Multiwavelength Detection of Cluster AGN

MISSING IMAGE

Material Information

Title:
Multiwavelength Detection of Cluster AGN
Physical Description:
1 online resource (187 p.)
Language:
english
Creator:
Klesman,Alison J
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Astronomy
Committee Chair:
Sarajedini, Vicki L
Committee Members:
Hamann, Fredrick
Gonzalez, Anthony
Guzman, Rafael L
Detweiler, Steven L

Subjects

Subjects / Keywords:
agn -- clusters -- galaxies -- infrared -- multiwavelength -- optical -- xray
Astronomy -- Dissertations, Academic -- UF
Genre:
Astronomy thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract:
We aim to study the effect of environment on the presence and fueling of Active Galactic Nuclei (AGN) in massive galaxy clusters. We first explore the use of different AGN detection techniques with the goal of selecting AGN across a broad range of luminosities, AGN/host galaxy flux ratios, and obscuration levels. We quantify the use of optical variability in comparison with X-ray detection and mid-IR power-law SED fitting to identify AGN in the GOODS-South field. We then analyze a sample of 12 galaxy clusters at redshifts 0.5 < z < 0.9 to identify cluster AGN candidates using optical variability utilizing multi-epoch HST ACS imaging in the I-band, X-ray point source detection in Chandra images, and mid-IR SED fitting for a power law slope through the Spitzer IRAC channels. We identify 178 optical variables, 74 X-ray point sources, and 64 IR power law sources within the 12 clusters, resulting in an average of ~25 AGN per cluster. We find no difference between the fraction of AGN among galaxies in clusters and the percentage of similarly-detected AGN in field galaxy studies (~2.5%). This result provides evidence that galaxies are still able to fuel accretion onto their supermassive black holes, even in dense environments. While IR AGN appear less concentrated in clusters, variables and X-ray detected AGN appear to be more centrally concentrated than normal galaxies, with X-ray sources dominating a significant overdensity in the central 0.5 Mpc of our clusters. The host galaxy colors of AGN in clusters are similar to those observed in field galaxy studies and appear to peak in the "green valley," indicating that the presence of an AGN is connected to the transitioning of the host galaxy from blue, star forming to redder, passively evolving galaxy types. We observe that while normal galaxies reveal a significant change in color as a function of cluster radius, the AGN host colors remain constant with cluster radius. This observation appears to support the theory that processes related to the accreting supermassive black hole have a more significant impact on the star forming properties of the host galaxy than does the intracluster medium and/or the local environmental density.
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility:
by Alison J Klesman.
Thesis:
Thesis (Ph.D.)--University of Florida, 2011.
Local:
Adviser: Sarajedini, Vicki L.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2013-08-31

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2011
System ID:
UFE0043202:00001


This item is only available as the following downloads:


Full Text

PAGE 1

MULTIWAVELENGTHDETECTIONOFCLUSTERAGN By ALISONJ.KLESMAN ADISSERTATIONPRESENTEDTOTHEGRADUATESCHOOL OFTHEUNIVERSITYOFFLORIDAINPARTIALFULFILLMENT OFTHEREQUIREMENTSFORTHEDEGREEOF DOCTOROFPHILOSOPHY UNIVERSITYOFFLORIDA 2011

PAGE 2

c r 2011AlisonJ.Klesman 2

PAGE 3

ThisthesisisdedicatedtoeveryonewhobelievedIcouldact uallydoit,whichIamlucky tosayisaratherlonglist,butmostespeciallymyfamily.Yo ualwayssaidIwouldendup withmoredegreesthanathermometer! DedicatedalsotoSashi,whowaseveraloyal,lovingfriend. 3

PAGE 4

ACKNOWLEDGMENTS Iwouldliketorstlythankmyadvisorandcommitteechair,V ickiSarajedini,for beingsosupportiveanddedicatedthroughoutmygraduateca reer.Iwouldalsoliketo thanktherestofmycommittee:AnthonyGonzalez,FredHaman n,RafaelGuzman,and StevenDetweiler,fortheirconsiderationandguidance.Ad ditionally,Iwanttothankmy UndergraduateandGraduateadvisorsatMIT,JimElliotandR ickBinzel,withoutwhomI neverwouldhavemadeittoUFintherstplace. ThanksalsototheAstronomyDepartmentITstaff,KenSallot andMattGlover, forbeingsoutterlyhelpfulwithmymany(andsometimesbiza rre)requests.Thanksto CatherineCassidy,foralwaysbeingontopofeverythingIne ededtogetdoneforthe schoolandforbeingsoverysupportive-andforgettingmean icebigdesk! IwouldalsoliketothankKerenSharonforprovidingherlist ofsupernovaeinthese clusters,aswellasKim-VyTranandSimonEllisforsokindly providingtheirunpublished catalogsofredshiftsuponrequest.Ialsowanttoacknowled getheworkofElizabeth Barrett,withoutwhosethesisthisprocesswouldhavebeens omuchmoredifcult. SpecialthankstoScottFleming,whoisnotonlyafantastico fcemate,butwithout whomIwouldbelostinaseaofunintelligibleIDLcode-thewo rdGundamisinherefor you.AlsothankstoNathanDeLee,forprovidingalotofCokeZ ero,M&Ms,andsome greattelevision-watchingcompany(nottomentionagoodso undingboardwhenIfelt likecomplainingaboutthisverydocument).Iwouldliketot hankElizabethTaskerforher wonderfuladviceonbothwritingandastronomy,aswellasal loftheexcellentswagshe broughtmebackfromJapan.AndaveryfondthankstoValerieM iklesandTyler(the Mighty)DesjardinsforsomeamazingtimesattheAASandsome ofthebestfriendsI couldeveraskfor.WithouttheirsupportandfriendshipIwo uldbelost. ThanksaswelltoJenWatson,StaciaSwanson,andDianaOwen. Allofyouhave beensuchwonderfulfriendstomeforsuchaverylongtime,an dIappreciatethe supportandgoodtimes,especiallywhenIneededabreakfrom academics.Andnally, 4

PAGE 5

thankyoutomyfantasticgroupoffriendshereinGainesvill e,andtomyRPfriends,all ofwhomcertainlykeptmesane(andupoftenlaterthanIshoul dbe). 5

PAGE 6

TABLEOFCONTENTS page ACKNOWLEDGMENTS .................................. 4 LISTOFTABLES ...................................... 9 LISTOFFIGURES ..................................... 10 ABSTRACT ......................................... 12 CHAPTER 1INTRODUCTION ................................... 14 1.1ActiveGalacticNuclei ............................. 14 1.2MultiwavelengthDetectionofAGN ...................... 16 1.2.1OpticalVariability ............................ 16 1.2.2X-rayEmission ............................. 17 1.2.3InfraredEmission ............................ 17 1.3AGNInGalaxyClusters ............................ 18 1.3.1Background ............................... 18 1.3.2EnvironmentalEffectsonAGN .................... 19 2OPTICALVARIABILITYOFINFRAREDPOWERLAW-SELECTEDGALAX IES ANDX-RAYSOURCESINTHEGOODS-SOUTHFIELD ............ 22 2.1Introduction ................................... 22 2.2AGNCandidates ................................ 23 2.2.1IR-SelectedGalaxies .......................... 23 2.2.2X-ray-SelectedGalaxies ........................ 24 2.2.3ACSGOODSDataandVariability ................... 24 2.3Discussion ................................... 26 2.4Conclusions:VariableAGNintheGOODS-SouthField .......... 31 3DATAANDTECHNIQUES .............................. 43 3.1GalaxyClusterSample ............................ 43 3.1.1CL0152-1357 .............................. 44 3.1.2CLJ1226.9+3332 ............................ 44 3.1.3MS0451.6-0305 ............................. 44 3.1.4MS1054.4-0321 ............................. 45 3.1.5SDSS1004+41 ............................. 46 3.2OpticalImageAnalysisandPhotometry ................... 46 3.2.1ImageCreation:MULTIDRIZZLE ................... 47 3.2.2CatalogCreation ............................ 48 3.2.3PhotometryofGalaxyClustersinMulti-EpochACSImag es .... 48 3.3DeterminationofVariabilityThreshold .................... 49 6

PAGE 7

3.3.1VariabilityDeterminationWith2Epochs ............... 50 3.3.2VariabilityDeterminationWith3Epochs ............... 51 3.3.3Summary:OpticalVariablesinClusterImages ........... 52 3.4X-rayDataReductionandPointSourceDetermination ........... 53 3.4.1DataAcquisitionandReduction .................... 53 3.4.2SourceDetection ............................ 55 3.4.3SourceProperties ........................... 57 3.4.4Summary:X-ray-DetectedSourcesinClusterImages ....... 58 3.5InfraredImageAnalysisandPhotometry ................... 59 3.5.1DataAcquisitionandReduction .................... 59 3.5.2SourceExtractionandPhotometry .................. 61 3.5.3Summary:IRPowerLawSourcesinClusterImages ........ 62 3.6ComparisonofOpticalVariables,X-rayPointSources,a ndMid-IRAGN Candidates ................................... 63 4THEPERCENTAGEOFAGNINGALAXYCLUSTERS ............. 103 4.1ClusterMembership .............................. 103 4.1.1SpectroscopicCatalogs ........................ 103 4.1.2FieldContamination .......................... 104 4.1.3ClusterRadialProlesandMembershipProbability ......... 105 4.1.4ColorSelectionandClusterMembershipProbability ........ 106 4.2PercentageofClusterAGN .......................... 107 4.2.1ComparisonwithFieldGalaxiesinGOODS ............. 108 4.2.2ComparisonwithClusterX-raySurveysforAGN .......... 109 4.2.3Summary:AGNPercentage ...................... 110 4.3ClusterProperties ............................... 111 4.3.1ClusterMorphology ........................... 113 4.3.2Summary:ClusterPropertiesandAGNPercentage ......... 115 5THEDISTRIBUTIONOFAGNINGALAXYCLUSTERS ............. 128 5.1RadialDistributionofAGN ........................... 128 5.1.1IndividualClusters ........................... 128 5.1.2CombinedClusterSample ....................... 130 5.2Discussion ................................... 132 5.3AGNPropertiesandClusterRadialDistance ................ 135 6HOSTGALAXYPROPERTIES ........................... 145 6.1OpticalColors .................................. 145 6.2ClusterMorphologyandHostGalaxyColor ................. 150 6.3RadialDistributionofGalaxyColors ..................... 151 7

PAGE 8

7SUMMARY ...................................... 159 7.1OpticalVariabilityintheGOODS-SouthField ................ 159 7.2AGNIdenticationinGalaxyClusters .................... 160 7.3ClusterMembershipandPercentageofAGNinGalaxyClust ers ..... 162 7.4RadialDistributionofAGN ........................... 163 7.5HostGalaxyOpticalColors .......................... 165 APPENDIX ACLUSTERACSIMAGES .............................. 167 BCLUSTERCHANDRAX-RAYIMAGES ...................... 169 CCLUSTERRADIALPROFILES ........................... 171 REFERENCES ....................................... 173 BIOGRAPHICALSKETCH ................................ 187 8

PAGE 9

LISTOFTABLES Table page 2-1Optical,IR,andX-rayPropertiesofGOODSSources .............. 37 3-1ClusterObservations ................................. 79 3-2GalaxyClusterProperties .............................. 79 3-3ACSObservations .................................. 80 3-4OpticalVariables:2Epochs ............................. 81 3-5OpticalVariables:3Epochs ............................. 86 3-6X-rayObservations .................................. 90 3-7X-rayPointSources ................................. 91 3-8Rest-FrameX-rayFittingParameters ........................ 96 3-9IRPowerLawGalaxies ............................... 97 3-10AGNOverlap ..................................... 101 3-11GalaxiesintheLacyWedge ............................. 102 4-1ClusterSpectroscopicCoverage .......................... 124 4-2AGNRedshifts .................................... 125 4-3PercentageofClusterAGN ............................. 125 4-4PercentageofClusterAGNByType ........................ 126 9

PAGE 10

LISTOFFIGURES Figure page 2-1OpticalVariables:StandardDeviationvs.AverageVMag nitude ......... 33 2-2VariabilitySignicancevs.AverageVMagnitude ................. 34 2-3OpticalVariabilitySignicancevs.Mid-IRSpectralIn dex ............. 35 2-4VariabilitySignicancevs.X-rayBandRatio .................... 36 3-1GalaxyNuclearMagnitudeinEpoch1 ....................... 69 3-2Group2:GalaxyNucleusMagnitudeDifferenceinTwoBins ........... 69 3-3Group2:Determinationof ............................ 70 3-4Group2:OpticalVariabilityDetermination ..................... 70 3-5Group3:OpticalVariabilityDetermination ..................... 71 3-6Group1:OpticalVariabilityDetermination ..................... 71 3-7DistributionofVariabilitySignicance ........................ 72 3-8X-rayFluxvs.ClusterRadius ............................ 73 3-9DistributionofX-rayFlux ............................... 74 3-10DistributionofX-rayHardnessRatio ........................ 74 3-11DistributionofIRMagnitude ............................. 75 3-12ExamplePowerLawFittoaMid-IRSED ...................... 75 3-13X-rayHardnessRatiovs.IRPowerLawSlope .................. 76 3-14OpticaltoX-rayFluxRatio .............................. 77 3-15Mid-IRColorsofGalaxies .............................. 78 4-1RadialDistributionofGalaxiesinCLJ1226 .................... 116 4-2V-IColorDistributionofConrmedClusterGalaxies ............... 117 4-3RadialDistributionofGalaxiesinMACSJ0717 .................. 118 4-4AGN%vs.ClusterRedshift ............................. 119 4-5AGN%vs.ClusterX-rayLuminosity ........................ 120 4-6AGN%vs.ClusterMass .............................. 121 10

PAGE 11

4-7AGN%vs.ClusterVirialRadius .......................... 122 4-8AGN%vs.ClusterVelocityDispersion ....................... 123 5-1CumulativeDistributionofGalaxiesin3Clusters ................. 137 5-2CumulativeDistributionofAllGalaxies ....................... 138 5-3RadialDistributionofAGN .............................. 139 5-4RadialDistributionofGalaxiesinRelaxedandDisturbe dClusters ........ 140 5-5RadialDistributionofGalaxiesinSDSS1004andCL0152 ............ 141 5-6OpticalVariabilitySignicancevs.ClusterRadius ................. 142 5-7X-rayHardnessRatiovs.ClusterRadius ..................... 143 5-8IRPowerLawSlopevs.ClusterRadius ...................... 144 6-1GalaxyColorsvs.Redshift ............................. 154 6-2GalaxyColorsvs.ClusterMembershipProbability ................ 155 6-3GalaxyColorsvs.ClusterMorphology ....................... 156 6-4GalaxyColorsByTypevs.ClusterRadius ..................... 157 6-5GalaxyColorsvs.ClusterRadius .......................... 158 A-1ACSImagesofClusters ............................... 167 B-1ChandraX-rayImagesofClusters ......................... 169 C-1RadialPlots:AllClusters .............................. 171 11

PAGE 12

AbstractofDissertationPresentedtotheGraduateSchool oftheUniversityofFloridainPartialFulllmentofthe RequirementsfortheDegreeofDoctorofPhilosophy MULTIWAVELENGTHDETECTIONOFCLUSTERAGN By AlisonJ.Klesman August2011 Chair:VickiL.SarajediniMajor:Astronomy Weaimtostudytheeffectofenvironmentonthepresenceandf uelingofActive GalacticNuclei(AGN)inmassivegalaxyclusters.Werstex ploretheuseofdifferent AGNdetectiontechniqueswiththegoalofselectingAGNacro ssabroadrangeof luminosities,AGN/hostgalaxyuxratios,andobscuration levels.Wequantifytheuseof opticalvariabilityincomparisonwithX-raydetectionand mid-IRpower-lawSEDttingto identifyAGNintheGOODS-Southeld.Wethenanalyzeasampl eof12galaxyclusters atredshifts0.5 < z < 0.9toidentifyclusterAGNcandidatesusingopticalvariab ility utilizingmulti-epochHSTACSimagingintheI-band,X-rayp ointsourcedetectionin Chandraimages,andmid-IRSEDttingforapowerlawslopeth roughtheSpitzerIRAC channels.Weidentify178opticalvariables,74X-raypoint sources,and64IRpower lawsourceswithinthe12clusters,resultinginanaverageo f 25AGNpercluster. WendnodifferencebetweenthefractionofAGNamonggalaxi esinclustersand thepercentageofsimilarly-detectedAGNineldgalaxystu dies( 2.5%).Thisresult providesevidencethatgalaxiesarestillabletofuelaccre tionontotheirsupermassive blackholes,evenindenseenvironments. WhileIRAGNappearlessconcentratedinclusters,variable sandX-raydetected AGNappeartobemorecentrallyconcentratedthannormalgal axies,withX-raysources dominatingasignicantoverdensityinthecentral0.5Mpco fourclusters.Thehost galaxycolorsofAGNinclustersaresimilartothoseobserve dineldgalaxystudiesand 12

PAGE 13

appeartopeakinthe“greenvalley,”indicatingthatthepre senceofanAGNisconnected tothetransitioningofthehostgalaxyfromblue,starformi ngtoredder,passively evolvinggalaxytypes.Weobservethatwhilenormalgalaxie srevealasignicant changeincolorasafunctionofclusterradius,theAGNhostc olorsremainconstantwith clusterradius.Thisobservationappearstosupportthethe orythatprocessesrelated totheaccretingsupermassiveblackholehaveamoresignic antimpactonthestar formingpropertiesofthehostgalaxythandoestheintraclu stermediumand/orthelocal environmentaldensity. 13

PAGE 14

CHAPTER1 INTRODUCTION 1.1ActiveGalacticNuclei ActiveGalacticNuclei(AGN)areenvisionedtobeaccreting supermassiveblack holesatthecentersofgalaxies,withmassesrangingfrom 10 6 – 10 10 M (e.g., Salpeter 1964 ; Lynden-Bell 1969 ).Supermassiveblackholesarenowbelievedtoresidein thecenterofallmassivegalaxies( Kormendy&Richstone 1995 ),andthereisa well-establishedcorrelationbetweenthemassofthecentr alblackholeandthemass andvelocitydispersionofthehostgalaxysspheroidalcomp onent( Magorrianetal. 1998 ; Ferrarese&Merritt 2000 ; Gebhardtetal. 2000 ).Thisrelationshipindicatesthat thereisafundamentallinkbetweenthegrowthofsupermassi veblackholesandthe formationofthegalaxiesinwhichtheyreside. TheUniedModelofAGN(e.g., Antonucci 1993 ; Urry&Padovani 1995 )describes theseobjectsasacentralsupermassiveblackholesurround edbyanaccretiondisk, fromwhichmostoftheluminosityoftheAGNarises.Thisenti resystemisitselfthought tobesurroundedbyatorusofobscuringgasanddust.Objects thatareviewedat anangleallowingamoredirectlineofsightintothecentral enginewillexhibitbroad emissionlines(BLAGNs),whichemanatefrommaterialmovin gathighspeedscloseto theblackhole.Narrowemissionlineobjects(NLAGNs)arevi ewedfromamoreoblique angleandthusthecentralaccretiondiskandthebroademiss ionlinesareobscured;the narrowlinesarisefromthere-radiationoflightfromthece ntralsourcebymoredistant material,andthusdonotshowmotionatthespeedofmaterial closertotheblackhole. TheUniedModel,however,doesnoteffectivelyaddressthe originorevolution oftheseobjects,andlittleisknownaboutthedutycyclesof AGN.Whilequasar lifetimescanbegenerallyconstrainedto 10 7 – 10 8 years(e.g., Martini 2004 ),AGN mayexperiencemorethanoneluminousphaseduringagalaxy' slifetimeandthereis likelyarangeofAGNlifetimesbasedontheirrangeinlumino sity( Hopkins&Hernquist 14

PAGE 15

2009 ).Inaddition,theexactmechanismthattriggersAGNfuelin gisstillunknown, thoughcurrentleadingtheoriessuggestthatmergersandot hergalaxyinteractionsmay playalargeroleinthegrowthofAGN,funnelinggasanddusti ntothenucleustofeed theblackhole( Hopkinsetal. 2005 ).Suchtheoriesplaceconstraintsontheobserved propertiesofAGNdependingonfuelavailabilityandhaveim plicationsfortheobserved numberandtypeofAGNbasedontheirenvironmentsandgalact ichistory. AGNcanbeclassiedbyluminosity,inwhichSeyfertgalaxie sarethedimmestand quasars(QSOs)arethebrightestoftheseobjects.Therefor e,QSOsmainlyrepresent thebrightendoftheAGNluminosityfunction,whichdescrib estheseobjects'behavior asafunctionoftheirdensityandluminosity.TheAGNlumino sityfunctionhasbeen well-characterizedforQSOsathighredshifts(e.g., Hartwick&Schade 1990 ).Not aswell-studied,butofequalimportance,isthefaintendof thisluminosityfunction. Whiletherehavebeenseveralstudiesfocusingonthelocall uminosityfunctionsof Seyfertgalaxies( Chengetal. 1985 ; Huchra&Burg 1992 ; Koehleretal. 1997 ; Londishetal. 2000 ; Maiaetal. 2003 ; Haoetal. 2005 ),theseobjectshavenot beenstudiedasextensively,inlargepartbecausetheirlow erluminositiesmakethem difculttodetectandidentifyatlargedistances,resulti nginapoorly-denedluminosity functionatthesemagnitudes.Theselower-luminosityAGNa relikelytorepresenta signicantcontributiontotheX-ray,IR,andUVbackground s,asAGNwhichareheavily obscuredbydustmayhavetheirX-raysblockedandre-emitte dbythedustatlonger wavelengths.Thus,knowledgeoftheirspaceandluminosity distributionatarangeof redshiftsisimportant.Theymayalsorepresentanimportan tphaseofgalaxyevolution, whichconnectsnormalandactivegalaxies( Crotonetal. 2005 ).Understandinghow lower-luminosityAGNaredistributedinspaceandluminosi ty,aswellashowthat relationshiphasevolvedovertime,willprovideimportant cluesaboutthetotalAGN densityatearlierepochs. 15

PAGE 16

1.2MultiwavelengthDetectionofAGN ThereareanumberofdetectionmethodsbywhichAGNhavebeen identiedin galaxysurveys.Inadditiontotheidenticationofbroadan dnarrowopticalemission lineswithdistinctuxratios( Baldwinetal. 1981 ; Veilleux&Osterbrock 1987 ),AGN canbeidentiedviaUVandopticalcolorselection( Markarian 1967 ; Schmidt&Green 1983 ; Richardsetal. 2002 ),X-rayemission( Fabbianoetal. 1992 ; Alexanderetal. 2003 ),infraredcolorselection( Lacyetal. 2004 ; Sternetal. 2005 ),radioemission ( Smith&Wright 1980 ),andvariability( Hooketal. 1994 ; MacLeodetal. 2010 ). However,manyofthesetechniquesproduceanincompletepic tureofAGN,astheycan bebiasedagainstgalaxiesinwhichtheAGNlightisnotasign icantpercentageofthe totalgalaxylight.OpticalandUVsurveyscanalsomissmore heavilyobscuredAGN andhostgalaxies.Furthermore,ithasbeensuggested(e.g. Hickoxetal. 2009 )that AGNselectedviadifferenttechniquescouldrepresentAGNa ndgalaxiesindifferent stagesofevolution.Therefore,inordertopresentamoreco mpletepictureoftheentire classofobjectswhichcompriseAGN,itisnecessarytoutili zemorethanonetechnique toselectamorecompletesampleofAGNtostudythelinksbetw eentheseobjectsand thegalaxiesinwhichtheyreside1.2.1OpticalVariability AGNareknowntoshowvariabilityinmultiplewavelengthreg imesonavariety oftimescales:theopticallineemissionandcontinuumuxf romquasarshasbeen observedtovaryontimescalesofmonthstoyears,whileX-ra yuxfromthesesources variesonshortertimescalesofhourstodays( Peterson2001 &referencestherein). Monitoringvariabilityinthesesourcesisanimportantway toextractinformationabout thecompactobjectpoweringtheAGN.Thetimescalesoverwhi chvariabilityisseen canprovideinformationaboutthesizeandstructureofther egionfromwhichthe emissionoriginatesinordertoplacephysicalconstraints onthecentralenginewithin. ThemechanismbehindtheopticalvariabilityinAGNisstill uncertain,thoughmany 16

PAGE 17

theorieshavebeenformulatedtoexplainthisphenomenon.T heleadingexplanation fortheobservedvariabilityisthatchangesinmagnitudear erelatedtoaccretionevents, suchaschangesinaccretionrateordiskinstabilities( Pereyraetal. 2006 ).Other explanationsforvariabilityincludechangesinmagnitude duetosupernovae,starbursts, ormicrolensingeventsinthenucleusofthegalaxy(e.g., Aretxaga&Terlevich 1994 ). Ithasbeenshownthat80–100%ofAGNcandidatesshowrandomv ariationsupto severalpercentintheopticaloverthecourseofseveralyea rs(e.g., Kooetal. 1986 ). Webb&Malkan ( 2000 )foundthat60%ofLow-LuminosityAGNs(LLAGNs;thosewith luminosities M B > -23)variedwhenstudiedoveraperiodofmonths.Thus,optic al variabilityisanimportantandsuccessfulmethodforident ifyingAGNinimagingsurveys. Indeed,thereissomeevidencethatLLAGNsexhibitgreatera mplitudeuxchangesthan dotheirbrightercounterparts( Bershadyetal. 1998 ),makingvariabilityaparticularly effectivemeansfortheiridentication.1.2.2X-rayEmission AGNareknowntobeluminousX-raysourcesandcanberoutinel yselectedfrom deepX-ray( < 10keV)surveys.TheX-rayemissionisbelievedtooriginate verycloseto thecentralblackholeandisproducedbyComptonupscatteri ngofsofterphotonsbya hot“corona”aroundtheaccretiondisk(e.g., Sunyaev&Titarchuk 1980 ).AGNarethe dominantcontributorstotheluminoushard(2–10keV)X-ray pointsourcepopulation ( Martinietal. 2002 ).Inparticular,hardX-rayemissionislessaffectedbyobs curing materialsuchasdustthanUVoropticallight,allowingAGNt obeidentiedwitharange ofobscurationlevels.InadditiontotheX-raypointsource population,obscuredAGNare thoughttocontributesignicantlytotheX-raybackground atlowerenergies,i.e.,below 10keV(e.g., Worsleyetal. 2005 ; Szokolyetal. 2004 ; Barger 2003 ). 1.2.3InfraredEmission AGNalsocontributetotheIRandsubmillimeterbackgrounds ( Fabian&Iwasawa 1999 ; Almainietal. 1999 ).X-ray,UV,andopticalemissioncanbeabsorbedbydust 17

PAGE 18

surroundingthenucleus,andthenreprocessedandre-radia tedasinfraredlight. SelectingAGNviatheirIRSEDsrevealsacatalogofobjectst hatmaynotbereadily identiedviaopticalorX-raymeans,astheymaybesubjectt olargeamountsof obscuration. AGNappeartoexhibitpowerlaw-likeemissionintheirconti nuaacrossabroad rangeofwavelengths,rangingfromtheUVandopticaltothen ear-,mid-andfar-infrared ( Neugebaueretal. 1979 ; Elvisetal. 1994 ).Thispowerlawarisesfromthecombination ofanumberofcomponentsthatbroadenthesourcesSED,rathe rthanfromanyactual powerlawemission,andcanbedescribedbytheexpression f / ,where isthe spectralindex(e.g., Rieke&Lebofsky 1981 ).X-ray-andoptically-selectedSeyfert1 galaxiesexhibitarelativelyatpowerlawSEDin f ,whileoptically-selectedSeyfert 2galaxiestendtohavesteeperslopesintheIR( Carletonetal. 1987 ; Edelsonetal. 1987 ).ThispowerlawemissionisalsopresentintheopticalandI RcontinuaofX-rayandnear-IR-selectedquasars( Kuraszkiewiczetal. 2003 ; Wilkesetal. 2005 ). 1.3AGNInGalaxyClusters 1.3.1Background Galaxyclusterspresentacompletelydifferentenvironmen tthantheeld.The denseclusterenvironmentisknowntoaffecttheevolutiono fclustergalaxies,producing alargerfractionofearly-typegalaxiesintheseenvironme nts,incontrasttothe lower-densityeld,whichismainlycomprisedoflate-type galaxies(e.g., Hubble& Humason 1931 ; Morgan 1961 ; Abell 1965 ; Oemler 1974 ).Furthermore,galaxiesin clustersencountereachotherathigherrelativespeeds,re sultinginfewermergersin galaxyclustersthaninmorelooselyboundgroups(e.g., Makino&Hut 1997 ). Earlyspectroscopicstudies( Gisler 1978 ; Dressleretal. 1985 )presented evidencethatAGNmaybelesscommoninrichclustersthanint heeld.Thesestudies concludedthatthefractionofAGNinclustersisonlyabout1 %,comparedwith5% observedintheeld( Dressleretal. 1985 ).Morerecently,theaccuracyandresolution 18

PAGE 19

oftheChandraX-RayObservatoryhasallowedfordetectiono fX-raypointsources withingalaxyclustersandrecentstudieshavedetectedlar genumbersofoptically normalclustergalaxieswhoseluminousX-rayemissionwoul dindicateAGNactivity (e.g., Martinietal. 2002 2006 ).UsingX-rayobservations, Martinietal. ( 2002 )found aclusterAGNfractionapproximately5timeshigherthantha tfoundby Dressleretal. ( 1985 ).X-RayimaginghasalsorevealedalargenumberofAGNsinth eoutskirtsof clustersrelativetotheeld(e.g., Henry&Briel 1991 ; Cappietal. 2001 ; Martinietal. 2006 ). Kauffmannetal. ( 2004 )foundthatwhilehigh-luminosityAGNstendtoavoid highdensityregionssuchasclusters,low-luminosityAGNs donot.However,only high-luminosityAGNsareeasilydetectableviaopticalemi ssionlinesurveysandthe numberoflow-luminosityAGNsinclustersmaycurrentlybea nunderestimate. ThisrecentworkindicatesthatthefractionofclusterAGNm aybehigherthan previouslybelieved,andmaybeashighorhigherthanthatin theeld.Theseestimates, however,relyonX-raysourcedetectionswhichoftencontai nedfewcountsand havebeenhardtoclassifyasAGNversusX-raysourcessuchas X-raybinaries, starbursts,ordiffusegas( Martinietal. 2006 ).Additionally,itisclearthatoptical,largely spectroscopicsearchesmaybemissingmanyAGN,asemission fromanAGNmay bepartiallyhiddenduetoweaklinestrengthsordilutionof thesourcebydustorhost galaxylight.AtrueunderstandingofthefractionofAGNing alaxyclustersthusrelieson thesuccessfulidenticationofAGNoverarangeofluminosi tiesandobscurationlevels. 1.3.2EnvironmentalEffectsonAGN ThefractionofclusterAGNhasasignicantimpactonourund erstandingofthe clusterenvironmentanditseffectonmembergalaxies.Thet womostimportantfuel sourcesforAGNarebelievedtobecoldgasreservoirsnearth ecentralblackholeand galaxymergers,whichcauseinowsofgasintothenucleus(e .g., Barnes&Hernquist 1992 ; Springeletal. 2005 ).ThefractionofAGNsinrichclustersversustheeldwas assumedtobelowermainlyduetothelackofavailablefuelin gepisodesinclusters– 19

PAGE 20

fewclustergalaxieshavesignicantreservoirsofcoldgas availabletofueltheblack hole(e.g., Giovanelli&Haynes 1985 ),andmajormergersarerareinclustersdueto highrelativespeedsbetweenclustergalaxies.Thus,fueli ngargumentssuggestthat smallreservoirsofcoldgasandinfrequentmergersbetween clustergalaxiesshould resultinfewerclusterAGNsthanobservedintheeld.Theex istenceofanAGNisan indicatorthatsuchareservoirofavailablegasexistsinth ecentralregionofthehost galaxy.AGNcanthusprovideinformationabouttheefcienc yofcoldgasstrippingin galaxyclusters,aswellastheextenttowhichthecentralsu permassiveblackholemay growinsuchanenvironment.ThenumberofAGNsfoundingalax yclustersislinked totheevolutionoftheclustergalaxies,andmaybeabletopr ovideconstraintsonthe fuelingprocessesofnuclearactivityindenseenvironment s. Additionally,clusterAGNscanbeusedtoprobebothAGNandg alaxyevolution indenseenvironments.VariationsinthenumberandtypeofA GNsdetectedin clusterenvironmentsversustheeldcanleadtofurtherund erstandingoftheeffects ofenvironmentontheevolutionofgalaxies.Understanding theimportantfactorsin transformingstar-forminggalaxiesintopassivegalaxies andtheeffectsofenvironment onthisprocesswillhelppresentaclearerpictureofgalaxy evolution.Alargernumber ofAGNsinclustergalaxieswouldindicatethatmorecluster galaxiesthanpreviously thoughtareabletosupportnuclearactivity–i.e.,theyare abletoretainorreplenish areservoirofcoldgasneartheircentralblackhole.Furthe rmore,feedbackfromAGN activitymayalsosignicantlycontributetotheheatingof theintraclustermedium(ICM) andbeanimportantfactoringalaxyevolutionwithintheclu ster.Understandingthelink betweenactivegalaxiesandtheirenvironmentisthusanimp ortantstepintheprocess ofunderstandinggalaxyclustersandtheirevolutionaswel l. Inthisthesis,wewillexploretheissueofAGNidenticatio ningalaxiesacrossa rangeofluminositiesandobscurationlevelsandapplythes edetectiontechniquesto investigateAGNinthedenseclusterenvironment.InChapte r 2 ,wepresentastudy 20

PAGE 21

tocompare3AGNidenticationtechniquesinthewell-studi edGOODS-Southeld: opticalvariability,X-rayemission,andIRpowerlawSEDt ting.Wethenapplythese techniquestoidentifyAGNinasampleof12massivegalaxycl ustersatz=0.5–0.9in Chapter 3 .Wedetermineclustermembershipprobabilitiesforthegal axiesinourcluster imagesandcalculatethepercentageofAGNinclustersinCha pter 4 ,andinChapter 5 weexaminetheradialdistributionofAGNandgalaxiestoinv estigatetheimpactoflocal environmentonthepresenceofAGN.InChapter 6 ,weexploretheopticalpropertiesof theAGNhostgalaxiesandsummarizeourresultsinChapter 7 21

PAGE 22

CHAPTER2 OPTICALVARIABILITYOFINFRAREDPOWERLAW-SELECTEDGALAXI ESAND X-RAYSOURCESINTHEGOODS-SOUTHFIELD 2.1Introduction InordertoidentifyAGNamonggalaxiesintheeldorinclust ers,itisnecessary tousearangeofwavelengthsandtechniquestominimizesurv eybiasesagainstfaint, obscuredAGNorthosethatdonotdominatethelightofthehos tgalaxy.X-rayand mid-IRsurveysforAGNareimportanttoolstominimizethelo ssofobscuredAGN. Additionally,measuringopticalnuclearvariabilitycani dentifyvaryingAGNwithin brighterhostgalaxies.Theuseofsmallaperturephotometr ywithhigh-resolutionHST imagesallowsforthedetectionofAGNcomprisingaslittlea s 5%ofthetotaloptical galaxylight( Sarajedinietal. 2003 ). Inthischapter,weinvestigatetheuseofthesemultiwavele ngthtechniquesby performinganuclearopticalvariabilitystudyonapre-sel ectedsampleofX-rayand mid-IRAGNintheGreatObservatoriesOriginsDeepSurvey(G OODS)Southeld ( Klesman&Sarajedini 2007 ).Theaimsofthisstudyare1)toconrmtheAGNnature ofcandidatesidentiedinmultiwavelengthsurveysviaopt icalvariability,and2)to quantifytheuseofopticalvariabilityinidentifyingvari oustypesofAGNforusein interpretingtheresultsofalargersurveyofallopticalso urcesintheGOODSSouth andNorthelds( Sarajedinietal. 2011 ).Wechoseasampleof112pre-selectedAGN candidatesfromtwoseparatecatalogs.Therstisasampleo finfrared-selected AGNcandidatesfrom Alonso-Herreroetal. ( 2006 )withopticalcounterpartsin theGOODS-Southeld.ThesegalaxieswereselectedusingSp itzer/MIPS24 m observationsandarewell-twithapowerlawspectralenerg ydistribution(SED)through theSpitzerIRACbandsat3.6,4.5,5.8,and8 m.ThesecondsampleconsistsofX-ray sourcescompiledby Alexanderetal. ( 2003 )usingChandraobservations.TheChandra DeepFieldSouth(CDFS),whichoverlapstheGOODS-Southel d,isacombinationof 22

PAGE 23

11individualpointingstakenwiththeAdvancedCameraImag ingSpectrometer(ACIS-I) coveringarangeof0.2–10keVwithatotalexposuretimeof94 2ks. Candidateswerecross-matchedwithopticalsourcesintheA CSimagesofthe GOODS-Southeldandaperturephotometrywasperformedona llsourcesineach epochtolookforopticalvariabilityusinganapertureradi usof2.5pixels(0 00 .075).This radiusisapproximately2xtheFWHMofanunresolvedsourcei ntheGOODSimages, andisappropriateforidentifyingvaryingAGNgiventhatwe expecttheemissionto comefromanunresolvednuclearregion.Variabilitysigni cancewasthencompared withoptical,IR,andX-raypropertiestosearchforcorrela tionsbetweenopticalvariability andpropertiesatotherwavelengthsinordertoinvestigate theopticalvariabilityofAGN candidateshavingarangeofobscuringproperties,fromsof tX-raysourcestohighly obscuredmid-IRsourceslackingX-rayemission. 2.2AGNCandidates 2.2.1IR-SelectedGalaxies Alonso-Herreroetal. ( 2006 )identied92galaxieswithintheCDFSwhichwere detectedat24 manddisplaypowerlaw-likeemissionfrom3.6to8 m.Thesegalaxies wereselectedat24 mbecauseatz > 1alargefractionoftheseobjectsshouldbe Ultra-LuminousInfraRedGalaxies(ULIRGs,withL IR > 10 12 L ).ULIRGsareknownto haveasteeppowerlaw-likeSEDintheinfrared( Sandersetal. 1988 ; Klaasetal. 2001 ) andmanyareclassiedasQSOs. Alexanderetal. ( 2005 )showedthatatleast40%of IR-luminoushigh-redshiftgalaxiescontainAGNandsimila rresultsatz=1–2havebeen foundwithrecentSpitzerobservations( Yanetal. 2005 ).Allofthesamplegalaxiesare luminousintheIR:30%arehyperluminous(L IR > 10 13 L ),41%areULIRGs(L IR 10 12 –10 13 L ),andallbutoneareLuminousIRGalaxies(LIRGs,L IR 10 11 –10 12 L ). Alonso-Herreroetal. ( 2006 )havefurtherquantiedtheIRSEDstoclassifytheirsource s aseitherBroadLineAGNs(BLAGNs),NarrowLineAGNs(NLAGNs ),orULIRGs, groupinggalaxieswithIRSEDsresemblingULIRGswiththeNL AGN. 23

