1: \documentclass{emulateapj}
2:
3: \shorttitle{Galaxy Groups in the SDSS-DR3}
4: \shortauthors{Merch\'an and Zandivarez}
5:
6: \begin{document}
7:
8: \def\mpc{h^{-1} {\rm{Mpc}}}
9: \def\up{h^{-3} {\rm{Mpc^3}}}
10: \def\uk{h {\rm{Mpc^{-1}}}}
11: \def\lsim{\mathrel{\hbox{\rlap{\hbox{\lower4pt\hbox{$\sim$}}}\hbox{$<$}}}}
12: \def\gsim{\mathrel{\hbox{\rlap{\hbox{\lower4pt\hbox{$\sim$}}}\hbox{$>$}}}}
13:
14: \title{Galaxy Groups in the Third Data Release of the SDSS}
15:
16: \author{Manuel E. Merch\'an\altaffilmark{1}, Ariel Zandivarez\altaffilmark{1}}
17: \affil{Grupo de Investigaciones en Astronom\'{\i}a Te\'orica y Experimental,
18: IATE, Observatorio Astron\'omico, Laprida 854, C\'ordoba, Argentina.}
19:
20: \altaffiltext{1}{Consejo de Investigaciones Cient\'{\i}ficas y T\'ecnicas de la Rep\'ublica Argentina.}
21:
22: \begin{abstract}
23: We present a new sample of galaxy groups identified in the Sloan Digital
24: Sky Survey Data Release 3. Following previous works we use the well tested
25: friend-of-friend algorithm developed by Huchra \& Geller which take into
26: account the number density variation due to the apparent magnitude limit of
27: the galaxy catalog. To improve the identification we
28: implement a procedure to avoid the artificial merging of small systems in high
29: density regions and then apply an iterative method to
30: recompute the group centers position.
31: As a result we obtain a new catalog with 10864 galaxy groups
32: with at least four members. The final group sample has a mean redshift
33: of 0.1 and a median velocity dispersion of $230 \ km \ s^{-1}$.
34: \end{abstract}
35:
36: \keywords{galaxies: clusters : general}
37:
38: \section{Introduction}
39: Galaxy group samples have become a very important issue in cosmology.
40: Since hierarchical clustering drives the structure formation in the
41: universe, galaxy groups can be considered a fundamental piece of the chain
42: joining galaxies and clusters of galaxies.
43:
44: The information obtained from these
45: systems allows us to understand what the internal processes
46: ruling the intragroup medium are and what the most suitable cosmological
47: model to describe their distribution is.
48:
49: The first large samples of groups comprised around $\sim1000$
50: galaxy groups and were constructed from different redshift surveys
51: (Merch\'an, Maia \& Lambas 2000, Giuricin et al 2000, Tucker et al. 2000,
52: Ramella et al. 2002).
53: More recently larger galaxy group catalogs were constructed from
54: different releases of the 2dF Galaxy Redshift Survey (hereafter, 2dFGRS).
55: The first one was constructed from the 2dFGRS 100K Data Release by
56: Merch\'an \& Zandivarez (2002) with a total of $\sim2200$ groups while
57: using the final release of the 2dFGRS, Eke et al. (2004a)
58: constructed a catalog containing $\sim7000$ galaxy groups with at least
59: four members.
60: Research involving these samples comprise studies ranging from local physical
61: properties (Mart\'{\i}nez et al. 2002, Dom\'{\i}nguez et al. 2002,
62: D\'{\i}az et al. 2004, Eke et al. 2004b, Ragone et al. 2004) to large
63: scale structure (Zandivarez, Merch\'an \& Padilla 2003, Padilla et al. 2004).
64:
65: At the present, the largest galaxy redshift survey is the Third Data
66: Release of the Sloan Digital Sky Survey (hereafter, SDSS-DR3).
67: This release has approximately $\sim530000$ spectra of
68: which $\sim380000$ are galaxies with a redshift accuracy of $30 \ km \
69: s^{-1}$.
70: The size and deepness of this sample, turns it an ideal source of information
71: to obtain a new galaxy group sample.
72:
73: The aim of this work is to construct a group catalog from the
74: galaxies in the Third Data Release of the SDSS.
75: The group identification is performed using the algorithm developed by
76: Huchra \& Geller (1982).
77: We also apply a technique to improve the identification for groups
78: with at least ten members.
