Characterizing Dust Attenuation in Local Star-Forming Galaxies: Near-Infrared Reddening and Normalization

TL;DR

This study measures near-infrared dust attenuation in about 5500 local star-forming galaxies, finding a generally consistent attenuation curve with some variations, useful for broad galaxy studies but less so for individual galaxy analysis.

Contribution

It provides a detailed characterization of the NIR dust attenuation curve for local star-forming galaxies, including normalization and sub-population variations, based on combined UKIRT, 2MASS, and UV-optical data.

Findings

Attenuation curve is slightly lower in the far-UV than starburst galaxies.

The normalization at V-band is R_V=3.67+/-0.44.

Broad consistency between inferred energy and infrared dust emission.

Abstract

We characterize the near-infrared (NIR) dust attenuation for a sample of ~5500 local (z<0.1) star-forming galaxies and obtain an estimate of their average total-to-selective attenuation $k(\lambda)$. We utilize data from the United Kingdom Infrared Telescope (UKIRT) and the Two Micron All-Sky Survey (2MASS), which is combined with previously measured UV-optical data for these galaxies. The average attenuation curve is slightly lower in the far-UV than local starburst galaxies, by roughly 15%, but appears similar at longer wavelengths with a total-to-selective normalization at V-band of $R_V=3.67\substack{+0.44 \\ -0.35}$. Under the assumption of energy balance, the total attenuated energy inferred from this curve is found to be broadly consistent with the observed infrared dust emission ($L_{\rm{TIR}}$) in a small sample of local galaxies for which far-IR measurements are available. However, the significant scatter in this quantity among the sample may reflect large variations in the attenuation properties of individual galaxies. We also derive the attenuation curve for sub-populations of the main sample, separated according to mean stellar population age (via $D_n4000$), specific star formation rate, stellar mass, and metallicity, and find that they show only tentative trends with low significance, at least over the range which is probed by our sample. These results indicate that a single curve is reasonable for applications seeking to broadly characterize large samples of galaxies in the local Universe, while applications to individual galaxies would yield large uncertainties and is not recommended.