1: \documentclass[usegraphicx,usenatbib]{mn2e}
2: \usepackage{times}
3: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
4: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
5: \def\mpc{h^{-1} {\rm{Mpc}}}
6: \def\kms {\rm{km~s^{-1}}}
7: \def\apj {ApJ}
8: \def\apjl {ApJL}
9: \def\apjs {ApJS}
10: \def\aj {AJ}
11: \def\aap {A\&A}
12: \def\mnras {MNRAS}
13: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
14: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
15: \begin{document}
16: \title[Galaxies in clusters]
17: {On the relationship between environment and galaxy properties in clusters
18: of galaxies}
19: \author[H.J. Mart\'{\i}nez, V. Coenda \& H. Muriel]
20: {H\'ector J. Mart\'\i nez,\thanks{E-mail: julian@oac.uncor.edu}
21: Valeria Coenda \& Hern\'an Muriel\\
22: Instituto de Astronom\'{\i}a Te\'orica y Experimental,
23: IATE, CONICET$-$Observatorio Astron\'omico, Universidad Nacional de
24: C\'ordoba,\\ Laprida 854, X5000BGR, C\'ordoba, Argentina}
25: \date{\today}
26: \pagerange{\pageref{firstpage}--\pageref{lastpage}}
27: \maketitle
28: \label{firstpage}
29: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31: \begin{abstract}
32: We study the correlation between different properties of bright ($L>L^{\ast}$)
33: galaxies in clusters and the environment in the Sloan Digital Sky Survey
34: (SDSS). Samples of clusters of galaxies used in this paper are those selected
35: by Coenda \& Muriel that are drawn from the Popesso et al. and Koester
36: et al. samples. Galaxies in these clusters have been identified in the Main
37: Galaxy Sample of the Fifth Data Release (DR5) of SDSS. We analyse which galaxy
38: properties correlate best with either, cluster mass or cluster-centric
39: distance using the technique by Blanton et al. We find that galaxy properties
40: do not clearly depend on cluster mass for clusters more massive than
41: $M\sim10^{14}M_{\odot}$. On the other hand, galaxy properties correlate
42: with cluster-centric distance. The property most affected by the
43: cluster-centric distance is $g-r$ colour, closely followed by the $u-r$
44: colour. These results are irrespective of the cluster selection criteria.
45: The two samples of clusters were identified based on the X-ray emission and
46: the galaxy colours, respectively. Moreover, the parameter that best predicts
47: environment (i.e. cluster-centric distance) is the same found by Mart\'\i nez
48: \& Muriel for groups of galaxies and Blanton at al. for the local density
49: of field galaxies.
50: \end{abstract}
51: \begin{keywords}
52: galaxies: fundamental parameters -- galaxies: clusters: general --
53: galaxies: evolution
54: \end{keywords}
55: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
56: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
57: \section{Introduction}
58: Galaxy properties depend on environment, the high fraction of early type
59: galaxies in rich cluster being one of the best examples. The fact that this
60: fraction also evolves with time, i.e., an increasing of the S/S0 rate with
61: redshift \citep{dressler97,fasano00}, strongly suggests that galaxy
62: morphologies are being altered by the physical processes that act in the
63: cluster environment. Depending on the latter, i.e., low or high local density,
64: different physical mechanisms will affect galaxy properties in different ways.
65: The fact that most galaxy properties are correlated (morphology, luminosity,
66: colours, etc), makes it difficult to know which properties are affected most
67: by the environment. \citet{blanton05} developed a test to evaluate which
68: property, or pair of properties, are most predictive of the local density
69: using galaxies in the Sloan SDSS \citep{sdss}. \citet{hm2} extended the
70: analysis to galaxies in groups correlating galaxy properties with both, the
71: mass of groups and the position in the system. \citet{blanton05} and
72: \citet{hm2} found that galaxy colour is the property most predictive of the
73: environment. This is particularly surprising taking into account the
74: significant differences between field and group environments.
75:
76: In high mass systems, the hot intracluster medium (ICM) becomes important.
77: Mechanisms like ram pressure stripping \citep{gg72} and
78: starvation/strangulation can affect both the gaseous content of galaxies
79: \citep{abadi99} and the star formation history \citep{fujita99}. From the
80: dynamical point of view, high speed encounters between galaxies are more
81: frequent, producing morphological transformations \citep{moore98,moore99}.
82: Galaxies can also suffer the stripping of gas and stars due to the
83: interaction with the cluster potential (e.g. \citealt{moore99}). It is not
84: clear which of these, or some other proposed mechanisms, are dominant. The
85: fact that different processes affect different galaxy components or
86: properties, means that the correlation between these properties can also
87: vary with the environment. Therefore, the most predictive galaxy properties
88: could also depend on the type of system considered. \citet{yo02},
89: \citet{hm2}, \citet{weinmann06} and \citet{zmm06} found a clear dependence
90: between the galaxy properties and the mass of the host group. For high mass
91: clusters, the correlation is not clear. Several studies did not find
92: dependence between the star formation rate or the fraction of early types with
93: masses or velocity dispersions of clusters (see for instance \citealt{goto03}).
