably due to ongoing merging events either in the central ?eld of the cluster observed by Chandra,or in its outer regions.
In conclusion,Chandra observations con?rm that A521 is made up by several sub-clusters and groups of galaxies converging towards the centre of the cluster and observed in di?erent phases of their merging process.The higher resolution density and temperature maps allow to corrob-orate and re?ne the merging scenario of the group hosting the BCG(i.e.clump A),and to identify new signatures of other possible interactions(i.e.the groups W and NE, the arcs A1and A2).A deeper and wider optical spectro-scopic coverage is now necessary to understand the most puzzling regions of this system and clarify its extremely complex multiple merging scenario.
Appendix A:Radio emission in A521
New VLA observations at1.4GHz,with angular resolu-tion of12”×12”and sensitivity of0.025mJy/beam(1σ) have revealed the presence of a faint radio relic in the South/East region of the cluster(Fig.A.1and Ferrari 2003),thus supporting the perturbed dynamics of A521. Such low-brightness extended radio sources are indeed only detected in cluster mergers(Feretti2003).Detailed results on this radio source will be presented elsewhere (Ferrari et al.in preparation).
A Wide Angle Tail(“WAT”)radio source has also been detected in the North of the cluster(Fig.A.1and Ferrari 2003).The WAT is located in the clump E1,which is probably merging with the main cluster(see Sect.4.3and Fig.9).Further spectroscopic observations could reveal if the optical galaxy associated to the WAT is at the cluster redshift.The relative motion between the host galaxy and the ICM,due to the infall of the clump E1towards the cluster centre,would then be responsible for the observed bend of the radio jets(Feretti and Venturi2002). Acknowledgements.We warmly thank Wolfgang Kapferer, Magdalena Mair and Jean-Luc Sauvageot for intensive and fruitful discussions on the merging scenario of the cluster.We are very grateful to Luigina Feretti for her helpful contribu-tion to the analysis of the radio properties of A521.The au-thors thank the anonymous referee for his/her suggestions that improved the presentation of the paper.This research was sup-ported in part by Marie Curie individual fellowship MEIF-CT-2003-900773(CF).
References
Abell,G.O.,Corwin,H.G.,&Olowin,R.P.,1989,ApJS,70,1 Abell,G.O.,1958,ApJS,3,211
Adami, C.,Biviano, A.,Durret, F.,&Mazure, A.,2005, astro-ph/0507542
Anders,E.,&Grevesse,N.,1989,GeCoA,53,197
Arnaud,M.,Maurogordato,S.,Slezak,E.,&Rho,J.,2000, A&A,355,461
Bardelli,S.,Venturi,T.,Zucca,E.,De Grandi,S.,Ettori,S., &Molendi,S.,2002,A&A,396,65
Bardelli,S.,Zucca, E.,Zamorani,G.,Moscardini,L.,& Scaramella,L.,2000,MNRAS,312,540
Barrena,R.,Biviano,A.,Ramella,M.,Falco,E.E.,&Seitz,S., 2002,A&A,386,861B
Belsole,E.,Sauvageot,J.-L.,Pratt,G.W.,&Bourdin,H., 2005,A&A,430,385
Belsole,E.,Pratt,G.W.,Sauvageot,J.-L.,&Bourdin,H., 2004,A&A,415,821
Bertschinger E.,1998,ARA&A,36,599
Borgani,S.,Murante,G.,Springel,V.,Diaferio,A.,Dolag,K., Moscardini,L.,Tormen,G.,Tornatore,L.,Tozzi,P.,2004, MNRAS,348,1078
Boschin,W.,Girardi,M.,Barrena,R.,Biviano,A.,Feretti,L., &Ramella,M.,2004,A&A,416,839
Czoske,O.,Moore,B.,Kneib,J.P.,&Soucail,G.,2002,A&A, 386,31
Demarco,R.,Rosati,P.,Lidman,C.,et al.,2005,A&A,432, 381
Dickey,J.M.,&Lockman,F.J.,1990,ARA&A,28,215 Donnelly,R.H.,Forman,W.,Jones,C.,Quintana,H.,Ramirez,
A.,Churazov,E.,Gilfanov,M,2001,ApJ,562,254 Dupke,R.,&White,R.E.2003,ApJL,583,13
Durret,F.,Lima Neto,G.B.,&Forman,W.,2005,A&A,432, 809
12Ferrari,C.et al.:Chandra observations of Abell
521
Fig.10.Cartoon showing the possible merging scenario in the central region of A521.The main merging subclusters (yellow)and their infall direction (white)are superimposed on the image of A521X-ray di?use emission.
