Loose groups of galaxies in the Perseus-Pisces survey

We present a large catalog(1) of loose groups of galaxies in the Southern Galactic Hemisphere, selected from the Perseus-Pisces redshift Survey (PPS). Particular care is taken in order to obtain group samples as homogeneous as possible to previously published catalogs. All our catalogs contain about 200 groups, significantly more than in most previous studies where group samples were obtained from galaxy data sets of comparable quality to (but smaller extent than) PPS. Groups are identified with the adaptive Friends-Of-Friends (FOF) algorithm of Huchra & Geller (1982): with suitable normalizations D-0 = 0.231 h(-1) Mpc and V-0 = 350 km s(-1) at cz(0) = 1000 km s(-1). The luminosity function (LF) normalization phi* = 0.02 h(3) Mpc(-3) appropriate for PPS yields a number density threshold delta n/n approximate to 180 for the adopted D-0, instead of delta n/n approximate to 80 used in previous studies of other samples. However, the customary choice of Do obtained (through the LF) from a fixed mass overdensity delta p/p = 80, well motivated in theory, suffers from important observational uncertainties and sample-to-sample variations of the LF normalization, and from major uncertainties in the relation between galaxy density n and mass density p. We discuss how to self-consistently match FOF parameters among different galaxy samples. We then separately vary several FOF and sample parameters, and discuss their effect on group properties. Loose groups in PPS nicely trace the large scale structure (LSS) in the parent galaxy sample. The group properties vary little with different redshift corrections, redshift cut-off, and galaxy LF, but are rather sensitive to the adopted links D-0 and V-0. More precisely, the typical group size (velocity dispersion) is linearly related to the adopted distance (velocity) link, while it is rather insensitive to the adopted velocity (distance) link. Physical properties of groups in PPS and in directly comparable samples show good agreement. There is a complex interplay among LSS features, sample depth, FOF grouping procedure, and group properties.