Stellar Velocity Dispersion: Linking Quiescent Galaxies to their Dark Matter Halos

TL;DR

This study uses the Illustris-1 simulation to demonstrate that stellar velocity dispersion reliably indicates dark matter halo mass in quiescent galaxies, even accounting for observational aperture effects and tidal stripping.

Contribution

It establishes a robust proportionality between stellar velocity dispersion and dark matter halo mass for both central and satellite galaxies, accounting for observational and tidal effects.

Findings

Stellar velocity dispersion correlates tightly with dark matter halo mass (<0.2 dex scatter).

The relation holds for both central and satellite galaxies, with some deviations due to tidal stripping.

Systematic uncertainties are small (<0.1 dex) when applying the relation statistically.

Abstract

We analyze the Illustris-1 hydrodynamical cosmological simulation to explore the stellar velocity dispersion of quiescent galaxies as an observational probe of dark matter halo velocity dispersion and mass. Stellar velocity dispersion is proportional to dark matter halo velocity dispersion for both central and satellite galaxies. The dark matter halos of central galaxies are in virial equilibrium and thus the stellar velocity dispersion is also proportional to dark matter halo mass. This proportionality holds even when a line-of-sight aperture dispersion is calculated in analogy to observations. In contrast, at a given stellar velocity dispersion, the dark matter halo mass of satellite galaxies is smaller than virial equilibrium expectations. This deviation from virial equilibrium probably results from tidal stripping of the outer dark matter halo. Stellar velocity dispersion appears insensitive to tidal effects and thus reflects the correlation between stellar velocity dispersion and dark matter halo mass prior to infall. There is a tight relation (<0.2 dex scatter) between line-of-sight aperture stellar velocity dispersion and dark matter halo mass suggesting that the dark matter halo mass may be estimated from the measured stellar velocity dispersion for both central and satellite galaxies. We evaluate the impact of treating all objects as central galaxies if the relation we derive is applied to a statistical ensemble. A large fraction (>2/3) of massive quiescent galaxies are central galaxies and systematic uncertainty in the inferred dark matter halo mass is <0.1 dex thus simplifying application of the simulation results to currently available observations.