Publications (6)30.35 Total impact

[Show abstract] [Hide abstract]
ABSTRACT: Spectra of highredshift QSOs show deep GunnPeterson absorptions on the blue sides of the Lyα emissions lines. They can be decomposed into components called Lyα leaks, defined to be emissive regions in complementary to otherwise zeroflux absorption gaps. Just like Lyα absorption forests at low redshifts, Lyα leaks are easy to find in observations and contain rich sets of statistical properties that can be used to study the early evolution of the intergalactic medium (IGM). Among all properties of a leak profile, we investigate its equivalent width in this paper, since it is weakly affected by instrumental resolution and noise. Using 10 Keck QSO spectra at z ~ 6, we have measured the number density distribution function n(W, z) , defined to be the number of leaks per equivalent width W and per redshift z in the redshift range 5.46.0. These new observational statistics, in both the differential and cumulative forms, fit well to hydrodynamic simulations of uniform ionizing background in the ΛCDM cosmology. In this model, Lyα leaks are mainly due to lowdensity voids. It supports the early studies that the IGM at z 6 would still be in a highly ionized state with a neutral hydrogen fraction 10−4. Measurements of n(W, z) at z > 6 would be effective to probe the reionization of the IGM.The Astrophysical Journal 12/2008; 671(2):L89. DOI:10.1086/525279 · 6.28 Impact Factor 
[Show abstract] [Hide abstract]
ABSTRACT: The calculation of the transmission power spectrum of QSO Lyα absorption requires two parameters for the normalization: the continuum Fc and mean transmission τ. Traditionally, the continuum is obtained by a polynomial fitting truncating it at a lower order, and the mean transmission is calculated over the entire wavelength range considered. The flux F is then normalized by Fcτ. However, the fluctuations in the transmitted flux are significantly correlated with the local background flux on scales for which the field is intermittent. As a consequence, the normalization of the entire power spectrum by an overall mean transmission τ will overlook the effect of the fluctuationbackground correlation upon the powers. In this paper we develop a selfnormalization algorithm of the transmission power spectrum based on a multiresolution analysis. This selfnormalized power spectrum estimator needs neither a continuum fitting nor a predetermining of the mean transmission. With simulated samples, we show that the selfnormalization algorithm can perfectly recover the transmission power spectrum from the flux regardless of how the continuum varies with wavelength. We also show that the selfnormalized power spectrum is also properly normalized by the mean transmission. Moreover, this power spectrum estimator is sensitive to the nonlinear behavior of the field. That is, the selfnormalized power spectrum estimator can distinguish between fields with or without the fluctuationbackground correlation. This cannot be accomplished by the power spectrum with the normalization by an overall mean transmission. Applying this analysis to a real data set of Q1700+642 Lyα forest, we demonstrate that the proposed power spectrum estimator can perform correct normalization and effectively reveal the correlation between the fluctuations and background of the transmitted flux on small scales. Therefore, the selfnormalized power spectrum would be useful for the discrimination among models without the uncertainties caused by free (or fitting) parameters.The Astrophysical Journal 12/2008; 561(1):94. DOI:10.1086/323216 · 6.28 Impact Factor 
[Show abstract] [Hide abstract]
ABSTRACT: We study the GunnPeterson effect of the photoionized intergalactic medium (IGM) in the redshift range 5 < z < 6.4 using semianalytic simulations based on the lognormal model. Assuming a rapidly evolved and spatially uniform ionizing background, the simulation can produce all the observed abnormal statistical features near redshift z 6. They include (1) a rapid increase of absorption depths, (2) large scatter in the optical depths, (3) longtailed distributions of transmitted flux, and (4) long dark gaps in spectra. These abnormal features are mainly due to rare events, which correspond to the longtailed probability distribution of the IGM density field, and therefore they may not imply significant spatial fluctuations in the UV ionizing background at z 6.The Astrophysical Journal 12/2008; 645(1):L1. DOI:10.1086/506149 · 6.28 Impact Factor 
Article: Lyα leaks and reionization
[Show abstract] [Hide abstract]
