The 10k zCOSMOS: morphological transformation of galaxies in the group environment since z~1
K. Kovac, S. J. Lilly, C. Knobel, M.Bolzonella, A. Iovino, C.M. Carollo, C. Scarlata, M. Sargent, O. Cucciati, G. Zamorani, L. Pozzetti, L.A.M. Tasca, M. Scodeggio, P. Kampczyk, Y. Peng, P. Oesch, E. Zucca, A. Finoguenov, T. Contini, J.-P. Kneib, O. Le Fevre, V. Mainieri, A. Renzini, S. Bardelli, A. Bongiorno, K. Caputi, G. Coppa, S. de la Torre, L. de Ravel, P. Franzetti, B. Garilli, F. Lamareille, J.-F. Le Borgne, V. Le Brun, C. Maier, M. Mignoli, R. Pello, E. Perez Montero, E. Ricciardelli, J.D. Silverman, M. Tanaka, L. Tresse, D. Vergani, U. Abbas, D. Bottini, A. Cappi, P. Cassata, A. Cimatti, M. Fumana, L. Guzzo, A.M. Koekemoer, A. Leauthaud, D. Maccagni, C. Marinoni, H. J. McCracken, P. Memeo, B. Meneux, C. Porciani, R. Scaramella, N.Z. Scoville
aa r X i v : . [ a s t r o - ph . C O ] S e p Draft version October 29, 2018
Preprint typeset using L A TEX style emulateapj v. 03/07/07
THE 10K ZCOSMOS: MORPHOLOGICAL TRANSFORMATION OF GALAXIES IN THE GROUPENVIRONMENT SINCE z ∼ K. Kovaˇc , S. J. Lilly , C. Knobel , M. Bolzonella , A. Iovino , C. M. Carollo , C. Scarlata , M. Sargent ,O. Cucciati , G. Zamorani , L. Pozzetti , L. A. M. Tasca , M. Scodeggio , P. Kampczyk , Y. Peng ,P. Oesch , E. Zucca , A. Finoguenov , T. Contini , J.-P. Kneib , O. Le F`evre , V. Mainieri , A. Renzini ,S. Bardelli , A. Bongiorno , K. Caputi , G. Coppa , S. de la Torre , L. de Ravel , P. Franzetti ,B. Garilli , F. Lamareille , J.-F. Le Borgne , V. Le Brun , C. Maier , M. Mignoli , R. Pello , E. PerezMontero , E. Ricciardelli , J. D. Silverman , M. Tanaka , L. Tresse , D. Vergani , U. Abbas ,D. Bottini , A. Cappi , P. Cassata , A. Cimatti , M. Fumana , L. Guzzo , A. M. Koekemoer ,A. Leauthaud , D. Maccagni , C. Marinoni , H. J. McCracken , P. Memeo , B. Meneux ,C. Porciani , R. Scaramella , N. Z. Scoville Draft version October 29, 2018
ABSTRACTWe study the evolution of galaxies inside and outside of the group environment since z = 1 usinga large well defined set of groups and galaxies from the zCOSMOS-bright redshift survey in theCOSMOS field. The fraction of galaxies with early-type morphologies increases monotonically with M B luminosity and stellar mass and with cosmic epoch. It is higher in the groups than elsewhere,especially at later epochs. The emerging environmental effect is superposed on a strong global mass-driven evolution, and at z ∼ . M ∗ /M ⊙ ) ∼ .
2, the “effect” of group environment isequivalent to (only) about 0.2 dex in stellar mass or 2 Gyr in time. The stellar mass function of galaxiesin groups is enriched in massive galaxies. We directly determine the transformation rates from lateto early morphologies, and for transformations involving colour and star-formation indicators. Thetransformation rates are systematically about twice as high in the groups as outside, or up to 3-4times higher correcting for infall and the appearance of new groups. The rates reach values, formasses around the crossing mass 10 . M ⊙ , as high as 0.3 - 0.7 Gyr − in the groups, implyingtransformation timescales of 1.4 - 3 Gyr, compared with less than 0.2 Gyr − , i.e. timescales > M ⊙ which we cannot well probe. The rates involving colourand star-formation are consistently higher than those for morphology, by a factor of about 50%. Ourconclusion is that the transformations which drive the evolution of the overall galaxy population since z ∼ Subject headings: galaxies: clusters: general - galaxies: evolution - galaxies: high-redshift - galaxies:luminosity function, mass function - galaxies: structure Based on observations obtained at the European Southern Ob-servatory (ESO) Very Large Telescope (VLT), Paranal, Chile, aspart of the Large Program 175.A-0839 (the zCOSMOS Spectro-scopic Redshift Survey) Institute of Astronomy, ETH Z¨urich, CH-8093, Z¨urich,Switzerland; [email protected] INAF Osservatorio Astronomico di Bologna, via Ranzani 1, I-40127, Bologna, Italy INAF Osservatorio Astronomico di Brera, Milan, Italy Spitzer Science Center, 314-6 Caltech, Pasadena, CA 91125,USA Max-Planck-Institut f¨ur Astronomie, K¨oningstuhl 17, D-69117Heidelberg, Germany Laboratoire d’Astrophysique de Marseille, Marseille, France INAF - IASF Milano, Milan, Italy Max-Planck-Institut f¨ur extraterrestrische Physik, D-84571Garching, Germany Laboratoire d’Astrophysique de Toulouse-Tarbes, Universitede Toulouse, CNRS, 14 avenue Edouard Belin, F-31400 Toulouse,France European Southern Observatory, Karl-Schwarzschild- Strasse2, Garching, D-85748, Germany INAF - Osservatorio Astronomico di Padova, Padova, Italy Dipartimento di Astronomia, Universita di Padova, Padova,Italy INAF Osservatorio Astronomico di Torino, Strada Osservato-rio 20, I-10025 Pino Torinese, Torino, Italy Dept. of Astronomy, University of Massachusetts at Amherst Dipartimento di Astronomia, Universit´a di Bologna, via Ran-zani 1, I-40127, Bologna, Italy INTRODUCTION
Galaxies exhibit a range of morphologies that reflect,at least in part, basic structural differences, such asthe presence of disks, spheroids and bars, as well asdifferences in the visibility of the spiral arms due tofor instance the rate of star-formation. The commonlyused morphological classification follows the Hubble se-quence of elliptical, S0, spiral, barred spiral and ir-regular galaxies, with various subclasses (Hubble 1926, Space Telescope Science Institute, 3700 San Martin Drive,Baltimore, MD 21218 Physics Division, MS 50 R5004, Lawrence Berkeley NationalLaboratory, 1 Cyclotron Rd., Berkeley, CA 94720, USA Centre de Physique Theorique, Marseille, Marseille, France Institut d’Astrophysique de Paris, UMR 7095 CNRS, Univer-sit´e Pierre et Marie Curie, 98 bis Boulevard Arago, F-75014 Paris,France Universit¨ats-Sternwarte, Scheinerstrasse 1, Munich D-81679,Germany Argelander-Institut f¨ur Astronomie, Auf dem H¨ugel 71, D-53121 Bonn, Germany INAF, Osservatorio di Roma, Monteporzio Catone (RM),Italy California Institute of Technology, MS 105-24, Pasadena, CA91125, USA z ∼
1, early type galaxiesare observed to be redder, older, more luminous andmore massive than spiral and irregular galaxies (e.g.Bell et al. 2004; Bundy et al. 2006; Pozzetti et al. 2009;Bolzonella et al. 2009). The fraction of rotationally sup-ported early type galaxies is about 60% and it doesnot change significantly between z ∼ z ∼ HI/M ∗ ra-tios in ellipticals compared with later type galaxies (e.g.Haynes & Giovanelli 1984; Noordermeer et al. 2005).It has been known for many years that the observedmix of different morphological classes at the presentepoch also depends on the galaxian environment. Earlytype morphologies are preferentially found in dense,cluster-like regions. This idea can be traced back toHubble, but the major impact on the astronomicalcommunity started after Dressler’s study of 55 nearbygalaxy clusters (Dressler 1980). The existence of themorphology-density relation is presumably linked to thephysical processes which govern the star formation (andmorphological transformation) of galaxies and the con-nection of these to the environments of individual galax-ies.After the establishment of the morphology-density re-lation in clusters (e.g. Dressler 1980), the existenceof the relation was extended to the group environ-ments (e.g. Postman & Geller 1984) at low redshift.Later, the evolution with redshift was measured (e.g.Dressler et al. 1997; Postman et al. 2005; Fasano et al.2000; Desai et al. 2007). There is evidence that theevolution in intermediate-density environments has oc-curred more recently than in high-density environments(Smith et al. 2005). Also, it appears that the morpho-logical mix of cluster early-types has changed with time(Fasano et al. 2000; Postman et al. 2005). Ultimately,understanding the origin of the morphology-density re-lation can help to disentangle the relative roles of nature(conditions at epoch of galaxy formation) and nurture(environment dependent processes) in determining theproperties of galaxies.It is known that galactic mass also plays a major rolein determining the properties of galaxies, since bothcolour, specific star-formation rates and morphologiesare strongly correlated with the stellar mass of galax-ies (e.g. Kauffmann et al. 2003). The recent evidencethat the stellar mass function of galaxies itself dependson the environment (Baldry et al. 2006; Bolzonella et al.2009) complicates the separation of the relative roles ofmass (which could be regarded as “nature”) and of envi-ronment (“nurture”). This linkage also complicates theinterpretation of environmental effects in any sample ofgalaxies that contains a significant range of masses, sincedifferent environments will contain, within such a sam-ple, a different mix of masses. This in turn can easilyproduce the appearance of a “spurious” environmental dependence through the dependence of galaxian prop-erties on mass (see Tasca et al. 2009 and Cucciati et al.submitted a for a discussion).Several physical processes have been suggested whichcould enhance the transformation of late to early typegalaxies (and star-forming to quiescent) in dense regions(see Boselli & Gavazzi 2006 for a recent review of theseprocesses). Mergers and interactions with other galax-ies or with the cluster potential (e.g. Toomre & Toomre1972; Barnes 1988) are purely gravitational processes.Galaxy harrasment combines the cumulative effect ofthe interaction of galaxy with close neighbours with theinteraction of the total cluster potential (Moore et al.1998). Ram pressure stripping may be responsible forthe removal of the cold gas when a galaxy infalls withhigh velocity into a dense medium (Gunn & Gott 1972;Quilis et al. 2000). While ram-pressure stripping is ex-pected to be most effective in the dense intraclustermedium, Rasmussen et al. (2006) have shown that gasstripping by the intragroup medium may also be rapidenough to transform late types to S0s over a few Gyrs.Strangulation (or starvation) is a process that may re-move a halo of hot diffuse gas after a galaxy becomesa satellite in a larger dark matter halo (Larson et al.1980; Balogh & Morris 2000). Bamford et al. (2009)summarised these processes in to two broad categories:(a) those processes which will quench star formation andact on morphology passively, through the fading of spiralarms and reduction of the prominence of the disc afterstar formation has stopped (e.g. strangulation), and (b)those processes which will directly affect the stellar kine-matics and galactic structure and also lead to a cessationof star formation (e.g. mergers). While both categoriescan produce S0s, only the second one can produce trueelliptical galaxies.One way to assess the relevance of a type-transformation process is to constrain its timescale.However, there are number of timescales related to agiven process, and the interpretation of an observedtransformation rate is not necessarily unique. We heredifferentiate between the following timescales. The“physical timescale t p ” is a measure for a duration ofthe transformation of an individual galaxy. For exam-ple, the t p of the morphological transformation would bethe average time which elapses since the beginning of themorphological transformation, when a galaxy would stillbe classified as late, till the point in its evolution when itwill be classified as early. Sometimes it is more relevantto express the transformation in a probabilistic sense, asin radioactive decay. We can then quantify how long itwill take (on average) for a galaxy of a given class beforeit gets transformed to a galaxy of some other class. Thisis a “statistical timescale” t s , which is just the inverseof the average transformation rate. Finally, there is anobserved “rate timescale” t η (obtained as the inverse ofthe observed rate), based on the real observations, whosemeaning depends on its relation to the other timescales.If t p ≪ t η , then t η will be of order t s . However, if t p isof order t η then t η will be more a measure of t p . The in-terpretation of the observed t η therefore depends on thephysical timescale t p .In this paper, the main aim is to study the changes inthe morphological mix of galaxies in group environmentsthat have occurred in the time interval between z ∼ z ∼ .
