Red sequence technique for finding clusters of galaxies

The red-sequence technique is described in detail in Gladders & Yee 2000. It uses the fact that all galaxy clusters, however they were discovered, possess a population of galaxies which exhibit a tight relationship in colour--magnitude space, the red-sequence.

The figure to the right shows an example red-sequence from a composite of low-redshift clusters. The colour of the red-sequence is dependent on the cluster redshift. This means that the colour can be used to estimate the redshift of the cluster. The typical accuracy obtained in the Red-sequence Cluster Surveys using R-z colour is ~0.05.

The red-sequence method involves constructing many colour slices from the survey data and searching for overdensities of galaxies in these slices. Once significant overdensities are found, the slice containing the peak signal for the overdensity gives the cluster candidate's most probable redshift. To illustrate this, the figure to the left shows two panels showing galaxies in a simulated field containing a high redshift cluster. The leftmost panel shows all galaxies in a magnitude selected sample, the right panel shows the same data after filtering the data using a colour-slice at the redshift of the cluster. The galaxy overdensity due to the cluster becomes obvious only after this colour-filtering is applied. This illustrates the power of the technique compared with traditional overdensity selection, which could detect only much richer systems, and be much more sensitive to unrelated systems projected along the same line of sight.

The image to the right is a significance map of a z~1 colour slice showing a spectroscopically confirmed z=0.95 cluster described in Barrientos et al. 2004. This animated gif is a fly-through of this region, covering all the redshift slices used in the survey. Note that the redshift slices are interpolated over 10 times to make a smoother movie, and that the real separation of the slices are coarser, set by the accuracy of the colour measurements as described in Gladders & Yee 2000.

White peaks represent significant detections in the survey (above ~3-sigma). The peak near the centre of the image is above 4-sigma. The blank patch in the lower left corner is due to a non-functioning chip on CTIO-Mosaic, when this area was observed. The image is ~20 arcmins across, or ~10 Mpc at the redshift of the cluster.