arXiv:1203.1840v1 [astro-ph.GA] 8 Mar 2012 · Rines et al. 2007; McIntosh et al. 2008; Liu et al. 2009), although some studies of BCGs in the more distant universe disagree with this

Mon Not R Astron Soc000 1ndash15 (2012) Printed 7 September 2018 (MN LATEX style file v22)

Star formation activities in early-type brightest cluster galaxies

F S Liu12⋆ Shude Mao23 X M Meng21College of Physical Science and Technology Shenyang Normal University Shenyang 110034 PRChina2National Astronomical Observatories Chinese Academy of Sciences A20 Datun Road Beijing 100012 PRChina3Jodrell Bank Centre for Astrophysics University of Manchester Manchester M13 9PL UK

Accepted 2012 March 8 Received 2012 March 7 in original form 2011 July 28

ABSTRACTWe identify a total of 120 early-type Brightest Cluster Galaxies (BCGs) at 01 lt z lt 04 intwo recent large cluster catalogues selected from the SloanDigital Sky Survey (SDSS) Theyare selected with strong emission lines in their optical spectra with both Hα and [O II]λ3727line emission which indicates significant ongoing star formation They constitute aboutsim05 of the largest optically-selected low-redshift BCG sample and the fraction is a strongfunction of cluster richness Their star formation historycan be well described by a recentminor and short starburst superimposed on an old stellar component with the recent episodeof star formation contributing on average only less than 1 percent of the total stellar massWe show that the more massive star-forming BCGs in richer clusters tend to have higher starformation rate (SFR) and specific SFR (SFR per unit galaxy stellar mass) We also comparetheir statistical properties with a control sample selected from X-ray luminous clusters andshow that the fraction of star-forming BCGs in X-ray luminous clusters is almost one order ofmagnitude larger than that in optically-selected clusters BCGs with star formation in coolingflow clusters usually have very flat optical spectra and show the most active star formationwhich may be connected with cooling flows

Key words galaxies clusters general - galaxies elliptical and lenticular cD - galaxiesstarburst - galaxies cooling flows

1 INTRODUCTION

The activity of star formation (SF) in a present-day galaxy isstrongly related to the local galaxy density and stellar mass(Kauffmann et al 2004) Massive early-type galaxies lie in higherdensity environments (Dressler 1980) and are dominated by red-der older stars than less massive ones The specific star formationrate (ie SFR per unit stellar mass) of galaxies tends to belowerin denser environments (Kauffmann et al 2004) pointing to a pic-ture where more massive galaxies form stars at a lower rate per unitmass than less massive ones Therefore the bulk of stars in present-day massive galaxies must have formed at earlier epochs thanstarsin less massive galaxies (eg Kauffmann et al 2003 Thomas et al2005) The standard models of galaxy formation have difficulty re-producing these red and dead massive galaxies unless feedbackmechanisms (eg by active galactic nuclei-AGN) are introducedthat prevent the gas from cooling and forming stars The starfor-mation history of massive galaxies is not yet fully understood

The Brightest Cluster Galaxies (BCGs) are at the most lumi-nous and massive end of galaxy population They are usually lo-cated at or close to the centres of dense clusters of galaxies(egJones amp Forman 1984 Smith et al 2005) Most of them are dom-

⋆ E-mail lfsnaocascn

inated by old stars without prominent ongoing star formation Ithas been shown that BCGs are different from other massive galax-ies (non-BCGs) in the surface brightness profiles and some basicscaling relations (eg Matthews et al 1964 Oemler 19731976Schombert 1986 1987 1988 Graham et al 1996 Patel et al 2006Lauer et al 2007 Bernardi et al 2007 von der Linden et al 2007Desroches et al 2007 Liu et al 2008) which may indicate a dis-tinct formation mechanism

Recent studies from numerical simulations and semi-analyticmodels in the cold dark matter hierarchical structure formationframework indicate that a large part of stellar mass in BCGs mayhave formed before redshift three and later dry (dissipationless)mergers play an important role in their stellar mass assembly(Gao et al 2004 De Lucia amp Blaizot 2007) This picture is largelyconsistent with observations For examples many examplesof drymergers involving central galaxies in groups and clusters at zlt1have been reported (eg Lauer 1988 van Dokkum et al 1999Mulchaey et al 2006 Jeltema et al 2007 Tran et al 2005 2008Rines et al 2007 McIntosh et al 2008 Liu et al 2009) althoughsome studies of BCGs in the more distant universe disagree withthis scenario (eg Whiley et al 2008 Stott et al 2011)

The inclusion of AGN feedback in De Lucia amp Blaizot (2007)can efficiently truncate the initial starburst and ensure that theprogenitor of BCGs experiences virtually no star formationin

ccopy 2012 RAS

2 F S Liu et al

any evolution However some recent studies from the ultravio-let luminosities infrared emission or line emission show increas-ing evidence for ongoing star formation and post-starbursts insome BCGs (eg Allen 1995 Cardiel et al 1998 Crawford etal1999 Edge 2001 Hicks amp Mushotzky 2005 McNamara et al2006 Egami et al 2006 Wilman et al 2006 Edwards et al 2007OrsquoDea et al 2008 Cavagnolo et al 2008 Pipino et al 2009OrsquoDea et al 2010 Liu et al 2012) The existence of blue cores andUV excess in some BCGs are also interpreted as evidence for on-going star formation (eg Bildfell et al 2008 Pipino et al 2009Wang et al 2010 Hicks et al 2010) Although active star forma-tion in these BCGs is compelling the starbursts may have veryshort timescales (shorter than 200 Myr) and only contributea smallmass fraction (less than 1 percent Pipino et al 2009)

The BCGs with ongoing SF studied in previous works aremostly selected from X-ray cluster samples and usually residein cooling flow clusters It has been shown that star formationin these BCGs is correlated with the cooling timescale (tcool) ofthe gas (Rafferty et al 2008) which is a strong indicator of theirconnections However previous studies are based on small sam-ples and are biased toward X-ray luminous clusters which maynot be a representative of this population It has also been shownnearby optically-selected local BCGs (eg zlt01) have little in-dication for enhanced active star formation (Edwards et al2007von der Linden et al 2007 Wang et al 2010) In this study wesearch for BCGs with ongoing SF in clusters at higher redshiftWe select a sample of 120 early-type BCGs at 01 lt z lt 04from two large optically-selected cluster catalogues of SDSS-WHL(Wen et al 2009) and GMBCG (Hao et al 2010) This sample isroughly an order of magnitude larger than previous ones They areselected with strong emission lines in their optical spectra withboth Hα and [O II]λ3727 line emission which indicates significantongoing star formation We investigate their statistical propertiesand make a comparison with a control sample selected from X-rayluminous clusters For the first time we probe the dependence ofSF activities in these BCGs on their stellar masses and cluster en-vironments We also reconstruct their star formation history usingstellar population synthesis models and discuss their physical con-nections with the cooling flows and galactic cannibalism

The structure of the paper is as follows We describe our sam-ple selection and data analysis insect2 andsect3 and present our resultsin sect4 A summary and discussion are given insect5 Throughout thispaper we adopt a cosmology with a matter density parameterΩm =

03 a cosmological constantΩΛ = 07 and a Hubble constant ofH0 = 70 km sminus1Mpcminus1 ieh = H0(100 km sminus1Mpcminus1) = 07

2 SAMPLE SELECTION

We identify early-type BCGs with significant star formationfromtwo large optically-selected cluster catalogues of SDSS-WHL(Wen et al 2009) and GMBCG (Hao et al 2010) The SDSS-WHLcluster catalogue was constructed from the SDSS DR6 photomet-ric galaxy catalogue which includes 39668 clusters in theredshiftrange 005 lt z ltsim 06 with more than eight luminous (Mr 6 minus21)member galaxies within a radius of 05 Mpc and a photometric red-shift intervalzplusmn004(1+z) The GMBCG catalogue was constructedfrom the SDSS DR7 photometric catalogue by identifying the redsequence plus BCG feature which includes 55424 clusters in theredshift range 01 lt z lt 055 We use the data at 01 lt z lt 04 be-cause both catalogues are relatively complete out toz sim 04 Thereis another cluster catalogue by Szabo et al (2011) we do notuse

this sample since it is not yet public at the time of writing Thesethree cluster samples overlap but also differ in their lists of clustersSDSS-WHL and GMBCG samples give fully consistent results (seebelow) concerning SF activities and thus our conclusions shouldnot be much affected by which catalogue we use

In our selection we require BCGs to have spectroscopicobservations and their spectra have been parameterised by theMPAJHU team1 We discard the BCGs with concentration indexC = R90R50 lt 25 in the iminusband (to select early-type objectsBernardi et al 2003) and those with a median signal-to-noise ratio(SN) per pixel of the whole spectrum smaller than 3 As a resultwe obtain 10996 and 15181 early-type BCGs at 01 lt z lt 04from the SDSS-WHL and GMBCG catalogues respectively

Baldwin et al (1981) proposed a suite of three diagnostic di-agrams to classify the dominant energy source in emission-linegalaxies which are commonly known as the Baldwin-Phillips-Terlevich (BPT) diagrams and are based on four emission lineratios [OIII]λ5007Hβ [NII] λ6584Hα [SII]λ6716+6731Hα and[OI]λ6300Hα Here we only use the emission line diagnostic di-agram of [OIII]λ5007Hβ versus [NII]λ6584Hα because the otherlines ([SII]λ67166731 and [OI]λ6300) are usually very weak inearly-type galaxies (Huang amp Gu 2009) We select the emission-line BCGs requiring (1) the lines of [NII]λ6584 Hα [OIII] λ5007Hβ and [O II]λ3727 are detected as emission lines and havethe SNgt3 (2) the equivalent widths (EWs) of both Hα and [OII] λ3727 lines are greater than 3Aring It should be noted that someprevious studies (eg Crawford et al 1999 Donahue et al2010)have shown that the line emission in BCGs with large equivalentwidth of Hα is dominated by star formation The cut of Hα EWgt 3Aring is thus strongest in our criteria to select sources with SFac-tivities The cut of [O II] EWgt 3Aring only rejectsim 1 of sources withlarge Hα EW which does not affect our statistical result Howeverthe inclusion of the [O II]λ3727 line here allows us to investigatethe origin of [O II]λ3727 line emission in star-forming BCGs (seesect31) Our criteria inevitably reject weak emission-line BCGs Itis acceptable since the ability to distingush their types bythe BPTdiagram will be poor (eg Hao et al 2005) In total 159 and201objects satisfy our criteria in the selected SDSS-WHL and GMBCGearly-type BCGs respectively The fraction issim 14 (15910996)for SDSS-WHL samplesim 13 (20115181) for GMBCG sam-ple respectively

