The WARPS Survey. VIII. Evolution of the Galaxy Cluster X-ray Luminosity Function

TL;DR

This study measures the evolution of the galaxy cluster X-ray Luminosity Function using the WARPS survey, finding a decrease in high luminosity clusters with redshift and confirming the evolution's significance through Bayesian analysis.

Contribution

It provides the first detailed measurement of the XLF evolution from the WARPS survey, including Bayesian analysis to account for uncertainties and systematic effects.

Findings

Significant negative evolution of the XLF with decreasing high luminosity clusters at higher redshift.

Nearly unchanged cluster population out to redshift z ~ 1.1, consistent with a low density universe.

Detected excess of clusters at specific redshift and luminosity ranges, not explained by biases or variance.

Abstract

We present measurements of the galaxy cluster X-ray Luminosity Function (XLF) from the Wide Angle ROSAT Pointed Survey (WARPS) and quantify its evolution. WARPS is a serendipitous survey of the central region of ROSAT pointed observations and was carried out in two phases (WARPS-I and WARPS-II). The results here are based on a final sample of 124 clusters, complete above a flux limit of 6.5 10E-15 erg/s/cm2, with members out to redshift z ~ 1.05, and a sky coverage of 70.9 deg2. We find significant evidence for negative evolution of the XLF, which complements the majority of X-ray cluster surveys. To quantify the suggested evolution, we perform a maximum likelihood analysis and conclude that the evolution is driven by a decreasing number density of high luminosity clusters with redshift, while the bulk of the cluster population remains nearly unchanged out to redshift z ~ 1.1, as expected in a low density Universe. The results are found to be insensitive to a variety of sources of systematic uncertainty that affect the measurement of the XLF and determination of the survey selection function. We perform a Bayesian analysis of the XLF to fully account for uncertainties in the local XLF on the measured evolution, and find that the detected evolution remains significant at the 95% level. We observe a significant excess of clusters in the WARPS at 0.1 < z < 0.3 and LX ~ 2 10E42 erg/s compared with the reference low-redshift XLF, or our Bayesian fit to the WARPS data. We find that the excess cannot be explained by sample variance, or Eddington bias, and is unlikely to be due to problems with the survey selection function.