2008ApJ...674L..13C

We compare environmental effects in two analogous samples of galaxies, one from the Sloan Digital Sky Survey (SDSS) and the other from a semianalytic model (SAM) based on the Millennium Simulation (MS), to test to what extent current SAMs of galaxy formation are reproducing environmental effects. We estimate the large-scale environment of each galaxy using a Bayesian density estimator based on distances to all 10 nearest neighbors, and we compare broadband photometric properties of the two samples as a function of environment. The feedbacks implemented in the semianalytic model produce a qualitatively correct galaxy population with similar environmental dependence as that seen in SDSS galaxies. In detail, however, the colors of MS galaxies exhibit an exaggerated dependence on environment: the field contains too many blue galaxies, whereas clusters contain too many red galaxies, compared to the SDSS sample. We also find that the MS contains a population of highly clustered, relatively faint red galaxies with velocity dispersions comparable to their Hubble flow. Such high-density galaxies, if they exist, would be overlooked in any low-redshift survey, since their membership to a cluster cannot be determined because of the ``fingers-of-God'' effect.