PAGE 24

WematchedsourcesidentiedintheACSGOODSeldfromtheve rsion1.1 sourcecatalogderivedfromthestackedV-bandimagewithth eIRcatalogandfound37 matcheswithina3 00 .5radius,30ofwhichfellwithin1 00 .Ofthesematches,11weretoo faintforopticalphotometry(fainterthan27thmagnitudei nasingleV-bandepoch)and5 werenotvisibleinallveGOODSepochs.Thisleft22objects ,2ofwhichhadpossible multiplenuclei,yieldingatotalof24targetsforphotomet ry.Allnalopticalsources matchedtheIRcoordinateswithin1 00 .1. 2.2.2X-ray-SelectedGalaxies AGNasaclassareknowntobeX-raysourcesandarecommonlyse lectedvia X-rayimaging( Brandt&Hasinger 2005 ).Wematchedthecoordinatesofobjectsin the Alexanderetal. ( 2003 )ChandraX-raycatalogwiththeGOODSversion1.1source catalogandfound200matcheswithina1 00 .9radius.Oftheseopticalcounterparts, 80weretoofaintforopticalphotometry,16werenotvisible inallveepochs,1was contaminatedbyacosmicray,and1sourcewasaddedasadoubl enucleus.Finally, 1additionalsourcewasfoundbymatchingtheGOODScatalogc oordinateswiththe Giacconietal. ( 2002 )X-raycatalog,whichalsocontainsthe102X-raysourcesfr om Alexanderetal. ( 2003 ).Thus,intotal104X-ray-selectedobjectswerephotomete red tolookforopticalvariability.Fourteenoftheseobjectso verlaptheIR-selectedsources discussedinSection 2.2.1 2.2.3ACSGOODSDataandVariability WeobtainedarchivalimagesfromtheGOODS-Southsurveytak enwiththeHubble SpaceTelescopeAdvancedCameraforSurveys(ACS)overvee pochsseparatedby 45-dayintervals.Thisallowedforopticalvariabilitytob emonitoredoverasix-month period.Itwasrstnecessarytoestimatetheopticalmagnit udetowhichphotometry couldbeaccuratelyandrobustlyobtained.Todothis,weobt ainedphotometryofoptical sourcesinfourepoch1regionswhichcoveralargespatialra ngeintheGOODS-Seld. WeusedtheIRAFtoolDAOFINDtolocateallsourcesineachepo chtoconstructa 24

PAGE 25

histogramofnumbervs.nuclear(r=0 00 .075or2.5pixels)Vmagnitude.Wemeasured themagnitudeofthegalaxynucleiusingthesameprocedurea swouldbeusedto identifynuclearvariabilityintheselectedsamples.This histogramrevealedadrop innumbercountsbeginningatanuclearVmagnitudeof27.Wet hereforechosea magnitudelimitof27andanyobjectswithaveragemagnitude sfainterthanV nuc = 27werenotconsideredinouranalysis.Thislimitisalsocon sistentwiththelimiting magnitudetowhichnuclearvariabilitymayexpecttobedete ctedusingxedaperture photometryforgalacticnucleiextendingtoz 1.Inthecurrentdataset,weexpectto becomelesssensitivetoavaryingnucleuswithinahostgala xyatnuclearVmagnitudes of 27.5. 1 Aperturephotometrywasperformedonallsourcesineachepo choftheversion 1.0GOODS-SouthV-bandimagesusinganapertureradiusof2. 5pixels(0 00 .075).All imageswerevisuallyinspectedtoensureconsistentcenter ingacrossallveepochs. Thephotometryyieldedalightcurveforeachsource;thefor malerrorbarsoneach magnitudeinthelightcurvegetlargerforfainterobjectsd uetomorerandomerror associatedwiththemeasurement.Inordertoquantifyvaria tionandpickoutgalaxies varyingsignicantlyabovethephotometricnoise,wecalcu latedthemeanmagnitude, thestandarddeviationofthemean,andanerroronthestanda rddeviationusingthe followingformula: error = r ( error mag ) 2 N (2–1) Here,error mag istheformalmagnitudeerrorforagalaxynucleusineachepo chandN isthetotalnumberofmeasurements(inthiscaseN=5,for5ep ochs).Error isthusthe RMSofthemagnitudeerrors(error mag ),essentiallygivinganerrorbaronthestandard 1 Seediscussionin Sarajedinietal. ( 2003 ),Section3,forfurtherdiscussionofthis effect 25

PAGE 26

deviation( ).Wethendenedthe“signicanceparameter”ofeachobject svariabilityin thefollowingway: Signicance = error (2–2) BydividingthestandarddeviationbytheRMSofitserrors,w ecomparetheamountof anobject'svariationarounditsaveragemagnitudebyitsty picalphotometricerror.Thus, thesignicanceparameterofeachobject'svariabilityiss implyameasureofitsstandard deviationnormalizedbytheRMSofthephotometricerror.Th issignicanceparameter shouldnotbeinterpretedintermsofstatisticalprobabili ties,sincethenumberofpoints (5epochsofdata)isnotenoughtodetermineaGaussiandistr ibution.Nonetheless, thisquantitydoesprovideavalidmeasurementofthelevelo fvariabilityfoundforeach galaxyinoursample.Wecomparethesignicancevaluesagai nstoptical,IR,andX-ray propertiestosearchforcorrelationsbetweenopticalvari abilityandpropertiesatother wavelengthsinthenextsection.Table 2-1 listsallofourobjectsandtheiroptical,IR, andX-rayproperties. 2.3Discussion Figure 2-1 showsthestandarddeviationforeachgalaxynucleus( )vs.average nuclearVmagnitude.Theerrorbarsareerror foreachsource,whichistheRMSofthe photometricerrorsineachepoch,andthesolidlineis3 error .Objectswhichhave largestandarddeviationsandrelativelysmallvaluesofer ror clearlystandoutabove thesolidline.Wehavechosenasignicancevalueof3asourv ariabilitythreshold, althoughthroughoutthepaperandgureswecarryalongthes ignicancevalueforeach source.Asignicanceof3orgreatermeansthatthestandard deviationisatleastthree timestheamountofchangeinthephotometricerrorsoverthe veepochs.Figure 2-2 showsvariabilitysignicancevs.averagenuclearVmagnit udeforallobjects.Thesolid lineatasignicancevalueof3isthesameasthe3 error lineinFigure 2-1 26

PAGE 27

Wendatotalof29variablesoutofthe112galaxiesselected viaX-rayand mid-IRemission,resultingina26 % variabilityrateforourcandidates.Eightofthe14 objects(57 % )whicharebothmid-IRpowerlaw-selectedAGNcandidatesan dX-ray sourcesshowsignicantopticalvariability.Manyofthese arealsoamongthebrightest opticalsources.DetectioninbothX-raysandthemid-IRind icatesthepresenceof anAGNwithsomedustpresenttoreprocessaportionofthelig htandre-emititin theinfrared.Amongtheothervariableobjects,thosedetec tedonlyinX-raysare likelyAGNenshroudedbylittledust,resultinginnosigni cantreprocessedlightand werethusnotidentiedinthemid-IR-selectedsample.Twoo ftheobjectsdetected asopticalvariableswereidentiedviamid-IRpowerlawSED sonlyanddonotshow X-rayemission(squares).Thesesourcesarelikelyobscure dAGNsthatdonotshow upinX-raysbecausemuchoftheX-ray,UV,andopticallighti sreprocessedintothe mid-IR.Thus,whiletheseobjectswouldnotbeselectedasAG NviaX-rayemission, theydoshowopticalandIRevidenceoftheirAGNnature.Into tal,wendthat10out of22mid-IR-selectedAGNcandidates(45 % )and27outof104X-ray-selectedAGN candidates(26 % )showopticalvariabilityoverthecourseof6months. Wefurtherexplorethepropertiesofthemid-IR-selectedAG NinFigure 2-3 Alonso-Herreroetal. ( 2006 )separateBLAGN-likeSEDsfromNLAGN-likeSEDs usingthesteepnessofthepowerlawttotheirinfraredemis sion f / ,where is thespectralindex.Aseparationaround -0.9roughlydividesNLAGN-likeSEDS withsteeper(i.e.,morenegative)slopevaluesfromBLAGNlikeSEDSintheirstudy. Figure 2-3 revealsthat7outof11galaxiesclassiedasBLAGNareoptic alvariables (64 % ),2outof4borderlinesourcesarevariable(50 % ),andonlyoneof7NLAGNs showsopticalvariability(14 % ).Traditionallyweexpectthatopticalvariabilityshould bemostobservableamongType1AGN,sincetheleadingtheory forthecauseof variabilityisbasedoninstabilitieswithintheaccretion disk( Pereyraetal. 2006 ).These instabilitieswouldbemostapparentinType1AGN,whichgiv eusaclearviewofthis 27

PAGE 28

partofthesystemintheuniedmodel.Therefore,ournding sareconsistentwiththis expectation,butalsoshowthatmanyborderlineandasmallf ractionofNLAGNscan alsobeidentiedasopticalvariables. OftheX-raysourcesinourstudy,only26 % arefoundtobeopticalvariables.To interpretthesestatistics,werstcalculatedtheX-ray-t o-opticaluxratiosfortheX-ray sourcestodeterminethattheyareindeedAGN.WeusetheR-ba ndmagnitudesfor theX-raysourcesfrom Giacconietal. ( 2002 ),takenwiththeFORS1cameraatthe VLTandtheWideFieldImager(WFI)ontheESO-MPG2.2meterte lescopeatLa Silla.Thereare93objectswithpublishedRmagnitudesinou rsampleof104X-ray objects(89 % ).WendthattheX-raysourcesinoursurveyclusteraroundl og(F x /F opt ) =0andcovertherangebetweenlog(F x /F opt )=1to-2,whichisthegeneralrangeof valuesobservedforAGN( Comastrietal. 2002 ; Szokolyetal. 2004 ).Wealsond foursourceswithlog(F x /F opt ) < -2(AID182,189,192,and207).Thesesourceshave opticalvariabilitysignicancevaluesof0.83,4.88,2.09 ,and4.64,respectively.Thetwo withhighsignicancevalues( > 3)maybeAGNwhichsimplyhavelowF x /F opt values, consistentwithsomeopticalvariablesidentiedintheHDF -N( Sarajedinietal. 2003 ). Theothertwo,havingbothlowvariabilitysignicanceandl owF x /F opt ,mayinfactnotbe AGN.Thus,98 % oftheX-raysourceshaveF x /F opt valuesconsistentwithAGN. ToinvestigatetrendsamongvariabilityandX-raybandrati o,weusedthepublished valuesof Alexanderetal. ( 2003 ),wherethebandratio(BR)istheratioofthehardX-ray band(2–8keV;HB)sourcecountstothesoftX-rayband(0.5–2 keV;SB)counts.We arbitrarilyseparate“hard”sourcesfrom“soft”sourcesat abandratioofBR=0.5,with softsourceshavingBR < 0.5.Ofthe100X-ray-selectedAGNcandidatesforwhicha bandratioismeasured,37 % aresoftsourcesand63 % arehardsources.Figure 2-4 showsvariabilitysignicancevs.X-raybandratio.Wendt hat19outofthe37X-ray sourceswithsoftbandratiosareopticallyvariable(51 % ).Thereareveryfewvariable objectswithbandratiosgreaterthan0.5andtherearenovar iablesamongthose 28

PAGE 29

sourceswithbandratiosharderthan2.Wendonly8variable soutof62sourceswith abandratiogreaterthan0.5.Thesendingsareconsistentw iththeexpectationthat harderbandratiosindicateamoreobscuredsource,whichwi lllikelyshowlessoptical variabilityasdustmaybeobscuringandreprocessingtheop ticalAGNlight.Objects whichexhibitsoftbandratiosarelessobscuredandtherefo remucheasiertodetectas variables.Insummary,51 % ofsoftX-raysources(BR < 0.5)areopticalvariables,16 % ofsourceshavingbandratiosintherange0.5 < BR < 2arevariables(8of48),andno objectswithBR > 2aredetectedasopticalvariables.Weclearlyobserveadec reasing levelofopticalvariabilitywithincreasingX-rayhardnes s. Webb&Malkan ( 2000 )ndthattheamplitudeofopticalvariabilityinAGNincrea ses ontimescalesof1–100days,with60 % ofAGNsvaryingonmonth-to-monthtimescales. TheresultsfromFigure 2-4 areconsistentwiththisexpectation,aswesee51 % of objectswithsoftbandratiosshowingvariability.The49 % ofsoftsourcesthatdonot showvariabilityarethuslikelytobeAGNthatdonothavevar iabilitytimescalestowhich oursurveyissensitive–ourbaselineofmonthsisnotsufci enttopickupvariabilityin thesesources. Webb&Malkan ( 2000 )alsofoundthattheobservedvariabilitydoesnot appeartodependonotherAGNproperties;asimilarstudyove rthesametimescale ofthesamegalaxieswouldrevealthesamepercentageofvari ability,thoughthe specicgalaxiesfoundtoshowthegreatestvariabilitywou ldchange.Thus,among thesoftX-raysources,itisplausiblethatallwouldshowop ticalvariabilityifobserved overalongertimebaseline(i.e.,severalyears).Thisinte rpretationisalsovalidforthe mid-IR-selectedsources,amongwhichwefoundthat50–63 % oftheBLAGN/borderline SEDsareopticalvariables.Again,allmaybevariablewheno bservedoverlongertime baselines.Thedropinthefractionofopticalvariablesamo ngharderX-raysourcesand theNLAGNSEDsamongIR-selectedsources,however,islikel yduetoincreasedlevels ofobscuringmaterialinthesetypesofAGN. 29

PAGE 30

Forthemanynon-variableX-raysources,acoexistingpossi bilityisthatsome fallintotheclassofX-rayBrightOpticallyNormalGalaxie s(XBONGs),objectswhich appearasAGNsviaX-rayuxbutlackopticalevidenceforacc retion( Comastrietal. 2002 ).SuchobjectstendtohavehardX-rayuxratiosindicative ofobscuredaccretion. SomeoftheseobjectscanbeexplainedasAGNwhicharedomina tedbytheirhost galaxylight,whichwoulddilutetheAGNlightandmaskoptic alevidenceforaccretion. Additionally, Rigbyetal. ( 2006 )studiedhardX-ray-selectedAGNsinhostgalaxies havingawiderangeofinclinationangles.Theyconcludedth attheopticaldullnessin someAGNwithinhostgalaxiesthatarenotface-onorspheroi dalmaybearesultof obscuringmaterialalignedwiththehostgalaxyandfarfrom theionizingnuclearactivity. Thisisconsistentwithourndingoffeweropticalvariable samongharderX-raysources. Finally,wehaveinvestigatedwhetherthereisanyrelation shipamongthepublished X-raybandratiosandthemultiwavelengthSEDclassicatio nsof Alonso-Herreroetal. ( 2006 ),andhowthatmightrelatetothedetectionofopticalvaria bility.Interestingly,of thosesourcesselectedviabothX-rayandmid-IRcriteria,m ost(10outof14)havesoft bandratios.Somewhatsurprisingly,thehardestsourcefro mthissubsetofgalaxies(AID 179)isclassiedwithaBLAGNSED.Alloftheopticalvariabl esdetectedinbothX-rays andviamid-IRselectionaresoftX-raysourceswithBR < 0.5.Thus,wedonotobserve aclearcorrelationbetweentheX-rayandmultiwavelengthS EDAGNclassications. Similarly, Barmbyetal. ( 2006 )ndonlymarginalagreementbetweenAGNclassication determinedseparatelybasedonmid-IRSEDsandX-raybandra tios,althoughthe amountofobscurationpresentshouldinprincipleaffectbo thwavelengthregimes.They listvariationsinthegas-to-dustratioorarangeofintrin sicAGNpropertiesaspossible reasonsthattheIRandX-rayclassicationsdonottendtoag ree.Suchreasonsmay alsoexplainthemanynon-X-ray-detectedgalaxieswithIRp ropertiessimilartothose thataredetectedinX-raysurveys. 30

PAGE 31

2.4Conclusions:VariableAGNintheGOODS-SouthField Thisstudyfoundthat26%ofallX-ray-orIR-selectedAGNcan didatesshow opticaluxvariationsaroundtheirmeanmagnitudethatare atleast3timesthetypical photometricerror.Themostsignicantvariablesarethose thatappearedinboth theX-rayandIRcatalogsandwerebright(M v > 21)opticalsources.Mostofthe variablesaresoftX-raysourceswitharatioofhardX-rayba nd(2–8keV)tosoftX-ray band(0.5–2keV)countsoflessthan0.5,indicatingarelati velyunobscuredAGN withnegligibleamountsofdustneartheionizingsource.Th isisconsistentwiththe expectationthatopticalvariabilityprimarilyselectsTy peI,relativelyunobscuredAGN ( Sarajedinietal. 2006 ). Wendseveralindications,however,thatopticalvariabil ityisalsoobservable amongmoreobscuredAGN,albeitwithamuchlowerdetectionr ate.Eightofthe opticallyvariableX-raysourcesinoursurveyhavebandrat iosgreaterthan0.5(ahigher bandratioindicateshighercolumndensitiesofobscuringm aterialalongthelineof sight).Wendthattheopticalvariabilitysignicancedec reaseswithincreasingamounts ofobscuration.UsingthemultiwavelengthSEDclassicati onsof Alonso-Herreroetal. ( 2006 ),wendthatwhilemost(70 % )oftheopticalvariablesareclassiedasBLAGNs, 20 % areborderlineNLAGN/BLAGNSEDsand10 % (1of10)haveaNLAGNSED. Finally,wedetectedopticalvariabilityfortwomid-IRpow erlaw-selectedAGNthatare notdetectedinX-rays.Suchsourcesmaybeheavilyobscured AGN,whereX-raysand muchoftheoptical/UVlightisblockedbydustwhichre-emit sthelightinthemid-IR. Thesesourcesareamongthefaintestopticalsourcesinours urveyandalsoliecloseto thevariabilitythreshold. AlargefractionoftheAGNcandidates(74 % ),however,donotshowoptical variabilitysignicantenoughtoproduceuxchangesgreat erthan3timesthetypical photometricerror.Amongtherelativelyunobscuredsource s(i.e.,thosedetectedinthe X-raywithbandratioslessthan0.5),about50 % arenotvariable.Thisisconsistent 31

PAGE 32

withtheexpectationthatonly 60 % ofAGNshowopticalvariabilityonmonth-to-month timescales( Webb&Malkan 2000 ).Thus,themajorityofunobscuredAGNthatdonot showopticalvariabilitymaybeopticalvariableswhenobse rvedonlongertimeintervals ofyears.Theinclusionofadditionalepochsofimagingdata forGOODS-Swillhelpto answerthisquestion.Anotherpossibilityisthatsomeofth esourcesareintheclass of“opticallydull”galaxieswhichexhibitAGNX-raylumino sitiesbutshownooptical evidenceforaccretion.Somemaybedominatedbylightfromt henon-varyinghost galaxy,whileothersmayresideingalaxieswithlargeamoun tsofobscuringmaterialin thehost. Theseresultsunderscoretheimportanceofmultiwavelengt hsurveystoidentify morecompletesamplesofAGN.Inparticular,wendthatopti calvariabilityisan importantselectioncriterionforidentifyingX-rayweako rlow-luminosityAGNwhichmay fallbelowthedetectionthresholdinX-rayormid-IRsurvey sandwouldotherwisebe missed. 32

PAGE 33

Figure2-1.Standarddeviationvs.averageVmagnitude.The errorbarsarevaluesof error ,theRMSofthephotometricerrorsoverveepochs.Thelinet races outvaluesofthreetimestheRMSofthephotometricerrorsat each magnitude.Objectswithstandarddeviationsgreaterthant hisvalueare consideredtobevaryingintheoptical. 33

PAGE 34

Figure2-2.Signicancevs.averageVmagnitude(r=2.5pixe ls)forallobjectsinthis study.TrianglesrepresentobjectsdetectedonlyinX-rays ,squaresrepresent objectsselectedviaIRpowerlawbehavior,anddiamondsrep resentobjects thatappearinbothcatalogs.Thesolidlineatsignicance3 separatesthe variableobjectsfromthenonvariableobjectsasdiscussed inSection 2.3 34

PAGE 35

Figure2-3.Opticalvariabilitysignicancevs.spectrali ndexformid-IR-selectedgalaxies. Squaresrepresentthosesourcesfoundonlyviamid-IRselec tionand diamondsrepresentobjectsidentiedasbothIRandX-rayso urces.The verticallineat =-0.9showstheapproximatedividinglinebetweenSEDs classiedasNLAGNs(steeperSEDshavingmorenegativespec tralindices) andBLAGNs,andthesolidhorizontallineatsignicance3se parates variablesfromnon-variables. 35

PAGE 36

Figure2-4.Signicancevs.X-raybandratio,denedasBR=H B/SB.Trianglesrepresent sourcesdetectedonlyinX-rays,whilediamondsaresources detectedin bothX-raysandselectedinthemid-IR.Blacklledpointsar eobjectsfor whichthebandratioisonlyanupperlimit;graylledpoints areobjectsfor whichitisalowerlimit.Theverticallineat0.5separatesh ard(BR > 0.5) andsoft(BR < 0.5)bandratios,andthesolidhorizontallineatsignican ce3 separatesvariablesfromnon-variables. 36

PAGE 37

Table2-1.Optical,IR,andX-rayPropertiesofGOODSSource s AID a MIPSID b GOODSIDRADec h V nuc i Error Signicance c BandRatio d 113 1 -J033217.14-275402.553.0714194-27.900702525.830.030 .030.95--1.22 -mips003133J033246.84-275121.253.1951520-27.8558940 25.250.060.022.76-0.60.2-mips003149J033214.55-275256.653.0606141-27.8823767 25.710.020.030.61-0.90.26a-J033216.16-274941.753.0673266-27.828261024.630.0 30.021.55-< 0.37 -mips003485J033225.24-275226.653.1051776-27.8740506 25.810.060.031.73-1.40.2-mips003528J033234.46-275005.053.1436200-27.8347720 26.480.050.050.93-0.90.2-mips003618J033235.71-274916.053.1487963-27.8211171 26.660.060.061.09-1.20.29a-J033218.45-274555.953.0768751-27.765525824.940.0 10.020.48-< 1.27 -mips003537J033240.75-274926.553.1699410-27.8240740 26.310.380.057.31-0.60.2-mips003871J033233.02-274200.453.1375810-27.7001130 25.810.060.032.02-0.50.2-mips010818J033227.19-274051.453.1133032-27.6809373 26.980.310.074.24-0.90.239-J033201.58-274327.053.0065900-27.724162624.430.0 20.011.31--0.42 43-J033203.04-274450.153.0126503-27.747240026.330.1 30.042.83--0.30 44-J033203.65-274603.753.0152257-27.767682823.600.0 20.011.96--3.56 53-J033206.27-274536.753.0261253-27.760183024.590.0 30.022.10-< 0.67 60mips004281J033207.98-274239.553.0332562-27.710959 325.320.020.020.91-1.10.20.89 62-J033208.00-274657.353.0333350-27.782570626.090.0 10.040.33--0.51 63-J033208.27-274153.553.0344441-27.698206525.840.3 30.039.74-< 0.37 65-J033208.53-274648.353.0355495-27.780079723.280.0 00.010.35-> 1.85 66mips013611J033208.66-274734.453.0360946-27.792881 019.820.100.0050.11-0.80.20.34 37

PAGE 38

Table2-1.Continued AID a MIPSID b GOODSIDRADec h V nuc i Error Signicance c BandRatio d 68-J033209.45-274806.853.0393617-27.801887021.040.0 80.0025.78--0.93 73-J033210.76-274234.653.0448370-27.709605522.990.0 10.011.22-< 1.59 73-J033210.76-274234.653.0448370-27.709605522.920.0 30.012.97-< 1.59 75-J033210.91-274343.153.0454675-27.728627325.650.0 50.031.64-< 1.39 76mips004258J033210.91-274414.953.0454704-27.737484 623.390.110.0112.86-0.90.20.35 80-J033211.41-274650.053.0475311-27.780548124.360.0 20.011.17--0.98 83-J033212.20-274530.153.0508213-27.758356924.180.0 20.011.24-< 1.51 84-J033212.22-274620.653.0509337-27.772401725.510.0 10.030.43--1.37 88-J033213.25-274240.953.0551915-27.711350722.430.0 10.011.63--1.22 91-J033213.85-274248.953.0577260-27.713579225.720.0 50.041.25-> 2.33 94-J033214.00-275100.753.0583525-27.850204425.130.0 40.021.73--1.84 95-J033214.08-274230.453.0586652-27.708438225.460.2 30.038.40-< 1.04 96-J033214.43-275110.853.0601151-27.853005524.820.0 40.022.25--2.61 97-J033214.45-274456.653.0601915-27.749053826.420.0 60.051.19-< 2.16 102-J033214.98-274224.953.0624129-27.706910125.020. 020.021.24--0.62 106-J033215.80-275324.753.0658257-27.890200425.800. 060.031.89-< 0.27 107-J033215.79-274629.753.0657934-27.774928424.620. 020.020.94--113-J033216.74-274327.553.0697345-27.724311624.110. 010.010.94-< 1.20 114-J033216.85-275007.553.0702116-27.835416026.140. 070.041.74-< 1.71 115-J033217.06-274921.953.0710682-27.822740122.780. 020.012.78--1.46 117-J033217.14-274303.353.0714326-27.717586423.120. 140.0119.5--0.32 38

PAGE 39

Table2-1.Continued AID a MIPSID b GOODSIDRADec h V nuc i Error Signicance c BandRatio d 121-J033218.01-274718.553.0750617-27.788485924.120. 010.011.00-< 0.27 122-J033218.24-275241.453.0760024-27.878160624.620. 140.028.94--0.38 123-J033218.34-275055.253.0764119-27.848659925.790. 030.031.03--5.28 125-J033218.83-275135.553.0784670-27.859854526.600. 050.060.90--2.57 126-J033219.00-274755.553.0791461-27.798740825.770. 030.031.12--1.17 129-J033219.81-274122.753.0825338-27.689650422.390. 020.011.95-< 0.76 131-J033220.05-274447.253.0835448-27.746448426.170. 070.041.69--4.35 132-J033220.31-274554.753.0846122-27.765181325.190. 020.040.69--134-J033220.48-274732.353.0853209-27.792309325.200. 030.021.16--1.80 137-J033221.42-274231.253.0892636-27.708659925.140. 030.021.15-> 2.89 144-J033222.54-274804.353.0939086-27.801189626.400. 060.051.35-< 0.39 145-J033222.54-274603.853.0939227-27.767732825.760. 070.032.35--1.96 146-J033222.55-274949.853.0939514-27.830509326.110. 050.041.30--1.18 148-J033222.58-274425.853.0941038-27.740511024.870. 020.020.75--149-J033222.76-275224.053.0948392-27.873322022.100. 010.012.02--0.62 155-J033224.26-274126.453.1010655-27.690671624.190. 010.010.83--0.37 158-J033224.86-274706.453.1035834-27.785106423.710. 010.010.96-< 0.73 160-J033224.96-275008.053.1039901-27.835568826.260. 040.041.04--3.23 161-J033224.98-274101.553.1040892-27.683752823.670. 020.011.64--2.46 162-J033225.11-275043.353.1046078-27.845348925.090. 030.021.47--2.44 163mips010787J033225.17-274218.853.1048578-27.70521 9724.070.020.011.49-0.70.20.28 39

PAGE 40

Table2-1.Continued AID a MIPSID b GOODSIDRADec h V nuc i Error Signicance c BandRatio d 164-J033225.16-275450.153.1048319-27.913925925.290. 020.020.73-> 5.53 167-J033225.74-274936.453.1072675-27.826767023.940. 010.010.50-< 1.37 173mips013087J033226.49-274035.553.1103938-27.67653 9920.270.070.0034.21-1.30.20.23 174mips010823J033226.67-274013.453.1111149-27.67038 3825.800.030.030.97-1.10.20.53 177mips000309J033227.01-274105.053.1125287-27.68472 3819.750.110.0067.25-0.60.20.31 179mips003888J033227.62-274144.953.1150981-27.69580 5323.760.030.012.63-0.70.23.11 181-J033228.74-274620.453.1197554-27.772331624.680. 030.021.60--0.20 182-J033228.81-274355.653.1200615-27.732125021.880. 000.010.83--0.29 188-J033229.85-275105.953.1243696-27.851633125.620. 050.031.66--1.58 189-J033229.88-274424.453.1244949-27.740124821.200. 030.014.88-< 0.12 191mips014635J033229.98-274529.953.1249148-27.75830 1321.500.050.0014.81-0.70.20.36 192-J033229.99-274404.853.1249588-27.734675320.870. 010.012.09--0.29 193-J033230.06-274523.553.1252547-27.756535023.630. 090.019.63--0.28 195-J033230.22-274504.653.1258995-27.751274922.450. 140.0126.77--0.28 196-J033231.36-274725.053.1306522-27.790271025.710. 030.040.77-< 1.05 197mips003938J033231.46-274623.253.1310696-27.77310 6024.770.020.021.13-1.20.2 < 1.04 199-J033232.04-274451.753.1335033-27.747704225.140. 060.022.65--203-J033232.99-274117.053.1374396-27.688058426.410. 060.051.21-< 3.86 207-J033233.46-274312.853.1394133-27.720214320.940. 020.004.64--0.90 212-J033234.34-274350.153.1430925-27.730581324.640. 030.022.16--0.72 214-J033234.95-275511.253.1456346-27.919773225.140. 010.020.51--0.49 40

PAGE 41

Table2-1.Continued AID a MIPSID b GOODSIDRADec h V nuc i Error Signicance c BandRatio d 215-J033235.04-274932.653.1459896-27.825733326.720. 060.060.91-< 0.76 216-J033235.10-274410.653.1462691-27.736272625.430. 030.031.33-< 0.70 227-J033236.72-274406.453.1529803-27.735123725.030. 040.021.98--0.45 229-J033237.46-274000.153.1560803-27.666691923.630. 090.018.82--0.30 230mips010697J033237.76-275212.353.1573435-27.87008 5325.450.070.032.61-1.90.20.33 234mips003915J033238.12-273944.853.1588307-27.66244 4420.790.080.0035.21-0.80.20.23 236-J033238.87-274733.253.1619461-27.792548725.870. 100.042.67-< 0.79 237-J033238.76-275121.653.1615043-27.856005726.880. 110.071.66--1.27 242-J033239.09-274601.853.1628593-27.767160221.250. 090.0030.07--0.70 244-J033239.46-275031.853.1644211-27.842169324.200. 010.011.15--0.35 248-J033241.40-274717.153.1725205-27.788094724.370. 010.010.80-< 1.66 250-J033241.87-274359.953.1744526-27.733299625.710. 230.037.30--0.36 251mips003108J033241.85-275202.553.1743886-27.86735 3323.100.050.016.80-0.50.20.47 254-J033242.86-274702.753.1785932-27.784082025.270. 090.023.93--0.52 256mips009708J033243.24-274914.253.1801493-27.82060 4623.240.370.0145.95-0.60.20.36 258-J033244.01-274635.053.1833644-27.776381725.380. 030.031.24--0.23 260-J033244.27-275141.153.1844471-27.861419622.170. 020.013.74-< 0.17 261-J033244.31-275251.353.1846401-27.880917324.250. 210.0115.99--0.29 262-J033244.44-274819.053.1851517-27.805275423.810. 020.011.56--0.36 263-J033244.45-274940.253.1852272-27.827836525.290. 160.027.06--1.47 264-J033244.60-274835.953.1858335-27.809968425.480. 030.031.23--0.55 41

PAGE 42

Table2-1.Continued AID a MIPSID b GOODSIDRADec h V nuc i Error Signicance c BandRatio d 265-J033245.02-275439.653.1875770-27.910995023.150. 020.011.25-< 0.85 266-J033245.11-274724.053.1879518-27.789998623.280. 010.011.45-< 0.66 267-J033245.68-275534.453.1903431-27.926223122.790. 010.011.34--0.41 269-J033246.41-275414.053.1933716-27.903877423.880. 020.011.49--1.04 273-J033247.18-275147.553.1965772-27.863206725.950. 160.044.35-< 1.08 276-J033248.18-275256.653.2007350-27.882390723.580. 030.013.74--1.88 286-J033252.88-275119.853.2203537-27.855509924.130. 190.0115.25--0.40 292-J033256.71-275319.153.2362800-27.888631623.510. 010.011.40-< 0.62 a AIDstandsforAlexanderID( Alexanderetal. 2003 ) b MIPSIDistheMIPS24 mname( Alonso-Herreroetal. 2006 ) c isthespectralindexofIR-selectedgalaxies( Alonso-Herreroetal. 2006 ) d BandRatioisdenedasHardBand/SoftBand( Alexanderetal. 2003 ) 1 Thisisanadditionalsourcefromthe Giacconietal. ( 2002 )catalog;theIDnumberlistedisthatgivenbyGiacconietal .,notthatof Alexander etal. ( 2003 ) 42

PAGE 43

CHAPTER3 DATAANDTECHNIQUES 3.1GalaxyClusterSample ToinvestigatetheroleofenvironmentontheAGNphenomenon ,weselected severalgalaxyclusterstoperformacensusofAGN.Wechosec lustersforwhich archivaldatawasavailablethatwouldallowAGNtobedetect edviaopticalvariability, X-rayemission,andmid-Infraredproperties.Allofthegal axyclustersusedinthis workarepartofasampleobservedmultipletimeswiththeHub bleSpaceTelescope (HST)AdvancedCameraforSurveys(ACS)inasurveyforsuper novaeinmassive high-redshiftclusters(HSTprogramsGO10493,cycle14and GO10793,cycle15, P.I.AvishayGal-Yam).Oursampleconsistsof12clusterswi thredshiftsrangingfrom 0.50–0.89.AllhavealsobeenobservedwiththeChandraX-ra yobservatory,and sevenhavebeenimagedinthemid-IRwiththeSpitzerIRACins trument.Table 3-1 lists theclusters,theirredshifts,thenumberofACSepochsandw hetherSpitzerdataare available. SevenoftheclustersinthisworkarepartoftheMassiveClus terSurvey(MACS) ( Ebelingetal. 2001 ).Thissurveytargetsdistant(z > 0.3),luminous(L X > 10 44 erg/s),andthereforemassivegalaxyclustersselectedfro mtheROSATAll-SkySurvey. DetailedopticalandX-rayobservationswerecarriedoutby Barrett ( 2006 ),including spectroscopicconrmationofclustermembers,mass,veloc itydispersion,andvirial radiusdeterminations. Stottetal. ( 2007 )reporttheX-rayluminositiesfortheseclusters. Inaddition,HSTACSobservationsintwoopticallters,F55 5WandF814W,are availableandtwohavebeenobservedwiththeSpitzerIRACin strument.Anadditional veclustersincludedinoursamplearedescribedbelow.The physicalpropertiesofall sampleclustersaresummarizedinTable 3-2 43

PAGE 44

3.1.1CL0152-1357 CL0152hasabolometricX-rayluminosityofL X =1.6x10 45 erg/s( Maughanetal. 2003 ). Jeeetal. ( 2005a )measuredatotalprojectedclustermassof4.92x10 14 M and avirialradiusr 200 of1.14Mpc.Theclustershowssignicantsubstructureinit sX-ray morphologyindicativeofanongoingmerger,withtwopeaksi nitsX-raymorphology thatcorrespondtotwosubclustersrstidentiedby Ebelingetal. ( 2000 )andlater conrmedby Maughanetal. ( 2003 ).Bothsubclustersshowsimilartemperaturesof 5.6keV( Maughanetal. 2003 ).Acatalogof102conrmedclustermembershasbeen publishedby Demarcoetal. ( 2005 ),andtheclusterhasbeenimagedbyHSTACSin F775WandF625W.TherearearchivalChandraX-rayandSpitze rIRACobservations availableforthiscluster.3.1.2CLJ1226.9+3332 CLJ1226hasabolometricX-rayluminosityofL X =5.3x10 45 erg/s,anX-ray temperatureofT X =11.5keV,andatotalmassM=1.4x10 15 M withavirialradiusof 1.66Mpc( Maughanetal. 2004 ). Ellisetal. ( 2006 )identied45spectroscopicallyconrmedclustermembers.CLJ1226showslittledynamicact ivityandrelaxedX-ray emission,althoughXMMandChandraobservationshaverevea ledanasymmetryin thetemperaturedistributionindicativeofamergereventi nthecluster( Maughanetal. 2007 ).InadditiontotheChandraX-raydata,thereisalsoSpitze rIRACdataavailable aswell,andtheclusterhasbeenimagedinF814WwithHSTACS.3.1.3MS0451.6-0305 MS0451hasabolometricX-rayluminosityofL X =2.01x10 45 erg/sandanX-ray temperatureofT X =10.0keV( Donahueetal. 2003 ).Theclusterhasanelongatedbut smoothdistributionofgalaxies,and Moranetal. ( 2007 )identied319spectroscopicallyconrmedclustermembersbetween0.52 < z < 0.56.Ithasavirialradiusof2.6Mpc ( Moranetal. 2007 )andaM 200 of1.4x10 15 M ( Donahueetal. 2003 ). Moranetal. ( 2007 )speculatethatICM-relatedphysicalprocessesimportant intheevolutionof 44