94: Nevertheless, \citet{goto05} and \citet{margo01} found indications of a
95: correlation between blue galaxy fraction with the cluster richness. More
96: recently, \citet{hansen07} have found that the fraction of red galaxies
97: increases with the cluster mass, although only weakly for cluster more massive
98: than $\sim 10^{14} M_{\odot}$. \citet{popesso07} found that a luminous X-ray
99: intracluster medium can affect the colour of galaxies. On the other hand, the
100: local density-morphology \citep{huhu31,oemler74,dressler80} or the
101: cluster-centric distance-morphology relation \citep{whitmore91} has been
102: confirmed by several authors in different environments (see
103: \citealt{dominguez01} for a discussion between these two approaches).
104:
105: In this paper we systematically explore the ability of different galaxy
106: properties to predict the total mass of the cluster and the normalised
107: cluster-centric distance. We have considered two samples of high mass clusters
108: based on different selection criteria.
109:
110: This paper is organised as follows: in section 2 we describe the sample of
111: galaxies in clusters; the analyses of the dependence of galaxy properties on
112: mass and on the cluster-centric distance are carried out in sections 3 and 4
113: respectively. We summarise our results and discuss them in section 5.
114:
115: Galaxy magnitudes used throughout this paper have been corrected for Galactic
116: extinction using the maps by \citet{sch98}, absolute magnitudes have been
117: computed assuming a flat cosmological model with parameters $\Omega_0=0.3$,
118: $\Omega_{\Lambda}=0.7$ and $H_0=100~h~{\rm km~s^{-1}~Mpc^{-1}}$
119: and $K-$corrected using the method of \citet{blanton03}~({\small KCORRECT}
120: version 4.1). All magnitudes are in the AB system.
121: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
122: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
123: \section{The sample of galaxies in clusters}
124: \subsection{The cluster sample}
125: Clusters of galaxies used in this paper has been taken from the cluster
126: catalogue constructed by \citet{coenda08}. This catalogue was drawn from two
127: cluster catalogues based on SDSS: ROSAT-SDSS Galaxy Cluster Survey of
128: Popesso et al. (2004, hereafter P04), which is a X-ray selected cluster sample
129: and the MaxBGC Catalogue of Koester et al. (2007b, hereafter K07), which is an
130: optically selected cluster sample. Briefly, the ROSAT-SDSS catalogue provides
131: X-ray properties of the clusters derived from the ROSAT data, and optical
132: parameters computed from SDSS data. P04 includes clusters with masses from
133: $10^{12.5}M_{\odot}$ to $10^{15}M_{\odot}$ in the redshift range
134: $0.002\leq z \leq 0.45$. On the other hand, the optical MaxBGC catalogue relies
135: on the observation that the galaxy population of rich clusters is dominated by
136: the bright red galaxies tightly clustered in colour (the E/S0 ridgeline). The
137: K07 catalogue comprises galaxy clusters with velocity dispersions
138: $\sigma \ge 400 \kms$ and redshifts $0.1\le z\le 0.3$.
139:
140: The subsamples from P04 and K07 selected by \citet{coenda08}, labelled
141: as C-P04 and C-K07, comprise galaxy clusters in the redshift range
142: $0.05<z<0.14$. For the K07 they also applied a restriction in the richness
143: selecting clusters with $N_{\rm gal}\geq 20$ in order to have cluster masses
144: comparable to those in the P04 sample. To select clusters members and estimate
145: the physical properties of clusters they used the Main Galaxy Sample (MGS;
146: \citealt{mgs}) of the Fifth Data Release (DR5) of SDSS \citep{dr5} that is
147: complete down to a \citet{petro76} magnitude $r=17.77$. To identify cluster
148: members, \citet{coenda08} use the friends-of-friends (\textit{fof}) algorithm
149: developed by Huchra \& Geller (1982) with percolation linking length values
150: according to D\'iaz et al. (2004). As a result, they get for each field a list
151: of substructures with at least 10 members identified by \textit{fof}. The
152: second step consists in a eyeball examination of the structures detected by
153: \textit{fof}, a comparison between them and the listed cluster position and
154: redshift to determine which coordinates and redshift fit best the observed
155: galaxy over-density, i.e., the cluster centre. According to \citet{coenda08},
156: for $\sim 40\%$ of the clusters the angular position of the centre given by
157: \textit{fof} is better than the original value, whereas for $\sim 17\%$ of the
158: clusters the redshift according to \textit{fof} is a better match to the
159: observed distribution than the listed one. From the redshift distribution of
160: galaxies within $|cz-cz_{\rm cluster}|\le 3000\kms$ the authors determine the
161: line-of-sight extension of the cluster, i.e., a maximum and a minimum redshift
162: for the cluster. They then consider as cluster members all galaxies in the
163: field that lie within that redshift range.
164:
165: Through visual inspection \citet{coenda08} classified clusters based on their
166: substructure. For the purposes of this work, we will only consider the
167: subsamples C-P04-I and C-K07-I that comprise regular clusters (type I in
168: \citealt{coenda08}) and exclude systems that have two or more close
169: substructures of similar size in the plane of the sky and/or in the redshift
170: distribution.
171:
172: Once the members of each cluster are selected \citet{coenda08} compute some
173: cluster physical properties we are interested in. Namely, they compute the
174: line-of-sight velocity dispersion $\sigma$, virial radius and mass and the
175: radius which encloses a mean over-density 200 times the mean density of the
176: universe, $r_{200}$.
