Tracking Down a Critical Halo Mass for Killing Galaxies through the Growth of the Red-Sequence

0.12 MB·English

by  David G. Gilbank

Mon.Not.R.Astron.Soc.000,1–5(2007) Printed2February2008 (MNLaTEXstylefilev2.2)
Tracking Down a Critical Halo Mass for Killing Galaxies through
the Growth of the Red-Sequence
⋆
David G. Gilbank and Michael L. Balogh†
8 AstrophysicsandGravitationGroup,DepartmentofPhysics&Astronomy,UniversityofWaterloo,Waterloo,Ontario,CanadaN2L3G1
0
0
2
2February2008

n
a
J
ABSTRACT

4 Red-sequence galaxies record the history of terminated star-formation in the Universe and
1 canthusprovideimportantcluestothemechanismsresponsibleforthistermination.Wecon-

structcompositesamplesof publishedclusterandfield galaxyphotometryinordertostudy
]
h the build-up of galaxies on the red-sequence, as parameterised by the dwarf-to-giant ratio
p (DGR). We find thatthe DGR in clusters is higherthan thatof the field at all redshifts,im-
- plyingthatthefaintendofthered-sequencewasestablishedfirstinclusters.Wefindthatthe
o
DGRevolveswithredshiftforbothsamples,consistentwiththe“down-sizing”pictureofstar
r
t formation. We examine the predictions of semi-analytic models for the DGR and find that
s
neitherthemagnitudeofitsenvironmentaldependencenoritsevolutioniscorrectlypredicted
a
[ in the models.Red-sequenceDGRs are consistentlytoo highin the models,the mostlikely
explanation being that the strangulation mechanism used to remove hot gas from satellite
1
galaxiesis too efficient. Finally we presenta simple toy model includinga thresholdmass,
v
belowwhichgalaxiesarenotstrangled,andshowthatthiscanpredicttheobservedevolution
0
ofthefieldDGR.
3
9 Keywords: galaxies:clusters:general–galaxies:evolution–galaxies:general
1
.
1
0
8
1 INTRODUCTION galaxieswhichareredaretoohighincurrentsemi-analyticmodels.
0
Withtherecentabundanceofsurveystargetingred-sequencegalax-
: Red-sequencegalaxiesareanimportantpopulationforunderstand-
v iesboth in thecluster and thefield, the timeisnow ripe todraw
i inggalaxyformationandevolutioningeneral.Theyrepresentsys- togethertheseworkstobuildapictureofthebuild-upofgalaxies
X temswithverylittleornoon-goingstar-formationandthusarea
onthered-sequenceasafunctionofenvironment andtoconfront
r uniquetracerofthepastactivityofgalaxies.DeLuciaetal.(2004)
a foundadeficitoffaintred-sequence galaxiesinclustersatz∼0.8 thesewithpredictionsofgalaxyformationmodels.
relative to local clusters. This would suggest that star formation
endedearlierforthemostluminous/massivegalaxiesathighred-
shift and would support the “down-sizing” picture first proposed
2 OBSERVATIONALDATA
byCowieetal.(1996),inwhichtheterminationofstar-formation
progressesfromthemostmassivetotheleastmassivegalaxiesas We aim to construct the most uniform sample possible using
theuniverseages.Similarresultsforclustersspanning arangeof colour-selected red-sequence galaxy data from the literature. To
redshifts have been found by Tanakaetal. (2005), deLuciaetal. thisend,wetransformallourdataontothesamesystem.Asimpler
(2007),Stottetal.(2007),Gilbanketal.(2007) andothers. Simi- quantity than the luminosity function (LF), which has been em-
larly, surveys of field galaxies have attempted to characterise the ployed in thepast, isthe red-sequence dwarf-to-giant ratio, here-
evolution of the red-sequence luminosity function. The relative after DGR. This is essentially just a LF reduced to two bins and
contributions of passive evolution/termination of star-formation avoidsthecomplicationsofhavingtofitananalyticfunction(usu-
andnumberdensityevolution/drymergingarestillopenquestions allywithdegeneracies between thefittedparameters) tothedata.