PAGE 45

currentlyinfallinggalaxies,suchasgasstarvationandra mpressurestripping,beginto affectgalaxiesatalargeradius(e.g.,r 3.5–5.5Mpcformerging,0–2.5Mpcforram pressurestripping,0–3.5Mpcforstarvation,and0–5Mpcfo rharassment). Molnaretal. ( 2002 )identied14unresolvedX-raypointsourcesintheeldofM S0451,thoughthey claimthisnumberiswithin1 ofthenumberexpectedfromanon-clusterbackground eld.InadditiontoarchivalChandraobservations,therei sHSTACSdataavailable forMS0451inF814WandF555W,andithasbeenobservedwithth eSpitzerIRAC instrument.3.1.4MS1054.4-0321 MS1054isanAbellclass3clusterwithabolometricX-raylum inosityL X =23.3x 10 44 erg/s( Stottetal. 2007 )andavirialmassM 200 =1.1x10 15 M ( Jeeetal. 2005b ). Substructureinthedistributionofgalaxiesmatchesdiffu sesoftX-rayemission,and weaklensingindicatesthattheclusterisyoung,massive,a ndstillrelaxing( Hoekstra etal. 2000 ). Tranetal. ( 1999 )foundavirialradiusfortheclusterof1.8Mpc.Inaddition toACSandground-basedopticalandIRimaging,therehavebe enspectroscopic surveysofthisclusterusingKeck(e.g., vanDokkumetal. 2000 ; Tranetal. 2007 )to determineclustermembershipofmanygalaxiesintheACSima ges. Deep5GHzradioobservationsoftheclusterperformedby Bestetal. ( 2002 )have revealed6–7radio-loudAGNinthecluster,allofwhicharec onrmedclustermembers. Chandraobservationsby Johnsonetal. ( 2003 )revealedanothertwoconrmedcluster AGN,oneofwhichwasalsodetectedintheradio. Baietal. ( 2007 )combinedMIPS 24 mdatawithspectraofover400clustergalaxiesandverydeep K-bandimagingin ordertostudytheclustersIRproperties;theyfoundthatth etwoX-rayAGNfoundby Johnsonetal. ( 2003 )andoneoftheradiosourcesfrom Bestetal. ( 2002 )appearas IRsourcesaswell.MS1054wasimagedbyHSTACSinF775WandF6 06W,andhas SpitzerIRACdataavailableinadditiontoChandraX-raydat a. 45

PAGE 46

3.1.5SDSS1004+41 SDSS1004containsthequadruply-lensedbackgroundquasar SDSSJ1004+4112 atz=1.734( Inadaetal. 2003 );thereisalsoafaintfthimageprojectedthroughthe clusterBrightestCentralGalaxy(BCG)( Inadaetal. 2005 ). Williams&Saha ( 2004 ) modeledtheclusterandcalculatedavirialradiusof1–1.5M pcandavirialmassM 200 = 4.2x10 14 M Oguri ( 2010 )determinethatthecoreofSDSS1004ishighlyevolvedand acomparisonbetweenSDSS1004andMACSJ1423by Limousinetal. ( 2010 )shows thatthetwoclustersareverysimilarbasedonthedistribut ionoftheirmass,light,and gas.Thereforewechooseavirialradiusof1.35MpcforSDSS1 004forthisstudy,as itisthevalueofthevirialradiusofMACSJ1423andfallsint hemiddleoftherange suggestedby Williams&Saha ( 2004 ).SDSS1004wasobservedintwoHSTACSlters (F814WandF555W),andhasalsobeenobservedwithSPITZERIR ACaswellas Chandra. 3.2OpticalImageAnalysisandPhotometry Multi-epochopticaldataisrequiredtomonitorgalaxiesfo revidenceofanAGNvia opticalvariability.Furthermore,thehighresolutionofH STallowsthenucleiofgalaxies tobeaccuratelytargetedforsmallaperturephotometry(r aperture =0 00 .09),whichallows ustominimizethecontributionoflightfromthehostgalaxy whensearchingfornuclear variabilityinlowerluminosityAGNwhichmightotherwiseb emaskedbytheirhost. Eachclusterinoursamplehas2–3epochsofACSobservations ineitherF775Wor F814W(roughlyequivalenttoJohnson-CousinsI-band).Ele venofthetwelveclusters werealsoobservedforoneepochinasecondopticallter(F6 25W,F606W,orF555W, roughlyequivalenttoVintheJohnson-Cousinssystem).Tab le 3-3 liststherelevant informationfortheACSexposures.Allvariabilitydetermi nationwasdonewiththeI-band observationstakenineitherF775WorF814W;observationst akenintheV-bandwere usedtocharacterizetheAGNhostgalaxycolorswhereavaila ble(seeChapter 6 ). 46

PAGE 47

Exposuretimesrangefrom 4000–9000sandepochsaretypicallyspacedoneyear apart.3.2.1ImageCreation:MULTIDRIZZLE ThecalibrateddataforeachepochwereobtainedfromtheHST ScienceCatalog 1 asseveral t.tsles,whicharedatacubescontainingthereadoutfro mthetwohalves oftheACSchipseparately,aswellasadataqualityimagefor each.Typically,asingle epochiscomprisedof4to6 t.tsles.Forimagecombinationandanalysis,packages withinthePython-basedversionoftheImageReductionandA nalysisFacility(IRAF; PyRAF)wereused;specically,theMULTIDRIZZLEandTWEAKS HIFTSscriptsinthe stsdaspackageofPyRAFwereusedtocreatenalimagesforan alysis.Theeldof view(FOV)ofatypicalobservationontheskyis 5.2arcminutesonasideandthe imagesarescaledtoaresolutionof0 00 .03perpixel,correspondingto1.91–2.42Mpc fromredshift0.504–0.888. MULTIDRIZZLEisdesignedtocombineditheredHSTimageswhi lealsoperforming cosmicrayremoval.TWEAKSHIFTSisascriptusedinintermed iatestepstocalculate theresidualshiftsbetweenthemultipleexposurestakendu ringthesameepoch. MULTIDRIZZLEwasusedtocombinetherawimagesbyidentifyi ngbadpixels, performingskysubtraction,anddrizzlingtheinputimages ontoseparateoutputframes usingtherstimageinthelistasareferencepoint,produci ngdistortion-corrected, sky-subtractedimagesthathavebeenshiftedinx,y,rotati on,andscaletomatchthe rstimageinthelist.WethenusedTWEAKSHIFTStodetermine anyshiftsamongthe clusterimagesinx,y,rotation,andpixelscale.Theseshif tsweresubsequentlyusedas inputforprocessingtheimagesasecondtimewithMULTIDRIZ ZLEandthisprocess wasrepeateduntilallshiftsdeterminedbyTWEAKSHIFTSwer esub-pixel(generally requiringoneortwoiterations). 1 http://archive.eso.org/wdb/wdb/hst/hst_meta_science _classic/form 47

PAGE 48

3.2.2CatalogCreation Oncetheimageshadbeenalignedtosub-pixelprecision,asi ngledeepimage ofeachclusterwascreatedbycombiningdatafromallepochs ofobservationsofthat clusterintoasinglestackedimagewith0 00 .03/pixelresolutionusingMULTIDRIZZLE. Thisdeepimagewascreatedwiththemaximumexposuretimefo reachclusterand usedtocreateasourcecatalog.WeusedSourceExtractor(SE xtractor)todetect objectsforphotometryinthedeepclusterimages.Theresul tingcatalogwasthen inspectedtoremoveanyspuriousdetectionsofcosmicrays, saturatedstars,orobjects whichfellontheedgeoftheframe.Duetoslightchangesinth ecenteringoftheACS observationsbetweenepochs,thecatalogwascomparedbetw eenthetwo(orthree) separateepochsandanyobjectswhichdidnotappearinallob servationswerealso removedfromthecatalog. SExtractordetectsobjectsbydeterminingabackgroundand whethereachpixel belongstothatbackgroundortoanobject.Itthensplitsupt hepixelsdeterminedtobe non-backgroundintoseparateobjectsandwritestheircoor dinatestoacatalog.This resultingcatalogmaynotbecenteredexactlyonthebrighte stnuclearcomponent ofeachgalaxy,soinordertoensurethatthecoordinateswer ecenteredonthe brightestcentralcomponentwere-centeredcoordinatesus ingthePyRAFtaskPHOT (calgorithm=“centroid”,cbox=“5.0”,maxshift=“3.0”).T hisallowedtheprogramtochoose thebrightestregionofeachgalaxy,allowingforamaximums hiftfromtheinputposition of3pixels(0 00 .09)toavoidconfusionofthenucleuswithanyotherbrightn on-nuclear featureswhenrunningphotometry.3.2.3PhotometryofGalaxyClustersinMulti-EpochACSImag es Eachepochoftheclusterimagingisseparatedintimebyappr oximatelyayear. WeexpectmostAGNtobevariableonthesetimescales(e.g., Kooetal. 1986 ; Webb &Malkan 2000 ).Individualepochswereconstructedusingthedeepstacke dimage asareferencesothattheSourceExtractorcatalogwouldbep roperlyalignedforeach 48

PAGE 49

epoch.Eachindividualepochwasmultipliedbytheexposure timeandtheABMAG zeropointwasdeterminedfromthePHOTFLAMandPHOTPLAMkey wordsFITSheader asspeciedintheHubbleACSWebsite/ACSDataHandbook 2 : zeropoint = 2.5 log ( photram ) 21.10 5 log ( photplam )+18.6921 (3–1) Smallaperturephotometrywasperformedoneachepochusing thePHOTtaskin PyRAFusingapertureswithradius2,2.5,and3pixels(r=0 00 .06,0 00 .075,and0 00 .09). Basedontheradialprolesofstarsandcompactgalaxiesint heelds,r=2.5is 2x theFWHMofanunresolvedsource,whichisappropriateforus ehereastheAGN emissionwearesearchingforcomesfromanunresolvednucle arregion.Aslightly largerapertureofradius3.0pixels(0 00 .09)waschosenforthenalphotometriccatalog toensurethatchangesinthePSFwouldnotresultinspurious detectionsofvariables (e.g., Sarajedinietal. 2011 ).WelimitedourphotometricsurveytosourceswithI nuc 27.Atmagnitudesfainterthanthislimit,thenumbercounts ofsourcesquickly decline,indicatingrapidlyincreasingincompleteness(F igure 3-1 ).Wealsondthat thephotometricnoiseatmagnitudesfainterthanthislimit increasesgreatly,making variabilitydetectionlessreliable.Thismagnitudelimit correspondstoanabsolute magnitudeofM I =-15.25atz=0.5andM I =-16.79atz=0.9. 3.3DeterminationofVariabilityThreshold Inordertomeasureopticalvariability,thegalaxycluster samplewassplitintothree groupsbyexposuretimetomaximizethetotalnumberofsourc esusedtomeasurethe photometricerroranddetermineathresholdforvariabilit y.Group1consistsofthose clusterswiththreeepochsofobservationsandexposuretim es 3000sperepoch (MACSJ0257,MACSJ0717,MACSJ0744,MACSJ0911,SDSS1004). Group2consists ofclusterswithtwoepochsofobservationsandexposuretim es 3300sperepoch 2 http://www.stsci.edu/hst/acs/analysis/zeropoints 49

PAGE 50

(MACSJ1149,MACSJ1423,MACSJ2214).Group3consistsofclu sterswithtwoepochs ofobservationswithexposuretimes 2100sperepoch(CL0152,CLJ1226,MS0451, MS1054).InSection 3.3.1 wedescribethemethodofdeterminingvariabilityforclust ers with2epochsofACSobservations(Groups2and3),andinSect ion 3.3.2 wedescribe variabilitydeterminationforclusterswith3epochsofobs ervations(Group1). 3.3.1VariabilityDeterminationWith2Epochs InthecaseofclusterswithonlytwoepochsofACSobservatio ns,two“fake”epochs wereconstructedusingMULTIDRIZZLEinthesamemannerasth erealepochs.These fakeepochsweremadebymixingtherawimages,takingtwofro mepoch1andtwo fromepoch2–resultingintwo“average”representationsof theeldwithsimilarnoise anddepthasthetrueindividualepochs.Theseimagesandthe resultingphotometry carrythroughtheeffectsofobjectandskyPoissonnoiseand readoutnoise.However, nointrinsicvariationsshouldexistandthustheapparentv ariationsinphotometry betweenthemcanbeusedtodenethenoise-magnituderelati onuponwhichwebase ourvariabilitythreshold.Aperturephotometrywasperfor medonallsourcesinboth “fake”epochsandthemagnitudedifferenceofeachobjectwa smeasured.Wethen combinedthemeasuredmagnitudedifferencesofthesources fromallclustersinGroup 2and3todeterminethephotometricnoiseandsetathreshold forvariabilityinthereal epochsforeachgroup. Toquantifyphotometricnoiseinourimagesasafunctionofm agnitude,the magnitudedifferenceswerebinnedintomagnitudeinterval sandaGaussianfunction wasttothedistributionofmagnitudedifferencesineachb in.Thebinsizeswere chosensuchthattherewereanadequatenumberofsourcesine achbintodetermine anaccurateGaussiant.Thebinsizesgenerallyspannedafu llmagnitudeinthe brighterrange( 22–24)andhalfamagnitudeamongdimmerobjects( 24–27),as therearefarmorefaintobjectsthanbright.Figure 3-2 showsanexamplettotwo 50

PAGE 51

magnitudebinsforGroup2,illustratingtheincreasedphot ometricerrorobservedinfaint sourcescomparedtobrightsources. WemeasurethestandarddeviationoftheGaussiantstothem agnitudedifference histograms(called hereafter).Wethentapolynomialfunctiontothe value toproduceasmoothlychangingvariabilitythresholdasafu nctionofmagnitude. Themagnitudedifferenceofeachobjectwasthendividedbyt hevalueof atthat magnitude.ForaperfectGaussiandistribution,ahistogra mofthesevalueswouldbe welltwithaGaussianfunctionhavingawidth( )of1andthusanyobjectshowing achangeinmagnitudegreaterthan3 couldbeidentiedasvaryingat3-sigma signicance.WefoundthatthewidthoftheGaussiantto m/ forthephotometry ofallsourcesinthefakeepochswasgenerallyslightlylarg erthan1.Thisvalue, wasthenadoptedasamoreconservativevariabilitythresho ldbydeningthethreshold as3.0x x ,wherethevalueof varieswithmagnitude.Figure 3-3 illustratesthe histogramof m/ inthefakeepochsforallsourcesinGroup2. Thevariabilitythresholddeterminedusingthefakeepochs canthenbeappliedto thetruetime-differenceepochs,inwhich mag ( mag 1 mag 2 )wascalculatedandthe objectssortedintobinsusingthesamecriteriaasthatfort hefakeepochs.Variables werethenidentiedasobjectswithmagnitudedifferencesi ntherealepochsgreater thanthevariabilitythreshold3 .Figures 3-4 and 3-5 showthevariabilityplotsforall clustersinGroups2and3,andTable 3-4 liststheopticalvariablesidentiedinclusters with2epochsofHSTACSdataandtheirproperties.3.3.2VariabilityDeterminationWith3Epochs InthecasewherethreeepochsofACSdatawereavailable(Gro up1),variability wasdeterminedinadifferentway,takingadvantageofthead ditionalepochofimaging data.Todeterminethethresholdofvariability,theaverag emagnitudeandstandard deviationwerecalculatedforeachobjectusingthephotome tryfromallthreeepochs. ObjectsweredividedintomagnitudebinsandaGaussianfunc tionwasttothevalues 51

PAGE 52

ofthestandarddeviationofthemagnitudeineachbin.Apoly nomialfunctionwast tothehistogramcentralvalues(i.e.,themeanvalueofthes tandarddeviationasa functionofmagnitude).WealsotapolynomialtotheGaussi an valuesasafunction ofmagnitude.Thevariabilitythresholdwasthendenedast hecenteroftheGaussian +3x (where isthestandarddeviationofthet).Thus,anobjectwasden edas variableifthemeasurementofitsmagnitudeshowedastanda rddeviationabovethis thresholdwithinitsmagnitudebin.Thisapproachissimila rtothatusedinChapter 2 andin Sarajedinietal. ( 2011 )fortheir5-epochvariabilityanalysisoftheGOODSelds. Figure 3-6 showsthevariabilityplotsforallclustersinGroup1,andT able 3-5 lists theopticalvariablesidentiedinclusterswith3epochsof HSTACSdataandtheir properties.3.3.3Summary:OpticalVariablesinClusterImages Wecomparedouropticalvariableswiththesupernovacatalo gof Sharonetal. ( 2010 )andtwoobjectsidentiedassupernovaewereremovedfromo ursample,asour aimisonlytostudytheAGNpopulation.Wend178opticalvar iablesinoursampleof 12clusterimages.Ninety(51%)ofthesevariableshavegrea terthan4 signicance. Wedetectanaverageof15variablespercluster,witharange of8–24variablesper cluster.Wefoundthatintotal,1.1%(178/15,849)ofallgal axiesinourclusterimages arevariabledowntoI nuc =27.Wecomparethepercentageofvariablesinclustersto thatfoundamongeldgalaxiesinChapter 4 Wealsoestimatepossiblesourcesofincompletenessofourv ariabilitysurvey. WeconsiderincompletenessduetoundersamplingoftheAGNl ightcurve.Previous variabilitysurveyswhichhaveimagedthesameeldseveral timesoverthecourse ofseveralyears(e.g., Treveseetal. 1994 ; Hawkins 2002 )havefoundthatvirtually allAGNvaryoveroversuchatemporalbaselinewhenseverale pochsofdataare analyzed.Inoursurvey,wesampleonly2or3pointsontheAGN lightcurveover 1 year,whichshouldresultinsomeincompletenessinourvari abilitysurvey.Inastudyof 52

PAGE 53

theHubbleDeepFieldusing2epochsofdataseparatedbyafew years, Sarajedinietal. ( 2003 )estimatethat 75%ofvariableAGNwouldbedetected.Weexpectthissame levelofcompletenessinoursurvey,aswehaveasimilarnumb erofepochsandsimilar temporalcoverage. Weestimatethenumberoffalsepositivesorspuriousvariab lesourcesinour variabilityanalysis.Figure 3-7 showsahistogramofvariabilitysignicanceforall galaxiesinourclustersample.Thedataarewell-tbyaGaus siandistributionoutto 3 ,withasignicantlyseparatepopulationofvariablesexte ndingtohighervaluesof .Inanormaldistribution,wewouldexpectthat 48sourceswouldhavevariations > 3 .BasedonGaussianstatistics,weestimatethat 30–40sources(17–22%ofour variables)couldbespuriousdetections. 3.4X-rayDataReductionandPointSourceDetermination AGNareknowntobeluminousX-raysourcesandcanberoutinel yselected fromdeepX-rayimages.AlthoughmanypreviousX-raysurvey sforpointsourcesin galaxyclustershavebeenhinderedbytheemissionofhotint raclustergas,thecurrent generationofX-rayobservatoriesofferthesensitivity,r esolution,andpositionalaccuracy neededtodetectpointsourceseveninthecoresofclusters, andwaveletdetection techniques(e.g., Freemanetal. 2002 )allowforaccurateseparationofpointsource emissionfromthediffusebackground.Thus,theChandraX-r ayObservatoryandits supportedsoftwarepackage,CIAO,allowustosearchforX-r aypointsourcesinthe coresofthegalaxyclustersinoursurveysample.3.4.1DataAcquisitionandReduction ArchivalX-rayobservationstakenwiththeChandraX-rayOb servatorywere obtainedfromtheChandraDataArchiveviatheChandraWebCh aSeRsite. 3 When available,multipleepochsofX-raydatawerecombinedtoma ximizeexposuretimeand 3 http://cda.harvard.edu/chaser/ 53

PAGE 54

allowdetectionoffaintersources.Theeffectiveexposure timeusedforpointsource extractionislessthanthetotalexposuretimeduetoremova lofbackgroundare events(discussedbelow).Table 3-6 liststheobservationIDs,totalexposuretimes,and effectiveexposuretimesforeachcluster. DatawerepreparedforanalysisusingCIAO4.1.1andfollowi ngthestepsinthe ACISDataPreparationintheAnalysisGuidefoundontheChan draX-rayCenter website 4 .Stepsweretakentoremovetheoriginal acis detect afterglow correctionandcreateanewACISbadpixelle,customizingt hisnewlefordata analysisbeforere-applyingthebadpixelcorrectionstoth edata.Thedataarethen lteredbyenergyintofull(0.5–8keV),hard(2–8keV),ands oft(0.5–2keV)bands, andmultipleobservations(whenavailable)weremergedint ooneimagetomaximize exposuretimeforthedetectionofX-raypointsources.Thes estepsaredescribedin moredetailhere. X-rayafterglowsareresidualchargesleftoverfromtheint eractionofcosmicrays withtheCCD,someofwhichcanbecaughtbythechargetrapand lingerforuptoafew dozensubsequentframes.Thiseffectcancausespuriousdet ectionsoffaintsourcesif notproperlyremoved.FordatatakenandprocessedwiththeC handrastandarddata processingpipelinepriortoversionDS7.4.0,thetool acis detect afterglow was usedtorejectbadpixelsfromthedata,whichsufferedfroma 3–5%systematicand non-uniformrejectionofvalidsourcephotonsasafterglow sfromdiffractedspectra.As ofDS7.4.0,amoreaccuratemethodwasdevelopedfortheiden ticationandremovalof afterglows;thus,therststepwastouseCIAOtoresettheaf terglowstatusofthedata bitsprocessedwiththeearlierversionbeforetheycouldbe re-analyzed. Oncetheerroneousafterglowcorrectionhadbeenremoved,a newlecontaining thebadpixelswascreatedusinganumberoftools: acis build badpix 4 http://cxc.harvard.edu/ciao/guides/acis_data.html 54

PAGE 55

acis find hotpix ,and acis classify hotpix .Thesetoolssearchforpixelswith biasvaluesthataretoolowortoohigh,aswellasthoseconta ininganunusuallyhigh orlownumberofevents.Theeventsonpixelslabeledassuspi ciousareclassiedinto afterglows,hotpixels,orrealsources,andbadpixelsaswe llasthesurroundingpixels arewrittenouttoabadpixelle. Usingthenewbadpixelle,thedatawerereprocessedtocrea tealevel=2event le,whichltersoutbadeventsbasedontheirgradeandstat us.Thegradeisavalue assignedtoeventsbasedonwhetherpixelsina3x3areasurro undingitareabove thethresholdvalue;aninitialgradeisgivenduringonboar dprocessing.Thestatusisa valuesettoindicateissueswithanevent,suchascosmicray s,hotpixels,streaking,or otherpossiblesourcesofcontamination.Whenaneweventl eiscreated,theimage islteredforbadgradesandstatuscolumnsinwhichalltheb itsaresetto0(good). TheimageisalsolteredtocontainonlyGoodTimeIntervals (GTIs),duringwhichall missionparameterswerewithintheiracceptableranges.Du ringthisprocess,thedata wereexaminedforbackgroundareevents,whichwerethenal soremoved,resultingin alowernaleffectiveexposuretimethanthetotaltimespen tobservingtheobject,as seeninTable 3-6 Finally,theobservationswerelteredbyenergyintofull( 0.5–8keV),hard(2–8 keV),andsoft(0.5–2keV)bandsandseparateepochswererep rojectedandmerged intoasingleobservationtoobtainthedeepestX-raydatapo ssible,resultinginthenal effectiveexposuretimeslistedinTable 3-6 .Thedataweretrimmedtothesizeofthe ACSeldofviewinordertorunsourcedetectionononlythear eacoveredbytheACS opticaldataavailableforcomparisonwithvariability-se lectedAGN. 3.4.2SourceDetection WedetectedX-raypointsourcesusingtheCIAOtool wavdetect inthefull (0.5–8keV)bandofthemergedX-rayimages.Thetoolworksin twostepstocorrelate theinputimagewithwaveletsatscalesspeciedbytheuser. First wtransform 55

PAGE 56

locatesprobablesourcesbycorrelatingtheimagepixelswi thMexicanHatwaveletsof differentscales.Then wrecon createsasourcelistofthesourcepropertiesinsideacell containingthemajorityofthesourceuxforeachdetection ForeachobservationID,wecreatedanexposuremapusingCIA Otoallow wavdetect toconvertthemeasuredsourcecountsintouxunits.Theexp osure mapisanimageoftheeffectiveareaateachpixelpositionon thesky,whichaccounts forditheringeffects.Itiscreatedusingtheaspecthistog ramle,whichmeasures theproductofthequantumefciencyoftheCCDandtheeffect iveareaofthemirror projectedontothesurfaceofthedetector.Theexposuremap istheninputinto the wavdetect toolparameterswhenitisrun;thewaveletscalesandsigni cance thresholdwerealsouser-specied. ThewaveletscalesparameterspeciestheradiusoftheMexi canHatfunctionused forsourcedetection. wtransform createsasetofoutputsforeachoftheinputscales, whichareinunitsofpixels.Thesignicancethresholdpara meterissimplythethreshold speciedforsourcedetection;itischosentobeapproximat elytheinverseofthetotal numberofimagepixels,suchthattheexpectednumberoffals esourcesperimageis one. The wavdetect toolwasrunwiththeexposuremapforeachimagespeciedwit h thefollowingparameters: psetwavdetectscales=“1.02.03.04.08.010.016.032.064. 0” psetwavdetectsigthresh=“1e-06” wavdetect alsoreportsasignicanceforeachdetectioninsigmaabove the backgroundlevel.Onlydetectionswith3 condenceorhigherabovethesurrounding areaareconsideredinthiswork. 56

PAGE 57

3.4.3SourceProperties Sourceuxesinthefull,hard,andsoftbandswerecalculate dusingtheCIAOtool eff2evt inCIAOversion4.3. eff2evt calculatestheuxinerg/cm 2 /sforaspecied locationonthechip,takingintoaccountthequantumefcie ncyandeffectivearea. Flux =( Energy = QE EA LIVETIME )) ( eV = erg ) (3–2) whereEnergyistheenergyofthephotonsobserved,QEistheq uantumefciency,EA istheeffectivearea,andLIVETIMEistheexposuretimeofth eimage.Theconversion factorfromeVtoergis1.60217646x10 12 Theuxwasmeasuredwithinacircularaperturecreatedarou ndeachsourcewith aradiusequivalenttothesourceradiusdeterminedby wavdetect .Thebackground uxatthepositionofeachsourcewasmeasuredusinganannul us2–5timesthesize ofthesourceradius,andsubsequentlysubtractedfromthes ourceux.Fluxeswere calculatedineachband(full,hard,soft)ineachindividua lobservationID;theresults weresummedtoobtainthetotalsourceux,withweightingba sedontheexposuretime ofeachobservation.Forexample,intheclusterCLJ1226(tw oobservationIDs:3180 and5014)theuxofeachobjectineachbandwascalculatedin thefollowingway: f tot =( f 3180 t exp ,3180 + f 5014 t exp ,5014 ) = ( t exp ,3180 + t exp ,5014 ) (3–3) Forthepurposeofcomparisonwithotherwork,wecalculated rest-frameX-ray uxesassuminga1DpowerlawplusGalacticabsorptionmodel andassumedallX-ray sourcestobeattheclusterredshift.Weusedtheforeground Galacticextinctiontoward eachclusterfrom Dickey&Lockman ( 1990 )andassumeda =1.7 powerlawtypical ofX-ray-selectedAGN(e.g., Martinietal. 2006 ),where istheslopeofthepowerlaw photonuxdensity N E / E .K-correctionstotheobserveduxeswerecalculated usingthe calc kcorr functionoftheSherpapackageinCIAO4.3.Figure 3-8 shows theX-rayuxvs.distancefromthecenterofthecluster(de nedtobeattheposition 57

PAGE 58

oftheBrightestClusterGalaxy,BCG)asafractionofthevir ialradius.Whilethegure showsboththemeasuredsourceux(plussymbols)andthebac kground-subtracted sourceux(triangles),onlythebackground-subtractedu xisconsideredthroughout thiswork.Itcanbeseenthatthebackgroundsubtractioniss ignicantintheinner partsoftheclusters,whereX-rayemissionfromhotintracl ustergasissignicant;the backgroundsubtractionissmallintheouterpartsoftheclu sters,wherethecontribution fromthisgasisminimal.Thisguredemonstratesourabilit ytodetectX-raysources downtoapproximatelythesamecompletenesslimitintheinn erregionsoftheclusters, whereextendedclusterX-rayemissionispresent.3.4.4Summary:X-ray-DetectedSourcesinClusterImages X-raypointsourcesineachclusterandtheirpropertiesare listedinTable 3-7 Wend74X-raypointsourcesinour12clusters,resultingin asourcedensityof822 X-raypointsourcespersquaredegree.Wendanaverageof6X -raypointsourcesper cluster,witharangeof1–12perclusteroverourentiresamp le. Figure 3-9 showsahistogramoftheX-rayuxinourthreebands.Weareab leto detectsourcesinthefullbandtoauxof 7x10 16 erg/cm 2 /s.Assumingastandard cosmology(H 0 =70, n M =0.3, n =0.7),thisuxcorrespondstoanX-rayluminosityof 6x10 40 erg/sataredshiftof0.5and 3x10 42 erg/sataredshiftof0.9.Oursurvey isthussensitivetomostAGNacrosstheredshiftrangeofour clustersample.Table 3-8 liststheparametersusedtocalculateX-rayluminositiesf orourpointsourcesample. WecalculateX-rayhardnessratiosforobjectsdetectedinb oththesoft(0.5–2keV), andhard(2–8keV)bands.Followingthesameprocedureasfor themergedfullband images, wavdetect wasusedtoanalyzethemergedsoftandhardbandimages.For objectsdetectedinbothbands,wecalculateahardnessrati oHR(e.g., Martinietal. 2002 ): HR = F X (2 8 keV ) F X (0.5 2 keV ) (3–4) 58

PAGE 59

Figure 3-10 showsthedistributionofhardnessratiosforthe69X-raypo intsources whicharedetectedinboththehardandsoftbands.Wendthat mostsourceshaveHR valuesonthesoftendofthedistribution,with46sources(6 7%)havinghardnessratios 5. 3.5InfraredImageAnalysisandPhotometry IfdustispresentaroundtheAGN,muchoftheX-ray,UV,andop ticalemissioncan beabsorbed,reprocessed,andre-radiatedasinfraredligh t.Galaxieswhoseemission isdominatedbyanobscuredAGNtendtohavemid-IRSEDswhich followapowerlaw ( f / ,where isthespectralindex). Donleyetal. ( 2008 )ndthatapowerlawt tomid-IRphotometryproducesamorereliableAGNsamplecom paredwithcolor-color selectionaswellasmid-IRexcessselection.SelectingAGN viaapowerlawttothe SEDinthemid-IRproducesacatalogofobjectsthatmaynotbe identiedviaopticalor X-rayemissionduetoobscuration.3.5.1DataAcquisitionandReduction ArchivalSpitzerobservationstakenwiththeIRAC(Infrare dArrayCamera) instrumentwereavailableforsevenclusters(indicatedin Table 3-1 )withexposure timesrangingfrom 900–6000s.AllobservationswereprocessedusingtheIRAC pipeline(versionS18.0orlater),andthuscorrectedBCD(b asiccalibrateddata)les knownascBCDleswereavailablefromtheSpitzerScienceCe nter.Theseimages werecorrectedforartifactssuchasmuxbleed(residualsig nalinmultiplexers),column pulldown(biasshiftincolumnscontainingbrightsources) ,banding(biasshiftinrows containingbrightsources),andrst-framecorrection(te mporaldependenceofthebias ofeacharray). ThedatawereobtainedusingtheSpitzerPrideprogramLeopa rdandmosaicked intoimagesforphotometryusingtheMOPEX(MOsaickerandPo intsourceEXtractor) 59

PAGE 60

toolavailablefromtheSpitzerScienceCenterwebsite 5 ,whichprocessescalibrated imagesintoasciencegrademosaicforpointsourceextracti onandphotometry.Point sourceextractioncanalsobeperformedwithMOPEX,butwasi nthiscaseperformed withSourceExtractorinordertoincorporatecatalogsderi vedfromtheopticalandX-ray observations. MosaicswerecreatedfollowingtheSpitzerDataAnalysisCo okbook. 6 TheMOPEX modules“Overlap”and“Mosaic”wereruntomatchthebackgro und,astrometrically registerandresampletheimages,removeoutliers,andcomb inetheindividualdata intothenalmosaickedimage.AFatalMaskBitPatternwasin puttospecifythatpixels aggedasbad(i.e.,affectedbyartifacts)inthecBCDsshou ldnotbeused. TheOverlapmodulecorrectsvariationsinthebackgroundle velbycalculatingand applyingacorrectiontocreateaconsistentbackgroundina llimages.Themodule rstproducesaninitialmosaic,thendetectsandmasksobje ctsthatcontributetothe backgroundandmeasuresthebackgroundlevelinoverlappin gframes.Fromthis information,thebackgroundcorrectioniscalculatedanda pplied. TheMosaicmoduleidentiesandremovesoutliersfromtheim ages,then resamplesthemontoacommonreferenceframeandcombinesth emintothenal mosaickedimage.Italsoprovidesaweightmapforeachmosai c.Forthepurposesof thiswork,the“MosaicOutlier”and“MosaicBoxOutlier”rou tineswereusedtosearch forpixelsthatdeviatedfromthemedianintimeandspaceusi ngmultipleoverlapping exposures.Cosmicrayswerealsorejected,andthenalimag eswerestackedand reprojectedusingthe“Drizzle”algorithmintoanalmosai cwithascaleof0 00 .6/pixel. ThemosaicsweretrimmedtothesizeoftheACSeldofviewfor eachclustertoinclude onlyobjectswithopticalvariabilitydata. 5 http://ssc.spitzer.caltech.edu/ 6 http://ssc.spitzer.caltech.edu/dataanalysistools/co okbook/ 60

PAGE 61

3.5.2SourceExtractionandPhotometry Sourceidenticationandphotometrywereperformedineach wavebandwith SourceExtractor,usingthe3.6 mbandasareferencecatalog.Photometrywas performedwithaperturesofradius3 00 .6(6pixels,or3“native”IRACpixels)anda backgroundannulusfrom3 00 .6–8 00 .4(8pixels,3–7“native”IRACpixels),followingthe suggestionsintheIRACinstrumenthandbook. 7 Theweightmapcreatedalongwith themosaicwasalsospecied.Fluxesweremeasuredwithinsp eciedxedcircular apertures(6pixels)withSourceExtractorandthecatalogw asfurtherrenedtoinclude onlyobjectsdetectedinallfourIRACwavebandsusingtheSP HEREMATCHroutine inIDL.TheuxofeachsourceinmicroJywasobtainedbyconve rtingtheuxfrom SExtractor(inMJy/sr)tomicroJy/pixandapplyingaspeci edaperturecorrectionfrom Table4.7inversion1.0(February2010)oftheSpitzerIRACI nstrumentHandbook. 8 ThezeropointstoconvertuxintoVegamagnitudeswereobta inedfromAppendixB (PerformingPhotometryonIRACImages)oftheInstrumentHa ndbookandarelisted below: Channel1(3.6 m):18.80 Channel2(4.5 m):18.32 Channel3(5.8 m):17.83 Channel4(8.0 m):17.20 Figure 3-11 showsahistogramofobjectmagnitudesineachoftheIRACcha nnels.A magnitudeof18inChannel4(8 m)correspondstoanIRluminosityofL IR ,8 m =1.4x 10 42 erg/sataz=0.5andL IR ,8 m =6.0x10 42 erg/satz=0.9. 7 http://ssc.spitzer.caltech.edu/irac/iracinstrumenth andbook/ 8 http://ssc.spitzer.caltech.edu/irac/iracinstrumenth andbook/home/ 61

PAGE 62

InordertoidentifyAGNcandidateswithpowerlawemissiont hroughthemid-IR wavelengths,theuxoverallfourIRACchannelswastwitha powerlawoftheform f / (3–5) andminimized 2 ,selectinggalaxiesthatarewellt(0.5 < 2reduced < 1.1)withaspectral index < -0.5within1 oftheirerrors,followingtheselectioncriteriaof Alonso-Herrero etal. ( 2006 ).Thisfollowsfromtheworkof Ivezi cetal. ( 2002 ),whichshowsthatthe opticalspectralindicesofoptically-selectedquasarsin theSloanDigitalSkySurvey areintherangeof-0.5 << -2. Alonso-Herreroetal. ( 2006 )alsondthattheslope ofthepowerlawtrelatestotheAGNtype,wheresteeper(i.e .,morenegative)values representNLAGNsandshallowerpowerlawSEDsareclassied asBLAGNs. Figure 3-12 isanexampleofoneofourmid-IRpowerlaw-selectedAGN.Tab le 3-9 liststheIRACpowerlawAGNcandidatesidentiedintheseve nclustersinourstudy withSpitzerobservationsalongwiththeirproperties.3.5.3Summary:IRPowerLawSourcesinClusterImages Wend64sourceswhichshowpowerlawSEDemissionovertheSp itzerIRAC bandsinthesevenclustersinoursamplewithSpitzerobserv ations.Wendanaverage of9IRpowerlawsourcesperclusterandarangeof2–18.Thear eacoveredbythese seveneldsis0.05squaredegrees;thus,wendadensityof1 ,360IRpowerlaw sourcespersquaredegree.Ofour64sources,20(31%)havesl opessteeperthan = -0.9andaresimilartotheNLAGNSEDsof Alonso-Herreroetal. ( 2006 ).Theremaining 44(69%)haveshallowerslopesandmorecloselymatchtheBLA GNSEDtemplateof Alonso-Herreroetal. ( 2006 ).Weexplorethemid-IRcolorsoftheseandAGNselected viaopticalvariabilityandX-rayemissioninthefollowing section. 62

