Description
Star clusters (SCs) are ubiquitous to most galaxies and are often used as tracers of galaxy formation and assembly. Their properties, e.g. metallicities, ages, spatial distribution, and masses, provide information about their formation environment. Understanding them at their birth, and subsequent evolution, is fundamental for their use as probes of galaxy evolution. Such understanding requires the cosmological environment, ideally from simulations, capturing the non-linear dynamics, internal and external, of the systems involved. However, modeling SCs in full cosmological simulations is challenging given the wide range of scales involved. Sub-grid or semi-analytic approaches have struggled to match observable counterparts, e.g. the Globular Cluster Mass Function (GCMF). In this work we present the implementation of a physically motivated sub-grid model for the formation and evolution of SCs in the Auriga galaxy simulation suite, that has been shown to produce realistic galaxies in the LCDM cosmological context. The formation of SCs is constrained to fully compressive media, and their evolution is shaped by their environment, as well as compact object remnants effects accelerating their evaporation via two-body interactions. We find that enhanced mass loss via two-body scattering from compact objects is critical to produce GCMFs compatible with observations (e.g. MW or M31), complementary to the findings of previous simulations with a similar treatment of the interstellar medium. This model is applied in the Auriga galaxy formation model at negligible extra computational cost making it ideal to make predictions for GC populations in a suite of MW-mass galaxies with a range of formation/assembly histories.
