LoCuSS: A DYNAMICAL ANALYSIS OF X-RAY ACTIVE GALACTIC NUCLEI IN LOCAL CLUSTERS

We present a study of the distribution of X-ray active galactic nuclei (AGNs) in a representative sample of 26 massive clusters at 0.15 < z < 0.30, combining Chandra observations sensitive to X-ray point sources of luminosity L-X similar to 10(42) erg s (1) at the cluster redshift with extensive and highly complete spectroscopy of cluster members down to similar to M-K*+2. In total we identify 48 X-ray AGNs among the cluster members, with luminosities 2 x 10(41)-1 x 10(44) erg s(-1). Based on these identifications, we estimate that 0.73% +/- 0.14% of cluster galaxies brighter than M-K = -23.1 (M-K*+1.5) host an X-ray AGN with L-X > 10(42) erg s(-1). In the stacked caustic diagram that shows (v(los) - < v >)/sigma(v) versus r(proj)/r(500), the X-ray AGN appear to preferentially lie along the caustics, suggestive of an infalling population. They also appear to avoid the region with lowest cluster-centric radii and relative velocities (r(proj) < 0.4r(500); |v - < v >|/sigma(v) < 0.8), which is dominated by the virialized population of galaxies accreted earliest into the clusters. The line-of-sight velocity histogram of the X-ray AGN shows a relatively flat distribution, and is inconsistent with the Gaussian distribution expected for a virialized population at 98.9% confidence. Moreover, the velocity dispersion of the 48 X-ray AGNs is 1.51 times that of the overall cluster population, which is consistent with the root 2 ratio expected by simple energetic arguments when comparing infalling versus virialized populations. This kinematic segregation is significant at the 4.66 sigma level. When splitting the X-ray AGN sample into two according to X-ray or infrared (IR) luminosity, both X-ray bright (L-X > 10(42)) and IR-bright (L-TIR > 2 x 10(10) L-circle dot) subsamples show higher velocity dispersions than their X-ray dim and IR-dim counterparts at > 2 sigma significance. This is consistent with the nuclear activity responsible for the X-ray and IR emission being slowly shut down as the host galaxies are accreted into the cluster. Overall, our results provide the strongest observational evidence to date that X-ray AGNs found in massive clusters are an infalling population, and that the cluster environment very effectively suppresses radiatively efficient nuclear activity in its member galaxies. These results are consistent with the view that for galaxies to host an X-ray AGN they should be the central galaxy within their dark matter halo and have a ready supply of cold gas.