The Contribution of Radio Galaxy Contamination to Measurements of the Sunyaev-Zel'dovich Decrement in Massive Galaxy Clusters at 140 GHz with Bolocam

TL;DR

This study characterizes radio galaxy contamination in 140 GHz observations of galaxy clusters, finding it generally negligible but more significant in cool-core clusters, with implications for Sunyaev-Zel'dovich measurements.

Contribution

It provides a detailed analysis of radio source contamination at 140 GHz in galaxy clusters, especially highlighting the role of cool-core systems and the intrinsic scatter in flux densities.

Findings

Radio contamination causes up to 20% change in SZ signal in some clusters.

Most clusters have less than 1% contamination, which is negligible for current measurements.

Contamination is more significant in cool-core clusters.

Abstract

We describe in detail our characterization of the compact radio source population in 140 GHz Bolocam observations of a set of 45 massive galaxy clusters. We use a combination of 1.4 and 30 GHz data to select a total of 28 probable cluster-member radio galaxies and also to predict their 140 GHz flux densities. All of these galaxies are steep-spectrum radio sources and they are found preferentially in the cool-core clusters within our sample. In particular, 11 of the 12 brightest cluster member radio sources are associated with cool-core systems. Although none of the individual galaxies are robustly detected in the Bolocam data, the ensemble-average flux density at 140 GHz is consistent with, but slightly lower than, the extrapolation from lower frequencies assuming a constant spectral index. In addition, our data indicate an intrinsic scatter of 30 percent around the power-law extrapolated flux densities at 140 GHz, although our data do not tightly constrain this scatter. For our cluster sample, which is composed of high-mass and moderate-redshift systems, we find that the maximum fractional change in the Sunyaev-Zel'dovich signal integrated over any single cluster due to the presence of these radio sources is 20 percent, and only 1/4 of the clusters show a fractional change of more than 1 percent. The amount of contamination is strongly dependent on cluster morphology, and nearly all of the clusters with more than 1 percent contamination are cool-core systems. This result indicates that radio contamination is not significant compared to current noise levels in 140 GHz images of massive clusters and is in good agreement with the level of radio contamination found in previous results based on lower frequency data or simulations.