THE ROLE OF DRY MERGERS FOR THE FORMATION AND EVOLUTION OF BRIGHTEST CLUSTER GALAXIES

Using a resimulation technique, we perform high-resolution cosmological simulations of dry mergers in a massive (10(15) M-circle dot) galaxy cluster identified in the Millennium Run. Our initial conditions include well resolved compound galaxy models consisting of dark matter halos and stellar bulges that are used to replace the most massive cluster progenitor halos at redshift z = 3, allowing us to follow the subsequent dry merger processes that build up the cluster galaxies in a self-consistent cosmological setting. By construction, our galaxy models obey the stellar mass-size relation initially. Also, we study both galaxy models with adiabatically contracted and uncompressed halos. We demonstrate that the brightest cluster galaxy (BCG) evolves away from the Kormendy relation as defined by the smaller mass galaxies (i.e., the relation bends). This is accompanied by a significantly faster dark matter mass growth within the half-light radius of the BCG compared to the increase in the stellar mass inside the same radius. As a result of the comparatively large number of mergers the BCG experiences, its total mass-to-light ratio becomes significantly higher than in typical elliptical galaxies. We also show that the mixing processes between dark matter and stars lead to a small but numerically robust tilt in the fundamental plane and that the BCG lies on the tilted plane. Our model is consistent with the observed steepening of the logarithmic mass-to-light gradient as a function of the stellar mass. As we have not included effects from gas dynamics or star formation, these trends are exclusively due to N-body and stellar dynamical effects. Surprisingly, we find only tentative weak distortion in the Faber-Jackson relation that depends on the aperture size, unlike expected based on studies of isolated merger simulations. This may be due to differences in the distribution of galaxy orbits, which is given in our approach directly by the cosmological context while it has to be assumed in isolated merger simulations, and the fact that the BCG is located deep in the cluster potential well. Another uncertainty in both approaches lies in the definition of the spatial extent of the BCG.