Modeling the co-evolution of supermassive black holes and galaxies: I. Black hole scaling relations and the Active Galactive Nuclei luminosity function

We model the cosmological co-evolution of galaxies and their central supermassive black holes (BHs) within a semi-analytical framework developed on the outputs of the Millennium Simulation. This model, described in detail by Croton et al. and De Lucia and Blaizot, introduces a `radio mode' feedback from active galactic nuclei (AGN) at the centre of X-ray emitting atmospheres in galaxy groups and clusters. Thanks to this mechanism, the model can simultaneously explain: (i) the low observed mass dropout rate in cooling flows; (ii) the exponential cut-off in the bright end of the galaxy luminosity function and (iii) the bulge-dominated morphologies and old stellar ages of the most massive galaxies in clusters. This paper is the first of a series in which we investigate how well this model can also reproduce the physical properties of BHs and AGN. Here we analyse the scaling relations, the fundamental plane and the mass function of BHs, and compare them with the most recent observational data. Moreover, we extend the semi-analytic model to follow the evolution of the BH mass accretion and its conversion into radiation, and compare the derived AGN bolometric luminosity function with the observed one. While we find for the most part a very good agreement between predicted and observed BH properties, the semi-analytic model underestimates the number density of luminous AGN at high redshifts, independently of the adopted Eddington factor and accretion efficiency. However, an agreement with the observations is possible within the framework of our model, provided it is assumed that the cold gas fraction accreted by BHs at high redshifts is larger than at low redshifts