DescriptionCurrent and near-future galaxy cluster surveys at a variety of wavelengths are expected to provide a promising way to obtain precision measurements of the growth of structure over cosmic time. This in turn would serve as an important precision probe of cosmology. However, to realize the full potential of these surveys, systematic uncertainties arising from, for example, cluster mass estimates and sample selection must be well understood. This work follows several different approaches towards alleviating these uncertainties.
Cluster sample selection is investigated in the context of arcminute-resolution millimeter-wavelength surveys such as the Atacama Cosmology Telescope (ACT) and the South Pole Telescope (SPT). Large-area, realistic simulations of the microwave sky are constructed and cluster detection is simulated using a multi-frequency Wiener filter to separate the galaxy clusters, via their Sunyaev-Zel'dovich signal, from other contaminating microwave signals. Using this technique, an ACT-like survey can expect to obtain a cluster sample that is 90% complete and 85% pure above a mass of 3 x 10^14 Msun.
Cluster mass uncertainties are explored by comparing X-ray and weak-lensing mass estimates for shear-selected galaxy clusters in the Deep Lens Survey (DLS) to study possible biases in using cluster baryons or weak-lensing shear as tracers of the cluster total mass. Results are presented for four galaxy clusters that comprise the top-ranked shear-selected system in the DLS, and for three of these clusters there is agreement between X-ray and weak-lensing mass estimates. For the fourth cluster, the X-ray mass estimate is higher than that from weak-lensing by 2-sigma, and X-ray images suggest this cluster may be undergoing a merger with a smaller cluster, which may be biasing the X-ray mass estimate high.
The feasibility of measuring galaxy cluster peculiar velocities using an ACT-like instrument is also investigated. Such a possibility would allow one to measure structure growth via large-scale velocity fields and circumvent the uncertainties associated with measuring cluster masses. We show that such measurements are possible and yield statistical uncertainties of roughly 100 km/sec given either a temperature prior with 1-sigma errors of less than 2 keV or additional lower frequency millimeter-band observations.