F. H. Briggs, PASA, 14 (1), 31.
Next Section: Where to find ``HI Title/Abstract Page: Local HI: Constraints on Previous Section: Evolution of the HI | Contents Page: Volume 14, Number 1 |
Distribution of Column Densities
Another place where future surveys can play an important role will be in exploring lower HI column densities than have been observed in the 21 cm line in the past. QSO absorption line statistics over a wide range of Lyman- line strengths specify that the incidence of absorption becomes increasingly prevalent toward lower column densities, so that along a randomly chosen sight line, the probability of interception rises by roughly a factor of 10 for every decrease by a factor of 100 in . This behavior is quantified by the distribution, defined so that is the number of lines detected within a range centered on over a ``normalized absorption distance'' dX, where is the normalized absorption distance from zero redshift to z for . The function is roughly proportional to over the range to cm, although there is evidence for subtle structure possibly related to opacity in the Lyman continuum that occurs for layers with cm (Petitjean et al 1993). At high redshift, the frequency of absorption for these optically thick absorbers is per unit X (Petitjean et al 1993); for cm, the frequency is roughly halved. The derivation of the at these is especially uncertain, since the entire Lyman series is heavily saturated in the regime where the Lyman continuum is optically thick, and the damping wings, which permit an unambiguous measure of column density, do not become readily observable until is well in excess of 10 cm. Thus, column density measurements in this regime are very uncertain, leading Petitjean et al to plot only one point on their diagram for to cm.
Recent large HI surveys with filled aperture telescopes are routinely capable of detecting column densities below cm (Schneider 1996, Zwaan et al 1996, Briggs et al 1996), provided the emission fills the telescope beam. The Arecibo survey by Sorar (1994; see also Briggs & Sorar 1996) was optimized to be sensitive to cm (), and the survey observed over 5000 independent beam areas to a depth of 7500 kmá s, covering a total absorption path . The Arecibo beam subtends á kpc at 3á Mpc and 70á kpc at 75á Mpc, which is a reasonable match to the cloud sizes deduced for the Lyman- forest (cf. Smette et al 1995b, 1992). To date, only one of the 61 detections (Zwaan et al 1996) has not been identified with a high column density layer of the type associated with the neutral intergalactic medium of a galaxy, implying that a separate population of low column density objects can add only a small fraction of the current HI content already identified with galaxies.
The high redshift distribution would imply of order 50 interceptions in the range to cm for the pathlength X explored by the Arecibo survey. Where are they? At least a part of the discrepancy is likely due to evolution of Lyman- forest cloud population. An additional observational problem is that the high column density end of the forest cloud distribution (around 10 cm and above) has associated metal lines, such as CIV and MgII, which has historically caused then to be identified with hypothetical galaxy halos; single dish observations seldom have the resolution to reliably separate the HI signal from a halo of a spiral galaxy from the bright signal originating in the main body of the galaxy, unless there are strong kinematic effects that create a difference in gas velocity as a function of radius and the halo gas is very extended.
Further considerations in the study of this intermediate column density range are the theoretical models that consider ionization of extended gas around galaxies by the extragalactic ionizing background (Sunyaev 1969, Corbelli et al 1989, Maloney 1993, Charlton et al 1994). Many of these models predict a strong dip in between 10 and 10 cm. If similar arguments apply to a population of intergalactic clouds or super-LSB galaxies, then an interesting experiment now coming into the realm of possibility will be to push the sensitivity limits of the local HI emission observations down to cm where ionized layers of high column density might be detected.
Next Section: Where to find ``HI Title/Abstract Page: Local HI: Constraints on Previous Section: Evolution of the HI | Contents Page: Volume 14, Number 1 |
á© Copyright Astronomical Society of Australia 1997