Environmental effects on the bright end of the galaxy luminosity function in galaxy clusters

0.41 MB·English

by  R. Barrena

Astronomy&Astrophysicsmanuscriptno.lf2 (cid:13)c ESO2012
January19,2012
Environmental effects on the bright end of the galaxy luminosity
function in galaxy clusters
R.Barrena1,2,M.Girardi3,4,W.Boschin5,andF.Mardirossian3,4
1 InstitutodeAstrof´ısicadeCanarias,C/V´ıaLa´cteas/n,E-38205LaLaguna(Tenerife),CanaryIslands,Spain
2 DepartamentodeAstrof´ısica,UniversidaddeLaLaguna,Av.delAstrof´ısicoFranciscoSa´nchezs/n,E-38205LaLaguna(Tenerife),
CanaryIslands,Spain
3 DipartimentodiFisicadell’Universita`degliStudidiTrieste-SezionediAstronomia,viaTiepolo11,I-34143Trieste,Italy
2
1
4 INAF-OsservatorioAstronomicodiTrieste,viaTiepolo11,I-34143Trieste,Italy
0
2
5 Fundacio´nGalileoGalilei-INAF,RamblaJose´AnaFerna´ndezPerez7,E-38712Bren˜aBaja(LaPalma),CanaryIslands,Spain

n
a
J Received/Accepted

8
1 ABSTRACT

] Context.Thedependenceoftheluminosityfunction(LF)ofclustergalaxiesontheevolutionarystateoftheparentclusterisstillan
O openissue,inparticularasconcerntheformation/evolutionofthebrightestclustergalaxies.
Aims.WeplantostudythebrightpartoftheLFsofasampleofveryunrelaxedclusters(“DARC”clustersshowingevidenceofmajor,
C
recentmergers)andcomparethemtoareferencesampleofrelaxedclustersspanningacomparablemassandredshiftrange.
h. Methods.OuranalysisisbasedontheSDSSDR7photometricdataoften,massive,andX-rayluminousclusters(0.2 . z . 0.3),
p alwaysconsideringphysicalradii(R200oritsfractions).Weconsiderr′bandLFsandusethecolor-magnitudediagrams(r′−i′,r′)to
- cleanoursamplesaswelltoconsiderseparatelyredandbluegalaxies.
o Results.Wefind that DARC and relaxed clusters give similar LFparameters and blue fractions. Thetwo samples differ for their
tr contentofbrightgalaxiesBGs,Mr′ <−22.5,sincerelaxedclustershavefewerBGs,inparticularwhenconsideringtheoutercluster
s region0.5R200 <R<R200(byafactortwo).However,thecumulativelightinBGsissimilarforrelaxedandDARCsamples.
a Conclusions.WeconcludethatBGsgrowinluminosityanddecreaseinnumberastheparentclustersgrowhierarchicallyinagree-
[ mentwiththeBGformationbymergingwithotherluminousgalaxies.

1 Keywords.Galaxies:clusters:general–Galaxies:luminosityfunctions–Cosmology:observations
v
6
9
1. Introduction local density and thus the clustercentric radius (Dressler 1978;
7
3 Whitmore 1991) make more complex the study of the cluster
The study of the formation and evolution of galaxies is a fun-
. LF.Inthiscontext,severalstudiesaboutLFconcernthepossible
1 damental avenue of research in the process of understanding
variationoftheclusterLFwiththelocalprojectedgalaxydensity
0 astrophysical and cosmological issues. The galaxy luminosity
and/orcluster-centricradius.Inparticular,thereareclaimsfora
2 function (LF) – the number of galaxies per unit volume in the
radius-dependent steepening of the galaxy LF at faint magni-
1 luminosityintervalLandL+dL–isoneofthemostdirectob-
: tudes,likelyduetoredgalaxies(e.g.,dePropris1995;Driveret
v servationaltestoftheoriesofgalaxyformationandevolution. al.1998;Popessoetal.2006;Barkhouseetal.2007).However,
i Clustersofgalaxiesareidealsystemswithinwhichtomea-
X the few cases of LFs based on deep spectroscopy exhibit shal-
surethegalaxyLFforthelargenumberofgalaxiesatthesame
lower slopes (e.g., Rines & Geller 2008 and refs. therein)sug-
r distance.Ontheotherhand,clustersrepresentanextremeenvi-
a gesting that the field contamination might be the cause of the
ronmentforgalaxyevolution,eitherinsituorthroughtheaccre-
observedsteep LFs based on photometryonly.Thus, the ques-
tionofgalaxieswithingroups,whicharesituatedinthefilamen-
tionisstillfarfromadefinitiveconclusion.
tarystructureofthehierarchicalUniverse(e.g.,Poggiantietal.
2005). Another issue concerns the possible dependence of the LF
WhileintroducingthemodernformoftheLF,theso–called on global cluster properties. Some studies interest a few well
“SchechterFunction”,Schechter(1976)suggestedthattheclus- studied clusters and/or cluster complexes (e.g., Abell 209 in
ter LF is universalin shape (with a turnoverat M∗ = −20.6+ Mercurio et al. 1996, Coma vs. Abell 1367 in Iglesias-Pa´ramo
B
5logh andafaint-endslopeofα=−1.25).Muchprogresshas etal.2003)andotherstudiesuseinsteadalargestatisticalbase.
50
thenbeenmadeinlookingforpossibleobservationalsignatures In particular, there are indicationsin favour of correlationsbe-
of a non universality of the cluster LF. That different morpho- tweentheLFparametersandtheclustermassoritsproxies(e.g.,
logicaltypesare characterizedby differentLFs (Binggeliet al. richness, luminosity, galaxy velocity dispersion). For instance,
1988) and the well known morphologicalsegregation with the Valotto et al. (1997) find that poorer clusters have a flatter LF.
But,whentheLFisproperlycalculatedwithintheclusterphys-
Sendoffprintrequeststo:R.Barrena,e-mail:[email protected] ical sizes, given by R200 or R500, the correlation between the
1
R.Barrenaetal.:Environmentaleffectsonthebrightendofthegalaxyluminosityfunctioningalaxyclusters
dwarf-to-giantratioandX-ray(oroptical)luminositydisappear better (e.g., Ruszkowski & Springel 2009). Moreover, galaxy-
(cf.Popessoetal.2006withPopessoetal.2004).Thissuggests galaxymergersaremostefficientwithinsmallhaloswithsmall
thatthe radialbehaviorof the LF should be taken into account velocitydispersionandthemergerofBCGsinclustersrequires
toobtainunbiasedconclusions.However,alsowhenconsidering alongtime(Dubinski1998).However,thecorrelationbetween
an appropriatecluster physicalsize, the conclusionsare notall BCG luminosity and parent cluster mass strongly suggest that
consistent. For example, Hilton et al. (2005) find that clusters BCGsco-evolveswiththecluster(Lin&Mohr2004).
withlowerX–rayluminosityhaveabrighterM∗,butBarkhouse Inthiscontext,weplantostudytheLFsofasampleofmas-
et al. (2007) find no correlation.Recentstudies also attemptto siveandveryunrelaxedclustersandcomparethemtoareference
analyze more distant clusters, but again with no conclusive re- sampleofrelaxedclustersspanningacomparablemassandred-
sults(Gilbanketal.2008;Rudninicketal.2009anddiscussion shiftrange.
