A NEW METHOD FOR MEASURING THE H-I-GUNN-PETERSON EFFECT AT HIGH-REDSHIFT

As part of an undergoing program to carry out a systematic and thorough measurement of the neutral hydrogen Gunn-Peterson effect at high redshift, a quantitatively testable and repeatable procedure-in particular, a robust statistical weighting technique-is developed. It is applied to an echelle spectrum of resolution similar to 15 km s(-1) of the quasar PKS 1937 - 101 with z(em) = 3.787, to demonstrate its capability. Based on good low-resolution spectrophotometric data, the continuum is extrapolated from redward of the Ly alpha emission line in an objective way so that the systematic fractional deviation is minimized and estimated to be less than +/-1%. A weighted intensity distribution which is derived overwhelmingly from pixels close to the continuum level in the Ly alpha forest region is constructed by the evaluation of how closely correlated each pixel is with its neighboring pixels. The merit of such a weighted distribution is its stronger and narrower peak compared to the unweighted distribution, as well as its smaller dependence on the uncertainty of strong absorption lines and noise spikes. It is more robust in locating the continuum level, especially at higher redshift when the unweighted intensity distribution becomes flat due to a rapidly increasing number of Ly alpha absorbers with redshift. By comparison to the weighted intensity distribution of synthetic Ly alpha forest spectra with various chosen diffuse neutral hydrogen opacities tau(GP), a chi(2) fit is performed between the observed and model distributions. In addition to a weak-line population with power-law column density N-H distribution with beta = 1.7 extrapolated down to 10(12) cm(-2) a best chi(2) fit requires a component of tau(GP) = 0.115 +/- 0.025 at [z(abs)] approximate to 3.4, with estimation of the contribution from the variance of the parameter beta. However, although no evidence of more than 1%-2% error is seen in the continuum extrapolation, the uncertainty attributed to the possible systematic overestimation due to such an extrapolated continuum slope can be as high as the level of the accuracy of the chi(2) fit, which is investigated by splitting the Ly alpha forest region into subsamples to check the continuum drop's dependence on absorber redshift. Quasars at high redshift with narrow-winged emission lines are needed to improve the extrapolation and clarify any such systematic effect. Possible contributions to the continuum drop from unresolved weak lines are discussed, and it is shown that in some cases the weighted intensity distribution has the potential to distinguish the continuum drop due to tau(GP) from that due to weak lines. The future prospects of applying this technique on a larger data set and higher redshift quasars are discussed.