177:
178: The mean values of these parameters are shown in table \ref{tb:mean} where it
179: can be seen that the subsample drawn from P04 includes on average clusters
180: slightly more massive and larger than the subsample taken from K07. The C-P04-I
181: and C-K07-I galaxy cluster samples comprise 49 and 209 clusters respectively.
182: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
183: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
184: \begin{table}
185: \center
186: \caption{Mean values of the cluster physical properties of our cluster
187: samples.}
188: \begin{tabular}{ccccc}
189: \hline
190: & $\sigma$ & $R_{200}$ & $M_{vir}$ & $R_{vir}$ \\
191: & [$\kms$] & [$\mpc$] & [$h^{-1}M_{\odot}$] &[$\mpc$] \\
192: \hline
193: C-P04-I Sample & $715$ & $1.77$ & $7\times10^{14}$ & $1.75$ \\
194: \hline
195: C-K07-I Sample & $675$ & $1.67$ & $6\times10^{14}$ & $1.59$ \\
196: \hline
197: \end{tabular}
198: \label{tb:mean}
199: \end{table}
200: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
201: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
202: \begin{table}
203: \caption{Parameters' cut-offs that define our samples of galaxies in clusters}
204: \begin{tabular}{lrr}
205: \hline
206: Property & Minimum Value & Maximum Value \\
207: \hline
208: $M_{^{0.1}r}-5\log(h)$ & $-22.5$ & $-20.4$ \\
209: $^{0.1}(u-r)$ & $1.7$ & $3.2$ \\
210: $^{0.1}(g-r)$ & $0.55$ & $1.1$ \\
211: $\mu_{50}$ & $19.25$ & $22$ \\
212: $C=r_{90}/r_{50}$ & $1.9$ & $3.5$ \\
213: $\log(n)$ & $-0.5$ & $1.0$ \\
214: eclass & $-0.25$ & $0.1$ \\
215: \hline
216: \label{cutoffs}
217: \end{tabular}
218: \end{table}
219: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
220: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
221: \subsection{Galaxy parameters}
222: As in \citet{hm2} and to avoid systematic effects, we have constructed volume
223: limited samples of galaxies instead of using flux limited samples with a galaxy
224: weighting scheme to account for Malmquist bias. This is crucial for a fair
225: comparison of galaxies in clusters at different redshifts. Thus we basically
226: deal with galaxies brighter than $M_{^{0.1}r}-5\log(h)=-20.4\simeq M^{\ast}$
227: and $z<0.14$. With this restriction our samples of galaxies comprise: 786
228: galaxies from the C-P04-I and 3041 from the C-K07-I samples.
229:
230: Among the available data for each object in the MGS, we have used in our
231: analyses parameters that are related to different physical properties of the
232: galaxies: luminosity, star formation rate, light distribution inside the
233: galaxies and the dominant stellar populations. The galaxy parameters we have
234: focused our study on are:
235: \begin{enumerate}
236: \item $^{0.1}r-$band absolute magnitude, $M_{^{0.1}r}$.
237: \item $^{0.1}(u-r)$ and $^{0.1}(g-r)$ colours.
238: \item The mono-parametric spectral classification based on the eigentemplates
239: expansion of galaxy's spectrum ${\rm eclass}$. This parameter ranges from about
240: $-0.35$ for early-type galaxies to $0.5$ for late-type galaxies
241: (\citealt{yip04}).
242: \item $r-$band surface brightness, $\mu_{50}$, computed inside the radius that
243: encloses 50\% of Petrosian flux, $r_{50}$.
244: \item $r-$band concentration parameter defined as the ratio between the radii
245: that enclose 90\% and 50\% of the Petrosian flux, $C=r_{90}/r_{50}$.
246: Typically, early-type galaxies have $C>2.5$, while for late-types $C<2.5$
247: \citep{st01}.
248: \item The S\'ersic index $n$ (taken from \citealt{blanton05}).
249: \end{enumerate}
250:
251: It should be noted that in our samples we do not have galaxies with
252: $r_{50}<2\arcsec$, i.e. $0.5\arcsec$ greater than the average seeing in SDSS
253: (the average seeing in SDSS is below a conservative value of $1.5\arcsec$),
254: therefore those galaxy parameters that involve any measure of the galaxy size
255: should not be affected by the effect of seeing.
256:
257:
258: We have introduced some further cut-offs in these galaxy properties besides
259: luminosity, a complete list is shown in Table \ref{cutoffs}. This excludes only
260: a few galaxies from our analyses but is necessary in order to properly bin
261: these quantities in the statistics we perform in next section. It should be
262: mentioned that we do not introduce further cut-offs in the cluster mass nor in the
263: cluster-centric distance. In Figure \ref{fig1} we show the distributions of
264: galaxy parameters.
