My primary research focus is galaxy formation and evolution, which I study using high performance computer simulations with and without hydrodynamics. They have led to many key insights, including the demonstration that adding dissipative gas dynamics to simulations of CDM cosmological models leads to the formation of systems with (approximately) the sizes and masses of observed galaxies, the demonstration that the Lyman Alpha forest arises naturally from inhomogeneities in the diffuse intergalactic medium in CDM cosmology and that the Lyman Alpha forest is a powerful cosmological probe in its own right, prediction of abundant and highly clustered populations of high redshift galaxies, and prediction of luminosity and age-dependent galaxy bias in good agreement with observed trends.
While direct confrontation of theory and observations is always one of our goals, we have concentrated throughout our work on using simulations to illuminate the physical processes that govern structure formation, turning on different physical effects one at a time to understand their impact, connecting our simulation results to those of simpler numerical or analytic approaches, and systematically investigating resolution and other numerical issues. A result of particular significance is the demonstration that galaxies gain much of their mass via "cold'' accretion, in which the gas does not undergo a virial shock but instead flows in along filamentary structures and remains cold. The distinction between cold accretion and hot accretion from a shock heated, quasi-static gaseous halo is one of the fundamental dichotomies in the physics of galaxy formation.