1. Our primary goal is to understand therole of the group environment in the transformation ofgalaxies over this redshift range. We carry out the analy-sis by measuring the redshift evolution of the fraction ofgalaxies with early type morphologies in the total popu-lation of galaxies of a given luminosity or stellar mass inthree different environmental samples: group, field andisolated galaxies. This leads us to construct the morpho-logical transformation rate, normalised to the numberof galaxies in the “pre-transformation” state (i.e. latemorphological type) in different environments and quan-tify the timescales of the morphological transformationin different environments. We then compare these rateswith the equivalent ones defined by overall colour and bystar-formation rate indicators, that reflect the quenchingof star-formation activity. Determination of the transfor-mation rates, i.e. of the flux of galaxies crossing througha chosen dividing line, should be less sensitive to the ex-act values chosen to divide galaxies into different classesthan for example the quantity which reflects directly thepartitioning into these classes of galaxies (such as cross-ing or transition mass defined as the mass at which twoclasses are equally partitioned).The analysis is based on the COSMOS field(Scoville et al. 2007a), relying heavily on the zCOSMOSdata (Lilly et al. 2007, 2009) necessary for the precise,small-scale definition of environment on the scale ofgroups. This paper is closely related to the numberof other studies based on the data in the COSMOSfield that have been aimed to address the complex in-terplay between galaxy properties and environment upto z ∼ − .
5. A complementary analysis exploringgalaxy colours in the group catalogue of Knobel et al.(2009) is presented in Iovino et al. (2009). Scoville et al.(2007b); Capak et al. (2007a); Guzzo et al. (2007);Cassata et al. (2007) and Ideue et al. (2009) study thegalaxy property-environment relation using the datasetsbased on the photometric redshifts. Tasca et al. (2009);Cucciati et al. (submitted a) and Bolzonella et al. (2009)extend this study by using high precision spectroscopicredshifts and the continuous density field of Kovac et al.(2009). Moreover, the environmental dependence of spe-cific galaxy populations has been addressed in separatestudies, e.g. infrared-luminous galaxies in Caputi et al.(2009), AGNs in Silverman et al. (2009) and post-starburst galaxies in Vergani et al. (2009). The rela-tion between the galaxy distribution and that of the un-derlying matter density field is explored in Kovaˇc et al.(submitted).Throughout the paper we use a concordance cosmologywith Ω m = 0 .
25, Ω Λ = 0 .
75 and H = 70 km s − Mpc − .Magnitudes are given in the AB system. THE ZCOSMOS SURVEY
Details of the survey
COSMOS Scoville et al. (2007a) is a multi-wavelengthsurvey of a 2 square degree equatorial field, observed sofar by the major earth and space based facilities, e.g.Hubble, Spitzer, Galex, Chandra, Subaru. The zCOS-MOS (Lilly et al. 2007) is a spectroscopic redshift sur-vey in this field carried out with the VIMOS spectro-graph on the ESO UT3 8-m VLT. The zCOSMOS-brightpart of the survey is flux limited at I AB < .
5, andwill ultimately obtain over 20,000 spectra. At this flux limit, the redshifts of the majority of the observed galax-ies fall in the range 0 < z < .
4. A higher redshiftrange, 1 . < z < . z ) orbetter. These are used to understand the spectroscopiccompleteness of the survey (see Lilly et al. 2009).The rest-frame absolute magnitudes of all I AB < . M ∗ are ob-tained from fitting stellar population synthesis modelsto the SED of the observed magnitudes as described inBolzonella et al. (2009) and Pozzetti et al. (2009), usingthe best available redshift. Stellar masses which we useare obtained with the Bruzual & Charlot (2003) librariesand the Chabrier initial mass function (Chabrier 2003),the Calzetti et al. (2000) extinction law with 0 < A V < τ , taking values from the interval 0 . < τ <
30 Gyr.The stellar mass is obtained by integrating the SFH overthe galaxy age, from which the mass of gas processedby stars and returned to the interstellar medium duringtheir evolution (“return fraction”) has been subtracted.See Bolzonella et al. (2009) for more details.
The zCOSMOS 10k group catalogue
The details of our group-finding algorithm and thegroup catalogue is presented in Knobel et al. (2009).Here, we outline only the most important points regard-ing the group catalogue.Knobel et al. (2009) combine two standard groupsearching algorithms: Friends-Of-Friends (FoF) and theVoronoi-Delaunay method (VDM) to define groups in the10k zCOSMOS. The group finding parameters are opti-mised based on the tests on the COSMOS mock cata-logues (Kitzbichler & White 2007) designed to match thegeometrical and sampling properties of the 10k zCOS-MOS catalogue. Knobel et al. (2009) also developed anew “multi-pass” strategy that optimises group findingover a range of richnesses by first optimising for thelargest groups and then subsequently moving down togroups with a smaller number of observed members.The group catalogue which we use in this work is theso-called “one-way-matched” catalogue, which is the in-tersection of the independently created FoF and VDMcatalogues. In total, there are 800 groups with at leasttwo members in the current “10k-sample”. Over 100 ofthese groups have at least 5 spectroscopically observedmembers.Extensive comparisons with the mock catalogues(Kitzbichler & White 2007) has established that thezCOSMOS 10k group catalogue has a relatively high pu-rity and completeness - 82% and 81% respectively forgroups with N ≥
5, and only marginally lower for thepoorer groups. It should however be appreciated that
Fig. 1.—
Properties of the zCOSMOS groups (0 . < z <
1) used for the current analysis. Left: Distribution of the “corrected richness”,i.e. the number of members corrected for sampling incompleteness, above M B < − . − z . Right: Distribution of the estimated halomasses for the zCOSMOS groups shown on the left. The halo masses are estimated by comparison with the COSMOS mock catalogues ofKitzbichler & White (2007). The majority of the groups have less than 10 corrected M B < − . − z members and masses in the range12 < log( M DM /M ⊙ ) <
14, emphasizing that the sample involves cosmic structures well below the masses of rich clusters of galaxies. these numbers are based on comparisons with the idealgroup catalogues that could have been constructed (withtwo or more spectroscopic members per group) from asparsely sampled “10k-like” sample, rather than the ac-tual set of all groups in the COSMOS volume. As aresult, some of the galaxies that are not currently placedin a group are nevertheless likely to be in a group whichwill be identified in the future with the higher samplingof the full survey. The current “non-group” sample willtherefore contain some “group” galaxies, and this con-tamination will act to reduce observed differences be-tween group and non-group populations. Partly for thisreason, as discussed below, we will usually use the entiregalaxy population that we have observed spectroscop-ically, selected without regard to environment, as ourcontrol sample. This also reflects the traditional defini-tion of a “field sample”.Due to the non-uniform sampling scheme of the current10k zCOSMOS survey (Lilly et al. 2007, 2009), the frac-tion of group members observed (as well as the chance ofrecognising the poorer groups) will be a function of posi-tions on the sky. We therefore correct the observed num-ber of members for each group by the overall samplingfraction of the survey at that location, taking into ac-count both the spatially variable spectroscopic samplingrate and the redshift-dependent success rate in yieldinga high confidence redshift (using the corrections derivedin Knobel et al. 2009). We will refer to this as the cor-rected richness, which is the estimated true number ofmembers above the flux-limit of the survey. For anygiven physical group, this corrected richness will varywith redshift due to the changing luminosity limit that isassociated with the observed I -band limit. We can dealwith this by considering groups with a given correctedrichness above some fixed luminosity-limit, allowing thislimit to evolve with redshift to account, as best we can,for the luminosity evolution of individual galaxies. We refer to such a set of groups as a “volume-limited” sam-ple in that it should contain all groups of a given type,but still this sample could be potentially affected by theso called “progenitor bias” since infall of new memberswill produce new groups in this volume-limited sampleat later epochs. Moreover, given that the original groupcatalogue is generated from a flux limited sample, therewill be a slight bias with redshift because poor groups aremore difficult to detect at high z than at low z . How-ever, the distribution of corrected richness with redshiftis rather flat, reassuring us that this effect is minor.The distribution of corrected richnesses using the lu-minosity limit M B < − . − z in 0 . < z < M B < − . − z .Knobel et al. (2009) also give estimates for the ob-served velocity dispersion for groups with N ≥ M B < − . − z detected in 0 . < z < < log( M DM /M ⊙ ) <
14. This emphasises thatthe zCOSMOS volume to z ∼ Isolated galaxies
In parallel, we measure the morphological mix of galax-ies in a sample of isolated galaxies, to further highlightthe possible environmental segregation of galaxy mor-phological types. We use for this purpose the sample ofisolated galaxies defined by Iovino et al. (2009) as fol-lows. First, the Voronoi volume is estimated for eachgalaxy in the zCOSMOS flux limited I AB < . . < RA < .
