Constraining the last 7 billion years of galaxy evolution in semi-analytic models

TL;DR

This study evaluates a semi-analytic galaxy evolution model over the last 7 billion years, identifying its limitations and suggesting directions for future improvements to better match observed galaxy properties.

Contribution

The paper applies Monte-Carlo Markov Chain techniques to constrain model parameters, revealing the need for extreme supernova feedback efficiencies and proposing targeted enhancements for high-redshift massive halos.

Findings

Model struggles to fit observed stellar mass functions without extreme feedback

Current model may be incomplete, requiring modifications

Future models should increase star formation efficiency in massive high-redshift halos

Abstract

We investigate the ability of the Croton et al. (2006) semi-analytic model to reproduce the evolution of observed galaxies across the final 7 billion years of cosmic history. Using Monte-Carlo Markov Chain techniques we explore the available parameter space to produce a model which attempts to achieve a statistically accurate fit to the observed stellar mass function at z=0 and z~0.8, as well as the local black hole-bulge relation. We find that in order to be successful we are required to push supernova feedback efficiencies to extreme limits which are, in some cases, unjustified by current observations. This leads us to the conclusion that the current model may be incomplete. Using the posterior probability distributions provided by our fitting, as well as the qualitative details of our produced stellar mass functions, we suggest that any future model improvements must act to preferentially bolster star formation efficiency in the most massive halos at high redshift.