THE LUMINOSITY FUNCTION OF FIELD GALAXIES IN THE CNOC1 REDSHIFT SURVEY

We have computed the luminosity function for a sample of 389 held gala xies from the Canadian Network for Observational Cosmology cluster red shift survey (CNOCl) over the redshift range z = 0.2-0.6. We find Sche chter parameters M(r) - 5 log h = -20.8 +/- 0.4 and alpha = -1.3 +/- 0.2 in rest-frame Gunn r, and M(BAB) - 5 log h = -19.6+/-0.3 and alph a = -0.9 +/- 0.2 in rest-frame B-AB. We have also split our sample at the color of a redshifted but nonevolving Sbc galaxy and find distinct ly different luminosity functions for red and blue galaxies. Red galax ies have a shallow slope alpha approximate to -0.4 and dominate the br ight end of the luminosity function, while blue galaxies have a steep alpha approximate to -1.4 and prevail at the faint end. Comparisons of the CNOCl results to analogous intermediate-redshift luminosity funct ions from the Canada-France (CFRS) and Autofib redshift surveys show b road agreement among these independent samples, but there are also sig nificant differences which will require larger samples to resolve. Als o, in CNOCl the red galaxy luminosity density stays about the same ove r the range z = 0.2-0.6, while the blue galaxy luminosity density incr eases steadily with redshift. These results are consistent with the tr end of the luminosity density versus redshift relations seen in the CF RS, although the normalizations of the luminosity densities appear to differ for blue galaxies. Comparison to the local luminosity function from the Las Campanas redshift survey (LCRS) shows that the luminosity density at z approximate to 0.1 is only about half that seen at z app roximate to 0.4. A change in the luminosity function shape, particular ly at the faint end, appears to be required to match the CNOCl and LCR S luminosity functions, if galaxy evolution is the sole cause of the d ifferences seen. However, it should be noted that the specific details of the construction of different surveys may complicate the compariso n of results and so may need to be considered carefully.