265: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
266: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
267: \begin{figure}
268: \includegraphics[width=90mm]{f1.eps}
269: \caption{The distributions of galaxy properties in our samples. Dashed blue
270: line: C-K07-I sample, continuous red line: C-P04-I sample. We also show the
271: distribution of cluster mass and the distribution of the projected
272: cluster-centric distance in units of $r_{200}$.}
273: \label{fig1}
274: \end{figure}
275: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
276: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
277: As can be seen in Table \ref{tb:mean} and Figure \ref{fig1}, despite the
278: clusters derived from the Popesso sample are, on average, slightly more massive
279: than those drawn from the Koester sample, galaxy properties in the C-P04-I
280: sample are very similar to those in C-K07-I having the latter a slightly
281: smaller fraction of red galaxies (85\% and 89\% of galaxies redder than
282: $g-r=0.82$, respectively).
283: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
284: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
285: \section{Which galaxy property correlates best with the environment?}
286: In order to determine which galaxy properties are more correlated with, either,
287: cluster mass or the projected distance to the centre of the cluster, we perform
288: here the same analysis carried out by \citet{blanton05} and \citet{hm2}.
289: Details of how to compute the quantities $\sigma_X$ and $\sigma_{XY}$ are given
290: in \citet{hm2}, thus, we will just briefly summarise here their meaning. The
291: galaxy property $X$ which correlates best with a quantity $\delta$ that
292: characterises the environment (either, cluster mass or cluster-centric distance
293: in this work), is the one that minimises the variance of $\delta$ after
294: subtracting the global trend of $\delta$ as a function of $X$. That is, the
295: property $X$ that minimises the expression:
296: \begin{equation}
297: \sigma^2_X=\frac{1}{n-1}\sum_{j=1}^m
298: \sum_{|X_i-\overline{X}_j|\leq\Delta X}(\delta_i-\overline{\delta}_j)^2,
299: \label{sigmaX}
300: \end{equation}
301: in which $X$ has been binned into $m$ bins $2\Delta X$ wide and centred in
302: $\overline{X}_j$ ($j=1,...,m$) and for each of these bins the mean value of
303: $\delta$ is $\overline{\delta}_j$. Clearly, for any property $X$, the quantity
304: $\sigma_X$ will be smaller than the corresponding variance of $\delta$ with no
305: trend subtraction. We label this variance $\sigma$ and present all of our
306: results as the difference $\sigma_X^2-\sigma^2$.
307:
308: The quantity $\sigma_X$ is independent of the units of the physical quantity
309: $X$, but it can be sensitive to the choice of binning. To have robust results
310: we take care that each bin is larger than the mean errors in the considered
311: parameter, is smaller than the features in the parameter's distribution, and
312: contains a large enough number of galaxies. It can be straightforwardly
313: generalised to two properties $X$ and $Y$ if one wants to analyse which pair
314: of properties is most closely correlated with mass \citep{blanton05}. However
315: we do not attempt to do so in this work since we do not have enough objects to
316: split them into as many bins as would be required for a proper computation of
317: $\sigma_{XY}^2$ while still obtaining a reliable outcome.
318: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
319: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
320: \begin{table}
321: \center
322: \begin{tabular}{lcccc}
323: \hline
324: & \multicolumn{2}{c}{C-K07-I sample} & \multicolumn{2}{c}{C-P04-I sample} \\
325: \hline
326: Property $X$ & $\sigma_X^2-\sigma^2$ & Significance& $\sigma_X^2-\sigma^2$ &Significance\\
327: \hline
328: $M_{r}^{0.1}$ & $-4$ &50\%& $-4$ &45\% \\
329: $(u-r)^{0.1}$ & $-3$ &49\%& \fbox{$-8$} &49\% \\
330: $(g-r)^{0.1}$ & $-1$ &46\%& $-3$ &44\% \\
331: $\mu_{50}$ & $-3$ &48\%& $-6$ &47\% \\
332: $C$ & \fbox{$-5$} &53\%& $-7$ &49\% \\
333: $\log(n)$ & $-1$ &45\%& $ 0$ &41\% \\
334: $eclass$ & $-3$ &49\%& $-6$ &48\% \\
335: \hline
336: \end{tabular}
337: \label{tabM}
338: \caption{Galaxy parameters as cluster mass indicators, i.e., in this case
339: $\delta\equiv\log_{10}(M)$ (see text for details). Quoted $\sigma^2$ values are
340: expressed in units of $10^{-4}$. Boxes highlight the lowest values of
341: $\sigma_X^2-\sigma^2$, i.e., those corresponding to the galaxy parameter that
342: predicts best the mass of the clusters. Quoted significances are assessed using
343: the bootstrap technique as described in the text.}
344: \end{table}
345: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
346: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
347: \subsection{Cluster mass}
348: \begin{figure}
349: \includegraphics[width=90mm]{f2.eps}
350: \caption{The mean value of cluster mass as a function of the galaxy properties
351: considered in our analysis. Error bars are errors in the mean value obtained by
352: the bootstrap re-sampling technique. Shaded areas correspond to the mean value
353: obtained in the bootstrap re-samplings plus/minus 1$\sigma$ error bar.}
354: \label{figM}
355: \end{figure}
356: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
357: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
358: In this subsection we focus on how the galaxy properties relate to cluster
359: mass. There is evidence that for groups of galaxies, galaxy colour is, among a
360: set of galaxy properties similar to the ones considered here, the property that correlates best with the mass of the system \citep{hm2,weinmann06}. It is
361: interesting to test whether this is also true for more massive systems. We show
362: in figure \ref{figM} the mean mass of the clusters as a function of galaxy
363: properties. There are some hints of what we would expect, for instance a redder mean colour of galaxies with increasing mass due to the higher fraction of red
364: galaxies. But the trends are not as well defined and strong as one may think
365: {\em a priori}. Shaded areas in figure \ref{figM} correspond to the mean value
366: plus/minus 1 standard deviation error bar for 500 bootstrap re-samplings in
367: which we assign to each galaxy the mass of the cluster to which another,
368: randomly chosen, galaxy in the sample belongs to. It is clear that the trends
369: are, with the exception of a few points, contained in the shaded areas. We
370: should keep in mind that we are dealing here with bright galaxies, that in
371: order to have volume limited samples we have only galaxies brighter than
372: $M^{\ast}$. In Table \ref{tabM} we list the values of the differences
373: $\sigma_X^2-\sigma^2$ for our samples of galaxies in clusters. Results differ
374: from one sample to another. They do not even agree in the single property most
375: predictive of mass, let alone the second or the third.
