CASCO: Cosmological and AStrophysical parameters from Cosmological simulations and Observations: II. Constraining cosmology and astrophysical processes with early- and late-type galaxies

Physical processes can influence the formation and evolution of galaxies in diverse ways. It is essential to validate their incorporation into cosmological simulations by testing them against real data encompassing various types of galaxies and spanning a broad spectrum of masses and galaxy properties. For these reasons, in this second paper of the CASCO series, we compare the structural properties and dark matter content of early-type galaxies taken from the CAMELS IllustrisTNG cosmological simulations to three different observational datasets (SPIDER, ATLAS^3D, and MaNGA DynPop), to constrain the value of cosmological and astrophysical feedback parameters, and we compare the results with those obtained comparing the simulation expectations with late-type galaxies. We consider the size-mass, internal DM fraction-mass, and internal DM mass-stellar mass relations for all the simulations, and search for the best-fit simulation for each set of observations. For SPIDER, we find values for the cosmological parameters in line with both the literature and the results obtained from the comparison between simulations and late-type galaxies; results for the supernovae feedback parameters are instead opposite with respect to the previous results based on late-type galaxies. For ATLAS^3D, we find similar values as from SPIDER for the cosmological parameters, but we find values for the supernovae feedback parameters more in line with what we found for late-type galaxies. From MaNGA DynPop, we find extreme values for the cosmological parameters, while the supernovae feedback parameters are consistent with ATLAS^3D results. When considering the full MaNGA DynPop sample, including both late- and early-type galaxies, no single simulation can reproduce the full variety in the observational datasets. The constraints depend strongly on the specific properties of each observational trend, making it difficult to find a simulation matching all galaxy types, indicating the existence of limitations in the ability of simulations in reproducing the observations.