PAGE 63

3.6ComparisonofOpticalVariables,X-rayPointSources,a ndMid-IRAGN Candidates ThecatalogsofX-rayandmid-IRAGNcandidateswerematched againsttheoptical catalogusingtheIDLroutineSPHEREMATCHwitha2 00 (67pixels)maximummatch radius.Whiletherewasgenerallygoodalignmentbetweenth eX-rayandIRdata,linear offsetswerequantiedandappliedtothecatalogswherenee ded. Table 3-10 liststheIR-andX-ray-selectedAGNwithopticalcounterpa rtsinour clusterphotometricsurvey,aswellasthenumberofAGNcand idatesdetectedwith morethanoneofthethreetechniquesusedhere(opticalvari ability,X-rayemission, ormid-IRpowerlawsources).Wendthat2/48(4%)ofmid-IRp owerlawsources aredetectedasopticalvariablesandoneoftheseisvariabl eat > 4 condence. TheoverlapisgreateramongtheX-raypopulation,wherewe ndthat12/50(24%) X-raypointsourcesareopticalvariablesandtwo-thirdsof thesearevariableat > 4 condence. Wecancomparetheseresultstothoseofrecenteldstudiess uchas Sarajedini etal. ( 2011 )andChapter 2 ofthisthesis( Klesman&Sarajedini 2007 ),whichexamine thepropertiesofAGNcandidatesidentiedusingsimilarte chniquesasthoseused inthisstudy.InChapter 2 ,wepresentaV-bandopticalvariabilityanalysisforX-ray andmid-IR-selectedAGNintheGOODS-Seld.Wendthatopti calvariabilitywas observedin26%ofX-raysourcesand45%ofmid-IRpowerlawso urces.Thesevalues aresimilartothosefoundinthecombinedGOODS-NandGOODSSeldspresented in Sarajedinietal. ( 2011 ),inwhich24%ofX-raysourcesand48%ofmid-IRpowerlaw sourceswereidentiedasvariable.Thesevaluesdonotchan geconsiderablywhen welimittheGOODSsurveystoincludeonlyX-raysourcesthat wouldbebrightenough (F X & 2x10 15 erg/s/cm 2 inthefullband)tobedetectedintheshallowerChandra observationsoftheseclusters.AmongtheX-raysources,we ndasimilarpercentage (24%)whichexhibitopticalvariabilityintheclustersast hatfoundintheeldpopulation, 63

PAGE 64

whileasignicantlylowerfractionofmid-IRpowerlawsour ces(4%)areidentiedas variableinoursurveythantheeldsurveys( 45%). Animportantdifferencebetweentheeldsurveyandtheclus tersurveypresented inthischapteristhetypeandtemporalsamplingofthedatau sedtodetermineoptical variability.IntheGOODSelds,variabilitywasmeasuredu singV-band(F606W) photometricdatawith5epochscoveringa6-monthtimespan. Intheclusteranalysis, I-band(F775WandF814W)photometrywasusedsincemulti-ep ochimagingwas onlyavailableatthiswavelength.Inaddition,variabilit ywasdeterminedfromjust2 or3epochs,thoughgenerallyoveralongertimebaselineof1 or2years.Weexpect ourvariabilityanalysistobesomewhatlesssensitivetova ryingnucleiatthislonger wavelength,sinceithasbeenshownthatopticalvariabilit yamplitudesincrease withdecreasingwavelength( VandenBerketal. 2004 ).Basedonacomparisonof Sarajedinietal. ( 2000 )and Sarajedinietal. ( 2003 ),weestimatethatabout 50%of V-band-detectedvariableswouldbeidentiedasvariables inourI-bandsurvey.Thus, thelowerpercentageofmid-IRsourcesidentiedasvariabl einoursurveymaybe partiallyexplainedbylesssensitivitytoweakly-varying nucleiintheI-band.Indeed,this isseeninFigure 2-2 ,whichshowsthatthevariabilitysignicanceofmid-IRsou rcesis generallyquitelow,indicatinglow-levelamplitudesofva riability.TheX-raysourcesdo notshowthesamelevelofsensitivity,whichmaybeduetothe factthattheygenerally displayoverallgreatervariabilityamplitudes(asdemons tratedinthelargenumberof > 4 condencevariablesamongthispopulationandalsoshownin Figure 2-2 ). Wecanalsoexaminethepercentageofvariablesthatareeith erX-rayormid-IR sourcesandndthat 7%ofvariablesareidentiedasAGNcandidateseitherthrou gh X-rayemissionorasmid-IRpowerlawsources.Thisislowert hanthefractionofeld variablesthatarealsoX-ray/mid-IRsources(36%ofGOODSv ariableswhencorrected fortheX-rayuxlevelachievedintheclustersurvey).This differencemayalsobe duetothelowersensitivityoftheI-bandvariabilitysurve yinwhichthelightmaybe 64

PAGE 65

moredominatedbystarsandincompletenessduetothesmalln umberofepochsas discussedinSection 3.3.3 WhencomparingourIRandX-raycatalogs,wendthat6/64(9% )ofallIR powerlawsourcesarealsodetectedasX-raypointsources;w henconsideringonly objectswithopticalcounterpartsthispercentageremains roughlythesame(4/48,8%). ApproximatelythesamepercentagesarefoundamongX-raypo intsourcesthatarealso IRpowerlawsources( 8%).TheoverlaphereislessthanthatfoundamongdeeperIR powerlawsourcesurveysconductedintheGOODSelds( Alonso-Herreroetal. 2006 ; Donleyetal. 2007 ),whereabouthalfofallgalaxiesdisplayingpowerlawbeha vior acrosstheIRACchannelsarealsoX-ray-detectedindeepX-r aysurveysoftheseelds. Atleeetal. ( 2011 )recentlyreportonastudyoflow-zclustersandndthatjus t 20%of AGNdetectedviaIRSEDsand/orX-rayemissionareidentied usingbothtechniques. TheyhypothesizethatthelackofX-rayemissionfrommostof theIRAGNsisdueto thelargercolumndensitiesofcoldgasinthehostsoftheseg alaxies,havingfoundthat IRAGNhostshavelargerspecicstarformationratesthanth ehostsofX-rayAGN.A moderatelylargecolumndensityofcoldgascouldsuppresst heX-rayemissionfrom theIRAGNs,makingthemundetectableinoursurvey.Additio nally,wenotethatthe sensitivityofclusterX-raysurveysislikelyimpededbyth eunderlyingX-rayemitting clustergas,whichmayresultinsomedifferencesinthenumb erofIRAGNdetectedin X-raysinclustersthanintheeld. Figure 3-13 plotsX-rayhardnessratiovs.IRpowerlawslopeforthe7AGN wedetectviabothmethods.OftheAGNwedetectinthemid-IRa ndX-rays,4/7 havepowerlawindicesindicativeofBLAGN( > -0.9)and6/7havesofterX-ray hardnessratios(HR < 5).TheobjectwiththehardestX-rayuxisalsoclassiedwi tha NLAGN-likeSED.Bothoftheobjectsalsodetectedviaoptica lvariabilityhaveshallow, BLAGN-likeSEDsandaresoftX-raysources.Thisagreeswith ourresultsfromthe surveyoftheGOODS-SeldinSection 2.3 .WendnoclearcorrelationbetweenX-ray 65

PAGE 66

hardnessandmid-IRSEDclassication,whichisinagreemen twithpreviouswork(e.g., Barmbyetal. 2006 ; Klesman&Sarajedini 2007 ).Variationsinthegas-to-dustratio orarangeofintrinsicAGNpropertiesmayexplainthereason forthepoorcorrelation betweenmeasuresofX-rayandIRobscuration,eventhoughob scurationshouldin principlehaveaneffectonbothwavelengthregimes( Barmbyetal. 2006 ). Next,weexaminetheX-raytoopticaluxratiosofourX-raydetectedAGN.From ourcatalogof74X-raypointsources,50haveopticalcounte rpartsforwhichtheI-band magnitudewasmeasured.WeplottheX-raytoopticaluxrati oofthesesourcesin Figure 3-14 ;overplottedarelinesofconstant log (F X /F opt ).AGNsareknowntooccupy thespace log (F X /F opt )= 1(e.g., Comastrietal. 2002 ; Rigbyetal. 2006 ; Georgakakis etal. 2004 ).Most(70%)ofourX-raysourcesliewithinthisregionofth ediagram, consistentwiththepresenceofanAGN.Bothgalaxieswedete ctviaallthreeAGN detectionmethodsandallgalaxiesdetectedinbothX-raysa ndviaamid-IRpowerlaw SEDarewithinthisrange.Oftheeightobjectsdetectedviab othvariabilityandX-rays,5 (63%)alsooccupyinthisregion. HighratiosofX-raytoopticaluxcanindicatethepresence ofanobscuredAGN, evenwithoutopticalevidencefornuclearactivity(e.g., Fioreetal. 2000 ; Hornschemeier etal. 2001 ; Giacconietal. 2001 ; Bargeretal. 2001 ).LowvaluesindicateX-ray-weak sourcesthatmaybeopticallybright.Objectswith log (F X /F opt ) -2areconsistentwith valuesofF X /F opt foundinstar-forminggalaxies(e.g., Moranetal. 1999 ; Alexander etal. 2002 ; Baueretal. 2002 ; Georgakakisetal. 2003 ). Georgakakisetal. ( 2004 ) ndsthatobjectswithF X /F opt belowthisvaluehaveopticalspectradominatedbythe hostgalaxy,thoughthepresenceofalow-luminosityAGN(LL AGN)cannotberuledout. X-raybrightoptically-normalgalaxies(XBONGs)inwhicht helightfromthehostgalaxy dominatestheopticalspectrumandmayhideopticalevidenc eofanLLAGNandtend tooccupytheregion-2 log (F X /F opt ) -1( Comastrietal. 2002 ; Rigbyetal. 2006 ). Whilemostofoursourceswith log (F X /F opt ) < -1areonlydetectedinX-rays,threeAGN 66

PAGE 67

candidatesarealsodetectedasopticalvariables,thusinc reasingthelikelihoodthatthey hostanAGNandtheX-rayemissionwedetectisduetoanaccret ingsupermassive blackholeratherthanprocessesrelatedtostarsandstarfo rmation.Suchsourcesare alsofoundineldstudiesofopticalvariability-andX-ray -detectedsources( Trevese etal. 2008 ; Sarajedinietal. 2011 ). Finally,weconsidertheInfraredpropertiesofoursampleo fAGNselectedusing varioustechniquesandmultiwavelengthdatatodeterminew hetherthemid-IRSED isdominatedbyemissionfromtheAGNorthehostgalaxylight Lacyetal. ( 2004 ) examinetheIRACcolorsofknownquasarsidentiedintheSlo anDigitalSkySurvey (SDSS)todenearegionincolorspacewhereobjectshavemid -IRcolorsthatare consistentwithhostinganAGN.TheseAGNareenergetically dominantintheirhost galaxies,affectingtheintegratedgalaxycolorssuchthat theyaretypicallyfoundinthis regionoftheIRACcolor-colordiagram.( Lacyetal. 2007 )ndthatformanygalaxies, classicationbasedonopticalemissionlineratiosagrees fairlywellwithclassication basedonmid-IRcolorthroughtheIRACchannels.WeusetheIR colorselectioncriteria of Lacyetal. ( 2007 )todeterminethenumberofourAGNcandidateswhosemid-IR colorsaredominatedbytheAGN(Figure 3-15 ). Table 3-11 liststhenumberofAGNcandidatesinourclustersamplewith mid-IR colorsindicativeofAGN(i.e.,foundwithintheLacywedge) .Wendthat14%(57/411) ofgalaxiesintheLacywedgeareidentiedasIRpowerlawsou rces,2%(10/411)are identiedasopticalvariables,and4%(18/411)areidenti edasX-raypointsources. Wendthat89%ofIRpowerlawsourcesareintheLacywedge,wh ile16%of opticalvariablesand44%ofX-raypointsourceslieintheLa cywedge.Conversely, 11%ofIRpowerlawsources,84%ofopticalvariables,and56% ofX-raypointsources arenotintheLacywedge,andthustheirmid-IRlightislikel ydominatedbythehost galaxyandstarformationratherthantheAGN. Atleeetal. ( 2011 )alsoconsiderthe mid-IRcolorsoftheX-ray-andmid-IR-selectedAGNintheir sampleusingasimilar 67

PAGE 68

color-selectiontechniquedevelopedby Sternetal. ( 2005 ),whichagreeswellwiththe selectionregiondevelopedby Lacyetal. ( 2007 ).Theyndthatgalaxieswhichfallin themid-IRAGNregionmusthavemorethan50%oftheirlightin theIRACchannels contributedbytheAGNcomponent.Aboutonequarterofthega laxiesintheirsample ofAGNhavecolorswithintheAGNregion.Approximately75–8 0%oftheirsampleof bothX-ray-andIR-selectedAGNdonotshowcolorsindicativ eofanAGNbasedon theirmid-IRcolors,whichisingeneralagreementwithour ndings.Thisalsoagrees wellwiththeresultsof Hickoxetal. ( 2009 )foreldgalaxies,inwhichtheyndthat32% ofX-raysourceshaveIRcolorswithintheSternwedge.Thus, wendnosignicant differencebetweenthepercentageofX-rayobjectswithIRc olorsindicativeofAGN emissionbetweentheeldandclusterpopulation. Thisillustratestheimportanceofamultiwavelengthappro achtoidentifyinga completecatalogofAGN.Mid-IRSED-basedselectionmaybea reliablemethodto detectmoreobscuredAGN,butitisclearthatinclustersasw ellastheeld,manyAGN arenotrevealedthroughthemid-IRcolorsofgalaxies.This maybepartlyduetoalow AGN/hostgalaxyuxratio,asislikelythecaseformostofth eopticallyvariableand X-ray-selectedAGN. Wendthatopticalvariability,X-rayemission,andmid-IR powerlawSEDs producesarobustcatalogofAGNcandidatesinourclustersa mple.Considering AGNdetectedinallthreetechniques,wendanaverageof25A GNcandidatesper clusterwitharangeof12–49overtheclustersample.Weexpl orethissampleofcluster AGNinthefollowingchapters,aswellastheimplicationsof environmentontheAGN population. 68

PAGE 69

Figure3-1.Histogramofnuclearmagnitude(F814W,I-band) inEpoch1forallgalaxies intheclustersample.Themagnitudelimitat27appliedinth isworkis indicatedbytheverticaldashedline. Figure3-2.Gaussiantstothehistogramofmagnitudediffe rencesintwomagnitude binsforGroup2.Bin3(left)isthedifferenceinmagnitudes forobjects betweenI=24.5–25.5andBin7(right)isthedifferenceinma gnitudefor objectswithmagnitudesbetweenI=27–27.5.Thewidthofthi sGaussians plottedhere, ,isusedtocalculatethevariabilitythresholdforthe magnitudebin. 69

PAGE 70

Figure3-3.Gaussianttoassessthephotometricnoiseinth efakeepochsinGroup2. Thevalueof forthistis1.06(thisis .Thusany“3 -condence” variabilitydetectionmustactuallydisplaymagnitudevar iationsof3.18times the valuedeterminedatitsmagnitude.ThetforGroup3wasfoun dto haveavaluefor of1.07. Figure3-4.Plotsofmagnitudedifferencevs.magnitudeine poch1forallclustersin Group2.Smalldotsarethedata;crossesare > 3 variablesandasterisks are > 4 variables.Theshortandlongdashedlinesillustratethe3 and 4 condencevariabilitythresholds. 70

PAGE 71

Figure3-5.Plotsofmagnitudedifferencevs.magnitudeine poch1forallclustersin Group3.Smalldotsarethedata;crossesare > 3 variablesandasterisks are > 4 variables.Theshortandlongdashedlinesillustratethe3 and 4 condencevariabilitythresholds. Figure3-6.Plotsofstandarddeviationvs.averagemagnitu deforallclusterswith3 epochsofACSdata(Group1).Smalldotsarethedata;crosses are3 variablesandasterisksare4 variables.Theshortandlongdashedlines illustratethe3 and4 variabilitythresholds,respectively. 71

PAGE 72

Figure3-7.Distributionof forallgalaxiesinourclusters.Thex-axisshowsvariabili ty signicanceforallgalaxies,normalizedtoremove inGroups2and3.The y-axisshowsthelogarithmofthenumberofgalaxiesinbinso f0.3.Error barsrepresentthePoissonstatisticalerrorsineachbin.T hesolidlineisa Gaussianttothedatawithin3 72

PAGE 73

Figure3-8.X-rayuxvs.distancefromthecenteroftheclus ter(asafractionofthe virialradiusindegrees)inthefull(top),hard(middle),a ndsoft(bottom) bands.Plussymbolsarethemeasureduxes,whiletriangles representthe background-subtractedux(usedthroughoutthiswork). 73

PAGE 74

Figure3-9.HistogramofthelogofX-rayux(erg/cm 2 /s)inthefull,hard,andsoftbands. Figure3-10.HistogramshowingthedistributionofX-rayha rdnessratiosforoursample ofX-ray-detectedAGN. 74

PAGE 75

Figure3-11.HistogramsofIRmagnitudesineachofthefourS pitzerIRACchannels. Figure3-12.ExampleofagalaxywhichshowsanSEDoverthefo urIRACchannels whichtapowerlawmodelwithaspectralindex between-0.5and-2. 75

PAGE 76

Figure3-13.X-rayhardnessratiovs.IRpowerlawslopeforA GNdetectedinbothIR andX-rays(squares).Starsdenotethe2objectswhichalsos howoptical variability. 76

PAGE 77

n nr n n Figure3-14.OpticaltoX-rayuxofX-raypointsourceswith identiableoptical counterparts;thelinesindicateconstantuxratiosof log (F X /F opt )=1,0,-1, and-2.Objectstowardthetopofthisplotaremoreoptically obscured,and objectstowardthebottomareopticallybrightandX-raywea k. 77

PAGE 78

Figure3-15.IRACcolor-colorplotforgalaxiesinall7clus terswithSpitzerobservations. TheregionwithinthedashedlineisthatusedtoselectAGNsv iathecolor selectioncriteriaof Lacyetal. ( 2007 ).SmallpointsaregalaxieswithIR emissioninallfourIRACchannels,anddifferentsymbolsde notetheAGN detectedviadifferentwavelengthtechniques(seelegend) 78

PAGE 79

Table3-1.ClusterObservations ClusterRedshiftRADec#ACSEpochsObs.w = Spitzer? CL 0152 1357 0.83101:52:43.00-13:57:20.002Y CLJ 1226.9+3332 0.88812:26:58.2133:32:49.42Y MACSJ 0257 2325 0.50602:57:08.83-23:26:03.33MACSJ 0717+3745 0.54807:17:32.9337:45:05.43Y MACSJ 0744+3297 0.68607:44:52.5839:27:26.73MACSJ 0911+1746 0.50409:11:11.1817:46:34.83MACSJ 1149+2223 0.54411:49:35.5122:24:04.22MACSJ 1423+2404 0.54514:23:48.6024:04:49.12Y MACSJ 2214 1359 0.50422:14:57.34-14:00:12.22MS 0451.6 0305 0.55004:54:10.48-03:01:38.52Y MS 1054.4 0321 0.83010:57:00.20-03:37:27.02Y SDSS 1004+41 0.68010:04:34.7241:12:44.983Y Table3-2.GalaxyClusterProperties ClusterL X (10 44 erg = s )M X ,200 (10 15 M ) r v (Mpc) (km/s) CL 0152 1357 160.451.141600 CLJ 1226.9+3332 531.41.66997 MACSJ 0257.6 2209 15.41.411.95 0.42735 MACSJ 0717.5+3745 27.42.751.93 0.281588 MACSJ 0744.8+3927 25.90.871.47 0.181060 MACSJ 0911.2+1746 13.20.841.61 0.501062 MACSJ 1149.5+2223 17.31.132.64 0.141789 MACSJ 1423.8+2404 15.00.781.35 0.191373 MACSJ 2214.9 1359 17.01.221.54 0.161393 MS 0451.6 0305 20.31.42.6650 MS 1054.4 0321 23.31.11.81170 SDSS 1004+41 -0.421.3579

PAGE 80

Table3-3.ACSObservations Cluster#ACSEpochsFilter(s)TotalExp.TimeinI(s) CL 0152 1357 2F775W,F625W4252 CLJ 1226.9+3332 2F814W4296 MACSJ 0257 2325 3F814W,F555W8858 MACSJ 0717+3745 3F814W,F555W8893 MACSJ 0744+3297 3F814W,F555W8893 MACSJ 0911+1746 3F814W,F555W8825 MACSJ 1149+2223 2F814W,F555W6774 MACSJ 1423+2404 2F814W,F555W6774 MACSJ 2214 1359 2F814W,F555W6640 MS 0451.6 0305 2F814W,F555W4198 MS 1054.4 0321 2F775W,F606W4232 SDSS 1004+41 3F814W,F555W9492 80

PAGE 81

Table3-4.OpticalVariables:2Epochs RADecMagnitudeMagnitudeDifference inEpoch1 mag E 1 mag E 2 Group2 MACSJ1149 177.391254622.374179619.7610.0624.11177.372773022.413181522.1690.1248.22177.385378322.415810022.1780.0624.11177.412953122.391234722.6840.0805.30177.403794322.404581223.0720.0463.05177.375922422.401495723.3600.22314.78177.368679922.404175123.9490.0593.36177.395469122.410094924.1880.0884.46177.407038122.408348124.6420.0833.34177.395201722.379885524.9000.1093.84177.377333222.401810226.2510.2173.48177.371768422.387904426.9130.2993.03 MACSJ1423 215.956699424.090813320.7860.39526.18215.965912824.092844221.3460.1006.63215.943562624.110371422.1460.50133.20215.920709424.085876322.9780.0765.04215.935161324.071543622.9800.0513.38215.978050224.093010323.1070.0463.05215.956785524.106728023.3150.0754.97215.965547724.086861723.3820.0865.70215.959715524.061909323.3840.21814.45215.935989024.105789423.3950.0503.31 81

PAGE 82

Table3-4.Continued RADecMagnitudeMagnitudeDifference inEpoch1 mag E 1 mag E 2 215.949592424.061261423.4470.0483.18215.976908724.091359123.4550.0503.31215.946353424.079487723.6360.0885.75215.920189124.083535323.8060.0875.29215.944076924.096772323.9940.1317.30215.950435824.059524524.0770.0593.16215.954228624.113251324.1810.19810.07215.988968424.080007524.5710.1435.97215.939852924.098254325.2840.39911.45215.950922924.110495925.3380.3028.41215.931937324.079226626.0260.4087.57215.948074424.116693826.0300.2444.52215.958758224.094324926.4070.2343.38215.940879424.085559026.8110.3213.50 MACSJ2214 333.7385399-14.010446022.1950.0714.71333.7499718-13.971977823.2940.0674.44333.7583932-13.991063423.6070.0624.09333.7527829-13.994906723.8680.1619.53333.7520584-14.026073123.8770.1106.48333.7341329-14.020537423.9590.0764.31333.7235526-14.004234124.4390.0803.57333.7239097-13.999556725.8810.1563.17 Group3 CL0152 28.1688786-13.945558619.5860.23114.70 82

PAGE 83

Table3-4.Continued RADecMagnitudeMagnitudeDifference inEpoch1 mag E 1 mag E 2 28.1607732-13.957999319.5890.18811.9728.1644026-13.939773520.6320.29818.9728.1822742-13.983721821.5890.0543.4428.1895690-13.976160522.5750.1197.5728.1803432-13.982578122.8570.21813.8828.1880417-13.947655123.0560.0593.7628.2001421-13.951998523.9870.1184.6028.1794506-13.922224724.5720.57215.1728.1900472-13.950217326.0560.2673.2828.1769691-13.958815626.7610.3913.0528.1775526-13.944382426.9910.5083.35 CLJ1226 186.731284233.525383221.3670.41526.41186.737709233.524743022.2040.16510.50186.713430933.568169822.4500.0935.92186.752300433.580129623.1390.0784.96186.724397033.516582123.5790.0683.64186.725085433.531065023.9940.0803.10186.774020033.568542224.6520.1293.27186.766870333.546143824.7050.42010.34186.739460833.554369324.8080.1363.17186.734348433.545417025.2790.1663.06186.752511033.567969125.3430.1943.46186.723101133.522807425.5700.2173.46186.769489733.537489525.7100.2163.21186.753528633.576475725.9020.2383.19 83

PAGE 84

Table3-4.Continued RADecMagnitudeMagnitudeDifference inEpoch1 mag E 1 mag E 2 186.760249633.546995426.0430.2443.02186.757656033.550984826.2400.3634.00186.746418733.579281726.2440.3964.35186.745347333.551151926.3070.3073.24186.754510733.542844726.4920.3463.25186.722811633.552483226.7750.4173.22 MS0451 73.5310816-3.024281719.9730.0774.9073.5470245-3.018921020.2310.0593.7673.5602406-3.021703422.5370.0875.5473.5400210-3.055474222.7410.1066.7573.5469272-3.008106122.8580.0573.6373.5429796-3.012845622.8660.0905.7373.5219066-3.009423223.2520.0593.9373.5352979-3.021808823.3170.0845.4173.5418986-3.011654123.3300.0754.7973.5359322-3.015049523.5210.0653.6473.5515820-3.037974623.6410.0763.8873.5400868-3.009790623.7100.1316.3373.5583410-3.046373023.8150.0773.4373.5440042-3.014233524.2830.1113.5273.5396628-3.006749424.4650.1083.0573.5386585-3.007009324.4830.1283.58 MS1054 164.2418732-3.616653920.7540.0543.44164.2872055-3.618083721.0340.0613.88 84

PAGE 85

Table3-4.Continued RADecMagnitudeMagnitudeDifference inEpoch1 mag E 1 mag E 2 164.2709896-3.627628021.5840.0664.20164.2512256-3.603636221.7070.1107.00164.2512507-3.625979522.0440.0523.31164.2641578-3.620169222.7030.0483.06164.2268129-3.638923822.7320.0966.11164.2510498-3.602329323.5430.0814.46164.2618582-3.598535923.8330.0773.38164.2892100-3.618512625.8990.2243.01164.2578790-3.621016426.1320.2873.37 85

PAGE 86

Table3-5.OpticalVariables:3Epochs RADecAverageStandard MagnitudeDeviation MACSJ0257 44.3055608-23.438953121.2700.24419.3944.2958934-23.444012622.4290.0826.5144.3086292-23.434909122.9140.0775.8844.2831760-23.461525323.8370.0665.0144.2698898-23.455810823.8510.0453.4244.2980084-23.400962024.2370.0473.4044.2826153-23.462528624.5040.0453.0244.3141248-23.430669724.7410.0543.3444.2738603-23.467920024.7520.0563.4844.2790992-23.407007126.1850.1834.8844.2930862-23.426325126.5340.1603.3744.3088571-23.448405926.5800.1503.0644.2747144-23.454500926.6250.1943.83 MACSJ0717 109.384423437.742602120.7600.0645.08109.391810337.745779721.0970.0675.33109.383344237.765178021.7550.0493.90109.409644537.741169321.9250.0443.47109.404932837.771370022.4460.0755.94109.409346537.780552922.7730.0493.76109.397585237.726285822.9860.0665.04109.366248337.725753323.3820.0544.11109.376952637.769620124.0020.0523.87109.396577937.761496324.3110.0634.43 86

PAGE 87

Table3-5.Continued RADecAverageStandard MagnitudeDeviation 109.372064737.729950724.3290.0896.29109.405548537.760348124.3600.0594.15109.394379937.754383524.3930.0614.22109.415691337.774677324.5970.17411.37109.410263137.781892724.7860.0503.06109.409228737.772386625.1110.0643.33109.412264137.744167525.3680.0713.23109.402745137.760036925.3950.0693.09109.391179137.757223026.7210.1743.22 MACSJ0744 116.237023939.452180620.1310.0786.21116.191348139.465465020.7090.18314.54116.197841339.470371220.9090.19415.40116.242874839.446286421.8010.0826.53116.235562939.463065522.1260.11910.45116.251790039.448417022.1520.18515.99116.247220939.444849022.9140.0624.75116.217053239.461449923.0600.0765.83116.213086139.456013823.4160.19915.19116.247806039.448939023.4500.0856.44116.262087539.457968823.4630.0634.81116.236578439.457973923.4910.14410.99116.242194639.456657323.9180.13710.28116.216087439.464855524.3510.1097.65116.208995439.451667324.5280.0493.30116.191310339.468045824.6760.0553.53 87

PAGE 88

Table3-5.Continued RADecAverageStandard MagnitudeDeviation 116.219935939.461817026.1780.1223.27116.198456439.473278126.7130.1763.29 MACSJ0911 137.794073717.781575622.5070.0524.10137.797592817.773748623.0940.0876.64137.818284817.794281223.6840.15211.52137.776497117.759349524.5540.0857.65137.798200017.799165724.6720.1815.62137.797620617.764681425.4240.11811.56137.797011317.777040825.5580.0833.41137.794685717.758056825.8320.1015.18137.782821517.755411326.1610.1373.36137.811078117.791532226.3120.1263.71137.781789017.767416626.5520.1513.09137.803801617.799267926.7640.1703.22137.799232017.791220226.8480.1783.14137.803946317.792273926.9120.1973.03 SDSS1004 151.130367541.204110820.2010.13510.75151.130193141.209447421.9060.0624.89151.134880941.222903522.0280.0997.87151.154955641.209028122.8130.29322.56151.152232841.209969523.0830.0403.08151.149116041.228738723.5480.0765.80151.132147541.196924223.9750.0503.71151.139320641.201858224.6630.0673.32 88

PAGE 89

Table3-5.Continued RADecAverageStandard MagnitudeDeviation 151.127679341.223138924.6650.2603.28151.136198041.201397124.7130.0537.65151.141092341.202604624.7430.0534.28 89

PAGE 90

Table3-6.X-rayObservations ClusterObsIDsTotalExp.Time(ks)EffectiveExp.Time(ks) CLJ 0152 1357 91336.9534.79 CLJ 1226.9+3332 3180,501465.2756.6 MACSJ 0257 2325 1654,351838.8336.27 MACSJ 0717+3745 1655,420080.0674.98 MACSJ 0744+3297 3197,3585,611190.7986.49 MACSJ 0911+1746 3587,501242.2339.11 MACSJ 1149+2223 1656,358939.0736.04 MACSJ 1423+2404 1657,4195135.88131.65 MACSJ 2214 1359 3259,501138.535.07 MS 0451.6 0305 529,90290.2184.59 MS 1054.4 0321 51258.8455.81 SDSS 1004+41 579481.0873.89 90

PAGE 91

Table3-7.X-rayPointSources RADecSourceFullBandHardBandSoftBandFullBandHardBand SoftBandHardness SignicanceFluxFluxFluxLuminosityLuminosityLuminosi tyRatio CL0152 28.1822599-13.9837439119.62.502.133.0383.1370.71100 .450.704 28.1658054-13.961427999.72.974.591.2798.65152.5542. 013.631 28.2054645-13.94944002.90.020.390.060.7213.001.946. 713 28.1794661-13.92222643.40.050.0010.051.620.051.630. 029 CLJ1226 186.774519133.537113039.11.540.960.5160.4437.4019.9 91.871 186.730677933.56192124.50.070.020.042.590.831.590.5 22 186.764829233.54867896.80.150.080.056.043.321.931.7 16 186.771341433.57082744.30.600.5623.0721.95186.747037033.57264193.70.200.030.527.791.020.630.0 48 186.742668733.54732506.00.060.030.032.431.261.211.0 47 MACSJ0257 44.2864042-23.434719522.30.330.180.223.281.752.11 MACSJ0717 109.396551737.761485915.20.740.610.158.847.211.794. 023 109.372344037.72992763.40.170.160.012.041.880.1512. 598 109.373396837.735590523.80.750.520.258.916.232.962. 107 109.404942437.739491614.00.590.410.186.974.882.132. 288 91

PAGE 92

Table3-7.Continued RADecSourceFullBandHardBandSoftBandFullBandHardBand SoftBandHardness SignicanceFluxFluxFluxLuminosityLuminosityLuminosi tyRatio 109.409542937.77109368.00.200.160.062.371.940.732.6 69 109.409413037.780448316.80.520.340.196.224.072.261. 802 109.382736637.75850137.80.0030.03109.374706437.75994735.50.020.070.280.82109.380477737.75593345.30.070.070.0060.890.850.0711 .429 MACSJ0744 116.212990939.455973318.80.270.100.155.552.153.010. 716 116.235555739.462984272.82.080.901.1042.9818.5522.7 60.815 116.233855039.44144468.90.590.560.0612.2711.521.169 .928 116.226173939.44639558.20.090.0060.081.930.121.670. 071 116.219711439.4570427113.34.382.481.7890.4651.2836. 661.399 116.225548139.459079710.20.110.060.042.181.280.761. 694 116.182531639.46811584.40.090.030.051.760.671.080.6 18 116.223506139.47221305.60.120.070.042.421.390.851.6 34 116.238641939.47829763.00.040.040.750.80116.206163839.47956733.00.040.0050.050.750.110.950. 114 116.210134639.48286723.40.050.061.131.25116.219106239.455542615.50.060.131.292.71 MACSJ0911 137.773591217.75493563.30.060.010.040.560.140.390.3 61 92

PAGE 93

Table3-7.Continued RADecSourceFullBandHardBandSoftBandFullBandHardBand SoftBandHardness SignicanceFluxFluxFluxLuminosityLuminosityLuminosi tyRatio 137.776459417.75934295.90.420.280.134.122.671.312.0 48 137.782840817.75735405.00.170.070.071.700.730.631.1 51 137.813382617.76648736.90.470.230.244.522.242.330.9 62 137.781933817.802001841.21.830.930.8917.809.018.681 .037 MACSJ1149 177.391245922.374169038.02.902.210.7633.9725.878.92 2.901 177.431720122.40840663.10.240.230.012.862.650.1123. 785 MACSJ1423 215.949597224.06123897.70.390.400.0034.544.680.0313 6.351 215.937455324.061291312.30.790.790.019.299.250.1753 .388 215.939592324.06203356.10.240.210.032.772.470.366.8 47 215.949948424.0778332468.76.833.513.2980.1941.1938. 691.065 215.924496924.07833969.90.260.230.043.092.700.446.0 87 215.949625824.0783947453.27.313.174.0385.8737.2747. 350.787 215.920699124.08589745.10.060.050.020.670.550.291.9 00 215.960473724.10984964.50.010.010.0020.170.150.028. 260 215.970675724.10405003.90.070.050.020.840.610.262.2 98 215.971353024.08244523.00.030.020.010.340.200.141.4 27 MACSJ2214 333.7058150-13.992715257.83.322.091.2232.2520.3611. 811.724 93

PAGE 94

Table3-7.Continued RADecSourceFullBandHardBandSoftBandFullBandHardBand SoftBandHardness SignicanceFluxFluxFluxLuminosityLuminosityLuminosi tyRatio 333.7037453-13.99878594.60.530.510.025.164.930.1827 .601 333.7527550-13.96619113.20.140.080.061.320.820.571. 448 333.7385399-14.00344197.60.190.200.041.842.00.355.6 10 MS0451 73.5532701-3.013156020.00.540.270.266.463.243.141.0 33 73.5666479-3.04263216.30.660.710.037.878.580.4120.9 23 73.5525164-3.02101335.70.610.590.037.367.060.3420.8 25 73.5441260-2.99458542.90.140.110.031.651.360.344.01 3 73.5451998-3.02365204.20.050.060.620.7373.5469876-3.018391814.80.170.150.042.061.760.473.7 17 73.5477995-3.015963910.60.480.450.075.795.430.906.0 53 MS1054 164.2608788-3.662017815.00.410.390.0513.6313.071.57 8.307 164.2446619-3.6473165147.83.862.911.00127.7696.4333 .252.900 164.2728766-3.64709933.80.110.100.0023.793.290.0744 .849 164.2700806-3.63901657.00.370.380.0112.3212.570.492 5.758 164.2510953-3.61359714.30.090.070.022.842.340.653.6 09 164.2343522-3.60992105.80.090.080.023.072.530.634.0 22 164.2305684-3.62946114.30.170.170.015.635.580.3814. 882 164.2359270-3.62261734.80.260.250.028.478.240.6812. 167 94

PAGE 95

Table3-7.Continued RADecSourceFullBandHardBandSoftBandFullBandHardBand SoftBandHardness SignicanceFluxFluxFluxLuminosityLuminosityLuminosi tyRatio SDSS1004 151.142938041.2057254258.59.076.972.02183.31140.844 0.813.451 151.154989041.208986852.90.900.400.5218.138.1810.46 0.782 151.140430941.218771210.80.390.330.057.876.651.076. 227 151.127614141.235010810.90.070.111.352.28151.126744241.23745557.10.250.250.0025.114.990.0599 .886 151.161431541.21397384.40.270.240.0085.434.910.1630 .132 95

PAGE 96

Table3-8.Rest-FrameX-rayFittingParameters ClusterRedshiftn H (10 20 cm 2 )D L (10 3 Mpc) CLJ 0152 1357 0.8311.585.2675 CLJ 1226.9+3332 0.8881.375.7207 MACSJ 0257 2325 0.5062.162.8636 MACSJ 0717+3745 0.5487.113.1546 MACSJ 0744+3297 0.6865.624.1540 MACSJ 0911+1746 0.5043.642.8499 MACSJ 1149+2223 0.5442.213.1266 MACSJ 1423+2404 0.5452.283.1336 MACSJ 2214 1359 0.5043.302.8499 MS 0451.6 0305 0.5505.033.1686 MS 1054.4 0321 0.8303.625.2596 SDSS 1004+41 0.6801.114.1092 96

PAGE 97

Table3-9.IRPowerLawGalaxies RADecChannel1Channel2Channel3Channel4 Flux( Jy)Flux( Jy)Flux( Jy)Flux( Jy) CL0152 28.1822554-13.983726384.382 0.0388.462 0.0396.722 0.2199.831 0.16-0.220 28.1657731-13.9614409316.814 0.03341.394 0.03425.256 0.21552.300 0.16-0.534 28.2023977-13.96093166.262 0.037.842 0.0312.860 0.2011.498 0.16-0.948 28.2118833-13.94450339.623 0.0310.218 0.0316.433 0.2013.709 0.16-0.429 28.2002399-13.96855994.550 0.035.340 0.034.764 0.205.670 0.16-0.463 28.1785907-13.94014081.416 0.031.910 0.033.398 0.213.143 0.16-1.241 28.1902671-13.954680015.310 0.0317.536 0.0316.802 0.2115.846 0.16-0.313 28.1497103-13.95531424.759 0.034.949 0.038.497 0.205.837 0.16-0.307 28.1889414-13.97041903.667 0.033.725 0.036.175 0.205.173 0.16-0.331 CLJ1226 186.783134633.54228957.047 0.029.017 0.0411.268 0.1615.481 0.16-1.049 186.764860133.548666313.674 0.0216.264 0.0421.723 0.1623.397 0.16-0.772 186.762204833.58104143.532 0.024.024 0.045.789 0.166.286 0.16-0.720 186.772381833.54049829.742 0.0211.192 0.049.535 0.1614.330 0.16-0.467 186.777032933.56743922.420 0.022.819 0.043.957 0.163.208 0.16-0.606 186.740677233.57418328.603 0.0210.191 0.049.839 0.1610.474 0.16-0.477 186.758261233.54934974.683 0.026.803 0.047.283 0.1616.658 0.16-1.542 186.780755333.530565211.133 0.0212.925 0.0413.438 0.1611.405 0.16-0.365 97