The diagnostic diagram mentioned above then classifies theseemission-line BCGs into different types which are shown in Fig-ure 1 We here detect 3 purely star-forming objects 59 composites(SF+ AGN) and 98 AGNs in the SDSS-WHL objects (see the leftpanel of Figure 1) and 7 purely star-formings objects 71 com-posites and 123 AGNs in the GMBCG objects (right panel) Weselect those classified as purely star-forming objects or compos-ites as our targets which constitute a very rare populationrelativeto the whole sample (6210996sim 056 for SDSS-WHL objects7815181 sim 05 for GMBCG objects) Notice that 20 of thesetargets overlap in these two catalogues Thus we finally obtain atotal of 120 early-type BCGs with significant ongoing star forma-tion (9 purely star-forming objects and 111 composites) which arelisted in Table 1

We also identify a control BCG sample from X-ray luminousclusters TheROSAT All Sky Survey detected 18806 bright sources(Voges et al 1999) and 105924 faint sources (Voges et al 2000) inthe 01ndash24 keV band of which 378 extended sources in the north-

1 httpwwwmpa-garchingmpgdeSDSSDR7

ccopy 2012 RAS MNRAS000 1ndash15

Star formation activities in early-type BCGs 3

Figure 1 The diagnostic diagrams of [NII]λ6584Hα versus [OIII]λ5007Hβ for emission-line early-type BCGs in the SDSS-WHL catalogue (left panel) andGMBCG catalogue (right panel) respectively The purely star-forming galaxies composites and AGNs are shown with solid circles open circles and crossesrespectively The sources in X-ray luminous clusters are shown with red symbols The solid line is from Kewley et al (2006) and the dashed line is fromKauffmann et al (2003)

ern hemisphere and 447 extended sources in the southern hemi-sphere have been identified as clusters of galaxies (Bohringer et al2000 2004) However many objects in recent catalogues of SDSSclusters may also be unidentified X-ray clusters (Wen et al 2009)since SDSS have detected many new clusters of galaxies We fol-low Wen et al (2009) to cross-identify theROSAT X-ray bright andfaint sources with two spectroscopic catalogues to construct a newsample of X-ray cluster candidates We first select X-ray sourceswith a projected separation ofrp lt 03 Mpc from the BCGs andhardness ratios of 0ndash1 as our targets (Wen et al 2009) In total weobtain 112 targets in theROSAT bright source catalog and 194 tar-gets in theROSAT faint source catalog for SDSS-WHL clustersrespectively We also obtain 125 targets in theROSAT bright sourcecatalog and 236 targets in theROSAT faint source catalog for GM-BCG clusters respectively

We first cross-correlate selected emission-line BCGs withthese X-ray candidates There are 24 SDSS-WHL emission-lineBCGs and 25 GMBCG emission-line BCGs inROSAT brightsource catalog respectively Only 3 emission-line BCGs and5 GM-BCG emission-line BCGs are found in theROSAT faint sourcecatalog The fractions of emission-line BCGs relative to the X-ray luminous samples (24112 sim 21 for SDSS-WHL objects25125sim 20 for GMBCG objects) are almost one order of mag-nitude higher than that in optically-selected sample (sim 14) Thederived fraction in X-ray luminous sample (sim 20) is slightlylower than previous results (eg 27 from Crawford et al 199922 from Donahue et al 2010) The difference may be a re-sult of different selection criteria or sample sizes Donahue et al(2010) analysed a small sample (with 32 objects) and the detectedemission-line BCGs by Hα EW gt 1Aring showed the typical forbid-den line emission as well Our incidence rate is close to their re-sult Crawford et al (1999) used a relatively large statistical sam-ple (with 256 central dominant galaxies) and selected emission-linegalaxies based on Hα line emission only If we select emission-lineBCGs by Hα emission line only and apply the same Hα luminos-ity detection limits as Crawford et al (1999) namely extrapolatingtheir slope to our redshift range according to the expectedL prop z2

relation (see Crawford et al 1999 for details) and correcting the

Figure 2 Top panels the incidence rates (fractions) of identified emission-line BCGs (dots connected with dashed line) and BCGs with SF (dots con-nected with solid line) as a function of cluster richness forSDSS-WHLclusters (left) and GMBCG clusters (right) respectivelyBottom panels thenormalised distributions of cluster richness of optically-selected samples(dots connected with dashed line) and X-ray luminous sample(dots con-nected with solid line) for SDSS-WHL clusters (left) and GMBCG clusters(right) respectively Poisson errors are shown

difference in cosmological parameters As a result We obtain 29SDSS-WHL Hα emitters and 35 GMBCG Hα emitters in X-rayluminous clusters The fractions (29112 sim 26 for SDSS-WHLobjects 35125 sim 28 for GMBCG objects) are almost the samewith that of Crawford et al (1999)

We then cross-correlate those BCGs with SF with these X-raycandidates There are 11 SDSS-WHL BCGs with SF and 10 GM-

4 F S Liu et al

BCG BCGs with SF inROSAT bright source catalog Notice that 8sources with SF overlap in these two X-ray samples Therefore 13out of a total of 120 early-type BCGs with SF are likely to be inX-ray luminous clusters In fact 11 of these 13 sources are known X-ray luminous clusters according to the NASAIPAC ExtragalacticDatabase (NED) which have been indicated in Table 1 The frac-tions of BCGs with SF relative to the whole X-ray luminous sam-ples (11112sim 98 for SDSS-WHL objects 10125sim 80 forGMBCG objects) are also one order of magnitude higher than thatin optically-selected sample (sim 05) We show the fractions ofthese BCGs with SF (solid line) and selected emission-line BCGs(dashed line) as a function of cluster richenss for the SDSS-WHLobjects (top-left panel) and GMBCG objects (top-right panel) inFigure 2 respectively Notice that the relations with cluster richnessfor sources in these two catalogues are shown separately becausethe cluster richness in these catalogues is estimated with differentalgorithms It can be seen that the more massive clusters tend tohabour higher fractions of emission-line BCGs and SF BCGs Thefractions are usually the highest in the richest clusters It indicatesthat the incidence rates of emission-line BCGs and SF BCGs inacluster sample may be much higher above some minimum clusterrichness (mass) It can thus be understood that the incidence ratesof emission-line BCGs and BCGs with SF are higher in X-ray lu-minous clusters than optically-selected ones since X-ray selectedclusters are usually more massive (see bottom panels of Figure 2)

3 DATA ANALYSIS

31 SFR estimates

It has been known that Hα emission is sensitive to the most recentstar formation The SFR based on Hα emission is an indicator ofthe nearly instantaneous SFR since it is produced by ionization bythe hottest and youngest stars We derive the SFRs of our targetBCGs by the Hα line following Kennicutt (1998)

SFR(Hα) = 79times 10minus42LHα M⊙yrminus1 (1)

whereLHα is the extinction-corrected luminosity of Hα emissionin units of 1042 ergs sminus1 The derived SFRs(Hα) range from 016M⊙yr to 1299M⊙yr with an average value of 77M⊙yr

We also estimate their SFRs by the [OII]λ3727 line and makea comparison with SFRs(Hα) In order to estimate the contributionsby AGN on the [O II] emission we follow Wang amp Wei (2008)to investigate the luminosity of [O III] as a function of the lineratio of [O II][O III] for our BCGs (see Figure 3) The symbolsare the same as in Figure 1 except that BCGs in X-ray luminousclusters are shown with red symbols Notice that hereafter we donot show the overlapping objects in the GMBCG catalog in thetotal sample It is clear that our composite BCGs show enhanced[O II] [O III] ratios just like purely star-forming BCGs as these twotypes of objects reside in the same region

We can estimate the [O II] luminosities emitted from the HIIregions by assuming the enhanced [O II][O III] ratios are causedby star formations (Kim et al 2006) The mean value of the [OII] [O III] ratios is sim 70 for our BCGs Given that the average [OIII] luminosity L[OIII] sim 78times 1040 ergs sminus1 the [O II][O III] ratiopredicted by the regression line (see Figure 3) given by Kim et al(2006) is onlysim 043 It means that on averagesim 92 of the [O II]emission for our BCGs can be attributed to star formation

We use the recent calibration of Kewley et al (2004)

Figure 3 The line ratio [O II][O III] versus the luminosity of [O III]The solid line shows the least-square regression for type I (narrow line)AGNs given by Kim et al (2006) The symbols are the same as in Figure 1Our targets clearly show enhanced [O II][O III] line ratios relative to theKim et al (2006) line

Figure 4 The comparison between the estimated SFR(Hα) from theHα line and SFR([O II]) from the [O II] line The symbols are the sameas in Figure 1 It shows that these two estimates are roughly consistent witheach other

SFR(O II)= 79timesC1timesL[O II] 42

1673minus 175[log(OH) + 12]M⊙ yrminus1 (2)

to estimate the SFR(O II) whereL[OII] 42 is the extinction-correctedluminosity of [O II] emission in units of 1042 ergs sminus1 C1 is the cor-rection factor due to the enhanced [O II][O III] ratio The metal-licity is fixed to be log(OH)+ 12= 89 corresponding to the solar