ABSTRACT: Lyα absorption spectra of QSOs at redshifts z≃ 6 show complete Gunn–Peterson absorption troughs (dark gaps) separated by tiny leaks. The dark gaps are from the intergalactic medium (IGM) where the density of neutral hydrogen are high enough to produce almost saturated absorptions, however, where the transmitted leaks come from is still unclear so far. We demonstrate that leaking can originate from the lowest density voids in the IGM as well as the ionized apatches around ionizing sources using semianalytical simulations. If leaks are produced in lowest density voids, the IGM must already be highly ionized, and the ionizing background should be almost uniform; in contrast, if leaks come from ionized patches, the neutral fraction of IGM should be still high, and the ionizing background is significantly inhomogeneous. Therefore, the origin of leaking is crucial to determining the epoch of inhomogeneoustouniform transition of the ionizing photon background. We show that the origin could be studied with the statistical features of leaks. Actually, Lyα leaks can be well defined and described by the equivalent width W and the full width of halfarea WH, both of which are less contaminated by instrumental resolution and noise. It is found that the distributions of W and WH of Lyα leaks are sensitive to the modelling of the ionizing background. We consider four representative models: uniform ionizing background (model 0), the photoionization rate of neutral hydrogen ΓH i and the density of IGM are either linearly correlated (model I), or anticorrelated (model II), and ΓH i is correlated with highdensity peaks containing ionizing sources (model III). Although all of these models can match to the mean of the observed effective optical depth of the IGM at z≃ 6, the distributions of W and WH are very different from each other. Consequently, the leak statistics provides an effective tool to probe the evolutionary history of reionization at z≃ 5–6.5. Similar statistics will also be applicable to the reionization of He ii at z≃ 3Monthly Notices of the Royal Astronomical Society 01/2008; 383(4):1459  1468. DOI:10.1111/j.13652966.2007.12642.x · 5.23 Impact Factor 
[Show abstract] [Hide abstract]
ABSTRACT: Spectra of high redshift QSOs show deep GunnPeterson absorptions on the blue sides of the Lya emissions lines. They can be decomposed into components called Lya leaks, defined to be emissive regions in complementary to otherwise zerofluxed absorption gaps. Just like Lya absorption forests at low redshifts, Lya leaks are both easy to find in observations and containing rich sets of statistical properties that can be used to study the early evolution of the IGM. Among all properties of a leak profile, we investigate its equivalent width in this paper, since it is weakly affected by instrumental resolution and noise. Using 10 Keck QSO spectra at z 6, we have measured the number density distribution function n(W,z), defined to be the number of leaks per equivalent width W and per redshift z, in the redshift range 5.4  6.0. These new observational statistics, in both the differential and cumulative forms, fit well to hydro numerical simulations of uniform ionizing background in the LCDM cosmology.In this model, Lya leaks are mainly due to low density voids.It supports the early studies that the IGM at z 6 would still be in a highly ionized state with neutral hydrogen fraction 104. Measurements of n(W,z) at z>6; would be effective to probe the reionization of the IGM. 
[Show abstract] [Hide abstract]
ABSTRACT: We study the baryonic gas clouds (the IGM) in the universe before the reionization with the lognormal model which is shown to be dynamcially legitimate in describing the fluctuation evolution in quasilinear as well as nonlinear regimes in recent years. The probability distribution function of the mass field in the LN model is long tailed and so plays an important role in rare events, such as the formation of the first generation of baryonic objects. We calculate density and velocity distributions of the IGM at very high spatial resolutions, and simulate the distributions at resolution of 0.15 kpc from z=7 to 15 in the LCDM cosmological model. We performed a statistics of the hydrogen clouds including column densities, clumping factors, sizes, masses, and spatial number density etc. One of our goals is to identify which hydrogen clouds are going to collapse. By inspecting the mass density profile and the velocity profile of clouds, we found that the velocity outflow significantly postpones the collapsing process in less massive clouds, in spite of their masses are larger than the Jeans mass. Consequently, only massive (> 10^5 M_sun) clouds can form objects at higher redshift, and less massive (10^410^5) collapsed objects are formed later. For example, although the mass fraction in clouds with sizes larger than the Jeans length is already larger than 1 at z=15, there is only a tiny fraction of mass (10^{8}) in the clouds which are collapsed at that time. If all the ionizing photons, and the 10^{2} metallicity observed at low redshift are produced by the first 1% mass of collapsed baryonic clouds, the majority of those first generation objects would not happen until z=10.The Astrophysical Journal 09/2003; 598(1). DOI:10.1086/378793 · 6.28 Impact Factor
Publication Stats
32  Citations  
30.35  Total Impact Points  
Top Journals
Institutions

2008

Government of the People's Republic of China
Peping, Beijing, China 
The University of Arizona
 Department of Physics
Tucson, AZ, United States
