Devouring the Milky Way Satellites: Modeling Dwarf Galaxies with Galacticus

Abstract: Dwarf galaxies are ubiquitous throughout the universe and are extremely sensitive to various forms of internal and external feedback. Over the last two decades, the census of dwarf galaxies in the Local Group and beyond has increased markedly. While hydrodynamic simulations (e.g. FIRE II, MINT Justice League) have reproduced the observed dwarf properties down to the ultra-faints, such simulations require extensive computational resources to run. In this work, we constrain the standard physical implementations in the semi-analytic model Galacticus to reproduce the observed properties of the Milky Way satellites down to the ultra-faint dwarfs found in SDSS. We run Galacticus on merger trees from our high-resolution N-body simulation of a Milky Way analog. We determine the best-fit parameters by matching the cumulative luminosity function and luminosity-metallicity relation from both observations and hydrodynamic simulations. With the correct parameters, the standard physics in Galacticus can reproduce the observed luminosity function and luminosity-metallicity relation of the Milky Way dwarfs. In addition, we find a multi-dimensional match with half-light radii, velocity dispersions and mass-to-light ratios at z = 0 down to M_V <= -6 (L >= 10^4 L_solar). In addition to successfully reproducing the properties of the z = 0 Milky Way satellite population, our modeled dwarfs have star formation histories which are consistent with those of the Local Group dwarfs.