Evidence for Inconsistencies in Galaxy Luminosity Functions Defined by Spectral Type

Abstract

Galaxy morphological and spectroscopic types should be nearly independent of apparent magnitude in a local, magnitude-limited sample. Recent luminosity function surveys based on morphological classification of galaxies are substantially more successful at passing this test than surveys based on spectroscopic classifications. Among spectroscopic classifiers, those defined by small aperture fibers (ESP, LCRS) show far stronger systematic classification biases than those defined by large apertures (APM). This effect can be easily explained as an aperture bias, whereby galaxies with globally late-type spectra are assigned earlier spectral types which depend on the redshift and luminosity of the galaxy. The effect is demonstrated by extracting successively larger aperture spectra from long-slit spectroscopy of nearby galaxies. This systematic classification bias is generic, and the 2dFGRS and SDSS surveys will show similar systematic biases in their spectral types. We suggest several methods for correcting the problem or avoiding it altogether. If not corrected, this aperture bias can mimic galaxy evolutionary effects and distort estimates of the luminosity function.