therein). As for the unrelaxed clusters, we are going on with a long
Possible correlations with the presence/absence of a cD term project to study clusters exhibiting large, diffuse radio
galaxy or with the Bautz-Morgan type (the BM-type classifi- sources,i.e.radiohalosandrelics, basedonspectroscopicdata
cation based on the brightest cluster galaxies, BCGs, Bautz & mostlyacquiredatTNG(DARC-DynamicalAnalysisofRadio
Morgan1970),are,maybe,alsomoreinterestingfortheirdirect Clusters1-e.g.,Girardietal.2007).Theseradiosourcesarerare
connectionwiththeclusterevolutionarystate.Infact,BCGsor phenomena and are due to the synchrotron radio emission of
cDgalaxiesareexpectedtobecreatedi)viathemergerofgiant widespread relativistic particles embedded in the cluster mag-
galaxies in an early phase during the cluster collapse (Merritt netic field (e.g., see Feretti et al. 2002b for a review). Cluster
1984) or ii) in a following phase due to the dynamical fric- mergershavebeenproposedtoprovidethelargeamountofen-
tion acting on late-comer galaxies (e.g., Ostriker & Tremaine ergynecessaryforelectronreaccelerationtorelativisticenergies
1975)oriii)viathedisruptionandcannibalizationofmanyfaint andformagneticfieldamplification(e.g.,Tribble1993;Ensslin
galaxiesin thecluster cores(Lo´pez-Cruzetal. 1997). Thefirst etal.1998;Brunettietal.2009)orrequiredtoexplainthetime-
two mechanisms might reduce the number of bright galaxies, dependenceofthemagneticfieldsinrecentmodelsbasedonsec-
thus shifting M∗ to a fainter value and, indeed, Barkhouse et ondary electrons (Keshet & Loeb 2010; Keshet 2011). Indeed,
al.(2007)findaweakcorrelationwiththeBM-type(butseede fromtheobservationalpointofview,thereisgrowingevidence
Proprisetal.2003fornocorrelation).Anotherconsequencecan of the connection between diffuse radio emission and cluster
be the increment of the gap between the first and the second merging, since up to now diffuse radio sources have been de-
brightestgalaxiesineachclusterand,infact,thismagnitudegap tectedonlyinmergingsystems.Inmostpaststudiesthecluster
isshowntobelargerinclusterswithoutsubstructureandsmaller dynamicalstatewasderivedfromX–rayobservations(seeBuote
inclusterswithsubstructure(Ramellaetal.2007). 2002;Feretti2006and2008;Cassanoetal.2010).Inagreement,
Asforthesecondmechanism,Lo´pez-Cruzetal.(1997)find ouranalyses,basedonopticaldataandthekinematicsofmem-
thatsevenrich,massive,cDclusterswithsymmetricX-raymor- ber galaxies, detect strong subclustering and find evidence of
phology show a flat faint-end slope while a steeper faint-end cluster mergers. Summarizing, clusters with radio halos/relics
slope is detected in poorer clusters but also in a rich, binary aretheidealcasestostudytheeffectoftheclusterformationon
cluster like the Coma cluster. Somewhat in agreement, Driver theLF.
et al. (1998) find that the dwarf-to-giant ratio increases with HereweanalyzeasubsampleoffiveDARCclusterswehave
larger BM-types(i.e., likely less evolved clusters). In this con- analyzedin the pastand all havingavailablephotometryin the
text,Valottoetal.(2004)pointedouttheimportanceofthepro- theSloanDigitalSkySurvey(SDSS).TheDARCclustersform
jectioneffectsindeterminingthefaint-endslope.Infact,inthe our sample of unrelaxed clusters. For comparison, we analyze
largeclustersampleanalyzedbyValottoetal.itistheX-rayse- a sample of relaxed clusters of comparable mass and redshift.
lectionand notthe cluster dominationby centralgalaxieswhat Consideringthe magnitudelimit of SDSS photometry,the typ-
determinestheflatnessofthefaint-endslopethussuggestingthat ical redshift of the samples, z ∼ 0.25, limits our study to the
real, compact clusters are less affected by the field contamina- bright part of the LF. We plan to extend this study to a fainter
tion. However,althoughthis bias couldexplainpossible differ- magnitudesinthenextfuture.
ences between rich and poor clusters, the reason for the differ- Unless otherwise stated, we give errors at the 68% confi-
encebetweenthesevencDrelaxedclustersofLo´pez-Cruzetal. dencelevel(hereafterc.l.).
(1997)andtheComaclusterisnotclearyet. Throughoutthispaper,we use H = 70km s−1 Mpc−1 in a
0
As for the possible dependence of the LF on the dynami- flat cosmologywith Ω0 = 0.3 and ΩΛ = 0.7. We define h70 =
calstatusofthecluster,thenoticeablestudyofdeProprisetal. H /(70kms−1Mpc−1).
0
(2003) analyze 60 clusters and find that the LF is similar for
clusterswithandwithoutsubstructure.Veryfewdetailedworks
2. Clustersample
aredevotedtostudytheLFsofsubclumpsinindividualclusters
to find possible evidence of cluster mergers (e.g., Durret et al.
Amongthe DARCclusters ofwhich we havealreadyanalyzed
2011).
the internal dynamics using a large number of member galax-
Asreportedabove,itisparticularlyinterestingthebrightend
ies, five clusters (with 62-113 spectroscopic member galaxies)
of the LF since there is a long debate on the BCG formation.
have also available photometry in the SDSS (Data Release 7):
For instance, semianalytical models suggest that BCGs assem-
Abell 697, Abell 773, Abell 959, Abell 1240, and Abell 2219,
ble surprisingly late with the stars formed very early in many
hereafter A697, A773, A959, A1240, A2219 (see Table 1 for
smallgalaxies(e.g.,50%atz=0.5,80%atz=0.3;see deLucia
therespectivereferencesources).Theirredshiftsspantherange
& Blaizot 2007). However, photometric and chemical observ- z = 0.19−0.29andtheyarecharacterizedbyhighline-of-sight
ables suggest that it is difficult to explain giant elliptical by a
galaxy(LOS)velocitydispersionσ ,X–raytemperatureT ,and
v X
pure sequence of multiple minor dry mergers or via major dry
mergers(e.g.,Stottetal. 2010;Ascaso etal.2011)andnumer- 1 other information are given in the web site
ical simulations are continuously improved to reproduce them http://adlibitum.oat.ts.astro.it/girardi/darc.
2
R.Barrenaetal.:Environmentaleffectsonthebrightendofthegalaxyluminosityfunctioningalaxyclusters
mass M (see Table 1). Evidence for their unrelaxeddynamical Asfortherelaxedclusters,weestimateσ fromtheobserved
v
stateisreportedinourpreviousstudies.Inparticular,theyspan T assuming the equipartition of energy density between ICM
X
differentangle of views for the mergingaxis. For instance, the and galaxies, i.e. β = 1, where β = σ2/(kT/µm ) with
spec spec V p
clustermergeroccurredlargelyintheplaneoftheskyinA1240 µ = 0.58 the mean molecular weight and m the proton mass.
p
andlargelyalongtheLOSinA773. Thisassumptionisparticularlyappropriateformassiveclusters
As forcomparison,we selectfromtheliteraturea fewwell (e.g.,Girardietal.1996,1998).R andM arethusrecovered
200 200
known examples of very relaxed clusters spanning a similar z withtheaboveequations.
range(0.2.z.0.3),characterizedbyhighT ,andsampledby For each relaxed cluster, Table 2 lists the main properties:
X
SDSS.FromtherelaxedclusterslistedbyAllenetal.(2004)we the cluster center (Col. 2); the cluster redshift, z (Col. 3); the
select2 Abell 383, Abell 963, and Abell 1835 (hereafter A383, X-raytemperature,T (Col.4);theestimatedvalueofσ from
X v
A963,A1835).Theseclustersareallclassifiedascoolcoreclus- T (Col.5;)thevaluesofR and M (Cols.6and7);useful
X 200 200
tersaccordingtoBaldietal.(2007).Thisagreeswiththatobser- references(Col.8).