Table 1Basic parameters for the 120 identified early-type BCGs withongoing star formation

Cluster Name X-ray Cluster BCG RA BCG Dec BCGz logMlowastfib logMlowasttot f[NII] λ6584 fHα f[OIII] λ5007 fHβ f[OII] λ3727 SFR(Hα) SFR([OII]) Dn(4000) HδA Type

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17)

WHLJ0047213+005239 1183866 087768 01025 1066 1125 24118 39726 11832 14505 37228 084 056 172 -088 compWHLJ0154359+002641 2864954 044486 02288 1081 1141 3124 5085 1411 1774 7805 062 073 193 -321 compWHLJ0242536minus065742a 4072351 -696174 03503 1108 1167 2780 3263 1234 1138 3541 107 085 175 -446 compWHLJ0806414+494628 12169920 4979065 02434 1089 1165 14948 199036860 7128 43216 281 501 158 -100 compWHLJ0817281+065903a 12436710 698433 02565 1088 1147 2096 4557 616 20574150 073 050 185 -182 sbWHLJ0854112+190702a 13354660 1911724 01813 1094 1153 18132 26325 119939583 65258 192 386 185 -322 compWHLJ0909406+105005 13741920 1083483 01402 1079 1131 94718 187365 63554 71886 824088 777 2911 145 312 compWHLJ0920188+370618a 14007820 3710510 02348 1079 1155 22131 22238 5769 8022 42254 290 452 178 075 compWHLJ0922437+351448 14068201 3524673 02307 1097 1146 70167 139030 23058 51079 114188 1740 1187 141 396 compWHLJ0926094+670407 14153909 6706886 01211 1097 1140 184734 303438 51255 111611 268482 917 667 141 185 compWHLJ0929543+002752 14247610 046474 01457 1051 1124 6035 8282 1835 2919 5832 037 017 169 131 compWHLJ1016521+135938 15421809 1397771 01455 1054 1115 13706 239313882 8537 33511 108 120 152 -075 compWHLJ1023396+041110ab Z3146 15591521 418628 02897 1106 1173 465055 617213125103 224514 1089693 12985 19685 118 494 compWHLJ1028318+151511a 15713229 1525325 03046 1092 1159 14372 13680 2965 4920 21256 323 411 176 014 compWHLJ1035455+152435 15893961 1540980 02582 1111 1157 11678 144835463 5179 34051 234 450 189 072 compWHLJ1049498+054629 16245770 577494 02640 1110 1174 27952 37897 11506 13919 130880 644 1884 186 -021 compWHLJ1051588+082222 16299510 837297 01884 1094 1136 10895 13129 3950 4691 22786 104 142 179 -159 compWHLJ1113205+173541ab A1204 16833540 1759474 01705 1062 1139 29318 29827 6469 10725 68154 190 358 173 -026 compWHLJ1121546+305515 17047729 3092111 02432 1088 1143 3767 4362 1678 1536 5162 061 052 165 147 compWHLJ1125368+592155 17140331 5936544 03104 1103 1132 76228 110091 15679 39699 61592 2715 1257 121 358 compWHLJ1126128minus005130 17155341 -085854 02553 1110 1173 11673 136702873 4865 22130 215 283 179 -024 compWHLJ1127149+482220 17181219 4837249 01679 1082 1135 10622 244629986 8691 29373 151 137 157 080 compWHLJ1135207+491127 17383611 4919104 01314 1061 1111 14734 254674364 9206 26507 092 073 148 082 compWHLJ1139571+681118 17487270 6817191 01543 1084 1145 64791 8992218426 32760 132314 460 558 151 284 compWHLJ1212562+272657ab 18324640 2745126 01797 1037 1137 47091 48982 4403 17600 46133 350 270 175 -070 compWHLJ1222050minus013609 18552090 -160274 02005 1085 1155 34295 4934520899 18158 114112 450 862 166 -104 compWHLJ1252300+035803 19312500 396769 01942 1090 1142 24915 47929 9708 17467 56697 408 392 154 226 compWHLJ1313304+320039a 19838640 3203464 03042 1098 1164 3419 5393 2510 1919 14849 127 282 192 -048 compWHLJ1314517+383418a 19871539 3857187 02360 1089 1153 6715 9261 4578 3296 23494 122 247 199 -153 compWHLJ1324146+041803b RX J13242+0419 20108200 431862 02631 1062 1148 23338 22490 2491 8019 38076 379 538 169 068 compWHLJ1328583minus005343 20227600 -089491 02373 1100 1152 5589 7110 2368 2516 10855 095 111 163 156 compWHLJ1353215+395909a 20833971 3998603 01057 1075 1125 4823 9643 2078 3365 12341 022 019 197 -168 compWHLJ1357428+303505 20932860 3059559 02085 1085 1131 44809 104045 12538 37933 92091 1037 762 136 350 sbWHLJ1401021+025242ab A1835 21025861 287847 02520 1118 1175 246191 33904267954 122692 526839 5181 6871 115 542 compWHLJ1415200+240036 21381760 2402097 01386 1032 1119 7694 8780 920 3070 9100 036 027 190 -111 compWHLJ1418355+020507a 21464799 208549 02697 1094 1145 28008 39497 7862 14519 41203 705 603 138 188 compWHLJ1418377+374624a 21465691 3777348 01349 1068 1133 7626 12925 3546 4515 16096 049 044 190 -254 compWHLJ1423555+262623ab RX J14239+2626 21598109 2643979 01482 1056 1136 34525 39916 8542 14330 78969 187 303 164 038 compWHLJ1424243+251427 21610139 2524108 02331 1063 1139 35165 378227603 13626 52047 485 549 144 334 compWHLJ1433409minus014503 21842059 -175099 02194 1090 1152 9088 11979 2976 4257 9748 134 080 185 -219 compWHLJ1440488+150625 22014830 1513004 01141 1041 1104 89017 236553 23756 86568 169703 628 368 131 341 sbWHLJ1446212+381525 22158850 3825720 02344 1081 1131 41779 8498525357 31654 71750 1103 754 145 377 compWHLJ1452134+053857 22305569 564942 02303 1102 1149 14702 19547 9266 6986 27304 244 266 170 -065 compWHLJ1457151+222034ab MS 14550+2232 22431300 2234288 02576 1096 1168 51778 60164 6233 21592 116608 967 1594 155 346 compWHLJ1504075minus024816b RXC J15041-0248 22603130 -280460 02169 1106 1157 423969 779580 202740 278825 1346217 8491 12545 116 498 compWHLJ1513038+252550 22826570 2543079 01835 1076 1124 26715 4271717700 15721 25942 320 140 160 085 compWHLJ1532538+302059b RX J15328+3021 23322411 3034983 03620 1099 1157 209019 28435553197 102270 485162 10034 14685 118 611 compWHLJ1539153+422950 23481390 4249735 02335 1099 1149 9926 12047 4395 4270 6906 155 060 156 167 compWHLJ1546066+120650a 23652740 1211406 01849 1095 1151 13732 17202 7383 6190 52295 131 325 205 -297 compWHLJ1612283+113547 24309700 1159213 02716 1119 1177 158266 244776 79020 91663 583431 4440 9044 134 -132 compWHLJ1616217+441914 24409019 4432083 01951 1093 1139 14992 233206458 8539 21813 200 143 170 010 compWHLJ1620442+125214 24518410 1287080 01904 1096 1142 6221 9090 4016 3219 17738 074 110 173 -311 compWHLJ1623021+475939 24575150 4796708 01993 1083 1137 20754 322698634 11960 58209 291 429 157 119 compWHLJ1623094+440441 24578931 4407832 01333 1089 1157 109792 185304 84461 68920 445219 689 1381 170 014 compWHLJ1639364+370501a 24990150 3708373 01826 1057 1115 26264 50106 6652 18096 36331 371 214 138 191 compWHLJ1700463+222141 25519141 2239863 02003 1111 1154 36865 9582661092 35442 203318 873 1533 165 043 compWHLJ1720100+263732ab RX J17202+2637 26004181 2662557 01601 1040 1140 53732 65943 6768 23429 86097 366 397 157 171 compWHLJ1724230+273242 26109579 2754511 02330 1087 1152 32610 454738177 16637 166645 582 1816 183 -175 compWHLJ2101013minus070019 31525549 -700530 01363 1084 1141 7476 9610 2969 3394 19798 038 059 185 -143 compWHLJ2129400+000521ab RX J21296+0005 32241650 008921 02339 1038 1128 21852 23198 2675 8207 33273 300 356 183 -038 compWHLJ2154033+000950 32851370 016403 02144 1091 1140 101989 190520 60840 70294 131039 2022 1130 123 566 compWHLJ2323161+005922 35081711 098978 01092 1067 1134 6445 10368 1680 3649 11299 025 019 194 -280 comp

value The derived SFRs(O II) ranges from 012M⊙yr to 1969M⊙yr with an average value of 109M⊙yr

The comparison between the SFR(Hα) and SFR([O II]) forour BCGs is shown in Figure 4 which shows that these two esti-mates are consistent with each other and most of the [OII]λ3727line emission in star-forming BCGs can be attributed to starforma-tion It should be noted that our SFR estimates are from the SDSSspectra within 3primeprime fiber diameter which corresponds to an averagesize ofsim 10 kpc for our BCGs at 01 lt z lt 04 It has been known

that the majority of the line emissions in BCGs may be containedwithin this aperture (eg Hatch et al 2007) We thus do notcor-rect the aperture effect for our BCGs as in previous studies (egCrawford et al 1999 OrsquoDea et al 2008)

32 Spectral synthesis

We use the spectral synthesis codestarlight (Cid Fernandes et al2005) to derive stellar populations of our BCGsstarlight fits the

6 F S Liu et al

Table 1 - continued

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17)

