LYMAN-ALPHA ABSORPTION-SPECTRUM OF THE PRIMORDIAL INTERGALACTIC MEDIUM

We investigate and simulate Lyalpha absorption by the primordial hydrogen (the IGM) in the cold dark matter cosmology taking into account the Doppler effect of peculiar velocity. The IGM density is found to follow n(x1) is-proportional-to exp [(1/4pix(b)2) integral e(-xb/x12)(1/x12)delta(DM)(x2)dx2] with delta(DM) being the CDM density contrast and x(b) being the Jeans length of the IGM. Four IGM models are simulated according to their mean density OMEGA(H), temperature T, and clumping factor xi(IGM)(0) at the redshift range 2.5-3.0. For the first model (the reference model), we take OMEGA(H) = 0.015 h-2, T = 2 x 10(4) K, and xi(IGM) = 1.72. Random fluctuation clumps in the medium produce a forest-like Lyalpha absorption spectrum compatible to those observed by fitting the probability distribution function of line equivalent widths that have a mean slighty larger than 22 km s-1, and the two-point correlation function of the lines. Models 2 and 3 have either an increased clumping factor xi(IGM) = 8102 or a decreased denSitY OMEGA(H) = 0.0015 h-2 both of which might resemble the IGM at lower redshifts. Forest-like absorptions appear in these models too, but the total number of W greater-than-or-equal-to 0.16 A lines is decreased to 30% or 21% of model 1. Model 4 uses T = 8000 K and then produces a very noisy spectrum like those observed at higher redshifts. In order to get a reasonable Gunn-Peterson effect, we have assumed that an external UV radiation of J(c)(13.6 eV) = 3.2 x 10(-21) h-1 ergs-1 cm-2 Hz-1 (or 3.2 x 10(-22) h-1 ergs-1 CM-2 Hz-1 for model 3) is responsible for the ionization. Parallel nearby Lyalpha absorption spectra, like those of two images of a gravitationally lensed quasar, is then simulated under model 1. If the separation of the spectra is less than 20 h-1 kpc, they will look very similar to each other. Spectra having separations greater-than-or-equal-to 200 h-1 kpc do not show apparent similarity, but there are still some dependences left between the forests. A time series technique (the auto- and cross-correlation coefficients) is then suggested to discriminate different IGM or Lyalpha cloud models and to measure the cross-dependence quantitatively. Finally, we discuss that some fundamental physical processes of the IGM neglected in the present study should follow the same dynamical behavior as in traditional cloud models; also, random clumps in the diffuse IGM can be progenitors of bound Lyalpha clouds formed later. The present models improve the cloud models in the sense that the linear evolved primordial IGM can also produce a forest-like Lyalpha absorption at high redshift, so it avoids making another assumption about the origin of clouds and thus is a simple but natural interpretation of Lyalpha forests at z greater-than-or-equal-to 3 when structures were not yet formed in general.