The origin of structure in the Universe is a central problem in Physics. Its solution will not only inform our understanding of the processes by which matter became organized into galaxies and clusters, but it will also help uncover the identity of the dark matter, offer insights into events that happened in the early stages of the Big Bang and provide a useful check on the values of the fundamental cosmological parameters estimated by other means.
Because of its non-linear character, lack of symmetry and general complexity, the formation of cosmic structure is best approached theoretically using numerical simulations. The problem is well posed because the initial conditions - small perturbations in the density and velocity field of matter - are, in principle, known from Big Bang theory and observations of the early Universe, while the basic physical principles involved are understood. The behaviour of the dark matter is governed primarily by gravity, while the formation of the visible parts of galaxies involves gas dynamics and radiative processes of various kinds. Using cosmological simulations it is possible to follow the development of structure from primordial pertubations to the point where the model can be compared with observations.
Over the past few years, there has been huge progress in quantifying observationally the properties of galaxies not only in the nearby universe, but also in the very distant universe. Since the clustering pattern of galaxies is rich with information about physics and cosmology, much effort is invested in mapping the distribution of galaxies at different epochs. Two large ongoing surveys, the US-based Sloan Digital Sky Survey (York et al. 2000), and the Anglo-Australian "2-degree field galaxy redshift survey" (2dFGRS, Colless et al. 2001), are revolutionizing our view of the nearby universe with order of magnitude increases in the amount of available data. Similarly, new data collected in the past five years or so have, for the first time, opened up the high redshift universe (1) to detailed statistical study (Steidel et al. 1996).
The advent of large computers, particularly parallel supercomputers, together with the development of efficient algorithms, has enabled the accuracy and realism of simulations to keep pace with observational progress. With the wealth of data now available, simulations are essential to interpret astronomical data and to link them to physical and cosmological theory.
1 In cosmology, distances to galaxies are estimated from the redshift of their spectral lines; higher redshifts correspond to more distant galaxies and thus to earlier epochs. Back.