2015ApJ...799...96G

We use a sample of 36 galaxies from the KINGFISH (Herschel IR), HERACLES (IRAM CO), and THINGS (Very Large Array H I) surveys to study empirical relations between Herschel infrared (IR) luminosities and the total mass of the interstellar gas (H₂+ H I). Such a comparison provides a simple empirical relationship without introducing the uncertainty of dust model fitting. We find tight correlations, and provide fits to these relations, between Herschel luminosities and the total gas mass integrated over entire galaxies, with the tightest, almost linear, correlation found for the longest wavelength data (SPIRE 500). However, we find that accounting for the gas-phase metallicity (affecting the dust to gas ratio) is crucial when applying these relations to low-mass, and presumably high-redshift, galaxies. The molecular (H₂) gas mass is found to be better correlated with the peak of the IR emission (e.g., PACS160), driven mostly by the correlation of stellar mass and mean dust temperature. When examining these relations as a function of galactocentric radius, we find the same correlations, albeit with a larger scatter, up to a radius of r ∼ 0.7 r₂₅(containing most of a galaxy's baryonic mass). However, beyond that radius, the same correlations no longer hold, with increasing gas (predominantly H I) mass relative to the infrared emission. The tight relations found for the bulk of the galaxy's baryonic content suggest that total gas masses of disk-like (non-merging/ULIRG) galaxies can be inferred from far-infrared continuum measurements in situations where only the latter are available, e.g., in ALMA continuum observations of high-redshift galaxies.