Cosmological filaments and minivoids: the origin of intergalactic absorption

Soon after the first QSO was identified, Gunn & Peterson searched for the expected characteristic absorption trough on the blueward side of Lya in the spectrum of the QSO due to an Intergalactic Medium (IGM). They failed to find it, placing a constraint on the density of neutral hydrogen in the IGM that was less than 1 part in 10^5 of that residing in galaxies, and concluded that the IGM must be highly ionized. Soon afterwards absorption by Lya was detected in the IGM; not by a diffuse component, but by a clumpy component of intergalactic gas clouds, the Lya forest. The goal then became to search for `excess' absorption beyond that expected from the forest. No clear absorption by a diffuse HI component, however, was ever detected. The results of recent numerical hydrodynamics computations of the formation of the Lya forest appear to indicate that the division between a clumpy component and a diffuse one may be inappropriate. We find that in a CDM-dominated cosmology, no substantial diffuse medium should be expected. Instead, the medium condenses into a network of complex structures that reveal themselves as discrete absorption systems in the spectra of QSOs. The lowest column density lines arise from the fine structure in minivoids -- small regions with densities below the cosmic mean. These results suggest that the long sought for diffuse HI Gunn-Peterson effect may not exist