The luminosity function of the Virgo Cluster from MB=-22 to −11

Abstract

We measure the galaxy luminosity function (LF) for the Virgo Cluster between blue magnitudes M B = -22 and -11 from wide-field charge-coupled device (CCD) imaging data. The LF is only gradually rising for -22 < M B < -16.Between M B = -16 and - 14 it rises steeply, with a logarithmic slope of α∼ - 1.6. Fainter than M B = -14, the LF flattens again. This LF is shallower (although turning up at brighter absolute magnitudes) than the R-band LF measured recently by Phillipps et al., who found α ∼ -2.2 to be fainter than M R = -13. It is similar, however, to the LF determined from the Virgo Cluster Catalog by Sandage et al. A few faint galaxies are found that Sandage et al. missed because their surface brightness threshold for detection was too high, but these do not dominate the luminosity function at any magnitude. Most of the faint galaxies we find are dwarf elliptical, alternatively called dwarf spheroidal, galaxies. The most important potential source of systematic error is that we may have rejected some high surface brightness galaxies from the cluster sample because we think that they are background galaxies. This is quite different from what has conventionally been regarded as the most serious source of systematic error in this kind of study: that we are missing many low surface brightness galaxies because they are never visible above the sky. There are ∼ 2.5 times more dwarfs per giant galaxy in Virgo than in the Ursa Major Cluster, a diffuse group of ∼ 80 spiral galaxies at the same distance as Virgo, or the Local Group. The Virgo and Ursa Major Cluster LFs are inconsistent with each other at a high level of significance. These results add weight to the hypothesis that is developing that dwarf galaxies are more common relative to giant galaxies in dense environments than diffuse ones. Both LFs are highly inconsistent with cold dark matter theory, which has been so successful at reproducing observations on large scales. Possible theoretical explanations for this discrepancy, and for the detailed shape of the Virgo Cluster LF, are investigated.