The DEEP2 Galaxy Redshift Survey: environments of post-starburst galaxies at z ∼ 0.1 and ∼0.8

Post-starburst (also known as K+A) galaxies exhibit spectroscopic signatures indicating that their star formation was recently quenched; they are candidates for galaxies in transition from a star-forming phase to a passively evolving phase. We have spectroscopically identified large samples of post-starburst galaxies both in the Sloan Digital Sky Survey (SDSS) at z similar to 0.1 and in the DEEP2 Galaxy Redshift Survey at z similar to 0.8, using a uniform and robust selection method based on a cut in H beta line emission rather than the more problematic [O ii] lambda 3727. Based on measurements of the overdensity of galaxies around each object, we find that post-starburst galaxies brighter than 0.4L*(B) at low redshift have a similar, statistically indistinguishable environment distribution as blue galaxies, preferring underdense environments, but dramatically different from that of red galaxies. However, at higher-z, the environment distribution of post-starburst galaxies is more similar to red galaxies than to blue galaxies. We conclude that the quenching of star formation and the build-up of the red sequence through the K+A phase is happening in relatively overdense environments at z similar to 1 but in relatively underdense environments at z similar to 0. Although the relative environments where quenching occurs are decreasing with time, the corresponding absolute environment may have stayed the same along with the quenching mechanisms, because the mean absolute environments of all galaxies has to grow with time. In addition, we do not find any significant dependence on luminosity in the environment distribution of K+As. The existence of a large K+A population in the field at both redshifts indicates that cluster-specific mechanisms cannot be the dominant route by which these galaxies are formed. Our work also demonstrates that studying post-starburst-environment relations by measuring the K+A fraction in different environments, as is the common practice, is highly non-robust; modest changes in the comparison population used to define the fraction can drastically alter conclusions. Statistical comparisons of the overall environment distributions of different populations are much better behaved.