The APM Galaxy Survey III: An Analysis of Systematic Errors in the Angular Correlation Function and Cosmological Implications

Abstract

We present measurements of the angular two-point galaxy correlation function, $w(theta)$, from the APM Galaxy Survey. The performance of various estimators of $w$ is assessed using simulated galaxy catalogues and analytic arguments. Several error analyses show that residual plate-to-plate errors do not bias our estimates of $w$ by more than $10^{-3}$. Direct comparison between our photometry and external CCD photometry of over 13,000 galaxies from the Las Campanas Deep Redshift Survey shows that the rms error in the APM plate zero points lies in the range 0.04-0.05 magnitudes, in agreement with our previous estimates. We estimate the effects on $w$ of atmospheric extinction and obscuration by dust in our Galaxy and conclude that these are negligible. We use our best estimates of the systematic errors in the survey to calculate corrected estimates of $w$. Deep redshift surveys are used to determine the selection function of the APM Galaxy Survey, and this is applied in Limber's equation to compute how $w$ scales as a function of limiting magnitude. Our estimates of $w$ are in excellent agreement with the scaling relation, providing further evidence that systematic errors in the APM survey are small. We explicitly remove large-scale structure by applying filters to the APM galaxy maps and conclude that there is still strong evidence for more clustering at large scales than predicted by the standard scale-invariant cold dark matter (CDM) model. We compare the APM $w$ and the three dimensional power spectrum derived by inverting $w$, with the predictions of scale-invariant CDM models. We show that the observations require $Gamma=Omega_0 h$ in the range 0.2-0.3 and are incompatible with the value $Gamma=0.5$ of the standard CDM model.