A Physical Model for the Coevolution of QSOs and Their Spheroidal Hosts

We present a physically motivated model for the early co-evolution of massivespheroidal galaxies and active nuclei at their centers. Within dark matterhalos, forming at the rate predicted by the canonical hierarchical clusteringscenario, the gas evolution is controlled by gravity, radiative cooling, andheating by feedback from supernovae and from the growing active nucleus.Supernova heating is increasingly effective with decreasing binding energy inslowing down the star formation and in driving gas outflows. The more massiveproto-galaxies virializing at earlier times are thus the sites of the fasterstar-formation. The correspondingly higher radiation drag fastens the angularmomentum loss by the gas, resulting in a larger accretion rate onto the centralblack-hole. In turn, the kinetic energy carried by outflows driven by activenuclei can unbind the residual gas, thus halting both the star formation andthe black-hole growth, in a time again shorter for larger halos. For the mostmassive galaxies the gas unbinding time is short enough for the bulk of thestar-formation to be completed before type Ia supernovae can substantiallyincrease the $Fe$ abundance of the interstellar medium, thus accounting for the$\alpha$-enhancement seen in the largest galaxies. The feedback from supernovaeand from the active nucleus also determines the relationship between theblack-hole mass and the mass, or the velocity dispersion, of the host galaxy,as well as the black-hole mass function. These and other model predictions arein excellent agreement with observations for a number of observables whichproved to be extremely challenging for all the current semi-analytic models,including the sub-mm counts and the corresponding redshift distributions, andthe epoch-dependent K-band luminosity function of spheroidal galaxies.Comment: 23 pages, 15 figures, revised after ApJ referee repor