79:
80: A description of the group identification algorithm is given in Section 2
81: while the identification on galaxies and the subsequent improvements are
82: carried out in section 3. Finally, in section 4 we summarize our results.
83:
84: \section{The group-finding algorithm}
85: Group identification is performed using the friend-of-friend algorithm
86: developed by Huchra \& Geller (1982).
87: According to these authors, given a pair of galaxies with mean radial
88: comoving distance $D=(D_1+D_2)/2$ and angular separation $\theta_{12}$,
89: the algorithm links galaxies satisfying the following conditions:
90: \begin{equation}
91: D_{12}=2 \sin\left(\frac{\theta_{12}}{2}\right)D \leq D_L = D_0 R
92: \end{equation}
93: and
94: \begin{equation}
95: V_{12}=|V_1-V_2| \leq V_L=V_0 R
96: \end{equation}
97: where $D_{12}$ is the projected distance and $V_{12}$ is the line-of-sight
98: velocity difference. Comoving distances $D_i$ are estimated using the
99: relation corresponding to an Einstein-de Sitter model
100: \begin{equation}
101: D_i(z)=\frac{c}{H_0}\int_0^z \frac{dz'}{\Omega_M (1+z')^3+\Omega_{\Lambda}}
102: \end{equation}
103: with density parameters $\Omega_M=0.3$, $\Omega_{\Lambda}=0.7$ and
104: $H_0=100 h \ km \ s^{-1} \ Mpc^{-1}$.
105: The transverse ($D_L $) and radial ($V_L$) linking lengths
106: scale with $R$, to compensate for the number density variation due to
107: the apparent magnitude limit of the survey.
108: The scaling factor $R$ is computed using the galaxy luminosity function of the
109: sample $\phi(M)$:
110: \begin{equation}
111: R = \left[\frac{\int_{-\infty}^{M_{12}}\phi(M)dM}{\int_{-\infty}^{M_{lim}}\phi(M)dM}\right]^{-1/3}
112: \end{equation}
113: where $M_{lim}$ and $M_{12}$ are the absolute magnitude of the brightest
114: galaxy visible at the fiducial $D_f$ and mean galaxy $D$ distances,
115: respectively.
116: Usually $D_0$ is chosen in order to obtain the desired overdensity
117: $\delta\rho/\rho$, which is given by
118: \begin{equation}
119: \frac{\delta \rho}{\rho}=\frac{3}{4\pi D_0^3}\left(\int_{-\infty}^{M_{lim}}\phi(M)dM\right)^{-1}-1
120: \end{equation}
121: and correspond to a fixed overdensity contour surrounding a group.
122: \begin{figure*}
123: \epsscale{0.85}
124: \plotone{f1.ps}
125: \caption{
126: Pieplots showing the galaxy groups in the SDSS-DR3. The
127: galaxy groups distribution is showed in three different ranges of right
128: ascension and for redshifts $z\leq0.15$.
129: \label{fig1}
130: }
131: \end{figure*}
132:
133:
134: \section{Galaxy groups sample}
135: \subsection{The galaxy catalog}
136: The Sloan Digital Sky Survey has validated and made publicly
137: available the Third Data Release (Abazajian et al. 2004).
138: This catalog is a photometric and spectroscopic survey covering
139: $4188 \ deg^2$
140: with five-band ($u \ g \ r \ i \ z$) imaging data and 528640
141: spectra of galaxies, quasars and stars.
142: In this work we use the main spectroscopic galaxy sample which
143: comprises $\sim 300000$ galaxies with redshifts
144: $z \leq 0.3$ and an upper apparent magnitude limit of 17.77 in the r-band.
145:
146:
147: \subsection{Group Identification}
148: The adopted linking lengths values are motivated by the
149: group identification analysis performed by
150: Merch\'an \& Zandivarez (2002) for the 2dFGRS.
151: Using mock catalogs, they explore a wide range of linking lengths values
152: in order to maximize the group identification accuracy.
153: Since the SDSS-DR3 and 2dFGRS have similar redshift distributions and
154: luminosity functions (Norberg et al. 2002),
155: we choose identical values for the group identification:
156: a transversal linking length corresponding to an overdensity of
157: $\delta \rho/\rho=80$, a line-of-sight linking length of
158: $V_0=200 \ km \ s^{-1}$ and a fiducial distance $D_f=10 \ h^{-1} Mpc $.
