8.2 QSO-Absorption Line Systems
Spectra of high redshift QSOs display absorption lines from intervening HI clouds, arising when the QSO continuum is absorbed at the local rest wavelength of the resonant Ly line. The QSO absorption systems come with a range of column densities from below 1014 cm-2, the Ly forest, to more than 1020 cm-2, the Damped Ly systems (DLAs).
Metal absorption lines associated with QSO absorption line systems can be used to study the metallicity and its temporal evolution in interstellar gas. This field of research has experienced rapid progress recently thanks to the new generation of 10m class telescopes. Now even low density Ly forest systems can be studied (but not those with too low NH) and have been found to contain heavy elements, on average [C/H] ~ -2.5, although the metallicities are still very uncertain due to the limited understanding of the physical condition in these systems (Pettini 1999).
The DLAs are better understood in terms of physics though their true nature is still not clear, and are on average more metal rich. At redshifts above z = 1.5 Pettini et al. (1997) and Prochaska and Wolfe (1999) determined an average zinc (where problems with dust depletion are expected to be small) abundance of [Zn/H] = -1.15. There is a large scatter at all redshifts, and no strong trend in metallicity with redshift (but see also Lu et al. 1996). It is generally assumed that DLAs are associated with galaxies at an early stage of evolution. It is not clear whether all remote galaxies at redshift larger than 2 are chemically young systems or whether damped Ly lines preferentially pick out unevolved systems.
Pettini et al. (1995) explore the possibility to measure the N/O ratio at a metallicity lower than those of the most metal-poor dwarf galaxies known. They find that relative to the Sun, N is more underabundant than O by at least a factor of 15 in agreement with Lu et al. (1996, 1998). Izotov and Thuan (1999) attribute this low N/O to uncertainties related to unknown physical conditions in the interstellar medium of high-redshift galaxies that make the correction factors for unseen low-ionisation species difficult to assess. On the other hand Pilyugin (1999) manages to account for the discrepant behaviour of DLAs and low metallicity BCGs, if assuming (among other things) that DLAs have had their last star formation event less than 1 Gyr ago, whereas the observed metals in the most metal-poor BCGs have been produced in a previous event, more than 1 Gyr ago.
Vladilo (1998) proposes that the abundance pattern in DLAs is more consistent with them being dwarf galaxies rather than disc galaxies. By dwarf galaxy, an LMC like galaxy was considered, i.e. much more massive than the most metal poor dwarfs which have narrow H I profiles, but consistent with the most luminous BCGs and dIs. However it is not clear whether this agrees with the observed kinematics of DLAs (Prochaska and Wolfe 1997). Observations of galaxies associated with lower redshift DLAs present a wide variety of morphologies (Le Brun et al. 1997). Thus it may be the case that DLAs are not linked with a specific kind of galaxy, but with all sorts of galaxies.
Although the nature of DLA galaxies and their relation to galaxies seen at low redshift is far from clear (Pettini 1999), they are among the most metal poor galaxies known. With a dozen 10m class telescopes available in the beginning of the next decade, there is good hope that our understanding of these systems will significantly improve.