We have obtained deep and wide field imaging of the Coma cluster of galaxies with the CFH12K camera at CFHT in the B, V, R and I filters. In this paper, we present the observations, data reduction, catalogs and first scientific results. We investigated the quality of our data by internal and external literature comparisons. We also checked the realisation of the observational requirements we set. Our observations cover two partially overlapping areas of 42 × 28 arcmin 2 , leading to a total area of 0.72 × 0.82 deg 2 . We have produced catalogs of objects that cover a range of more than 10 mag. and are complete at the 90% level at B ∼ 25, V ∼ 24, R ∼ 24 and I ∼ 23.5 for stellar-like objects, and at B ∼ 22, V ∼ 21, R ∼ 20.75 and I ∼ 20.5 for faint low-surface-brightness galaxy-like objects. Magnitudes are in good agreement with published values from R ∼ 16 to R ∼ 25. The photometric uncertainties are of the order of 0.1 mag at R ∼ 20 and of 0.3 mag at R ∼ 25. Astrometry is accurate to 0.5 arcsec and also in good agreement with published data. Our catalog provides a rich dataset that can be mined for years to come to gain new insights into the formation and evolution of the Coma cluster and its galaxy population. As an illustration of the data quality, we examine the bright part of the Colour Magnitude Relation (B−R versus R) derived from the catalog and find that it is in excellent agreement with that derived for galaxies with redshifts in the Coma cluster, and with previous CMRs estimated in the literature.

Abstract.We have analyzed the galaxy number density and luminosity density profiles of rich clusters of galaxies from redshifts z ∼ 0 to z ∼ 0.5. We show that the luminosity profile computed with bright galaxies (MR < −21) is significantly cusped in the center of the clusters, whatever the redshift. This is in agreement with the dark matter profiles predicted by numerical simulations. The galaxy number density profile for the bright galaxies is fitted equally well with a core model or a cusped model. In contrast, the luminosity and the galaxy number density profiles of the fainter galaxies are significantly better fitted by a core model. We did not detect any statistically significant different fits when applied to data in the range from z ∼ 0 to z ∼ 0.5. The difference in profile between faint and bright galaxies may be due to the rapid (relative to the age of the universe at z = 0 versus z = 0.5) destruction of the faint galaxies by tidal forces and merging events in the denser central regions of the clusters. This process could erase the cusp by turning faint galaxies into diffuse light. In this case, the galaxies (with a cusp visible in the bright galaxy number density and mainly in luminosity profiles) would trace the total mass distribution.