Do gravitational lens galaxies have an excess of luminous substructure?

TL;DR

This study compares luminous satellite galaxies around lens and non-lens galaxies to test if lens galaxies are representative of the general galaxy population, finding no significant differences in satellite populations.

Contribution

First measurement of satellite counts as a function of host stellar mass for both lens and non-lens galaxies, enabling direct comparison.

Findings

Satellite numbers are consistent between lens and non-lens galaxies.

Satellite spatial distributions are similar across both galaxy samples.

Predicted satellite counts around lens galaxies match observed counts.

Abstract

Strong gravitational lensing can be used to directly measure the mass function of their satellites, thus testing one of the fundamental predictions of cold dark matter cosmological models. Given the importance of this test it is essential to ensure that galaxies acting as strong lenses have dark and luminous satellites which are representative of the overall galaxy population. We address this issue by measuring the number and spatial distribution of luminous satellites in ACS imaging around lens galaxies from the Sloan Lens Advanced Camera for Surveys (SLACS) lenses, and comparing them with the satellite population in ACS imaging of non lens galaxies selected from COSMOS, which has similar depth and resolution to the ACS images of SLACS lenses. In order to compare the samples of lens and non lens galaxies, which have intrinsically different stellar mass distributions, we measure, for the first time, the number of satellites per host as a continuous function of host stellar mass for both populations. We find that the number of satellites as a function of host stellar mass, as well as the spatial distribution are consistent between the samples. Using these results, we predict the number of satellites we would expect to find around a subset of the Cosmic Lens All Sky Survey (CLASS) lenses, and find a result consistent with the the number observed by Jackson et al. 2010. Thus we conclude that within our measurement uncertainties there is no significant difference in the satellite populations of lens and non lens galaxies.