The DEEP2 Galaxy Redshift Survey: Color and Luminosity Dependence of Galaxy Clustering at <IMG SRC="eq-00001.gif" ALT="z\sim 1"/> z ? 1

We present measurements of the color and luminosity dependence of galaxy clustering at <IMG SRC="eq-00002.gif" ALT="z\sim 1"/> z [?] 1 in the DEEP2 Galaxy Redshift Survey. Using volume-limited subsamples in bins of both color and luminosity, we find the following: (1) The clustering dependence is much stronger with color than with luminosity and is as strong with color at <IMG SRC="eq-00003.gif" ALT="z\sim 1"/> z [?] 1 as is found locally. We find no dependence of the clustering amplitude on color for galaxies on the red sequence, but a significant dependence on color for galaxies within the blue cloud. (2) For galaxies in the range <IMG SRC="eq-00004.gif" ALT="L/L^{*}\sim 0.7 - 2"/> L/L* [?] 0.7-2, a stronger large-scale luminosity dependence is seen for all galaxies than is seen for red and blue galaxies separately. The small-scale clustering amplitude depends significantly on luminosity for blue galaxies, with brighter samples having a stronger rise on scales <IMG SRC="eq-00005.gif" ALT="r_{p} &lt; 0.5"/> rp < 0.5 <IMG SRC="eq-00006.gif" ALT="h^{-1}"/> h[?]1 Mpc. (3) Redder galaxies exhibit stronger small-scale redshift-space distortions ("fingers of god"), and both red and blue populations show large-scale distortions in <IMG SRC="eq-00007.gif" ALT="\xi (r_{p},\pi) "/> x (rp,p) due to coherent infall. (4) While the clustering length, <IMG SRC="eq-00008.gif" ALT="r_{0}"/> r0, increases smoothly with galaxy color (in narrow bins), its power-law exponent, g, exhibits a sharp jump from the blue cloud to the red sequence. The intermediate-color "green" galaxy population likely includes transitional galaxies moving from the blue cloud to the red sequence; on large scales green galaxies are as clustered as red galaxies but show infall kinematics and a small-scale correlation slope akin to the blue galaxy population. (5) We compare our results to a semianalytic galaxy formation model applied to the Millennium Run simulation. Differences between the data and the model suggest that in the model star formation is shut down too efficiently in satellite galaxies.