Ultraviolet spectra of extreme nearby star-forming regions: evidence for an overabundance of very massive stars

TL;DR

This study uses ultraviolet spectroscopy of nearby star-forming regions to investigate the presence of very massive stars and tests stellar models, revealing an overabundance of such stars likely caused by binary interactions, which impacts galaxy evolution understanding.

Contribution

It provides new ultraviolet spectra of young stellar populations and highlights the need to include binary interactions in stellar models to explain observed features.

Findings

Detection of nebular C III] with low equivalent widths at high metallicity.

Strong stellar wind features indicating an overabundance of very massive stars.

Models with binary interactions better reproduce observed stellar wind features.

Abstract

As deep spectroscopic campaigns extend to higher redshifts and lower stellar masses, the interpretation of galaxy spectra depends increasingly upon models for very young stellar populations. Here we present new HST/COS ultraviolet spectroscopy of seven nearby ($<120$ Mpc) star-forming regions hosting very young stellar populations ($\sim$ 4-20 Myr) with optical Wolf-Rayet stellar wind signatures, ideal laboratories in which to test these stellar models. We detect nebular C III] in all seven, but at equivalent widths uniformly $< 10$ {\AA}. This suggests that even for very young stellar populations, the highest equivalent width C III] emission at $\geq 15$ {\AA} is reserved for inefficiently-cooled gas at metallicities at or below that of the SMC. The spectra also reveal strong C IV P-Cygni profiles and broad He II emission formed in the winds of massive stars, including some of the most prominent He II stellar wind lines ever detected in integrated spectra. We find that the latest stellar population synthesis prescriptions with improved treatment of massive stars nearly reproduce the entire range of stellar He II wind strengths observed here. However, we find that these models cannot simultaneously match the strongest wind features alongside the optical nebular line constraints. This discrepancy can be naturally explained by an overabundance of very massive stars produced by a high incidence of binary mass transfer and mergers occurring on short $\lesssim 10$ Myr timescales, suggesting these processes may be crucial for understanding the highest-sSFR galaxies in the early Universe. Reproducing both the stellar and nebular light of young systems such as these will be a crucial benchmark for the next generation of stellar population synthesis models.