No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

A semi-analytic galaxy formation model, N-body GALFORM, is developed which uses outputs from an N-body simulation to follow the merger histories of dark matter halos and treats baryonic processes using the semi-analytic model of Cole et al. We find that, apart from limited mass resolution, the only significant differences between this model and the Monte-Carlo based model of Cole et al. are due to known inaccuracies in the distribution of halo progenitor masses in the Monte-Carlo method. N-body GALFORM is used to compare Smooth Particle Hydrodynamics (SPH) and semi-analytic calculations of radiative cooling in the absence of star formation. We consider two cases: firstly, a simulation of a representative volume of the Universe with relatively poor mass resolution, and, secondly, a high resolution simulation of the formation of a single galaxy. We find good agreement between the models in terms of the mass of gas which cools in each halo, the masses of individual galaxies, and the spatial distribution of the galaxies. The semi-analytic model is then compared with a realistic, high-resolution galaxy simulation which includes prescriptions for star formation and feedback. A semi-analytic model without feedback is found to best reproduce the masses of the simulated galaxy and its progenitors. This model is used to populate a large volume with semi-analytic galaxies. The resulting luminosity function has an order of magnitude too many galaxies at high and low luminosities. We conclude that, while SPH and semi-analytic cooling calculations are largely consistent and therefore likely to be reasonably reliable, current numerical models of galaxy formation still contain major uncertainties due to the treatment of feedback, which will lead them to predict very different galaxy populations. Further work is required to find simulation algorithms which can simultaneously produce realistic individual galaxies and a population with reasonable statistical properties.

We study the environmental dependence of the formation history of dark matter haloes in a large dark matter simulation, the Millennium Run. Adopting a sensitive test of this dependence —— the marked correlation function —— reveals highly significant evidence that haloes of a given mass form earlier in denser regions. We explore the effect further using a new variant of this statistic, and confirm our results using some simpler tests made possible by the size and resolution of the simulation. We go on to study the effect of this environmental dependence on the galaxy population generated by a recent semi-analytic model run in the Millennium Run. We show that environmentally dependent halo formation imparts a small but cleanly detected change to the correlation function and void probability function of galaxies. We can model this change by applying a modulation based on local density to the halo occupation distribution of galaxies. We also note that having the correct placement scheme for galaxies within haloes is at least as important as correctly accounting for environmental effects. Two more dark matter simulations are run, and their outputs are appropriately relabelled and rescaled to represent different cosmologies. We generate consistent semi- analytic galaxy populations in these simulations, using two versions of each of three variants of our semi-analytic model. We compare the predictions for the galaxy clustering from these models to the projected two-point correlation function of the SDSS, obtaining a constraint on the amplitude of the fluctuations in the mass, σ(_8) = 0.96 士 0.05. We find that environmental effects do not significantly affect this estimate, but discuss other possible effects which might. We remark on how this result compares to other recent determinations of σ(_8).

Much can be learned about the formation of galaxies by taking a census of the present-day population. As a first step towards such an endeavour, I present luminosity and surface brightness distributions of 36 659 galaxies with K-band photometry from the UKIRT Infrared Deep Sky Survey (UKIDSS) Large Area Survey (LAS) Data Release 3 and optical photometry from Data Release 5 of the Sloan Digital Sky Survey (SDSS).