PAGE 98

Table3-9.Continued RADecChannel1Channel2Channel3Channel4 Flux( Jy)Flux( Jy)Flux( Jy)Flux( Jy) MACSJ0717 109.392829137.72458852.935 0.062.923 0.083.148 0.323.255 0.35-0.071 109.362999637.733721712.555 0.0612.858 0.0818.233 0.3318.016 0.35-0.366 109.355829637.75352422.964 0.063.207 0.082.244 0.333.146 0.35-0.087 109.384030337.74603769.658 0.0611.164 0.089.666 0.3322.502 0.35-0.730 109.416352737.729286021.710 0.0627.736 0.0834.780 0.3328.593 0.35-0.768 109.396534237.761489480.344 0.0695.472 0.0874.035 0.3390.152 0.35-0.359 109.393482937.764365753.588 0.0652.944 0.0838.442 0.3388.598 0.35-0.138 109.410688937.739450413.860 0.0612.896 0.0814.578 0.3224.543 0.35-0.191 109.379389237.74853356.242 0.066.358 0.0812.673 0.338.346 0.35-0.430 109.390812837.76944485.769 0.064.976 0.085.664 0.337.951 0.350.110 109.373837437.72526181.882 0.062.846 0.089.206 0.3310.313 0.35-2.232 109.381026437.76761571.138 0.061.625 0.084.445 0.322.915 0.35-1.636 109.385583237.76829475.787 0.064.886 0.088.234 0.336.579 0.350.130 109.373282737.77239071.101 0.061.459 0.081.939 0.331.111 0.35-0.723 109.381854337.76836551.468 0.062.047 0.081.703 0.338.049 0.35-1.596 109.356755837.75438113.573 0.063.656 0.086.319 0.3210.675 0.35-0.828 109.358054637.76579234.023 0.064.209 0.089.123 0.3311.918 0.35-0.948 109.408650337.77127523.446 0.064.218 0.0812.289 0.336.895 0.35-1.183 98

PAGE 99

Table3-9.Continued RADecChannel1Channel2Channel3Channel4 Flux( Jy)Flux( Jy)Flux( Jy)Flux( Jy) MACSJ1423 215.951757924.09738085.609 0.055.880 0.065.963 0.306.611 0.31-0.200 215.939796224.057549013.764 0.0515.769 0.0619.672 0.3019.487 0.31-0.566 215.935475524.07796167.590 0.057.427 0.067.531 0.308.625 0.31-0.017 215.945847124.091178915.937 0.0517.022 0.0615.636 0.3017.905 0.31-0.192 215.926847724.097565015.264 0.0515.830 0.0615.460 0.3012.725 0.310.012 215.968816124.084106629.734 0.0554.470 0.0684.057 0.30183.671 0.31-2.498 215.934218924.06831761.626 0.052.238 0.062.127 0.304.656 0.31-1.281 215.968378724.08525785.594 0.056.819 0.0610.164 0.307.092 0.31-0.711 215.950404224.082015314.820 0.0515.812 0.0610.566 0.3013.276 0.310.019 215.972940424.09576111.448 0.052.457 0.062.809 0.307.077 0.31-2.107 215.926283724.09276058.800 0.058.833 0.069.290 0.3018.253 0.31-0.392 215.977509024.07391971.611 0.051.586 0.061.110 0.302.476 0.30-0.074 215.961773924.104893415.508 0.0517.456 0.0612.332 0.3012.817 0.30-0.085 215.959151524.10981089.437 0.0512.721 0.0626.109 0.3017.275 0.31-1.232 215.946031124.081627320.359 0.0524.766 0.0618.337 0.3022.185 0.30-0.425 215.932122124.089928124.573 0.0525.250 0.0613.636 0.3026.745 0.310.073 215.973238324.098372426.372 0.0529.005 0.0639.993 0.3016.673 0.30-0.090 MS0451 73.5525335-3.015106921.817 0.0222.596 0.0322.395 0.1231.604 0.13-0.288 99

PAGE 100

Table3-9.Continued RADecChannel1Channel2Channel3Channel4 Flux( Jy)Flux( Jy)Flux( Jy)Flux( Jy) 73.5266952-3.03652441.198 0.021.983 0.032.600 0.125.976 0.13-2.053 MS1054 164.2609994-3.6619864125.759 0.02163.272 0.04191.572 0.15327.143 0.17-1.148 164.2343897-3.609961311.703 0.0214.129 0.0422.537 0.1525.869 0.17-1.001 164.2447098-3.647323139.975 0.0257.745 0.0466.941 0.15120.088 0.17-1.468 164.2591409-3.59595432.566 0.022.951 0.042.441 0.153.777 0.17-0.457 164.2702649-3.65056581.533 0.022.035 0.043.923 0.156.180 0.17-1.590 164.2772592-3.61907671.692 0.022.392 0.045.149 0.158.099 0.17-1.862 164.2814489-3.638149711.625 0.0215.880 0.0416.480 0.1522.703 0.17-1.083 SDSS1004 151.147012241.19223623.002 0.013.511 0.023.766 0.074.691 0.07-0.612 151.138388841.215474029.172 0.0133.091 0.0235.597 0.0738.805 0.07-0.456 151.159902141.22611948.808 0.019.754 0.0210.441 0.0715.124 0.07-0.545 100

PAGE 101

Table3-10.AGNOverlap Cluster#Variables#IRPowerLaw#IRPowerLaw#X-rayPS#X-r ayPSVar = IRX-ray = IRVar = X-ray TotalTotalw = opt.counterpartsTotalw = opt.counterparts CL 0152 1296431 a 2 a 2 a CLJ 1226 208666010 MACSJ 0257 13--11--0 MACSJ 0717 191813971 a 1 a 3 a MACSJ 0744 18--126--2 MACSJ 0911 14--54--1 MACSJ 1149 13--22--1 MACSJ 1423 241715109002 MACSJ 2214 8--13--0 MS 0451 16217-0-MS 1054 117483030 SDSS 1004 113366001 a Oneobjectdetectedbyall3techniques. 101

PAGE 102

Table3-11.GalaxiesintheLacyWedge ClusterGalaxiesIRPowerLawOpticalVariablesX-rayPoint Sources CL 0152 65/2679/92/54/4 CLJ 1226 90/3328/83/162/6 MACSJ 0717 49/28815/182/82/5 MACSJ 1423 55/24713/172/121/8 MS 0451 42/2932/20/82/6 MS 1054 54/3197/70/64/6 SDSS 1004 56/2513/31/93/6 102

PAGE 103

CHAPTER4 THEPERCENTAGEOFAGNINGALAXYCLUSTERS 4.1ClusterMembership Aprimarygoalofthisthesisistodeterminehowtheclustere nvironmentaffectsthe onsetandmaintenanceofaccretionontoagalaxy'ssupermas siveblackholesuchthat itisobservedasanAGNviaeitherX-ray,mid-IR,oropticalv ariabilityselection.Inthis chapterwediscussthemethodsbywhichweestimateclusterm embershipamongthe galaxiesineachclustereldandcomparethepercentageofc lusterAGNwiththeeld, aswellasexaminetheeffectofclusterphysicalproperties onthenumberofAGNwe observe.4.1.1SpectroscopicCatalogs InordertoaccuratelydeterminethepercentageofAGNamong clustergalaxies, informationabouttheeldandclusterpopulationsreprese ntedineachACSeldof viewisrequired.Spectroscopiccatalogsofvaryingcomple tenessareavailablefromthe literaturefortheclustersinoursample.Clustermembersw eredeterminedusingthe publishedredshiftrangeofeachcluster.InthecaseoftheM ACSclusters,thisrange wascalculatedusingthemeasuredvelocitydispersionofth ecluster( Barrett 2006 ). Table 4-1 liststhenumberofsourceswithpublishedredshiftsthatfa llwithintheACS FOV,aswellastheredshiftrangespannedbytheclustermemb ers. Thesecatalogswerematchedwithourphotometriccatalogs( seeSection 3.2.2 )to identifyobjectsinthisstudywithspectroscopicredshift s.Thecatalogswerematched usingtheIDLroutineSPHEREMATCHwitha2 00 (67pixels)maximummatchradius.In somecasesalinearoffsetbetweentheWCSoftheACSimagesan dthespectroscopic catalogwasidentiedandcorrectedusingalinearshifttoa lignthetwocatalogsbefore searchingformatches.Allresultswerevisuallyinspected toremoveanyspurious matches.Table 4-2 liststheAGNcandidatesineachACSimagehavingameasured 103

PAGE 104

spectroscopicredshift,conrmingwhetheritisinoroutof theclusterredshiftrange.A dashindicatesthatdataisnotavailableforthiscluster.4.1.2FieldContamination Eachofourclustereldscontainsacombinationofthegalax yclusterpopulation togetherwithaforegroundandbackgroundeldgalaxypopul ation.Inorderto supplementtheexistingspectroscopicclustermembership informationgivenin Table 4-1 andbetterestimatecontaminationfromtheeldgalaxypopu lation,we estimatethenumberdensityoftheeldpopulationacrossea chofourclusters.Todo this,weobtainedarchivalHSTACSimagesoftheGOODS-North andGOODS-South eldstakenintheF775Wlter.ThreeACSimagesnearthecent eroftheGOODS-N andGOODS-Seldswereselectedfromthearchivewithexposu retimessimilartothe depthoftheclusterdata:Field1(totalexposuretime5000s ),Field2(totalexposure time7028s),andField3(totalexposuretime8350s).Thedat awerereducedusing MULTIDRIZZLEandthesametechniqueasusedfortheclusterA CSimages,outlinedin Section 3.2.1 .Imagesweredrizzledtothesameresolution(0 00 .03/pix)andweightmaps werealsocreated. SourceextractionandphotometryusingaKronexibleellip ticalapertureofall galaxiesineachtilewasperformedwithSExtractor.Zeropo intswereobtainedfrom theheaderkeywordsusingEquation 3–1 .Theresultinggalaxyphotometrycatalogfor theseeldswasusedtodeterminethemeannumberofgalaxies /arcmin 2 thatshouldbe observedatthedepthoftheACSimagesinourclustersurvey: Field1(5000s):15.6galaxies/arcmin 2 Field2(7028s):16.6galaxies/arcmin 2 Field3(8350s):18.4galaxies/arcmin 2 Foreachcluster,wenowhaveanestimateofthedensityofthe eldgalaxy populationbyusingthevaluedeterminedfromtheGOODSeld imagesofsimilar 104

PAGE 105

exposuretimedepth.Weusethistoestimateclustermembers hipprobabilitiesfor sourceswithoutspectroscopicinformationasdescribedin thenextsection. 4.1.3ClusterRadialProlesandMembershipProbability Assumingthatthecenterofeachcluster(denedastheBrigh testClusterGalaxy) containsthehighestconcentrationofclustergalaxieswit hadecreasingcontribution fromthegalaxyclusterandincreasingcontributionfromth eeldwithincreasing distancefromthecenter,wecalculatethenumberofgalaxie s/arcmin 2 vs.radiusin binsofvaryingsizefrom0.3–0.01arcminstartingfromthec enterofeachcluster.We thenassumeateachradiusaconstantdensityfortheeldpop ulationasdeterminedin Section 4.1.2 .Bysubtractingtheelddensity,aeld-decontaminatedde nsityproleof theclustercanbeproduced.Anexampleofthegalaxydensity asafunctionofcluster radiusisshowninthetoppanelofFigure 4-1 forCLJ1226.Theaveragevalueofthe galaxydensityoftheeldisalsoshown,aswellastheeld-s ubtractedradialproleof thecluster. Atanygivenradius,thetotalnumberofgalaxies( T )isequaltothenumberof galaxiesintheeld( F )plusthenumberofgalaxiesinthecluster( C ).Therefore, T = F + C andbysubtractingtheeldfromthetotal,weestimatedthen umberof galaxiesinthecluster( C )atanygivenradius.Fromthis,wecalculatethepercentage of galaxiesateachradiusintheclusterandtheeld. % galaxiesintheeld = F = T (4–1) % galaxiesinthecluster = C = T (4–2) Theresultingvaluesweretwithapolynomialtomodelthera dialprobabilityprole ofeachcluster.Wethendeterminetheprobabilityfrom0(no tinthecluster)to1(in thecluster)foreachgalaxyinourimagebasedonitsdistanc efromthecenterofthe cluster.AnexampleisshowninthebottompanelofFigure 4-1 .Basedontheseradial probabilityproles,eachgalaxyisassignedaweightequal toitsclustermembership 105

PAGE 106

likelihood.Galaxieswithspectroscopicdataareassigned aweightof1or0basedon whethertheirspectroscopicredshiftlieswithintherange ofclusterredshiftvaluesgiven inTable 4-1 4.1.4ColorSelectionandClusterMembershipProbability Inadditiontousingradialdistanceintheclustertodeterm ineprobabilityofcluster membership,wealsousegalaxycolortoincreaseclustermem bershipprobability forgalaxieswithcolorsmatchingthoseofknownclustermem bersatthecluster redshiftintheclusterimages.Thegalaxyclustersampleca nberoughlydividedinto threeredshiftbinsatz=0.5,0.7,and0.85.Foreachoftheth reeredshiftgroups, weexaminedtheV-Icolorsfortheclusterwiththemostspect roscopically-conrmed clustermembers(MACSJ0717atz 0.5,MACSJ0744atz 0.7,andMS1054atz 0.85)andcomparedthepeakoftheclustermembergalaxycolo rdistributionwiththe expectedgalaxycolorsforearlytypegalaxies(E/S0)atthe seredshiftsin Fukugitaetal. ( 1995 ).TheobservedpeakinV-I(F555W-F814W)colorsofclusterm embersgalaxiesin MACSJ0717is2.54,whichverycloselymatchestheexpectedg alaxycolorofV-I=2.48 givenin Fukugitaetal. ( 1995 )forearlytypegalaxies. InthecaseofMS1054,theV-IcolorwehaveavailableisF606W -F775W,which cannotbedirectlycomparedwiththeworkof Fukugitaetal. ( 1995 ),inwhichcolors areprovidedindifferentlters.Wethereforeusethepubli shedvaluesof Tranetal. ( 2007 ),whondameancolorofF606W-F775W=1.61forred,brightel lipticalgalaxies inMS1054.Thisisconsistentwiththevalueof1.69thatweme asureforthecluster members. Fukugitaetal. ( 1995 )doesnotprovidegalaxyV-Icolorsataredshiftof0.7for comparisonwithourclustersatthatredshift.However,ali nearextrapolationwastto theexpectedV-Igalaxycolorsforellipticalgalaxiesbetw eenaredshiftof0.5and0.8. Thevalueof2.79weobtainfromthistisconsistentwiththe observedpeakvalueof 2.66forclustermembersinMACSJ0744. 106

PAGE 107

Figure 4-2 showstheV-IgalaxycolordistributionsforMACSJ0717,MAC SJ0744, andMS1054forspectroscopically-conrmedclustermember s(solidline),non-cluster members(dottedline),andallgalaxiesintheclusterimage FOV(dashedline).Based onthecolordistributionofclustermembers,weestimateth atallgalaxieswithcolors redderthantheblueendofthisdistributionhaveahighprob abilityofresidingin thecluster.WedeterminetheV-IcolorthresholdtobeF555W -F814W=2.36for clustersatredshiftz 0.5,F555W-F814W=2.43forclustersatredshiftz 0.7,and F606W-F775W=1.49forclustersatredshiftz 0.85basedonthecolordistribution forclustermembersinthesethreerepresentativeclusters .Inordertodeterminethe probabilitythatshouldbeassignedtogalaxiesredderthan thislimit,weexaminedthe radialprolesofclustermembersinallclustersanddeterm inedtheaveragevalueof theirradially-determinedclustermembershipprobabilit y.Thiswasgenerallyfoundto be 80%forallclusters.Wethereforeincreasetheclustermemb ershipprobabilityto 80%forgalaxiesredderthanthecolorthresholdsdetermine dhere.Ifaredgalaxyhad alreadybeenassignedaclustermembershipprobabilityhig herthan80%basedonits radiuswithinthecluster,itretainedthehigherprobabili tyvalue.Figure 4-3 showsan exampleofgalaxyclustermembershipprobabilityvaluesfo rtheclusterMACSJ0717. Becausethepopulationofconrmedclustermembersinthese clustersconsistof mainlyred,earlytypegalaxies,wecouldnotdetermineacol or-basedprobabilityforblue galaxies.Thus,whilebluergalaxiesdidnotreceivedecrea sedmembershipprobability values,theirprobabilitiesarebasedsolelyontheirradia ldistancefromthecenterofthe clusterorspectroscopicinformationwhenavailable,asst atisticalinformationaboutthe bluegalaxypopulationisnoteasilydetermined. 4.2PercentageofClusterAGN TodeterminethepercentageofAGNinclusters,wedividethe totalnumberof AGNineachclusterbythetotalnumberofgalaxiesintheclus ter.Wedothisinseveral 107

PAGE 108

waysbasedupontheinformationwehave.First,wecomputeth epercentageforonly spectroscopically-conrmedclustermembersonly: % AGN spec = # spectroscopicallyconrmedclusterAGN # spectroscopicallyconrmedclustergalaxies (4–3) Sincespectroscopiccoveragevariesfromclustertocluste randissometimessparse, wealsocalculateanAGNpercentageusingallsourcesintheA CSFOVforeach cluster,whichcoversroughlyhalfofthevirialradius.Wec allthisthenon-weighted AGNpercentage.Finally,wecomputeourweightedAGNpercen tageusingthecluster membershipprobabilitiesdescribedinSections 4.1.3 and 4.1.4 basedonradialdistance andcolorinformation: % AGN weighted = P ( AGNclustermembershipprobabilities ) P ( galaxyclustermembershipprobabilities ) (4–4) Inaddition,wealsocomputeaclusterAGNpercentageusings ourceswhichhavea 50%orgreaterclustermembershipprobability. Table 4-3 liststheweightedandnon-weightedtotalpercentageofAGN inourgalaxy clustersaswellasthepercentageofAGNdetectedamongspec troscopically-conrmed clustermembers.Wendthatourclustershavearangeof0.8– 3.75%AGN(weighted), withamedianvalueof2.27 1.5%.Ifweconsiderconrmedclustermembersonly,we ndthatforthe11clusterswithredshiftinformation,25/5 30or4.7%ofconrmedcluster membersshowevidenceofnuclearactivity.Table 4-4 liststhepercentageofcluster AGNdetectedusingeachofthe3techniques.Inbothtablesad ash(-)indicatesthat dataisnotavailable,andanxindicatestherearenoAGNdete cted,thoughdataexist. 4.2.1ComparisonwithFieldGalaxiesinGOODS Wecompareourresultsforthe12clusterswithasimilarmult iwavelengthapproach toAGNidenticationcarriedoutin Sarajedinietal. ( 2011 ).Thisstudyidentied85 variablegalaxiesfromatotalof4174galaxiesobservedint heGOODS-NorthandSouth eldsusingveepochsofHSTACSimaginginF606W(V-band)ov erthecourseof 108

PAGE 109

6months.Thevariability-selectedAGNarecombinedwith25 9X-raysourcesfrom the2MsChandraDeepFieldsurveysand22IRpowerlawsources identiedinthese elds.Tocomparethiseldstudywithourclusterresults,w eimposeseverallimits tothedifferentsurveysamples.First,weconsideronlythe sixty-eightX-raysources inGOODSdetectabledowntoalimitof 2x10 15 erg/cm 2 /s,similartotheaverage depthreachedintheX-rayobservationsofourclustersampl e.Wefurtherlimittheeld sampletotheredshiftrangeofourclusters(z=0.4–0.9).Wi ththeserestrictions,we nd31/1235galaxiesintheGOODSeldshostAGN(2.5 1.6 %).Thisiscomparable to,thoughslightlyhigher,thanthemedianweightedvaluew endof2.27 1.5%in ourclusters.Ifwefurtherlimitourclustersampletoonlyt he7clusterswhereall3AGN selectioncriteriaarepossible(i.e.,thosewithmid-IRob servations),theAGNpercentage is2.5 1.6%,exactlyequaltotheeldAGNpercentage. Itisalsonecessarytoensurethatthegalaxymagnitudedist ributionsofthe GOODSeldsandourclustergalaxiesarecomparable.Sincet heGOODSsurvey individualepochimagesareshallowerthanourclusterimag es(1000s,comparedwith 2100–3000s),thevariabilityanalysisintheGOODSeldsex tendsonlyto 24.5,while ourclusterphotometryextendstoI 26.Wethereforeimposeagalaxymagnitudelimit ofI=24.5onallobjectsinourclustersampletofurthermatc htheGOODSsample.This resultsinamedianweightedAGNpercentageamongallourclu stersof4.01%,which risesto4.94 2.2%ifweconsideronlythoseclusterswithIRobservations (andthusall threeAGNselectiontechniquesarepossible).Ifweconside ronlythenumberofknown clustermembersinthesesevenclusters,theAGNpercentage is5.5 2.3%(20/363). ThesendingsindicatethatthenumberofAGNamonggalaxies inclusters( 5%)is greaterthanthatintheeld( 2.5%),thoughonlyat 1 signicance. 4.2.2ComparisonwithClusterX-raySurveysforAGN Martinietal. ( 2002 )foundalowerlimitof 5%AGNintheclusterA2104(z=0.154) withopticalcounterpartsdowntoanabsolutemagnitudeofR < 20.Thiscorresponds 109

PAGE 110

tosixX-ray-detectedAGN,onlyoneofwhichshowsopticalem issionlinesinitsspectra indicativeofAGNactivity.Thisisconsistentwithemissio nlinesurveyssuchas Dressler etal. ( 1999 ),andsuggeststhatthefractionofAGNinclustersmaybehig herthan previouslydeterminedifadditionalAGNdetectiontechniq uesareemployed. Martinietal. ( 2009 )conductedasurveyoftheluminousAGNpopulationincluste rs outtoz=1.3.Theirsampleincludestwoofourclusters,MS04 51andMS1054.They requiredthattheirsources1)musthaveahardX-rayluminos ityL X H 10 43 erg/s,2)the redshiftmustbewithin3timesthecluster'svelocitydispe rsionoftheclusterredshift,3) thesourcemustliewithintheprojectedvirialradiusofthe cluster,and4)theabsolute magnitudeofthehostgalaxymustbeM R =M (z)+1.TheydetectnoX-rayAGNin MS0451meetingthiscriteria,andwhenthesecriteriaareap pliedtoourX-raysources inthatcluster,wendthesameresult. Martinietal. ( 2009 )detectoneX-raypoint sourcemeetingtheirrequirementsintheclusterMS1054,bu titfallsoutsideofourACS FOVandthereforewasnotincludedinoursurvey.Wealsodono tdetectanyotherX-ray sourceswithintheACSFOVthatmeetthesecriteria. WealsocompareourresultswiththeX-raysourcecatalogof Johnsonetal. ( 2003 ), inwhichtheauthorsidentify47X-raypointsourcesinMS105 4downtouxessimilar tothoseachievedinourdata.Theyestimatea 2 excessofX-raypointsourcesin thisparticularclusterdowntoanX-rayuxof5x10 15 erg/s/cm 2 ,whichisconsistent withanexcessof 6AGNsrelativetotheeld.Theirsurveyareais8.3x8.3arcm in 2 afactorof2.5timeslargerthantheareausedinourvariabil itysurveyperclustereld. Sevenofthesourcesof Johnsonetal. ( 2003 )fallwithintheACSFOV,andwedetect sixoftheseobjectsinoursurvey,consistentwiththeirnd ingsthatthisclusterhasan excessofAGNwhencomparedwiththeeld.4.2.3Summary:AGNPercentage Comparingourresultswiththesimilarmultiwavelengthsur veyof Sarajedinietal. ( 2011 ),wendthatthepercentageofAGNisenhancedinclustersre lativetotheeld. 110

PAGE 111

WealsondthatthenumberofX-ray-detectedAGNinoursurve yisconsistentwith othersurveysforX-raypointsourcesingalaxyclusters,wh ichreportanenhancement ofX-raypointsourcesingalaxyclustersrelativetotheel d.Earlierindicationsthat5% ofgalaxiesinclustersmaybeAGNbasedonX-raypointsource s(e.g., Martinietal. 2002 )areconsistentwithourndings,whichalsoincludevariab ility-andIR-detected AGN.However,theactualfractionappearstobeheavilydepe ndentongalaxysurvey magnitudelimitsandX-rayuxlimits. ThisenhancementinthepercentageofAGNinclustersoverth eeldisevidence thatgalaxiesarestillabletofuelaccretionontotheirsup ermassiveblackholes,even indenserenvironments. Martinietal. ( 2004 )pointsoutthatwhilemajormergersmay beperhapstheonlyreasonablecandidatetotriggerandsust ainluminousAGN,lower luminosityAGNsuchasthoseidentiedbyoursurveymayhave signicantlymore physicalprocessescapableoffuelingacentralsupermassi veblackhole,includingbar structures,minormergers,galaxyharassment,andstellar massloss–allofwhichstill playasignicantroleinagalaxyclusterenvironment.Ifth isisinfactthecase,there maybeacomparablenumberoflowerluminosityAGNincluster sandintheeld,which isconsistentwithourresults,asopticalvariabilityandI Rpowerlawdetectionarelikely topickoutlowerluminosityAGN.Thisisalsoconsistentwit hthepicturethatmore luminousAGNmaybelesscommonindenserenvironments(e.g. Kauffmannetal. 2004 ; Popesso&Biviano 2006 ),aswendnoluminousX-raysourcesinMS0451or MS1054,consistentwiththesurveyof Martinietal. ( 2009 ). 4.3ClusterProperties Wealsolookforcorrelationsbetweenclusterpropertiessu chasredshift,mass, luminosity,velocitydispersion,andvirialradiuswithth epercentageofAGNfound amongclustermembers.WeexamineboththeweightedAGNperc entage,withthe clustersdividedintotwogroups(thosewithIRobservation sandthosewithout),aswell 111

PAGE 112

astheeffectsofclusterpropertiesonthetypeofAGNdetect ed(opticalvariables,X-ray pointsources,IRpowerlawsources). Therstclusterpropertyweexamineisredshift.Studiessu chas Galametz etal. ( 2009 )and Martinietal. ( 2009 )searchforevolutioninthenumberandradial distributionofAGNinclustersasafunctionofredshift.Bo thofthesestudiesnd evidenceforanincreaseinthenumberofAGNwithredshift.F igure 4-4 showsthe totalweightedpercentofAGNasafunctionofclusterredshi ftforclusterswithand withoutIRobservations,aswellastheweightedpercentofA GNbydetectiontypeasa functionofredshift.Itcanbeseenthatovertheredshiftra nge0.5–0.9,thepercentage ofAGNwedetectisfairlyconstant.Thepercentagesofvaria bles,X-raypointsources, andIRpowerlaw-likesourcesallremainroughlyconstantre gardlessoftheredshiftof thecluster.Theseresultsindicatenotonlythatthereappe arstobenochangeinthe numberofAGNdetectedwithinourredshiftrange,butthatwe areabletoreachsimilar levelsofcompletenessindetectingAGNacrosstheredshift rangeofourclustersample. Theseresultsarenotinconsistentwiththendingsofprevi ousstudies,astheredshift rangecoveredbyourclustersismuchsmallerthantherangeo fthesestudies.Infact, ourclusterredshiftrangefrom 0.5to 0.9iswithinasingleredshiftbinintheprevious studies( Galametzetal. 2009 ; Martinietal. 2009 ). NextweconsidertheX-rayluminosityofourclustersample. TheX-rayluminosity measuredfortheseclusterscomesfromthermalbremsstrahl ungemissionfromhot (T 10 8 K)intraclustergasboundtotheclusterpotentialandisthe reforerelatedto theclustermass.AsshowninFigure 4-5 ,thereappearstobenocorrelationbetween thetotalweightedpercentageofAGNandthecluster'sX-ray luminosity.However, ourclustersamplecoversarelativelysmallrangeofX-rayl uminosity,lessthanan orderofmagnitude.Therealsoappearstobenosignicantco rrelationbetweenthe percentageofopticalvariables,IRpowerlawsources,orXraypointsourceswiththe X-rayluminosityofthecluster. 112

PAGE 113

Figure 4-6 showstheweightedpercentageofAGNdetectedasafunctiono fcluster mass(M 200 ).Thereappearstobenocorrelationbetweenthetotalperce ntageofAGN detectedandtheclustermass,andthepercentageofoptical variables,X-raypoint sources,andIRpowerlaw-likesourcesallremainconstanto vertherangeofcluster massesweexamineinourstudy.Thisisconsistentwiththere sultsweobservedwhen consideringclusterX-rayluminosity,whichisalsoameasu reofclustermass.Thus,the fractionofAGNamonggalaxiesinclustersdoesnotappearto dependonmassand/or hotgascontentoveranorderofmagnitudeinmass. InFigure 4-7 welookforcorrelationsbetweenthepercentageofAGNdetec ted andtheclustervirialradius(orsize).Thoughourclusters coveralargerangeofvirial radii,from 1.1–2.6Mpc,thetotalpercentageofAGNdetectedintheclus tersremains roughlyconstantoverthisrange.Thisisalsothecaseforth eopticalvariablesandthe X-raypointsources,thoughtheIRpowerlawsourcesshowwha tmightbeaveryslight declineinthepercentageofIRsourcesdetectedasafunctio nofclustervirialradius. Finally,inFigure 4-8 ,welookatthepercentageofAGNdetectedversusthecluster velocitydispersion.Thereappearstobenosignicantcorr elationbetweenthetotal percentageofAGNdetectedandtheclustervelocitydispers ion,evenoverourrangeof 600–1800km/s.ThisisalsotrueforopticalvariablesandXraypointsources,though theIRpowerlawsourcesshowaslightincreaseinthepercent agedetectedasthe velocitydispersionoftheclusterincreases.4.3.1ClusterMorphology Ebelingetal. ( 2007 )assignseachoftheMACSclustersinoursamplea morphologycode,dependentontheagreementbetweentheX-r ayandopticalemission inthecluster,aswellassignsofdisturbancesinthecluste rsubstructure.Their morphologycodesareassignedbasedonthefollowingcriter ia: 113

PAGE 114

1:Relaxed(pronouncedcoolcore,perfectalignmentofX-ra ypeakandsinglecD galaxy) 2:Semi-Relaxed(goodoptical/X-rayalignment,concentri ccontours) 3:Semi-Disturbed(nonconcentriccontours,obvioussmall -scalesubstructure) 4:Disturbed(pooroptical/X-rayalignment,multiplepeak s,nocDgalaxy) Basedonthesedenitions,theygavethegalaxyclusterstha tappearinoursamplethe followingmorphologyclassications: MACSJ0257:2(Semi-Relaxed)MACSJ0717:4(Disturbed)MACSJ0744:2(Semi-Relaxed)MACSJ0911:4(Disturbed)MACSJ1149:4(Disturbed)MACSJ1423:1(Relaxed)MACSJ2214:2(Semi-Relaxed)Usingtheclassicationcriteriaof Ebelingetal. ( 2007 ),weattempttoassigna morphologycodetoeachofourremainingclusters: CL0152:4(Disturbed)CLJ1226:1(Relaxed)MS0451:1(Relaxed)MS1054:3(Semi-Disturbed)SDSS1004:1(Relaxed) Astherearesuchasmallnumberofclustersineachofthesemo rphologicalgroups, wehavecombinedtherelaxedandsemi-relaxedclusters(7cl usters),andalsothe disturbedandsemi-disturbedclusters(5clusters).Acomp arisonofthemedianvalueof 114

PAGE 115

theweightedpercentageofAGNshowsthatbothgroupshaveve rysimilarpercentages ofAGN:therelaxedclustershaveamedianvalueof2.13%andt hedisturbedclusters haveamedianvalueof2.27%.IfweexcludeclusterswithnoIR observations,wehave 3clustersinthedisturbedgroupand4clustersintherelaxe dgroup.Wendthatthe disturbedclustershaveamedianpercentageof2.29%cluste rAGN,whiletherelaxed clustershaveamedianAGNpercentageof3.13%.Thus,wedono tndevidencethat theclusterdynamicalstatehasanimpactonthenumberofAGN detectedinoursample ofclusters.4.3.2Summary:ClusterPropertiesandAGNPercentage Insummary,wendnoobvioustrendsbetweenthepercentageo fclusterAGN detectedandclusterproperties,includingmass,X-raylum inosity,virialradius,velocity dispersion,andredshift.Wedonotethatmanyoftheseprope rtiesareinterrelated,and ingeneraltheclustersinoursamplecoverasmallrangeofma ssandX-rayluminosity asafunctionoftheirdiscoverycriteria.Wendthatthenum berofAGNwedetectis roughlyconstantregardlessofredshift,conrmingthatwe areabletoreachsimilar levelsofcompletenessovertherangeofclusterredshiftsi noursample.Finally,wedo notobserveasignicantlinkbetweenclustermorphologyan dthenumberofAGNwe detect,ndingsimilarpercentagesofclusterAGNinbothdi sturbedandrelaxedclusters. 115

PAGE 116

Figure4-1. TopPanel :RadialplotofCLJ1226showinggalaxies/arcmin 2 vs.radial distancefromthecenterinarcminutes.Thesolidhistogram isthetotal numberofobjectscountedineachradialbin,andthedottedh istogramisthe eld-subtractedradialproleofthecluster.Theaveragen umberof galaxies/arcmin 2 isshownasthehorizontaldashedline. BottomPanel : RadialplotofCLJ1226showingtheprobabilitythatagalaxy atthisradius residesinthecluster(plussymbols)andtheeld(asterisk s)asafunctionof radialdistancefromthecenterinarcminutes.Thedashedan ddottedlines arepolynomialtstotheclusterandeldprobabilities. 116

PAGE 117

Figure4-2.V-IcolordistributionsofgalaxiesinMACSJ071 7(topleft),MACSJ0744(top right),andMS1054(bottomleft).Thesolidhistogramshows the spectroscopically-conrmedclustermembersandthedotte dhistogram showsspectroscopically-conrmednon-clustermembers.T hesolidand dottedhistogramshavebeenscaledbyafactorof5–10forvis ualization purposes.TheverticaldashedlineindicatestheV-Icolorl imitforgalaxiesin thecluster.Galaxiesredderthanthislimithaveincreased probabilityof residinginthecluster. 117

PAGE 118

nr rn nnrn nrr Figure4-3.RadialproleoftheclusterMACSJ0717,showing clustermembership probabilityvs.distancefromthecenteroftheclusterwith cluster membershipprobabilitycategoriesindicated. 118

PAGE 119

Figure4-4.AGN%asafunctionofclusterredshift.Thetoptw opanelsshowthetotal percentageofAGNdetected;theleftpanelshowsclusterswi thIR observationsandtherightpanelshowsthosewithout.Thela sttworows showthepercentageofIRpowerlawsources,opticalvariabl es,andX-ray pointsourcesineachcluster. 119

PAGE 120

Figure4-5.WeightedPercentageofAGNdetectedasafunctio nofclusterX-ray Luminosity.ThetoptwopanelsshowthetotalpercentageofA GNdetected; theleftpanelshowsclusterswithIRobservationsandtheri ghtpanelshows thosewithout.ThelasttworowsshowthepercentageofIRpow erlaw sources,opticalvariables,andX-raypointsourcesineach cluster. 120

PAGE 121

Figure4-6.WeightedPercentageofAGNdetectedasafunctio nofclustermass.The toptwopanelsshowthetotalpercentageofAGNdetected;the leftpanel showsclusterswithIRobservationsandtherightpanelshow sthosewithout. ThelasttworowsshowthepercentageofIRpowerlawsources, optical variables,andX-raypointsourcesineachcluster. 121

PAGE 122

Figure4-7.WeightedPercentageofAGNdetectedasafunctio nofclustervirialradius (Mpc).ThetoptwopanelsshowthetotalpercentageofAGNdet ected;the leftpanelshowsclusterswithIRobservationsandtheright panelshows thosewithout.ThelasttworowsshowthepercentageofIRpow erlaw sources,opticalvariables,andX-raypointsourcesineach cluster. 122

PAGE 123

Figure4-8.WeightedPercentageofAGNdetectedasafunctio nofclustervelocity dispersion(km/s).Thetoptwopanelsshowthetotalpercent ageofAGN detected;theleftpanelshowsclusterswithIRobservation sandtheright panelshowsthosewithout.Thelasttworowsshowthepercent ageofIR powerlawsources,opticalvariables,andX-raypointsourc esineachcluster. 123

PAGE 124

Table4-1.ClusterSpectroscopicCoverage Cluster#Redshifts#Members#FieldRedshiftClusterRedsh iftRange CL 0152 a 8148330.8310.8201-0.8672 CLJ 1226 b 4225170.8880.8793-0.9087 MACSJ 0257 c 161330.5060.4974-0.5146 MACSJ 0717 c 125108170.5480.5265-0.5635 MACSJ 0744 c 343130.6860.6846-0.7090 MACSJ 0911 c 242130.5040.4926-0.5174 MACSJ 1149 c 444220.5440.5231-0.5649 MACSJ 1423 c 453690.5450.5270-0.5590 MACSJ 2214 c 555230.5040.4857-0.5183 MS 0451 d 12768590.550.52-0.56 MS 1054 e 206108980.830.80-0.86 SDSS 1004 33-0.68a Demarcoetal. ( 2005 ) b Ellisetal. ( 2006 ) c Barrett ( 2006 ) d Moranetal. ( 2007 ) e Tranetal. ( 2007 ),privatecommunication 124