GMBCGJ05750891+0006656 5750892 006657 01314 1092 1142 6805 10365 3325 3634 12435 037 031 184 -205 compGMBCGJ12187813+3401156 12187810 3401156 02079 1055 1150 32736 31873 7328 11570 84925 316 704 180 -237 compGMBCGJ12222965+4681917 12222970 4681917 01259 1073 1125 44518 60343 7165 21951 45293 198 116 149 196 compGMBCGJ12988700+2726509 12988699 2726509 02839 1120 1168 19518 28661 15087 10591 130843 575 2222 208 -340 compGMBCGJ13054583+5992378 13054581 5992378 01278 1088 1135 173454 385276 36060 141817 182902 1308 506 144 424 sbGMBCGJ13553378+3726415 13553380 3726415 02931 1110 1167 21491 22436 5781 8094 24733 485 432 183 -153 compGMBCGJ13855769+0311440 13855769 311440 01420 1090 1129 30965 58208 18550 21230 62594 248 208 138 227 compGMBCGJ13936943+0770896 13936940 770896 01297 1075 1134 5306 6690 2649 2347 12161 023 030 199 -259 compGMBCGJ14061017+5189892 14061020 5189892 02016 1090 1148 32818 45100 6318 16538 50611 417 384 159 162 compGMBCGJ14135050+1733344 14135049 1733344 02145 1090 1142 6179 10795 3228 3781 13672 115 111 196 -049 compGMBCGJ14168858+2956935 14168860 2956935 02026 1095 1142 4054 42192029 2828 3345 039 019 177 -224 compGMBCGJ14256414+2522437 14256410 2522437 03167 1131 1168 7043 97673270 3548 20558 252 434 195 -162 compGMBCGJ14515501+2034164 14515500 2034164 02474 1109 1159 3618 44611685 1571 10398 065 119 192 -156 compGMBCGJ14995279+1263987 14995280 1263987 02162 1085 1140 32232 53442 27116 20012 229238 578 2095 174 167 compGMBCGJ15029590+3396550 15029590 3396550 01997 1090 1147 5425 73822503 2618 13925 067 097 191 -313 compGMBCGJ15082925+0639045 15082930 639045 01841 1076 1137 33675 51635 17762 19177 92844 390 574 178 -144 compGMBCGJ15196927+2750322 15196930 2750323 01482 1101 1146 107010 233407 28003 85657 178168 1093 689 146 247 sbGMBCGJ15624967+3994290 15624969 3994291 03097 1110 1162 3149 3040747 1195 5566 075 107 188 -111 compGMBCGJ15685843+4796615 15685851 4796611 01317 1072 1142 11166 15828 2413 5577 20105 057 056 196 -150 compGMBCGJ16238906+2415104 16238910 2415105 02488 1100 1169 21109 26986 10921 9914 92085 401 1146 199 -147 compGMBCGJ16259096+3718131 16259100 3718132 01691 1089 1142 14498 26550 5931 9583 18026 166 083 168 -049 compGMBCGJ16659455+0923113 16659450 923113 02220 1087 1136 6782 111334442 3974 20595 128 186 172 -001 compGMBCGJ16706517+2319665 16706520 2319666 02223 1097 1147 4243 58701616 2086 12449 068 113 192 -341 compGMBCGJ16905610+5702011 16905611 5702011 01624 1071 1124 27035 44503 3874 16052 24449 255 109 161 090 compGMBCGJ17668298+5120676 17668300 5120676 02792 1099 1154 3787 47641916 1769 7900 092 117 190 034 compGMBCGJ17831993+2351160 17831990 2351162 02580 1105 1155 40135 55953 22172 20252 135689 903 1839 156 134 compGMBCGJ18132281+3382819 18132280 3382819 01834 1066 1118 3470 61022047 2174 4238 046 020 165 026 compGMBCGJ18310487+4603118 18310490 4603118 03433 1092 1154 2800 5230965 1849 4566 163 109 197 024 compGMBCGJ18768381+3655121 18768381 3655122 01173 1068 1130 35208 71368 6796 25930 40545 201 088 158 080 sbGMBCGJ18785496+1952685 18785500 1952687 03289 1107 1168 7887 12396 3470 4402 20949 350 484 210 -140 compGMBCGJ18824063+4290074 18824060 4290074 01681 1106 1159 80524 124919 54801 47221 442496 771 2311 177 -099 compGMBCGJ19118836+3324089 19118840 3324089 01298 1048 1112 3332 45771199 1731 6528 016 015 194 -116 compGMBCGJ19158395+3291053 19158400 3291054 01334 1074 1143 10301 15865 7353 5593 24464 059 066 197 -244 compGMBCGJ19283454+4835832 19283450 4835832 01479 1068 1125 33459 72667 18538 26061 82362 339 307 141 122 compGMBCGJ19365889+0703035 19365891 703036 03444 1110 1157 1842 3032 643 1072 2259 095 051 156 -089 compGMBCGJ20134167+0398025 20134171 398024 02548 1104 1164 27714 32276 10099 11648 33703 506 422 163 -011 compGMBCGJ20482740+3385023 20482739 3385023 02543 1096 1148 10613 33140 9469 12034 41261 517 523 162 174 sbGMBCGJ21045775+1650052 21045770 1650051 02194 1106 1152 25091 30832 4352 10959 49936 345 464 181 -216 compGMBCGJ21077165+1005610 21077170 1005609 01202 1050 1113 4478 53202154 1862 6736 016 012 177 -100 compGMBCGJ21506365+0157007 21506360 157007 01905 1090 1136 8138 113223462 4003 11432 092 068 174 -124 compGMBCGJ21574059+2734423 21574060 2734423 01596 1100 1145 22644 32550 10267 11882 99877 179 455 195 -261 compGMBCGJ21645610+1090254 21645610 1090254 02400 1103 1155 3399 35421432 1668 3770 048 035 184 -032 compGMBCGJ22185842+0847364b 22185840 847365 03755 1127 1180 123013 159509 27250 56813 239022 6130 7852 125 388 compGMBCGJ22457182+5096220 22457179 5096219 02726 1090 1141 4129 89674608 3189 22336 164 329 201 017 compGMBCGJ22462785+5803770 22462790 5803770 02819 1114 1166 7840 10668 4370 3833 23500 211 376 179 -135 compGMBCGJ22503764+2165720 22503760 2165720 01503 1091 1131 67611 190993 16311 70753 130182 923 519 137 368 sbGMBCGJ22956532+5868779 22956531 5868780 01877 1085 1148 4192 52551847 1911 4994 041 026 183 -214 compGMBCGJ22995329+2636325 22995329 2636326 01206 1064 1110 21019 47762 5502 17228 19796 143 042 151 082 sbGMBCGJ23061689+0575409 23061690 575410 01825 1103 1138 7857 101463329 3605 22387 075 131 202 -374 compGMBCGJ23533215+1881746 23533220 1881746 03052 1130 1178 15007 16498 3805 5891 29447 391 578 192 -244 compGMBCGJ24100574+2988749 24100571 2988749 02927 1109 1161 2136 43401613 1598 3640 093 054 174 030 compGMBCGJ24518841+2631922 24518840 2631922 02267 1095 1155 15090 20061 9914 7331 83461 241 838 210 -314 compGMBCGJ24535810+2462761 24535809 2462761 01878 1075 1132 8467 11888 3504 4258 11164 094 064 165 -059 compGMBCGJ25106514+4586376 25106509 4586376 01547 1085 1135 4633 55812306 1947 8021 029 028 198 -110 compGMBCGJ25886623+5628854 25886621 5628856 02903 1094 1136 23054 38325 5637 13795 20020 810 336 132 122 compGMBCGJ25971676+3173790 25971680 3173790 01890 1042 1111 4974 51511121 1889 5951 041 034 167 -066 compGMBCGJ31589493+0912085 31589490 912084 01458 1086 1136 84536 147122 52645 52597 292953 665 1101 194 -085 compGMBCGJ35527875+0030927b RX J23411+0018 35527869 030927 02768 1080 1158 250419 294764 82403 107783 611038 5584 9899 128 770 comp

NoteCol(1) SDSS-WHL Cluster Name or GMBCG Cluster NameCol(2) 11 known X-ray luminous clusters out of 13 candidatesCol(3) BCG RA(J20000) in units of degreesCol(4) BCG Dec(J20000) in units of degreesCol(5) The spectroscopic redshift of the BCGCol(6) Logarithm of the stellar mass inside the fiber aperture in units ofM⊙ Col(7) Logarithm of the total stellar mass in units ofM⊙ Col(8) The flux of [NII]λ6584 line in units of 10minus17erg sminus1cmminus2Col(9) The flux of Hα line in units of 10minus17erg sminus1cmminus2Col(10) The flux of [OIII]λ5007 line in units of 10minus17erg sminus1cmminus2Col(11) The flux of Hβ line in units of 10minus17erg sminus1cmminus2Col(12) The flux of [OII]λ3727 line in units of 10minus17erg sminus1cmminus2Col(13) The derived SFR(Hα) by the Hα emission line in units ofM⊙yrCol(14) The derived SFR([OII]) by the [OII] emission linein units ofM⊙yrCol(15) The amplitude of the 4000 Balmer breakDn(4000)Col(16) The absorption line index HδACol(17) Galaxy type classified by the BPT diagnosic diagrams rsquosbrsquominusstarburst rsquocomprsquominuscompositeThe SDSS-WHL objects marked with lsquoarsquo are also in the GMBCG catalogueThe objects marked with lsquobrsquo are the identified 13 candidates of X-ray luminous clustersThe flux of all emission lines used in this paper have been applied the dust correcton