41 degand 1 . < DEC < .
68 deg in 0 . < z <
1. TheseVoronoi volumes are normalised by the median Voronoivolume in ∆ z = 0 . ∼
10% of the preliminary sample of isolated galax-ies), most probably affected by the survey borders, andgalaxies detected to be in groups ( ∼
14% of the pre-liminary sample of isolated galaxies) are removed fromthe sample. Based on the I AB < . Field galaxies
Following original practice, we refer in this paper toa “field” sample as being all galaxies selected withoutregard of environment. In the comparisons of the variousrelations, the field sample will always be chosen to matchexactly the selection of the group and isolated samplesof galaxies as regards all non-environmental parameters,e.g. luminosity, mass, type, redshift etc.
Morphologies
The COSMOS field is the largest contiguous fieldcovered by HST ACS images. The HST observations(Koekemoer et al. 2007) were taken in the F814W filter.The ACS imaging allows the possibility to derive robuststructural parameters and to confidently morphologicallyclassify all galaxies into ’Hubble types’ down to about 24mag (and possibly deeper), well below the flux limit ofzCOSMOS-bright.Here we use the structural parameters and morpho-logical classification based on the
Zurich Estimator ofStructural Types (ZEST) and presented in Scarlata et al.(2007). The ZEST classification scheme is based on aprincipal component analysis of five non-parametric di-agnostics of galaxy structure: asymmetry A, concentra-tion C, Gini coefficient G, the second-order moment ofthe brightest 20% of galaxy pixels M and the ellipticity ǫ . The morphological classification is performed in thespace of the first three eigenvectors, which contain mostof the variance of the original non-parametric quanti-ties. At each point in the three-dimensional eigenvectorspace, a morphological class is assigned based on eye-ballinspection and the median Sersic index of all the galaxiesin that cell.Galaxies are classified by ZEST into ellipticals, diskand irregular galaxies. The disk galaxies are further splitin four bins, namely bulgeless, small-bulge, intermediate-bulge and bulge-dominated disk galaxies.In the following we consider as “early-types” all galax-ies that are classified by ZEST as either ’elliptical’ or’bulge-dominated’ systems, i.e. specifically ZEST Classes1 and 2.0. All other ZEST types are together consideredto be “late-type” galaxies. In terms of the conventional Hubble classification scheme, our division would proba-bly be between Sa and Sb. SAMPLE DEFINITION
While there are about 40,000 galaxies in the zCOS-MOS field which satisfy the survey selection criteria I AB < . z < .
2, increasing to 95%for 0 . < z < .
8, with an overall redshift reliability of98.6%. We refer to this sample as the high confidenceclass 10k sample (HCC 10k).Due to the inhomogeneous sampling of galaxies withspectroscopic redshifts, we limit our analysis to thezCOSMOS region defined by 149 . ≤ RA ≤ .
42 degand 1 . ≤ DEC ≤ .
70 deg, in which the sampling isboth higher and more homogeneous. The area is slightlysmaller for the isolated galaxies, see Section 2.3). This isthe same region as used by Iovino et al. (submitted) fora parallel analysis which investigate the colour segrega-tion of galaxies in groups (e.g. see Figure 1 in Iovino etal. submitted).
Correction for the spectroscopic incompleteness
In zCOSMOS, the targets to be observed spectroscopi-cally are selected independently of the galaxy properties,except for the requirement of 15 . < I AB < .
5. Thefailure to assign a redshift to an observed spectra mayhave several causes, some of which may depend directlyor indirectly on the properties of the target galaxies. Wetherefore adopt a weighting scheme to correct for the ob-jects observed spectroscopically which have insufficientlygood redshifts to be included in the HCC 10k sample(including complete failures) taking into account onlygalaxy properties. We define a multidimensional spaceby the selection magnitude I AB , the rest-frame ( U − V ) colour, the morphological type and the redshift, relyingon the values calculated using the high quality photo-metric redshifts. For each galaxy, the multi-parameterspace is sampled for galaxies of the same morphologicalclass within an interval, centred on the galaxy in ques-tion, that is 0.25 mag in I AB magnitude, 0.25 mag in therest-frame ( U − V ) colour, and 0.1 in redshift. We thendefine the inverse weight w i of each galaxy i in the HCC10k sample as the fraction: w i = N iHCC k N i k , (1)where N iHCC k is the number of galaxies in the HCC10k sample and N i k is the number of galaxies in thetotal 10k sample which are neighbours of the i -th galaxyin the above defined manner in the multidimensional I AB − ( U − V ) − z phot -morphology space. In those caseswhere a galaxy in the 10k sample could not be assignedany of the properties considered ( I AB , ( U − V ) , z phot andmorphology) the galaxy is excluded from the analysis.We can neglect these galaxies given that their exclusionis based on the absence of data and not directly to theproperties of galaxies. Moreover, they constitute a neg-ligible fraction ( < . < z < Computing the fraction of galaxies and errors
Having defined the inverse weights of the galaxies inthe HCC 10k sample using equation 1, the fraction ofgalaxies of a given morphological type within any givengalaxy sample is straightforwardly computed from theHCC sample in terms of the sum of these individualweights. Specifically, the total number of galaxies of agiven morphological type T (using the ZEST morpho-logical classification) in a given sample of k galaxies iscalculated using N KT = Σ k w k ( T ) (2)where the summation goes over all galaxies of the T typein the K-subsample of the HCC 10k sample. Similarly,the total (weighted) number of galaxies of any type canbe calculated using N KALL = Σ k w k (3)summing over all galaxies in the K-subsample of the HCC10k sample. The fraction of galaxies of a given morpho-logical type T is then straightforwardly given by f KT = N KT N KALL . (4)The index K refers to the sample in which the morpho-logical fraction of galaxies is calculated and is defined byenvironment, as well luminosity or stellar mass. In con-sidering sets of galaxies defined by luminosity or mass, itwas decided to only consider intervals in either quantityin which we are essentially complete over a given redshiftrange, i.e. not to attempt a V /V max type correction forincompleteness.Uncertainties in the observed fractions are estimatedusing a bootstrap resampling of the given subsample ofgalaxies, setting the error bars in the f KT as the standarddeviation in the distribution of fractions resulting from1000 Monte Carlo bootstrap realisations. RESULTS
In this section, we measure and compare the morpho-logical mix of galaxies residing in different environments(groups, isolated and field) lying within the zCOSMOSvolume in the redshift interval 0 . < z <
1. As notedabove, we refer to the morphological types 1 and 2.0 to-gether as the “early type” galaxies and the remainder as“late type”.
Fig. 2.—
The rest frame B -band magnitude as a function ofredshift for the 10k zCOSMOS sample. Galaxies of early morpho-logical type (ZEST types 1 and 2.0) are coloured red. The greenlines define the four luminosity-complete samples used in the anal-ysis. Given that the zCOSMOS groups contain typicallyonly a few members, we cannot measure a meaningfulmorphological fraction of galaxies in individual groups,as is common when studying similar relations in rich clus-ters of galaxies, but rather combine all galaxies residingin groups with similar properties to build a statisticallymeaningful sample of galaxies. It will become clear that,even with this large set of high redshift groups, the se-lection of sub-samples of groups and galaxies for study isprimarily guided by the need to have statistically usablenumbers of galaxies in each of the bins.Furthermore, the current sparse sampling of zCOS-MOS means that we cannot meaningfully distinguish be-tween central and satellite galaxies within the groups.Much recent work aimed to address the redshift evolu-tion of a morphological mix of galaxies targeted rich clus-ters individually which could be easily identified at high z (see recent work by Desai et al. 2007 and referencestherein). For the precise reconstruction of the groupsone needs high resolution spectroscopic redshifts in a rel-atively large continuous volume. Needless to say, theadditional requirement is high resolution imaging overthe full survey area in order to reliably measure galaxianmorphologies. The zCOSMOS groups are identified in alarger volume than in the two previous high redshift ( z ∼
1) spectroscopic surveys, VVDS (Le F`evre et al. 2005)and DEEP2 (Davis et al. 2003). Moreover, only colourproperties of galaxies in group environments have beenstudied in 0 . < z < . . < z < . Morphological fraction of galaxies in the luminositycomplete samples
To study the dependence of morphological segrega-tion of galaxies in groups on the luminosity of galax-ies, it is useful to first define ”luminosity-complete”, i.e. M B,compl limit M B < − − z M B < − . − z M B < − − z M B < − . − zz range . < z < . . < z < . . < z < . . < z < Group ( M B < M B,compl ) 564 665 605 329
Group ( M B < . − z ) 275 321 386 329 F ield
Isolated
240 334 340 206
TABLE 1Number of (non-weighted) galaxies in the luminosity complete samples. “volume-limited”, samples of galaxies. We do this pri-marily to compare with the results in literature, beforeurning to mass-selected samples below.We define four galaxy samples which satisfy the fol-lowing criteria: M B < − − z at 0 . < z < . M B < − . − z at 0 . < z < . M B < − − z at0 . < z < . M B < − . − z at 0 . < z < M B magnitude be-cause it is well matched to the observed I -band selec-tion of zCOSMOS-bright at high redshifts, with an exactmatch at z ∼ .