(e.g.,Belletal.2004,Cimattietal.2006). Following DeLuciaetal. (2004), we define luminous, or giant,
Red-sequence galaxies can thus place constraints on galaxy galaxiesasthosebrighterthanMV = −20.0andfaint,ordwarfs,
formation prescriptions in theoretical models. Recent observa- asthosewith−20 < MV 6 −18.2(k-andevolution-corrected).
tions in the local universe using SDSS data (Baldryetal. 2006; Where data are presented on a different photometric system, we
Weinmannetal. 2006) have found that the fractions of satellite use Bruzual&Charlot (2003) stellar population models to calcu-
late the transformation using a z = 3, Solar metallicity, single
f
stellar population with a Chabrier IMF. Where DGRs have been
⋆ Email:[email protected] given in the original paper using these limits (as is the case for
† Email:[email protected] muchoftheclusterdata)weusethesenumbersandtheirassociated
2 Gilbank& Balogh.
errorsdirectly.Forotherworks,wetaketheM⋆andαvaluesfrom errors. As found by e.g., Gilbanketal. (2007) and Hansenetal.
theSchechterfunctionfit,transformtoMVandintegratethetrans- (2007),thefaintendslopeoftheclusterRSLFdependsoncluster
formedLFtocalculateDGR.Weestimateconservativeerrorsusing richness/mass(atleastforz<∼0.5),inthesensethatlowmassclus-
aMonteCarlomethodtakingtheerrorsinthefittedparametersand ters/richgroups areexpectedtohavesystematicallylowerDGRs.
allowingfortheircovariancebyemployingtheerrorellipsegiven Inthis compilation, wehave attempted toutilisethe richest/most
intheoriginal reference. In addition wepropagate a0.05 magni- massivesystemsateachredshiftfortheclustersample,tomaximise
tudeuncertainty(discussedbelow)inthemagnitudelimitstoallow thedifferencebetweentheclustersandthefield.Insomuchas,on
foranyresidualmismatchintheconvertedphotometricsystem. average,themostmassivesystemsatoneredshift evolveintothe
For the field samples, we carefully examine all of the mostmassivesystemsatlowerredshift,suchaselectionisappro-
Schechterfunctionfitsineachredshiftbinandonlyusemeasure- priateforevolutionarystudies.Ifweconsiderthethreedatapoints
ments where the data fully cover the dwarf and giant range. For nearesttoz∼0.5,theseweredrawnfromverydifferentlyselected
example,theBrownetal.(2007,NOAODWFS)datadoagoodjob samples:astatisticalsampleof54red-sequenceselectedclusters,
of constrainingtheshape oftheLFoverthemagnitude rangewe havingrichnessescorrespondingtoamasslimitof>∼6×1013M⊙
require for z60.5 but at higher redshift are insufficiently deep to (Gilbanketal.2007);asampleof11clustersselectedviatheirun-
constrainthenumberofdwarfs.Conversely,theZuccaetal.(2006, resolvedclusterlightfromshallowopticalimaging,havingvarious
VVDS)data aresufficiently deep to measure thefaint end of the masses(deLuciaetal.2007);andasampleoftenofthemostX-
LFathigherredshift,butatz60.6aresubjecttoconsiderableun- rayluminousclusters(Stottetal.2007).Thegoodagreementofthe
certaintyaroundM⋆,presumablyduetotherelativelylimitedarea pointsfromtheseensembleclusterssuggeststhattheDGRisrela-
ofthesurvey.Theremainderofourfieldsampleusespointsfrom tivelyinsensitivetomassorselectionmethodoverthismassrange.
