Cloud and star formation in disk galaxy models with feedback

We include feedback in global hydrodynamic simulations in order to study the star formation properties, and gas structure and dynamics, in models of galactic disks. In previous work we studied the growth of clouds and spiral substructure due to gravitational instability. We extend these models by implementing feedback in gravitationally bound clouds; momentum (due to massive stars) is injected at a rate proportional to the star formation rate. This mechanical energy disperses cloud gas back into the surrounding interstellar medium, truncating star formation in a given cloud and raising the overall level of ambient turbulence. Propagating star formation can however occur as expanding shells collide, enhancing the density and triggering new cloud and star formation. By controlling the momentum injection per massive star and the specific star formation rate in dense gas, we find that the negative effects of high turbulence outweigh the positive ones, and in net, feedback reduces the fraction of dense gas and, thus, the overall star formation rate. The properties of the large clouds that form are not, however, very sensitive to feedback, with cutoff masses of a few million M_⊙, similar to observations. We find a relationship between the star formation rate surface density and the gas surface density with a power-law index ∼2 for our models with the largest dynamic range, consistent with theoretical expectations for our model of disk flaring. We point out that the value of the "Kennicutt-Schmidt" index found in numerical simulations (and likely in nature) depends on the thickness of the disk, and therefore, a self-consistent determination must include turbulence and resolve the vertical structure. With our simple feedback prescription (a single combined star formation event per cloud), we find that global spiral patterns are not sustained; less correlated feedback and smaller scale turbulence appear to be necessary for spiral patterns to persist.