Galaxy quenching timescales from a forensic reconstruction of their colour evolution

TL;DR

This study reconstructs galaxy colour evolution to measure quenching timescales, revealing differences between observations and models, and exploring the influence of mass and environment on galaxy evolution.

Contribution

It introduces a consistent method to measure galaxy quenching timescales from both observations and simulations, and compares physical insights across datasets.

Findings

GAMA shows increasing quenching timescales over cosmic time.

SHARK model predicts shorter, constant quenching timescales.

Environment impacts low-mass galaxy quenching in GAMA.

Abstract

The timescales on which galaxies move out of the blue cloud to the red sequence ($\tau^{}_\mathrm{Q}$) provide insight into the mechanisms driving quenching. Here, we build upon previous work, where we showcased a method to reconstruct the colour evolution of observed low-redshift galaxies from the Galaxy And Mass Assembly (GAMA) survey based on spectral energy distribution (SED) fitting with ProSpect, together with a statistically-driven definition for the blue and red populations. We also use the predicted colour evolution from the SHARK semi-analytic model, combined with SED fits of our simulated galaxy sample, to study the accuracy of the measured $\tau^{}_\mathrm{Q}$ and gain physical insight into the colour evolution of galaxies. In this work, we measure $\tau^{}_\mathrm{Q}$ in a consistent approach for both observations and simulations. After accounting for selection bias, we find evidence for an increase in $\tau^{}_\mathrm{Q}$ in GAMA as a function of cosmic time (from $\tau^{}_\mathrm{Q}\sim1$ Gyr to $\tau^{}_\mathrm{Q}\sim2$ Gyr in the lapse of $\sim4$ Gyr), but not in SHARK ($\tau^{}_\mathrm{Q}\lesssim1$ Gyr). Our observations and simulations disagree on the effect of stellar mass, with GAMA showing massive galaxies transitioning faster, but is the opposite in SHARK. We find that environment only impacts galaxies below $\sim10^{10}$ M$_\odot$ in GAMA, with satellites having shorter $\tau^{}_\mathrm{Q}$ than centrals by $\sim0.4$ Gyr, with SHARK only in qualitative agreement. Finally, we compare to previous literature, finding consistency with timescales in the order of couple Gyr, but with several differences that we discuss.