Title:

Nonlinear effects in the evolution of galaxy clustering

The evolution of largescale structure in the Universe is well understood in the linear regime, where the density is close to the mean density. Where the density contrast is large, the linearization of the equations of motion is no longer valid, and new techniques are needed. To this end, analytic arguments are combined with nbody simulations in Chapter 2, resulting in an analytic correction for the nonlinear evolution of clustering. This method, and models of bias and redshiftspace distortion, are then applied to a number of observational power spectra, in order to reconstruct the linear power spectrum of cosmic mass fluctuations. Constraints are put on the values of bias parameters, and a high degree of redshiftspace distortion is required, Ω^{0.6}/b_{IRAS} = 1.0±0.2. A Cold Dark Matter power spectrum can be fit to the data, provided Ωh = 0.255. Chapter 3 is concerned with the formation of galaxy clusters through gravitational collapse. The nonlinear techniques developed in Chapter 2 are used to set up the initial conditions for numerical nbody simulations such that the final power spectra are nearly the same for two different cosmological models, Ω = 1 and Ω = 0.2. Galaxy clusters formed in these simulations are identified, and a mean density profile calculated. It is shown that although differences in power spectra have been largely eliminated, significant differences remain in the density profile under different cosmological conditions. In Chapter 4, the angular correlation function, w(θ), of faint blue galaxies is considered. Simple models of the evolution of clustering are unable to reproduce the observed w(θ) of the faint galaxies, overpredicting the amplitude of w(θ) by nearly an order of magnitude. The nonlinear evolution model of Chapter 2 is applied to the present epoch correlation function, and it is found that the agreement with the observations is significantly improved, and that the model predictions are consistent with the observations, provided that the faint blue galaxies lie at the highest redshift allowed by the observations. Low Ω models are disfavoured, as they are unable to reproduce the observed shape of w(θ), approximately described by a powerlaw. A Cold Dark Matter model, with Ω = 1, is well able to reproduce this shape.
