Dust attenuation of the nebular regions and optical emission lines of $z\sim2$ star-forming galaxies

TL;DR

This study investigates dust attenuation differences between nebular regions and stellar continua in z~2 star-forming galaxies, revealing a correlation with star formation rate that helps clarify previous discrepancies.

Contribution

It introduces a spectral energy distribution fitting approach that accurately predicts nebular line strengths and demonstrates the need for higher dust correction for nebular emission at high SFRs.

Findings

Higher dust correction needed for nebular emission in high SFR galaxies.

Correlation between SFR and difference in nebular and stellar dust attenuation.

Model successfully reproduces UV slopes, IR luminosities, and emission lines.

Abstract

We use a sample of 149 spectroscopically confirmed UV-selected galaxies at $z\sim 2$ to investigate the relative dust attenuation of the stellar continuum and the nebular emission lines. For each galaxy in the sample, at least one rest-frame optical emission line (H$\alpha$/[NII]$\lambda6583$ or [OIII]$\lambda5007$) measurement has been taken from the litterature, and 41 galaxies have additional Spitzer/MIPS 24$\mu$m observations that are used to infer infrared luminosities. We use a spectral energy distribution (SED) fitting code that predicts nebular line strengths when fitting the stellar populations of galaxies in our sample, and we perform comparisons between the predictions of our models and the observed/derived physical quantities. We find that on average our code is able to reproduce all the physical quantities (e.g., UV $\beta$ slopes, infrared luminosities, emission line fluxes), but we need to apply a higher dust correction to the nebular emission compared to the stellar emission for the largest SFR ($\log\mathrm{(SFR/M}_\odot\mathrm{yr}^{-1})>1.87$, Salpeter IMF). We find a correlation between SFR and the difference in nebular and stellar color excesses, which could resolve the discrepant results regarding nebular dust correction at $z\sim2$ from previous results.