Massive and red objects predicted by a semianalytical model of galaxy formation

We study whether hierarchical galaxy formation in a concordance Lambda CDM universe can produce enough massive and red galaxies compared to the observations. We implement a semianalytical model in which the central black holes gain their mass during major mergers of galaxies and the energy feedback from active galaxy nuclei (AGNs) suppresses the gas cooling in their host halos. The energy feedback from AGNs acts effectively only in massive galaxies when supermassive black holes have been formed in the central bulges. Compared with previous models without black hole formation, our model predicts more massive and luminous galaxies at high redshift, agreeing with the observations of K20 up to z similar to 3. Also, the predicted stellar mass density from massive galaxies agrees with the observations of GDDS. Because of the energy feedback from AGNs, the formation of new stars is stopped in massive galaxies with the termination of gas cooling and these galaxies soon become red with color R - K > 5 (Vega magnitude), comparable to the extremely red objects (EROs) observed at redshift z similar to 1-2. Still, the predicted number density of very EROs is lower than observed at z similar to 2, and it may be related to inadequate descriptions of dust extinction, star formation history and AGN feedback in those luminous galaxies.