Ionization correction factors and dust depletion patterns in giant HII regions

TL;DR

This paper introduces new ionization correction factors for key elements in giant H II regions, derived from photoionization models calibrated with extensive observational data, and explores dust depletion patterns and their implications.

Contribution

It provides the first analytical expressions for uncertainties of ICFs and applies these to determine elemental abundances and dust depletion patterns in giant H II regions.

Findings

Oxygen is depleted into dust grains up to 0.12 dex at solar metallicity.

New ICFs are robust and based on a large observationally calibrated model grid.

Dust depletion patterns vary with metallicity and impact elemental abundance estimates.

Abstract

We provide new ionization correction factors (ICFs) for carbon, nitrogen, neon, sulfur, chlorine, and argon in giant H II regions. The ICFs were computed using the most representative photoionization models from a large initial grid. The models were selected using an observational sample of 985 giant H II regions (GHR) in spiral galaxies and blue compact galaxies (BCG). The observational sample was also used to assign a weight to each model describing how well it agrees with observations in the [O III]/Hbeta versus [N II]/Halpha diagram. In addition to the ICFs we provide, for the first time, analytical expressions for their formal uncertainties. We use our ICFs to compute the abundances of nitrogen, neon, sulfur, and argon in our samples. Our abundances are robust within the adopted framework, but may require revision in the case of important changes in atomic data or in the spectral energy distribution of the ionizing radiation in H II regions. Considering the abundance patterns we obtained for the BCG sample (abundances for the GHR sample are less reliable) we find that oxygen is depleted into dust grains at a rate increasing with metallicity and reaching 0.12 dex at solar abundances. The discussion of possible depletion of sulfur and argon requires considering recent Type Ia Supernova yields, which are still uncertain.