SOFIA observations of far-IR fine-structure lines in galaxies to measure metallicity

TL;DR

This study uses SOFIA far-IR spectroscopy to measure galaxy metallicities, revealing discrepancies with optical methods likely due to different ionization regions and dust effects, and proposing gas infall as a possible explanation.

Contribution

It introduces IR-based metallicity diagnostics for galaxies and compares them with optical estimates, highlighting systematic differences and potential physical causes.

Findings

IR and optical O/H--N/O relations are similar.

IR-based N/O abundances are systematically lower than optical.

No correlation found between N/O differences and ionization or dust extinction.

Abstract

We present new and archival SOFIA FIFI-LS far-IR spectroscopic observations of 25 local galaxies of either the [OIII]52um and/or the [NIII]57um lines. Including other 31 galaxies from Herschel-PACS, we discuss a local sample of 47 galaxies, including HII region, luminous IR, low-metallicity dwarf and Seyfert galaxies. Analyzing the mid- to far-IR fine-structure lines of this sample, we assess the metallicity and compare with the optical spectroscopy estimates. Using the IR, we find a similar O/H--N/O relation to that known in the optical. As opposite, we find systematically lower N/O IR abundances when compared to the optical determinations, especially at high values of N/O (log(N/O) > -0.8). We explore various hypotheses to account for this difference: (i) difference in ionization structure traced by optical (O+, N+ regions) versus IR lines (O++, N++ regions); (ii) contamination of diffuse ionized gas affecting the optical lines used to compute the N/O abundance; (iii) dust obscuration affecting the optical-based determinations. However, we have not found any correlation of the Delta(N/O)= (N/O)_OPT-(N/O)_IR with either ionization, or electron density, or optical extinction. We speculatively suggest that accretion of metal-poor gas from the circumgalactic medium could provide an explanation for this difference, because the rapid decrease of total abundances during infall is followed by a N/O ratio decrease due to primary production of young - possibly embedded - massive stars, are preferentially traced by the IR diagnostics, while optical diagnostics would better trace the secondary production, when both N/O and O/H abundance ratios will increase.