8, thereby assuring that the completenessof the zCOSMOS galaxies is not colour dependent (seeIovino et al. 2009). The “-z” term accounts, at best ap-proximately, for the luminosity evolution of individualgalaxies ∆ M B = − × ∆ z (see Kovac et al. 2009 for dis-cussion). The number of galaxies in the luminosity com-plete samples is given in Table 1. As shown in Figure 2,these galaxy samples should be statistically complete.Correspondingly, for each of these luminosity-completegalaxy samples, we can also define a set of “luminosity-complete” groups that satisfy the group criteria, i.e. thattheir corrected richness is at least two above these samegalaxy luminosity limits. These groups should also be“volume-limited” in a sense, although it is clear that theaccretion of new members may lead to a change in thegroup population through the relevant redshift range.With these considerations and caveats in mind, Fig-ure 3 shows the fraction of early type galaxies in dif-ferent environments as a function of (evolving) absolute M B magnitude. The black points show the field sam-ple (all galaxies) while the green points show the set ofisolated galaxies. The yellow points in each panel showthe subset of galaxies that appear in the correspondingcomplete set of groups at that redshift, i.e. with morethan two corrected members above a luminosity thresh-old for which we are complete, i.e. M B < − − z for the 0 . < z < . M B < − . − z for the0 . < z < . M B < − − z for the 0 . < z < . M B < − . − z , for whichwe are complete across the whole redshift range. Apartfrom the caveats above about infall and progenitor bias,the group galaxies that are represented by the red pointsshould therefore be directly comparable across the wholeredshift range.It is clear from Figure 3 that the fraction of early typegalaxies is higher for brighter galaxies in all redshift bins,covering the range from z = 0 . z = 1. This reflectsthe strong and well-known dependence of morphology onmass or luminosity. In addition, however, the fraction ofearly type galaxies at fixed M B is higher for galaxies re-siding in groups compared with the overall populationof field galaxies. This holds for every luminosity sampleand at every redshift bin considered, even though thedifference is sometimes within the errors. Similarly, the fractions of isolated galaxies with early type morpholo-gies tend to be lower than in the sample as a whole. Inthe highest redshift bin (0 . < z < M B which are explored over three redshift in-tervals: 0 . < z < .
5, 0 . < z < . . < z < z = 1, whereas the yellow points represent themembers of a set of groups that is only volume limitedwithin each individual panel. The redshift evolution ofthe early type fraction seen in Figure 4 indicates a clearbuild up over time of the population of luminous earlytype galaxies in the group environment. For the isolatedgalaxies the build-up of early type galaxies is detectedonly for the faintest galaxies. For the isolated galaxieswith − − z < M B < − − z the fraction of early typesstays more or less constant since z = 0 .
9, but the erroron the measured fraction at the lowest redshift is ratherlarge and we do not have a sufficient number of isolatedgalaxies with − − z < M B < − − z to reliably carryout this analysis.As in Figure 3, it can be seen that the fraction of earlytype galaxies in the group environment is higher thanthe fraction of early type galaxies in the field (all galax-ies), which is again higher than the early-type fractionin the isolated galaxies. By fitting a simple linear rela-tion between the fraction of early type galaxies and thelook-back time, the data suggest that the morphologi-cal transformation of galaxies from late to early type hasgenerally happened earlier for galaxies residing in groupsthan for galaxies of the same luminosity in the field, cou-pled with an overall trend for the transformation to occurearlier for more luminous galaxies. We return to quantifythis further below.The evolution of the early type fraction over cosmictime and its dependence on luminosity qualitatively mir-rors (i.e. it is inverse to) the evolution and its environ-mental and luminosity dependencies of the fraction ofblue galaxies in the zCOSMOS group, field and isolatedsamples, investigated by Iovino et al. (2009). Comparison with literature
An exact comparison of our results with the previouslypublished studies of the morphological mix of galaxies in0 . < z < Fig. 3.—
The fraction of early type galaxies at a given M B magnitude, evolving the luminosity to z = 0. The red and yellow pentagonsrepresent the group galaxies, the black circles represent the field galaxies and the green triangles represent the isolated galaxies. The groupgalaxies are included in the analysis only if the group has a corrected richness of at least two in the given luminosity complete sample(empty yellow pentagons) or if the group has the effective richness of at least two in M B < − . − z sample (solid red pentagons, whichare therefore complete across all panels). The fraction of galaxies in the different panels are calculated using only those galaxies fromluminosity complete samples in the narrow redshift slices (indicated at the top of each panel). The fraction of early type galaxies is higherfor brighter galaxies in all redshift bins. The fractions of early type galaxies at fixed M B are also higher for galaxies residing in groupscompared with the overall population of field galaxies. The fractions of isolated galaxies with early type morphologies tend to be the lowest. ent selection of galaxies and the generally much denserenvironments. However, we can qualitatively place ourresults within the previous results obtained with the lu-minosity complete samples.Galaxies which reside in the zCOSMOS groups residealso in the regions of the highest density in the con-tinuously defined zCOSMOS density field (Kovac et al.2009). The higher fractions of the early type galaxiesin the groups relative to the field and isolated (whenmeasured) samples are consistent with the morphology-density relations of (Tasca et al. 2009; Capak et al.2007a; Guzzo et al. 2007; Cassata et al. 2007) at similarluminosity limits and similar investigated environments.The ZENS project (Carollo et al. in prep.) is a B and I wide-field imaging survey of galaxies in a statistically andluminosity-complete sample of 185 groups at z ∼ . z ∼ .
05 ZENS groups down to M B = −
17 (Cibinel et al.in prep.). Going to higher redshift, Wilman et al. (2009)measured the morphological mix of M r < −
19 galaxiesin the group and non-group environments using a sampleof optically selected groups in 0 . ≤ z ≤ .
55 from theCanadian Network for Observational Cosmology redshiftsurvey (Carlberg et al. 2001; Wilman et al. 2005). Theyconclude that the fraction of S0 galaxies in groups at z ∼ . z ∼ . . < z < . M V = −
20 (using cosmology with Ω m = 1 and H = 50 km s − Mpc − ), Desai et al. (2007) concludethat the fraction of S0 galaxies in clusters increases about2 times from z = 0 . z = 0, while the fraction of ellip-ticals stays constant over the same redshift interval. TheEdisCS results ( M V < −
20) combined with the equiva-lent studies of seven clusters from Postman et al. (2005)at 0 . < z < .
27 ( M V < − . − . z ) are consistentwith no evolution in the morphological mix in clustersover the entire 0 . < z < .
25 interval (using Ω m = 0 . Λ = 0 . H = 70 km s − Mpc − , Desai et al. 2007), Fig. 4.—
The redshift/look-back time evolution of the fraction of early type galaxies in different environments. As in Figure 3, the redand yellow symbols represent the group galaxies, black circles represent the field galaxies and the green triangles represent the isolatedgalaxies in the indicated M B -bins of galaxies. We consider two following types of groups for this analysis: with the corrected richness of atleast two in M B < − − z (yellow symbols left panel), in M B < − − z (yellow symbols middle panel) or in M B < − . − z sample ofgalaxies (red symbols, therefore complete across all panels). For the analysis, we consider only redshift intervals in which galaxies of a givenluminosity are drawn from the luminosity complete samples: 0 . < z < .
5, 0 . < z < . . < z < M B , also in the isolated sample. suggesting that z ∼ . z = 1 or z = 0 . − − z < M B < − − z or − − z < M B < − − z galaxies, respectively. Ig-noring the differences in the luminosity selection (themedian B − V colour of the field M B < − . − z or M V < −
20 zCOSMOS galaxies in 0 . < z < . < z <
1, groups are preferred environment forthe morphological transformation of galaxies over clus-ters.
Evolution of the morphological fraction as afunction of the properties of the group
In the previous Section we have seen the changes inthe morphological mix of galaxies in groups over cosmictime. Comparison of the yellow and red points in thepanels on the left of Figure 3 and Figure 4 suggests a de-pendence also on the group properties, since the yellowpoints represent groups that extend to poorer levels thanthe red points, which represent galaxies in (rich) groupsthat have at least two members above M B < − . − z .Desai et al. (2007) have shown that morphological evolu-tion of galaxies at z < z ∼ z ∼ Group richness
First, we look at the dependence of the morphologicalfraction of galaxies as a function of the corrected rich-ness of the host group. We carry out this study for thefour luminosity complete samples defined above, usingthe group corrected richness defined in each of the lu-minosity complete samples, i.e. the corrected numberof members above the given galaxian luminosity limit.We split the group galaxies in bins using the effectiverichness of a group (above the quoted luminosity limit)between 2 and 5, between 5 and 8 and larger than 8 forthe groups in the luminosity complete samples up to theluminosity of M B < − − z . For the brightest sample M B < − . − z we define bins of group effective rich-ness between 2 and 7 and larger than 7, considering only0 Fig. 5.—
The time-evolution of the fraction of early type galaxiesin the groups of various richnesses. The solid squares represent thegroup population in luminosity complete samples as indicated inthe individual panels, which are also used to define the correctedrichness. The orange, violet and blue squares are for the galax-ies residing in the groups with effective richness between 2 and 5,between 5 and 8 and larger than 8, respectively, for the top threepanels. In the bottom panel, the light red and violet-blue squaresare for the groups with the effective richness between 2 and 7, andlarger than 7, respectively. The points are plotted at the medianlook-back time of galaxies in that bin. The fits are carried out atthe median look-back time in the given redshift bin. For a compar-ison, the continuous line represents the linear fit to the f early − t LB relation of the field galaxies in the corresponding luminosity com-plete sample. At lower luminosities and/or lower redshifts there isa clear trend of higher early-type fraction in the richer groups. galaxies and groups in 0 . < z < Velocity dispersion
This analysis can be repeated using the observed ve-locity dispersion for the groups. Due to the large un-certainties related to velocity dispersion measurements(see Knobel et al. 2009), only groups with at least 5 ob-served members (above the I AB < . < σ <
500 km s − and σ >
500 km s − . Thefraction of early type galaxies in these two group bins areshown in Figure 6.Even though of a lower significance, the same pictureemerges. The fraction of early type galaxies is largerin the groups with larger velocity dispersion for galaxiesfainter than M B = − . − z . For brighter galaxies wedo not detect any significant difference between fractionsof early type galaxies in groups of different velocity dis-persion, possibly due to the limited number of galaxies. Fig. 6.—
The fraction of early type galaxies in different environ-ments defined by the velocity dispersion of a group, as a functionof redshift/time. The yellow and violet solid squares are for thegalaxies residing in the groups with the velocity dispersion between250 and 500 km s − , and larger than 500 km s − , respectively.The additional requirement is that a group has at least 5 observed( I AB < .
