The CNOC2 field galaxy luminosity function. I. A description of luminosityfunction evolution

We examine the evolution of the galaxy luminosity function (LF) using a sam ple of over 2000 galaxies, with 0.12 < z < 0.55 and 17.0 < R-c < 21.5, draw n from the Canadian Network for Observational Cosmology Field Galaxy Redshi ft Survey (CNOC2), at present the largest such sample at intermediate redsh ifts. We use UBVR,I, photometry and the spectral energy distributions (SEDs ) of Coleman, Wu, and Weedman to classify our galaxies into early, intermed iate, and late types, for which we compute LFs in the rest-frame B, R-c, an d U bandpasses. In particular, we adopt a convenient parameterization of LF evolution including luminosity and number density evolution and take care to quantify correlations among our LF evolution parameters. We also careful ly measure and account for sample selection effects as functions of galaxy magnitude and color. Our principal result is a clear quantitative separation of luminosity and d ensity evolution for different galaxy populations and the finding that the character of the LF evolution is strongly dependent on galaxy type. Specifi cally, we find that the early- and intermediate-type LFs show primarily bri ghtening at higher redshifts and only modest density evolution, whereas the late-type LF is best fit by strong number density increases at higher z wi th little luminosity evolution. We also confirm the trend seen in previous smaller z less than or similar to 1 samples of the contrast between the str ongly increasing luminosity density of late-type galaxies and the relativel y constant luminosity density of early-type objects. Specific comparisons a gainst the Canada-France and Autofib redshift surveys show general agreemen t among our LF evolution results, although there remain some detailed discr epancies. In addition, we use our number count and color distribution data to further confirm the validity of our LF evolution models to z similar to 0.75, and we also show that our results are not significantly affected by p otential systematic effects such as surface brightness selection, photometr ic errors, or redshift incompleteness.