Heidelberg Joint Astronomical Colloquium

I will review and discuss very recent progress in modeling the formation of galaxies in a CDM Universe. In the last few years, improved resolution and sub-grid models of star formation and feedback processes have made it possible to form galaxies with realistic disks and global structural properties in numerical simulations. Yet, simulated galaxies generally suffer from an excessive central baryonic density, boast massive bulges that do not match galaxies of type later than Sa, including our own Milky Way, and have steep rotation curves. However, we show that once the numerical resolution is high enough to resolve the inhomogeneous structure of the ISM at scales of tens of parsecs, including the cold, star forming molecular gas phase, supernovae outflows can remove efficiently the low angular momentum material during galaxy assembly, preventing the formation of bulges in low-mass galaxies. The resulting galaxy exhibits a pure exponential profile and a slowly rising rotation curve that matches that of a low-mass disk galaxy. As a by product, the cuspy dark matter halo is impulsively heated when baryons are rapidly ejected via outflows, producing a shallower dark matter profile. We argue that our findings finally solve the small-scale crisis of CDM.

At larger mass scales, comparable or above that of the Milky Way, outflows would be more inefficient, and a bulge can still form. In massive galaxies feedback from AGNs could be the key to prevent the excessive concentration of baryons seen in the simulations, producing flat rotation curves in agreement with observations. While a diverse array of sub-grid recipes in semi-analytical models and simulations have been widely used to incorporate AGN feedback, we are still missing a key ingredient in the picture - we do not know how supermassive black holes form in the first place. I will describe an exciting new scenario in which massive black hole seeds (M > 10^5 Solar masses) form by direct collapse in gas-rich disk galaxy mergers at relatively high redshift (z > 6) and rapidly grow to form the billion solar mass black holes that power the bright QSOs detected by the Sloan Digital Sky Survey at z > 6. I conclude by discussing the implications of this scenario on bulge formation.