vations suggest that cool cores are destroyed by cluster merg- AsfortheDARCclusterswehavetoadoptamorecomplex
ers and cool core clusters are in the final phase of cluster re- procedure.Duringaclustermerger,theglobalvalueofvelocity
laxation (see e.g., Allen et al. 2001; Buote 2002; Sanderson et dispersion,σ ,mightstronglyvaryandnotbeaverygoodin-
v,tot
al. 2006). Among the cool core clusters listed by Baldi et al. dicators of the real cluster potential (e.g., Pinkney et al. 1996)
(2007) we considered other two cool core clusters to complete and the same problem could afflict T (e.g., Ricker & Sarazin
X
ourcomparisonsample:Abell2261(hereafterA2261)andZwCl 2001;Mastropietro&Burkert2008).Asdiscussedinourpapers
1021.0+0426(hereafterandbetterknownasZwCl3146). forthefiveDARCclusters,wehavedetectedthemainsubclus-
Toavoidtheinherentbiasthathasplaguednumerousstudies ters and obtain an estimate of their individual velocity disper-
(seethesectionabove),theclusterLFswillbecomparedbased sion, σ . Thus, an alternative, likely more reliable value of
v,subs
onscalingrelativetothedynamicalradius,R .Asanestimate themassM ofthecluster,canthenbeobtainedbyaddingthe
200 200
of R , we use R definition by Girardi & Mezzetti (2001), massesofallsubclusters.Then,invertingtheabovescalingrela-
200 vir
Eq.1withthescalingofH(z), tioneq.2,wecanobtainavalueofthevelocitydispersion,σ ,
v
which should be a better indicator of the potential. The use of
R200 =0.17×σv/(kms−1)/H(z)h−701Mpc, (1) eq.1leadstotheestimateofR200.
ForeachDARCcluster,Table1liststhemainproperties:the
where σv is the LOS velocity dispersion. The same result was clustercenter(Col.2);theclusterredshift,z(Col.3);theX-ray
obtained by Carlberg et al. (1997, see their Eq. 8 for R200) temperature,TX (Col. 4);the globalobservedvalue ofvelocity
using a different, more theoretical, approach. Biviano et al. dispersion,σ (Col.5);the(rounded)valuesofvelocitydis-
v,obs
(σ2v0/0(6k)mosb−t1a)i/nHed(za)h1−7001%MspmcaollnertheestbimasaeteofofNR-b2o00d,yRsvim=ul0at.1io5n×s. pmearsssiofnrsomforthseuabdcdluitsitoenrsoσfvs,usubbcsl(uCstoelr.s6,)M; t2h00e(eCstoilm.7at)e;dthveacluoerroef-
Wealsoconsiderinternalandexternalregions,i.e.R < 0.5R200 spondingvaluesofσvandR200(Cols.8and9).Thevaluesofthe
and 0.5R200 < R < R200, respectively. The centers for relaxed clustercenter,z,TX,σv,tot,andσv,subs canbeobtainedfromthe
clusters are taken from X-ray studies, while those for DARC listed references. In particular, the notes list the source for the
clustersarethoseusedinourpreviouspapers.Notethat,dueto valuesofσ ,hererounded.Othervaluesarehomogeneously
v,subs
the unrelaxedstate of DARC clusters, the choice of the cluster computedinthisstudy.
center is not obvioussince gas and galaxy spatial distributions
aregenerallydifferent.ForA697,A773andA2219clusterswe
used the position of the brightest dominant galaxy, very close 3. Magnitudedataandbackgroundcontamination
totheX-raycentroid(orX-raymainpeak),whileforA959and The magnitudedata (r′,i′) are taken from SDSS (Data Release
A1240weuseX–raycentroid.However,inthesetwocases,X–
7). We retrieve model mag values. For each cluster, we create
ray centroid is located in between the two brightest dominant
a galaxy “cluster galaxy sample” by selecting the galaxies in
galaxies. A radius of 0.5R is sufficient to contain the cores
200 an area centered on the cluster center listed in Tables 2 and 1
of galaxy subclusters, the most critical case being the bimodal
and withinthe respectiveradiusR . We verifythatthe whole
clusterA1240,stronglyelongatedintheplaneofthesky. 200
clusterareaiscoveredbySDSSdatabyvisualinspection.
Asforthemassestimate,inarelaxedcluster,onecanusethe
For each cluster, we also consider a galaxy sample select-
measuredglobalvalueofvelocitydispersion,σ todetermine
v,tot ing galaxies in one square degree field at 1.5 degree toward
R andthenM withinR .Theestimationofmassisbased
200 200 200 the west of each cluster obtaining ten individualfield samples.
onthevirialtheoremand,inparticular,we followtheprescrip-
Together these samples, for a total area of ten square degrees,
tionsofGirardietal.(1998;seealsoGirardi&Mezzetti2001).
forms the “field galaxy sample” and are used to estimate the
Weprompttotheoriginalpapersforthedetails,but,inpractice,
field galaxy contamination. Fig. 1 shows the field counts for
onecanuse: r′ and i′ magnitudes, and the logarithmic function fit. This fit
M =A×[σ /(103kms)]3×h−11015M , (2) represents the count-magnitude relation expected for a homo-
200 v 70 ⊙ geneous galaxy distribution in a universe with Euclidean ge-
ometry. We obtain the relations log(N ) = 0.455(±0.043)r′ −
whereA∼1.4(heremedianA=1.416).Thusbothourestimates r
(6.17±0.59) 0.5mag−1deg−2 for 15 < r′ < 22 and log(N) =
ofR andM dependontheestimateoftheobservablevalue i
200 200 0.451(±0.046)i′−(5.70±0.58) 0.5mag−1deg−2 for 15 < i′ <
ofσ .However,theydonotaremutuallyconsistentaccordingto
v
thedefinition:M =200×ρ ×4/3×π×R3 ,whereρ is 21.5. We used these fits with two purposes: i) to estimate the
200 crit 200 crit completenessofoursample;ii)tosubtractfieldgalaxycontam-
thecriticaldensityoftheUniverseattheclusterredshift.Rather,
inationineachcluster.
accordingtothedefinition,theycorrespondtoR andM .
194 194
By comparison with the logarithmic fits, the completeness
2 NotethatweexcludeAbell611duetoitsunrelaxedappearancein magnitudecanbederived.So,weestimatethephotometricsam-
theSDSSimage. ple is complete down to the magnitude counts are lower than
3
R.Barrenaetal.:Environmentaleffectsonthebrightendofthegalaxyluminosityfunctioningalaxyclusters
Table1.Unrelaxedclusters(DARCsample).
Name α,δ(J2000) z T σ σ M σ R
X v,tot v,subs 200 v 200
keV kms−1 (kms−1) 1015M⊙ (kms−1) Mpc
A697a 084257.55,+362159.9 0.2815 10.2 1334+114 200,600,400 0.40 660 1.39
−95
A773b 091753.26,+514336.5 0.2197 7.8 1394+84 950,500 1.39 994 2.16
−68
A959c 101735.04,+593327.7 0.2883 7.0 1170+83 600,700 0.76 824 1.73
−73
A1240d 112337.60,+430551.0 0.1948 6.0 870+91 700,1000 1.93 1103 2.43
−79
A2219e 164019.87,+464241.3 0.2254 10.3 1438+109 1000 1.41 1000 2.17
−86
a Ref.:Girardietal.2006;σ fromKMM4g1-2-4inTable4.
v,subs
b Ref.:Barrenaetal.2007;σ fromthemainandsecondarysubclustersinSect.5.
v,subs
c Ref.:Boschinetal.2008;σ fromtheaveragesofKMM1andV1,andofKMM2andV2inTable2.Herewenotconsiderthesubclump
v,subs
V3(=KMM3=DS-NE)alsodetectedinX-rayandthuslikelyinapremergingphaseandnotresponsableoftheongoingmerger.
d Ref.:Barrenaetal.2009;σ fromA1240NandA1240SinTable2.
v,subs
e Ref.:Boschinetal.2004;σ fromNWinTable2.Theseparationofsuclumpsisnotobvious.Here,sincetheprimarysubclumpliesatthe
v,subs
SEoftheclustercenter(seealsothemorerecentpaperMillion&Allen2009),weconsiderthevalueofσ ofNWsector,wherethemain
v
clusterislikelyfreefromthesubclump.
