The magnitude-size relation of galaxies out to z∼11,2

As part of the Deep Extragalactic Evolutionary Probe (DEEP) survey, a sample of 190 held galaxies (I-814 less than or equal to 23.5) in the "Groth Survey Strip," has been used to analyze the magnitude-size relation over the range 0.1 < z < 1.1. The survey is statistically complete to this magnitude limit. All galaxies have photometric structural parameters, including bulge fractions (B/T), from Hubble Space Telescope images, and spectroscopic redshifts from the Keck 'Telescope. The analysis includes a determination of the survey selection function in the magnitude-size plane as a function of redshift, which mainly drops faint galaxies at large distances. Our results suggest that selection effects play a very important role. A first analysis treats disk-dominated galaxies with B/T < 0.5. If selection effects are ignored, the mean disk surface brightness (averaged over all galaxies) increases by similar to 1.3 mag from z = 0.1 to 0.9. However, most of this change is plausibly due to comparing bw-luminosity galaxies in nearby redshift bins to high-luminosity galaxies in distant bins. If this effect is allowed for, no discernible evolution remains in the disk surface brightness of bright (M-B < - 19) disk-dominated galaxies. A second analysis treats all galaxies by substituting half-light radius for disk scale length, with similar conclusions. Indeed, at all redshifts, the bulk of galaxies is consistent with the magnitude-size envelope of local galaxies, i.e., with little or no evolution in surface brightness. In the two highest redshift bins (z > 0.7), a handful of luminous, high surface brightness galaxies appears that occupies a region of the magnitude-size plane rarely populated by local galaxies. Their wide range of colors and bulge fractions points to a variety of possible origins.