Compact groups in theory and practice - IV. The connection to large-scale structure

We investigate the properties of photometrically selected compact groups (CGs) in the Sloan Digital Sky Survey. In this paper, the fourth in a series, we focus on understanding the characteristics of our observed CG sample with particular attention paid to quantifying and removing contamination from projected foreground or background galaxies. Based on a simple comparison of pairwise redshift likelihoods, we find that approximately half of CGs in the parent sample contain one or more projected (interloping) members; our final clean sample contains 4566 galaxies in 1086 CGs. We show that half of the remaining CGs are associated with rich groups (or clusters), i.e. they are embedded sub-structure. The other half have spatial distributions and number-density profiles consistent with the interpretation that they are either independently distributed structures within the field (i.e. they are isolated) or associated with relatively poor structures. Comparisons of late-type and red-sequence fractions in radial annuli show that galaxies around apparently isolated CGs resemble the field population by 300 to 500 kpc from the group centre. In contrast, the galaxy population surrounding embedded CGs appears to remain distinct from the field out beyond 1 to 2 Mpc, consistent with results for rich groups. We take this as additional evidence that the observed distinction between CGs, i.e. isolated versus embedded, is a separation between different host environments.