Abstract:
The Dark Energy Survey (DES) is an ongoing 5-year program that covers 5000 deg2 of the sky area in five optical-near infrared photometric bands (grizY), designed to measure the dark energy and dark matter densities and the dark energy equation of state by combining multiple probes. The vast dataset generated by DES contains multi-wavelength information about hundreds of million galaxies in an unprecedented combination of area and depth. As a result, DES is a goldmine for a broad list of science cases far beyond the dark energy studies. We present a study of the galaxy luminosity function (LF) evolution using preliminary data from the three first years of DES observations. Here, we discuss and validate a new methodology to be applied at the completion of the survey and also for upcoming projects, as the Large Synoptic Survey Telescope (LSST). We introduce the probability density function of photometric redshifts calculated with the DNF algorithm directly as additional weight in the traditional 1/Vmax estimator. Absolute magnitudes are recomputed for each redshift slice using template-fitting algorithm LePhare, with a unique combination of synthetic SED templates optimized for the DES dataset. We observed dimming of global characteristic luminosity (measured in terms of absolute magnitude, M) and luminosity density (j), accompanied by a modest increase in the characteristic number density (f) with cosmic time, in agreement with previous results available in the literature. We split the sample into two populations based on morphology and studied their LF evolution separately. We found that bulgedominated systems increased in number density and in their contribution to the total luminosity density with cosmic time. Our results are compatible with the transformation of part of the disk-dominated population into bulge-dominated galaxies. We also observed pieces of evidence of a “downsizing” effect in galaxy assembly from the lack of evolution at the bright-end of bulge-dominated LF, suggesting that the brightest early-type objects are already in place at z 1.