On the impact of empirical and theoretical star formation laws on galaxy formation

We investigate the consequences of applying different star formation laws in the galaxy formation model GALFORM. Three broad star formation laws are implemented: the empirical relations of Kennicutt and Schmidt and Blitz & Rosolowsky and the theoretical model of Krumholz, McKee & Tumlinson. These laws have no free parameters once calibrated against observations of the star formation rate (SFR) and gas surface density in nearby galaxies. We start from published models, and investigate which observables are sensitive to a change in the star formation law, without altering any other model parameters. We show that changing the star formation law (i) does not significantly affect either the star formation history of the universe or the galaxy luminosity functions in the optical and near-infrared, due to an effective balance between the quiescent and burst star formation modes, (ii) greatly affects the cold gas contents of galaxies and (iii) changes the location of galaxies in the SFR versus stellar mass plane, so that a second sequence of 'passive' galaxies arises, in addition to the known 'active' sequence. We show that this plane can be used to discriminate between the star formation laws.