Baldryetal. (2004, SDSS), Belletal. (2003, SDSS), Belletal. Furthermore, it suggests that the evolution of the cluster DGR is
(2004,COMBO-17),Driveretal.(2006,MillenniumGalaxyCata- likelynotduetoasystematicallychangingmasslimitwithredshift.
logue)andScarlataetal.(2007,COSMOS).Fig.1showsDGRasa Additionally,thevirialradiusofaclusterdependsonitsmass
functionofredshiftforthesesurveys.Wheretheredshiftrangesof roughly as r ∝ M1/3 (e.g., Carlbergetal. 1997), so using a
vir
differentsurveysoverlap,wefindgoodagreementinourmeasured fixedphysicalradiuswithinwhichtomeasureapropertyofaclus-
DGRs(e.g.,VVDSandCOSMOSatz=0.7). terprobesadifferentfractionofthecluster’sextentdependingon
Clusters show considerable scatter in their individual DGRs themassofthecluster.TheeffectofthisonDGRissmallascan
(e.g., deLuciaetal. 2007) and so we only focus on works mea- beseenfromfig.5ofdeLuciaetal.(2007).Bylimitingthemass
suring DGR from ensemble averages of clusters in each redshift range of the clusters to the most massive systems, we have min-
bin (deLuciaetal. 2007, Stottetal. 2007, Gilbanketal. 2007, imisedthe possible impact of such aneffect, since thedifference
Hansenetal.2007,Barkhouseetal.2007). between a fixed physical radius and a fixed fraction of the virial
Tworesultsareimmediatelyapparentfromtheobservational radiusonlydependsonthemasstothe1/3power.
data in Fig. 1: the DGR of field red-sequence galaxies is always
lower than that of cluster galaxies at the same redshift, and the
2.2 Conflictingresults
DGR of both samplesevolves withredshift. Weparameterise the
evolution of the DGR with a fit of the form (1 + z)β finding Andreon(2007)recentlyargued,basedonasmallnumberofindi-
β = −1.8±0.5 and β = −3.4±1.1 for the cluster and field vidualclusters,thattheclusterDGRdoesnotappeartoevolvewith
samplesrespectively. redshift. Weindicate hisdatapoints onFig. 1,omitting thethree
z>1clusters for which thedata do not bracket the 4000A˚ break.
Alloftheindividualclustersareinreasonableagreementwiththe
2.1 Sourcesofsystematicerror
ensemble averages, and with our best fit line, except for one at
The first systematic to be addressed is that of transforming pub- z=0.89.Furthermore,hissampleonlycontainstwoz<0.5clus-
lishedphotometryondifferentsystemstoauniformone.Mostof ters,oneofwhichisComa,forwhichtheDGRisingoodagree-
the published cluster studies give rest-frame V magnitudes (e.g., ment with the deLuciaetal. (2007) value. Given the significant
deLuciaetal.2007),andmostfieldstudiesgiverest-frameB(e.g., cluster-to-clusterscatter,theensemblesamplescompiledherepro-
Belletal.2004).Thisisnotexclusivelythecase,however,andwe videabetterreflectionoftheglobalaverageinclusters,andallow
find good agreement where different studies measured in differ- ustomeasuresignificantevolutionwithredshift.
ent passbands inthesameenvironments atsimilarredshiftsover- Tanakaetal. (2005) and deLuciaetal. (2007) both con-
lap(e.g.,Hansenetal.2007;Stottetal.2007).Thuswebelievethe structed composite samples of local SDSS clusters from the
transformationisnot alargesourceof error.Wehavepropagated C4 cluster catalogue (Milleretal. 2005). However, Tanakaetal.