Figure 5 The spectral synthesis of an example BCG WHLJ1504075-024816 (RXC J15041-0248) The observed spectrumOλ (green) model spectrumMλ(red) and error spectrum (blue) ofOλ are shown in the top left panel respectively The residual spectrumEλ (purple) is shown in the bottom left panel Theflux intensities in the left two panels are both normalised at4020Aring by 45times 10minus16 ergs sminus1 cmminus2 The light and mass weighted stellar population fractionsxj

andmicroj are shown in the top right and bottom right panels respectively Several derived quantities (see text for details) from the fitting are shown at the topright corner

observed spectrumOλ with a model spectrumMλ which is madeup of a pre-defined set of base spectra It carries out the fitting witha simulated annealing plus Metropolis scheme to yield the mini-mumχ2 =

λ[(Oλ minus Mλ)wλ]2 wherewminus1λ is the error inOλ at each

wavelength It modelsMλ by a combination

Mλ = Mλ(~x AV v⋆ σ⋆) =N⋆sum

x jγ jλrλ (3)

whereγ jλ equiv bλ j otimesG(v⋆ σ⋆) bλ j equiv( Bλ j

Bλ0 j

is the normalised flux of

the jth spectrumBλ j is the jth component of base spectrumBλ0 j isthe value of thejth base spectrum at the normalisation wavelengthλ0 G(v⋆ σ⋆) is the Gaussian distribution centred at velocityv⋆ andσ⋆ is the line-of-sight velocity dispersionx j is the fraction of fluxdue to componentj atλ0 rλ equiv 10minus04(AλminusAV ) is the global extinctionterm represented byAV The residual spectrumEλ including emis-sion lines can be obtained by subtracting the model spectrumfromthe observed one asEλ = Oλ minus Mλ

In this work we take simple stellar populations (SSPs) fromthe BC03 evolutionary synthesis models (Bruzual amp Charlot 2003)as our base spectra We adopt the spectral templates withNlowast=42 SSPs ndash 3 metallicities (Z=02 1 and 25Z⊙) and 14 ages(3 5 10 25 40 100 280 500 900 Myr and 14 25 5 1013 Gyr) computed with the ldquoPadova 1994rdquo evolutionary tracks(Alongi et al 1993 Bressan et al 1993 Fagotto et al 1994abGirardi et al 1996) and the Chabrier (2003) initial mass function(IMF) We follow Meng et al (2010) to model the extinction usingthe dust extinction law given by Calzetti et al (1994 2000)andCalzetti (1997) We show a typical example of spectral fitting forour BCGs in Figure 5 The top left panel shows the observed spec-

trum Oλ (green) and the modelMλ (red) The bottom left panelgives the residual spectrumEλ = Oλ minus Mλ (purple) The light-weighted stellar population fractionsxj are shown in the top rightpanel The mass-weighted population fractionsmicroj are shown in thebottom right panelSTARLIGHT presents the current stellar mass and the fraction

of each stellar component Following Cid Fernandes et al (2005)we can derive the mean ages of the stellar population weighted bythe flux and stellar mass respectively

〈logt⋆〉L =N⋆sum

x j log t j (4)

wherex j is the fraction of flux contributed by certain SSP and

〈logt⋆〉M =N⋆sum

micro j log t j (5)

wheremicro j is the fraction of stellar mass contributed by each SSPThe results of the population synthesis are presented insect42

4 RESULTS

41 Dependence on stellar mass and environment

The stellar mass for each BCG inside the fiber aperture has beenestimated from our spectral synthesis which are roughly consis-tent with that obtained by fits on the photometry by the MPAJHUteam The specific SFR (inside the fiber aperture) for our BCGscan thus be derived The total stellar mass can be obtained bymul-tiplying the factorC2 equiv 10minus04(mpetrominusmfiber) between the fiber mag-

8 F S Liu et al

Figure 6 The derived total SFR and specific SFR (SSFR) by the Hα line versus the total stellar mass (logMlowasttot) and cluster richness respectively Thesymbols are the same as in Figure 1 except that 8 BCGs in knowncooling flow clusters are shown as red boxes The size of each box is inversely proportionalto its cooling time (tcool) (see Sec 44) At the top right corner of each panel we showthe correlation coefficient and corresponding significance level for thenull hypothesis of no correlation as given by the Spearman-Rank order test

nitudes and total (Petrosian) magnitudes We follow the methodof MPAJHU team to take the correction factor averaged over thefive SDSS bands weighted by 1∆C2

2 where∆C2 is the error inthe correction factorC2 (estimated from the errors in the photom-etry) The estimated SFR and specific SFR by the Hα line versusthe galaxy total stellar mass (log Mlowasttot) and the cluster richnessfor our BCGs are shown in Figure 6 (the relations are similar if theSFR estimated by the [O II] line is used) respectively The symbolsare the same as in Figure 1 except that 8 BCGs in known coolingflow clusters are shown as red boxes The size of each box is in-versely proportional to its cooling time (tcool) Notice that the rest5 BCGs in X-ray luminous sample are also likely to be in cooling

flow clusters (seesect44 for discussions) We perform the Spearman-Rank order correlation test for each relation The correspondingcorrelation coefficient and the significance level of the null hypoth-esis that there is no correlation are given in each panel of Figure 6 Itcan be seen that there is an obvious trend that more massive BCGswith SF and those in richer clusters tend to have higher SFR andspecific SFR but with large scatters BCGs with SF in X-ray lumi-nous clusters are often located in the densest environment and havethe highest SFR and specific SFR which shows they appear to beforming stars at a higher rate BCGs with SF in cooling flow clus-ters (red boxes) usually have the most active star formation(alsoseesect44)

Table 2 The normalised percentage of flux and mass contributed on average by stellar populations of different age for the whole optically-selected 120 BCGswith SF 13 BCGs with SF in X-ray luminous clusters 200 optically-selected quiescent BCGs and 116 quiescent (without distinct emission-lines) BCGs inX-ray luminous clusters respectively The 1σ (683) confidence interval for each average value is given in the parentheses

SSP SF (optical) SF (X-ray) quiescent(optical) quiescent(X-ray)

fburst (t lt 01Gyr) 763 (0001365) 1837 (1013211) 138 (000540) 316 (000712)fyoung (t lt 05Gyr) 1992 (3913910) 3977 (5356120) 1068 (0001734) 924 (0001796)

fmiddle(05Gyrlt t lt 25Gyr) 2431 (1254473) 2480 (6125004) 2053 (0004235) 2244 (0003654)fold(t gt 25Gyr) 5577 (22757705) 3544 (17225856) 6879 (42928118) 6832 (48238891)

mburst (t lt 01Gyr) 004 (000006) 017 (000023) 0002 (0000005) 0012 (0000016)myoung (t lt 05Gyr) 051 (000187) 168 (000297) 021 (000059) 013 (000034)

mmiddle(05Gyrlt t lt 25Gyr) 1173 (0002836) 1907 (2004012) 465(000960) 572 (0001221)mold(t gt 25Gyr) 8776 (57359306) 7925 (43239725) 9514(88209760) 9414 (84069701)

42 Star formation history

We investigate the star formation history of these BCGs withSFthrough spectral synthesis method We combine 42 SSPs into threeages young- middle- and old-age stellar populations Theyoung-age stellar population includes the SSPs with age less than 05 Gyrthe old-age population is SSPs with age larger than 25 Gyr andthe intermediate-age population is the SSPs between them We alsoshow a burst population defined as SSPs with age less than 01Gyrconnected with the most recent star formation activity We make adirect comparison between these BCGs with SF and a randomly-selected sample of 200 quiescent BCGs (without emission lines ofboth Hα and [O II] ie sources with the EWs of Hα and [O II]6 0 excluded at the 954 confidence level) in MPAJHU cata-logues These two samples are matched in total stellar massWehave shown that the majority (sim 80) of BCGs in two X-ray lumi-nous samples are not classified as emission-line BCGs They havealso no significant emission lines (lsquoquiescentrsquo) We make anothercontrol sample with 116 objects The normalised fractions of theirstellar populations are listed in Table 2 which provide a coarse starformation history of these BCGs

The flux-weighted average population fraction is sensitivetothe star formation activity Nearly 20 flux are from the young stel-lar population for the whole sample This fraction can increase tosim40 for BCGs with SF in X-ray luminous clusters A large frac-tion (sim40) of the young stellar population is from the recent burst(with age tlt 01Gyr) BCGs with SF and quiescent BCGs havecomparable fraction of intermediate-age stellar population whichindicates that SF activities in BCGs contribute mainly to the frac-tion of young stellar population and the timescale of SF activity isshort However the fractions of stellar mass in different age bins aresignificantly different from the flux-weighted ones The majority ofstellar mass of BCGs with SF is contributed by the old populationThe stellar mass of the young stellar population is small (sim 05 onaverage) which is consistent with the result of Pipino et al (2009)obtained from a smaller sample It shows that the stellar popula-tion will not have significant differences with that of normal BCGswhen their star formation is quenched The stellar population ofBCGs with SF are still predominantly old not very different withthe quiescent (normal) BCGs

Notice that aboutsim 12 of stars formed within the last 25Gyr or so for our BCGs with SF (the first column in Table 2) Thederived stellar mass inside the fiber aperture is 82 times 1010M⊙ onaverage If the typical SFR (sim 77M⊙ yrminus1) in bursts is the sameas the current one then the total star formation duration will be13 Gyr This is consistent with a scenario that the rejuvenation SFactivities in these BCGs may be sporadic (eg Schawinski et al

Figure 7 Top panel the Balmer line absorption index HδA plotted againstthe 4000Aring break strengthDn(4000) for the BCGs with SF and quiescentBCGs (blue dots) respectively Bottom panel specific SFR (SSFR) versusthe 4000Aring break strengthDn(4000) for the BCGs with SF The symbols arethe same as in Figure 1 The BCGs in X-ray luminous clusters are shownwith red symbols in particular 8 BCGs in known cooling flow clusters areshown as red boxes The sizes of boxes are inversely proportional to theircooling times (tcool)

