Abstract:
Semi-analytic (SA) galaxy formation models are widely used in mock catalogue building, model testing and parameter space investigation. In order to make all these applications robust, SA models should be as physical as possible. In this talk, I will focus on the modelling of gas cooling in SA models, which is an important process because it determines the mass and angular momentum delivered to central galaxies. I will first present a new gas cooling model, which, compared to previous models, has better estimation of gas thermal history, better consideration of hot gas halo evolution and better calculation of cooled down angular momentum. This new model is then compared with several major SA cooling models (GALFORM, L-GALAXIES and MORGANA), and with hydrodynamical simulations. It predicts more cooling and higher angular momentum than GALFORM in massive halos, but the predicted angular momentum is generally lower than L-GALAXIES and MORGANA. It generally agrees the best with hydrodynamical simulations among all SA models considered here. The comparison with simulations also provides some indications of cooling physics. At high redhsifts, the gas accretion in simulations is mainly through cold filaments, but the new cooling model with assumed spherical gas halo can still roughly capture the mass accretion rate, because both accretions are limited by free-fall; in the slow cooling regime, SA models ignore the compression work done on gas along with infall, leading to overestimation of cooling; low redshift rapid mass accretion onto halos tends to suppress cooling, but at high redshift this accretion has little effects on cooling.