Durret,F.,Forman,W.,Gerbal,D.,Jones,C.,Vikhlinin,A.,
1998,A&A,335,41
Feretti L.,2003,in Texas in Tuscany ,XXI Symp.on
Relativistic Astrophysics,eds.R.Bandeira,R.Maiolino,F.Mannucci,World Scientif.Publ.Singapore,p.209
Feretti L.,&Venturi,T.,2003,“Merging Processes in Galaxy
Clusters”,edited by L.Feretti,I.M.Gioia &G.Giovannini,(Kluwer,Dordrecht=,pp.163-195
Ferrari,C.,Benoist,C.,Maurogordato,S.,Cappi,A.,&Slezak,
E.,2005,A&A,430,19
Ferrari,C.,Maurogordato,S.,Cappi,A.,&Benoist,C.,2003,
A&A,399,813
Ferrari, C.,2003,PhD thesis,University of Nice-Sophia
Antipolis,France
Flores,R.A.,Quintana,H.,&Way,M.J.,2000,ApJ,532,206Forman,W.,Jones,C.,&Tucker,W.,1985,ApJ,293,102Gavazzi,G.,Cortese,L.,Boselli, A.,Iglesias-Paramo,J.,
V ′ilchez,J.M.,&Carrasco,L.,2003,ApJ,597,210Henriksen,M.J.,&Tittley,E.,2002,ApJ,577,701
Henriksen,M.J.,Donnelly,R.H.,&Davis,D.S.,2000,ApJ,
529,692
Henry,J.P.,Finoguenov,A.,&Briel,U.G.,2004,ApJ,615,
181
Hoessel,J.,1980,ApJ,241,493
Johnson,M.W.,Cruddace,R.G.,Wood,K.S.,Ulmer,M.P.,&
Kowalski,M.P.,1983,ApJ,266,425
Kowalski,M.P.,Cruddace,R.G.,Wood,K.S.,&Ulmer,M.P.,
1984,ApJS,56,403L′o pez-Cruz,O.,Yee,H.K.C.,Brown,J.P.,Jones, C.,&
Forman,W.,1997,ApJL,475,97
Lubin L.M.,Bahcall,N.A.,1993,ApJ,415,L17
Markevitch,M.,Govoni,F.,Brunetti,G.,&Jerius,D.,2005,
ApJ,627,733
Markevitch,M.,Gonzalez,A.H.,David,L.,Vikhlinin,A.,
Murray,S.,Forman,W.,Jones,C.,&Tucker,W.2002,ApJL,567,27
Markevitch,M.,&Vikhlinin,A.,2001,ApJ,563,95
Markevitch,M.,Vikhlinin,A.,&Mazzotta,P.2001,ApJL,
562,153
Markevitch,M.,Ponman,T.J.,Nulsen,P.E.J.,et al.,2000,
ApJ,541,542
Maurogordato,S.,Proust,D.,Beers,T.C.,Arnaud,M.,Pell,
R.,Cappi,A.,Slezak,E.,Kriessler,J.R.,2000,A&A,355,848
Mazzotta,P.,Markevitch,M.,Vikhlinin,A.,Forman,W.R.,
David,L.P.,&VanSpeybroeck,L.2001,ApJ,555,205Merritt,D.,1984,ApJ,276,26
Plionis,M.,Benoist, C.,Maurogordato,S.,Ferrari, C.,
Basilakos,S.,2003,ApJ,594,144
Ferrari,C.et al.:Chandra observations of Abell52113
Fig.A.1.Left:20cm radio map of A521obtained from new VLA observations(Ferrari2003).The contours levels are:-0.075,0.075,0.150,0.200,0.400,0.800,1,2,4,8,32mJy/beam.The r.m.s.noise level is0.025mJy/beam.The beam size is12”×12”.Right:contours of radio(red)and X-ray(blue)emission are overlayed on the deep I-band image of A521central?eld(～15′×13′).The contour levels are set as in the left panel(radio)and in Fig.2(X-ray). The newly detected radio relic(Ferrari2003)is the extended source in the South/East cluster region.The WAT radio source in the North of the cluster is also indicated.
Poggianti, B.M.,Bridges,T.J.,Komiyama,Y.,Yagi,M.,
Carter,D.,Mobasher,B.,Okamura,S.,&Kashikawa,N.,
2004,ApJ,601,197
Ricker,P.M.,Sarazin,C.L.,2001,ApJ,561,621
Roettiger,K.,Burns,J.,Loken,C.,1993,ApJ,407,53
Rose,J. A.,Gaba, A.E.,Christiansen,W.A.,Davis, D.S.,
Caldwell,N.,Hunstead,R.W.,&Johnston-Hollitt,M.,
2002,AJ,123,1216
Schindler,S.,&M¨u ller,E.,1993,A&A,272,137
Starck,J.L.,Murtagh, F.,Bijaoui, A.,1998,”Image and
Data Analysis:The Multiscale Approach”,Cambridge
University Press
Sun,M.,Murray,S.S.,Markevitch,M.,&Vikhlinin,A.2002,
ApJ,565,867
Takizawa,M.,1999,ApJ,520,514
Umeda,K.,Yagi,M.,Yamada,S.F.,et al.,2004,ApJ,601,805
Valtchanov,I.,Murphy,T.,Pierre,M.,Hunstead,R.W.,&
L′e monon,L.,2002,A&A,392,795
Vikhlinin,A.,Markevitch,M.,&Murray,S.S.2001,ApJ,551,
160
West,M.J.,Jones,C.,&Forman,W.,1995,ApJL,451,5
West,M.J.,1994,MNRAS,268,79
Zabludo?,A.I.,&Mulchaey,J.,1998,ApJ,496,39