agenerous0.05magnitudeerrorthroughouranalysistoallowfor (2005) found a low DGR, comparable withtheir two higher red-
this.Thisalsoallowsforthedominant transformationuncertainty shiftclusterz=0.55andz=0.83)andthereforeconcludedthatthe
which is that of the systematic calibration in the individual sur- clusterDGRmightnotevolvewithredshift.Bothoftheirhigher-z
veys(e.g.,Hansenetal.2007quotetheircalibrationuncertaintyas individualclusterslieonthebest-fitrelationofFig.1,sothedis-
0.03mag).Infact,comparingHansenetal.(2007)andStottetal. crepancy arises from the z∼0 point. deLuciaetal. (2007) found
(2007),themeasuredDGRsagreewithintheerrorsevenifthisad- a higher DGR (for their σ > 600 km s−1 sample) more consis-
ditionaluncertaintyisneglected.So,webelievethatourerrorbars tentwiththeresultsofHansenetal.(2007)fortherichestclusters;
areample. Stottetal. (2007) for the most X-ray luminous clusters; and also
Theimportantothersystematicforfieldstudiesiscosmicvari- fortheregionsofhighestdensityintheSDSS(Baldryetal.2006)
ance. The authors of thevarious works included here discuss the (opensquare). Wesuggest thatthereasonfor thelow DGRmea-
size of this effect for their surveys, and this is not likely to be a suredbyTanakaetal.(2005) maybethat theirsamplewas dom-
major contribution to the error bars in Fig. 1. Cluster studies re- inated by lower mass systems. We note that their measurement
quire more care in the sample selection to minimise systematic is just consistent with the lowest DGR point at z∼0.1 in Fig. 1
Red-SequenceDwarf-to-GiantRatios 3
A
A
A
A
A A
T A AAAT A
T A
1.0 A
R
G
D
1012M
O
Clusters (ensemble) •
Field 1012.5MO
Millennium clusters •
Millennium field
1013M
O
0.1 •
0.0 0.2 0.4 0.6 0.8 1.0
z
Figure1.TheevolutionoftheDwarf-to-Giant ratio,DGR,forvarioussamplestakenfromtheliterature andconverted toauniformsystem,asdescribed
inthetext.Filledcirclesandopendiamondsdenotecluster(Barkhouseetal.2007,deLuciaetal.2007,Gilbanketal.2007,Hansenetal.2007,Stottetal.
2007)andfield(Belletal.2003,Baldryetal.2004,Belletal.2004,Driveretal.2006,Zuccaetal.2006,Brownetal.2007,Scarlataetal.2007)samples
respectivelyfromobservationaldata,andlinesshow(1+z)βfits.Horizontalerrorbarsindicatetheredshiftrangecoveredbythedata.Nakederrorbarsare
measurementsforindividualclustersfromAndreon(2007)andTanakaetal.(2005),labelled‘A’and‘T’respectively. Onlyensembleclustersareincluded
inthefit.ThefilledsquareshowsthehighestdensityregionsintheSDSSfromBaldryetal.(2006)whichagreeswellwiththeclusterdata.Asterisksand
trianglesshowpredictionsfromtheBoweretal.(2006)semi-analyticmodel,connectedwithlinestoguidetheeye.Thedot-dashedlinesshowthepredicted
evolutionofthefieldDGRifisolatedgalaxieshaveaconstantDGR=0andtheobservedtrendisduetotheincreasingabundanceofgroups(havingthesame
DGRasclusters)abovethelabelledthresholdmass,M withcosmictime,asdescribedin§4.
th
whichcomesfromBarkhouseetal.(2007)andcomprisedasetof at agiven redshift. Themodel predicts approximately the correct
57Abellclustersoverarangeofrichnesses. DGRforclustersinthelocaluniverse,butthisDGRvaluethenre-
mainsconstantouttoz∼1.Theseresultsareinsensitivetotheexact
halomasschosentodefinetheclusters,ascanbeinferredfromthe
similaritybetweentheclusterandfieldpointsateachredshift.i.e.,
3 PREDICTIONSFROMTHEGALFORM theDGRisnotastrongfunctionofhalomassinthemodel.
SEMI-ANALYTICMODELS
WedrawsamplesfromtheMillenniumsimulationusingthesemi-
analyticprescriptionofBoweretal.(2006),hereafterB06.Galax-
iesareselectedbasedontheirhalomass(>1014 M⊙ forclusters) 4 DISCUSSION&CONCLUSIONS
and their observed colours and magnitudes in this prescription.
