The spectra of several high-redshift (z>6) quasars have shown evidence for a Gunn-Peterson (GP) damping wing, indicating a substantial mean neutral hydrogen fraction (x_HI > 0.03) in the z ~ 6 intergalactic medium (IGM). However, previous analyses as
sumed that the IGM was uniformly ionized outside of the quasars HII region. Here we relax this assumption and model patchy reionization scenarios for a range of IGM and quasar parameters. We quantify the impact of these differences on the inferred x_HI, by fitting the spectra of three quasars: SDSS J1148+5251 (z=6.419), J1030+0524 (z=6.308), and J1623+3112 (z=6.247). We find that the best-fit values of x_HI in the patchy models agree well with the uniform case. More importantly, we confirm that the observed spectra favor the presence of a GP damping wing, with peak likelihoods decreasing by factors of > few - 10 when the spectra are modeled without a damping wing. We also find that the Ly alpha absorption spectra, by themselves, cannot distinguish the damping wing in a relatively neutral IGM from a damping wing in a highly ionized IGM, caused either by an isolated neutral patch, or by a damped Ly alpha absorber (DLA). However, neutral patches in a highly ionized universe (x_HI < 0.01), and DLAs with the large required column densities (N_HI > few x 10^{20} cm^{-2}) are both rare. As a result, when we include reasonable prior probabilities for the line of sight (LOS) to intercept either a neutral patch or a DLA at the required distance of ~ 40-60 comoving Mpc away from the quasar, we find strong lower limits on the neutral fraction in the IGM, x_HI > 0.1 (at 95% confidence). This strengthens earlier claims that a substantial global fraction of hydrogen in the z~6 IGM is in neutral form.
UV radiation from early astrophysical sources could have a large impact on subsequent star formation in nearby protogalaxies. Here we study the radiative feedback from the first, short-lived stars using hydrodynamical simulations with transient UV ba
ckgrounds (UVBs) and persistent Lyman-Werner backgrounds (LWBs) of varying intensity. We extend our prior work in Mesinger et al. (2006), by studying a more typical region whose proto-galaxies form at lower redshifts, z~13-20, in the epoch likely preceding the bulk of reionization. We confirm our previous results that feedback in the relic HII regions resulting from such transient radiation, is itself transient. Feedback effects dwindle away after ~30% of the Hubble time, and the same critical specific intensity of J_UV~0.1 x 10^{-21} ergs/s/cm^2/Hz/sr separates positive and negative feedback regimes. Additionally, we discover a second episode of eventual positive feedback in halos which have not yet collapsed when their progenitor regions were exposed to the transient UVB. This eventual positive feedback appears in all runs, regardless of the strength of the UVB. However, this feedback regime is very sensitive to the presence of Lyman-Werner radiation, and notable effects disappear under fairly modest background intensities of J_LW>10^{-3} x 10^{-21} ergs/s/cm^2/Hz/sr. We conclude that UV radiative feedback in relic HII regions, although a complicated process, seems unlikely to have a major impact on the progress of cosmological reionization, provided that present estimates of the lifetime and luminosity of a PopIII star are accurate. More likely is that the build-up of the LWB ultimately governs the feedback strength until a persistent UV background can be established. [abridged]
The earliest generation of stars and black holes must have established an early Lyman-Werner background (LWB) at high redshift, prior to the epoch of reionization. Because of the long mean free path of photons with energies E<13.6 eV, the LWB was nea
rly uniform. However, some variation in the LWB is expected due to the discrete nature of the sources, and their highly clustered spatial distribution. In this paper, we compute the probability distribution function (PDF) of the LW flux that irradiates dark matter (DM) halos collapsing at high-redshift (z~10). Our model accounts for (i) the clustering of DM halos, (ii) Poisson fluctuations in the number of corresponding star forming galaxies, and (iii) scatter in the LW luminosity produced by halos of a given mass (calibrated using local observations). We find that > 99% of the DM halos are illuminated by a LW flux within a factor of 2 of the global mean value. However, a small fraction, ~1e-8 to 1e-6, of DM halos with virial temperatures above 1e4 K have a close luminous neighbor within < 10 kpc, and are exposed to a LW flux exceeding the global mean by a factor of > 20, or to J_(21,LW)> 1e3 (in units of 1e-21 erg/s/Hz/sr/cm^2). This large LW flux can photo--dissociate H_2 molecules in the gas collapsing due to atomic cooling in these halos, and prevent its further cooling and fragmentation. Such close halo pairs therefore provide possible sites in which primordial gas clouds collapse directly into massive black holes (M_BH~ 1e4 - 1e6 M_sun), and subsequently grow into supermassive (M_BH > 1e9 M_sun) black holes by z~6.
The ionizing ultraviolet background (UVB) during reionization can suppress the gas content of low-mass galaxies, even those capable of efficient atomic cooling, and thus lead to an extended reionization epoch. In this work, we explore the importance
of negative UV radiative feedback on Tvir > 10^4 K halos during the middle and late stages of reionization. We do not try to self-consistently model reionization; instead, we explore a large parameter space in an attempt to draw general, robust conclusions. We do this using a tiered approach. Using 1-D hydrodynamical simulations, we model the collapse of gas onto halos of various masses under UVBs of various intensities. We then generate realistic, parametrized maps of the inhomogeneous UVB, using large-scale semi-numeric simulations. By combining these results, we find that under all reasonably conservative scenarios, UV feedback on atomically-cooled halos is not strong enough to notably delay the bulk of reionization. Such a delay is only likely if ionizing efficiencies of z > 10 sources are much higher (~ two orders of magnitude) than z ~ 6 data seem to imply. We also find that feedback is very strongly dependent on halo mass. Our results suggest that the natural time-scale for the bulk of reionization is the growth of the global collapsed fraction contained in Tvir > 10^4 K halos. Finally, our results underscore the importance of taking into account extended dynamical ranges when modeling reionization.
One well-known way to constrain the hydrogen neutral fraction, x_H, of the high-redshift intergalactic medium (IGM) is through the shape of the red damping wing of the Lya absorption line. We examine this methods effectiveness in light of recent mode
ls showing that the IGM neutral fraction is highly inhomogeneous on large scales during reionization. Using both analytic models and semi-numeric simulations, we show that the picket-fence absorption typical in reionization models introduces both scatter and a systematic bias to the measurement of x_H. In particular, we show that simple fits to the damping wing tend to overestimate the true neutral fraction in a partially ionized universe, with a fractional error of ~ 30% near the middle of reionization. This bias is generic to any inhomogeneous model. However, the bias is reduced and can even underestimate x_H if the observational sample only probes a subset of the entire halo population, such as quasars with large HII regions. We also find that the damping wing absorption profile is generally steeper than one would naively expect in a homogeneously ionized universe. The profile steepens and the sightline-to-sightline scatter increases as reionization progresses. Of course, the bias and scatter also depend on x_H and so can, at least in principle, be used to constrain it. Damping wing constraints must therefore be interpreted by comparison to theoretical models of inhomogeneous reionization.
