HeCS-SZ: The Hectospec Survey of Sunyaev-Zeldovich Selected Clusters

TL;DR

This study compares optical spectroscopic measurements of galaxy clusters with Sunyaev-Zeldovich (SZ) mass estimates, finding good agreement and minimal bias, which supports the reliability of SZ mass proxies in cosmology.

Contribution

The paper presents the HeCS-SZ survey with new spectroscopic data for SZ-selected clusters, establishing a robust scaling relation between velocity dispersion and SZ mass, and assessing biases in SZ mass estimates.

Findings

Velocity dispersion scales well with SZ mass estimates.

SZ mass estimates are not significantly biased.

Galaxy velocity bias is small and consistent with dark matter expectations.

Abstract

We estimate cluster masses and velocity dispersions for 123 clusters from optical spectroscopy to compare the Sunyaev-Zeldovich (SZ) mass proxy and dynamical masses. Our new survey, HeCS-SZ (Hectospec Cluster Survey of SZ-selected clusters), includes 7,721 new or remeasured redshifts from MMT/Hectospec observations of 24 SZ-selected clusters at redshifts $z$=0.05-0.20 and not in previous surveys. We supplement the Hectospec data with spectra from the Sloan Digital Sky Survey (SDSS) and cluster data from the Cluster Infall Regions in SDSS (CIRS) project and the Hectospec Cluster Survey (HeCS), our Hectospec survey of clusters selected by X-ray flux. We measure the scaling relation between velocity dispersion and SZ mass estimates from the integrated Compton parameter for an SZ complete sample of 83 clusters. The observed relation agrees very well with a simple virial scaling from mass (based on SZ) to velocity dispersion. The SZ mass estimates (calibrated with hydrostatic X-ray mass estimates) are not significantly biased. Further, the velocity dispersion of cluster galaxies is consistent with the expected velocity dispersion of dark matter particles, indicating that galaxies are good dynamical tracers (i.e., velocity bias is small). Significant mass bias in SZ mass estimates could relieve tension between cosmological results from Planck SZ cluster counts and Planck CMB data. However, the excellent agreement between our measured velocity dispersions and those predicted from a virial scaling relation suggests that any SZ mass bias is too small to reconcile SZ and CMB results. In principle, SZ mass bias and velocity bias of galaxies could conspire to yield good agreement, but the required velocity bias is $\sigma_{galaxy}\approx 0.77\sigma_{DM}$, outside the range of plausible models of velocity bias in the literature.