Red-sequence galaxies were selected by fittinga Gaussian to the Previousworkshavesuggestedthatthereappearstobelittleorno
redpeakofthe(B−V)colourdistributionandlocatingthelocal evolution in the bright end of the RSLF, both in clusters and the
minimumonthebluesidetoisolatetheextentofthered-sequence. field (e.g., Gilbanketal. 2007; Scarlataetal. 2007). If this is the
Inordertocapturetheuncertaintiesinsamplingthesimulationin case, the evolution of the DGR can be attributed to the build-up
thesamewayastheobservations,wevarytheslopeandnormali- offaintgalaxiesonthered-sequencewithincreasingcosmictime.
sationofthisred-sequencelimit(by15%and3%respectively)as Thisimpliesthatstar-formationisterminatedingiantgalaxiesfirst
well asthemagnitude limitsused todefineDGR (0.1 mags each and later in dwarf galaxies, inagreement with the“down-sizing”
limit)andMonte-Carlotheresults.Thepointsforclustersandfield pictureofstar-formation.Furthermore,thelowervalueofDGRin
inthreeredshiftbins(0.0,0.5,1.0)areshownasasterisksandtri- the field compared with clusters implies that star-formation was
anglesrespectivelyinFig.1.Statisticalerrorsarenegligible,sothe completedinclustergalaxiesbeforefieldgalaxies.
errorbarsreflectthissystematicvariation. PerformingasimilaranalysisontheB06semi-analyticmodel
When confronted with the observational data, the semi- predictsanon-evolvingDGRouttoz=1,withlittledifferencebe-
analytic model fails to reproduce either the redshift evolution of tweenclustersandthefield.Sucharesultsuggeststhatfaintgalax-
the DGR or the relative difference between clusters and the field iesmoveontothered-sequence tooefficientlyinthemodel,with
4 Gilbank& Balogh.
this build-up of the faint end of the red-sequence already having oftheclusterDGRwehavehadtoincludeinourmodelmayreflect
beencompletedbytheearliestepochconsideredhere(z∼1). theeffectofastrangulationtimescale.
Disentangling the effects of a timescale and threshold mass
Themethodsavailableinthemodelsforthequenchingofstar-
for thequenching of satellitegalaxies isacomplex observational
formationinclude:feedbackfromsupernovaeorAGNandhothalo
problem. The colour difference between central and satellitered-
gas removal. This latter process only applies to galaxies which
sequence galaxies may be a useful test of models. For example,
are satellites (i.e. not the most massive, central object) in their
in the B06 prescription, it is predicted that satellitered-sequence
dark matter halo. The relative importance of the various quench-
galaxiesaresignificantlyredderthancentralred-sequencegalaxies.
ing mechanisms are summarised in fig. 4 of B06. At the bright
vandenBoschetal.(2007)recentlymeasuredtheaveragecolours
end of the red-sequence, most of the galaxies are the central ob-
of satelliteand central galaxies in the local universe from an ap-
jectintheirdarkmatterhaloandthusfeedbackfromAGNorsu-
proximatelyhalomass-selectedsampleandfoundthattheformer
pernovawindsarerequiredtoturnoffstar-formation.Atthefaint
wereindeedonaverageredder. Theyalsoshowedthatmeasuring
end(inthedwarfregime),red-sequencegalaxiesarepredominantly halomassesfromstatisticalsamplesfor∼1012M⊙systemsisnot
satellites.Wehaveverifiedthatthisisstillthecaseforallthered-
possible with high completeness and thus, if there is a threshold
shifts we consider here. In most galaxy formation models (e.g.,
massfor strangulationaround thislimit,suchstudieswillrequire
B06, DeLuciaetal. 2006), when any galaxy enters a more mas-
detailedtargetedobservationsofgroups(e.g.,Wilmanetal.2005).
