2019MNRAS.483.4020K

Despite the influx of unprecedented-quality data from the Fermi Gamma-Ray Space Telescope that have been collected over 9 yr of operation, the contribution of normal star-forming galaxies to the extragalactic gamma-ray background (EGRB) remains poorly constrained. Different estimates are discrepant both in their underlying physical assumptions and their results. With several detections and many upper limits for the gamma-ray fluxes of nearby star-forming galaxies now available, estimates that rely on empirical scalings between gamma-ray and longer wavelength luminosities have become possible and increasingly popular. In this paper, we examine factors that can bias such estimates, including (a) possible sources of non-trivial redshift dependence; (b) dependence on the choice of star formation tracer; (c) uncertainties in the slope and normalization of empirical scalings. We find that such biases can be significant, pointing towards the need for more sophisticated models for the star-forming galaxy contribution to the gamma-ray background, implementing more and more confident physics in their build-up. Finally, we show that there are large regions of acceptable parameter space in observational inputs that significantly overproduce the gamma-ray background, implying that the observed level of the background can yield significant constraints on models of the average cosmic gamma-ray emissivity associated with star formation.