Weak lensing measurements of the APEX-SZ galaxy cluster sample

TL;DR

This paper presents a weak lensing analysis of 39 galaxy clusters from the APEX-SZ survey, focusing on background galaxy selection, redshift distribution effects, and mass estimation accuracy.

Contribution

It introduces a multi-colour background selection method and quantifies the impact of source redshift variance on cluster mass estimates.

Findings

Background selection reduces redshift distribution variance impact by 20-40%.

Mass estimates are consistent with mass-concentration relations.

Deep imaging significantly decreases scatter in mass measurements.

Abstract

We present a weak lensing analysis for galaxy clusters from the APEX-SZ survey. For $39$ massive galaxy clusters that were observed via the Sunyaev-Zel\textquotesingle dovich effect (SZE) with the APEX telescope, we analyse deep optical imaging data from WFI(@2.2mMPG/ESO) and Suprime-Cam(@SUBARU) in three bands. The masses obtained in this study, including an X-ray selected subsample of 27 clusters, are optimised for and used in studies constraining the mass to observable scaling relations at fixed cosmology. A novel focus of our weak lensing analysis is the multi-colour background selection to suppress effects of cosmic variance on the redshift distribution of source galaxies. We investigate the effects of cluster member contamination through galaxy density, shear profile, and recovered concentrations. We quantify the impact of variance in source redshift distribution on the mass estimate by studying nine sub-fields of the COSMOS survey for different cluster redshift and manitude limits. We measure a standard deviation of $\sim 6$\% on the mean angular diameter distance ratio for a cluster at $z\!=\!0.45$ and shallow imaging data of $R\!\approx\!23$ mag. It falls to $\sim 1$\% for deep, $R=26$ mag, observations. This corresponds to 8.4\% and 1.4\% scatter in $M_{200}$. Our background selection reduces this scatter by $20-40$\%, depending on cluster redshift and imaging depth. We derived cluster masses with and without using a mass concentration relation and find consistent results, and concentrations consistent with the used mass-concentration relation.