Nonlinear stochastic biasing of galaxies and dark halos in cosmological hydrodynamic simulations

We perform an extensive analysis of nonlinear and stochastic biasing of gal axies and dark halos in a spatially flat, low-density cold dark matter univ erse (Omega (o) = 0.3, lambda (o) = 0.7, and sigma (8) = 1) using cosmologi cal hydrodynamic simulations. We identify galaxies by linking cold and dens e gas particles that satisfy the Jeans criterion. We compare their biasing properties with the predictions of an analytic halo biasing model. Dark hal os in our simulations exhibit reasonable agreement with the predictions onl y on scales larger than similar to 10 h(-1) Mpc; on smaller scales, the vol ume exclusion effect of halos due to their finite size becomes substantial. Interestingly, the biasing properties of galaxies are better described by extrapolating the halo biasing model predictions. The clustering amplitudes of galaxies are almost independent of the redshift between z = 0 and 3, as reported in previous simulations. This in turn leads to a rapidly evolving biasing factor; we find that b(cov) similar or equal to 1 at redshift z si milar or equal to 0 and b(cov) similar or equal to 3-4 at z = 3, where b(co v) is a biasing parameter defined from the linear regression of galaxy and dark matter density fields. Those values are consistent with the observed c lustering of Lyman break galaxies. We also find the clear dependence of gal axy biasing on formation epoch; the distribution of old populations of gala xies tightly correlates with the underlying mass density field while that o f young populations is slightly more stochastic and antibiased relative to dark matter. The amplitude of the two-point correlation function of old pop ulations is about 3 times larger than that of young populations. Furthermor e, the old population of galaxies resides within massive dark halos while t he young galaxies are preferentially formed in smaller dark halos. Assuming that the observed early- and late-type galaxies correspond to the simulate d old and young populations of galaxies, respectively, all of these segrega tions of galaxies are consistent with observational ones for early- and lat e-type galaxies such as, e.g., the morphology-density relation of galaxies.