10 F S Liu et al

2007) In any case it should be emphasized that the scattersin theseestimates are quite high (see Table 2)

It has been shown by Kauffmann et al (2003) that the planedefined by the 4000Aring break strengthDn(4000) and Balmer lineabsorption index HδA is also a powerful diagnostic for the starformation history of galaxies We show the HδA absorption indexas a function ofDn(4000) for these BCGs with SF in the top panelof Figure 7 and compare with quiescent BCGs (blue dots) We alsoshow the relation of the specific SFR versusDn(4000) for our BCGswith SF in the bottom panel of Figure 7 The values ofDn(4000)and HδA are taken from the MPAJHU catalog As can be seenBCGs with more active SF activities (higher specific SFR) tendto have lowerDn(4000) values and stronger HδA absorption thanquiescent BCGs It means they have a higher fraction of youngstars and are more likely to be experiencing sporadic star forma-tion events at the present day (Kauffmann et al 2003) This analy-sis is consistent with our results obtained through spectral synthesismethod

43 SF activity amp cooling flow

It has been shown that active star formation in BCGs may be con-nected with the cooling flow of intracluster medium (ICM) in manyX-ray clusters (eg Rafferty et al 2008 and reference therein)There are 11 known BCGs in our 13 targets in X-ray luminous clus-ters We collect their information from the literature andfind that 8out of 11 (sim 73) have been identified to be in cooling flow clusters(eg Dunn amp Fabian 2008 Rafferty et al 2008) The rest-frameoptical spectra and colour images of BCGs in these eight clus-ters (RXC J15041-0248 Zw 3146 A1835 RX J15328+3021 MS14550+2232 RX J17202+2637 RX J21296+0005 A1204) areshown in Figure 8 ordered with increasing cooling time of ICMTheir cooling time tcool are obtained from Rafferty et al (2008)andor Bauer et al (2005) which are marked in each panel of Fig-ure 8 Although not all objects have measured cooling time andthe derived values are somewhat different for the same source fromthe two studies it is still apparent in Figure 8 that the fourBCGswith shortest cooling times have bluer colours which is consis-tent with the result of Rafferty et al (2008) that bluer BCGs residein clusters with shorter cooling times These eight BCGs in cool-ing flow clusters are shown with red boxes in the plot of specificSFR versusDn(4000) (the bottom panel of Figure 7) with the sizeof each box inversely proportional to the cooling time It can beseen that BCGs in cooling flow clusters usually have more activeSF activities (higher specific SFR) the four BCGs (RXC J15041-0248 Zw 3146 A1835 RX J15328+3021) with the shortest tcool

have the most active SF activities In fact it has been showntwo(Zw 3146 and A1835) of them can even be classified as luminousinfrared galaxies (LIRGs) (Egami et al 2006) This is a strong in-dicator that the cooling flow and star formation in these BCGsareconnected Figure 8 also shows that BCGs in cooling flow clusters(in particular the four with the shortest tcool) usually have very flatoptical spectra (smaller 4000Aring break strength see the bottom panelof Figure 7)

It should be noted that we do not know whether or not theremaining 5 targets in our X-ray luminous sample are also in cool-ing flow clusters Their rest-frame spectra and colour images areshown in Figure 9 It can be seen that their images and opti-cal spectra are very similar to those of 8 BCGs in known cool-ing flow clusters particularly for two GMBCG objects (GMBCGJ22185842+0847364 J35527875+0030927) These two objectshave extremely blue images and very flat optical spectra with ex-

tremely high SFR amp SSFR (two red circles on the top of right pan-els of Figure 6) In any case the majority (even 100) of BCGswith SF in X-ray luminous sample are likely to be in cooling flowclusters

5 SUMMARY amp DISCUSSION

Only a few BCGs have been reported with ongoing star formationin previous studies and the majority of them are identified from X-ray cluster samples In this paper we identify a large sample of 120early-type BCGs at 01 lt z lt 04 from two large optically-selectedcluster catalogues of SDSS-WHL (Wen et al 2009) and GMBCG(Hao et al 2010) Their optical spectra show strong emission linesof both [O II]λ3727 and Hα indicating significant ongoing star for-mation This sample is not biased toward X-ray luminous clustersand is thus more representative of this population We investigatetheir statistical properties and make a comparison with a controlsample selected from X-ray luminous clusters We also investigatetheir star formation history using stellar population synthesis mod-els The main results can be summarised as follows

(i) The incidence rates of emission-line BCGs and BCGs withSF in X-ray luminous clusters are almost one order of magnitudehigher than those in optically-selected clusters

(ii) More massive BCGs with SF in richer clusters tend to havehigher SFR and specific SFR which shows they appear to be form-ing stars at a higher rate BCGs with SF in X-ray luminous clustersusually have more active SF activities

(iii) The star formation history of BCGs with SF can be welldescribed by a recent minor and short starburst superimposed onan old stellar component with the recent episode of star formationcontributinglt 1 percent of the total stellar mass (sim 05 on aver-age) The star formation history may be episodic lasting a substan-tial fraction of the time in the last 25 Gyr (see Table 2 and section42)

(iv) BCGs with SF in cooling flow clusters usually have very flatoptical spectrum and the most active SF activities Star formationin these BCGs and cooling flow are correlated

Although the short SF activity appears to be a rather commonphenomenon during the evolution of BCGs the source of the coldgas required to fuel the star formation is unclear The correlationbetween SF activities in BCGs in cooling flow clusters and thegascooling timescale (Rafferty et al 2008) suggests a clear (cooling)origin of the cold gas in these systems However the majority of ourstar-forming BCGs have less active SF activities They either lie innon-cooling flow (non-X-ray luminous) clusters or may be belowthe ROSAT detection limit Recent theoretical studies often assumea very efficient form of AGN feedback which may suppress the starformation completely On the other hand Bildfell et al (2008) sug-gest that AGN feedback may not fully compensate the energy lostvia radiative cooling allowing the gas to cool at a reduced rate Itremains to be seen how universal this mode operates in BCGs andwhat kind of SFR it can sustain Another possible and attractivemechanism is through the known galactic cannibalism that appearsfrequently in cluster environments due to dynamical friction Weindeed find a large fraction of sample BCGs in non-X-ray luminousclusters with distinct merger features (see 30 examples2 in Fig-ure 10) If there is some remaining cold gas in the captured satellite

2 Colour images and corresponding spectra for all 120 target BCGs areavailable at httpwwwjbmanacuk˜ smaoliutargz

Figure 8 Spectra and colour images of the eight known early-type BCGswith significant ongoing SF in cooling flow clusters Each spectrum is shifted tothe rest-frame wavelength corrected for the Galactic extinction and smoothed using a 15 Aring box The size of each colour image corresponds to 200 kpc by200 kpc The objects are ordered with increasing cooling time (tcool) in units of 108yr taken from Rafferty et al (2008 R08) andor Bauer et al (2005 B05)

12 F S Liu et al

Figure 9 Spectra and colour images of the other 5 early-type BCGs withongoing SF in 13 X-ray luminous clusters Each spectrum is shifted to the rest-framewavelength corrected for the Galactic extinction and smoothed using a 15 Aring box The size of each colour image corresponds to 200 kpc by 200 kpc

galaxy then the merger may supply fresh cold gas (sim few 108M⊙)that can trigger a new episode of star formation (Poole et al2006)It will be interesting to explore this issue further using our samplein a future study

ACKNOWLEDGMENTS

We thank X Y Xia Z G Deng Z L Wen Cheng Li Lin Yan JWang for useful discussions and comments We acknowledge theanonymous referee for a constructive report that much improvedthe paper This project is supported by the NSF of China 11103013

SM and XMM acknowledge the Chinese Academy of Sciences forfinancial support

Funding for the creation and distribution of the SDSS Archivehas been provided by the Alfred P Sloan Foundation the Partici-pating Institutions the National Aeronautics and Space Adminis-tration the National Science Foundation the US Department ofEnergy the Japanese Monbukagakusho and the Max Planck Soci-ety The SDSS Web site is httpwwwsdssorg The SDSS is man-aged by the Astrophysical Research Consortium (ARC) for thePar-ticipating Institutions The Participating Institutionsare The Uni-versity of Chicago Fermilab the Institute for Advanced Study the

Figure 10 Colour images of 30 sample early-type BCGs with distinct merger features in non-X-ray luminous clusters The size of each colour imagecorresponds to 200 kpc by 200 kpc

Japan Participation Group The Johns Hopkins University the Ko-rean Scientist Group Los Alamos National Laboratory the Max-Planck-Institute for Astronomy (MPIA) the Max-Planck-Institutefor Astrophysics (MPA) New Mexico State University University

of Pittsburgh Princeton University the United States Naval Obser-vatory and the University of Washington

14 F S Liu et al

REFERENCES

Allen S W 1995 MNRAS 276 947Alongi M Bertelli G Bressan A Chiosi C Fagotto F GreggioL Nasi E 1993 AampAS 97 851

Baldwin J A Phillips M M Terlevich R 1981 PASP 93 5Bauer F E Fabian A C Sanders J S Allen S W JohnstoneR M 2005 MNRAS 359 1481

Bernardi M Hyde J B Sheth R K Miller C J Nichol R C2007 AJ 133 1741

Bernardi M Sheth R K Annis J Burles S Eisenstein D JFinkbeiner D P Hogg D W Lupton R H Schlegel D J Sub-baRao M Bahcall N A Blakeslee J P 2003 AJ 125 1817

Bildfell C Hoekstra H Babul A Mahdavi A 2008 MNRAS389 1637

Bohringer H Schuecker P Guzzo L Collins C A Voges WCruddace R G Ortiz-Gil A Chincarini G De Grandi S EdgeA C MacGillivray H T Neumann D M Schindler S ShaverP 2004 AampA 425 367

Bohringer H Voges W Huchra J P McLean B Giacconi RRosati P Burg R Mader J Schuecker P Simic D KomossaS Reiprich T H Retzlaff J Trumper J 2000 ApJS 129 435

