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

TL;DR

This study compares simulated and observed galaxy properties across different types to constrain cosmological and astrophysical parameters, revealing limitations in current models to fully reproduce galaxy diversity.

Contribution

It provides a detailed comparison of early- and late-type galaxy properties from simulations and observations to refine feedback parameters and highlight modeling challenges.

Findings

Cosmological parameters for SPIDER align with literature.

Supernova feedback parameters differ between galaxy types.

No single simulation reproduces all observed galaxy trends.

Abstract

Physical processes impact galaxy formation and evolution in diverse ways, requiring validation of their implementation in cosmological simulations through comparisons with real data across various galaxy types and properties. In this second paper of the CASCO series, we compare the structural properties and dark matter (DM) content of early-type galaxies from the CAMELS IllustrisTNG simulations to three observational datasets (SPIDER, $\textrm{ATLAS}^{\textrm{3D}}$, and MaNGA DynPop), to constrain cosmological and astrophysical feedback parameters, contrasting these results with those obtained for late-type galaxies. We analyze the size-, internal DM fraction-, and DM mass-stellar mass relations, identifying the best-fit simulation for each dataset. For SPIDER, we find cosmological parameter values consistent with literature and results obtained from the comparison between simulations and late-type galaxies, with supernova feedback parameters differing from results derived for late-type galaxies. For $\textrm{ATLAS}^{\textrm{3D}}$, cosmological parameter results align with SPIDER, while supernova feedback parameters are more consistent with late-type galaxies results. MaNGA DynPop yields extreme cosmological parameter values but similar supernova feedback results to $\textrm{ATLAS}^{\textrm{3D}}$. However, no single simulation matches the full range of observational trends, especially when combining early- and late-type galaxies from MaNGA DynPop. These findings highlight the limitations of simulations in reproducing diverse galaxy properties, underscoring the challenge of capturing the complexity of galaxy formation across all types.