Differential galaxy evolution in cluster and field galaxies at z ≈ 0.3

We measure spectral indexes for 1823 galaxies in the Canadian Network for Observational Cosmology 1 (CNOC1) sample of 15 X-ray luminous clusters at 0.18 < z < 0.55 to investigate the mechanisms responsible for differential evolution between galaxy cluster and field environments. The radial trends of D4000, W-0(H delta), and W-0(O II) are all consistent with an age sequence, in the sense that the last episode of star formation occurred more recently in galaxies farthest from the cluster center. Throughout the cluster environment, galaxies show evidence for older stellar populations than field galaxies; they have weaker W-0(O II) and W-0(H delta) lines and stronger D4000 indexes. From our primary sample of 1413 galaxies, statistically corrected for incompleteness and selection effects, we identify a sample of K+A galaxies, which have strong H delta absorption lines [W-0(H delta) > 5 Angstrom] but no [O II] emission [W-0(O II) < 5 Angstrom], perhaps indicative of recently terminated star formation. The observed fraction of 4.4% +/- 0.7% in the cluster sample is an overestimate due to a systematic effect that results from the large uncertainties on individual spectral index measurements. Corrected for this bias, we estimate that K+A galaxies make up only 2.1% +/- 0.7% of the cluster sample and 0.1% +/- 0.7% of the field. From the subsample of galaxies more luminous than M-r = -18.8 + 5 log h, which is statistically representative of a complete sample to this limit, the corrected fraction of K+A galaxies is 1.5% +/- 0.8% in the cluster and 1.2% +/- 0.8% in the held. Compared with the z approximate to 0.1 fraction of 0.30%, the fraction of K+A galaxies in the CNOC1 field sample is greater by perhaps a factor of 4, but with only 1 sigma significance; no further evolution of this fraction is detectable over our redshift range. We compare our data with the results of PEGASE and GISSEL96 spectrophotometric models and conclude, from the relative fractions of red and blue galaxies with no [O II] lambda 3727 emission and strong H delta absorption, that up to 1.9% +/- 0.8% of the cluster population may have had its star formation recently truncated without a starburst. However, this is still not significantly greater than the fraction of such galaxies in the field, 3.1% +/- 1.0%. Furthermore, we do not detect an excess of cluster galaxies that have unambiguously undergone starbursts within the last 1 Gyr. In fact, at 6.3% +/- 2.1%, the A+em galaxies that Poggianti et al. have recently suggested are dusty starbursts are twice as common in the held as in the cluster environment. Our results imply that these cluster environments are not responsible for inducing starbursts; thus, the increase in cluster blue galaxy fraction with redshift may not be a strictly cluster-specific phenomenon. We suggest that the truncation of star formation in clusters may largely be a gradual process, perhaps due to the exhaustion of gas in the galactic disks over fairly long timescales; in this case differential evolution may result because field galaxies can refuel their disks with gas from extended halos, thus regenerating star formation, while cluster galaxies may not have such halos and so continue to evolve passively.