Bressan A Fagotto F Bertelli G Chiosi C 1993 AampAS 100647

Bruzual G Charlot S 2003 MNRAS 344 1000Cardiel N Gorgas J Aragon-Salamanca A 1998 ApampSS 26383

Cavagnolo K W Donahue M Voit G M Sun M 2008 ApJ682 821

Chabrier G 2003 PASP 115 763Cid Fernandes R Mateus A Sodre L Stasinska G GomesJ M 2005 MNRAS 358 363

Crawford C S Allen S W Ebeling H Edge A C Fabian A C1999 MNRAS 306 857

De Lucia G Blaizot J 2007 MNRAS 375 2Desroches L Quataert E Ma C West A A 2007 MNRAS377 402

Donahue M Bruch S Wang E Voit G M Hicks A KHaarsma D B Croston J H Pratt G W Pierini D OrsquoConnellR W Bohringer H 2010 ApJ 715 881

Dressler A 1980 ApJ 236 351Dunn R J H Fabian A C 2008 MNRAS 385 757Edge A C 2001 MNRAS 328 762Edwards L O V Hudson M J Balogh M L Smith R J 2007MNRAS 379 100

Egami E Misselt K A Rieke G H Wise M W NeugebauerG Kneib J Le Flocrsquoh E Smith G P 2006 ApJ 647 922

Fagotto F Bressan A Bertelli G Chiosi C 1994a AampAS104365

Fagotto F Bressan A Bertelli G Chiosi C 1994b AampAS10529

Gao L Loeb A Peebles P J E White S D M Jenkins A2004 ApJ 614 17

Girardi L Bressan A Chiosi C Bertelli G Nasi E 1996AampAS 117 113

Graham A Lauer T R Colless M Postman M 1996 ApJ 465534

Hao J McKay T A Koester B P Rykoff E S Rozo E AnnisJ Wechsler R H Evrard A Siegel S R Becker M Busha MGerdes D Johnston D E Sheldon E 2010 ApJS 191 254

Hao L Strauss M A Tremonti C A Schlegel D J HeckmanT M Kauffmann G Blanton M R Fan X Gunn J E 2005

AJ 129 1783Hatch N A Crawford C S Fabian A C 2007 MNRAS 38033

Hicks A K Mushotzky R 2005 ApJ 635 L9Hicks A K Mushotzky R Donahue M 2010 ApJ 719 1844Huang S Gu Q 2009 MNRAS 398 1651Jeltema T E Mulchaey J S Lubin L M Fassnacht C D 2007ApJ 658 865

Jones C Forman W 1984 ApJ 276 38Kauffmann G Heckman T M Tremonti C Brinchmann JCharlot S White S D M Ridgway S E Brinkmann JFukugita M Hall P B IvezicZ Richards G T SchneiderD P 2003 MNRAS 346 1055

Kauffmann G Heckman T M White S D M Charlot STremonti C Brinchmann J Bruzual G Peng E W Seibert MBernardi M Blanton M Brinkmann J Castander F Csabai I2003 MNRAS 341 33

Kauffmann G Heckman T M White S D M Charlot STremonti C Peng E W Seibert M Brinkmann J Nichol R CSubbaRao M York D 2003 MNRAS 341 54

Kauffmann G White S D M Heckman T M Menard BBrinchmann J Charlot S Tremonti C Brinkmann J 2004MNRAS 353 713

Kennicutt Jr R C 1998 ARAampA 36 189Kewley L J Geller M J Jansen R A 2004 AJ 127 2002Kewley L J Groves B Kauffmann G Heckman T 2006 MN-RAS 372 961

Kim M Ho L C Im M 2006 ApJ 642 702Lauer T R 1988 ApJ 325 49Lauer T R Faber S M Richstone D Gebhardt K TremaineSPostman M Dressler A Aller M C Filippenko A V GreenR Ho L C Kormendy J Magorrian J Pinkney J 2007 ApJ662 808

Liu F S Mao S Deng Z G Xia X Y Wen Z L 2009 MN-RAS 396 2003

Liu F S Wen Z L Han J L Meng X M 2012 ScChG 55354

Liu F S Xia X Y Mao S Wu H Deng Z G 2008 MNRAS385 23

Matthews T A Morgan W W Schmidt M 1964 ApJ 140 35McIntosh D H Guo Y Hertzberg J Katz N Mo H J van denBosch F C Yang X 2008 MNRAS 388 1537

McNamara B R Rafferty D A Bırzan L Steiner J WiseM W Nulsen P E J Carilli C L Ryan R Sharma M 2006ApJ 648 164

Meng X Wu H Gu Q Wang J Cao C 2010 ApJ 718 928Mulchaey J S Lubin L M Fassnacht C Rosati P JeltemaT E2006 ApJ 646 133

OrsquoDea C P Baum S A Privon G Noel-Storr J Quillen A CZufelt N Park J Edge A Russell H Fabian A C DonahueM Sarazin C L McNamara B Bregman J N Egami E2008 ApJ 681 1035

OrsquoDea K P Quillen A C OrsquoDea C P Tremblay G R SniosB T Baum S A Christiansen K Noel-Storr J Edge A CDonahue M Voit G M 2010 ApJ 719 1619

Oemler A 1973 ApJ 180 11Oemler Jr A 1976 ApJ 209 693Patel P Maddox S Pearce F R Aragon-Salamanca A ConwayE 2006 MNRAS 370 851

Pipino A Kaviraj S Bildfell C Babul A Hoekstra H Silk J2009 MNRAS 395 462

Poole G B Fardal M A Babul A McCarthy I G Quinn T

Wadsley J 2006 MNRAS 373 881Rafferty D A McNamara B R Nulsen P E J 2008 ApJ 687899

Rines K Finn R Vikhlinin A 2007 ApJ 665 L9Schawinski K Thomas D Sarzi M Maraston C Kaviraj SJooS-J Yi S K Silk J 2007 MNRAS 382 1415

Schombert J M 1986 ApJS 60 603Schombert J M 1987 ApJS 64 643Schombert J M 1988 ApJ 328 475Smith G P Kneib J Smail I Mazzotta P Ebeling H CzoskeO 2005 MNRAS 359 417

Stott J P Collins C A Burke C Hamilton-Morris V SmithG P 2011 MNRAS 414 445

Szabo T Pierpaoli E Dong F Pipino A Gunn J 2011 ApJ736 21

Thomas D Maraston C Bender R Mendes de Oliveira C 2005ApJ 621 673

Tran K Moustakas J Gonzalez A H Bai L Zaritsky DKautsch S J 2008 ApJ 683 L17

Tran K van Dokkum P Franx M Illingworth G D KelsonD D Schreiber N M F 2005 ApJ 627 L25

van Dokkum P G Franx M Fabricant D Kelson D D Illing-worth G D 1999 ApJ 520 L95

Voges W Aschenbach B Boller T Brauninger H Briel UBurkert W Dennerl K 2000 VizieR Online Data Catalog9029 0

Voges W Aschenbach B Boller T Brauninger H Briel UBurkert W Dennerl K Englhauser J Gruber R Haberl FHartner G 1999 AampA 349 389

von der Linden A Best P N Kauffmann G White S D M2007 MNRAS 379 867

Wang J Overzier R Kauffmann G von der Linden A Kong X2010 MNRAS 401 433

Wang J Wei J Y 2008 ApJ 679 86Wen Z L Han J L Liu F S 2009 ApJS 183 197Whiley I M Aragon-Salamanca A De Lucia G von der LindenA Bamford S P Best P Bremer M N Jablonka P JohnsonO Milvang-Jensen B Noll S Poggianti B M Rudnick GSaglia R White S Zaritsky D 2008 MNRAS 387 1253

Wilman R J Edge A C Swinbank A M 2006 MNRAS 37193

Yan R Newman J A Faber S M Konidaris N Koo D DavisM 2006 ApJ 648 281

1 Introduction

2 Sample Selection

3 Data Analysis

31 SFR estimates


4 Results




5 Summary amp Discussion

2 F S Liu et al

any evolution However some recent studies from the ultravio-let luminosities infrared emission or line emission show increas-ing evidence for ongoing star formation and post-starbursts insome BCGs (eg Allen 1995 Cardiel et al 1998 Crawford etal1999 Edge 2001 Hicks amp Mushotzky 2005 McNamara et al2006 Egami et al 2006 Wilman et al 2006 Edwards et al 2007OrsquoDea et al 2008 Cavagnolo et al 2008 Pipino et al 2009OrsquoDea et al 2010 Liu et al 2012) The existence of blue cores andUV excess in some BCGs are also interpreted as evidence for on-going star formation (eg Bildfell et al 2008 Pipino et al 2009Wang et al 2010 Hicks et al 2010) Although active star forma-tion in these BCGs is compelling the starbursts may have veryshort timescales (shorter than 200 Myr) and only contributea smallmass fraction (less than 1 percent Pipino et al 2009)

The BCGs with ongoing SF studied in previous works aremostly selected from X-ray cluster samples and usually residein cooling flow clusters It has been shown that star formationin these BCGs is correlated with the cooling timescale (tcool) ofthe gas (Rafferty et al 2008) which is a strong indicator of theirconnections However previous studies are based on small sam-ples and are biased toward X-ray luminous clusters which maynot be a representative of this population It has also been shownnearby optically-selected local BCGs (eg zlt01) have little in-dication for enhanced active star formation (Edwards et al2007von der Linden et al 2007 Wang et al 2010) In this study wesearch for BCGs with ongoing SF in clusters at higher redshiftWe select a sample of 120 early-type BCGs at 01 lt z lt 04from two large optically-selected cluster catalogues of SDSS-WHL(Wen et al 2009) and GMBCG (Hao et al 2010) This sample isroughly an order of magnitude larger than previous ones They areselected with strong emission lines in their optical spectra withboth Hα and [O II]λ3727 line emission which indicates significantongoing star formation We investigate their statistical propertiesand make a comparison with a control sample selected from X-rayluminous clusters For the first time we probe the dependence ofSF activities in these BCGs on their stellar masses and cluster en-vironments We also reconstruct their star formation history usingstellar population synthesis models and discuss their physical con-nections with the cooling flows and galactic cannibalism

