Evolution of central dark matter of early-type galaxies up to z ~ 0.8

TL;DR

This study examines how the central dark matter content of early-type galaxies evolves up to redshift 0.8, revealing that local galaxies are more dark matter dominated, with implications for galaxy formation models and the initial mass function.

Contribution

It provides the first detailed analysis of dark matter fraction evolution in early-type galaxies up to z ~ 0.8 using Jeans modeling and galaxy colors.

Findings

Local galaxies are more dark matter dominated than at higher redshifts.

Size and dark matter growth are inconsistent with passive evolution.

Minor mergers explain the observed evolution better than passive growth.

Abstract

We investigate the evolution of dark and luminous matter in the central regions of early-type galaxies (ETGs) up to z ~ 0.8. We use a spectroscopically selected sample of 154 cluster and field galaxies from the EDisCS survey, covering a wide range in redshifts (z ~ 0.4-0.8), stellar masses ($\log M_{\star}/ M_{\odot}$ ~ 10.5-11.5 dex) and velocity dispersions ($\sigma_{\star}$ ~ 100-300 \, km/s). We obtain central dark matter (DM) fractions by determining the dynamical masses from Jeans modelling of galaxy aperture velocity dispersions and the $M_{\star}$ from galaxy colours, and compare the results with local samples. We discuss how the correlations of central DM with galaxy size (i.e. the effective radius, $R_{\rm e}$), $M_{\star}$ and $\sigma_{\star}$ evolve as a function of redshift, finding clear indications that local galaxies are, on average, more DM dominated than their counterparts at larger redshift. This DM fraction evolution with $z$ can be only partially interpreted as a consequence of the size-redshift evolution. We discuss our results within galaxy formation scenarios, and conclude that the growth in size and DM content which we measure within the last 7 Gyr is incompatible with passive evolution, while it is well reproduced in the multiple minor merger scenario. We also discuss the impact of the IMF on our DM inferences and argue that this can be non-universal with the lookback time. In particular, we find the Salpeter IMF can be better accommodated by low redshift systems, while producing stellar masses at high-$z$ which are unphysically larger than the estimated dynamical masses (particularly for lower-$\sigma_{\star}$ systems).