A HIGH-REDSHIFT EXTENSION OF THE SURVEY FOR C-IV ABSORPTION IN THE SPECTRA OF QSOS - THE REDSHIFT EVOLUTION OF THE HEAVY ELEMENT ABSORBERS

We have obtained spectra with a resolution of 60 km s-1 of 11 QSOs with 3.045 ≤ zem ≤ 4.104. In general, the spectra cover the wavelength range between the C IV λ1549 and Lyα emission lines. The QSOs were chosen to supplement the large survey for C IV absorption in the spectra of high-redshift QSOs recently published by Sargent, Boksenberg, and Steidel. The combined sample of C IV λλ1548, 1550 doublets now contains 275 redshifts, spanning the redshift range 1.3 ≤ zabs ≤ 4.0. The new data considerably improve the statistics of the C IV absorbers for zabs > 2.5. Unbiased subsamples of C IV redshifts have been isolated in order to investigate the evolution of the properties and distribution of the absorbers with redshift, with the following principal results. The number of C IV absorption systems per unit redshift range N(z) decreases with increasing redshift in the range 1.3 ≤ zabs ≤ 3.7 in a manner which is inconsistent with a constant comoving density of absorbers. This inconsistency is at the 3.1 σ level for q0 = 1/2 and at the 4.0 σ level for q0 = 0. Thus, the properties of the absorbers are almost certainly evolving with time. The N(z) curve exhibits a gradual decline for z > 2, as opposed to the rapid fall-off suggested by earlier results. The peak in the C IV two-point correlation function on velocity scales 200 ≤ Δv ≤ 600 km s-1 appears to have the same power in high-redshift (zabs > 2.2) and low-redshift (zabs < 2.2) subsamples. There is marginal evidence that the width of the correlation peak on these small velocity scales is smaller in the high-redshift subsample. A 2 σ excess of power on velocity scales 1000 ≤ Δv ≤ 10,000 km s-1 exists in the low-redshift subsample, but not in the high-redshift subsample. Arguments are presented which appear to rule out a significant evolution of the gas velocity dispersion parameter b with redshift in the clouds giving rise to the C IV absorption, and therefore the evolution of the number of observed C IV systems is interpreted as a systematic change in the mean C IV column density of the clouds. A changing abundance of C, and not a systematically changing ionization state in the gas, is probably dominating the observed evolution. The evolution of the mean C IV doublet ratio and a changing distribution of equivalent width with redshift indicate that the typical C IV column density, and by extension the typical abundance of C, is increasing by a factor of ∼ 3 over the redshift range 3 ≥ z ≥ 1.5, corresponding to lookback times of 12.4h50-1 to 9.8h50-1 Gyr (q0=1/2) and an e-folding time for chemical enrichment of between 1 and 2 Gyr. An attempt is made to relate the chemical evolution of the absorbers to early enrichment in galactic halos.