sivedarkmatterhaloandbecomesasatellite,itinstantlyhasallof
Analternativeapproachtogetatthetimescaleforstrangulationis
itshotgasremovedinaprocessgenerallyreferredtoas‘strangula-
thestudyofthestarformationhistoriesofindividualgalaxiesun-
tion‘(Larsonetal.1980;Baloghetal.2000).Thus,onceithascon-
dergoingtransformation(e.g.,Moranetal.2007).Itispossiblethat
sumedthecoldgasalreadypresentinitsdisk,itsfuelcannolonger
theobservationsinFig.1,coupledwithnewstrangulationprescrip-
bereplenishedbyhotgasfromthehaloanditsstar-formationister-
tionsinthesemi-analyticmodelsmayimproveconstraintsonthe
minated.Thisprescriptionmaybetooefficient.Recentsimulations
likelyvaluesofthethresholdmassandtimescaleforstrangulation.
byMcCarthyetal.(2007)foundthattheremovaloftheinitialhot
galactichalogasdoesnotoccurinstantaneously,butlikelytakes∼
1Gyrtoremovethebulkofit,andthatasmuchas30%canstill
remainafter10Gyr.Inaddition,themodelsofDekel&Birnboim
ACKNOWLEDGMENTS
(2006)predicttheexistenceofacriticalhalomass(∼1012M⊙)for
haloes,belowwhichstrangulationmaynotoccur.Thus,weexpect Thisresearch wassupported byanEarlyResearcher Awardfrom
additional parametersinvolving strangulation efficiency/timescale theprovinceofOntario,andbyanNSERCDiscoverygrant.
and/or a threshold halo mass for strangulation to be important in
tuningthemodelstoreproducetheobservationspresentedinFig.1.
To illustratethe effect of a threshold halo mass for strangu-
REFERENCES
lation, we use the following toy model to predict the evolution
of the field DGR. Field surveys do not just probe isolated (i.e. Andreon,S.2007,MNRASsubmitted,astro-ph/710.2737
non-group) galaxies.Indeed,theyaredominatedbygroups,espe- Baldry, I. K., Balogh, M. L., Bower, R. G., Glazebrook, K.,
ciallytoward lower redshift ascosmic structuregrows. Thusit is Nichol, R. C., Bamford, S. P., & Budavari, T. 2006, MNRAS,
group-scalephysicswhichisimportantindeterminingtheproper- 373,469
tiesofgalaxiesinfieldsurveys.Weassumefurtherthatgalaxiesin Baldry,I.K.,Glazebrook,K.,Brinkmann,J.,Ivezic´,Zˇ.,Lupton,
haloesbelowsome(group-size)thresholdmass,M ,havenored- R.H.,Nichol,R.C.,&Szalay,A.S.2004,ApJ,600,681
th
sequence satellites, i.e. DGR = 0. We assume that haloes above Balogh,M.L.,Navarro,J.F.,&Morris,S.L.2000,ApJ,540,113
M haveaDGRgivenbythatoftheobservedclusterpopulation Barkhouse,W.,Yee,H.K.C.,&Lopez-Cruz,O.2007,submitted
th
at the same redshift (from the best-fit cluster line in Fig. 1). We Bell,E.F.,McIntosh,D.H.,Katz,N.,&Weinberg,M.D.2003,
consider Mth = (1012,1012.5,1013) M⊙ and measure the frac- ApJS,149,289
tional contribution of haloes in the appropriate mass range from Bell, E. F., Wolf, C., Meisenheimer, K., Rix, H.-W., Borch, A.,
theMillenniumSimulation.Weoverplottheexpectedevolutionof Dye, S., Kleinheinrich, M., Wisotzki, L., & McIntosh, D. H.
