We provide a consistent framework for estimating galaxy counts and variances in wide-field images for a range of photometric bands. The variances include both Poissonian noise and variations due to large-scale structure. We demonstrate that our statistical theory is consistent with the counts in the deepest multiband surveys available. The statistical estimates depend on several observational parameters (e.g. seeing, signal-to-noise ratio), and include a sophisticated treatment of detection completeness. The JAVAcalculator is freely available1 and offers the user the option to adopt our consistent framework or a different scheme. We also provide a summary table of statistical measures in the different bands for a range of different fields of view.
Reliable estimation of the background counts has profound consequences in many areas of observational astronomy. We provide two such examples. One is from a recent study of the Sculptor galaxy NGC 300 where stellar photometry has been used to demonstrate that the outer disc extends to 10 effective radii, far beyond what was thought possible for a normal low-luminosity spiral. We confirm this finding by a re-analysis of the background counts. Secondly, we determine the luminosity function of the galaxy cluster Abell 2734, both through spectroscopically determined cluster membership, and through statistical subtraction of the background galaxies using the calculator and offset fields. We demonstrate very good agreement, suggesting that expensive spectroscopic follow-up, or off-source observations, may often be bypassed via determination of the galaxy background with GALAXYCOUNT.