159: The scaling factor $R$ is estimated using a galaxy luminosity function
160: fitted using a Schecter function with parameters ($\alpha=-1.05\pm 0.01,
161: M_{\ast}-5\log{h}=-20.44\pm 0.01$) given by Blanton et al. (2003).
162: The main properties of the obtained groups sample
163: are summarized in Table \ref{tab1} (SDSS-DR3 First identification).
164:
165: \subsection{Rich groups identification improvement}
166: \begin{figure}
167: \epsscale{0.85}
168: \plotone{f2.ps}
169: \caption{
170: Velocity dispersion (upper panel) and redshift (lower panel)
171: distributions for rich groups. The solid line are the distributions
172: for groups with at least ten members after the first identification
173: while the dotted line shows the group distributions when the corrections
174: on the identification has been applied.
175: \label{fig2}
176: }
177: \end{figure}
178:
179:
180:
181: It should be taken into account that groups obtained from galaxy redshift
182: surveys have unavoidable contamination problems.
183: By instance, the method described previously, can not fully avoid the
184: interloper effect (i.e. spurious inclusion of non-member galaxies),
185: so an artificial merging of small groups with large systems is
186: likely to happen.
187: Recently, working with groups identified in the 2dFGRS Final Release
188: and the First Release of the SDSS, D\'{\i}az et al. (2004) have developed
189: a method to minimize this problem on group identification.
190: This method propose a second identification on galaxy groups with at least
191: ten members in order to split merged systems or eliminate spurious member
192: detections.
193: Analyzing group identifications on mock catalogs in real and redshift
194: space, they found that a higher value for $\delta \rho/\rho$ ($\sim315$)
195: produce a more reliable group identification.
196: Performing a redshift space identification with this density contrast
197: is equivalent to a group identification in real space corresponding to
198: $\delta \rho/\rho=80$.
199: Furthermore, as shown by M\'erchan \& Zandivarez (2002), adopting lower
200: transversal linking length values, inproves the identification accuracy
201: (see Figures 2 and 3 therein).
202: Therefore, following the suggestion of D\'{\i}az et al. (2004), we
203: perform a second identification for groups with at least ten members.
204:
205: A very accurate group center determination is required for some studies
206: related with the galaxy distribution inside groups.
207: Trying to measure density profiles of galaxy groups, D\'{\i}az et al. (2004)
208: have also proposed a method for correcting group center position for those
209: systems with at least ten members.
210: Firstly, their method define a new group center estimator by using the
211: projected local number density of each member galaxy for weighting their
212: group center distances.
213: The second part of the method is an iterative procedure to improve the
214: group center location by removing galaxies beyond a given distance and
215: recomputing the center position.
216: The procedure follows until the center location remains unchanged (for more
217: details see section 2 of D\'{\i}az et al. 2004).
218: Given that the described procedure needs to compute the projected local
219: number density with five galaxies, this method only can be applied
220: to groups with at least ten members.
221: After applying this correction, the full catalog comprise 9703 galaxy
222: groups with at least 4 members.
223: Figure \ref{fig1} shows the spatial distribution of groups identified
224: in the SDSS-DR3 with at least 4 members and $z \leq 0.15$.
225:
226: \subsection{Group physical properties}
227: \begin{figure}
228: \epsscale{0.85}
229: \plotone{f3.ps}
230: \caption{
231: Histograms of the redshift (upper-left panel), velocity dispersion
232: (upper-right panel), virial radius (lower-left panel) and virial mass
233: (lower-right panel) distributions of groups in the SDSS-DR3. Solid lines
234: represent the histograms for galaxy groups after the first identification,
235: while dotted lines show the histograms for the galaxy group sample after
236: improving the rich groups identification.
237: \label{fig3}
238: }
239: \end{figure}
240:
241:
242: In addition to the group identification, we also estimate its basic physical
243: properties such as velocity dispersion, virial radius and virial mass.
244:
245: The line-of-sight velocity dispersion $\sigma_{v}$,
246: is estimated using the methods described by
247: Beers, Flynn \& Gebhardt (1990).
