CASCO: Cosmological and AStrophysical parameters from Cosmological simulations and Observations – I. Constraining physical processes in local star-forming galaxies

Abstract

ABSTRACT We compare the structural properties and dark matter content of star-forming galaxies taken from the Cosmology and Astrophysics with MachinE Learning Simulations (camels) cosmological simulations to the observed trends derived from the SPARC sample in the stellar mass range $[10^{9}, 10^{11}]\, \textrm {M}_{\odot }$, to provide constraints on the value of cosmological and astrophysical (supernova- and active galactic nucleus-related) parameters. We consider the size–, internal DM fraction–, internal DM mass–, and total–stellar mass relations for all the 1065 simulations, all having different cosmological and astrophysical parameters, from the IllustrisTNG, SIMBA, and ASTRID suites of camels, and search for the parameters that minimize the χ2 with respect to the observations. For the IllustrisTNG suite, we find the following constraints for the cosmological parameters: $\Omega _{\textrm {m}} = 0.27_{-0.05}^{+0.01}$, $\sigma _{8} = 0.83_{-0.11}^{+0.08}$, and $S_{8} = 0.78_{-0.09}^{+0.03}$, which are consistent within 1σ with the results from the 9-yr WMAP observations. SN feedback-related astrophysical parameters, which describe the departure of outflow wind energy per unit star formation rate and wind velocity from the reference IllustrisTNG simulations, assume the following values: $A_{\textrm {SN1}} = 0.48_{-0.16}^{+0.25}$ and $A_{\textrm {SN2}} = 1.21_{-0.34}^{+0.03}$, respectively. Therefore, simulations with a lower value of outflow wind energy per unit star formation rate with respect to the reference illustrisTNG simulation better reproduce the observations. Variation of active galactic nucleus feedback parameters, on the other hand, shows negligible effects on the scaling relation trends in the mass range probed. Simulations based on SIMBA and ASTRID suites predict central dark matter masses substantially larger than those observed in real galaxies, which can be reconciled with observations only by requiring values of Ωm inconsistent with cosmological constraints for SIMBA, or simulations characterized by unrealistic galaxy mass distributions for ASTRID.