|Annu. Rev. Astron. Astrophys. 1998. 36:
Copyright © 1998 by . All rights reserved
4.3. Gravitational Confinement: Clouds Dominated by Hot Dark Matter Gravity
The advent of ab initio theories of gravitational structure formation made it possible to place the Ly forest in a larger frame, and investigate its relation to galaxy formation. In principle, the number density, sizes and physical parameters of the absorbers can be predicted as a consequence of cosmological models, though, until recently this has been wishful thinking.
ZELDOVICH PANCAKES The first such theory to explicitly address the Ly forest phenomenon was the hot dark matter (HDM) model. The formation of adiabatic pancakes suggested by Zeldovich (1970) is expected to produce a primordial gas phase with the right properties for detection in HI absorption (Doroshkevich & Shandarin 1977). Among the interesting consequences of this theory are the large (in fact: too large) sizes of the pancakes. Even after fragmentation, coherent absorption should be extending over Mpcs across the sky (Doroshkevich & Muecket 1985) thus explaining the large transverse sizes seen later in QSO pair studies. In a prescient paper, Oort (1981), at the time referring to HDM pancakes, suggested identifying Ly absorbers with collapsed but uncondensed gas in the "superclusters" of galaxies known from low redshift. He calculated the gravitational scale height of a sheet of gas, and noted the similarity of the mean free path between Ly clouds and the distances between superclusters. The very large transverse sizes expected (~ 20 Mpc, at z ~ 2) were not confirmed, however, by the QSO pair study of Sargent et al (1982), so such objects cannot be the rule.
The underlying HDM structure formation scenario has become somewhat unpopular, but the physical idea of Ly absorbers as flattened pancakes survives into the currently favored, CDM based picture (see below). However, the CDM pancakes (or sheets) are more than an order of magnitude smaller than the Zeldovich ones, and they form late, after denser structures like knots and filaments are already in place (Cen et al 1994; Bond et al 1996).