376:
377: The $\sigma_X$ analysis always provides a parameter that predicts best a given
378: environment. Nevertheless, it does not mean that this parameter is good in
379: predicting the environment. In order to test the significance of the results
380: quoted in Table 3, we again use the bootstrap re-samplings to compute different
381: values of $\sigma_X$, that we label as $\sigma_X^b$, and obtain a distribution
382: for $\sigma_X^b$. We then compute the fraction, $F$, of re-samplings that gave
383: values $\sigma_X^b<\sigma_X$. The significance level of the measured $\sigma_X$
384: is then $1-F$. We list these significance values in Table \ref{tabM}. They
385: confirm what is observed in figure \ref{figM}, that none of the galaxy
386: properties correlates significantly with cluster mass. This may be interpreted
387: as if bright galaxies in clusters were similar from cluster to cluster
388: irrespective of cluster mass.
389: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
390: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
391: \begin{figure}
392: \includegraphics[width=90mm]{f3.eps}
393: \caption{The mean value of cluster-centric distance in units of $r_{200}$ as a
394: function of the galaxy properties considered in our analysis. Error bars are
395: errors in the mean value obtained by the bootstrap re-sampling technique.
396: Shaded areas correspond to the mean value obtained in the bootstrap
397: re-samplings plus/minus 1$\sigma$ error bar.}
398: \label{figR}
399: \end{figure}
400: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
401: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
402: \subsection{Cluster-centric distance}
403: It is well known that spatial segregation occurs in galaxy clusters. We now
404: compute $\sigma_X$ of the galaxies projected distance to the cluster centre
405: (in units of $r_{200}$) as a function of the galaxy properties. For this
406: purpose, we assume that the cluster centres are those determined by
407: \citet{coenda08} as explained in section 2.1. In Figure \ref{figR} we show the
408: mean values of $r/r_{200}$ as a function of the galaxy parameters, as well as
409: the corresponding areas defined by the bootstrap re-samplings. In contrast to
410: what we found for the cluster mass in the figure \ref{figM}, there are clear
411: trends of $r/r_{200}$ that show what is well known, earlier galaxies inhabit
412: the inner regions of the clusters. The resulting values for $\sigma_X^2$ and
413: their significance are quoted in table \ref{tabR}.
414:
415: Now we do find agreement between the two samples for the parameter that
416: correlates best with cluster-centric distance. For both samples the $g-r$
417: colour ranks first. The colour $u-r$ is as good as $g-r$ in the C-P04-I sample,
418: while in the C-K07-I sample comes second, but by a small difference. From the
419: third position onwards rankings differ. The significance is greater than 68\%
420: only for 3 properties: the two colours and eclass. This result is consistent
421: with \citet{hm2}, who found that the $g-r$ colour is the single parameter that
422: correlates best with the distance from the centre of the system in groups of
423: galaxies.
424:
425: In the upper left panel of figure \ref{figR} the point corresponding to the
426: highest luminosities considered is well below the shaded area in both samples.
427: We tested whether this is due to the brightest galaxy (BCG) alone by repeating
428: the computations excluding the brightest galaxy of each cluster. In the
429: brightest luminosity bin, 88\% and 60\% of the galaxies are BCGs for the
430: C-K07-I and the C-P04-I samples, respectively. The resulting trend for the
431: C-K07-I sample is kept almost unchanged, while for the C-P04-I sample the
432: point lifts into the shaded area. However, it is important to take into account
433: that the actual brightest galaxy of each cluster is not always included in the
434: samples (this is explored in detail in \citealt{coenda08}).
435: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
436: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
437: \section{Discussion and conclusions}
438: In this paper we have extended the analyses by \citet{blanton05} and
439: \citet{hm2} to galaxies in clusters. We analyse which galaxy properties
440: correlate best with environment characterised by either cluster mass or
441: cluster-centric distance. We use a sample of bright ($L>L^{\ast}$) galaxies in clusters
442: of galaxies in SDSS identified by two different criteria: X-ray selected
443: clusters and clusters selected according to their red sequence. The two
444: sub-samples of clusters have some differences in the mean properties. X-ray
445: selected clusters tend to have a slightly higher mean mass and a higher
446: fraction of red galaxies than maxBCG systems.
