The Spitzer Survey of Stellar Structure in Galaxies (S$^4$G): Precise Stellar Mass Distributions from Automated Dust Correction at 3.6 microns

TL;DR

This paper presents a new pipeline using Independent Component Analysis to accurately separate stellar light from dust emission at 3.6 microns in over 1250 galaxies, enabling precise stellar mass mapping.

Contribution

The authors develop and apply an automated dust correction method for 3.6 micron data, improving stellar mass estimates across a large galaxy sample.

Findings

10-30% of 3.6 micron light is from dust emission, correlated with star formation.

Calibrated a mass-to-light ratio relation based on observed color.

Provided a large, publicly available set of stellar mass maps.

Abstract

The mid-infrared is an optimal window to trace stellar mass in nearby galaxies and the 3.6$\mu m$ IRAC band has been exploited to this effect, but such mass estimates can be biased by dust emission. We present our pipeline to reveal the old stellar flux at 3.6$\mu m$ and obtain stellar mass maps for more than 1600 galaxies available from the Spitzer Survey of Stellar Structure in Galaxies (S$^{4}$G). This survey consists of images in two infrared bands (3.6 and 4.5$\mu m$), and we use the Independent Component Analysis (ICA) method presented in Meidt et al. (2012) to separate the dominant light from old stars and the dust emission that can significantly contribute to the observed 3.6$\mu m$ flux. We exclude from our ICA analysis galaxies with low signal-to-noise ratio (S/N < 10) and those with original [3.6]-[4.5] colors compatible with an old stellar population, indicative of little dust emission (mostly early Hubble types, which can directly provide good mass maps). For the remaining 1251 galaxies to which ICA was successfully applied, we find that as much as 10-30% of the total light at 3.6$\mu m$ typically originates from dust, and locally it can reach even higher values. This contamination fraction shows a correlation with specific star formation rates, confirming that the dust emission that we detect is related to star formation. Additionally, we have used our large sample of mass estimates to calibrate a relationship of effective mass-to-light ratio ($M/L$) as a function of observed [3.6]-[4.5] color: $\log(M/L)=-0.339 (\pm 0.057) \times ([3.6]-[4.5]) -0.336 (\pm 0.002)$. Our final pipeline products have been made public through IRSA, providing the astronomical community with an unprecedentedly large set of stellar mass maps ready to use for scientific applications.