Faint galaxy photometry and cosmology
Deep galaxy catalogues have been constructed from automatic measurements of photographic plates by the COSMOS machine at the Royal Observatory Edinburgh. The plates were taken by the 1.2m UK Schmidt telescope (UKST) and 4m Anglo-Australian telescope (AAT), in both blue and red passbands. The UKST plates cover an area of sky of~170 square degrees, some four times larger than any previous study to these depths (B-21, R~20mag).By comparing the UKST and AAT galaxy number-magnitude counts and colour distributions with those predicted using computer models, evidence for luminosity evolution has been obtained. The red passband counts require less luminosity evolution than in the blue passband and at the faintest magnitudes reached here (R'-22mag) the cosmological parameter, q(_o), has as large an effect. The red count models are well enough determined to reject world models with q(_o)> 1. In order to further separate the effects of luminosity evolution and q^, the possibility of using a well determined Hubble diagram or faint "galaxy redshift surveys is considered. The galaxy two-point angular correlation function, w(ɵ), has been estimated from the UKST catalogues and shows evidence of a feature at large angular scales, corresponding to a spatial separation of ~3h Mpc (H(_o) = l00h kms Mpc ). In a study of the correlation function scaling relation it is found that the observed clustering amplitudes at AAT depths are lower than those predicted using well determined models that assume no clustering evolution. However, sampling errors are large and more 4m data is required in order to test the reality of this result. Also discussed is the possibility of discriminating between recent theories of galaxy formation using the w(#) observations. The method of Turner and Gott has been used to automatically detect groups and clusters of galaxies in the UKST catalogues. It is found that the cluster-cluster w(f) is several times higher than the galaxy-galaxy w(4) when scaled to the same depth. The implications of this result for galaxy formation theories are discussed. By using the average magnitude, m, of a cluster as a distance estimator the redshift distribution of the clusters has been obtained. Features present in these distributions^ may correspond to the effects of superclustering on scales < l50h (^-1)Mpc. The modelled m:distance relation has been used to set constraints on the galaxy luminosity function and hence help to more tightly constrain the number count and clustering models. The orientations of galaxies within clusters and the orientations and ellipticity distributions of the clusters themselves have been used in order to obtain further constraints on the theories of galaxy formation.