The MOSDEF Survey: [S III] as a New Probe of Evolving Interstellar Medium Conditions

We present measurements of [S III]λλ9069,9531 for a sample of z ∼ 1.5 star-forming galaxies, the first representative sample with measurements of these lines at z ≳ 0.1. We employ the line ratio S ₃₂ ≡ [S III]λλ9069,9531/[S II]λλ6716,6731 as a novel probe of evolving interstellar medium (ISM) conditions. Since this ratio includes the low-ionization line [S II], it is crucial that the effects of diffuse ionized gas (DIG) on emission-line ratios be accounted for in z ∼ 0 galaxy spectra, or else that comparisons be made to samples of local H II regions in which DIG emission is not present. We find that S ₃₂ decreases with increasing stellar mass at both z ∼ 1.5 and z ∼ 0, but with a shallow slope suggesting S ₃₂ has a weak dependence on metallicity, in contrast with [O III]/[O II] that displays a strong metallicity dependence. As a result, S ₃₂ only mildly evolves with redshift at fixed stellar mass. The z ∼ 1.5 sample is systematically offset toward lower S ₃₂ and higher [S II]/Hα at fixed [O III]/Hβ relative to z = 0 H II regions. We find that such trends can be explained by a scenario in which the ionizing spectrum is harder at fixed O/H with increasing redshift, but are inconsistent with an increase in ionization parameter at fixed O/H. This analysis demonstrates the advantages of expanding beyond the strongest rest-optical lines for evolutionary studies, and the particular utility of [S III] for characterizing evolving ISM conditions and stellar compositions. These measurements provide a basis for estimating [S III] line strengths for high-redshift galaxies, a line that the James Webb Space Telescope will measure out to z ∼ 5.5.

Based on data obtained at the W.M. Keck Observatory, which is operated as a scientific partnership among the California Institute of Technology, the University of California, and NASA, and was made possible by the generous financial support of the W.M. Keck Foundation.