Josh Nollenberg, Ph.D. 2004

Galaxy-QSO Cross-Correlations: Weak Gravitational Lensing and Systematics

Josh Nollenberg

Under the supervision of Dr. L.L.R. Williams


A disparity greater than an order of magnitude between theoretical expectations and observed galaxy-QSO cross-correlations has been a longstanding problem. This work investigates a number of avenues in an attempt to explain this discrepancy through theoretical modeling, the presentation of new cross-correlation measurements, and analysis of potential systematic effects.

In addition to the confirmation of a cross-correlation between bright, high-redshift LBQS QSOs and z ~ 0.25 APM UKST galaxies that were first observed by Williams and Irwin (1998), I present new results from correlations of Hamburg-ESO Bright QSO Survey and 2dF QSOs with APM UKST galaxies. Positive, null and negative correlations are observed in accordance with the theoretical expectation from magnification bias that the type of correlations are dependent on the slope of the QSO sample cumulative number magnitude counts. The integrated cross-correlation signals are also consistent, within uncertainties with the expected increase in the correlation amplitude with increasing source redshift. Yet the amplitudes of the cross-correlations are significantly greater than anticipated. Despite a search of a number of potential sources, this discrepancy remains unexplained.

Various cross-correlation estimators found in the literature are compared. Cross-correlation estimates are typically within ~5% of one another. Uncertainties due to QSO catalogue survey masks are also explored using the mask from the 2dF QSO survey. In this case, there was a 10% underestimation of the cross-correlation when correction for the mask was not implemented.

The presence of dust can also generate positive- or negative- associations, depending on whether the dust is located in Galactic or galaxy cluster environments. No significant amounts of dust were discovered in either case.