The study of groups of galaxies as dynamical systems is interesting not only per se , but also because groups can be used to set constraints on cosmological models (e.g. Frenk et al. 1990; Weinberg & Cole 1992, Zabludoff et al. 1993; Zabludoff & Geller 1994, Nolthenius et al. 1994, 1997) and on models of galaxy formation (Frenk et al. 1996; Kaufmann et al. 1997). Groups are also interesting sites where to look for interactions of galaxies with their environment, in order to obtain information on galaxy evolution processes (Postman & Geller 1984, Allington-Smith et al. 1993).
Group catalogs identified in redshift space are increasing both in number and size (CfA2N, RPG; SSRS2, Ramella et al. 1998; Perseus-Pisces, Trasarti Battistoni 1998; LCRS, Tucker et al. 1997). At the same time, cosmological n-body simulations are reaching the resolution required to allow replication of the observational techniques for the identification of groups. In particular, Frederic (1995a,b) uses n-body simulations to evaluate and compare the performances of commonly used group finding algorithms.
Among the main properties of groups, the velocity dispersion is of particular interest. It is easy to measure and it is well suited for comparison with the predictions of cosmological n-body models (Frenk et al. 1990; Moore et al. 1993; Zabludoff et al. 1993). Distributions of velocity dispersions of nearby groups are now well determined with rather small statistical uncertainties given the large size of the samples. Ramella et al. 1995 and Ramella et al. 1996 survey the redshifts of candidate faint members of a selection of nearby groups and find that the velocity dispersion of groups is stable against inclusion of fainter members. In other words, the velocity dispersion estimated on the basis of the fewer original bright members is a good indicator of the velocity dispersion obtained from a better sampling of the same group.
In this paper we identify and analyze groups of galaxies in the recently completed ESP survey (Vettolani et al. 1997). The ESP group catalog is interesting because of its depth () and because it samples a new independent region of the universe. ESP is a nearly bi-dimensional survey (the declination range is much smaller than the right ascension range), five times deeper than either CfA2 (Geller & Huchra 1989) or SSRS2 (da Costa et al. 1994). The volume of the survey is Mpc3 at the sensitivity peak of the survey, , and Mpc3 at the effective depth of the sample, . Even if the volume of ESP is of the same order of magnitude of the volumes explored with the individual CfA2, SSRS2, and Perseus-Pisces samples, it intercepts a larger number of structures. In fact, the strip geometry is very efficient for the detection of large scale structures within redshift surveys (de Lapparent et al. 1988).
In particular we determine the distribution of the velocity dispersions of groups and show that our result is reliable in spite of the particular geometry of the ESP survey (two rows of adjacent circular fields of radius arcmin, see Fig. 1 of Vettolani et al. 1997).
An important aspect that distinguishes the ESP group catalog from the other shallower catalogs is that we have the spectra of all galaxies with measured redshift. It is already well known that emission line galaxies are rarer in rich clusters than in the field (Biviano et al. 1997). The relation between the fraction of emission line galaxies and the local density is a manifestation of the morphology-density relationship observed for clusters (Dressler 1980), a useful tool in the study of galaxy evolution. With the ESP catalog we explore the extent of the morphology density relationship in the intermediate range of densities that are typical of groups at a larger depth than in previous studies.
We note that preliminary results of a search for groups in the Las Campanas Redshift Survey (Shectman et al. 1996) have been presented by Tucker et al. (1997). The properties of these groups, as distant as ours, are difficult to compare with those of our ESP groups and with those of shallower surveys because LCRS a) is a red band survey (ESP and shallower surveys are selected in the blue band), b) it is not simply magnitude limited, and c) it does not uniformly sample structures in redshift space. In particular, the different selection criteria could have a strong impact on the results concerning the morphology-density relation, the luminosity segregation, and the possible differences between the luminosity functions of member and non-member galaxies.
In Sect. 2 we briefly describe the data; in Sect. 3 we analyze the effect of the ESP geometry on the estimate of the velocity dispersion of groups; in Sect. 4 we summarize the group identification procedure; in Sect. 5 we present the ESP group catalog; in Sect. 6 we analyze properties of groups that are relevant to a characterization of the Large Scale Structure (LSS); in Sect. 7 we analyze the properties of galaxies in groups and compare them to the properties of "field" galaxies (i.e.galaxies that have not been assigned to groups) and "cluster" galaxies; in Sect. 8 we identify ESP counterparts of ACO and/or EDCC clusters (Lumsden et al. 1992). Finally, we summarize our results in Sect. 9.
© European Southern Observatory (ESO) 1999
Online publication: December 22, 1998