Table2.Relaxedcluster(Comparisonsample).
Name α,δ(J2000) z T σ R M Refs.a
X v 200 200
keV (kms−1) Mpc 1015M
⊙
A383 024803.50,−033145.0 0.188 3.9 802 1.78 0.74 Allenetal.2004,Maughanetal.2008
A963 101701.20,+390144.4 0.206 6.0 995 2.18 1.41 Allenetal.2004,Baldietal.2007
A1835 140102.40,+025255.2 0.252 8.1 1156 2.47 2.15 Allenetal.2004,Baldietal.2007
A2261 172227.60,+320757.2 0.224 7.2 1090 2.36 1.83 Baldietal.2007,Maughanetal.2008
ZwCl3146 102339.60,+041124.0 0.291 8.6 1191 2.49 2.31 Baldietal.2007,Baldietal.2007
a ReferencesforzandT arelistedinthefirstandsecondposition,respectively.ClustercentersaretakenfromEbelingetal.1996orEbeling
X
etal.1998.
parametersusing either a globalor localfield subtractiontech-
nique.Herewe considertheglobalfieldcorrection,subtracting
the field counts derived from the logarithmic fits to the cluster
counts. Note that the field counts are treated in a different and
appropriatewayforeachclusterbeforethesubtraction.
The magnitudesof each cluster are correctedfor the galac-
ticabsorptionusingtheSchlegel,FinkbeinerandDavisGalactic
reddeningmaps(Schlegelet al. 1998), derivedfrom IRAS and
COBE/DIRBEdata.
Thepossiblecontaminationofdistantclustersprojectedonto
Fig.1. Values of r′− and i′−band counts for the 10 square de- thefield-of-viewofthetargetclustercanprovetobeproblematic
greesconsideredasthefieldgalaxysample.Thesolidlineindi- inthederivationoftheLF,inparticularatthefaint-end.Onecan
catesthelogarithmicfunctionfittothehistograms.Thedashed usethatthespectroscopicstudiesshowthatthereareessentially
lineindicatesthecompletenessmagnitudes. no cluster galaxies significantly redward of the red sequence
(e.g.Rines & Geller 2008). We minimizethe contaminationof
backgroundclustersusingther′−i′vs.r′ colour-magnitudere-
5% of the fit. Therefore, we consider the r′- and i′-band pho- lationwhichisprovedtobethemostusefulfromtargetclusters
tometry complete down to r′ = 22.4 and i′ = 21.4, respec- areat0.2<z<0.4(Luetal.2009,theirFig.6andSect.3.4.1).
tively. We find that these values are in very good agreement For each cluster, we fit the red sequence (RS) r′ −i′ vs. r′ for
with those reportedby the Sloan Digital Sky Survey DR7 (see galaxies.Werejectgalaxiesthatare0.15magredwardoftheRS
http://www.sdss.org/dr7/). (i.e. & 3 times the average dispersion of the cluster RS). Fig.2
Twodifferentapproachestothestatisticalsubtractionofthe shows an example for the RS in the case of A383. After this
galaxy foreground/background are discussed in the literature: backgroundcorrection,theclustersappearmorecontrastedwith
several studies have examined the effect on the derived cluster respecttothesurroundingfield(seeFig.3forA383.)
LF using a global galaxy field correction versus one measured Tab.3lists,foreachcluster,theslopeaandtheinterceptb,
locally for each cluster. For instance, Popesso et al. (2004) has andstandarddeviations,oftheRSr′ −i′ vs.r′ obtainedwitha
shown, for a study of about 100 clusters based on SDSS data, linear fit within 0.5 h−1 Mpc from the center and for galaxies
70
thatthereisnosignificantdifferenceinthemeasuredclusterLF withr′ <20.2(Col.2and3).
4
R.Barrenaetal.:Environmentaleffectsonthebrightendofthegalaxyluminosityfunctioningalaxyclusters
Table 3. Color-magnitude relations (CMR) r′ −i′ vs. r′ linear
fits(az˙+b):RedSequences.
Name (a,δa) (b,δb)
A697 −0.0107,0.014 0.744,0.041
A773 −0.0213,0.007 0.861,0.037
A959 −0.0310,0.006 1.126,0.037
A1240 −0.0110,0.020 0.666,0.037
A2219 −0.0086,0.007 0.668,0.062
A383 −0.0160,0.001 0.763,0.047
A963 −0.0422,0.007 1.216,0.042
A1835 −0.0004,0.011 0.501,0.031
A2261 −0.0091,0.015 0.653,0.048
ZwCl3146 −0.0220,0.003 1.020,0.050
Fig.2.Colormagnitudediagramr′−i′vs.r′consideringgalaxies
within 0.5h−1 Mpc fromthe center of Abell383.TheRS (red
70
solidline)isfittedusinggalaxieswithr′ < 20.2.Galaxieswith
r′−i′ >RS+0.15(onthedashedline)areconsideredbackground
galaxiesandrejectedfromtheanalysis.
Fig.4.RSparametersvs.redshift:the(r′−i′)vs.r′slope(upper
panel)and(r′−i′)coloratr′ =17(lowerpanel).Blacksymbols
indicate DARC clustersand red symbolsindicate relaxedclus-
ters.Thesolidlineintheupperpanelshowsthemeanvaluefor
alltenclusters.Thesolidlineinthelowerpanelshowsthelinear
Fig.3. Galaxies with r′ −i′ <RS+0.15 in the region of A383. fit.Thedashedlinesshowone-sigmauncertaintiesofthelinear
TheredcircleindicatesR200forthiscluster. fit.
Foreachcluster,wetransformgalaxyapparentintoabsolute
magnitudesbyapplyingtherelation:
Fig. 4 shows the RS parametersvs. cluster redshift. As for
theRSslope,thereisnocorrelationwithzintheobservedrange M =r′−2.5−5log(D /1Mpc)−K(z)+E(z), (3)
r L
and we estimate a mean value of < a >= −0.017±0.008. As
for the r′ −i′ colors at r′ = 17, as computedfrom the RS fits, where r′ is the apparent magnitude (already corrected for the
there is no difference in the behaviour of DARC and relaxed Galactic absorption,see above), D is the luminosity distance;
L
clusters:weonlyobserveamarginallyhighervalueforthemost K(z)istheK-correctionandE(z)istheevolutionarycorrection.
distantcluster A697,A959andZwCl3146,whichis consistent K(z)andE(z)correctionsareappliedusingasingleparametriza-
withthek−correctioneffectonearly-typegalaxiesinthecentral tion based on early–type galaxies, which are dominant in the
coreofclusters(seeFukugitaetal.(1995,tables3,6,7and8). cluster galaxy population. As for K(z), we use the correction
Assuming a linear variation of the color with the redshift, we obtained from the Table 2, col. 3 in Roche et al. (2009). As
obtaintherelation(r′−i′) = 1.24∗z+0.23,withaglobal for the evolutionary correction, we use E(z) = 0.86z (Roche
r′=17
dispersionincolorof±0.02. et al. 2009). Given that the completeness of magnitude data is
5
R.Barrenaetal.:Environmentaleffectsonthebrightendofthegalaxyluminosityfunctioningalaxyclusters
Table4.M⋆andαvaluesobtainedfromSchechterfitfortheLF.
r
R<R R<0.5R
200 200
Name M⋆,∆M⋆ α,∆α M⋆,∆M⋆ α,∆α
r r r r
A697 −21.19,0.30 −0.96,0.20 −20.95,0.16 −0.70,0.21
A773 −21.84,0.26 −1.01,0.13 −21.43,0.24 −0.61,0.18
A959 −21.60,0.62 −0.96,0.39 −22.00,0.94 −0.96,0.46
A1240 −22.31,0.36 −1.28,0.12 −21.61,0.37 −0.95,0.20
Fig.5.LFsforDARCclusters(rightpanel)andrelaxedclusters A2219 −21.74,0.28 −1.09,0.15 −21.79,0.08 −0.97,0.08
(leftpanel)ascomputedwithinR .Errorbarsareonlyplotted A383 −21.89,0.25 −1.30,0.17 −21.92,0.27 −1.26,0.21
200
intheA1240andA383LFsforthesakeofclarity. A963 −21.31,0.28 −0.83,0.19 −20.93,0.22 −0.27,0.25
A1835 −21.78,0.21 −1.02,0.15 −22.06,0.33 −1.13,0.18
A2261 −21.56,0.22 −0.88,0.16 −21.62,0.25 −1.01,0.19
r′ = 22.4 (Fig. 1) and the most distant cluster of our catalogs ZwCl3146 −21.36,0.26 −0.97,0.24 −20.89,0.23 −0.51,0.29
is at z = 0.29,we expectthat the LFs we obtainin the present
studyareallreliableforabsolutemagnitudeM <−18.6.
r
Foreachcluster,beforethefieldsubtractionintheconstruc-
tion of the LF (see below), the field galaxies are treated in the rameterandremakethefit.Whenfixingα=−1.00fortheslope
same way of the respective cluster galaxies, i.e. we apply the parameter for all individual clusters, we find no longer differ-
same correctionfor the Galactic absorption;the same rejection ence (see Table 5). Inparticular,whenfixing the slope,we ob-
ofveryredgalaxies;thesametransformationinabsolutemagni- tain M∗ = −21.64±0.06and M∗ = −21.61±0.05within
DARC rel
tudes. R<R200andMD∗ARC =−21.82±0.14andMr∗el =−21.64±0.06
withinR < 0.5R .Fromthisresults,wecanconcludethatthe
200
globalLF profileis similar forDARC and relaxedclusters and
4. GalaxypopulationsinDARCandrelaxedclusters independentofthesamplingradius.
4.1.Individualluminosityfunctions
Table5. M⋆ valuesobtainedfromtheSchechterfitwhenfixing
Fig. 5 shows the individual LFs for DARC and relaxed clus- r
α=−1.00fortheLF.
ters within R , where the galaxy counts are computed as
200
N = N /N . N are the net counts, i.e. N =
i net,i net,<−20 net,i net,i
N −N wherethefieldcountsareopportunelynormal-
cluster,i field,i
ized to the cluster area. Nnet,<−20 is the number of net counts R<R200 R<0.5R200
for Mr < −20 and thus represents the “richness” here used to Name Mr⋆,∆Mr⋆ Mr⋆,∆Mr⋆
normalize.The errorbarsin Fig. 5 arethe richness-normalized A697 −21.30,0.12 −21.19,0.79
σ =(N +N +rms2)1/2,wherethefirsttwotermstakeintoac- A773 −21.85,0.11 −22.15,0.12
i i b,i f A959 −21.83,0.16 −22.25,0.19
countsthePoissonianerrorsoftheclusterandbackgroundfield
A1240 −21.65,0.17 −21.69,0.18
samples,andthethirdtermmeasuresthefield-to-fieldvariation
A2219 −21.57,0.12 −21.81,0.31
perbininthebackgroundfieldcounts(i.e.,thecosmicvariance
A383 −21.51,0.14 −21.65,0.21
obtainedusingthetenindividualfieldsamples).Inordertosee
A963 −21.54,0.12 −21.70,0.13
possible variations with the radius, we also compute the LFs
A1835 −21.75,0.09 −21.95,0.14
within0.5R andR .
200 200 A2261 −21.72,0.09 −21.59,0.11
Each LF was fitted to a Schechter function (see Schechter ZwCl3146 −21.53,0.13 −21.33,0.08
1976) using the “minuit” procedure (CERN Libraries) to min-
imize the three values, φ (richness), α (slope) and M∗ of the
Schechterprofile.Wedonotdiscusstheφparametersincehere
thecountsarealreadynormalizedbythe“richness”N for The mean LF will be studied in the following subsection,
net,<−20
each cluster. Table 4 shows M∗ and αvaluesobtainedforeach andwewilldetailpossibledifferencesconsideringblueandred
cluster,fittingSchechterfunctionwithinR and0.5R . galaxypopulationintheinnerandexternalregions.
200 200
Fig. 6 shows α vs. M∗ values estimated for each cluster
within R : the absence of difference between DARC and re-
200 4.2.Thecompositeluminosityfunctions
laxedclustersisshown.WeobtainthesameresultforLFswithin
0.5R andwhencomparingR and0.5R results. For both DARC and relaxed clusters separately, we construct
200 200 200
FortheDARCsampleweobtainameanvalueof M∗ = thecompositeLFsbycombiningintoasingle”ensemble”LF,in
DARC
−21.73±0.16 and −21.57±0.16, considering galaxies within ordertoimprovetheratherpoornumberstatisticsforeachindi-
R and0.5R ,respectively.Ontheotherhand,fortherelaxed vidualLF. The composite LF is built by averaging,in absolute
200 200
clusters, we derive M∗ = −21.58± 0.11 and −21.48± 0.12. magnitudebins,therichness-normalizednetcountsasobtained
rel
Fortheslopeparameter,weestimateα =−1.06±0.09and aboveforeachindividualLF.Weassumethestandarderrorson
DARC
−0.84±0.10 within R and 0.5R , respectively. For the re- theaverage.
200 200
laxedclusters,weobtainα = −1.00±0.12and−0.84±0.10. Fig.7showsthecompositeLFsforDARCandrelaxedclus-
rel
Duetothecorrelatednatureofαand M⋆ parametersinthefit- terswithinR and0.5R .Table 6summarizesthe M∗ andα
200 200
tingprocedure,whichlikelyleadstotheapparentcorrelationin parameters obtained for the composite LFs. In agreement with
Fig. 6, we preferto compareindividualclustersfixingonepa- the above section, there is no evidence of difference between
6
R.Barrenaetal.:Environmentaleffectsonthebrightendofthegalaxyluminosityfunctioningalaxyclusters
Fig.6.αvs. M⋆ derivedfromtheSchechterLFfit.Blacksym- Fig.8. Upper panels: LFs and Schechter fits for red and blue
r
bols indicate DARC clusters and red symbols indicate relaxed galaxiescomputedwithin R200. Lower panels: the same within
clusters. 0.5R200.SolidlinesindicateDARCclustersanddashedlinesin-
dicaterelaxedclusters.