447:
448: We find that the properties of bright galaxies do not clearly depend on cluster
449: mass for
450: systems more massive than $M\sim10^{14}M_{\odot}$. Although the mass range of
451: our sample of cluster is not very large, the lack of dependence between mass
452: and galaxy properties can be interpreted in terms of galaxy evolution. For
453: systems with masses between $10^{14}M_{\odot}$ and $10^{15}M_{\odot}$ bright
454: galaxies have experienced the same physical processes and therefore have
455: similar properties. This result is consistent with \citet{hansen07}. Using
456: clusters and groups identified in SDSS with the MaxBCG finder of K07, these
457: authors find that above a cluster mass $\sim10^{14}M_{\odot}$ the fraction of
458: red galaxies increases weakly with increasing mass. For bright galaxies, the
459: lack of a significant correlation between some galaxy properties such as colour
460: and concentration with halo mass for halo masses above $\sim 10^{14}M_{\odot}$
461: is also present in \citet{weinmann06} (their figure 11).
462: To check consistency with their findings we have computed
463: the median of $^{0.1}(g-r)$ colour and the median of the
464: concentration parameter as a function of mass for our samples. We have found
465: that they are remarkably independent on cluster mass, taking
466: values $\sim0.95$ and $\sim3.0$, respectively, in fully agreement with the
467: higher mass bins of \citet{weinmann06}.
468: For lower mass systems, the importance of processes like high
469: speed encounters or those produced by
470: the intra-cluster medium, tends to change faster with mass.
471: \citet{yo02} and \citet{hm2}
472: analysed groups of galaxies and found a clear dependence between mass and
473: galaxy properties that tends to flatten as the mass of the systems grows. The
474: higher masses considered by these authors are similar to the lower masses
475: analysed in this work.
476: All these results seem to imply that above a mass $\sim10^{14}M_{\odot}$
477: clusters have a similar population of bright ($L>L^{\ast}$) galaxies, that
478: arises as a result of the action of the same physical mechanisms with similar
479: relative impact.
480:
481: On the other hand, we find, as expected, that galaxy properties do correlate
482: with cluster-centric distance. The property most affected by the
483: cluster-centric distance is $g-r$ colour, followed closely by the $u-r$ colour.
484: These results are irrespective of the cluster selection criteria. This is also
485: in agreement with \citet{hansen07}. Moreover, the parameter that best
486: predicts the cluster-centric distance is the same found by \citet{hm2} for
487: groups of galaxies and \citet{blanton05} for the local density of field
488: galaxies as the most predictive of environment.
489:
490: Galaxy parameters considered in this work can be classified into two classes.
491: Those related to the physical properties of stars and those associated with the
492: light distribution. In the first set are colour, absolute magnitude and
493: spectral type. To the second group belong the concentration parameter, the
494: surface brightness and the S\'ersic index. Colour and spectral types of
495: galaxies strongly depend on the age and metallicity of the stars as well as
496: the present and recent star formation history. The luminosity of galaxies also
497: depends on the same properties, although in a more indirect way. The fact that
498: from field to massive clusters colour is the most sensitive property of
499: galaxies to the present time environment, suggests that what really matters is the
500: overall evolution of the environment where a galaxy and its progenitors
501: form the stars. Moreover, according to \citet{blanton05} and \citet{hm2}, the
502: other two parameters that appear as one of the pair of properties most
503: predictive of the environment for field and group's galaxies are the magnitude
504: and the spectral type, both belonging to the first group of parameters.
505:
506: Among
507: the phenomena that can affect the star formation history, we can mention the
508: suppression or stimulation of the star formation due to interactions with the
509: intergalactic medium or with other galaxies. Also, the past merger history of
510: galaxies can play a fundamental role. In the particular case of the cluster
511: environment, a natural segregation in the colour of galaxies arises as a
512: result of cluster-centric gradient in the age of the stellar population as is
513: observed in numerical simulations (e.g. \citealt{gao}). Galaxies in the inner
514: regions of clusters would have older stellar populations and therefore would be
515: redder. These galaxies would also have had a longer time to deplete their gas
516: reservoirs thus stopping star formation. In this scenario, the reddening as a
517: function of radius emerges naturally.
518: However, these processes might suffer a saturation resulting in a flattening
519: of the mean colour as a function of mass.
520:
521: The lack of a significant correlation between environment and the galaxy properties
522: in the second set described above, may imply that phenomena
523: like ram pressure stripping or galaxy harassment would have had a secondary
524: role in the evolution of bright galaxies.
525: Nevertheless, these physical processes would also have an impact in the first
526: set of parameters. For example, ram pressure striping can
527: remove an important fraction of the intragalactic gas producing a reddening
528: of galaxies as a consequence of the reduction of the star formation rate.
529: Numerical simulations show that galaxies can lose a high fraction of the gas
530: after a single passage through the inner regions of a cluster (see for instance
531: \citealt{abadi99}). Therefore, above a certain mass ($\sim10^{14}M_{\odot}$),
532: galaxies will experience the same physical processes acting with similar
533: relative effectiveness thus producing a saturation in the mass-colour
534: relationship.
