2001ApJS..132...19S

We use rest-frame Strömgren photometry to observe clusters of galaxies in a self-consistent manner from z=0 to z=0.8. Strömgren photometry of galaxies is intended as a compromise between standard broadband photometry and spectroscopy, in the sense that it is more sensitive to subtle variations in spectral energy distributions than the former, yet much less time-consuming than the latter. principal component analysis is used to facilitate extraction of information from the Strömgren data. By calibrating the principal components using well-studied galaxies, as well as models of stellar populations, we develop a purely empirical method to detect, and subsequently classify, cluster galaxies at all redshifts smaller than 0.8. Interlopers are discarded with unprecedented efficiency (up to 100%). The first principal component essentially reproduces the Hubble sequence and can thus be used to determine the global star formation history of cluster members. The (PC2, PC3) plane allows us to identify Seyfert galaxies (and distinguish them from starbursts) based on photometric colors alone. In the case of E/S0 galaxies with known redshift, we are able to resolve the age-dust-metallicity degeneracy, albeit at the accuracy limit of our present observations. We use this technique in later papers to probe galaxy clusters well beyond their cores and to fainter magnitudes than spectroscopy can achieve, because the faint end of the luminosity function as well as the outer cluster regions seem to exhibit the strongest evolutionary trends. We are able to directly compare these data over the entire redshift range without a priori assumptions because our observations do not require first-order k-corrections. The compilation of such data for different cluster types over a wide redshift range is likely to set important constraints on the evolution of galaxies and on the clustering process.