Galaxy assembly, stellar feedback and metal enrichment: the view from the GAEA model

TL;DR

This paper uses the GAEA semi-analytic model to analyze how different stellar feedback mechanisms influence galaxy formation, successfully reproducing observed galaxy evolution and metallicity trends.

Contribution

It introduces a self-consistent treatment of gas, metal, and energy recycling in galaxy formation models, demonstrating the importance of feedback schemes in matching observations.

Findings

Both ejective and preventive feedback models can reproduce observed trends.

Up to 60-70% of baryons are ejected and unavailable for cooling at high redshift.

Models align better with observed galaxy properties when feedback schemes are included.

Abstract

One major problem of current theoretical models of galaxy formation is given by their inability to reproduce the apparently `anti-hierarchical' evolution of galaxy assembly: massive galaxies appear to be in place since $z\sim 3$, while a significant increase of the number densities of low mass galaxies is measured with decreasing redshift. In this work, we perform a systematic analysis of the influence of different stellar feedback schemes, carried out in the framework of GAEA, a new semi-analytic model of galaxy formation. It includes a self-consistent treatment for the timings of gas, metal and energy recycling, and for the chemical yields. We show this to be crucial to use observational measurements of the metallicity as independent and powerful constraints for the adopted feedback schemes. The observed trends can be reproduced in the framework of either a strong ejective or preventive feedback model. In the former case, the gas ejection rate must decrease significantly with cosmic time (as suggested by parametrizations of the cosmological `FIRE' simulations). Irrespective of the feedback scheme used, our successful models always imply that up to 60-70 per cent of the baryons reside in an `ejected' reservoir and are unavailable for cooling at high redshift. The same schemes predict physical properties of model galaxies (e.g. gas content, colour, age, and metallicity) that are in much better agreement with observational data than our fiducial model. The overall fraction of passive galaxies is found to be primarily determined by internal physical processes, with environment playing a secondary role.