535: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
536: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
537: \begin{table}
538: \center
539: \begin{tabular}{lcccc}
540: \hline
541: & \multicolumn{2}{c}{C-K07-I sample} & \multicolumn{2}{c}{C-P04-I sample} \\
542: \hline
543: Property $X$ & $\sigma_X^2-\sigma^2$ & Significance& $\sigma_X^2-\sigma^2$ &Significance\\
544: \hline
545: $M_{r}^{0.1}$ & $-7$ &67\%& $-7$ &50\% \\
546: $(u-r)^{0.1}$ & $-13$ &84\%& \fbox{$-16$} &81\% \\
547: $(g-r)^{0.1}$ & $\fbox{$-14$}$ &85\%& \fbox{$-16$} &79\% \\
548: $\mu_{50}$ & $-2$ &55\%& $-8$ &61\% \\
549: $C$ & $-5$ &63\%& $-9$ &64\% \\
550: $\log(n)$ & $-4$ &60\%& $-3$ &49\% \\
551: $eclass$ & $-8$ &70\%& $-7$ &69\% \\
552: \hline
553: \end{tabular}
554: \label{tabR}
555: \caption{Galaxy parameters as cluster-centric distance indicators, i.e., in
556: this case $\delta\equiv r/r_{200}$ (see text for details). Quoted $\sigma^2$
557: values are expressed in units of $10^{-3}$. Boxes highlight the lowest values
558: of $\sigma_X^2-\sigma^2$, i.e., those corresponding to the galaxy parameter
559: that predicts best the cluster-centric distance. Quoted significances are
560: assessed using the bootstrap technique as described in the text.}
561: \end{table}
562: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
563: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
564: \section*{Acknowledgements}
565: We thank the anonymous referee for comments and suggestions that have
566: improved this paper.
567: HJM acknowledges the support of a Young Researcher's grant from Agencia
568: Nacional de Promoci\'on Cient\'\i fica y Tecnol\'ogica Argentina, PICT
569: 2005/38087. This work has been partially supported with grants from Consejo
570: Nacional de Investigaciones Cient\'\i ficas y T\'ecnicas de la Rep\'ublica
571: Argentina (CONICET) and Secretar\'\i a de Ciencia y Tecnolog\'\i a de la
572: Universidad de C\'ordoba.
573:
574: Funding for the Sloan Digital Sky Survey (SDSS) has been provided by the
575: Alfred P. Sloan Foundation, the Participating Institutions, the National
576: Aeronautics and Space Administration, the National Science Foundation, the U.S.
577: Department of Energy, the Japanese Monbukagakusho, and the Max Planck Society. The SDSS Web site is http://www.sdss.org/. The SDSS is managed by the
578: Astrophysical Research Consortium (ARC) for the Participating Institutions.
579: The Participating Institutions are The University of Chicago, Fermilab, the
580: Institute for Advanced Study, the Japan Participation Group, The Johns Hopkins
581: University, the Korean Scientist Group, Los Alamos National Laboratory, the
582: Max Planck Institut f\"ur Astronomie (MPIA), the Max Planck Institut f\"ur
583: Astrophysik (MPA), New Mexico State University, University of Pittsburgh,
584: University of Portsmouth, Princeton University, the United States Naval
585: Observatory, and the University of Washington.
586: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
587: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
588: \begin{thebibliography}{}
589: \bibitem[Abadi et al.(1999)]{abadi99}
590: Abadi, M.~G., Moore, B., \& Bower, R.~G.\ 1999, \mnras, 308, 947
591: \bibitem[Adelman-McCarthy et al.(2007)]{dr5}
592: Adelman-McCarthy, J.~K., et al.\ 2007, \apjs, 172, 634
593: \bibitem[Beers et al.(1990)]{beers90}
594: Beers, T.~C., Flynn, K., \& Gebhardt, K.\ 1990, \aj, 100, 32
595: \bibitem[Blanton et al.(2003)]{blanton03}
596: Blanton, M.~R., et al.\ 2003, \aj, 125, 2348
597: \bibitem[Blanton et al.(2005)]{blanton05}
598: Blanton, M.~R., Eisenstein, D., Hogg, D.~W., Schlegel, D.~J., \& Brinkmann,
599: J.\ 2005, \apj, 629, 143
600: \bibitem[Carlberg et al.(1997)]{carlberg97}
601: Carlberg, R.~G., Yee, H.~K.~C., \& Ellingson, E.\ 1997, \apj, 478, 462
602: \bibitem[Coenda \& Muriel (2008)]{coenda08}
603: Coenda, V., Muriel, H., \ 2008, in preparation.
