Present - day clustering of radio galaxies

This review discusses three ways in which radio galaxies and other high-redshift objects can give us information on the nature and statistics of cosmological inhomogeneities, and how they have evolved between high redshift and the present: (1) The present-day spatial distribution and clustering of radio galaxies; (2) The evolution of radio-galaxy clustering and biased clustering at high redshift; (3) Measuring density perturbation spectra from the abundances of high-redshift galaxies. 1 Present-day clustering of radio galaxies Radio galaxies are interesting probes of large-scale structure in the universe, and give a view of the galaxy distribution which differs significantly from that obtained in other wavebands. Radio selection is uniform over the sky and independent of galactic extinction, so that reliably complete catalogues can be obtained over large areas. As a result, the apparent uniformity of the distribution of radio sources in early surveys such as 4C and Parkes gave the first convincing evidence for the large-scale homogeneity of the Universe (Webster 1977). This uniformity arises because even relatively bright samples of radio galaxies are at redshifts z ≃ 1, so that projection effects give a huge dilution of any intrinsic spatial clustering. The 3D clustering of radio galaxies was first detected by Peacock & Nicholson (1991; PN91), using a redshift survey of 329 galaxies with z < 0.1 and S(1.4 GHz) > ∼ 500 mJy. The result was a correlation function measured in redshift space of ξ rg (s) = [s/11 h −1 Mpc] −1.8 (1) (h ≡ H 0 /100 km s −1 Mpc −1). This corresponds to a clustering amplitude intermediate between normal galaxies and rich clusters of galaxies, as seems appropriate give that radio galaxies are normally found in moderately rich groups (e.g. Allington-Smith et al. 1993). However, redshift surveys move on rapidly: in the region of 10 5 galaxy redshifts are now known, and it should be possible to do very much better than PN91 today. Particularly, the PN91 survey has a very low density 1