The structure of the paper is as follows We describe our sam-ple selection and data analysis insect2 andsect3 and present our resultsin sect4 A summary and discussion are given insect5 Throughout thispaper we adopt a cosmology with a matter density parameterΩm =

03 a cosmological constantΩΛ = 07 and a Hubble constant ofH0 = 70 km sminus1Mpcminus1 ieh = H0(100 km sminus1Mpcminus1) = 07

2 SAMPLE SELECTION

We identify early-type BCGs with significant star formationfromtwo large optically-selected cluster catalogues of SDSS-WHL(Wen et al 2009) and GMBCG (Hao et al 2010) The SDSS-WHLcluster catalogue was constructed from the SDSS DR6 photomet-ric galaxy catalogue which includes 39668 clusters in theredshiftrange 005 lt z ltsim 06 with more than eight luminous (Mr 6 minus21)member galaxies within a radius of 05 Mpc and a photometric red-shift intervalzplusmn004(1+z) The GMBCG catalogue was constructedfrom the SDSS DR7 photometric catalogue by identifying the redsequence plus BCG feature which includes 55424 clusters in theredshift range 01 lt z lt 055 We use the data at 01 lt z lt 04 be-cause both catalogues are relatively complete out toz sim 04 Thereis another cluster catalogue by Szabo et al (2011) we do notuse

this sample since it is not yet public at the time of writing Thesethree cluster samples overlap but also differ in their lists of clustersSDSS-WHL and GMBCG samples give fully consistent results (seebelow) concerning SF activities and thus our conclusions shouldnot be much affected by which catalogue we use

In our selection we require BCGs to have spectroscopicobservations and their spectra have been parameterised by theMPAJHU team1 We discard the BCGs with concentration indexC = R90R50 lt 25 in the iminusband (to select early-type objectsBernardi et al 2003) and those with a median signal-to-noise ratio(SN) per pixel of the whole spectrum smaller than 3 As a resultwe obtain 10996 and 15181 early-type BCGs at 01 lt z lt 04from the SDSS-WHL and GMBCG catalogues respectively

Baldwin et al (1981) proposed a suite of three diagnostic di-agrams to classify the dominant energy source in emission-linegalaxies which are commonly known as the Baldwin-Phillips-Terlevich (BPT) diagrams and are based on four emission lineratios [OIII]λ5007Hβ [NII] λ6584Hα [SII]λ6716+6731Hα and[OI]λ6300Hα Here we only use the emission line diagnostic di-agram of [OIII]λ5007Hβ versus [NII]λ6584Hα because the otherlines ([SII]λ67166731 and [OI]λ6300) are usually very weak inearly-type galaxies (Huang amp Gu 2009) We select the emission-line BCGs requiring (1) the lines of [NII]λ6584 Hα [OIII] λ5007Hβ and [O II]λ3727 are detected as emission lines and havethe SNgt3 (2) the equivalent widths (EWs) of both Hα and [OII] λ3727 lines are greater than 3Aring It should be noted that someprevious studies (eg Crawford et al 1999 Donahue et al2010)have shown that the line emission in BCGs with large equivalentwidth of Hα is dominated by star formation The cut of Hα EWgt 3Aring is thus strongest in our criteria to select sources with SFac-tivities The cut of [O II] EWgt 3Aring only rejectsim 1 of sources withlarge Hα EW which does not affect our statistical result Howeverthe inclusion of the [O II]λ3727 line here allows us to investigatethe origin of [O II]λ3727 line emission in star-forming BCGs (seesect31) Our criteria inevitably reject weak emission-line BCGs Itis acceptable since the ability to distingush their types bythe BPTdiagram will be poor (eg Hao et al 2005) In total 159 and201objects satisfy our criteria in the selected SDSS-WHL and GMBCGearly-type BCGs respectively The fraction issim 14 (15910996)for SDSS-WHL samplesim 13 (20115181) for GMBCG sam-ple respectively

The diagnostic diagram mentioned above then classifies theseemission-line BCGs into different types which are shown in Fig-ure 1 We here detect 3 purely star-forming objects 59 composites(SF+ AGN) and 98 AGNs in the SDSS-WHL objects (see the leftpanel of Figure 1) and 7 purely star-formings objects 71 com-posites and 123 AGNs in the GMBCG objects (right panel) Weselect those classified as purely star-forming objects or compos-ites as our targets which constitute a very rare populationrelativeto the whole sample (6210996sim 056 for SDSS-WHL objects7815181 sim 05 for GMBCG objects) Notice that 20 of thesetargets overlap in these two catalogues Thus we finally obtain atotal of 120 early-type BCGs with significant ongoing star forma-tion (9 purely star-forming objects and 111 composites) which arelisted in Table 1

We also identify a control BCG sample from X-ray luminousclusters TheROSAT All Sky Survey detected 18806 bright sources(Voges et al 1999) and 105924 faint sources (Voges et al 2000) inthe 01ndash24 keV band of which 378 extended sources in the north-

1 httpwwwmpa-garchingmpgdeSDSSDR7

4 F S Liu et al

3 DATA ANALYSIS

31 SFR estimates

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17)

6 F S Liu et al

Table 1 - continued

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17)

x jγ jλrλ (3)

Bλ0 j

x j log t j (4)

micro j log t j (5)

4 RESULTS

8 F S Liu et al

10 F S Liu et al

12 F S Liu et al

ACKNOWLEDGMENTS

14 F S Liu et al

REFERENCES

1 Introduction

2 Sample Selection

3 Data Analysis

31 SFR estimates


4 Results





4 F S Liu et al

3 DATA ANALYSIS

31 SFR estimates

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17)

6 F S Liu et al

Table 1 - continued

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17)

x jγ jλrλ (3)

Bλ0 j

x j log t j (4)

micro j log t j (5)

4 RESULTS

8 F S Liu et al

10 F S Liu et al

12 F S Liu et al

ACKNOWLEDGMENTS

14 F S Liu et al

REFERENCES

1 Introduction

2 Sample Selection

3 Data Analysis

31 SFR estimates


4 Results





4 F S Liu et al

3 DATA ANALYSIS

31 SFR estimates

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17)

6 F S Liu et al

Table 1 - continued

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17)

x jγ jλrλ (3)

Bλ0 j

x j log t j (4)

micro j log t j (5)

4 RESULTS

8 F S Liu et al

10 F S Liu et al

12 F S Liu et al

ACKNOWLEDGMENTS

14 F S Liu et al

REFERENCES

1 Introduction

2 Sample Selection

3 Data Analysis

31 SFR estimates


4 Results





(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17)

6 F S Liu et al

Table 1 - continued

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17)

x jγ jλrλ (3)

Bλ0 j

x j log t j (4)

micro j log t j (5)

4 RESULTS

8 F S Liu et al

10 F S Liu et al

12 F S Liu et al

ACKNOWLEDGMENTS

14 F S Liu et al

REFERENCES

1 Introduction

2 Sample Selection

3 Data Analysis

31 SFR estimates


4 Results





6 F S Liu et al

Table 1 - continued

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17)

x jγ jλrλ (3)

Bλ0 j

x j log t j (4)

micro j log t j (5)

4 RESULTS

8 F S Liu et al

10 F S Liu et al

12 F S Liu et al

ACKNOWLEDGMENTS

14 F S Liu et al

REFERENCES

1 Introduction

2 Sample Selection

3 Data Analysis

31 SFR estimates


4 Results





x jγ jλrλ (3)

Bλ0 j

x j log t j (4)

micro j log t j (5)

4 RESULTS

8 F S Liu et al

10 F S Liu et al

12 F S Liu et al

ACKNOWLEDGMENTS

14 F S Liu et al

REFERENCES

1 Introduction

2 Sample Selection

3 Data Analysis

31 SFR estimates


4 Results





8 F S Liu et al

10 F S Liu et al

12 F S Liu et al

ACKNOWLEDGMENTS

14 F S Liu et al

REFERENCES

1 Introduction

2 Sample Selection

3 Data Analysis

31 SFR estimates


4 Results





10 F S Liu et al

12 F S Liu et al

ACKNOWLEDGMENTS

14 F S Liu et al

REFERENCES

1 Introduction

2 Sample Selection

3 Data Analysis

31 SFR estimates


4 Results





10 F S Liu et al

12 F S Liu et al

ACKNOWLEDGMENTS

14 F S Liu et al

REFERENCES

1 Introduction

2 Sample Selection

3 Data Analysis

31 SFR estimates


4 Results





12 F S Liu et al

ACKNOWLEDGMENTS

14 F S Liu et al

REFERENCES

1 Introduction

2 Sample Selection

3 Data Analysis

31 SFR estimates


4 Results





12 F S Liu et al

ACKNOWLEDGMENTS

14 F S Liu et al

REFERENCES

1 Introduction

2 Sample Selection

3 Data Analysis

31 SFR estimates


4 Results





14 F S Liu et al

REFERENCES

1 Introduction

2 Sample Selection

3 Data Analysis

31 SFR estimates


4 Results





14 F S Liu et al

REFERENCES

1 Introduction

2 Sample Selection

3 Data Analysis

31 SFR estimates


4 Results





1 Introduction

2 Sample Selection

3 Data Analysis

31 SFR estimates


4 Results





Documents

arXiv:1203.1840v1 [astro-ph.GA] 8 Mar 2012 · Rines et al. 2007; McIntosh et al. 2008; Liu et al. 2009), although some studies of BCGs in the more distant universe disagree with this