Ultra-Diffuse Galaxies as Extreme Star-forming Environments I: Mapping Star Formation in HI-Rich UDGs

TL;DR

This study investigates star formation in ultra-diffuse galaxies (UDGs), revealing they form stars normally but with low efficiency, challenging some galaxy formation models and providing detailed spatial maps of star formation activity.

Contribution

It introduces a new joint SED fitting method combining optical, UV, and HI data to map star formation in HI-rich UDGs at high spatial resolution.

Findings

UDGs have low star formation efficiencies at 500 pc scales.

Stellar sizes of UDGs are comparable to other dwarfs at fixed HI mass.

Observations challenge models predicting stellar expansion from feedback.

Abstract

Ultra-Diffuse Galaxies are both extreme products of galaxy evolution and extreme environments in which to test our understanding of star formation. In this work, we contrast the spatially resolved star formation activity of a sample of 22 HI-selected UDGs and 35 low-mass galaxies from the NASA Sloan Atlas (NSA) within 120 Mpc. We employ a new joint SED fitting method to compute star formation rate and stellar mass surface density maps that leverage the high spatial resolution optical imaging data of the Hyper Suprime-Cam Subaru Strategic Program (HSC-SSP) and the UV coverage of GALEX, along with HI radial profiles estimated from a subset of galaxies that have spatially resolved HI maps. We find that the UDGs have low star formation efficiencies as a function of their atomic gas down to scales of 500 pc. We additionally find that the stellar mass-weighted sizes of our UDG sample are unremarkable when considered as a function of their HI mass -- their stellar sizes are comparable to the NSA dwarfs at fixed HI mass. This is a natural result in the picture where UDGs are forming stars normally, but at low efficiencies. We compare our results to predictions from contemporary models of galaxy formation, and find in particular that our observations are difficult to reproduce in models where UDGs undergo stellar expansion due to vigorous star formation feedback should bursty star formation be required down to $z=0$.