The Clustering of Faint Galaxies

Prepared by S. Matarrese and L. Moscardini

The direct observation of samples of galaxies at high redshift will allow to yield important information about the origin and evolution of cosmic structures. In particular, the possibility of observing `normal' galaxies (or their progenitors), rather than just quasars or other ultraluminous sources, may lead to an understanding of how these objects relate to the distribution of the dark matter which is assumed, in most theories, to dominate the density of the Universe. This, in turn, should allow observations of galaxy clustering to be used to gain insights about fundamental aspects of cosmological models, as well as learning about the way galaxies themselves evolve.

Different cosmological paradigms, by construction, display similar characteristics at the present epoch, but can show very different behaviours at high redshifts and are seriously challenged by the simultaneous knowledge of the luminosity function, star formation history and clustering evolution.

Both the angular and spatial correlation functions have been extensively used to describe the evolving distribution of galaxies of different classes. However, no conclusion on the growth of the structure can be drawn without knowledge on the light-to-mass ratio among different galaxy types and the star formation associated to each ones which is crucial to understand the bias effect introduced in magnitude limited samples. Theoretical predictions depend on the relation - the effective bias - between the luminous objects and the matter density field. Therefore, even at high redshifts, problems of the degeneracy between the choice of a cosmology and the models of galaxy formation may occur. (cf. Matarrese et al. 1997; Moscardini et al. 1998).

Semi-analytical techniques, which use fits of high-resolution N-body simulations together with a `local' extension of the Press-Schechter theory, allow to follow the evolution of dark matter halo clustering. Using this technique with the outputs of hydro code simulations one can study the clustering properties of galaxies at high redshift as a function of their intrinsic luminosity. This, in turn, allows a large variety of models to be analyzed having an accurate sampling of long wavelength modes, keeping the essential astrophysical aspects inherent in the formation of each single structure. Of fundamental importance in this problem is the `bias factor' connecting the matter distribution to that of the various classes of cosmic structures. The bias significantly depends on the catalog selection criteria; furthermore, its redshift dependence allows to discriminate among different galaxy formation and evolution scenarios (e.g. passive evolution vs. merging).

Aim of our proposal will be to use the two-point correlation of faint galaxies and its redshift dependence as a test of theoretical predictions for the clustering evolution in competing theory of galaxy formation. In particular, the opportunity offered by LBT of reaching very faint magnitudes ( $m\approx 27$) will allow to build a large database where it will be possible to study the galaxy clustering for the same class of objects in the whole redshift interval $0 < z \le 4$: in this way one can avoid the modeling of the physical connection between objects at different redshifts. Finally the large size of the wide field imager for the LBT will give the opportunity to reliably compute the galaxy counts in cells and their higher moments. Since the bias factor enters the theoretical relations for these statistics in a different way, this can help to discriminate between various models of biasing and consequently between models of galaxy formation.