A new method for galaxy cluster detection. I. The algorithm

Numerous methods for finding clusters at moderate to high redshifts have been proposed in recent years, at wavelengths ranging from radio to X-rays. In this paper we describe a new method for detecting clusters in two-band optical/near-IR imaging data. The method relies upon the observation that all rich clusters, at all redshifts observed so far, appear to have a red sequence of early-type galaxies. The emerging picture is that all rich clusters contain a core population of passively evolving elliptical galaxies that are coeval and formed at high redshifts, The proposed search method exploits this strong empirical fact by using the red sequence as a direct indicator of overdensity. The fundamental advantage of this approach is that with appropriate filters, cluster elliptical galaxies at a given redshift are redder than all normal galaxies at lower redshifts. A simple color cut thus virtually eliminates all foreground contamination, even at significant redshifts. In this paper, one of a series of two, we describe the underlying assumptions and basic techniques of the method in detail and contrast the method with those used by other authors, We provide a brief demonstration of the effectiveness of the technique using a real photometric sample with redshift data, and from this we conclude that the method offers a powerful yet simple way of identify galaxy clusters. We find that the method can reliably detect structures to masses as small as groups with velocity dispersions of only similar to 300 km s(-1), with redshifts for all detected structures estimated to an accuracy of similar to 10%.