5) members. In the given panels, all galaxies used forthe analysis satisfy the indicated magnitude limit, and the redshiftranges in which these samples are complete. For a comparison, thecontinuous line represents the linear fit to the f early − t LB relationof the field galaxies in the corresponding luminosity complete sam-ple. The bottom panel is an exception where we consider a sampleof M B < − . − z galaxies only in 0 . < z < M B = − . − z . Importance of the stellar mass distributions (stellarmass function in the groups and for the isolatedgalaxies)
For comparison with the literature, the above discus-sion was based on the rest-frame B -band luminosity. Aswell as the obvious sensitivity of the B -band luminosityto the recent star formation history, there are many rea-sons to prefer stellar mass as a yardstick of galaxian prop-erties. In particular, as mentioned in the introduction,the existence of a correlation between the stellar massfunction of galaxies and their environment (Baldry et al.2006; Bolzonella et al. 2009), plus the strong correla-tion between mass and galaxian properties, means thatany sample that contains a significant range of massesmay show a spurious environmental dependence sim-ply because the different distribution of masses in thedifferent environments. This will be true not only for B -selected samples, but also K -selected samples, andeven mass-limited samples (i.e. all galaxies with massesabove some threshold). The spurious environmental ef-fects produced by this difference in mass functions isshown by the clear flattening of the observed colour-density and morphology-density relations (Cucciati et al.submitted a; Tasca et al. 2009) when going from narrow M B luminosity bins to narrow mass bins.Using the overall zCOSMOS density field (Kovac et al.1 Fig. 7.—
The stellar mass function in different environments. The left-hand panels are obtained for the group galaxies residing in thegroups with at least 2 corrected M B < − . − z members. The right-hand panels are for the isolated galaxies. The symbols correspondto the 1 /V max measurements and the continuous lines correspond to the Shechter fits (see text for more details). Each four-panel figurecovers 0 . < z <
1, split in four redshift bins indicated on the top. The black, red and blue symbols/curves are results for the populationsof all, early type and late type galaxies, respectively. For a comparison, the stellar mass function of the field galaxies is presented with graysymbols/curve. The environmental dependence of the shape of the stellar mass function is explained by the different relative contributionsof galaxies of different types in different environments. See Bolzonella et al. (2009) for the detailed discussion of the role of environmenton the stellar mass function. z ∼ .
7, and probably at higherredshifts. The galaxian stellar mass function in the dens-est regions, defined by the highest quartile in the distri-bution of galaxy environments at that redshift bin, is ofbimodal shape at least up to z ∼ .
5, showing an upturnat low-mass end and with massive galaxies preferentiallyresiding in these densest regions. Bolzonella et al. (2009)explain the environmental dependence of the shape ofthe stellar mass function by the different relative con-tributions of galaxies of different types in different en-vironments. The individual stellar mass function foreach of the different types is well represented by theSchechter function in all environments (see Figure 4 inBolzonella et al. 2009). The higher contribution of thered/early objects in the densest environments producea characteristic “bump” at high-mass end in the stellarmass function in the densest environment.In this paper we are investigating the morpholog-ical properties of galaxies in the group environment(Knobel et al. 2009), corresponding to virialised struc-tures (i.e. single dark matter haloes) - and can be confi-dent from the comparison with the mocks that this is in-deed to a large degree the case (Knobel et al. 2009). Notsurprisingly, galaxies in the identified groups reside alsoin the denser environments (see Figure 15 in Kovac et al.2009), and the groups are themselves good tracers ofthe dense peaks in the density field (see Figure 16 inKovac et al. 2009). However, there is not a one-to-onecorrespondence between the group and the continuouslydefined environment.We have therefore recalculated the stellar mass func- tion for the samples of the group and isolated galaxiesused in this paper, following exactly the same methodol-ogy as described in Bolzonella et al. (2009). The stellarmass functions are calculated using the 1 /V max method,by using the additional weights for each galaxy to cor-rect for the average sampling and redshift success rate ofthe 10k HCC zCOSMOS sample to the full 40k sample(as the stellar mass function deals with the volume den-sities and not fractions of objects, similar to the weightsused to define the group richness). The errors corre-spond to Poissonian errors obtained from the 1 /V max method. The V max refer to the total zCOSMOS volume,and therefore the normalisation of all stellar mass func-tions, except of the one of the total sample of galaxies,is arbitrary. As for the rest of the analysis in this paper,galaxies are selected to reside in the central zCOSMOSregion, defined in Section 3.The resulting stellar mass functions for the group andisolated sample of galaxies for four redshift bins areshown in the left and right panel in Figure 7, respectively(black symbols and curves). The mass functions are fit-ted with a double Schechter function (with a unique valueof the exponential cut-off) in the two lower redshift binsand with a single Schechter function in the two higherredshift bins. To be able to compare similar group en-vironments at different redshifts, we select only groupgalaxies residing in the groups with at least 2 corrected M B < − . − z members. The shape of the stellarmass functions in the sample of group and isolated galax-ies is reminiscent of the shape of the stellar mass func-tions in the sample of galaxies in the highest and lowestquartile of overdensities in Bolzonella et al. (2009). Forthe group galaxies, similar results are obtained when us-2 M ∗ ,compl limit M ∗ > . M ∗ > . M ∗ > . M ∗ > . z range . < z < .
45 0 . < z < . . < z < . . < z < Group ( M B < M B,compl ) 422 390 294 132
Group ( M B < . − z ) 241 235 218 132 F ield
987 1045 858 428
Isolated
141 180 109 40
TABLE 2Number of (non-weighted) galaxies in the mass complete samples. ing all group members without restriction in the grouprichness - smoothing slightly the characteristic bimodalshape of the mass function in the dense region in the twolowest redshift bins, as expected when including in thegalaxy sample also members of the smaller groups.We also calculate stellar mass function per morpholog-ical type in different environments, using the morpholog-ical classification described in Section 2.5. As expected,the high mass end of the stellar mass function in bothenvironments is dominated by the early type galaxies,peaking at roughly 10 M ⊙ . Towards lower masses, thenumber of early type galaxies decreases and the numberof late type galaxies increases. Late type galaxies domi-nate below the crossing mass (the mass at which the twomass functions cross) and this crossing mass shifts to-wards lower values at later cosmic epochs. Since z ∼ . Morphological mix in different environments fromthe mass selected samples
We therefore select four stellar mass-complete sam-ples of galaxies to investigate the morphological mix in-side and outside groups in zCOSMOS. The selection ofthe complete stellar mass samples has been described inPozzetti et al. (2009), and is optimised to be completefor both early and late type galaxies following our defini-tion of morphological types. We use the samples whichare 85% complete for the early (1+2.0) type galaxies.The samples are defined as follows: log( M ∗ /M ⊙ ) > . . < z < .
45, log( M ∗ /M ⊙ ) > .
21 in 0 . 6, log( M ∗ /M ⊙ ) > . 68 in 0 . < z < . M ∗ /M ⊙ ) > . 96 in 0 . < z < . 95. The numbers ofgalaxies in the different environments within these masscomplete samples are given in Table 2.The morphological mix of galaxies within these mass-selected samples in different environments is shown in Figure 8. The stellar mass bins have a width of∆ log( M/M ⊙ ) = 0 . ∼ log( M/M ⊙ ) = 11 weobserve a steady increase with cosmic time in the frac-tion of early type galaxies in all environments. For thehighest mass bin, 10 . < log( M/M ⊙ ) < . 66, the num-ber of galaxies is small, and little can be conclusivelysaid regarding the groups and for the field. However, itseems clear that high fractions ∼ − 80% of early type ∼ log( M/M ⊙ ) > 11 galaxies in both the group and fieldare present before z = 1 and that the morphological mixstays more or less constant since then.Progressing to lower masses, we find that, at a givenstellar mass, the fraction of early type galaxies is consis-tently highest in the group environment, and lowest forthe isolated galaxies, at every redshift that we can exam-ine for that particular stellar mass. Fitting a linear rela-tion between the early type fraction and look-back time,the linear fits for galaxies with log( M/M ⊙ ) . 11 sug-gest that the fractions of early type galaxies were moreor less the same in all types of environment at higherredshifts and then, with cosmic time, diverged with pro-gressive morphological transformation from late to earlytype galaxies happening at different rates in different en-vironments at a given mass. As above, there is an indi-cation that the increase in the early type fraction overcosmic time is more rapid in the groups with higher rich-ness than in the poorer groups (e.g. difference betweenred pentagons and yellow squares and the correspondinglinear fits in Figure 8).Looking at the same data from another perspective,we can calculate the fractions of early type galaxies asa function of stellar mass in the narrow redshift bins:0 . < z < . 4, 0 . < z < . . < z < . 8. Theseare shown in Figure 9, using only groups with a correctedrichness of at least two in M B < − − z galaxies, as weextend this analysis only to z = 0 . 8. At every redshift,the fraction of early type galaxies increases strongly withthe stellar mass for galaxies residing in both the groupsand in the field. The isolated galaxies follow the sametrend in the lowest redshift bins up to z = 0 . 6, while inthe highest redshift bin they become very rare objectsand the obtained statistics is much less reliable. Thisfigure also illustrates the convergence of the morpholog-ical mix at a given (relatively high) mass at redshiftsapproaching z ∼ Fig. 8.— Redshift evolution of the fraction of early type galaxies of a given stellar mass in different environments. The stellar massesare from intervals of log( M ∗ /M ⊙ ) = 0 . M ∗ /M ⊙ ) units are given in the individual panels. Thelower stellar mass value represents the 85% completeness limit for the early (1 + 2 . 0) type galaxies in the redshift interval of consideration:0 . < z < . 45, 0 . < z < . 6, 0 . < z < . . < z < . 95 going from the left to the right, respectively. The solid yellow squaresand solid red pentagons represent group galaxies with an effective richness of at least two in the corresponding luminosity complete sampleand in the M B < − . − z sample of galaxies, respectively. The empty circles represent the field galaxies and the green solid trianglesrepresent the isolated galaxies. The symbols are plotted along horizontal axis at the median look-back time/redshift of galaxies in theconsidered bin. The dotted lines are the linear fits to the f early − t LB relation for the various samples, marked in the same colour as thecorresponding early type fractions. For galaxies with log( M ∗ /M ⊙ ) . 11, there is a steady increase with cosmic time in the fraction of earlytype galaxies in all environments. Moreover, the fraction of early type galaxies is highest in the group environment, and lowest for theisolated galaxies at every redshift probed by that particular log( M ∗ /M ⊙ ) . 