The formation and evolution of dwarf galaxies in diverse cosmological environments

The lambda Cold Dark Matter (lambda CDM) cosmological model is the currently-preferred paradigm for explaining the large-scale structure of the cosmos. However, reconciling its small-scale predictions with observations of dwarf galaxies poses outstanding challenges (e.g. Bullock and Boylan-Kolchin, 2017). Classified by stellar masses below ~10^9Msol and dark matter halo masses below ~10^11 Msol, dwarf galaxies represent some of the oldest and most dark matter dominated systems in the cosmos, and their shallow gravitational potential wells make them particularly sensitive to a diverse range of internal and external astrophysical processes. They are therefore sensitive probes for testing the predictions of galaxy formation models *and* the prevailing cosmogony.

Deep optical surveys have recently fostered the discovery of many more, ever-fainter, dwarf galaxies, and it is expected that the forthcoming Vera Rubin Observatory will dramatically accelerate the rate of new discoveries. Existing observations have already revealed remarkable diversity among the fundamental properties of the faintest dwarfs, including their kinematics, size, gas content and star formation histories. To date, reproducing these properties in numerical models has proven challenging. Moreover, the very formation mechanism of the faintest dwarfs remains uncertain, and poses a major question at the forefront of contemporary astrophysics. The leading hypothesis is, at present, that they are relics of the early universe, which formed at high redshift and were quenched during the epoch of reionization (Simon, 2019).

This PhD thesis aims to scrutinise this hypothesis using cosmological zoom-in simulations run with the next-generation, state-of-the-art galaxy formation model: COLIBRE. COLIBRE ncorporates a more sophisticated treatment of interstellar gas physics than the current state-ofthe-art, in addition to superior dynamics achieved by using equal mass baryon and dark matter particles. It is therefore ideally suited to examination of dwarf galaxies. We will study dwarf galaxies in a diverse range of environments by creating a series of simulations complementary to the flagship COLIBRE run: these will be "zoomed" spherical regions, selected from the (300Mpc)^3 EAGLE-XL parent volume. Each will have a radius of 5-10Mpc (dictated by computational limitations), and with various mean overdensities but devoid of massive haloes (M200 > 10^12 Msol). The latter condition mitigates computational cost and allows us to increase the resolution to 10^4Msol per baryonic particle whilst still simulating a large volume.