PAGE 125

Table4-2.AGNRedshifts ClusterVarVarIRIRX-rayX-ray ClusterFieldClusterFieldClusterField CL 0152 112021 CLJ 1226 100010 MACSJ 0257 00--10 MACSJ 0717 110022 MACSJ 0744 10--10 MACSJ 0911 01--00 MACSJ 1149 10--00 MACSJ 1423 520022 MACSJ 2214 00--10 MS 0451 3210-MS 1054 221110 SDSS 1004 -----Table4-3.PercentageofClusterAGN Cluster%AGN weighted %AGN non weighted %AGN spec %AGN 50% CL 0152 2.49(21/889)2.36(21/889)4.17(2/48)2.50(18/719) CLJ 1226 3.13(33/961)3.43(33/961)8.0(2/25)2.94(25/850) MACSJ 0257 1.0(14/1535)0.91(14/1535)7.7(1/13)0.99(14/1418) MACSJ 0717 2.27(42/1813)2.32(42/1813)2.78(3/108)2.33(40/1715) MACSJ 0744 2.13(28/1430)1.96(28/1430)3.23(1/31)2.12(28/1321) MACSJ 0911 1.35(18/1353)1.33(18/1353)x1.35(17/1259) MACSJ 1149 0.92(13/1512)0.86(13/1512)2.38(1/42)0.93(13/1399) MACSJ 1423 3.75(49/1337)3.66(49/1337)20.0(7/35)3.71(45/1213) MACSJ 2214 0.81(12/1373)0.87(12/1373)1.92(1/52)0.86(11/1284) MS 0451 2.65(25/1102)2.27(25/1102)5.88(4/68)2.57(23/895) MS 1054 2.29(23/893)2.58(23/893)2.78(3/108)2.17(16/737) SDSS 1004 1.43(20/1651)1.21(20/1651)-1.38(20/1446) Mean2.01 1.41.98 1.45.88 2.31.99 1.4 Median2.27 1.52.27 1.54.17 2.02.17 1.5 125

PAGE 126

Table4-4.PercentageofClusterAGNByType Cluster%AGN weighted %AGN 50% %AGN spec VariablesCL 0152 1.37(12/889)1.39(10/719)2.08(1/48) CLJ 1226 2.05(20/961)2.0(17/850)4.0(1/25) MACSJ 0257 0.92(13/1535)0.92(13/1418)MACSJ 0717 1.04(19/1813)1.05(18/1715)0.93(1/108) MACSJ 0744 1.38(18/1430)1.36(18/1321)3.23(1/31) MACSJ 0911 1.05(14/1353)1.03(13/1259)x MACSJ 1149 0.86(12/1512)0.86(12/1399)2.38(1/42) MACSJ 1423 1.89(24/1337)1.81(22/1213)14.29(5/35) MACSJ 2214 0.61(8/1373)0.62(8/1284)x MS 0451 1.65(16/1102)1.56(14/895)4.41(3/68) MS 1054 1.08(11/893)1.09(8/737)1.85(2/108) SDSS 1004 0.80(11/1651)0.76(11/1446)IRPowerLawCL 0152 1.16(9/889)1.11(8/719)4.17(2/48) CLJ 1226 0.53(8/961)0.47(4/850)x MACSJ 0257 --MACSJ 0717 0.98(18/1813)1.05(18/1715)x MACSJ 0744 --MACSJ 0911 --MACSJ 1149 --MACSJ 1423 1.36(14/1337)1.40(14/1213)x MACSJ 2214 --MS 0451 0.23(2/1102)0.22(2/895)1.47(1/68) 126

PAGE 127

Table4-4.Continued Cluster%AGN weighted %AGN 50% %AGN spec MS 1054 0.62(7/893)0.41(3/737)0.93(1/108) SDSS 1004 0.21(3/1651)0.21(3/1446)X-rayCL 0152 0.49(4/889)0.42(3/719)4.17(2/48) CLJ 1226 0.65(6/961)0.59(5/850)4.0(1/25) MACSJ 0257 0.08(1/1535)0.07(1/1418)7.69(1/13) MACSJ 0717 0.43(9/1813)0.41(7/1715)1.85(2/108) MACSJ 0744 0.93(12/1430)0.91(12/1321)3.23(1/31) MACSJ 0911 0.38(5/1353)0.40(5/1259)x MACSJ 1149 0.12(2/1512)0.14(2/1399)x MACSJ 1423 0.67(10/1337)0.66(8/1213)5.71(2/35) MACSJ 2214 0.20(4/1373)0.23(3/1284)1.92(1/52) MS 0451 0.77(7/1102)0.78(7/895)x MS 1054 1.0(8/893)1.09(8/737)0.93(1/108) SDSS 1004 0.42(6/1651)0.41(6/1446)127

PAGE 128

CHAPTER5 THEDISTRIBUTIONOFAGNINGALAXYCLUSTERS PreviousX-raysurveysofindividualorsmallsamplesofgal axyclustershave uncoveredanexcessofX-raypointsourcesingalaxycluster s( Cappietal. 2001 ; Ruderman&Ebeling 2005 ; Eckartetal. 2006 ).Morerecently, Galametzetal. ( 2009 ), Martinietal. ( 2009 ),and Atleeetal. ( 2011 )haveexaminedX-raypointsourcesand AGNidentiedinotherwavelengthsingalaxyclustersoutto redshiftsz 1.5to examinehowenvironmentaffectstheirspatialdistributio nwithinthecluster.Inthis chapter,weexaminetheradialdistributionofAGNinourclu sterstoinvestigatethe impactoflocalenvironmentonAGN.Thiscanhelpustoaddres sissuesrelatedto AGNfuelingingalaxyclustersbydeterminingwhatprocesse smightbedominantinthe regionsofclusterswheregalaxiesaremosthighlyconcentr ated. 5.1RadialDistributionofAGN 5.1.1IndividualClusters Foreachclusterinoursample,wecomparetheradialdistrib utionofAGNto theradialdistributionofgalaxies,usingvariouscluster membershipcriterion.First, weconsideredallsourceswithintheACSimageofthecluster (referredtointhis chapterasgroup1).Wealsocomparedthedistributionsofal lsourceswith > 50% probabilityofclustermembership,asdenedinChapter 4 (group2).Finally,we produceddistributioncomparisonsforallsourceswithapr obabilitygreaterthan thecolor-selected(i.e.,red)galaxiesineachcluster,wh ichwasgenerallygalaxies with > 80%clustermembershipprobability(group3).Theradialdi stributionof spectroscopically-conrmedclustermemberswasdifcult tocompareforindividual clusters,asmostclustersonlyhad2–3AGNwithspectroscop icredshifts.Therefore,we onlycomparespectroscopically-conrmedAGNbycombining clusterdatatoimprove statistics.Weinvestigatethecombinedradialdistributi onsinSection 5.1.2 128

PAGE 129

Foreverycluster,wecomparedtheradialdistributionofth eAGNtothatof thenormalgalaxypopulationineachofthe3groupsdened.W eperformeda Kolmogorov-Smirnovtest(KS-test)todeterminetheprobab ilitythattheAGNand galaxyradialdistributionscomefromthesameparentpopul ation.Largevaluesof theKS“D”statisticindicategreaterdifferencesbetweent wodistributions.Ourresults ndasurprisinglywiderangeofvaluesforthelikelihoodth atAGNandgalaxieshave differentdistributions.WendanaverageKS-statisticDf ortheclusterpopulationand theAGNpopulationof0.24witharangeof0.11–0.55forgroup 1,withprobabilities rangingfrom13–99%thatthedistributionscomefromdiffer entparentpopulations. WendanaverageKS-statisticof0.22witharangeof0.09–0. 50forgroup2,with probabilitiesrangingfrom1–99%thatthedistributionsco mefromdifferentpopulations. Finally,wendanaverageKS-statisticof0.28witharangeo f0.16–0.34forgroup3, withprobabilitiesrangingfrom12–99%thatthedistributi onscomefromdifferentparent populations.Aswecomparethedifferencesinthedistribut ionsbetweenthegroups,we expecttoremovemoreoftheeldgalaxypopulationasthegro upnumberincreases from1to3.Theoveralltrendshowsthatthedifferencebetwe entheAGNpopulation andtheclustergalaxiesaresomewhatlargerforgroup3,inw hichweconsidergalaxies with > 80%clustermembershipprobability.Theseareessentially thosegalaxiesclose tothecenterofthecluster,galaxieswithreddercolorslik elytobeinthecluster,and spectroscopically-conrmedclustermembers. Figure 5-1 showsthecumulativedistributionofAGNandgalaxiesintwo clusters withverydifferentdistributions,MACSJ0257andMS0451.B othclusterslieat approximatelythesameredshift(z 0.5)andhavesimilarvirialradii,withMS0451 beingsomewhatlargeratr v 2.6MpcandMACSJ0257havingavirialradiusr v 2 Mpc.MACSJ0257isanexampleofaclusterwithanAGNpopulati onthatveryclosely followsthegalaxyclusterpopulationwithamaximumdeviat ionof0.33with30% probabilitythattheAGNandclustergalaxiesaredrawnfrom differentpopulations. 129

PAGE 130

Incontrast,itcanbeseenthattheAGNinMS0451appearmorec entrallyclustered thantheunderlyinggalaxypopulation,withamaximumdevia tionof0.45and > 99% probabilitythattheAGNandclustergalaxiescomefromdiff erentparentdistributions. Ofthe12clustersinoursample,2haveAGNdistributionstha tsignicantlydiffer fromtheclustergalaxypopulationwith > 95%probabilityingroup1,2havesignicantly differentdistributionsofAGNatthe > 95%condencelevelingroup2,and3have signicantlydifferentdistributionsofAGNwith > 95%condenceingroup3.Thus,we ndthatoveralltheAGNpopulationisconsistentwiththedi stributionofclustergalaxies to > 3 condencein 75%ofclusters,whileonequarterrevealAGNmorecentrally concentratedthanthenormalgalaxydistribution.5.1.2CombinedClusterSample Next,wecombineourentireclustersampletocomparetherad ialdistributionsof AGNandclustergalaxieswithbetterstatisticalsignican ce.Figure 5-2 isthecumulative distributionofgalaxiesandAGNasafunctionofradialdist ance(asafractionofthe clustervirialradius)fromthecenterofeachcluster.Weca nseethatwhenconsidering allobjectsintheclusterimage(group1,leftpanelofFigur e 5-2 ),thedistributionof AGNcloselyfollowsthedistributionofgalaxiesintheclus ters.AK-Stestshowsthatthe maximumdeviationbetweenthetwopopulationsis0.06andth etwopopulationsare differentwithonly74%condence–i.e.,theyappeartobedr awnfromthesameparent population.Whenweconsiderspectroscopically-conrmed clustermembersonly(right panelofFigure 5-2 ),wendthattheconrmedclustermembersalsofollowthecl uster galaxydistribution,withamaximumdeviationbetweenthet wopopulationsof0.23anda probabilitythatthetwopopulationsaredifferentofonly8 5%condence. However,ifwelimitourcomparisontotheinnerregionsofth eclusters(radii lessthan20–30%ofthevirialradius),weobserveasignica ntlyincreasedcentral concentrationofAGNovernormalclustergalaxies.Thiscor respondstoadistance of 0.4–0.5Mpcfromtheclustercenter,giventhattheaveragev irialradiusforour 130

PAGE 131

clustersis 1.75Mpc.ThiscentralconcentrationofAGNissignicantat the99% condencelevelwithin0.4Mpc.Wealsoobserveanunderdens ityinthedistributionof AGNcomparedwithclustergalaxiesatradialdistancesof 30–70%ofthevirialradius (0.5–1.2Mpc),signicantatthe 95%condencelevel. InFigure 5-3 ,wefurtherdivideourAGNsampleintoopticalvariables,IR powerlaw sources,andX-raypointsources.Thetoppanelshowsthecom binedclusterdatafor allsourcesintheclusterimages(group1).Wendthattheva riablesaremorecentrally concentratedthanthenormalclustergalaxieswith98%con dence,whiletheIRpower lawgalaxiesarelesscentrallyconcentratedthanthenorma lclustergalaxieswith98% condence.ThedistributionofX-raypointsourcesfollows thatofthenormalcluster galaxies,withonly86%condencethatthetwodistribution saredrawnfromdifferent parentpopulations.However,whenweconsidertheinner 0.5Mpcoftheclusters,the X-raypointsourcesshowasignicantcentralconcentratio nabovethatofthenormal clustergalaxieswith > 99%condence,whiletheradialdistributionofvariablesa ndIR sourcesisnotsignicantlydifferentfromthatoftheclust ergalaxies. Inthemiddlepanel,weconsiderobjectswith > 50%probabilityofresidingwithin thecluster(group2).Herewendthatthedistributionofva riablesappearstomatch thatoftheclustergalaxies,withonlya74%probabilitytha tthevariablesandnormal clustergalaxiesaredrawnfromdifferentparentpopulatio ns.TheIRpowerlawsources ingroup2againappearlesscentrallyconcentratedthanthe normalgalaxies,with 96%probabilitythatthesetwodistributionsaredrawnfrom differentpopulations.The distributionofX-raypointsourcesdoesnotshowasignica ntdifferencefromthatof theclustergalaxies,withonly72%probabilitythatthedis tributionscomefromdifferent parentpopulations.Ifweconsidertheinner0.5Mpcofthecl usters,weseethataswith group1,thevariablesandIRpowerlawsourcesfollowthedis tributionofnormalcluster galaxies,whiletheX-raypointsourcesagainappearmorece ntrallyconcentratedthan theclustergalaxieswith > 99%condence. 131

PAGE 132

Finally,inthebottompanelweconsideronlyspectroscopic ally-conrmedcluster members.Inthiscase,thevariablesandX-raypointsources appeartofollowthe distributionofnormalclustergalaxies,with64%and88%pr obabilitythattheAGNand clustergalaxiesaredrawnfromdifferentparentdistribut ions,respectively.TheIRpower lawsourcesagainappearlesscentrallyconcentratedthant henormalclustergalaxies with97%probabilitythatthedistributionscomefromdiffe rentparentpopulations.Within theinner0.5Mpcoftheclusters,thedistributionofthevar iablesfollowsthenormal clustergalaxies,buttheX-raypointsourcesshowacentral concentrationoverthe clustergalaxieswith > 99%condence.ThereisonlyoneIRpowerlawsourcewithin thisradius,sowearenotabletocomputetheKS-statisticfo rthispopulation. Insummary,wendthatgalaxyclustersrevealamarginallyd ifferentradial distributionfortheAGNpopulationthanthenormalcluster galaxydistribution.Variables showaslightlyhighercentralconcentrationandmid-IRsou rcesshowalowercentral concentration(Figure 5-3 ),whichresultsinatotalAGNdistributionsimilartothato fthe normalgalaxypopulation(Figure 5-2 ).TheoverdensityofAGNintheverycenterofthe clusters,however,appearstobedominatedbytheX-rayAGNc andidates(Figure 5-3 ). 5.2Discussion Galametzetal. ( 2009 )examinedalarge,uniformly-selectedsampleofgalaxy clusterstoz 1.5.TheyidentifyAGNviamid-IRcolors,radioluminosity, andX-ray luminosity,andseparatetheirsampleofclustersintothre eredshiftbinsinordertostudy theevolutionofAGNintheseclusters.Wecompareourresult swiththeX-ray-selected AGNsampleingalaxyclustersintheirintermediatebin(0.5 z 1),inwhichthey detectasmall(1.2 )overdensityofX-raysourcesinthecentersoftheircluste rs(r < 0.5–1Mpc).ThisisconsistentwiththeX-raypointsourceov erdensitywendinour clusters.( Galametzetal. 2009 )ndalesssignicant(2 )overdensityofIRsources withinthecentral0.3Mpcofintermediateredshiftcluster s,whichwedonotndamong ourdata. 132

PAGE 133

Atleeetal. ( 2011 )examinetheradialdistributionofX-raypointsourcesand IR powerlawsourcesinasampleoflow-redshift(z 0.06–0.31)galaxyclustersfrom thesamplein Martinietal. ( 2006 )and Martinietal. ( 2007 ).Theyndthattheradial distributionsofbothsamplesareconsistentwiththedistr ibutionofclustergalaxies, thoughwithbetteragreementamongtheIRsourcesthantheXraysources.Amongthe IRpowerlawsourcesinoursample,wendthatthedistributi onofIRAGNisdifferent thanthatoftheclustergalaxieswith 98%condenceatradiilargerthan 30%ofthe virialradius,butfollowsthedistributionofclustergala xieswellwithin 0.5Mpcofthe clustercenter.WealsoseeanoveralltrendthattheIRsourc esarelessclusteredthan thenormalclustergalaxies.Thisresultisconsistentwith theobservationsof Hickox etal. ( 2009 ),whoexaminetheclusteringpropertiesofdifferenttypes ofAGN.They ndthatunlikeX-rayandradiosources,IRAGNshowaprefere nceforregionswithlow density,suggestingthatIRAGNactivitymaybearesultoflo calenvironmentaswellas hostgalaxyproperties.Theyalsondthatthistrendappear smorepronouncedatsmall radii(i.e.,IRAGNarelessclusteredatsmallradii 0.3–1 h 1 Mpc).Wealsoobserve thiseffect,asthedistributionofIRAGNatradiigreaterth an 50%ofthevirialradius matchesthatofthenormalgalaxies,withonly42–75%probab ilitythattheyaredrawn fromdifferentparentdistributionsingroups1and2. Ruderman&Ebeling ( 2005 )ndanexcessofX-raypointsourceswithintheinner r < 0.5Mpcoftheirsampleof51MACSclusterscomparedwithcont rolelds.We ndasimilarresultinourclustersfortheX-raypointsourc es,asshowninFigure 5-3 Ruderman&Ebeling ( 2005 )dividetheirsampleintorelaxedanddisturbedclusters toexaminetheeffectofclustermorphologyonthedistribut ionofX-raypointsources, andndthattheobservedcentralexcessispresentinbothsa mplesbutismore pronouncedforrelaxedclusters.Theseclustershaveacent ralcoolingcoredominated bytheirmassivecDgalaxies,andtheyspeculatethatintera ctions(e.g.,mergers,tidal 133

PAGE 134

interactions)withthecDgalaxiesandothergiantelliptic alsinthecoresoftheseclusters couldresultintheexcessofX-rayAGNtheydetect. Toinvestigatethispossibilityinourclustersample,wedi videourclustersinto relaxedanddisturbedclustersfollowingthedescriptioni nSection 4.3.1 andcompare theradialdistributionsofAGNandnormalclustergalaxies inFigure 5-4 .Wend thesameexcessofAGNintheinner < 0.5Mpcoftherelaxedclustersatthe > 98% condencelevelandthisexcessisdominatedbytheX-raypoi ntsources(upperleft paneloftheFigure).Thedisturbedclustersdonotshowthis centralconcentrationof AGNovernormalgalaxies.Disturbedclustersbynaturehave signicantsubstructure andtheircentersarenotwell-dened. Ruderman&Ebeling ( 2005 )explainthatthelack ofawell-denedexcessofAGNinthecentralregionsofthese clustersmaybedueto thisfact–becausedisturbedclustershavenowell-denedc oresorsphericalsymmetry, anychangeinthedistributionofAGNwithrespecttothatoft heclustergalaxieswill bespreadoutoverlargerradii.WealsondthevariableAGNt obemorecentrally concentratedinrelaxedclustersthandisturbedclusters, thoughwithoutthecentral overdensityatr < 0.5MpcobservedforX-raysources. Ruderman&Ebeling ( 2005 )alsoobservean“AGNdepletionzone”inrelaxed clustersresultinginareduceddensityofAGNatradii 0.5–2Mpc,andanincrease ofAGNatroughlytheclustervirialradiuswherethecluster outskirtsmeettheeld. TheyexplainthereduceddensityofAGNatintermediateradi iastheresultofthe cessationofactivityasgalaxiesapproachthecenterasthe timescalefordepletionof amerger-inducedaccretiondiskaroundthecentralsuperma ssiveblackholeisless thantheclustercrossingtime.Theyalsosuggestthatthein creaseatthecluster-eld interfaceistheresultofAGNtriggeredbymergersorgalaxy harassmentasgalaxies rstfallintothecluster. Weobserveasimilardepletioneffect,withthelargestdiff erencebetweentheAGN populationandtheclustergalaxypopulationoccurringatr adii 0.5–1.2Mpc.However, 134

PAGE 135

in10/12ofourclusters,theACSFOVdoesnotcoverdistances outtor 200 .InCL0152 andSDSS1004weareabletoreachthevirialradiuswithinthe ACSFOV;weplotthese twoclustersinFigure 5-5 .Theseclustersalsohappentohavedifferentmorphologica l classications:CL0152isadisturbedclusterandSDSS1004 isarelaxedcluster.The effectofmorphologyonthedistributionofAGNrelativetot henormalclustergalaxies isclearlyseenintheseclusters.Figure 5-5 showsallAGNandgalaxiesinthecluster (group1).ParticularlyinCL0152,weseeaslightincreaseo fAGNatthevirialradius thatcouldbearesultofinteractionsatthecluster-eldin terface.Whenweconsideronly spectroscopically-conrmedclusterAGN,ofwhichCL0152h astwo,oneisfoundata radiusof1.0Mpc,justbelowthecluster'svirialradiusof1 .14Mpc,whichisconsistent withthetrendobservedby Ruderman&Ebeling ( 2005 ). 5.3AGNPropertiesandClusterRadialDistance Wealsolookforcorrelationsbetweendistancefromtheclus tercenterandAGN propertiessuchasvariabilitysignicance,X-rayhardnes sratio,andIRpowerlawslope. Weconsiderfourclustermembershipgroups:1)allgalaxies intheACSFOV,2)galaxies with > 50%clustermembershipprobability,3)galaxieswith > 80%clustermembership probability,and4)spectroscopically-conrmedclusterm embersonly.Figure 5-6 plots variabilitysignicancevs.distancefromtheclustercent erforallopticalvariables.We seenoclearcorrelationbetweenvariabilitysignicancea nddistancefromthecluster center,thoughthemostsignicantopticalvariablesappea rtolieatdistancesof20to 60%oftheclustervirialradius. Figure 5-7 plotsX-rayhardnessratiovs.distancefromtheclustercen terforall X-raypointsourcesobservedinboththehardandsoftbands. Overall,wedetect bothhardandsoftX-raysourcesatallradii,thoughthehard estsourcesappear atradiilessthan 60%oftheclustervirialradius(wenotethatwedonotdetect manyX-raysourcesoutsideofthisradiusingeneral).There isnocleartrendwith radialdistancefromtheclustercenter,thoughwendthata llX-raypointsourceswith 135

PAGE 136

spectroscopically-conrmedclusterredshiftsareamongt hesoftersourcesandhavea smallrangeofX-rayhardnessratios,withonlytwoclusterm embershavinghardness ratios > 5. Finally,Figure 5-8 plotstheslopeofIRpowerlawvs.distancefromthecluster centerforallIRpowerlawsources.WeseethatsourceswithS EDsresemblingBLAGN ( > -0.9)arefoundatallradii,butNLAGN-likeSEDs( < -0.9)appearprimarilyat radiibetweenof 20–50%oftheclustervirialradius.ThereareonlyfourIRpo werlaw sourceswithspectroscopically-conrmedclusterredshif ts(bottomrightpanelofthe Figure),andonlyonefallsbetween0.2–0.6oftheclustervi rialradius,thoughitdoes haveasteeperIRSEDthantheotherthreeobjects. Insummary,wendnosignicantcorrelationsbetweenAGNpr opertiessuchas variabilitysignicance,X-rayhardness,orIRpowerlawsl opeandtheradiallocationof anAGNwithinthecluster. 136

PAGE 137

Figure5-1.CumulativedistributionofgalaxiesandAGNinM ACSJ0257(left)and MS0451(right)forthefollowingpopulations: Top :everythingwithintheACS FOV; Middle :objectswith > 50%probabilityofresidingthecluster; Bottom : color-selectedobjectsandobjectswith > 80%clustermembership probability. 137

PAGE 138

Figure5-2.CumulativehistogramofallgalaxiesandAGNint heclusterasafunctionof fractionofthevirialradius(left).AllgalaxiesandAGNwi th spectroscopically-conrmedredshiftswithinthecluster areshownonthe right.Thesolidlinerepresentsthenormalclustergalaxie sandthedashed linerepresentstheAGNinbothgures. 138

PAGE 139

Figure5-3.RadialdistributionofAGNinourentirecluster sample,separatedbytype (opticalvariables,X-raypointsources,andIRpowerlawga laxies).Thetop panelshowsallobjectsintheACSFOV,themiddlepanelshows objectswith > 50%chanceofclustermembership,andthebottompanelshows conrmedspectroscopicclustermembersonly. 139

PAGE 140

Figure5-4.RadialdistributionofgalaxiesandAGNinourcl ustersforallobjects(top) andclustermembersonly(bottom).Theclustershavebeensp litintorelaxed (leftpanels)anddisturbed(rightpanels)clusters.Wehav eidentied16 spectroscopically-conrmedclusterAGNintherelaxedsam pleand9inthe disturbedsample. 140

PAGE 141

Figure5-5.RadialprolesofgalaxiesandAGNSDSS1004(lef t,relaxed)andCL0152 (right,disturbed).Thesolidlinedenotesnormalclusterg alaxies,whilethe thickdashedlineshowsthedistributionoftheAGN.Thedist ributionofAGN isalsoshownbytype:thedottedlinerepresentsopticalvar iables,thethin dashedlineshowsIRpowerlawsources,andthedash-dotline showsthe X-raypointsources. 141

PAGE 142

Figure5-6.Variabilitysignicancevs.radialdistancefr omtheclustercenterinfraction ofthevirialradius. UpperLeft: allgalaxiesintheACSFOV, UpperRight: galaxieswith > 50%clustermembershipprobability, LowerLeft: galaxies with > 80%clustermembershipprobability, LowerRight: spectroscopically-conrmedclustermembersonly. 142

PAGE 143

Figure5-7.X-rayhardnessratiovs.radialdistancefromth eclustercenterinfractionof thevirialradius. UpperLeft: allgalaxiesintheACSFOV, UpperRight: galaxieswith > 50%clustermembershipprobability, LowerLeft: galaxies with > 80%clustermembershipprobability, LowerRight: spectroscopically-conrmedclustermembersonly. 143

PAGE 144

Figure5-8.IRpowerlawslope( )vs.radialdistancefromtheclustercenterinfraction ofthevirialradius. UpperLeft: allgalaxiesintheACSFOV, UpperRight: galaxieswith > 50%clustermembershipprobability, LowerLeft: galaxies with > 80%clustermembershipprobability, LowerRight: spectroscopically-conrmedclustermembersonly. 144

PAGE 145

CHAPTER6 HOSTGALAXYPROPERTIES ThelinkbetweenAGNactivityandtheevolutionofgalaxiesm aybereectedin thepropertiesofgalaxieswhichcurrentlyhostactivelyac cretingsupermassiveblack holes.Thereisalsoaclearcorrelationbetweenenvironmen tandgalaxytypeandstar formationrate.Inthischapter,weexaminecharacteristic softheAGNhostgalaxiesin ourclustersampleinordertotrytounderstandtherelation shipbetweenthepresenceof anAGN,thehostgalaxy,andtheclusterenvironment. 6.1OpticalColors WeexaminetheV-IcolorsoftheAGNcandidatesinaneffortto studytheirhost galaxypropertiesandcomparethemwiththeclustergalaxyp opulationtoobserve howthepresenceofAGNselectedindifferentwaysrelatesto thehostgalaxy'sstellar population.AsdescribedinSection 3.1 ,elevenofthetwelveclustersinoursample havebeenobservedintheV-bandinadditiontotheI-bandobs ervationsusedfor ourvariabilityanalysis.InordertocomparetheobservedV -Icolors,wedivideour clustersintothreeredshiftgroupsasdescribedinSection 4.1.4 :clusterswith0.504 < z < 0.550,0.680 < z < 0.686,and0.830 < z < 0.888.InFigure 6-1 ,weshowthe observedcolor-magnitudediagramforallgalaxiesinthecl usterimageFOVdivided intothese3redshiftgroupswiththeAGNcandidatesindicat edbycoloredsymbols. Theexpectedbimodaldistributionofa“redsequence”consi stingofbulge-dominated, passively-evolvinggalaxiesanda“bluecloud”ofstar-for ming,disk-dominatedgalaxies (e.g., Stratevaetal. 2001 ; Blantonetal. 2003 ; Kauffmannetal. 2003 )isclearlyseen. Manyrecentstudies(e.g., Hickoxetal. 2009 ; Cardamoneetal. 2010 ; Sarajedini etal. 2011 )indicatethattheopticalcolorsofAGNselectedviaX-ray, IR,andoptically varyingnucleilargelyreectthecolorsofthehostgalaxy. Cardamoneetal. ( 2010 )nd thatX-ray-selectedAGNshowlessthat0.1magofcontaminat ionbytheAGNonthe hostgalaxyrest-frameU-Vcolors. Hickoxetal. ( 2009 )ndthatonly0.4–0.5magof 145

PAGE 146

correctioninu-risrequiredtoremoveAGNcontaminationfr omthebluestX-rayand mid-IRsourcesintheireldAGNsurvey.Itisexpectedthatt hemostluminousblue AGNcandidatesaremostlikelytocontainsignicantamount sofcontaminationfrom thenuclearemissiononthehostgalaxylight.AnAGN-domina tedgalaxywouldliein thebluecloudatthebrightendofthedistributioninFigure 6-1 .Wendonly5AGN candidatesamongbluesourcesatI < 20(correspondingtoanabsolutemagnitudeof M I -22ataredshiftof0.5andM I -23.5ataredshiftof0.9).Thus,wedonotsee evidenceforalargeamountofAGNcontaminationamongtheho stgalaxiesinourAGN sample,andtheassumptionthatthecolorsarerepresentati veofthehostgalaxylightis areasonableapproximation. Figure 6-2 showshistogramsofobservedV-Icolorforbothnormalgalax iesand thosehostingAGNineachofthethreeredshiftgroups.Inadd ition,wefurtherdividethe datausingclustermembershipprobabilitycuts:1)everyth ingintheACSFOV,2)objects with > 50%probabilityofclustermembership,and3)spectroscopi cally-conrmedcluster members.Forspectroscopically-conrmedclustermembers only(bottomrow),wesee mainlyredgalaxies.Thisislikelyapartialaselectioneff ectcausedbythefactthatmost objectsforwhichspectrawereobtainedwerechosenbasedon theirredcolors( Barrett 2006 ). FocusingrstonthetoptworowsoftheFigure(allgalaxiesi ntheclustereldand thosewith > 50%clustermembershipprobability),inthelowestredshif tbin(z 0.5, leftcolumnoftheFigure)weseeageneraltrendthattheopti calvariablesextendover thewholecolorrange,peakinginthe“greenvalley”between theredsequenceandblue cloudatV-I 1.5andhavinganaveragecolorof h V-I i 2.TheIR-selectedAGNfollow asimilartrend,showingbothblueandredcolorsandapeakin the“greenvalley,”though theygenerallytendtowardbluercolorswith h V-I i =1.3.Finally,theX-raypointsources showsomebluehostsbutseemtoprefersomewhatreddergalax ies,peakingonthe blueendoftheredsequenceandhaving h V-I i =1.9,perhapsindicatingthattheyare 146

PAGE 147

morepreferentiallylocatedinslightlyolder,redgalaxie s.Thisagreeswiththendings of Atleeetal. ( 2011 ),whondthattheirsampleofX-rayAGNingalaxyclusterste ndto avoidthemostvigorouslystar-forminggalaxiesandisalso consistentwiththendings of Hickoxetal. ( 2009 ),aswediscussbelow. Inthemiddleandhighestredshiftbins(z 0.7,0.85,middleandrightcolumns oftheFigure),weseethatthevariablesonceagainextendac rossthewholecolor range,includingthe“greenvalley”betweentheredandblue galaxies.TheIR-and X-ray-selectedAGNhavesimilarcolordistributionsastho seobservedinthez=0.5 clusters,thoughtheX-raypointsourcesshowgenerallyblu ercolorsthanatz=0.5. Consideringthespectroscopically-conrmedclustermemb ersonly(bottomrowofthe Figure),wendthatforthelowestredshiftclustersthevar iablesandX-raypointsources resideinredgalaxies,whiletheyarefoundinsomewhatblue rhostgalaxiesamong higherredshiftclustergalaxies.TheIR-selectedAGNappe arinbluehostgalaxiesin allredshiftbinswheretheyareidentied,thoughoursampl eofAGNinallredshiftbins suffersfromsmallnumberstatisticsinthespectroscopica lly-conrmedclustersample. Hickoxetal. ( 2009 )examinethehostcolorsofapopulationofX-ray-and IR-selectedAGNinasampleof585radio,X-ray,andIRAGNint heeldfrom0.25 < z < 0.8.TheythatX-rayAGNtendtohavecolorslocatedinthe“gr eenvalley”witha smalltailofbluehostgalaxies.ThedistributionoftheirI RAGNsampleisalsosimilar tobothoursandthatof Atleeetal. ( 2011 ),whichshowIRAGNresidingingalaxiesthat havebluer,lessluminoushoststhantheX-rayAGNandalessp ronouncedgreenpeak. Wealsocompareourresultstotheeldstudyof Sarajedinietal. ( 2011 ),whondthat AGNhostgalaxycolorscoverarangeofU-Vcolorsandalsopea kinthegreenvalley regionbetweenthebluecloudandtheredsequence.Theyndt hattheirX-raypoint sourcesandopticalvariablesinhabithostgalaxieswithar angeofcolors,withtheX-ray pointsourcespeakinginthegreenvalleyandthevariabless howingaatterdistribution throughthegreenvalley.TheirsampleofIRpowerlawsource sshowsarelatively 147

PAGE 148

atdistributionacrosstherangeofU-Vcolors,thoughthey haveasmallsamplefor whichtheyexaminethehostgalaxycolors.Theseresultsagr eewithourndingsas well,indicatingthatthehostgalaxycolorsofAGNincluste rsandtheeldmaynot signicantlydiffer.Wenotethatthisconclusionisbasedp rimarilyonoursamplesdrawn fromthedistributionofAGNintheclusterwithsomeeldcon tamination(i.e.,thetoptwo rowsofFigure 6-2 ,whichcontainallgalaxiesintheimageFOVandthosewith50 %or greaterclustermembershipprobability).Thus,wecannotr uleoutclusterenvironment effectsasunimportant,thoughwecansaythattheeffectofe nvironmentmaybemore subtle,andthusconsistentwiththequenchingofstarforma tiontakingplaceoverlonger timescalesfoundinclusterstudiesasdiscussedlaterinth issection. InordertoexplaintheopticalcolorsofAGNselectedinX-ra yandmid-IRsurveys, Hickoxetal. ( 2009 )presentageneralpictureofgalaxyevolution,inaccordan ce withthemodelpresentedby Hopkinsetal. ( 2008 ):asgalaxiesevolve,theybeginas disk-dominatedgas-rich,star-formingsystemscharacter izedbyblueopticalcolors. Oncethegalaxyundergoessomesortoftriggeringevent,suc hasamajormerger, itdisplaysarelativelyshort( 10 8 yr)phaseofAGNactivitywhichisthenquenched, alongwiththestarformationinthegalaxy.Ifthetimescale forthisquenchingof activityisshort,asAGNandstarformationactivitydeclin eanddisappearthehost galaxy'sspectrumwillevolvetowardintermediatecolorst hatresultfromacomposite spectrumofemissionfromtheolderstellarpopulationanda decliningcontributionfrom youngerstars.Finally,after 1–2Gyr,thegalaxyevolvesintoared,passively-evolving spheroid-dominatedsystemintheredsequenceportionofth ecolor-magnitudediagram. Observationsofnearby(z < 0.1)galaxieshavefoundthatAGNdetectedthrough opticalemissionlinesseemtocoincidewithdecliningstar formationinthehostgalaxy andthetransitionfrombluetoredcolors( Schawinskietal. 2007 ).Thishasledtothe ideathatnuclearactivitymaysuppressorquenchstarforma tioninthehostgalaxy throughfeedbackmechanismssuchasradiationpressureand /orjets,whichcan 148

PAGE 149

injectenergy(eitherradiativeormechanical)intothegas surroundingthenucleus andpreventitfromcoolingandfuelinganyfurtherstarform ation(e.g., Fabianetal. 2003 ; McNamaraetal. 2005 ; Formanetal. 2007 ; McNamara&Nulsen 2007 ).Inthis scenario,thehostgalaxywouldtransitionfromablue,star -forminggalaxytoared, passively-evolvingone.Ingalaxyclusters,relativistic jetsfromradioAGN(suchasthe centralcDgalaxyinacluster)couldheatthesurroundingin traclustermediumandaid inquenchingstarformationinclustergalaxies(e.g., Bˆrzanetal. 2004 ; Raffertyetal. 2006 ; Crostonetal. 2008 ).Thisprocessmayaffectsomeofthehostsofthecluster AGNinoursample,alargefractionofwhichweobserveinthe“ greenvalley”andthus maybetransitioningfrombluetoredgalaxiesasstarformat ionisquenchedbyeither theAGNorthehotintraclustermedium. Thedenseclusterenvironmentitselfisexpectedtohavesom eeffectonthestar formationrates(andthuscolors)ofclustergalaxies.Stud ieswhichhaveinvestigated thestarformationratesofgalaxiesinmergingclustersatz 0.8havefoundevidence ofstarformationbeingtriggeredininfallinggalaxies,wh ichlatercoupleswithphysical mechanismssuchasrampressurestrippingorgalaxyharassm enttoquenchstar formation( Baietal. 2007 ; Marcillacetal. 2007 ). Chungetal. ( 2010 )ndevidence foranelevatedIRluminosityfunctionintheBulletCluster ,agalaxyclusteratz=0.296 currentlyundergoingamajorsupersonicmerger.Theobserv edexcessinstar-forming IRgalaxiescanbeassociatedwiththeinfallinggroupofgal axiesthathavenotyetbeen processedbytheclusterenvironmentintomorequiescentga laxies.Thissuggests thatprocesseswhichquenchstarformationoccuronalonger timescalethanthatof theinfallinggroup'saccretionintotheBulletCluster( 250Myr),andpointtowarda slowerprocesssuchas“strangulation”ofinfallinggalaxi esasanimportantprocessin galaxyclusters,throughwhichagalaxy'sloosely-boundha logasisgentlypushedaway throughinteractionswiththeintraclustermediumandcana ffectboththemorphology andstarformationofthegalaxy( Larsonetal. 1980 ; Baloghetal. 2000 ).Tofurther 149

