THE REDSHIFT DISTRIBUTION OF LY-ALPHA CLOUDS AND THE PROXIMITY EFFECT

The results of a systematic analysis of the redshift distribution of Ly-alpha forest clouds are presented. We form a Ly-alpha line sample from 38 QSO spectra which exhibit neither broad absorption lines systems nor the "associated" z(abs) approximately z(em) metal line systems. The absorption line sample includes only Ly-alpha lines which occur between Ly-beta and Ly-alpha emission, which have rest frame equivalent width greater than 0.36 angstrom, and which not associated with any known metal line systems. The sample contains 950 lines, the largest by far in this kind of study. We confirm that there is overwhelming evidence for the evolution in number density of Ly-alpha forest lines: for the functional form dN/dz proportional (1 + z)gamma, we find gamma = 2.37 + 0.26. Using two mutually independent samples we confirm the existence of an inverse effect or proximity effect in that the Ly-alpha line number density tends to be systematically lower near the QSO emission redshifts than elsewhere. We find no positive evidence that the proximity effect is correlated significantly with the QSO luminosities, contrary to previous results. Explanations are sought to reconcile this paradox. We also find that the proximity effect is uncorrelated with the radio properties of QSOs. We show that the data are entirely consistent with a model in which the inverse effect is due to the enhanced ionization of the Ly-alpha clouds by the nearby QSOs. Assuming the model is correct, the background mean intensity, J(upsilon), is found to be log J(upsilon) congruent-to 21.0, where J(upsilon) is the Lyman limit intensity in unit ergs cm-2 s-1 HZ-1 sr-1. This value of J(upsilon) is consistent with a previous result from a similar study (BDO). When the proximity effect is taken into consideration, the redshift index is found to be gamma = 2.75 +/- 0.29. This value should be considered as a more realistic estimate for the redshift evolution of Ly-alpha clouds than the previously quoted value of gamma = 2.37. In addition, we consider two other alternative explanations for the inverse effect, i.e., (1) the Ly-alpha clouds may not be big enough to cover the entire broad emission line regions, or (2) the functional form, (1 + z)gamma, may be inadequate to describe the redshift distribution of the Ly-alpha clouds, thus causing the inverse effect. We find no evidence in the data, however, that support either of them. Further tests are suggested.