11 stellar mass. tion with stellar mass in 0 . < z < . M/M ⊙ ) = 10 . 52. We have taken into ac-count the difference in the IMF used by Holden et al.(2007) and here, assuming that the stellar masses calcu-lated with the Chabrier IMF are larger by 0.08 than the“diet Salpeter” IMF used by Holden et al. (2007) (seee.g. Bolzonella et al. 2009 on the differences betweenthe stellar masses with the different IMFs). Holden et al.(2007) results are obtained for five massive X-ray clustersin 0 . < z < . 83 spanning a range in velocity disper-sion 865 < σ < − , or with 56 to 109 membersabove the quoted mass limit. The measured fractionsof early type galaxies are between 71% and 100%, com-parable to the fractions which we measure only for themost massive > log( M/M ⊙ ) ∼ 11 galaxies. This indi-cates that clusters are either more efficient in transform-ing late types to early types or that the bulk of mor-phological transformation of galaxies more massive thanlog( M/M ⊙ ) = 10 . 52 has happened earlier in the clus-ters than in any other environment, and specifically thepoorer group environments examined here.Figure 8 and Figure 9 also show one way of looking atthe relative importance of mass and environment in driv-ing morphological transformation. It can be seen that the morphological mix that is achieved in the groups at someredshift and at some mass, is achieved at only slightlylater times (or equivalently for slightly higher masses)in the field (all galaxies) and in the isolated galaxies. At z ∼ . M/M ⊙ ) ∼ . 2, the “effect” of group en-vironment is equivalent to only about 0.2 dex in stellarmass or to about 2 Gyr in time.Fitting the linear relation to the observed f early − log( M/M ⊙ ) trends at a given redshift bin, we can calcu-late the stellar mass at which 50% of galaxies in a givensample is of early type at that redshift. We call thisquantity the morphological crossing mass since it is alsoobviously the mass at which the individual mass func-tions of the two populations cross. The change with red-shift of the morphological crossing mass in the group andother environments is shown in Figure 10. The morpho-logical crossing mass decreases with decreasing redshift,indicating that the morphological transformation fromlate to early types starts in galaxies with higher stellarmasses and shifts to galaxies with lower stellar massesover cosmic time. This is another manifestation of theso-called downsizing scenario in the evolution of galaxies(e.g. Cowie et al. 1996). At z ∼ . M/M ⊙ ) ≈ . 85 arealready of the early type, independent of their environ-ment. Moving toward the lower redshifts, the morpho-logical crossing mass decreases more rapidly for galaxies4 Fig. 9.— Environmental dependence of the fraction of early type galaxies on stellar mass. The red solid pentagons represent the groupgalaxies with the corrected effective richness of at least two in M B < − − z sample of galaxies, the empty circles represent the fieldgalaxies and the green solid triangles represent the isolated galaxies. In the selected redshift intervals galaxies are extracted from the stellarmass complete samples (85% completeness limit for the early 1 + 2 . f early − log( M ∗ /M ⊙ ) relations, marked inthe same colour as the symbols in the corresponding environments. At every redshift, the fraction of early type galaxies increases stronglywith the stellar mass for galaxies residing in both the groups and in the field. The morphological mix that is achieved in the groups atsome redshift and at some mass, is achieved at only slightly later times (or equivalently for slightly higher masses) in the field and for theisolated galaxies. residing in groups than for the field and isolated galaxiesreflecting the emergence of morphological segregation. DISCUSSION The relative role of the stellar mass andenvironment in determining galaxian properties The above results have painted a consistent picture of amorphological segregation emerging between the typicalgroup and field environments (i.e. all galaxies) emerg-ing over the last several billion years, since z ∼ 1, andof being more prominent in galaxies of lower luminosityand/or stellar mass. Corresponding trends in the zCOS-MOS 10k sample relative to the larger scale density fieldwere shown in Tasca et al. (2009).On the one hand, it is clear that this emerging environ-mental dependence is superposed on top of a global andstrong mass-driven evolution that is seen in the wholepopulation. There is no doubt that the stellar, or pos-sibly total baryonic, mass (“nature”) is directly relatedto the processes responsible for this morphological trans-formation. The offset between the observed fractions ofdifferent morphological types of galaxies in different en-vironments since z ∼ 1, i.e. higher fractions of early typegalaxies for the group than the typical field population,which are again higher than the fractions for the isolatedgalaxies (Figure 8), is however tuned by the environment.As noted above, at a given mass, this environmental tun-ing corresponds to about 2 Gyrs in time: a certain mor-phological mix, a product of the efficient morphologicaltransformation, will be achieved first in the group, thenin the field and at the end in the isolated environments for galaxies of a given mass.However, it is not immediately clear if this is becauseof an offset in “starting time”, as the more massive galax-ies are expected to be formed first in the more mas-sive structures (again “nature”, e.g. Kauffmann 1995;Benson et al. 2001; Heavens et al. 2004), or because ofa difference in the rate of evolution in different environ-ments (“nurture”), caused by some of the dense envi-ronment related processes (e.g. ram pressure stripping,harrasment etc, see Introduction).Moreover, we do clearly detect an apparent increasein the environmental differences, at least for the lowermass galaxies which are transforming at the lower red-shifts. This produces the clear divergence of the mor-phological crossing masses from z ∼ . z ∼ . Fig. 10.— Morphological crossing mass of galaxies in differentenvironments. The red solid pentagons represent the group galax-ies, the empty circles represent the field galaxies and the green solidtriangles represent the isolated galaxies. The crossing mass is themass at which 50% of the selected galaxy population is of early (orequivalently of late) type. The stellar masses are estimated fromthe linear fit to the measured points in Figure 9 in the individualredshift intervals. The vertical error bars correspond to the rms val-ues in the crossing mass obtained by bootstrapping of the galaxysamples and repeating the fitting, using the previously estimatederrors on the real fractions. The horizontal error bars correspondto the upper and lower quartiles in the look-back time distributionof galaxies in the considered redshift bins, plotted at the medianlook-back time/redshift in that bin. The morphological crossingmass decreases with decreasing redshift, indicating that the mor-phological transformation from late to early types starts in galaxieswith higher stellar masses and shifts to galaxies with lower stellarmasses over cosmic time. Moving toward the lower redshifts, themorphological crossing mass decreases more rapidly for galaxiesresiding in groups than for the field and isolated galaxies. tween morphological transformation and star-formationquenching. Timescales for the type-transformations of galaxies One interesting question is whether it is possible todetermine whether one of morphological transformationand star-formation quenching preceded the other. Basedon the field stellar mass functions of the zCOSMOSgalaxies of different types, Pozzetti et al. (2009) find thatthe morphological crossing mass M sm,cross has highervalues than the equivalently defined crossing mass forthe populations of photometrically red and blue, andalso active and passive galaxies. From the differencesin the number density of photometrically and morpho-logically early type galaxies, the inferred delay time be-tween the colour and morphological transformation isabout 1-2 Gyr. A similar timescale is derived for thetransformation to red colours after switching of the star-formation. Bundy et al. (2006) find the qualitativelysame behaviour for the crossing mass of the DEEP2field galaxies when population of galaxies is split ac-cording to the morphology and colour, extending thecolour/morphology dichotomy of crossing mass to red-shift 1.4.Similar results have been obtained for the population of cluster galaxies. Wolf et al. (2009) study the popula-tion of galaxies in A901/2 cluster complex at z ∼ . M ⊙ ) to be red and spheroidal in all environments (out-skirt, centre and outside of cluster). Based on thelarge number of cluster spirals in the mass range of10 − M ⊙ , Wolf et al. (2009) conclude that quench-ing of star-formation in these galaxies is a slow process,where the timescale of morphological evolution must belonger than that of the spectral evolution. However,at even lower masses, not accessible by our study, star-formation rate decline and morphological changes appearto be more synchronised in A901/2. Using the sam-ple of 24 clusters from the EDiSCs (White et al. 2005),S´anchez-Bl´azquez et al. (2009) find that the fraction ofearly type red sequence galaxies decreases by ∼ 20% from z = 0 . 75 to z = 0 . 45, what can be expected if the clus-ter spiral galaxies first get redder, stopping their starformation, before they become of early type (note thatS´anchez-Bl´azquez et al. 2009 have done the analysis us-ing the luminosity complete samples).Using the marked correlation statistics on the SDSSdata, Skibba et al. (2008) conclude that z ∼ z ∼ 1) comparison betweenthe colour and morphological properties of galaxies to thegroup environment.To enable a closer comparison between the colour andmorphological properties of galaxies in the group envi-ronment, we calculate the colour crossing mass using thesame procedure and the same galaxy samples as for cal-culation of the morphological crossing mass, describedin Section 4.4 (see also Iovino et al. 2009 who calculatea similar quantity, t − , the cosmic time at which thefraction of red/blue galaxies is 50%). In addition, we alsocalculate the crossing mass for star-formation activity bypartitioning galaxies into active and passive classes us-ing the specific star-formation rate (sSFR) from the SEDfitting (see discussion in Pozzetti et al. 2009 on the agree-ment between the star-formation rates derived from theseSEDs and those from measurement of emission lines). Inthis work, we consider galaxies with U − B > . sSF R/yr ) < − . z ∼ . 6, this evolution is clearly environment dependent.The activity crossing mass is systematically lower thanthe colour crossing mass, which is systematically lowerthan the morphological crossing mass. This is observedboth in the field (in agreement with Pozzetti et al. 2009)and in the group environment. Extrapolating the plot,at a given mass 50% of population will turn to be pas-sive at earlier cosmic time than the time at which 50%of population will become red. 50% of population willbecome of early morphological type even later. More-6 Fig. 11.