Annu. Rev. Astron. Astrophys. 1998. 36: 267-316 Copyright © 1998 by Annual Reviews. All rights reserved

2. OBSERVATIONAL APPROACHES

2.1. Technical Possibilities, Observational Constraints

Observational progress with traditional Ly studies can conveniently be charted in terms of two limiting factors: The spectral resolution, and the signal-to-noise ratio. The early (photographic) spectroscopy in the 1960s typically had to rely on resolutions of 10 - 20 Å. At optical wavelengths this is barely sufficient to resolve velocity dispersions characteristic of galaxy clusters. Later, in particular the combination of echelle spectrographs and CCD detectors permitted QSO spectroscopy to be done with 4m telescopes at resolutions as high as R ~ 5 × 10⁴. QSO observers have been quick to make use of these advances as is obvious from the semantic drift of "high resolution": in 1979 it meant 0.8 Å (Sargent et al. 1979), in 1984, 0.25 Å (Carswell et al. 1984), and in 1990, 0.08 Å (Pettini et al. 1990). Naturally, the average signal-to-noise ratio per resolution element did not benefit from the increase in spectral dispersion, and a single high resolution QSO spectrum required an embarassingly large number of nights on a 4m telescope. Thus various ways of extracting information from the Ly forest have been developed in parallel. Some were tailored to detailed analysis of individual, expensive, high resolution spectra (line profile fitting), while others could be applied more automatically to larger samples of low resolution data (mean absorption; equivalent width measurements). The 10m Keck telescope with its powerful High Resolution Spectrograph (HIRES; Vogt et al. 1994) brought signal-to-noise ratios of ~ 100 within reach of high resolution (FWHM ~ 7 kms^-1) absorption line studies, rendering some of the low resolution approaches obsolete. The new limiting factor for large telescope spectroscopy is not resolution nor collecting area but manpower - coping (intelligently) with the continuous stream of large datasets already, or soon, available from the Keck telescopes, the Magellan, the Hobby-Eberly, the ESO-VLT, the MMT, etc. Recently, progress has also come from extending the wavelength regime into the UV band with the Hubble Space Telescope (HST) and the Hopkins Ultraviolet Telescope (HUT). In particular, the advent of the HST with its high resolution UV spectrographs has helped to compensate to some extent for the fact that optical Ly spectroscopy can only sample the universe at redshifts larger than ~ 2.5. We can now study the absorber properties and the absorber-galaxy connection in the local universe, and, at high redshift, the far-UV (rest frame) helium Ly forest.