The Sloan Digital Sky Survey ( SDSS ) and photometric/spectroscopic surveys in the GOODS-South field ( the Chandra Deep Field-South , CDF-S ) are used to construct volume-limited , stellar mass-selected samples of galaxies at redshifts 0 < z < 1 .
The CDF-S sample at 0.6 < z < 1.0 contains 207 galaxies complete down to M = 4 \times 10 ^ { 10 } ~ { } M _ { \odot } ( for a “ diet ” Salpeter initial mass function ) , corresponding to a luminosity limit for red galaxies of M _ { B } = -20.1 .
The SDSS sample at 0.020 < z < 0.045 contains 2003 galaxies down to the same mass limit , which corresponds to M _ { B } = -19.3 for red galaxies .
Morphologies are determined with an automated method , using the Sérsic parameter n and a measure of the residual from the model fits , called “ bumpiness , ” to distinguish different morphologies .
These classifications are verified with visual classifications .
In agreement with previous studies , 65 \% - 70 \% of the galaxies are located on the red sequence , both at z \sim 0.03 and at z \sim 0.8 .
Similarly , 65 \% - 70 \% of the galaxies have n > 2.5 .
The fraction of E+S0 galaxies is 43 \% \pm 3 \% at z \sim 0.03 and 48 \% \pm 7 \% at z \sim 0.8 ; i.e. , it has not changed significantly since z \sim 0.8 .
When combined with recent results for cluster galaxies in the same redshift range , we find that the morphology-density relation for galaxies more massive than 0.5 M ^ { * } has remained constant since at least z \sim 0.8 .
This implies that galaxies evolve in mass , morphology , and density such that the morphology-density relation does not change .
In particular , the decline of star formation activity and the accompanying increase in the stellar mass density of red galaxies since z \sim 1 must happen without large changes in the early-type galaxy fraction in a given environment .