604: \bibitem[Dom{\'{\i}}nguez et al.(2001)]{dominguez01}
605: Dom{\'{\i}}nguez, M., Muriel, H., \& Lambas, D.~G.\ 2001, \aj, 121, 1266
606: \bibitem[Dressler(1980)]{dressler80}
607: Dressler, A.\ 1980, \apj, 236, 351
608: \bibitem[Dressler et al.(1997)]{dressler97}
609: Dressler, A., et al.\ 1997, \apj, 490, 577
610: \bibitem[Fasano et al.(2000)]{fasano00}
611: Fasano, G., Poggianti, B.~M., Couch, W.~J., Bettoni, D., Kj{\ae}rgaard, P.,
612: \& Moles, M.\ 2000, \apj, 542, 673
613: \bibitem[Fujita \& Nagashima(1999)]{fujita99}
614: Fujita, Y., \& Nagashima, M.\ 1999, \apj, 516, 619
615: \bibitem[Gao et al.(2005)]{gao}
616: Gao, L., Springel, V., \& White, S.~D.~M.\ 2005, \mnras, 363, L66
617: \bibitem[Girardi et al.(1993)]{girardi93}
618: Girardi, M., Biviano, A., Giuricin, G., Mardirossian, F., \& Mezzetti,
619: M.\ 1993, \apj, 404, 38
620: \bibitem[Goto et al.(2003)]{goto03}
621: Goto, T., Yamauchi, C., Fujita, Y., Okamura, S., Sekiguchi, M., Smail, I.,
622: Bernardi, M., \& Gomez, P.~L.\ 2003, \mnras, 346, 601
623: \bibitem[Goto(2005)]{goto05}
624: Goto, T.\ 2005, \mnras, 356, L6
625: \bibitem[Gunn \& Gott(1972)]{gg72}
626: Gunn, J.~E., \& Gott, J.~R.~I.\ 1972, \apj, 176, 1
627: \bibitem[Hansen et al.(2007)]{hansen07}
628: Hansen, S.~M., Sheldon, E.~S., Wechsler, R.~H., \& Koester, B.~P.\ 2007,
629: ArXiv e-prints, 710, arXiv:0710.3780
630: \bibitem[Hubble \& Humason(1931)]{huhu31}
631: Hubble, E., \& Humason, M.~L.\ 1931, \apj, 74, 43
632: \bibitem[Koester et al.(2007a)]{koester07a}
633: Koester, B.~P., et al.\ 2007, \apj, 660, 221
634: \bibitem[Koester et al.(2007b)]{koester07b}
635: Koester, B.~P., et al.\ 2007, \apj, 660, 239
636: \bibitem[Margoniner et al.(2001)]{margo01}
637: Margoniner, V.~E., de Carvalho, R.~R., Gal, R.~R., \& Djorgovski, S.~G.\ 2001,
638: \apjl, 548, L143
639: \bibitem[Mart{\'{\i}}nez et al.(2002)]{yo02}
640: Mart{\'{\i}}nez, H.~J., Zandivarez, A., Dom{\'{\i}}nguez, M., Merch{\' a}n,
641: M.~E., \& Lambas, D.~G.\ 2002, \mnras, 333, L31
642: \bibitem[Mart{\'{\i}}nez \& Muriel(2006)]{hm2}
643: Mart{\'{\i}}nez, H.~J., \& Muriel, H.\ 2006, \mnras, 370, 1003
644: \bibitem[Merch{\'a}n \& Zandivarez(2005)]{mz05}
645: Merch{\'a}n, M.~E., \& Zandivarez, A.\ 2005, \apj, 630, 759
646: \bibitem[Moore et al.(1998)]{moore98}
647: Moore, B., Lake, G., \& Katz, N.\ 1998, \apj, 495, 139
648: \bibitem[Moore et al.(1999)]{moore99}
649: Moore, B., Lake, G., Quinn, T., \& Stadel, J.\ 1999, \mnras, 304, 465
650: \bibitem[Oemler(1974)]{oemler74}
651: Oemler, A.~J.\ 1974, \apj, 194, 1
652: \bibitem[Petrosian(1976)]{petro76}
653: Petrosian, V.\ 1976, \apjl, 209, L1
654: \bibitem[Popesso et al.(2007)]{popesso07}
655: Popesso, P., Biviano, A., Romaniello, M., \ B\"ohringer, H.\ 2007, \aap,
656: 461, 411
657: \bibitem[Schlegel et al.(1998)]{sch98}
658: Schlegel, D.~J., Finkbeiner, D.~P., \& Davis, M.\ 1998, \apj, 500, 525
659: \bibitem[Strateva et al. (2001)]{st01}
660: Strateva, I., et al.\ 2001, \aj, 122, 1861
661: \bibitem[Strauss et al.(2002)]{mgs}
662: Strauss, M.~A., et al.\ 2002, \aj, 124, 1810
663: \bibitem[Weinmann et al.(2006)]{weinmann06}
664: Weinmann, S.~M., van den Bosch, F.~C., Yang, X., \& Mo, H.~J.\ 2006, \mnras,
665: 366, 2
666: \bibitem[Whitmore \& Gilmore(1991)]{whitmore91}
667: Whitmore, B.~C., \& Gilmore, D.~M.\ 1991, \apj, 367, 64
668: \bibitem[Yip et al.(2004)]{yip04} Yip, C.~W., et al.\ 2004,
669: \aj, 128, 585
670: \bibitem[York et al.(2000)]{sdss}
671: York, D.~G., et al.\ 2000, \aj, 120, 1579
672: \bibitem[Zandivarez et al.(2006)]{zmm06}
673: Zandivarez, A., Mart{\'{\i}}nez, H.~J., \& Merch{\'a}n, M.~E.\ 2006, \apj, 650,
674: 137
675: \end{thebibliography}
676: \label{lastpage}
677: \end{document}
678: