The Small Scale Velocity Dispersion of Galaxies: A Comparison of Cosmological Simulations

Abstract

The velocity dispersion of galaxies on small scales ($r\sim1h^{-1}$ Mpc), $\sigma_{12}(r)$, can be estimated from the anisotropy of the galaxy-galaxy correlation function in redshift space. We apply this technique to ``mock-catalogs'' extracted from N-body simulations of several different variants of Cold Dark Matter dominated cosmological models to obtain results which may be consistently compared to similar results from observations. We find a large variation in the value of $\sigma_{12}(1 h^{-1} Mpc)$ in different regions of the same simulation. We conclude that this statistic should not be considered to conclusively rule out any of the cosmological models we have studied. We attempt to make the statistic more robust by removing clusters from the simulations using an automated cluster-removing routine, but this appears to reduce the discriminatory power of the statistic. However, studying $\sigma_{12}$ as clusters with different internal velocity dispersions are removed leads to interesting information about the amount of power on cluster and subcluster scales. We also compute the pairwise velocity dispersion directly and compare this to the values obtained using the Davis-Peebles method, and find that the agreement is fairly good. We evaluate the models used for the mean streaming velocity and the pairwise peculiar velocity distribution in the original Davis-Peebles method by comparing the models with the results from the simulations.