Fig.7.CompositeLFswithinR (leftpanel)and0.5R (right Fig.9.Errorcontoursatthe1and3σc.l.forthebest-fittingLF
200 200
panel). Solid lines show the counts for DARC clusters and parameters considering the galaxy population within R < R
200
dashed lines show the counts for the relaxed clusters. The red (left panel)and R < 0.5R (rightpanel).Black, red and blue
200
linesindicatethecorrespondingSchechterfits. lines show the composite LF for the whole galaxysample, red
andbluegalaxies,respectively.SolidlinesindicateDARCclus-
tersanddashedlinesindicaterelaxedclusters.
DARC and relaxed samples and there is no dependenceon the
samplingradius.
for DARC clusters and 0.35 for relaxed clusters (both within
R ) and χ2 = 0.33 for DARC clusters and 0.27 for relaxed
200
Table6.M∗ andαforthecompositeLFs. clusters (both within 0.5R ). On the contrary, the bright-end
200
(M < −22.5) is not well represented by Schechter functions.
r′
Inanycase,weobtainχ2 valueshigherthan1.4.Therefore,the
contributionofthesebrightgalaxies(hereafterBGs)havetobe
R<R200 R<0.5R200 analyzedseparately(seeSect.5).
Wholegalaxypopulation
Sample M∗,δM∗ α,∆α M∗,δM∗ α,∆α
4.3.BlueandredLFs
DARC −22.00,0.41 −1.20,0.17 −21.64,0.35 −0.86,0.27
Relaxed −21.63,0.24 −1.03,0.20 −21.74,0.29 −0.99,0.21 We also estimate LF for red and blue galaxieswithin R and
200
Redgalaxies 0.5R200.WeselectredandbluegalaxiesfromtheRS(seeabove).
WeassumeredgalaxiesthoseobjectswithintheRS,thatisRS-
DARC −21.62,0.25 −0.76,0.22 −21.23,0.28 −0.47,0.27
Relaxed −21.51,0.28 −0.76,0.23 −21.50,0.36 −0.70,0.30 0.15< r′ −i′ <RS+0.15, while blue galaxies were selected as
r′−i′ <RS-0.15.ResultsarepresentedifFig.8andsummarized
Bluegalaxies
inTable6.
DARC −22.55,0.27 −2.03,0.10 −22.98,0.37 −1.79,0.24
When considering the whole galaxy population, no differ-
Relaxed −21.66,0.52 −1.44,0.46 −23.63,0.53 −2.07,0.23
encesarefoundbetweenDARCandrelaxedclusterswithinR
200
orinsidethecoreoftheclusters,within0.5R .Thedifferences
200
inM⋆andαarenotsignificantandwithin3σerrors(seeFig.9).
Thesameisalsotruefortheredgalaxypopulation.Weonlynote
Fig. 7 shows as the Schechter functionsfit reasonablywell
thecountsforgalaxieswithM >−22.5.Weobtain3χ2 =0.45 a marginaldifferencein the LF parametersof blue galaxies, in
r′
particularinthecoreoftheclusters,inside0.5R .
200
3 The χ2 statistic has been calculated considering three degrees of To check the significance of these differences, we fit the
freedom,thenormalizationcoefficient,M⋆andtheslope. Schechter profiles to the composite LFs fixing the slope pa-
7
R.Barrenaetal.:Environmentaleffectsonthebrightendofthegalaxyluminosityfunctioningalaxyclusters
rameter. Table 7 lists the results obtained. Again, we confirm
no differences between DARC and relaxed LFs. As above, we
onlyfindsmalldifferencesforthebluegalaxypopulationwithin
0.5R .However,thisdifferenceshouldbetakenwithcaresince
200
thebluegalaxypopulationisverypoorintheinternalregionsof
theclustersthuslimitingnumberofconsideredcounts.
Table7.M∗ forthecompositeLFs(atfixedα).
R<R R<0.5R
200 200
Wholegalaxypopulation (α=−1.00)
Sample M∗,δM∗ M∗,δM∗
DARC −21.61,0.07 −21.83,0.06
Relaxed −21.59,0.05 −21.75,0.06
Redgalaxies (α=−0.70)
Fig.10. Fraction of blue galaxies within 0.5R for DARC
DARC −21.54,0.07 −21.52,0.05 200
(black symbols) and relaxed clusters (red symbols). The solid
Relaxed −21.44,0.05 −21.48,0.06
linedescribesthelinearfitofthisrelation.Thedashedlinesshow
Bluegalaxies (α=−2.00)
one-sigmauncertaintiesofthelinearfit.
DARC −22.74,0.52 −23.18,0.11
Relaxed −22.38,0.11 −23.56,0.06
4.4.Butcher-Oemlereffect
The window opened by the redshift dependence of the galaxy
properties has been used to set constraints on the time-scales
ofthe processesofgalaxyevolution(Butcher& Oemler,1978,
1984;Stanfordetal.1998).Inthiscontext,theobservationalev-
idenceof the environmentaleffectis still uncertain.In orderto
clarify whether merging processes in clusters can enhance the
star formationin the galaxy populations,we compare the frac-
tionof bluegalaxies, f , inDARC andrelaxedclusters. So we
B
quantify the Butcher-Oemler effect (Butcher & Oemler, 1978,
1984)intwohomogeneousclustersampleswithsimilarvelocity
dispersionsandmasses.
StudiesofAndreonetal.(2006)showthattheradialdepen-
Fig.11.NumberofbrightgalaxieswithinR .Verticaldashed
200
denceofthebluefractionisquiteshallow,anditsmoothlyand
lines separate DARC and relaxed cluster samples. Labels indi-
monotonicallyincreasesfromthecenteroftheclustertothefield
cateindividualclusters.Theredpointsrepresentthemeanvalue
(seeFig.10inAndreonetal.2006).However,thecontamination
computedforeachclustersampleseparately.
contribution of field galaxies is lower in the inner regions. For
thisreason,weestimate f within0.5R .
B 200
Fig. 10 shows as f increases towards high redshifts being
B
at the 95% c.l. according to the Kendall statistics. A linear fit
mean richness as computed on all clusters. Here we multiply
gives f (z) = 1.18(±0.07)z−0.07(±0.06)(see also Sect. 7 for
B forthemeanrichnesstoobtainmorerealisticBG counts.Error
further discussions). We find no significant difference between
barsare assumedto bethe Poissonianerrors.Thefigureshows
f ofDARCandrelaxedclusters.
B as the BG counts are higher in DARC clusters than in relaxed
clusters. Considering the cluster regions within R , the mean
200
values for both samples separately are N = 23.1±3.1
5. ThebrightendoftheLF BG,DARC
andN 12.0±2.7forDARCandrelaxedclusters.Therefore,
BG,rel
Herewedefinebrightgalaxies(BGs)thosegalaxieswith Mr′ < DARCclusterscontainafactortwomoreBGsthanrelaxedclus-
−22.5andanalyzewhetherDARCandrelaxedclustersdifferfor tersandthedifferenceissignificantatthe2.7σc.l..
theirBGs,incountsorluminosity. We also study the spatial distribution of BG galaxies. We
compute BG counts in two separate regions, R < 0.5R and
200
0.5R < R < R (see Fig. 12). For the inner cluster region,
5.1.CountingBGs 200 200
R < 0.5R , we obtain N = 10.9±2.7 and N =
200 BG,DARC BG,rel
Fig. 11 shows the number of BGs for each cluster N = 7.2±2.4.Forthe outercluster region,0.5R < R < R , we
BG 200 200
Nr′<−22.5/Nr′<−20<˙N >, where Nr′<−22.5 is the numberof galax- obtainNBG,DARC = 20.2±2.5andNBG,rel = 8.3±2.1.Thusthe
ies brighter than M < −22.5 (normalized to the cluster rich- differencein BG countsis restricted to the outer region,where
r′
ness, see the above Sect. 4) and < N >=< N > is the issignificantat3.6σc.l..
r′<−20
8
R.Barrenaetal.:Environmentaleffectsonthebrightendofthegalaxyluminosityfunctioningalaxyclusters
wherethesumisperformedoveralltheN magnitudebinswith
galaxynumber N(m) and meanluminosity l(m). The transfor-
i i
mationfromabsolutemagnitudestoabsoluteluminosityinunits
of solar luminosities is performed using the solar magnitude
M = 4.62. The luminosity associated to BGs, L , is com-
r⊙ BG
putedinthesame wayoutto Mr′ < −22.5.Table9liststhelu-
minositiesforeachcluster.Valueswithinparenthesisareerrors
estimatedat1σc.l..
