Galaxies inN‐Body Simulations: Overcoming the Overmerging Problem

We present analysis of the evolution of dark matter halos in denseenvironments of groups and clusters in dissipationless cosmologicalsimulations. The premature destruction of halos in such environments, known asthe overmerging, reduces the predictive power of N-body simulations and makesdifficult any comparison between models and observations. We analyze thepossible processes that cause the overmerging and assess the extent to whichthis problem can be cured with current computer resources and codes. Using bothanalytic estimates and high resolution numerical simulations, we argue that theovermerging is mainly due to the lack of numerical resolution. We find that theforce and mass resolution required for a simulated halo to survive in galaxygroups and clusters is extremely high and was almost never reached before: ~1-3kpc and 10^8-10^9 Msun, respectively. We use the high-resolution AdaptiveRefinement Tree (ART) N-body code to run cosmological simulations with theparticle mass of \approx 2x10^8/h Msun} and the spatial resolution of \approx1-2/h kpc, and show that in these simulations the halos do survive in regionsthat would appear overmerged with lower force resolution. Nevertheless, thehalo identification in very dense environments remains a challenge even withthe resolution this high. We present two new halo finding algorithms developedto identify both isolated and satellite halos that are stable (existed atprevious moments) and gravitationally bound. To illustrate the use of thesatellite halos that survive the overmerging, we present a series of halostatistics, that can be compared with those of observed galaxies. (Abridged)Comment: Accepted for publication in ApJ, substantional revisions after the first version, LaTeX 23 pages, 18 figs. (uses emulateapj.sty), Full-resolution version of Fig.9 is available upon reques