248: The biweight estimator is applied for groups
249: with richness $N_{tot}\ge 15$ whereas the gapper estimator is applied to
250: poor groups (Girardi et al. 1993, Girardi \& Giuricin 2000). Following
251: Eke et al. (2004) we also introduce a correction due to the error in the
252: redshift measurement. This is a second order correction since the
253: redshift measurement error in the SDSS is $30 \ km \ s^{-1}$.\\
254: The virial radius is estimated using the following equation
255: \begin{eqnarray}
256: R_V & = & \frac{\pi}{2}\frac{N_g(N_g-1)}{\sum_{i>j}R_{ij}^{-1}}
257: \end{eqnarray}
258: where $N_g$ is the number of galaxy members and $R_{ij}$ the galaxy
259: projected distances.\\
260: Finally the virial masses of galaxy groups is computed as
261: \begin{equation}
262: M_{V}=\frac{3\sigma_{v}^2 R_V}{G}
263: \end{equation}
264: where $G$ is the gravitational constant.
265:
266: Individual group velocity dispersion estimation, allow us to observe
267: rich group reideintification and recentering consequences.
268: Figure \ref{fig2}
269: illustrate the rich group ($N_g \ge 10 $) velocity dispersion behavior
270: (upper panel) for groups before and after correction (dotted and solid lines,
271: respectively).
272: As can be seen, the distribution of groups with high
273: velocity dispersions ($\sigma_v \gsim 300 \ km \ s^{-1}$) remains unchanged,
274: whereas artificially merged groups in the first identification, now appears as
275: a new population with low velocity dispersion.
276: The redshift distribution for groups with at least ten members is shown
277: in the lower panel of Figure \ref{fig2},
278: It should be noted that the correction applied to
279: rich groups do not introduce any bias with redshift.
280: Histograms showing the distribution of redshift, velocity dispersion,
281: virial radius and virial mass of groups with at least 4 members are plotted
282: in Figure \ref{fig3}.
283: Finally the median physical properties of the full group catalog
284: are quoted in Table \ref{tab1}.
285:
286: \section{CONCLUSIONS}
287: We present the largest galaxy groups catalog at the present constructed from
288: galaxies in the Third Data Release of the Sloan Digital Sky Survey.
289: The group identification is carried out using the friend-of-friend algorithm
290: developed by Huchra \& Geller (1982), plus the introduction of corrections for
291: rich groups which improve the group identification and the group center
292: location.
293: The resulting catalog comprise 10864 galaxy groups with
294: at least four members and it has a mean redshift of $0.1$.
295: The median basic
296: physical properties of our catalog are very similar to those obtained
297: in previous works (Table \ref{tab1}).
298:
299: The complete galaxy group catalog with the physical properties estimated in
300: this work is available in the World Wide Web at
301: {\it https://www.iate.oac.uncor.edu/SDSSDR3GG/}
302: or upon request at any of the following e-mail addresses: {\it manuel@oac.
303: uncor.edu, arielz@oac.uncor.edu}.
304:
305: \acknowledgments
306:
307: This work has been partially supported by Consejo de Investigaciones
308: Cient\'{\i}ficas y T\'ecnicas de la Rep\'ublica Argentina (CONICET), the
309: Secretar\'{\i}a de Ciencia y T\'ecnica de la Universidad Nacional de C\'ordoba
310: (SeCyT) and Fundaci\'on Antorchas, Argentina.
311:
312: \begin{thebibliography}{}
313: \bibitem[{Abazajian et al.}2004]{aba04}
314: Abazajian K., et al. (SDSS Team), 2004, AJ, in press, (astro-ph/0410239).
315: \bibitem[{Beers et al. }1990]{beers1990}
316: Beers T. C., Flynn K., Gebhardt K., 1990, AJ, 100, 32.
317: \bibitem[{Blanton et al.}2003]{bla03}
318: Blanton M.R., et al. (SDSS Team), 2003, ApJ, 592, 819.
319: \bibitem[{Diaz et al. }2004]{eu04}
320: D\'{\i}az E., Zandivarez A., Merch\'an M.E., Muriel H., 2004, ApJ, submitted.
321: \bibitem[{Dominguez et al. }2002]{dom02}
322: Dom\'{\i}nguez M., Zandivarez A., Mart\'{\i}nez H.J, Merch\'an M.E., Muriel H.,
323: Lambas D.G., 2002, MNRAS, 335, 825.
324: \bibitem[{Eke et al.}2004]{eke04a}
325: Eke V., et al. (2dFGRS Team), 2004a, MNRAS, 348, 866.