PAGE 150

exploretheinterplayofAGNandclusterenvironmenteffect sontheAGNhostgalaxies, weinvestigatethegalaxycolorsindifferenttypesofclust ersandtheirradialdistribution inthefollowingsections. 6.2ClusterMorphologyandHostGalaxyColor Figure 6-3 plotsahistogramoftheopticalV-Icolorsofthegalaxiesan dAGN, nowwithourclustersampledividedintomorphologicalgrou ps(relaxedanddisturbed clusters,seeSection 4.3.1 )toexaminetheeffectofclustermorphologyonthecolors ofAGNhostgalaxies.Thetwohighest-redshiftclustersatz 0.8–0.9wereremoved fromthedatainordertocomparecolorsonlyinF555W-F814W, leaving6clustersinthe relaxedsampleand3clustersinthedisturbedsample. InthetoprowoftheFigure(allgalaxiesintheACSFOV),thed isturbedclusters revealadistributionofopticalvariablespeakedinthegre envalleyatV-I=1.6with h V-I i =1.9.Inrelaxedclustersthevariableshaveroughlythesam emeanV-Icolor,butextend throughthegreenvalleyandpeakintheredsequenceatV-I=2 .3.Weobservethatthe X-rayandIRAGNhavehostgalaxieswithcolorsdistributedt hroughoutthefullrangeof colorsinbothrelaxedanddisturbedclusters,withsimilar h V-I i 2.Whenweconsider spectroscopically-conrmedclustermembersonly(bottom row),itisdifculttodraw conclusionsduetothesmallnumberofobjects. IfgalaxyinteractionsandsubstructureinthehotX-raygas aremorecommonin disturbedclustersthanrelaxedclusters,thismayexplain thefactthattheAGNhosts appeartotendtowardbluercolorsintheseclusters,atleas tfortheopticalvariables. DisturbedclustersarecharacterizedbyX-raysubstructur eandatleasttwoofthese clustersareeithercurrentlyundergoingamergerorhavere centlyexperiencedone inthepast.Minormergers,galaxyharassment,andinteract ionswithsubstructurein theintraclustermediumindisturbedclustersmaybemoreco mmonasaresultoftheir morphology,whichisnotpresentintheX-raysmooth,morest ructurally-symmetric relaxedclusters.Theseprocessesmayberesponsiblefortr iggeringormaintainingstar 150

PAGE 151

formationorAGNactivityinclustergalaxies,pushingthem towardbluercolorsinstead ofallowingtheirstarformationtobequenchedandtheircol orstotransitiontothose ofapassively-evolvingredsequencegalaxy.Incontrast,g alaxiesinrelaxedclusters mayundergofewerinteractionsastheyfallintothesmoothe rclusterdensityprole,and thusgalaxiesinrelaxedclustersmaybeallowedtoevolvemo repassivelythanthosein disturbedclusters. 6.3RadialDistributionofGalaxyColors WealsoconsidertheradialdistributionofAGNhostgalaxyc olorscomparedwith thatofthenormalclusterpopulation.Inadenseclusterenv ironment,starformation activityingalaxiesissubduedwhencomparedwitheldgala xiesandahigherfraction ofearly-typegalaxiesisobserved(e.g., Hubble 1936 ; Dressler 1980 ).Thereisalsoan observedmorphology-densityrelationinclusters,charac terizedbyahigherfractionof early-typegalaxiesfoundindenseclustercoresandanincr easingfractionoflate-type galaxieswithincreasingdistancefromtheclustercentera sdensitiesapproachthat oftheeld(e.g., Dressler 1980 ; Oemler 1974 ; Melnick&Sargent 1977 ).Agalaxy's distancefromthecenterofaclustercanalsocorrelatewith thetimesinceinfallinto thecluster( Gaoetal. 2004 ),asgalaxiesatlargeclustercentricradiihavenotyet encounteredthedensestpartsoftheclusterwhilegalaxies intheclustercorewere eitherformedinthedenseenvironmentorhavealreadycross edfromtheoutskirtstothe clustercoreatleastonce. GiventhatthepresenceofanAGNmayalsoaffectstarformati oninthehostgalaxy andthusimpactthegalaxycolor,weexaminetheradialcolor distributionofthehost galaxiesofourAGNcandidatescomparedwiththatofthenorm algalaxypopulationto determineiftheAGNhasanysignicanteffectonthehostgal axyasafunctionofradial distancefromtheclustercenter.InFigure 6-4 ,weshowtheradialcolordistributionof thehostgalaxiesforour3typesofAGNcandidatesandthenor malgalaxypopulation forallgalaxieswithin60%ofthevirialradiusofthecluste r.Hereweincludeallclusters 151

PAGE 152

exceptforthe2clustersbeyondz=0.8,sincethesewereobta inedinslightlydifferent HSTVandIltersandtheirexclusionallowsustoavoidsigni cantchangesinthe observedV-Icolorwithredshift.Thedistributionsweret withastraightlineusing 2 minimizationtodeterminethebesttandtheerroronthet. Wefoundthatcluster morphology(relaxedvs.disturbed)hadnodiscernibleeffe ctontheslopesofthetsto theAGNorthenormalgalaxydistribution.Theslopesofthe tstotheopticalvariables, IRpower-lawsources,andX-raysourcesarefoundtobe-0.16 0.6,1.95 1.8and -0.25 1.2,respectively.Theslopetothenormalgalaxydistribut ionis-1.2 0.10. Thus,thenormalgalaxiesdemonstrateasignicantlysteep erslopethanthehost galaxiesofanyAGNtype. FittingthecombinedAGNsamples,wendaslopeof-0.07 0.54andindicate thistwithasolidredlineinFigure 6-5 .Wealsoshowtheslopeofthenormalgalaxy populationforallgalaxiesintheclusterimageFOVwithaso lidblueline,asalsoshown inthelowerrightpanelofFigure 6-4 forcomparison.Thegalaxiesinthesetsrepresent theclusterpluseldpopulations,wheretheeldpopulatio nisexpectedtoremain constantwithclusterradius.Thus,anyobservedslopeofth islineispartlyduetothe truecolorgradientexpectedamongclustergalaxiesandpar tlyduetothedecreasing clusterdensityandincreasingdominanceofthebluereldp opulationasafunctionof increasingclusterradius.SinceourttotheAGNhostgalax iesalsoincludesallsources intheimages,thissampleissubjecttoroughlythesamechan gesinclusterpopulation densityasafunctionofradius.WendthatthettotheAGNho stsisconsistentwithno changeincolorwithradiustowithin1 ofthettingerror,whilethettotheclusterplus eldpopulationshowsanegativeslopewith > 3 condence. vonderLindenetal. ( 2010 )compileasampleof521low-redshift(z < 0.1)clusters fromtheSloanDigitalSkysurveyandexaminetheradialdist ributionofgalaxycolorsin theirsample.Theyndthattheradialdistributionofcolor sindicatesthatstarformation inclustergalaxiesisquenchedslowlyovertimescalesofaf ewGyr(i.e.,thecluster 152

PAGE 153

crossingtime).Similarly,aspreviouslymentioned Chungetal. ( 2010 )ndthatslow processessuchasstrangulationarelikelyimportantfortr ansformingthecolorsof infallingclustergalaxies,whichfollowswelltheassumpt ionofagradualdeclineinstar formationrateoverafewGyr( Baloghetal. 2000 ).Ourobservationsareconsistent withtheseresultsfortheclusterpluseldpopulation,giv enthattheslopeoftheblue lineinFigure 6-5 showsthetrendtowardreddergalaxiesinthecenterofthecl usterand bluergalaxiesatlargerradii.Thelackofsuchatrendamong theAGNhostsindicates thatAGNhostcolorsdonotdependonclusterradius.Thus,wh ilethecolorofanormal galaxywithinaclusterappearstodependtosomeextentonth eradialdistanceand consequentlythelocalenvironment,theAGNhostcolorsdon otrevealthisdependency. Thissupportsthetheorythatprocessesrelatedtotheaccre tingsupermassiveblack holemayhaveamoresignicantimpactonthestar-formingpr opertiesofthehost galaxythandoestheintraclustermediumand/orthelocalen vironmentaldensity. 153

PAGE 154

Figure6-1.Color-magnitudediagramsforthreegalaxyclus tergroupsatz 0.5,0.7, 0.85.Smallblackpointsarenormalgalaxies,greencircles areoptical variables,bluetrianglesareX-raypointsources,andreds quaresareIR powerlawsources. 154

PAGE 155

Figure6-2.HistogramofobservedV-Igalaxycolorsfordiff erentclustermembership cutsingroups1(left),2(middle),and3(right).Thetoprow isforallobjects withintheACSFOV,themiddlerowisobjectswith 50%cluster membershipprobability,andthebottomrowisspectroscopi cally-conrmed clustermembersonly.Theblackhistogramisfornormalgala xies,greenis opticalvariables,redisIRpowerlawsources,andblueisXraypoint sources. 155

PAGE 156

Figure6-3.HistogramofobservedV-Igalaxycolorsindistu rbed(leftcolumn)and relaxed(rightcolumn)clusters.Threeofourclustersatz < 0.8areclassied asdisturbed,whilesixclusterswithz < 0.8areclassiedasrelaxed.Both morphologicalclassicationshaveclusterswhichcoverth efullredshiftrange ofoursample.ThetoprowincludesallobjectswithintheACS FOV,andthe bottomrowincludesspectroscopically-conrmedclusterm embersonly.The blackhistogramisfornormalgalaxies,greenisopticalvar iables,redisIR powerlawsources,andblueisX-raypointsources. 156

PAGE 157

Figure6-4.V-Icolorvs.distancefromtheclustercenteras afractionofthevirialradius. Thesolidlineisalinearttothepoints. TopLeft :OpticalVariables; Top Right :IRPowerLawSources; BottomLeft :X-rayPointSources; Bottom Right :NormalGalaxies.Theerrorbarintheupperlefthandcorner ofthe plotsindicatestheaveragephotometricerror. 157

PAGE 158

Figure6-5.V-Icolorvs.distancefromtheclustercenteras afractionofthevirialradius. Thebluelineisalinearttothecolorsofthecluster+eldg alaxy populationandtheredlineisalinearttothecolorsoftheA GNhostgalaxy population. 158

PAGE 159

CHAPTER7 SUMMARY Inthisthesis,wehaveexploredseveralissuesconcerningt heAGNpopulationin denseclusterenvironments.Theenvironmentinmassiveclu stersisbelievedtohinder galaxymergersthatwouldtriggernuclearactivityandmake itdifcultforgalaxiesto retainareservoirofcoldgastofeedacentralsupermassive blackholeinthecluster potential.Whileearlystudiesconcludedthatthefraction ofopticalemissionlineAGN inclustersisonlyabout1%,morerecentworkwiththeChandr aX-rayobservatory hasallowedfordetectionofX-raypointsourceswithingala xyclustersandrecent studieshavedetectedseveralopticallynormalclustergal axieswhoseluminousX-ray emissionwouldindicateAGNactivity.Theemergingpicture ofahigherfractionofAGN ingalaxyclustersthanpreviouslythoughtthusreliesonth esuccessfulidentication oflow-luminosityand/orobscuredAGNthatarenotdiscerni blethroughtheiroptical spectra.ThefractionanddistributionofclusterAGNhasas ignicantimpactonour understandingoftheclusterenvironmentanditseffectonm embergalaxies. 7.1OpticalVariabilityintheGOODS-SouthField InordertoidentifyAGNamonggalaxiesinclustersortheles sdenseeld,itis necessarytousearangeofwavelengthsandtechniquestomin imizesurveybiases againstfaint,obscuredAGNorthosethatdonotdominatethe lightofthehostgalaxy. WeinvestigatetheuseofmultiwavelengthAGNidenticatio ntechniquesbyperforming anuclearopticalvariabilitystudyonapre-selectedsampl eofX-rayandmid-IRAGN intheGreatObservatoriesOriginsDeepSurvey(GOODS)Sout held.Asampleof 22mid-Infraredpowerlawsourcesand102X-raysourceswith opticalcounterparts intheHSTACSimagesoftheeldwereselected.Eachobjectis classiedwitha variabilitysignicancevaluerelatedtothestandarddevi ationofitsmagnitudeinve V-bandepochstakenwithHSTACSandseparatedby45-dayinte rvals.Thevariability signicanceiscomparedtotheoptical,mid-IR,andX-raypr opertiesofthesources. 159

PAGE 160

Wendthat26%ofallAGNcandidates(eitherX-ray-ormid-IR -selected)areoptical variables.Thefractionofopticalvariablesincreasesto5 1%whenconsideringsources withsoftX-raybandratios.Forthemid-IRAGNcandidateswh ichhavemultiwavelength SEDs,wendopticalvariabilityfor64%ofthoseclassiedw ithBLAGN-likeSEDs. WhileunobscuredAGNappeartohavethemostsignicantopti calvariability,someof themoreobscuredAGNsarealsoobservedasvariables. 7.2AGNIdenticationinGalaxyClusters Weanalyze12galaxyclustersatredshifts0.5 < z < 0.9todeterminethe AGNfractionandaddresstheissueofAGNfuelinginmassiveg alaxyclusters.We compile,forthersttime,amorecompletecatalogofcluste rAGNcandidatesusing acombinationofthreedetectiontechniquesexploredinCha pter 2 :opticalvariability utilizingmulti-epochHSTACSimagingintheI-band,X-rayp ointsourcedetectionin Chandraimages,andmid-IRSEDttingforapowerlawslopeth roughtheSpitzerIRAC channels. Weidentify178opticalvariablesamongthegalaxiesinourc lustersampleusing 2–3epochsofACSimaging,anaverageof15variablesperclus ter.Ninety(51%) ofthesevariableshave > 4 signicance.Wendthatintotal,1.1%ofallgalaxies surveyeddisplaynuclearopticalvariabilityingalaxiest oI nuc =27,correspondingtoan absolutemagnitudeofM I -15.3ataredshiftof0.5andM I -16.8ataredshiftof0.9. Wend74X-raypointsourcesdowntoafullbanduxof 7x10 16 erg/cm 2 /s usingChandraX-rayimaging,anaverageof6X-raypointsour cespercluster.Thisux correspondstoanX-rayluminosityof 6x10 40 erg/sataredshiftof0.5and 3x10 42 erg/sataredshiftof0.9.Mostofthepointsourceshavehard nessratiosonthesoft endofthedistribution,with46sources(67%)havingF X (2–8keV) = F X (0.5–2keV) 5. SeventypercentofourX-raysourcesliewithin log (F X /F opt )= 1,consistentwiththe presenceofanAGN. 160

PAGE 161

SpitzerIRACdataisavailablefor7clusters,inwhichweide ntifyatotalof64IR powerlawsourcesdowntoanIRluminosityofL IR ,8 m =1.4x10 42 erg/sataz=0.5and L IR ,8 m =6.0x10 42 erg/satz=0.9.Thisisanaverageof9IRpowerlawsourcesper cluster.Oftheseobjects,44%showBLAGN-likemid-IRSEDs, while31%havesteeper, NLAGN-likeSEDs.Wendthat87–100%ofIRpowerlawsourcesa reintheLacyAGN wedge,while13–16%ofopticalvariablesand38–47%ofX-ray pointsourceslieinthe Lacywedge.Thisindicatesthatthemajorityofopticallyva riableAGNandabouthalfof X-ray-selectedAGNdonotdominatethemid-IRlightoftheir hostgalaxies.Additionally, wendnodifferenceinthepercentageofX-rayobjectswithI RcolorsindicativeofAGN emissionbetweentheeldandclusterpopulations. Intotal,wendanaverageof25AGNcandidatesperclusterwi tharangeof12–49 perclusteroverthesample.Weidentify50X-raypointsourc esand48mid-IRpower lawsourceswithopticalcounterparts,andndthat4%ofmid -IRpowerlawsourcesand 24%ofX-raypointsourcesaredetectedasopticalvariables .Intotal, 7%ofvariables areidentiedasAGNcandidateseitherthroughX-rayemissi onorasmid-IRpowerlaw sources.AmongtheX-rayandIRobjectswendthat9%ofIRpow erlawsourcesare alsodetectedviaX-rays,while8%ofX-raypointsourcesals oshowapowerlawSED inthemid-IR.WendnoclearcorrelationbetweenX-rayhard nessandmid-IRSED classication. Wecompareourresultswithsurveysofeldgalaxiesandnda similardegreeof overlapbetweenvariability-andX-ray-detectedAGN,butl essoverlapbetweenvariable andIR-selectedobjects.Weattributesomeofthedifferenc esweobservetothefactthat theeldandclustervariabilityanalyseswereperformedat differentwavelengthsand overdifferenttemporalbaselines.Thelowerpercentageof mid-IRsourcesidentiedas variableinoursurveymaybepartiallyexplainedbylesssen sitivitytoweakly-varying nucleiintheI-bandandtheavailabilityoffewerepochsofi magingdata. 161

PAGE 162

7.3ClusterMembershipandPercentageofAGNinGalaxyClust ers Weusespectroscopiccatalogsofvariouslevelsofcomplete nesstodetermine clustermembershipwhenpossible,andcalculateclusterme mbershipprobabilities forgalaxieswithoutspectra.Weestimatethefractionofe ldgalaxiesinourcluster imagesbyusingthreeeldsintheGOODSsurveywithexposure timessimilartoour clusterACSimages.Weconstructradialprolesofourclust ersinordertoassigna clustermembershipprobabilitytoeachgalaxybasedonitsd istancefromthecenter ofthecluster.Additionally,weexaminetheV-Icolorofspe ctroscopically-conrmed clustermembersandaugmenttheradially-determinedclust ermembershipprobabilityof galaxieswithcolorsmatchingthoseofknownclustermember s. WecalculatethepercentageofclusterAGN,takingintoacco untthecluster membershipprobabilityforeachgalaxy.Wendthatourclus tershavearangeof 1–4%AGNwithamedianvalueof2.3 1.5%.Ifweconsiderspectroscopicallyconrmedclustermembersonly,wendthat4.7%ofclusterme mbersshowevidence ofnuclearactivity.WecomparethepercentageofclusterAG Nwiththepercentageof AGNamongeldgalaxiesintheGOODSelddetectedviaoptica lvariability,X-rays, ormid-IRSEDtting.WithinthesameredshiftandX-rayuxl imitsasourcluster data,2.5 1.6 %ofeldgalaxiesarefoundtohostAGN.Applyingagalaxymag nitude limittoourclusterdatatomatchtheshallowerexposuretim esinGOODS,wenda medianweightedAGNpercentageamongtheclustersof4.94 2.2%.Considering spectroscopically-conrmedclustermembersonly,theAGN percentageis5.5 2.3%. Earlierindicationsthat 5%ofgalaxiesinclustershostAGNbasedonX-ray pointsourcedetectionsareconsistentwithourndings,wh ichalsoincludeoptical variablesandIR-detectedAGN.However,theactualAGNfrac tionappearstobeheavily dependentongalaxysurveymagnitudelimitsandX-rayuxli mits.Wendthatthe numberofAGNamonggalaxiesinclustersisthesameasthatin theeld( 2.5%).Our resultsindicatethatgalaxiesareabletofuelaccretionon tothecentralsupermassive 162

PAGE 163

blackhole,evenindenserenvironments.Whilemajormerger smaybeperhapsthe onlyreasonablecandidatetotriggerandsustainluminousA GN,lowerluminosityAGN, suchasmanyofthoseidentiedbyoursurvey,maybetriggere dorsustainedthrough agreatervarietyofphysicalprocessesincludingbarstruc tures,minormergers,galaxy harassment,andstellarmassloss–allofwhichstillplayas ignicantroleinagalaxy clusterenvironment. WeinvestigatedcorrelationsbetweenthepercentageofAGN andclusterphysical propertiessuchasmass,X-rayluminosity,size,andredshi ft.Wendnoobvioustrends amongclusterpropertiesandthepercentageofAGNdetected ,thoughinmostcases ourclusterscoverarelativelysmallrangeofthesepropert ies.Wefurtherdividedour clustersbymorphology,classifyingthemaseitherrelaxed ordisturbedclusters,anddo notndasignicantdifferenceinthepercentageofcluster galaxiesthathostAGNasa functionofclusterdynamicalstate. 7.4RadialDistributionofAGN Inordertoexaminetheimpactoflocalenvironmentoncluste rAGN,wecompare theradialdistributionofAGNtotheradialdistributionof galaxiesinourclusters.A KS-testrevealsthattheAGNpopulationisconsistentwitht heradialdistributionof clustergalaxiesto > 3 condencein 75%ofclusters,whileonequarterrevealAGNto bemorecentrallyconcentratedthanthenormalgalaxydistr ibution.Combiningthedata fromourclusters,weobserveasignicantlyincreasedcent ralconcentrationofAGN overnormalgalaxieswithin 0.4–0.5Mpcoftheclustercenterwith99%condence. WealsoobserveanunderdensityinthedistributionofAGNco mparedwithcluster galaxiesatradialdistancesof 0.5–1.2Mpcsignicantatthe 95%condencelevel. ThecentraloverdensityofvariablesandX-raypointsource smaybeexplainedby interactions(e.g.,mergers,tidalinteractions)withthe cDgalaxiesandothergiant ellipticalsinthecoresoftheseclusters,whiletheunderd ensitymaybeexplainedby thefollowingscenario:asgalaxiesfallintotheclusterfo rthersttime,nuclearactivity 163

PAGE 164

maybetriggeredbyinteractionsatthecluster-eldinterf ace.TheAGNtriggeredasa resultwilleventuallydepletetheiraccretiondisksovera timescalelessthanthecluster crossingtime,thusresultinginanunderdensityatinterme diateradii. WeexploreradialtrendsamongAGNcandidatesdetectedviao urthree identicationmethods.Variablesappeartobemorecentral lyconcentratedthanthe normalclustergalaxieswith98%condence,whileIRpowerl awgalaxiesareless centrallyconcentratedthanthenormalclustergalaxieswi th98%condence.Thisis ingeneralagreementwitheldstudiesthatfoundIRAGNtobe lessclusteredthan otherAGNtypes.ThedistributionofX-raypointsourcessho wsasignicantcentral concentrationabovethatofthenormalclustergalaxieswit h > 99%condencewithinthe inner 0.5Mpcoftheclusters,consistentwithpreviousstudiesof X-raypointsources ingalaxyclusters.Wendnosignicantcorrelationsbetwe enAGNpropertiessuchas variabilitysignicance,X-rayhardness,orIRpowerlawsl opeandtheirradialdistribution withinthecluster. Wealsoexaminetheeffectofclusterdynamicalstateonther adialdistribution oftheAGNandnormalgalaxies.RelaxedclustersrevealX-ra ypointsourcesthat aremorecentrallyconcentratedthantheclustergalaxiesa t > 98%condence,which isnotobservedindisturbedclusters.ThevariableAGNtend tobemorecentrally concentratedinrelaxedclustersthandisturbedclusters, thoughwithoutthesignicant centraloverdensityatr < 0.5MpcobservedforX-raysources.Inbothtypesofclusters theIRpowerlawsourcesappearlesscentrallyconcentrated thanthenormalcluster galaxies.Thereasonwedonotobserveacentraloverdensity ofAGNindisturbed clustersmaybeduetothefactthattheyhavenowell-denedc oresorspherical symmetry,andthusanychangeinthedistributionofAGNwith respecttothatofthe clustergalaxieswillbespreadoutoverlargerradii. 164

PAGE 165

7.5HostGalaxyOpticalColors WeexaminetheV-IcolorsoftheAGNcandidatesinaneffortto studytheirhost galaxypropertiesandcomparethemwiththeclustergalaxyp opulation.Weseea generaltrendthatallAGNhavecolordistributionswhichdi fferfromthenormalbimodal galaxycolordistribution.Theopticalvariablescoverthe entiregalaxycolorrange, peakinginthe“greenvalley”betweentheredsequenceandbl uecloud.TheIR-selected AGNfollowasimilartrend,thoughtheytendtowardbluercol ors.TheX-raypoint sourcesalsopeakinthegreenvalley,butseemtogenerallyp refersomewhatredder, probablyolderhostgalaxies.Comparingourresultswithth eGOODSeld,wendthat thehostgalaxycolorsofAGNinclustersandtheeldarenots ignicantlydifferent. Whilewecannotruleoutclusterenvironmenteffectsasunim portant,wecansaythat theeffectmaybemoresubtle,andthusconsistentwiththequ enchingofstarformation takingplaceoverlongertimescalesfoundinrecentcluster studies. Ifwedivideoursampleintodisturbedandrelaxedclusters, weobservethatin disturbedclustersthedistributionofvariablespeaksint hegreenvalley,whereasin relaxedclustersthedistributionextendsthroughthegree nvalleybutpeaksinthered sequence.WendthatX-rayandIRAGNhavehostgalaxieswhic hseemtobeevenly distributedthroughoutthefullrangeofcolorsinbothrela xedanddisturbedclusters.If galaxyinteractionsorsubstructureinthehotX-raygasare morecommonindisturbed clustersthanrelaxedclusters,thenminormergers,galaxy harassment,andinteractions withsubstructureintheintraclustermediummaybemorecom monasaresultoftheir morphology,whichisnotpresentinthestructurally-symme tricrelaxedclusters.This mayberesponsiblefortriggeringmorestarformationinthe AGNhostswithindisturbed clusters,pushingthegalaxiestowardbluercolorsinstead ofallowingtheirstarformation tobemostlyquenchedandthehoststotransitiontopassivel y-evolvingredsequence galaxies. 165

PAGE 166

GiventhatthepresenceofanAGNaswellastheintraclusterm ediummayaffect starformationinthehostgalaxyandthusimpactthegalaxyc olor,weexaminethe radialcolordistributionofthehostgalaxiesofourAGNcan didatescomparedwiththat ofthenormalgalaxypopulation.Inthiswayweattempttodet ermineifthepresence ofanAGNhasanysignicanteffectonthehostgalaxyasafunc tionofradialdistance fromtheclustercenter.Wettheradialcolordistribution sofAGNandnormalgalaxies within60%ofthevirialradiusoftheclustersusing 2 minimization.Wendthatthet totheAGNhostsisconsistentwithnochangeincolorwithclu sterradius,whilethet tothenormalgalaxypopulationshowsanegativeslopewith > 3 condence.Since thegradualtransitiontowardsbluergalaxycolorswithinc reasingclusterradiusisa well-knowntrendfornormalgalaxiesrelatedtothecluster localenvironment,thelack ofsuchatrendamongtheAGNhostgalaxypopulationisimport ant.Thisobservation supportsthetheorythatprocessesrelatedtotheaccreting supermassiveblackhole mayhaveamoresignicantimpactonthestar-formingproper tiesofthehostgalaxy thandoestheintraclustermediumand/orthelocalenvironm entaldensity. 166

PAGE 167

APPENDIXA CLUSTERACSIMAGES nr nrrnrn FigureA-1.FinaldrizzledHSTACSI-bandimagesofthe12clu stersinoursample. 167

PAGE 168

nnn rr Figure A-1 .continued 168

PAGE 169

APPENDIXB CLUSTERCHANDRAX-RAYIMAGES n nnrrn FigureB-1.FinalmergedChandraX-rayimagesofthe12clust ersinoursample.The greensquareshowsthepositionoftheACSI-bandimageandci rclesshow theX-raypointsourcesidentiedinthisstudy. 169

PAGE 170

nr Figure B-1 .continued 170

PAGE 171

APPENDIXC CLUSTERRADIALPROFILES FigureC-1. Toppanelofeachgure :Radialplotofgalaxies/arcmin 2 vs.radialdistance fromthecenterinarcminutes.Thesolidhistogramisthetot alnumberof objectsineachradialbinandthedottedhistogramistheel d-subtracted radialproleofthecluster.Theaveragenumberofgalaxies /arcmin 2 is shownasthehorizontaldashedline. Bottompanelofeachgure :Radial plotshowingtheprobabilitythatagalaxyresidesintheclu ster(plus symbols)andtheeld(asterisks)asafunctionofradialdis tancefromthe centerinarcminutes.Thedashedanddottedlinesarepolyno mialtstothe clusterandeldprobabilities. 171

PAGE 172

Figure C-1 .continued 172

PAGE 173

REFERENCES Abell,G.O.(1965).ClusteringofGalaxies. ARA&A 3 ,1–+. Alexander,D.M.,Bauer,F.E.,Brandt,W.N.,Chary,R.,Dick inson,M.,Grogin,N.A., Koekemoer,A.M.,Padovani,P.,Urry,C.M.,Rosati,P.,Berg eron,J.,Fosbury,R., Ettori,S.,&GOODSTeam(2002).Tightconstraintsonthenat ureoftheopticallyfaint X-raysourcepopulationfromtheGOODSSouthernField.In AmericanAstronomical SocietyMeetingAbstracts ,vol.201of BulletinoftheAmericanAstronomicalSociety (pp.604–+). Alexander,D.M.,Bauer,F.E.,Brandt,W.N.,Schneider,D.P .,Hornschemeier,A.E., Vignali,C.,Barger,A.J.,Broos,P.S.,Cowie,L.L.,Garmir e,G.P.,Townsley,L.K., Bautz,M.W.,Chartas,G.,&Sargent,W.L.W.(2003).TheChan draDeepFieldNorth Survey.XIII.2MsPoint-SourceCatalogs. AJ 126 ,539–574. Alexander,D.M.,Smail,I.,Bauer,F.E.,Chapman,S.C.,Bla in,A.W.,Brandt,W.N., &Ivison,R.J.(2005).Rapidgrowthofblackholesinmassive star-forminggalaxies. Nature 434 ,738–740. Almaini,O.,Lawrence,A.,&Boyle,B.J.(1999).TheAGNcont ributiontodeep submillimetresurveysandthefar-infraredbackground. MNRAS 305 ,L59–L63. Alonso-Herrero,A.,P erez-Gonz alez,P.G.,Alexander,D.M.,Rieke,G.H.,Rigopoulou, D.,LeFloc'h,E.,Barmby,P.,Papovich,C.,Rigby,J.R.,Bau er,F.E.,Brandt,W.N., Egami,E.,Willner,S.P.,Dole,H.,&Huang,J.(2006).Infra redPower-LawGalaxies intheChandraDeepField-South:ActiveGalacticNucleiand UltraluminousInfrared Galaxies. ApJ 640 ,167–184. Antonucci,R.(1993).Uniedmodelsforactivegalacticnuc leiandquasars. ARAA 31 473–521. Aretxaga,I.,&Terlevich,R.(1994).Typetransitionsinst arburstspoweredAGN.In T.Courvoisier&A.Blecha(Ed.) Multi-WavelengthContinuumEmissionofAGN ,vol. 159of IAUSymposium ,(pp.438–+). Atlee,D.W.,Martini,P.,Assef,R.J.,Kelson,D.D.,&Mulch aey,J.S.(2011).A Multi-wavelengthStudyofLow-RedshiftClustersofGalaxi es.I.Comparisonof X-rayandMid-infraredSelectedActiveGalacticNuclei. ApJ 729 ,22–+. Bai,L.,Marcillac,D.,Rieke,G.H.,Rieke,M.J.,Tran,K.,H inz,J.L.,Rudnick,G.,Kelly, D.M.,&Blaylock,M.(2007).IRObservationsofMS1054-03:S tarFormationandIts EvolutioninRichGalaxyClusters. ApJ 664 ,181–197. Baldwin,J.A.,Phillips,M.M.,&Terlevich,R.(1981).Clas sicationparametersforthe emission-linespectraofextragalacticobjects. PASP 93 ,5–19. Balogh,M.L.,Navarro,J.F.,&Morris,S.L.(2000).TheOrig inofStarFormation GradientsinRichGalaxyClusters. ApJ 540 ,113–121. 173

PAGE 174

Barger,A.J.(2003).TheX-rayBackground.InV.Avila-Rees e,C.Firmani,C.S.Frenk, &C.Allen(Ed.) RevistaMexicanadeAstronomiayAstrosicaConferenceSer ies vol.17of RevistaMexicanadeAstronomiayAstrosicaConferenceSer ies ,(pp. 226–229). Barger,A.J.,Cowie,L.L.,Mushotzky,R.F.,&Richards,E.A .(2001).TheNatureofthe HardX-RayBackgroundSources:Optical,Near-Infrared,Su bmillimeter,andRadio Properties. AJ 121 ,662–682. Barmby,P.,Alonso-Herrero,A.,Donley,J.L.,Egami,E.,Fa zio,G.G.,Georgakakis, A.,Huang,J.-S.,Laird,E.S.,Miyazaki,S.,Nandra,K.,Par k,S.Q.,P erez-Gonz alez, P.G.,Rieke,G.H.,Rigby,J.R.,&Willner,S.P.(2006).MidInfraredPropertiesof X-RaySourcesintheExtendedGrothStrip. ApJ 642 ,126–139. Barnes,J.E.,&Hernquist,L.(1992).Dynamicsofinteracti nggalaxies. ARA&A 30 705–742. Barrett,E.(2006). Theassemblyofmassivegalaxyclusters:AjointX-ray-opti calstudy Ph.D.thesis,UniversityofHawai'iatManoa. Bauer,F.E.,Alexander,D.M.,Brandt,W.N.,Hornschemeier ,A.E.,Vignali,C., Garmire,G.P.,&Schneider,D.P.(2002).TheChandraDeepFi eldNorthSurvey.XII. TheLinkbetweenFaintX-RayandRadioSourcePopulations. AJ 124 ,2351–2363. Bershady,M.A.,Trevese,D.,&Kron,R.G.(1998).VariableE xtendedObjectsin SelectedArea57. ApJ 496 ,103–+. Best,P.N.,vanDokkum,P.G.,Franx,M.,&R ¨ ottgering,H.J.A.(2002). Jyradio sourcesintheclusterMS1054-03. MNRAS 330 ,17–34. Bˆrzan,L.,Rafferty,D.A.,McNamara,B.R.,Wise,M.W.,&N ulsen,P.E.J.(2004).A SystematicStudyofRadio-inducedX-RayCavitiesinCluste rs,Groups,andGalaxies. ApJ 607 ,800–809. Blanton,M.R.,Hogg,D.W.,Bahcall,N.A.,Baldry,I.K.,Bri nkmann,J.,Csabai,I., Eisenstein,D.,Fukugita,M.,Gunn,J.E.,Ivezi c, Z.,Lamb,D.Q.,Lupton,R.H., Loveday,J.,Munn,J.A.,Nichol,R.C.,Okamura,S.,Schlege l,D.J.,Shimasaku,K., Strauss,M.A.,Vogeley,M.S.,&Weinberg,D.H.(2003).TheB roadbandOptical PropertiesofGalaxieswithRedshifts0.02 < z < 0.22. ApJ 594 ,186–207. Brandt,W.N.,&Hasinger,G.(2005).DeepExtragalacticX-R aySurveys. ARA&A 43 827–859. 174

PAGE 175

Cappi,M.,Mazzotta,P.,Elvis,M.,Burke,D.J.,Comastri,A .,Fiore,F.,Forman, W.,Fruscione,A.,Green,P.,Harris,D.,Hooper,E.J.,Jone s,C.,Kaastra,J.S., Kellogg,E.,Murray,S.,McNamara,B.,Nicastro,F.,Ponman ,T.J.,Schlegel,E.M., Siemiginowska,A.,Tananbaum,H.,Vikhlinin,A.,Virani,S .,&Wilkes,B.(2001). ChandraStudyofanOverdensityofX-RaySourcesaroundTwoD istant(Z0.5) Clusters. ApJ 548 ,624–638. Cardamone,C.N.,Urry,C.M.,Schawinski,K.,Treister,E., Brammer,G.,&Gawiser,E. (2010).Dust-correctedColorsRevealBimodalityintheHos t-galaxyColorsofActive GalacticNucleiatz1. ApJL 721 ,L38–L42. Carleton,N.P.,Elvis,M.,Fabbiano,G.,Willner,S.P.,Law rence,A.,&Ward,M.(1987). Thecontinuumoftype1Seyfertgalaxies.II-Separatingthe rmalandnonthermal components. ApJ 318 ,595–611. Cheng,F.,Danese,L.,Franceschini,A.,&deZotti,G.(1985 ).Theopticalluminosity functionofSeyfert1nuclei. MNRAS 212 ,857–871. Chung,S.M.,Gonzalez,A.H.,Clowe,D.,Markevitch,M.,&Za ritsky,D.(2010). StarFormationintheBulletCluster.I.TheInfraredLumino sityFunctionandStar FormationRate. ApJ 725 ,1536–1549. Comastri,A.,Brusa,M.,Ciliegi,P.,Mignoli,M.,Fiore,F. ,Maiolino,R.,Severgnini,P., Baldi,A.,Molendi,S.,Vignali,C.,LaFranca,F.,Matt,G., &Perola,G.C.(2002).On thenatureofX-rayBrightOpticallyNormalgalaxies. ArXivAstrophysicse-prints Croston,J.H.,Hardcastle,M.J.,Birkinshaw,M.,Worrall, D.M.,&Laing,R.A.(2008). AnXMM-Newtonstudyoftheenvironments,particlecontenta ndimpactoflow-power radiogalaxies. MNRAS 386 ,1709–1728. Croton,D.J.,Farrar,G.R.,Norberg,P.,Colless,M.,Peaco ck,J.A.,Baldry,I.K., Baugh,C.M.,Bland-Hawthorn,J.,Bridges,T.,Cannon,R.,C ole,S.,Collins,C., Couch,W.,Dalton,G.,DePropris,R.,Driver,S.P.,Efstath iou,G.,Ellis,R.S.,Frenk, C.S.,Glazebrook,K.,Jackson,C.,Lahav,O.,Lewis,I.,Lum sden,S.,Maddox,S., Madgwick,D.,Peterson,B.A.,Sutherland,W.,&Taylor,K.( 2005).The2dFGalaxy RedshiftSurvey:luminosityfunctionsbydensityenvironm entandgalaxytype. MNRAS 356 ,1155–1167. Demarco,R.,Rosati,P.,Lidman,C.,Homeier,N.L.,Scannap ieco,E.,Bentez,N., Mainieri,V.,Nonino,M.,Girardi,M.,Stanford,S.A.,Tozz i,P.,Borgani,S.,Silk, J.,Squires,G.,&Broadhurst,T.J.(2005).AVLTspectrosco picsurveyofRX J0152.7-1357,aformingclusterofgalaxiesatz=0.837. A&A 432 ,381–394. Dickey,J.M.,&Lockman,F.J.(1990).HIintheGalaxy. ARAA 28 ,215–261. Donahue,M.,Gaskin,J.A.,Patel,S.K.,Joy,M.,Clowe,D.,& Hughes,J.P.(2003).The Mass,BaryonicFraction,andX-RayTemperatureoftheLumin ous,High-Redshift ClusterofGalaxiesMS0451.6-0305. ApJ 598 ,190–209. 175

PAGE 176

Donley,J.L.,Rieke,G.H.,P erez-Gonz alez,P.G.,&Barro,G.(2008).Spitzer's ContributiontotheAGNPopulation. ApJ 687 ,111–132. Donley,J.L.,Rieke,G.H.,P erez-Gonz alez,P.G.,Rigby,J.R.,&Alonso-Herrero,A. (2007).SpitzerPower-LawActiveGalacticNucleusCandida tesintheChandraDeep Field-North. ApJ 660 ,167–190. Dressler,A.(1980).Galaxymorphologyinrichclusters-Im plicationsfortheformation andevolutionofgalaxies. ApJ 236 ,351–365. Dressler,A.,Smail,I.,Poggianti,B.M.,Butcher,H.,Couc h,W.J.,Ellis,R.S.,&Oemler, A.,Jr.(1999).ASpectroscopicCatalogof10DistantRichCl ustersofGalaxies. ApJS 122 ,51–80. Dressler,A.,Thompson,I.B.,&Shectman,S.A.(1985).Stat isticsofemission-line galaxiesinrichclusters. ApJ 288 ,481–486. Ebeling,H.,Barrett,E.,Donovan,D.,Ma,C.-J.,Edge,A.C. ,&vanSpeybroeck,L. (2007).ACompleteSampleof12VeryX-RayLuminousGalaxyCl ustersatz > 0.5. ApJL 661 ,L33–L36. Ebeling,H.,Edge,A.C.,&Henry,J.P.(2001).MACS:AQuestf ortheMostMassive GalaxyClustersintheUniverse. ApJ 553 ,668–676. Ebeling,H.,Jones,L.R.,Perlman,E.,Scharf,C.,Horner,D .,Wegner,G.,Malkan,M., Fairley,B.W.,&Mullis,C.R.(2000).TheWARPSSurvey.III. TheDiscoveryofan X-RayLuminousGalaxyClusteratz=0.833andtheImpactofXRaySubstructureon ClusterAbundanceMeasurements. ApJ 534 ,133–145. Eckart,M.E.,Stern,D.,Helfand,D.J.,Harrison,F.A.,Mao ,P.H.,&Yost,S.A.(2006). TheSerendipitousExtragalacticX-RaySourceIdenticati on(SEXSI)Program.III. OpticalSpectroscopy. ApJS 165 ,19–56. Edelson,R.A.,Malkan,M.A.,&Rieke,G.H.(1987).Broad-ba ndpropertiesoftheCfA SeyfertGalaxies.II-Infraredtomillimeterproperties. ApJ 321 ,233–250. Ellis,S.C.,Jones,L.R.,Donovan,D.,Ebeling,H.,&Khosro shahi,H.G.(2006).The colour-magnituderelationsofClJ1226.9+3332,amassivec lusterofgalaxiesatz= 0.89. MNRAS 368 ,769–780. Elvis,M.,Wilkes,B.J.,McDowell,J.C.,Green,R.F.,Becht old,J.,Willner,S.P.,Oey, M.S.,Polomski,E.,&Cutri,R.(1994).Atlasofquasarenerg ydistributions. ApJS 95 1–68. Fabbiano,G.,Kim,D.-W.,&Trinchieri,G.(1992).AnX-rayc atalogandatlasofgalaxies. ApJS 80 ,531–644. Fabian,A.C.,&Iwasawa,K.(1999).Themassdensityinblack holesinferredfromthe X-raybackground. MNRAS 303 ,L34–L36. 176

PAGE 177

Fabian,A.C.,Sanders,J.S.,Allen,S.W.,Crawford,C.S.,I wasawa,K.,Johnstone, R.M.,Schmidt,R.W.,&Taylor,G.B.(2003).AdeepChandraob servationofthe Perseuscluster:shocksandripples. MNRAS 344 ,L43–L47. Ferrarese,L.,&Merritt,D.(2000).AFundamentalRelation betweenSupermassive BlackHolesandTheirHostGalaxies. ApJL 539 ,L9–L12. Fiore,F.,LaFranca,F.,Vignali,C.,Comastri,A.,Matt,G. ,Perola,G.C.,Cappi,M., Elvis,M.,&Nicastro,F.(2000).Spectroscopicidenticat ionoftenfainthardX-ray sourcesdiscoveredbyChandra. NewAstronomy 5 ,143–153. Forman,W.,Jones,C.,Churazov,E.,Markevitch,M.,Nulsen ,P.,Vikhlinin,A., Begelman,M.,B ¨ ohringer,H.,Eilek,J.,Heinz,S.,Kraft,R.,Owen,F.,&Pah re,M. (2007).Filaments,Bubbles,andWeakShocksintheGaseousA tmosphereofM87. ApJ 665 ,1057–1066. Freeman,P.E.,Kashyap,V.,Rosner,R.,&Lamb,D.Q.(2002). AWavelet-Based AlgorithmfortheSpatialAnalysisofPoissonData. ApJS 138 ,185–218. Fukugita,M.,Shimasaku,K.,&Ichikawa,T.(1995).GalaxyC olorsinVarious PhotometricBandSystems. PASP 107 ,945–+. Galametz,A.,Stern,D.,Eisenhardt,P.R.M.,Brodwin,M.,B rown,M.J.I.,Dey,A., Gonzalez,A.H.,Jannuzi,B.T.,Moustakas,L.A.,&Stanford ,S.A.(2009).The CosmicEvolutionofActiveGalacticNucleiinGalaxyCluste rs. ApJ 694 ,1309–1316. Gao,L.,White,S.D.M.,Jenkins,A.,Stoehr,F.,&Springel, V.(2004).Thesubhalo populationsof CDMdarkhaloes. MNRAS 355 ,819–834. Gebhardt,K.,Bender,R.,Bower,G.,Dressler,A.,Faber,S. M.,Filippenko,A.V.,Green, R.,Grillmair,C.,Ho,L.C.,Kormendy,J.,Lauer,T.R.,Mago rrian,J.,Pinkney,J., Richstone,D.,&Tremaine,S.(2000).ARelationshipbetwee nNuclearBlackHole MassandGalaxyVelocityDispersion. ApJL 539 ,L13–L16. Georgakakis,A.,Georgantopoulos,I.,Vallb e,M.,Kolokotronis,V.,Basilakos,S.,Plionis, M.,Stewart,G.C.,Shanks,T.,&Boyle,B.J.(2004).TheXMMNewton/2dFsurveyII.ThenatureofX-ray-faintopticallybrightX-raysource s. MNRAS 349 ,135–145. Georgakakis,A.,Hopkins,A.M.,Sullivan,M.,Afonso,J.,G eorgantopoulos,I., Mobasher,B.,&Cram,L.E.(2003).ThePhoenixDeepSurvey:X -rayproperties offaintradiosources. MNRAS 345 ,939–948. Giacconi,R.,Rosati,P.,Tozzi,P.,Nonino,M.,Hasinger,G .,Norman,C.,Bergeron,J., Borgani,S.,Gilli,R.,Gilmozzi,R.,&Zheng,W.(2001).Fir stResultsfromtheX-Ray andOpticalSurveyoftheChandraDeepFieldSouth. ApJ 551 ,624–634. 177

PAGE 178

Giacconi,R.,Zirm,A.,Wang,J.,Rosati,P.,Nonino,M.,Toz zi,P.,Gilli,R.,Mainieri, V.,Hasinger,G.,Kewley,L.,Bergeron,J.,Borgani,S.,Gil mozzi,R.,Grogin,N., Koekemoer,A.,Schreier,E.,Zheng,W.,&Norman,C.(2002). ChandraDeepField South:The1MsCatalog. ApJS 139 ,369–410. Giovanelli,R.,&Haynes,M.P.(1985).Gasdeciencyinclus tergalaxies-Acomparison ofnineclusters. ApJ 292 ,404–425. Gisler,G.R.(1978).Clustersofgalaxiesandthestatistic sofemission-linegalaxies. MNRAS 183 ,633–643. Hao,L.,Strauss,M.A.,Fan,X.,Tremonti,C.A.,Schlegel,D .J.,Heckman,T.M., Kauffmann,G.,Blanton,M.R.,Gunn,J.E.,Hall,P.B.,Ivezi c, Z.,Knapp,G.R.,Krolik, J.H.,Lupton,R.H.,Richards,G.T.,Schneider,D.P.,Strat eva,I.V.,Zakamska,N.L., Brinkmann,J.,&Szokoly,G.P.(2005).ActiveGalacticNucl eiintheSloanDigitalSky Survey.II.Emission-LineLuminosityFunction. AJ 129 ,1795–1808. Hartwick,F.D.A.,&Schade,D.(1990).Thespacedistributi onofquasars. ARAA 28 437–489. Hawkins,M.R.S.(2002).Variabilityinactivegalacticnuc lei:confrontationofmodels withobservations. MNRAS 329 ,76–86. Henry,J.P.,&Briel,U.G.(1991).TheX-raymorphologyofth erelaxedclusterof galaxiesA2256.II-Sourcesaroundtheextendedclusteremi ssion. A&A 246 L14–L16. Hickox,R.C.,Jones,C.,Forman,W.R.,Murray,S.S.,Kochan ek,C.S.,Eisenstein, D.,Jannuzi,B.T.,Dey,A.,Brown,M.J.I.,Stern,D.,Eisenh ardt,P.R.,Gorjian,V., Brodwin,M.,Narayan,R.,Cool,R.J.,Kenter,A.,Caldwell, N.,&Anderson,M.E. (2009).HostGalaxies,Clustering,EddingtonRatios,andE volutionofRadio,X-Ray, andInfrared-SelectedAGNs. ApJ 696 ,891–919. Hoekstra,H.,Franx,M.,&Kuijken,K.(2000).HubbleSpaceT elescopeWeak-Lensing Studyofthez=0.83ClusterMS1054-03. ApJ 532 ,88–108. Hook,I.M.,McMahon,R.G.,Boyle,B.J.,&Irwin,M.J.(1994) .TheVariabilityof OpticallySelectedQuasars. MNRAS 268 ,305–+. Hopkins,P.F.,&Hernquist,L.(2009).QuasarsAreNotLight Bulbs:TestingModels ofQuasarLifetimeswiththeObservedEddingtonRatioDistr ibution. APJ 698 1550–1569. Hopkins,P.F.,Hernquist,L.,Cox,T.J.,DiMatteo,T.,Mart ini,P.,Robertson,B.,& Springel,V.(2005).BlackHolesinGalaxyMergers:Evoluti onofQuasars. APJ 630 705–715. 178

PAGE 179

Hopkins,P.F.,Hernquist,L.,Cox,T.J.,&Kere s,D.(2008).ACosmologicalFramework fortheCo-EvolutionofQuasars,SupermassiveBlackHoles, andEllipticalGalaxies.I. GalaxyMergersandQuasarActivity. ApJS 175 ,356–389. Hornschemeier,A.E.,Brandt,W.N.,Garmire,G.P.,Schneid er,D.P.,Barger,A.J., Broos,P.S.,Cowie,L.L.,Townsley,L.K.,Bautz,M.W.,Burr ows,D.N.,Chartas,G., Feigelson,E.D.,Grifths,R.E.,Lumb,D.,Nousek,J.A.,Ra msey,L.W.,&Sargent, W.L.W.(2001).TheChandraDeepSurveyoftheHubbleDeepFie ld-NorthArea. II.ResultsfromtheCaltechFaintFieldGalaxyRedshiftSur veyArea. ApJ 554 742–777. Hubble,E.(1936). RealmoftheNebulae .NewHaven:YaleUniversityPress. Hubble,E.,&Humason,M.L.(1931).TheVelocity-DistanceR elationamong Extra-GalacticNebulae. ApJ 74 ,43–+. Huchra,J.,&Burg,R.(1992).Thespatialdistributionofac tivegalacticnuclei.I-The densityofSeyfertgalaxiesandliners. APJ 393 ,90–97. Inada,N.,Oguri,M.,Keeton,C.R.,Eisenstein,D.J.,Casta nder,F.J.,Chiu,K.,Hall, P.B.,Hennawi,J.F.,Johnston,D.E.,Pindor,B.,Richards, G.T.,Rix,H.,Schneider, D.P.,&Zheng,W.(2005).DiscoveryofaFifthImageoftheLar geSeparation GravitationallyLensedQuasarSDSSJ1004+4112. AstronomicalSocietyJapan Publications 57 ,L7–L10. Inada,N.,Oguri,M.,Pindor,B.,Hennawi,J.F.,Chiu,K.,Zh eng,W.,Ichikawa,S.,Gregg, M.D.,Becker,R.H.,Suto,Y.,Strauss,M.A.,Turner,E.L.,K eeton,C.R.,Annis,J., Castander,F.J.,Eisenstein,D.J.,Frieman,J.A.,Fukugit a,M.,Gunn,J.E.,Johnston, D.E.,Kent,S.M.,Nichol,R.C.,Richards,G.T.,Rix,H.,She ldon,E.S.,Bahcall, N.A.,Brinkmann,J.,Ivezi c, Z.,Lamb,D.Q.,McKay,T.A.,Schneider,D.P.,&York, D.G.(2003).Agravitationallylensedquasarwithquadrupl eimagesseparatedby 14.62arcseconds. Nature 426 ,810–812. Ivezi c, Z.,Menou,K.,Knapp,G.R.,Strauss,M.A.,Lupton,R.H.,Van denBerk,D.E., Richards,G.T.,Tremonti,C.,Weinstein,M.A.,Anderson,S .,Bahcall,N.A.,Becker, R.H.,Bernardi,M.,Blanton,M.,Eisenstein,D.,Fan,X.,Fi nkbeiner,D.,Finlator, K.,Frieman,J.,Gunn,J.E.,Hall,P.B.,Kim,R.S.J.,Kinkha bwala,A.,Narayanan, V.K.,Rockosi,C.M.,Schlegel,D.,Schneider,D.P.,Strate va,I.,SubbaRao,M., Thakar,A.R.,Voges,W.,White,R.L.,Yanny,B.,Brinkmann, J.,Doi,M.,Fukugita,M., Hennessy,G.S.,Munn,J.A.,Nichol,R.C.,&York,D.G.(2002 ).OpticalandRadio PropertiesofExtragalacticSourcesObservedbytheFIRSTS urveyandtheSloan DigitalSkySurvey. AJ 124 ,2364–2400. Jee,M.J.,White,R.L.,Bentez,N.,Ford,H.C.,Blakeslee ,J.P.,Rosati,P.,Demarco, R.,&Illingworth,G.D.(2005a).Weak-LensingAnalysisoft hez=0.8ClusterCL 0152-1357withtheAdvancedCameraforSurveys. ApJ 618 ,46–67. 179

PAGE 180

Jee,M.J.,White,R.L.,Ford,H.C.,Blakeslee,J.P.,Illing worth,G.D.,Coe,D.A., &Tran,K.(2005b).HubbleSpaceTelescopeAdvancedCameraf orSurveys Weak-LensingandChandraX-RayStudiesoftheHigh-Redshif tClusterMS 1054-0321. ApJ 634 ,813–832. Johnson,O.,Best,P.N.,&Almaini,O.(2003).Thecontentof activegalacticnucleiin thez=0.83clusterMS1054-0321. MNRAS 343 ,924–932. Kauffmann,G.,Heckman,T.M.,White,S.D.M.,Charlot,S.,T remonti,C.,Peng,E.W., Seibert,M.,Brinkmann,J.,Nichol,R.C.,SubbaRao,M.,&Yo rk,D.(2003).The dependenceofstarformationhistoryandinternalstructur eonstellarmassfor10 5 low-redshiftgalaxies. MNRAS 341 ,54–69. Kauffmann,G.,White,S.D.M.,Heckman,T.M.,M enard,B.,Brinchmann,J.,Charlot, S.,Tremonti,C.,&Brinkmann,J.(2004).Theenvironmental dependenceofthe relationsbetweenstellarmass,structure,starformation andnuclearactivityin galaxies. MNRAS 353 ,713–731. Klaas,U.,Haas,M.,M¨uller,S.A.H.,Chini,R.,Schulz,B., Coulson,I.,Hippelein,H., Wilke,K.,Albrecht,M.,&Lemke,D.(2001).Infraredtomill imetrephotometryof ultra-luminousIRgalaxies:Newevidencefavouringa3-sta gedustmodel. A&A 379 823–844. Klesman,A.,&Sarajedini,V.(2007).OpticalVariabilityo fInfraredPower-Law-Selected GalaxiesandX-RaySourcesintheGOODS-SouthField. ApJ 665 ,225–231. Koehler,T.,Groote,D.,Reimers,D.,&Wisotzki,L.(1997). Thelocalluminosityfunction ofQSOsandSeyfert1nuclei. A&A 325 ,502–510. Koo,D.C.,Kron,R.G.,&Cudworth,K.M.(1986).QuasarstoBg reaterthan22.5 inselectedarea57-Acatalogofmulticolorphotometry,var iability,andastrometry. PASP 98 ,285–306. Kormendy,J.,&Richstone,D.(1995).InwardBound—TheSear chForSupermassive BlackHolesInGalacticNuclei. ARAA 33 ,581–+. Kuraszkiewicz,J.K.,Wilkes,B.J.,Hooper,E.J.,McLeod,K .K.,Wood,K.,Bjorkman,J., Delain,K.M.,Hughes,D.H.,Elvis,M.S.,Impey,C.D.,Lonsd ale,C.J.,Malkan,M.A., McDowell,J.C.,&Whitney,B.(2003).TheFar-InfraredSpec tralEnergyDistributions ofX-Ray-selectedActiveGalaxies. ApJ 590 ,128–148. Lacy,M.,Petric,A.O.,Sajina,A.,Canalizo,G.,Storrie-L ombardi,L.J.,Armus,L., Fadda,D.,&Marleau,F.R.(2007).OpticalSpectroscopyand X-RayDetectionsofa SampleofQuasarsandActiveGalacticNucleiSelectedinthe Mid-InfraredfromTwo SpitzerSpaceTelescopeWide-AreaSurveys. AJ 133 ,186–205. 180

PAGE 181

Lacy,M.,Storrie-Lombardi,L.J.,Sajina,A.,Appleton,P. N.,Armus,L.,Chapman, S.C.,Choi,P.I.,Fadda,D.,Fang,F.,Frayer,D.T.,Heinric hsen,I.,Helou,G.,Im,M., Marleau,F.R.,Masci,F.,Shupe,D.L.,Soifer,B.T.,Surace ,J.,Teplitz,H.I.,Wilson, G.,&Yan,L.(2004).ObscuredandUnobscuredActiveGalacti cNucleiintheSpitzer SpaceTelescopeFirstLookSurvey. ApJS 154 ,166–169. Larson,R.B.,Tinsley,B.M.,&Caldwell,C.N.(1980).Theev olutionofdiskgalaxies andtheoriginofS0galaxies. ApJ 237 ,692–707. Limousin,M.,Ebeling,H.,Ma,C.-J.,Swinbank,A.M.,Smith ,G.P.,Richard,J.,Edge, A.C.,Jauzac,M.,Kneib,J.-P.,Marshall,P.,&Schrabback, T.(2010).MACS J1423.8+2404:gravitationallensingbyamassive,relaxed clusterofgalaxiesatz =0.54. MNRAS 405 ,777–782. Londish,D.,Boyle,B.J.,&Schade,D.J.(2000).Thespacede nsityoflow-redshift activegalacticnuclei. MNRAS 318 ,411–416. Lynden-Bell,D.(1969).GalacticNucleiasCollapsedOldQu asars. Nature 223 690–694. MacLeod,C.L.,Ivezi c, Z.,Kochanek,C.S.,Kozowski,S.,Kelly,B.,Bullock,E.,K imball, A.,Sesar,B.,Westman,D.,Brooks,K.,Gibson,R.,Becker,A .C.,&deVries,W.H. (2010).ModelingtheTimeVariabilityofSDSSStripe82Quas arsasaDamped RandomWalk. ApJ 721 ,1014–1033. Magorrian,J.,Tremaine,S.,Richstone,D.,Bender,R.,Bow er,G.,Dressler,A.,Faber, S.M.,Gebhardt,K.,Green,R.,Grillmair,C.,Kormendy,J., &Lauer,T.(1998).The DemographyofMassiveDarkObjectsinGalaxyCenters. AJ 115 ,2285–2305. Maia,M.A.G.,Machado,R.S.,&Willmer,C.N.A.(2003).TheS eyfertPopulationin theLocalUniverse. AJ 126 ,1750–1762. Makino,J.,&Hut,P.(1997).MergerRateofEqual-MassSpher icalGalaxies. ApJ 481 83–+. Marcillac,D.,Rigby,J.R.,Rieke,G.H.,&Kelly,D.M.(2007 ).StrongDustyBursts ofStarFormationinGalaxiesFallingintotheClusterRXJ01 52.7-1357. ApJ 654 825–834. Markarian,B.E.(1967).GalaxieswithanUltravioletConti nuum. Astrozika 3 ,55–68. Martini,P.(2004).Whydoeslow-luminosityAGNfuelingrem ainanunsolvedproblem? InT.Storchi-Bergmann,L.C.Ho,&H.R.Schmitt(Ed.) TheInterplayAmongBlack Holes,StarsandISMinGalacticNuclei ,vol.222of IAUSymposium ,(pp.235–241). Martini,P.,Kelson,D.D.,Kim,E.,Mulchaey,J.S.,&Athey, A.A.(2006).Spectroscopic ConrmationofaLargePopulationofActiveGalacticNuclei inClustersofGalaxies. ApJ 644 ,116–132. 181

PAGE 182

Martini,P.,Kelson,D.D.,Mulchaey,J.S.,&Athey,A.(2004 ).NewResultsontheAGN ContentofGalaxyClusters. ClustersofGalaxies:ProbesofCosmologicalStructure andGalaxyEvolution Martini,P.,Kelson,D.D.,Mulchaey,J.S.,&Trager,S.C.(2 002).AnUnexpectedlyHigh FractionofActiveGalacticNucleiinRedClusterGalaxies. ApJL 576 ,L109–L112. Martini,P.,Mulchaey,J.S.,&Kelson,D.D.(2007).TheDist ributionofActiveGalactic NucleiinClustersofGalaxies. ApJ 664 ,761–776. Martini,P.,Sivakoff,G.R.,&Mulchaey,J.S.(2009).TheEv olutionofActiveGalactic NucleiinClustersofGalaxiestoRedshift1.3. ApJ 701 ,66–85. Maughan,B.J.,Jones,C.,Jones,L.R.,&VanSpeybroeck,L.( 2007).Deep XMM-NewtonandChandraObservationsofClJ1226.9+3332:AD etailedX-Ray MassAnalysisofaz=0.89GalaxyCluster. ApJ 659 ,1125–1137. Maughan,B.J.,Jones,L.R.,Ebeling,H.,Perlman,E.,Rosat i,P.,Frye,C.,&Mullis, C.R.(2003).ChandraX-RayAnalysisoftheMassiveHigh-Red shiftGalaxyClusters ClJ1113.1-2615andClJ0152.7-1357. ApJ 587 ,589–604. Maughan,B.J.,Jones,L.R.,Ebeling,H.,&Scharf,C.(2004) .AnXMM-Newton observationofthemassive,relaxedgalaxyclusterClJ1226 .9+3332atz=0.89. MNRAS 351 ,1193–1203. McNamara,B.R.,&Nulsen,P.E.J.(2007).HeatingHotAtmosp hereswithActive GalacticNuclei. AR&AA 45 ,117–175. McNamara,B.R.,Nulsen,P.E.J.,Wise,M.W.,Rafferty,D.A. ,Carilli,C.,Sarazin,C.L., &Blanton,E.L.(2005).Theheatingofgasinagalaxycluster byX-raycavitiesand large-scaleshockfronts. Nature 433 ,45–47. Melnick,J.,&Sargent,W.L.W.(1977).Theradialdistribut ionofmorphologicaltypesof galaxiesinX-rayclusters. ApJ 215 ,401–407. Molnar,S.M.,Hughes,J.P.,Donahue,M.,&Joy,M.(2002).Ch andraObservationsof UnresolvedX-RaySourcesaroundTwoClustersofGalaxies. ApJL 573 ,L91–L94. Moran,E.C.,Lehnert,M.D.,&Helfand,D.J.(1999).X-Raysf romNGC3256: High-EnergyEmissioninStarburstGalaxiesandTheirContr ibutiontotheCosmic X-RayBackground. ApJ 526 ,649–664. Moran,S.M.,Ellis,R.S.,Treu,T.,Smith,G.P.,Rich,R.M., &Smail,I.(2007).A Wide-FieldSurveyofTwoz0.5GalaxyClusters:Identifyin gthePhysicalProcesses ResponsiblefortheObservedTransformationofSpiralsint oS0s. ApJ 671 1503–1522. Morgan,W.W.(1961).TheClassicationofClustersofGalax ies. Proceedingsofthe NationalAcademyofScience 47 ,905–906. 182

PAGE 183

Neugebauer,G.,Oke,J.B.,Becklin,E.E.,&Matthews,K.(19 79).Absolutespectral energydistributionofquasi-stellarobjectsfrom0.3to10 microns. ApJ 230 ,79–94. Oemler,A.,Jr.(1974).TheSystematicPropertiesofCluste rsofGalaxies.Photometryof 15Clusters. ApJ 194 ,1–20. Oguri,M.(2010).TheMassDistributionofSDSSJ1004+4112R evisited. PASJ 62 1017–. Pereyra,N.A.,VandenBerk,D.E.,Turnshek,D.A.,Hillier, D.J.,Wilhite,B.C.,Kron, R.G.,Schneider,D.P.,&Brinkmann,J.(2006).Characteris ticQSOAccretionDisk TemperaturesfromSpectroscopicContinuumVariability. ApJ 642 ,87–95. Peterson,B.M.(2001).VariabilityofActiveGalacticNucl ei.InI.Aretxaga,D.Kunth,& R.Mujica(Ed.) AdvancedLecturesontheStarburst-AGN ,(pp.3–+). Popesso,P.,&Biviano,A.(2006).TheAGNfraction-velocit ydispersionrelationin clustersofgalaxies. A&A 460 ,L23–L26. Rafferty,D.A.,McNamara,B.R.,Nulsen,P.E.J.,&Wise,M.W .(2006).The Feedback-regulatedGrowthofBlackHolesandBulgesthroug hGasAccretion andStarburstsinClusterCentralDominantGalaxies. ApJ 652 ,216–231. Richards,G.T.,Fan,X.,Newberg,H.J.,Strauss,M.A.,Vand enBerk,D.E.,Schneider, D.P.,Yanny,B.,Boucher,A.,Burles,S.,Frieman,J.A.,Gun n,J.E.,Hall,P.B.,Ivezi c, Z.,Kent,S.,Loveday,J.,Lupton,R.H.,Rockosi,C.M.,Schl egel,D.J.,Stoughton, C.,SubbaRao,M.,&York,D.G.(2002).SpectroscopicTarget SelectionintheSloan DigitalSkySurvey:TheQuasarSample. AJ 123 ,2945–2975. Rieke,G.H.,&Lebofsky,M.J.(1981).Spectralcomponentso fNGC4151. ApJ 250 87–97. Rigby,J.R.,Rieke,G.H.,Donley,J.L.,Alonso-Herrero,A. ,&P erez-Gonz alez,P.G. (2006).WhyX-Ray-selectedActiveGalacticNucleiAppearO pticallyDull. ApJ 645 115–133. Ruderman,J.T.,&Ebeling,H.(2005).TheOriginoftheSpati alDistributionof X-Ray-luminousActiveGalacticNucleiinMassiveGalaxyCl usters. ApJL 623 L81–L84. Salpeter,E.E.(1964).AccretionofInterstellarMatterby MassiveObjects. ApJ 140 796–800. Sanders,D.B.,Soifer,B.T.,Elias,J.H.,Neugebauer,G.,& Matthews,K.(1988).Warm ultraluminousgalaxiesintheIRASsurvey-Thetransitionf romgalaxytoquasar? ApJL 328 ,L35–L39. Sarajedini,V.L.,Gilliland,R.L.,&Kasm,C.(2003).AV-Ba ndSurveyforVariable GalacticNucleiintheHubbleDeepField. ApJ 599 ,173–184. 183

PAGE 184

Sarajedini,V.L.,Gilliland,R.L.,&Phillips,M.M.(2000) .VariableGalaxiesintheHubble DeepField. AJ 120 ,2825–2834. Sarajedini,V.L.,Koo,D.C.,Klesman,A.J.,Laird,E.S.,Pe rezGonzalez,P.G.,& Mozena,M.(2011).VariabilityandMultiwavelength-detec tedActiveGalacticNucleiin theGoodsFields. ApJ 731 ,97–+. Sarajedini,V.L.,Koo,D.C.,Phillips,A.C.,Kobulnicky,H .A.,Gebhardt,K.,Willmer, C.N.A.,Vogt,N.P.,Laird,E.,Im,M.,Iverson,S.,&Mattos, W.(2006).TheDEEP GrothStripSurvey.VI.Spectroscopic,Variability,andXRayDetectionofActive GalacticNuclei. ApJS 166 ,69–88. Schawinski,K.,Thomas,D.,Sarzi,M.,Maraston,C.,Kavira j,S.,Joo,S.-J.,Yi,S.K., &Silk,J.(2007).ObservationalevidenceforAGNfeedbacki nearly-typegalaxies. MNRAS 382 ,1415–1431. Schmidt,M.,&Green,R.F.(1983).Quasarevolutionderived fromthePalomarbright quasarsurveyandothercompletequasarsurveys. ApJ 269 ,352–374. Sharon,K.,Gal-Yam,A.,Maoz,D.,Filippenko,A.V.,Foley, R.J.,Silverman,J.M., Ebeling,H.,Ma,C.,Ofek,E.O.,Kneib,J.,Donahue,M.,Elli s,R.S.,Freedman,W.L., Kirshner,R.P.,Mulchaey,J.S.,Sarajedini,V.L.,&Voit,G .M.(2010).TheTypeIa SupernovaRateinRedshift0.5-0.9GalaxyClusters. ApJ 718 ,876–893. Smith,M.G.,&Wright,A.E.(1980).Aradiostudyofopticall yselectedQSOs. MNRAS 191 ,871–886. Springel,V.,DiMatteo,T.,&Hernquist,L.(2005).Modelli ngfeedbackfromstarsand blackholesingalaxymergers. MNRAS 361 ,776–794. Stern,D.,Eisenhardt,P.,Gorjian,V.,Kochanek,C.S.,Cal dwell,N.,Eisenstein,D., Brodwin,M.,Brown,M.J.I.,Cool,R.,Dey,A.,Green,P.,Jan nuzi,B.T.,Murray,S.S., Pahre,M.A.,&Willner,S.P.(2005).Mid-InfraredSelectio nofActiveGalaxies. ApJ 631 ,163–168. Stott,J.P.,Smail,I.,Edge,A.C.,Ebeling,H.,Smith,G.P. ,Kneib,J.,&Pimbblet,K.A. (2007).AnIncreaseintheFaintRedGalaxyPopulationinMas siveClusterssincez 0.5. ApJ 661 ,95–101. Strateva,I.,Ivezi c, Z.,Knapp,G.R.,Narayanan,V.K.,Strauss,M.A.,Gunn,J.E. Lupton,R.H.,Schlegel,D.,Bahcall,N.A.,Brinkmann,J.,B runner,R.J.,Budav ari,T., Csabai,I.,Castander,F.J.,Doi,M.,Fukugita,M.,Gy ory,Z.,Hamabe,M.,Hennessy, G.,Ichikawa,T.,Kunszt,P.Z.,Lamb,D.Q.,McKay,T.A.,Oka mura,S.,Racusin,J., Sekiguchi,M.,Schneider,D.P.,Shimasaku,K.,&York,D.(2 001).ColorSeparationof GalaxyTypesintheSloanDigitalSkySurveyImagingData. AJ 122 ,1861–1874. Sunyaev,R.A.,&Titarchuk,L.G.(1980).Comptonizationof X-raysinplasmacloudsTypicalradiationspectra. A&A 86 ,121–138. 184

PAGE 185

Szokoly,G.P.,Bergeron,J.,Hasinger,G.,Lehmann,I.,Kew ley,L.,Mainieri,V.,Nonino, M.,Rosati,P.,Giacconi,R.,Gilli,R.,Gilmozzi,R.,Norma n,C.,Romaniello,M., Schreier,E.,Tozzi,P.,Wang,J.X.,Zheng,W.,&Zirm,A.(20 04).TheChandraDeep Field-South:OpticalSpectroscopy.I. ApJS 155 ,271–349. Tran,K.,Franx,M.,Illingworth,G.D.,vanDokkum,P.,Kels on,D.D.,Blakeslee,J.P.,& Postman,M.(2007).AKeckSpectroscopicSurveyofMS1054-0 3(z=0.83):Forming theRedSequence. ApJ 661 ,750–767. Tran,K.,Kelson,D.D.,vanDokkum,P.,Franx,M.,Illingwor th,G.D.,&Magee,D. (1999).TheVelocityDispersionofMS1054-03:AMassiveGal axyClusteratHigh Redshift. ApJ 522 ,39–45. Trevese,D.,Boutsia,K.,Vagnetti,F.,Cappellaro,E.,&Pu ccetti,S.(2008). Variability-selectedactivegalacticnucleifromsuperno vasearchintheChandra deepeldsouth. A&A 488 ,73–81. Trevese,D.,Kron,R.G.,Majewski,S.R.,Bershady,M.A.,&K oo,D.C.(1994).The ensemblevariabilitypropertiesoffaintQSOs. ApJ 433 ,494–509. Urry,C.M.,&Padovani,P.(1995).UniedSchemesforRadioLoudActiveGalactic Nuclei. PASP 107 ,803–+. vanDokkum,P.G.,Franx,M.,Fabricant,D.,Illingworth,G. D.,&Kelson,D.D.(2000). HubbleSpaceTelescopePhotometryandKeckSpectroscopyof theRichClusterMS 1054-03:Morphologies,Butcher-OemlerEffect,andtheCol or-MagnitudeRelationat Z=0.83. ApJ 541 ,95–111. VandenBerk,D.E.,Wilhite,B.C.,Kron,R.G.,Anderson,S.F .,Brunner,R.J.,Hall, P.B.,Ivezi c, Z.,Richards,G.T.,Schneider,D.P.,York,D.G.,Brinkmann ,J.V.,Lamb, D.Q.,Nichol,R.C.,&Schlegel,D.J.(2004).TheEnsemblePh otometricVariabilityof 25,000QuasarsintheSloanDigitalSkySurvey. ApJ 601 ,692–714. Veilleux,S.,&Osterbrock,D.E.(1987).Spectralclassic ationofemission-linegalaxies. ApJS 63 ,295–310. vonderLinden,A.,Wild,V.,Kauffmann,G.,White,S.D.M.,& Weinmann,S.(2010). StarformationandAGNactivityinSDSSclustergalaxies. MNRAS 404 ,1231–1246. Webb,W.,&Malkan,M.(2000).RapidOpticalVariabilityinA ctiveGalacticNucleiand Quasars. APJ 540 ,652–677. Wilkes,B.J.,Pounds,K.A.,Schmidt,G.D.,Smith,P.S.,Cut ri,R.M.,Ghosh,H., Nelson,B.,&Hines,D.C.(2005).XMM-NewtonObservationso fRedAGNs. ApJ 634 ,183–192. Williams,L.L.R.,&Saha,P.(2004).ModelsoftheGiantQuad rupleQuasarSDSS J1004+4112. AJ 128 ,2631–2641. 185

PAGE 186

Worsley,M.A.,Fabian,A.C.,Alexander,D.M.,Bauer,F.E., Brandt,W.N.,Hasinger, G.,&Lehmer,B.D.(2005).ResolvingtheX-rayBackground.I nT.Bulik,B.Rudak,& G.Madejski(Ed.) AstrophysicalSourcesofHighEnergyParticlesandRadiati on ,vol. 801of AmericanInstituteofPhysicsConferenceSeries ,(pp.51–56). Yan,L.,Chary,R.,Armus,L.,Teplitz,H.,Helou,G.,Frayer ,D.,Fadda,D.,Surace,J.,& Choi,P.(2005).SpitzerDetectionofPolycyclicAromaticH ydrocarbonandSilicate DustFeaturesintheMid-InfraredSpectraofz 2UltraluminousInfraredGalaxies. ApJ 628 ,604–610. 186

PAGE 187

BIOGRAPHICALSKETCH AlisonKlesmanwasborninSouthBend,Indiana,andspenther childhoodin Illinois,rstinOlympiaFieldsandlaterinOrlandPark,wh eresheattendedCarl SandburgHighSchool,graduatingin1999.Whenshewasyoung ,sheprofesseda terribledilemma,unabletodecidewhethershewantedtogro wuptobeamarine biologist...orahairdresser.EventuallyAlisondiscover edStarTrek:TheNext Generationanddecidedthatshewantedtobeanengineerlike GeordiLaForge-that is,untilsherealizedthatphysicsandastronomywerealotm orefun(andsupposedly easier).Theinuenceofsomewonderfullysupportive(andm aybealittlecrazy)high schoolteachersalsohelpedtoinuenceherdecisionsigni cantly. AlisonreceivedaB.S.inphysicsandanM.S.ingeosystemsfr omthe MassachusettsInstituteofTechnology,whereshealsoserv edasacoxswainonthe women'screwteamandanofcerintheMITAnimeClub.Shecame totheUniversity ofFloridainthefallof2004andreceivedanM.S.inastronom yin2006beforenally completingherPhDinastronomyin2011. Alisonloveslotsofvariedthings,mostofwhicharebooksor science-ctionTV shows,butalsoincludecoffee,running,sewing,andcats.S heisanavidwriterand haswonanumberofprizesforhershortction,andshehopest osomedaybecomea successfully-publishednovelist. 187