— Evolution of the class-dependent crossing mass in thegroup and field environments. The filled symbols are for the groupand the empty symbols are for the field samples. The red circles,green squares and violet squares correspond to the morphologi-cal, colour and star-formation activity crossing mass, respectively.Error bars for each value have the same meaning as described inFigure 10. The effective transformation of galaxies from active topassive, from blue to red and from late to early moves progressivelytowards lower mass galaxies as cosmic time passes or as the envi-ronment is denser. The crossing masses in morphology are higher,and change less rapidly, than those for colour and/or star-formationactivity. over, this will generally happen earlier for a group galaxythan for a field galaxy. A similar sequence of the colourand morphological crossing mass in the zCOSMOS hasbeen presented in Bolzonella et al. (2009) for the envi-ronments of extreme densities, when quantifying environ-ment using the local overdensity of neighbouring galaxies(Kovac et al. 2009). Bundy et al. (2006) find a tentativeevidence for the rise of the quiescent population in dense,continuously defined environments.While the exact values of the crossing mass are stronglydependent on how galaxies were divided into types, thereis a clear and progressive (with epoch) environmentaldependence in these values. The effective transforma-tion of galaxies from active to passive, from blue to redand from late to early moves progressively towards lowermass galaxies as cosmic time passes or as the environ-ment is denser. Derived values of the various crossingmasses are consistent with a scenario in which, for themajority of galaxies, the timescale for the morphologi-cal transformation is longer than the timescale for thecolour transformation, both in the field and group envi-ronment, assuming that all these processes started to actapproximately at the same time in a given environment.Unfortunately, it is hard to use arguments such as thelocation and rate of change of the above-defined crossingmass to infer quantitative information on the transfor-mation rates, since these may be sensitive to the choiceof the location of the dividing line between the two prop-erties and also on the shape and relative normalisation ofthe mass functions of different components of the popu-lation. The dividing line may be clear for colour (e.g. theso-called “green valley”) but more arbitrary in the caseof morphological classification. The relative amplitude of the mass functions of early or late type galaxies, and ofblue or red galaxies, or of active or passive galaxies, willalso change the location and movement of M cross . Useof the redshift/epoch axis directly to determine transfor-mation rates, i.e. of the flux of galaxies crossing througha chosen dividing line, is likely to be less sensitive to theprecise location of that divide and to allow a more directstudy of these questions. Role of group environment in the general galaxytype-transformation rate In the following, we explore more directly the “trans-formation rate” of late type galaxies into early typegalaxies for different masses and different environmentsin order to try to better understand the physical causes.It should be noted in the following discussion that thisis really a “net” transformation rate from late to early,since it is in principle possible for early type galaxies toreacquire a gaseous disk and become of later morpho-logical type, although whether this would be sufficientto cross our defined morphological divide between ZESTclasses 2.0 and 2.1 is unclear.Assuming from now on that galaxy transformationsare dominated by one-way transformation late to early(and blue/active to red/passive), we proceed to definea normalised transformation rate η in which the rate ofchange in the number of “progenitor” galaxies (eitherlate type, blue or active) is normalised by the number ofthese progenitors: η = − N C dN C dt = − d ln N C dt (5)where N C is a measure of the number of progenitors of agiven class C , i.e. the mean comoving density averagedover suitably large volumes of the Universe.The inverse of η gives the transformation timescale t η for a typical “progenitor galaxy” to be transformed. Thetransformation rate η might be expected to be a functionof mass, environment and probably epoch.Unfortunately, an accurate determination of N C ( z )and dN C /dt for a particular class C of galaxies is ob-servationally challenging. It requires an accurate mea-surement of galaxian properties, e.g. the stellar mass,over a range of redshifts, careful treatment of large scaledensity variations within surveys due to large scale struc-ture, and, ideally, knowledge of the changes in the num-ber of galaxies in a particular class due to, for instance,mass growth through star-formation and/or merging. Inthe case of galaxies defined by environment, i.e. those ingroups, there will also be the change in the populationdue to the hierarchical growth of the groups through theaccretion of galaxies onto the virialised structures.In cases where it can be assumed that the total num-ber of galaxies in a given bin (defined by mass and en-vironment) is more or less constant, then some of thesedifficulties can be alleviated by considering the fractionof galaxies, rather than their absolute number, i.e. bydividing by a fixed total number of galaxies N . In thiscase: η = − f C df C dt = − d ln f C dt (6)If the evolution of ln f C is linear with cosmic time, we7simply obtain η as the slope to this relation for any sam-ple defined by mass and environment. With the limitedbaseline in redshift and limited number statistics, we willhenceforth consider the rate η as single time-averagedrate over the accessible range of epochs, from a linear fitto ln f C .For group galaxies, we know that the total numberof galaxies is certainly not conserved, because galaxieswill be accreting onto the group from the surroundingregions. It is clear that this migration into the groups islikely to further amplify any environmental difference intransformation rates, because the new arrivals will haveto “catch up” with the existing group members whichalready have a more evolved state.In the following, we derive a simple first order correc-tion accounting for the galaxies infalling to the groupsby assuming (a) that the total number of galaxies in thesample (at a given mass) is conserved, i.e. we do notconsider galaxy mergers and neglect the effects of star-formation, and (b) that the infalling galaxies can be con-sidered to be a representative subset of the non-grouppopulation. This latter condition may not be true, andthis infall correction is therefore likely to represent anupper bound to the actual transformation rate.In some small increment of time dt , the number ofclass-transformations dN in groups will be given by dN = − N G df CG − ( f CG − f CN ) dN G (7)where N G is a number of the group galaxies, f CG and f CN are fractions of the C -class galaxies in the groupsand non-groups, respectively, and where the class C isa “progenitor” type of galaxy (i.e. assumed to be latemorphological type or active/blue type). The equationabove can be further written as dN = − d ( N G f CG ) + f CN dN G . (8)By normalising the number of transformations with N G f CG , which is the “pool” of potential objects in thegroup to be transformed, the rate η IN,CORR includingthe effects of infall can be written as: η IN,CORR = − f CG df CG dt + (cid:18) f CN f CG − (cid:19) f G df G dt . (9)In the case of no infall, where f G is constant over time, orin the case where the group and non-group populationshave the same properties at a given epoch, it can be seenthat the rate η will be given just with the first term onthe left hand side, which is the same as Equation 6 above.What about progenitor bias? By considering two sim-ple complementary categories, group and non-group andthe change in the fraction of galaxies associated witheach, we automatically deal with both additional mem-bership of pre-existing groups and the emergence of newgroups where previously none existed (above the fixedthreshold in richness). We should however recognise thatthe rates so derived will be averages across the entire,evolving, group population and not, necessarily, repre-sentative of what is happening within a single group.Because of the attraction of using two complementarypopulations, group and non-group, we no longer considereither the isolated galaxies nor the field (all galaxies)populations in this analysis. Fig. 12.— Evolution of the fraction of galaxies in our group sam-ple. Different symbols correspond to the different ∆ log( M ∗ /M ⊙ )mass bins, indicated in the right corner of the plot. The dottedlines are the linear fits to the time evolution of the fractions ofgroup galaxies, extrapolated to all epochs. At a given stellar mass,the fraction of galaxies residing in the group environment increaseswith cosmic time, as individual groups grow through infall. It is important to note that, based on extensive com-parisons with the mock catalogues, neither the complete-ness nor purity of the group catalogue are believed todepend significantly on redshift (Knobel et al. 2009, seeFigure 9). The same is also true of the individual groupmembers. This allows us to use the observed fractionof galaxies in a volume-limited sample of groups as agood estimate of the infall rate from the non-group intothe group population, and the observed morphologicalmix of galaxies in and outside of those groups as rep-resentative of both populations. The redshift evolutionof the fraction of galaxies residing in the groups with atleast two corrected members above M B < − . − z isshown in Figure 12. At a given stellar mass, the fractionof galaxies residing in the group environment increasesmarkedly with cosmic time, building up these structures.At a given time, a larger fraction of the more massivegalaxies are already present in the groups reflecting thedifferent mass functions shown above (see Section 4.3 andFigure 7).The derived rates of the morphological transformation η L − >E in the group environment are shown in the left-hand panel in Figure 13. We obtain the infall correctedrates taking for the term f CN f CG the average of the measuredfractions at different times at a given mass, discussed inSubsection 4.4 (see Figure 8). The errors are obtainedby the error progression of Equation 6 and Equation 9,using the individual bootstrap errors on the fractions.The red filled squares represent the rates corrected forinfall using Equation 9, while the red open circles repre-sent those computed directly from Equation 6, which arelower, as expected. The equivalently derived rates forthe non-group galaxies, which are galaxies not residingin the groups with the corrected richness of at least two M B < − . − z members, are shown in the right-hand8 Fig. 13.— Derived transformation rates for morphology, colour and star-formation activity in the group environment (left) and in thenon-group environment (right). The solid symbols correspond to the final rates, which are corrected in the case of the group galaxies, forthe infall of new members onto existing groups and the appearance of new groups above the selection threshold (see Equation 9). It shouldbe noted that this is “worst-case” correction which assumes that the new group members are representative of the previous non-grouppopulation. Using this same assumption, the rates in the non-group population need not be corrected for the loss of these galaxies (seeEquation 6). The open symbols in the left hand plot also correspond to the rates in the groups with no infall correction applied (Equation 6)for the group galaxies. The morphological, colour and star-formation activity rates are shown in red, green and violet, respectively. Thetransformation rates are consistently higher in the group environment than outside of it. The rate of transformation is also consistentlyhigher for the transformations involving star-formation than those involving morphological transformation. Finally the rate is higher forgalaxies around 10 . M ⊙ than for higher mass ones. The rates must also fall again at lower masses, which cannot yet be studied. panel in Figure 13 (red filled squares). We use the simpleEquation 6 above, since the infall is a one-way depletionof the non-group population, again assuming that the in-falling migrants are representative of the population asa whole.In a similar manner, we derive the rates of the colourtransformation from blue to red galaxies η B − >R and fromactive to passive galaxies η A − >P using the same groupand non-group samples of galaxies as for measurementsof η L − >E . Galaxies are divided in different classes asin the previous subsection. The corresponding rates areoverplotted in Figure 13 (using the green symbols for η B − >R and violet symbols for η A − >P ). Generally speak-ing, the rates obtained for the colour and activity type-transformation also show similar dependence on the massand environment as the rates of the morphological trans-formation.Comparison of the two panels of Figure 13 shows threeinteresting facts:(a) The rate of the class-transformation is consistentlyhigher in the group environment than outside of it. (b)The rate of transformation is consistently higher for thetransformations involving star-formation than for thoseinvolving morphological transformation. (c) The rate ishigher for lower mass galaxies than for higher mass ones.Indeed the rates, and the differences between the ratesinside and outside of groups, and between morphologyand star-formation, are all small and consistent with zerofor the most massive galaxies above log( M ∗ /M ⊙ ) ∼ . M ⊙ . It should be noted that these rates must fall at stilllower masses, beyond the range explored by zCOSMOS.For typical galaxies with log( M ∗ /M ⊙ ) ∼ . 