ProGeny II: the impact of libraries and model configurations on inferred galaxy properties in SED fitting

Abstract

We use a volume-complete sample of ~8,000 galaxies from the GAMA survey to characterise the impact of stellar population libraries (SPLs) and model configurations on the resulting inferred galaxy properties from Spectral Energy Distribution (SED) fitting. We compare a fiducial SPL from ProGeny (a new tool that can generate SPLs quickly and flexibly) against five other commonly used SPLs using the SED-fitting code ProSpect. The impact of selecting each SPL is compared to the consequence of changing the model implementation in the SED fitting process, including the implementation of metallicity evolution versus a fixed or constant metallicity, and a functional parametric star formation history (SFH) versus a stepwise parametric (or "non-parametric") SFH. Furthermore, we use ProGeny to assess the impact of sub-SPL choices, including isochrone selection, stellar spectra selection, and IMF selection. Through a comparison of derived stellar masses, star formation rates, metallicities, ages, and the inferred cosmic star formation history (CSFH), we rank the impact of varying choices. Overall the assumption of a solar metallicity creates the greatest biases, with a substantial impact also caused by the choice of a specific SPL. To recover a CSFH most consistent with observations, we advocate for the use of the fiducial implementation with a skewed Normal functional form for the SFH, and an evolving metallicity, although we note that all studied SPLs underestimate the peak in the CSFH.

arXiv comment26 pages, 28 figures, accepted for publication to MNRAS, companion paper to Robotham & Bellstedt (ProGeny I)