Fig.12.BG countsintheinnerregion,R < 0.5R (leftpanel)
200
Table9.LuminositycontentofBGs.
and in the outer region, 0.5R < R < R (right panel).
200 200
SymbolsarethesameasinFig.11.
Name L L
global BG
WecorrelateNBG withtheredshiftandthemassoftheclus- (1012L⊙) (1012L⊙)
ter. We do not observe any clear relation in both N − z and
BG A697 4.38(0.24) 0.68(0.04)
N −M relations.
BG 200 A773 10.71(2.33) 3.43(0.75)
Our above results are based on photometricBG counts op- A959 7.20(2.11) 2.53(0.74)
portunelycorrectedwiththeaveragefieldcounts(seeSect.4.1). A1240 16.14(4.74) 10.78(3.17)
Due to the poornumberstatistic of BGs in the individualclus- A2219 12.08(2.96) 3.32(0.81)
ters, we also use another approach to check the above results. A383 3.03(0.63) 2.90(0.60)
We also resortto the SDSS spectroscopicsurveysince redshift A963 6.29(0.67) 0.78(0.08)
datacanprovidethecorrectclustermembership.Unfortunately, A1835 16.95(2.74) 7.04(1.14)
SDSS surveyis notspectroscopicallycomplete.Indeed,within A2261 12.31(2.52) 2.55(0.53)
theclusterregionsofoursample,SDSSDR8provideredshiftsz ZwCl3146 6.95(3.52) 1.74(0.09)
for60%ofgalaxieshavingM <−22.5.Thus,wecannotassess
r′
the cluster membershipfor all BGs. Here we considera mixed
approach: we counts BGs in the photometric sample and then
weightthesenumbersonthebaseofthespectroscopicinforma- Note that LBG represents a non-negligible fraction of the
tion. globalluminosityfor Mr′ < −20andtheten clustersspanover
As for photometric data, Table 8 lists the mean number of a wide range of luminosities. We observe two dramatic cases,
BGsforDARCandrelaxedcluster(Cols.2and3);therespective these are A1240 in the DARC cluster sample and A383 in the
differenceandratiobetweenN inDARCandrelaxedclusters relaxedsample,wheretheluminosityfractionassociatedto the
BG
(Cols.4and5).Thesevaluesconfirmthatthestrongdifference BGs, fL,BG,areabout67%and96%ofthetotalluminosity,re-
betweenDARCandrelaxedclustersareduetotheoutercluster spectively.Consideringthewholesamples,wefindthatthemean
region,whilenothingcanbesaid aboutthe innerregiondueto luminosity fractions of BGs are fL,BG,DARC = 0.35±0.19 and
the poornumberstatistics. Note thathere we do notnormalize fL,BG,rel = 0.39 ± 0.33 for DARC and relaxed clusters. If we
with the respectivecluster richnessas in the abovesection:the donotconsiderthetwoextremecasesofA1240andA383,we
neteffectwouldbeofslightlyreducingtheobserveddifference find fL,BG,DARC = 0.27±0.09 and fL,BG,rel = 25±0.13. From
between DARC and relaxed clusters. Therefore, the results of these values, we conclude that DARC and relaxed clusters do
thissectionshouldbelookedatasbeingveryconservative. not significantly differ for the luminosity content associated to
We use the spectroscopic information to weight the above theirBGs.
ratios. We consider galaxy members those galaxies with a dif-
ference of δz = 0.01 (δz = 0.007) from the cluster red-
shift, i.e. δcz ∼ 3000 km s−1(∼ 2000 km s−1) from the mean 7. Discussion&conclusions
cluster velocity. Table 8 lists the number of cluster BGs over
We compare five unrelaxed clusters (DARC sample) with re-
the number of all BGs having redshift, N /N , for DARC
mem z laxedclusters.Asfaraspossiblewehavepotentialbiasesunder
and relaxed cluster (Cols. 6 and 7), and the values of the ra-
control.Therelaxedclustersaresimilartounrelaxedclustersfor
tiosDARC/relaxednowweightedwiththerespective N /N ,
mem z theirredshiftrange,X-raytemperature,andmass(see Tables1
i.e.WRatio(DARC/rel)=Ratio(DARC/rel)*[N /N (DARC)]×
mem z and 2).Moreover,we alwaysworkwithinphysicalratherthan
[N /N (rel)]−1 (Col. 8). When comparing the values of
mem z fixedradii.
“WRatio”withthoseofabove“Ratio”wesee thattheresultof
Wefindthatveryunrelaxedarecomparabletorelaxedclus-
DARC clusters beingricherin BGs than relaxedcluster within
ters for the statistical properties, i.e. the parameters of the LF
0.5R isconfirmed.
200 andtherelativebehaviourbetweenredandbluegalaxies.Inpar-
ticular,weagreewithaseriesofpreviousresults.
6. BGsLuminosity We compare M∗ and α parameters of the composite
SchechterluminosityfunctionswithresultsobtainedbyPopesso
WeinvestigatewhetherthecontributioninluminosityofBGsis etal.(2006).Transformingtheir M∗ estimates(seetheirTables
differentbetweenDARC and relaxedclusters. We computethe 1 and 2) to our cosmology, applying the factor 5log(h ), we
70
“global”luminosityfor Mr′ < −20 usingthe countsof Sect. 4, obtainvaluesinperfectagreement.
i.e.: In addition, in agreement with Popesso et al. (2006, their
Fig. 10) and de Filippis et al. (2011, their Fig. 11) the LF pro-
N
file does not depend on the considered cluster region within
Lglobal =XNi(m)li(m), (4) M <−19.
i r
9
R.Barrenaetal.:Environmentaleffectsonthebrightendofthegalaxyluminosityfunctioningalaxyclusters
Table8.BGsinphotometricsamplesweightedwithspectroscopicsamples.
fromphotometricsurvey fromspectroscopicsurvey photo+spec
N N Diff. Ratio N /N(DARC)a N /N(rel)a WRatioa
BG,DARC BG,rel mem z mem z
R<R 65 45 20±10 1.4 10/41(8/41) 8/25(6/25) 1.1(0.9)
200
R<0.5R 13 16 3±6 0.8 4/13(4/13) 4/8(4/8) 0.5(0.5)
200
R>0.5R 52 29 23±9 1.8 6/28(4/28) 4/17(2/17) 1.6(2.2)
200
a Wheremembergalaxiesarethosehavingδz=0.01(δz=0.007).
Wealsofindevidenceofthevariationofthebluefractionof Institutions. TheSDSSWebsiteis http://www.sdss.org/.Thisworkhas
galaxies with redshift. It is very noticeably that we can detect beenfundedbytheSpanishMinistryofScienceandInnovation(MICINN)under
the effect even if our study is limited to a smallredshiftrange. thecollaborationgrantsAYA2010-21887-C04-04andAYA2010-21322-C03-02.
MGacknowledgesfinancialsupportfromPRIN-INAF-2010.
