Title:

Clustering and redshiftspace distortions in QSO and galaxy surveys

In this thesis, we exploit the potential of existing QSO and galaxy surveys for investigating the nature of the large scale structure in the Universe. A detailed analysis of clustering and redshift space distortions allows us to constrain cosmological parameters. We model the anisotropies due to dynamical and geometrical effects in the measured clustering pattern of distant QSOs from the 2dF QSO Survey (2QZ) and also Lyman break galaxies. The 2QZ is then combined with the QSO sample from the 2dF SDSS LRG and QSO Survey (2SLAQ) to study the luminosity dependence of QSO clustering. Using ATbody simulations, we estimate the statistical gains in the determination of cosmological parameters from future LRG surveys. We measure the clustering of distant QSOs from the 2QZ survey by performing a correlation analysis of redshiftspace (zspace) distortions. To interpret the zspace correlation function measured in orthogonal directions, ع(σ,π)， we require an accurate model for the QSO realspace correlation function, ع (r). Motivated by the form for ع (r) seen in the 2dF Galaxy Redshift Survey (2dFGRS) and in standard ACDM predictions, we use a double powerlaw model for ع (r). which gives a good fit to the zspace and projected correlation functions. By fitting functional forms of ع(σ,π) which include both dynamical and geometrical modelling, we find, as expected, that β (which parameterises the infall into overdense regions) and the density of the Universe (Ω(^0_M)) are degenerate. However, this degeneracy can be lifted by using linear theory predictions under different cosmological scenarios. Using the 2QZ survey, we obtain: βQAO (z=1.4) = 0.50 (^0.13_0.15), Ω(^0_M) = 0.35 (^+0.19_0.13). The modelling of geometrical and dynamical anisotropies in the measured ع(σ,π) pattern is then applied to a sample of distant Lymaribreak galaxies. The presence of feedback mechanisms in these z〜 3 starforming galaxies heightens the importance of understanding the effects of zspace distortions. Despite the limited size of the fields, which hampers the determination of ع(r) at large scales, we find that a doublepower law ع(s) parameterisation is consistent with the correlation function measurements. This double power law model is then used as an input for the ع(σ,π) fitting and subsequent constraining of cosmological parameters. This investigation reveals that: ßLBG (z= 3) = 0.25± (^+0.05_0.05) and Ω(^0_M) = 0.55(^+0.45_0.16). The combination of the 2QZ with the fainter 2SLAQ QSO sample reveals that QSO clustering does not depend strongly on luminosity. This result is consistent with models which predict that haloes of similar mass can harbour QSOs of different luminosities. By assuming ellipsoidal models for the collapse of density perturbations, we test this hypothesis and estimate the mass of the dark matter haloes which the QSOs inhabit. We find that halo mass does not seem to evolve strongly with redshift nor depend on QSO luminosity. Having determined the black hole mass associated with the QSOs, we investigate how it correlates with luminosity and redshift and ascertain the relation between Eddington efficiency and black hole mass. Our results suggest that: (i) black hole mass does not depend strongly on accretion efficiency and (іі) black holes associated with QSOs of different luminosities have similar masses. Finally, the Hubble Volume simulation is used to construct a mock sample of a future ΑΑΩ Luminous Red Galaxy (LRG) survey. The ultimate aims of this survey are to identify the baryon acoustic features in the LRG clustering signal and to determine the equation of state of dark energy. We apply the zspace distortion analysis developed previously in the thesis to infer the statistical gain in terms of determinations of ß(_LRG) (z~ 0.7) and Ω(+0_m) .This thesis exploits the wealth of information contained in cosmological surveys, and demonstrates how the use of tools such as clustering statistics or zspace distortion analyses permit the extraction of such information.