thefieldDGRforthedifferentthresholdmassesinFig.1.Itcanbe 2004,ApJ,608,752
seenthatboththeMth = 1012 and1012.5 M⊙ modelsgivegood Bower,R.G.,Benson,A.J.,Malbon,R.,Helly,J.C.,Frenk,C.S.,
agreementwiththeobservedfieldDGR.Inordertoreproducethe Baugh,C.M.,Cole,S.,&Lacey,C.G.2006,MNRAS,370,645
observations so well, our model requires both an increasing con- Brown,M.J.I.,Dey,A.,Jannuzi,B.T.,Brand,K.,Benson,A.J.,
tributiontothefieldbygroups due togrowth ofstructureand an Brodwin,M.,Croton,D.J.,&Eisenhardt,P.R.2007,ApJ,654,
intrinsic evolution of the DGR in haloes above M . If either of 858
th
thesefactorsisremoved(e.g.ifaconstantgroupfractionatallred- Bruzual,G.&Charlot,S.2003,MNRAS,344,1000
shifts or a constant cluster DGR at all redshifts is assumed) then Carlberg,R.G.,etal.1997,ApJ,485,L13
thepredictedevolutionofthefieldDGRistooflatwithrespectto Cimatti,A.,Daddi,E.,&Renzini,A.2006,A&A,453,L29
the observations. If groups do not exhibit exactly the same DGR Cowie,L.L.,Songaila,A.,Hu,E.M.,&Cohen,J.G.1996,AJ,
as clusters, but instead are intermediate between clusters and the 112,839
field(thisiscurrentlyanopenquestion),thenthiswouldeffectively DeLucia,etal.2004,ApJ,610,L77
shiftourpredictedlinesdownandalowerM wouldthenrepro- DeLucia,G.,Springel,V.,White,S.D.M.,Croton,D.,&Kauff-
th
ducetheobservations.Ofcourse,ascenarioinwhichthetimescale mann,G.2006,MNRAS,366,499
fortheremovalofhotgasismuchlongermayfittheobservations deLucia,eta;.2007,MNRAS,374,809
equallywellandthisorsomealternativecombinationofthreshold Dekel,A.&Birnboim,Y.2006,MNRAS,368,2
mass/timescalecannotberuledout.Indeed,theobservedevolution Driver,S.P.,etal.2006,MNRAS,368,414
Red-SequenceDwarf-to-GiantRatios 5
Gilbank,D.G.,Yee,H.K.C.,Ellingson,E.,Gladders,M.D.,Loh,
Y..,Barrientos,L.F.,&Barkhouse,W.A.2007,ApJaccepted,
astro-ph/710.2351
Hansen,S.M.,Sheldon,E.S.,Wechsler,R.H.,&Koester,B.P.
2007,ApJsubmitted,astro-ph/710.3780
Larson,R.B.,Tinsley,B.M.,&Caldwell,C.N.1980,ApJ,237,
692
McCarthy,I.G.,Frenk,C.S.,Font,A.S.,Lacey,C.G.,Bower,
R.G.,Mitchell,N.L.,Balogh,M.L.,&Theuns,T.2007,MN-
RASaccepted,astro-ph/710.0964
Miller,C.J.,etal.2005,AJ,130,968
Moran, S. M., Loh, B. L., Ellis, R. S., Treu, T., Bundy, K., &
MacArthur,L.A.2007,ApJ,665,1067
Scarlata,C.,etal.2007,ApJS,172,494
Stott,J.P.,Smail,I.,Edge,A.C.,Ebeling,H.,Smith,G.P.,Kneib,
J..,&Pimbblet,K.A.2007,ApJ,661,95
Tanaka,M.,Kodama,T.,Arimoto,N.,Okamura,S.,Umetsu,K.,
Shimasaku, K.,Tanaka, I.,&Yamada, T. 2005, MNRAS,362,
268
vandenBosch,F.C.,Aquino,D.,Yang,X.,Mo,H.J.,Pasquali,
A.,McIntosh,D.H.,Weinmann,S.M.,&Kang,X.2007,ArXiv
e-prints,710
Weinmann,S.M.,vandenBosch,F.C.,Yang,X.,Mo,H.J.,Cro-
ton,D.J.,&Moore,B.2006,MNRAS,372,1161
Wilman, D. J., Balogh, M. L., Bower, R. G., Mulchaey, J. S.,
Oemler, A., Carlberg, R. G., Morris, S. L., & Whitaker, R. J.
2005,MNRAS,358,71
Zucca,E.,etal.2006,A&A,455,879