Ages and metallicities of central and satellite galaxies: implications for galaxy formation and evolution

Using the stellar ages and metallicities of galaxies in the Sloan Digital Sky Survey (SDSS) obtained by Gallazzi et al. and the SDSS galaxy group catalogue of Yang et al., we study how the stellar ages and metallicities of central and satellite galaxies depend on stellar mass, M*, and halo mass, Mh. We find that satellites are older and metal richer than centrals of the same stellar mass, and this difference increases with decreasing M*. In addition, the slopes of the age–stellar mass and metallicity–stellar mass relations are found to become shallower in denser environments (more massive haloes). This is due to the fact that the average age and metallicity of low-mass satellite galaxies (M*≲ 1010h−2 M⊙) increase with the mass of the halo in which they reside. In order to gain understanding of the physical origin of these trends, we compare our results with the semi-analytical model of Wang et al. The model, which predicts stellar mass functions and two-point correlation functions in good overall agreement with observations, also reproduces the fact that satellites are older than centrals of the same stellar mass and that the age difference increases with the halo mass of the satellite. This is a consequence of the fact that satellites are stripped of their hot gas reservoir shortly after they are accreted by their host haloes (strangulation). The ensuing quenching of star formation leaves the stellar populations of satellites to evolve passively, while the prolonged star formation activity of centrals keeps their average ages younger. The resulting age offset is larger in more massive environments because their satellites were accreted earlier. The model does not reproduce he halo mass dependence of the metallicities of low-mass satellites, yields metallicity–stellar mass and age–stellar mass relations that are too shallow and predicts that satellite galaxies have the same metallicities as centrals of the same stellar mass, in disagreement with the data. We argue that these discrepancies are likely to indicate the need to (i) modify the recipes of both supernova feedback and active galactic nucleus feedback, (ii) use a more realistic description of strangulation and (iii) include a proper treatment of the tidal stripping, heating and destruction of satellite galaxies.