5, we con- clude that the transformation rates are 3-4 times higherin the groups than outside, and are typically 50% higherfor colour transformations than for morphological ones.The implied transformation timescales η − , are of order1.6 Gyr (colour transformation) or 2.5 Gyr (morphologytransformation) in the groups, and about 6 Gyr and 10Gyr, respectively, outside of groups.Needless to say, the exact values of the derived ratesand corresponding timescales should be taken as indica-tive only. Not least, the quoted errors include only thepurely statistical uncertainty. There are various otherissues which one needs to be aware of. Although wehave tried to deal with variations in N ( z ) due to thelarge scale structure by considering fractions, the tem-poral gradients may be affected by the choice of redshiftbins, since this will be affected by large scale structure.At low z , the volume used for the analysis is the small-est, and the estimated fractions could be less representa-tive of the universal values than the fractions estimatedat higher redshifts, modifying the the true slopes in theoverall fractional evolution. Moreover, we have assumedconstant rates (i.e a linear evolution in the logarithmicfraction), and have furthermore estimated these at dif-ferent epochs for the different mass ranges. We believethat the relative values should be more robust than theabsolute ones.It is fairly clear, e.g. comparing with Figures 7 and 8that the difference in the transformation rates computedhere stem as much from the different fractions in thedifferent different environments (in the denominator) asfrom differences in the rate of change of the fractions(in the numerator), which are often rather similar. It is9also clear that this difference in the fractions between thedifferent environments amplifies the difference in impliedtransformational rates in the groups through the infallcorrection, increasing the difference by about ∼ M ∗ /M ⊙ ) ∼ . ± . 5. At higher masses the transformation rates arelower because the population has saturated with mosttransformations completed, more than compensating forthe small number of potential progenitor galaxies. Atlower stellar masses, log( M ∗ /M ⊙ ) . 10, not presentlyaccessible by the zCOSMOS observations, the transfor-mation rates must drop again because the vast majorityof galaxies are still in the pre-transformation state.The effect of the environment modifying the trans-formation rates for star-formation activity and colouris larger than the environment’s effect the morpho-logical rates, and we conclude that the environmentseems to be more closely related to the star-formationand colour transformation processes than to the mor-phological transformation of galaxies. Similar con-clusions have been obtained at z ∼ z ∼ 1, especially around the transition or crossing mass,these processes are evidently much more effective ingroups than outside of them. The transformation rateoutside of groups is by no means zero, but might be lowerthan we have estimated because of residual contamina-tion of the non-group population by group galaxies, asdiscussed above.Our physical interpretation of the shorter timescalesthat we attribute to the colour and/or star-formationtransitions, compared with those of the morpholog-ical transformations, is hindered by the uncertaintyabout whether the timescales involved are the physi-cal timescales of the transitions (i.e. where t p ∼ t η ),or what we have called the statistical timescales (where t p ≪ t s ∼ t η ), as discussed in the Introduction. If deal-ing with prompt phenomena which have the same phys-ical timescale, e.g. say the merging of galaxies, then wewould want to interpret the difference in terms of dif-ferent rates of this event, e.g. different merger rates.On the other hand, in the case of longer term processes,such as strangulation, the difference would be more nat-urally interpreted in terms of the speed with which thegiven process acts. Regardless, the fact that the trans-formation timescales vary rather strongly with galacticmass and with group/non-group environment, and also,but less strongly, with transition type, points towards apicture where a variety of physical processes may be inoperation.The current sampling in the 10k zCOSMOS is about ∼ 30% on average, which is not sufficient to meaningfullysplit the population of the group galaxies into central andsatellite galaxies, because two-thirds of the central galax-ies will not have been observed spectroscopically. Thiswill be much easier as the spectroscopic completenessdoubles in the final 20k sample, especially if we use alsothe high quality photometric redshifts. Comparison withthe mock group catalogue (Knobel et al. 2009) suggestthat both group and non group samples (defined using M B < − . − z ) are dominated by central galaxies. CONCLUSIONS We have used the sample of ∼ ∼ 110 km s − ) to study and compare the evolu-tion of galaxies in 0 . < z < B -band magnitudes. There is a systematic increase inthe fraction of early type galaxies in the group and fieldenvironment since z = 1. The build-up of the early typepopulation depends on M B and environment: brighterearly type galaxies are formed before the fainter earlytype galaxies, while early-type galaxies of a given B -bandmagnitude appear earlier in the group than in the fieldenvironment.2. There is some indication within the group popula-tion that the fraction of early types increases with the0group richness and velocity dispersion, at least at red-shifts z < . 65 and for M B < − . − z .3. Mirroring the analysis in Bolzonella et al. (2009),the stellar mass function exhibits a different shape forsamples of galaxies in different environments (groups,field and isolated) at least up to z ∼ . 7. The stellarmass function shows an upturn at low masses and red-shifts in the group environment and more assive galaxiespreferentially reside in the groups. The characteristicshape of the stellar mass functions in the group or forthe isolated galaxies reflects the different relative contri-butions of galaxies of different types (e.g. early and late)in different environments. Due to this mass function dif-ference, the environmental effect on galaxian propertiesmust be examined in narrow bins of stellar mass.4. Looking at narrow mass bins, there is again asystematic build-up of early type galaxies with stellarmasses below ∼ M ⊙ since z = 1. At a given cos-mic time, the fraction of early type galaxies in a givenenvironment increases with stellar mass (“morphologicaldownsizing”). For galaxies with masses below ∼ M ⊙ , the fraction of early type galaxies (at a given stel-lar mass) is always higher in the group than in the fieldenvironment, which is higher still than in isolated galax-ies. Galaxies with masses above ∼ M ⊙ show littleevolution since z = 1, having a roughly constant fractionof early types of ∼ . − . z ∼ . 5, the net effect of the group envi-ronment is equivalent to about 0.2 dex in stellar mass,or 2 Gyr in cosmic epoch.Broadening our analysis to compare directly the nor-malised transition rates in morphology with those involv-ing galaxian colour and star-formation activity, we findthe following additional points:6. The normalised transformation rates are systemat-ically higher in the groups than outside, by a factor of ∼ ∼ − 4. The rates reach val-ues, for masses around the crossing mass 10 . M ⊙ , as high as 0.3 - 0.7 Gyr − in the groups, implying transfor-mation timescales of 1.4 - 3 Gyr, compared with less than0.2 Gyr − , i.e. timescales > . M ⊙ , andwe infer they must also drop at the lower masses below10 M ⊙ that are not presently accessible by our spec-troscopic data.8. The transformation rates are consistently higher,by about 50%, for those transformations involving starformation and colour, than for those involving morpho-logical changes. Although these rate/timescale differ-ences are not trivial to interpret (mostly because of un-certainties in the physical processes involved), they sug-gest faster star-formation quenching than morphologi-cal transformation, and indicate a tighter relation be-tween the environment and star-formation related pro-cesses than with morphology-dynamical processes.8. A clear and important conclusion is that, what-ever their physical origin, those transformational pro-cesses which have driven the evolution of the galaxy pop-ulation (around the characteristic transition or crossingmass of 10 . M ⊙ ) since z ∼ 1, occur faster, or moreefficiently, in the group environment, by a factor of order3. Although the transformation rate is not zero outsideof groups, this suggests that the group environment hasplayed a very significant role in driving the evolution ofthe overall population.The next step in the analysis will involve the forth-coming higher sampled 20k zCOSMOS catalogue. Thiswill not only increase the group sample by a factor of 2.5or so, but will also allow (together with improved photo-z) a much better discrimination between the central andsatellite galaxies within the groups themselves. This willallow a more direct study of their properties (e.g. mor-phologies, colours, star-formation rates) which will putstronger constrain on the transformation mechanisms -both the type of processes and their timescales. Uti-mately, this will provide an answer whether the lifepathsof a galaxy are really different depending on if a galaxy iscentral or satellite (e.g. Bower et al. 2006; Croton et al.2006), or if the differences between the centrals and satel-lites should be more gradual (e.g. Simha et al. 2008). ACKNOWLEDGMENTS This work has been supported in part by a grant fromthe Swiss National Science Foundation and by grantASI/COFIS/WP3110I/026/07/0. REFERENCESBaldry, I. K., Balogh, M. L., Bower, R. G., Glazebrook, K., Nichol,R. C., Bamford, S. P., & Budavari, T. 2006, MNRAS, 373, 469Balogh, M. L. & Morris, S. L. 2000, MNRAS, 318, 703Bamford, S. P., et al. 2009, MNRAS, 393, 1324Barnes, J. E. 1988, ApJ, 331, 699Bell, E. F., Wolf, C., Meisenheimer, K., Rix, H.-W., Borch, A., Dye,S., Kleinheinrich, M., Wisotzki, L., & McIntosh, D. H. 2004, ApJ,608, 752Benson, A. J., Frenk, C. S., Baugh, C. M., Cole, S., & Lacey, C. G.2001, MNRAS, 327, 1041Blanton, M. R., Eisenstein, D., Hogg, D. W., Schlegel, D. J., &Brinkmann, J. 2005, ApJ, 629, 143Bolzonella, M., et al. 2009, arXiv:0907.0013Boselli, A. & Gavazzi, G. 2006, PASP, 118, 517 Bower, R. G., Benson, A. J., Malbon, R., Helly, J. C., Frenk, C. S.,Baugh, C. M., Cole, S., & Lacey, C. G. 2006, MNRAS, 370, 645Bruzual, G. & Charlot, S. 2003, MNRAS, 344, 1000Bundy, K., et al. 2006, ApJ, 651, 120Calzetti, D., Armus, L., Bohlin, R. C., Kinney, A. L., Koornneef,J., & Storchi-Bergmann, T. 2000, ApJ, 533, 682Capak, P., Abraham, R. G., Ellis, R. S., Mobasher, B., Scoville,N., Sheth, K., & Koekemoer, A. 2007a, ApJS, 172, 284Capak, P., et al. 2007b, ApJS, 172, 99Caputi, K. I., et al. 2009, ApJ, 691, 91Carlberg, R. G., Yee, H. K. C., Morris, S. L., Lin, H., Hall, P. B.,Patton, D. R., Sawicki, M., & Shepherd, C. W. 2001, ApJ, 552,427Carollo, C. M.et al. in prep.1