The star formation histories of quiescent ultra-diffuse galaxies and their dependence on environment and globular cluster richness

TL;DR

This study analyzes the stellar populations of 25 quiescent ultra-diffuse galaxies, revealing links between their formation timescales, environment, and globular cluster richness, and compares observations with simulations.

Contribution

It provides the largest spectroscopic sample of UDGs, identifying diverse formation histories and the existence of 'failed-galaxies' with specific properties, and evaluates simulation models.

Findings

UDGs with rapid formation have high [Mg/Fe] ratios.

High-density environments host earlier-forming UDGs.

Some UDGs resemble 'failed-galaxies' with high metallicity and GC richness.

Abstract

We derive the stellar population parameters of 11 quiescent ultra-diffuse galaxies (UDGs) from Keck/KCWI data. We supplement these with 14 literature UDGs, creating the largest spectroscopic sample of UDGs to date (25). We find a strong relationship between their $\alpha$-enhancement and their star formation histories: UDGs that formed on very short timescales have elevated [Mg/Fe] abundance ratios, whereas those forming over extended periods present lower values. Those forming earlier and faster are overall found in high-density environments, being mostly early infalls into the cluster. No other strong trends are found with infall times. We analyze the stellar mass-metallicity, age-metallicity and [Mg/Fe]-metallicity relations of the UDGs, comparing them to other types of low mass galaxies. Overall, UDGs scatter around the established stellar mass--metallicity relations of classical dwarfs. We find that GC-rich UDGs have intermediate-to-old ages, but previously reported trends of galaxy metallicity and GC richness are not reproduced with this spectroscopic sample due to the existence of GC-rich UDGs with elevated metallicities. In addition, we also find that a small fraction of UDGs could be 'failed-galaxies', supported by their GC richness, high $\alpha$-abundances, fast formation timescales and that they follow the mass-metallicity relation of z~2 galaxies. Finally, we also compare our observations to simulated UDGs. We caution that there is not a single simulation that can produce the diverse UDG properties simultaneously, in particular the low metallicity failed-galaxy like UDGs.