Inparticular,notethatextrapolatingour f (z)relationtoz=0.35,
B
weobtaina f (z=0.35)=0.34±0.05.Thisestimatewellagrees
B
withthoseobtainedbyAndreonetal.(2006).Theyfind f (z =
B
0.35)=0.33±0.05withclusterslocatedinaredshiftrange0.3− References
0.4.Whenextrapolatingtoz=0.5,weobtain f (z=0.5)∼0.5.
B Allen,S.W.,Ettori,S.,&Fabian,A.C.2001,MNRAS,324,877
Thisfindingisinverygoodagreementwithcosmologicalgalaxy Allen, S.W.,Schmidt,R.W.,Ebeling, H.,Fabian,A.C.,vanSpeybroeck, L.
formationmodelsforclusterwithsimilarmass(seeMencietal. 2004,MNRAS,353,457
2008,theirFig.3). Andreon,S.,Quintana,H.,Tajer,M.,Galaz,G.,Surdej,J.2006,MNRAS,365,
915
Our new first result is that relaxed clusters contain fewer
Ascaso,B.,Aguerri,J.A.L.,Varela,J.,etal.2011,ApJ,726,69
BGs and these BGs are more concentrated in cluster inner re-
Baldi,A.,Ettori,S.,Tozzi,P.&Borgani,S.2007,ApJ,666,835
gions.Onthecontrary,unrelaxedclusterspresentmoreBGsand Barkhouse,W.A.,Yee,H.K.C.,&Lo´pez-Cruz,O.2007,ApJ,671,1471
more homogeneouslydistributed within the whole cluster. The Barrena,R.,Boschin,W.,Girardi,M.,&Spolaor,M.2007a,A&A,467,37
BG richnessof DARC clusters is the robustand significantre- Barrena,R.,Girardi,M.,Boschin,&Das´ı,M.2009,A&A,503,357
Bautz,L.P.,&Morgan,W.W.1970,ApJ,162,149
sultofSect.5.
Binggeli,B.,Sandage,A.,Tamman,G.A.1988,ARA&A,26,509
Our result is not in contrast with that of de Propris et al. Biviano, A.,Murante, G.,Borgani,S.,Diaferio, A.,Dolag,K.,&Girardi, M.
(2003) who found similar LF between substructured and non 2006,A&A,456,23
substructuredclusters since: i) their study is devotedto LF pa- Boschin,W.,Barrena,R.,Girardi,M.,&Spolaor,M.2008,A&A,487,33
Boschin,W.,Girardi,M.,Barrena,R.,etal.2004,A&A,416,839
rameters, for which we find no difference, too; ii) our sample
Butcher,H.,&Oemler,A.1978,ApJ,219,18
ofDARC clustersare likely to bemuchmorefar fromdynam- Brunetti,G.,Cassano,R.,Dolag,K.,&Setti,G.2009,A&A,507,661
ical equilibrium than their sample of substructured clusters. In Buote,D.A.2002,in“MergingProcessesinGalaxyClusters”,eds.L.Feretti,
fact,DARCclustersarefoundtobecasesofmajor,veryimpor- I.M.Gioia,&G.Giovannini(TheNetherlands,KluwerAc.Pub.):Optical
tant mergers, in agreement with the rarity of the diffuse radio AnalysisofClusterMergers
Butcher,H.,&Oemler,A.1984,ApJ,285,426
sources phenomena (∼ 10% of clusters, Giovannini & Feretti
Carlberg,R.G.,Yee,H.K.C.,Ellingson,E.,etal.1996,ApJ,462,32
2002). Moreover, DARC clusters should all be recent mergers Carlberg,R.G.,Yee,H.K.C.,&Ellingson,E.1997,ApJ,478,462
sinceradioemissionsareexpectedtohaveashortlifetime,i.e. Cassano,R.,Ettori,S.,Giacintucci,S.,etal.2010,ApJ,721,82
oftheorderofafew108years(e.g.,Giovannini&Feretti2002; deFilippis,E.,Paolillo,M.,Longo,G.,etal.2011,MNRAS,414,277
deLucia,G.&Blaizot,J.2007,MNRAS,375,2
Skillman et al. 2010), in agreement with times estimated in a
dePropris,R.,Pritchet,C.J.,Harris,W.E.,McClure,R.D.1995,ApJ,450,534
few individualclusters (e.g.,Markevitchet al. 2002; Girardiet dePropris.,R.,etal.2003,MNRAS,342,725
al.2008). Dressler,A.1978,ApJ,223,765
Oursecond,newresultisthattheluminositycontentinBGs Dressler,A.1980,ApJ,236,351
Driver,S.P.,Couch,W.J.,&Phillipps,S.1998,MNRAS,301,369
isthesameinDARCandrelaxedclusters.Theconsequentsce-
Dubinski,J.1998,ApJ,502,141
narioisthatthe(morenumerous)BGslyingintheouterregions
Durret,F.,Lagana´,T.F.,&Haider,M.2011,A&A,529,A38
of merging clusters will merge to form (more luminous) BGs Ebeling,H.,Voges,W.,Bo¨hringer,H.,etal.1996,MNRAS,281,799
in the inner regions of relaxed clusters. This combined forma- Ebeling,H.,Edge,A.C.,Bo¨hringer,H.,etal.1998,MNRAS,301,881
tion/evolution of BGs and the parent clusters well agree with Ensslin,T.A.,Biermann,P.L.,Klein,U.,&Kohle,S.1998,A&A,332,395
Feretti,L.,GioiaI.M.,andGiovanniniG.eds.,2002b,AstrophysicsandSpace
the resultsof Lin& Mohr(2004) whoanalyzedthe correlation
ScienceLibrary,vol.272,“MergingProcessesinGalaxyClusters”,Kluwer
betweenBCG luminosityandparentclustermass,supportinga AcademicPublisher,TheNetherlands
recent formation of the brightest galaxies in the context of the Feretti, L. 2006, Proceedings of the XLIst Rencontres de Moriond, XXVIth
hierarchicalscenario. Astrophysics Moriond Meeting: ”From dark halos to light”, L.Tresse, S.
MaurogordatoandJ.TranThanhVan,Eds,e–printastro–ph/0612185
We plan to extend our study to a larger sample of DARC
Feretti,L.2008,Mem.SAIt,79,176
clusters,byexaminingtheirdynamicstocomputemorereliable
Fukugita,M.,Shimasaku,K.,&Ichikawa,T.1995,PASP,107,945
massestimates.Anotherpossibleextensionofthisworkislook- Gilbank,D.G.,Yee,H.K.C.,Ellingson,E.,etal.2008,ApJ,673,742
ingatdeepermagnitudesthusto investigatepossible effectson Giovannini,G.,&Feretti,L.2002,in“MergingProcessesinGalaxyClusters”,
thefaintendofthegalaxyLFconnectedwiththeclusterevolu- eds.L.Feretti,I.M.Gioia,&G.Giovannini(TheNetherlands,KluwerAc.
Pub.):DiffuseRadioSourcesandClusterMergers
tion.
Girardi,M.,Boschin,W.,&Barrena,R.2006,A&A,455,45
Girardi,M.,Fadda,D.,Giuricin,G.etal.1996,ApJ,457,61
Acknowledgements. Weare indebt with Andrea Biviano forhis useful com- Girardi,M.,Giuricin,G.,Mardirossian,F.,Mezzetti,M.,&Boschin,W.1998,
mentsonthiswork.WealsothanksforthecreationanddistributionoftheSDSS ApJ,505,74
Archive, provided by the Alfred P. Sloan Foundation and other Participating Girardi,M.,&Mezzetti,M.2001,ApJ,548,79
10