We present results from a low-resolution spectroscopic survey for 21 galaxy clusters at $0.4 < z < 0.8$ selected from the ESO Distant Cluster Survey. We measured spectra using the low-dispersion prism in IMACS on the Magellan Baade telescope and calc
ulate redshifts with an accuracy of $sigma_z = 0.007$. We find 1763 galaxies that are brighter than $R = 22.9$ in the large-scale cluster environs. We identify the galaxies expected to be accreted by the clusters as they evolve to $z = 0$ using spherical infall models and find that $sim30%$ to $sim70%$ of the $z = 0$ cluster population lies outside the virial radius at $z sim 0.6$. For analogous clusters at $z = 0$, we calculate that the ratio of galaxies that have fallen into the clusters since $z sim 0.6$ to those that were already in the core at that redshift is typically between $sim0.3$ and $1.5$. This wide range of ratios is due to intrinsic scatter and is not a function of velocity dispersion, so a variety of infall histories is to be expected for clusters with current velocity dispersions of $300 lesssimsigmalesssim 1200$ km s$^{-1}$. Within the infall regions of $z sim 0.6$ clusters, we find a larger red fraction of galaxies than in the field and greater clustering among red galaxies than blue. We interpret these findings as evidence of preprocessing, where galaxies in denser local environments have their star formation rates affected prior to their aggregation into massive clusters, although the possibility of backsplash galaxies complicates the interpretation.
We analyze GALEX UV data for a system of four gravitationally-bound groups at z=0.37, SG1120, which is destined to merge into a Coma-mass cluster by z=0, to study how galaxy properties may change during cluster assembly. Of the 38 visually-classified
S0 galaxies, with masses ranging from log(M_*)~10-11, we detect only one in the NUV channel, a strongly star-forming S0 that is the brightest UV source with a measured redshift placing it in SG1120. Stacking the undetected S0 galaxies (which generally lie on or near the optical red-sequence of SG1120) still results in no NUV/FUV detection (<2 sigma). Using our limit in the NUV band, we conclude that for a rapidly truncating star formation rate, star formation ceased *at least* ~0.1 to 0.7 Gyr ago, depending on the strength of the starburst prior to truncation. With an exponentially declining star-formation history over a range of time-scales, we rule out recent star-formation over a wide range of ages. We conclude that if S0 formation involves significant star formation, it occurred well before the groups were in this current pre-assembly phase. As such, it seems that S0 formation is even more likely to be predominantly occurring outside of the cluster environment.
We reinvestigate the dramatic rise in the S0 fraction, f_S0, within clusters since z ~ 0.5. In particular, we focus on the role of the global galaxy environment on f_S0 by compiling, either from our own observations or the literature, robust line-of-
sight velocity dispersions, sigmas, for a sample of galaxy groups and clusters at 0.1 < z < 0.8 that have uniformly determined, published morphological fractions. We find that the trend of f_S0 with redshift is twice as strong for sigma < 750 km/s groups/poor clusters than for higher-sigma, rich clusters. From this result, we infer that over this redshift range galaxy-galaxy interactions, which are more effective in lower-sigma environments, are more responsible for transforming spiral galaxies into S0s than galaxy-environment processes, which are more effective in higher-sigma environments. The rapid, recent growth of the S0 population in groups and poor clusters implies that large numbers of progenitors exist in low-sigma systems at modest redshifts (~ 0.5), where morphologies and internal kinematics are within the measurement range of current technology.