326: \bibitem[{Eke et al.}2004]{eke04b}
327: Eke V., et al. (2dFGRS Team), 2004b, MNRAS, in press.
328: \bibitem[{Girardi et al.}1993]{girardi93}
329: Girardi M., Biviano A., Giuricin G., Mardirossian F. \& Mezzetti M., 1993,
330: ApJ, 404, 38.
331: \bibitem[{Girardi et al.}2000]{girardi00}
332: Girardi M., Giuricin G., 2000, ApJ, 540, 45.
333: \bibitem[{Giuricin et al.}2000]{giuricin00}
334: Giuricin G., Marinoni C., Ceriani L., Pisani A., 2000, ApJ, 543, 178.
335: \bibitem[{Huchra \& Geller }1982]{huchra82}
336: Huchra J.P., Geller M.J., 1982, ApJ, 257, 423.
337: \bibitem[{Martinez et al. }2002]{mart02}
338: Mart\'{\i}nez H.J, Zandivarez A., Merch\'an M.E., Dom\'{\i}nguez M., 2002,
339: MNRAS, 337, 1441.
340: \bibitem[{Merchan et al. }2000]{merchan00}
341: Merch\'an M.E., Maia M.A.G., Lambas D.G., 2000, ApJ, 545, 26.
342: \bibitem[{Merchan et al. }2002]{merchan02}
343: Merch\'an M.E., Zandivarez A., 2002, MNRAS, 335, 216.
344: \bibitem[{Norberg et al.}2002]{norb02}
345: Norberg P., et al. (2dFGRS Team), 2002, MNRAS, 336, 907.
346: \bibitem[{Padilla et al.}2004]{pad04}
347: Padilla N.D., et al. (2dFGRS Team), 2004, MNRAS, 352, 211.
348: \bibitem[{Ragone et al. }2004]{cin04}
349: Ragone C., Merch\'an M.E., Muriel H., Zandivarez A., 2004, MNRAS, 350, 938.
350: \bibitem[{Ramella et al.}2002]{ramella02}
351: Ramella M., Geller M.J., Pisani A., da Costa L.N., 2002, AJ, 123, 2976.
352: \bibitem[{Tucker et al.}2000]{tucker00}
353: Tucker D.L., Oemler A.Jr., Hashimoto Y., Shectman S.A., Kirshner R.P.,
354: Lin H., Landy S.D., Schechter P.L., Allam S.S., 2000, ApJS, 130, 237.
355: \bibitem[{zandivarez et al. }2003]{z03}
356: Zandivarez A., Merch\'an M.E., Padilla N.D., 2003, MNRAS, 344, 247.
357:
358: \end{thebibliography}
359:
360:
361:
362: \begin{table*}
363: \footnotesize
364: \begin{center}
365: \caption{Median physical properties of groups}
366: \label{tab1}
367: \vskip 0.5cm
368: \begin {tabular}{lcccccc}
369: \tableline
370: Sample & $N$ & $N_{gal}$ & $\bar{z}$ & $\bar{\sigma}_v(km \ s^{-1})$ & $\bar{M}(h^{-1} M_{\odot})$ & $\bar{R}_V(h^{-1}Mpc)$\\
371: \tableline
372: USGC (Ramella et al., 2002) & $1168$ & $6846$ & --- & $264$ & $4.7\times 10^{13}$ & $1.06$\\
373: LCRS Loose Groups (Tucker et al., 2002) & $1495$ & $\sim 9250$ & --- & $164$ & $1.9\times 10^{13}$ & $\sim 1.2$ \\
374: 2dFGRS (Mech\'an \& Zandivarez, 2002) & $2209$ & $14634$ & $0.1$ & $232$ & $3.7\times 10^{13}$ & $1.02$\\
375: 2dFGRS (Eke et al., 2004a) & $7020$ & $55753$ & $0.11$ & $260$ & $3.3\times 10^{13}$ & ---\\
376: SDSS-DR3 First identification & $10152$ & $78410$ & $0.1$ & $240$ & $4.2\times 10^{13}$ & $1.21$\\
377: SDSS-DR3 Improved identification & $10864$ & $66517$ & $0.1$ & $230$ & $3.8\times 10^{13}$ & $1.12$\\
378: \tableline
379: \end{tabular}
380: \end{center}
381: \end{table*}
382:
383:
384: \end{document}
385:
386:
387:
388:
389: