Large-scale structure in the Lyman-  forest: a new technique

Abstract

A B S T R A C TWe present a new technique for detecting structure on Mpc scales in the Lya forest. The technique is easy to apply in practice since it does not involve absorption-line fitting, but rather is based on the statistics of the transmitted flux. It identifies and assesses the statistical significance of regions of overdense or underdense Lya absorption and is fairly insensitive to the quality of the data. Using extensive simulations, we demonstrate that the new method is significantly more sensitive to the detection of large-scale structure in the Lya forest than a traditional two-point correlation function analysis of fitted absorption lines.

The proximity effect in a close group of QSOs

“…The reason is that a given underdensity of absorption is more significantly detected when the ‘background’ absorption‐line density is higher than when it is lower (cf. LWC, fig. 6b).…”

“…Essentially, we use the method of Liske, Webb & Carswell (1998, hereafter LWC) to look for regions of underdense absorption (i.e., ‘voids’) in Ly α forest spectra near the positions of suspected sources of ionizing radiation. These sources may be the background QSOs themselves, or foreground sources near the line of sight to the background QSOs.…”

“…σ n denotes the noise of a spectrum, σ LSF the width of the instrumental line spread function, ps the pixel size in Å, and . q is the intrinsic width of the autocovariance function of a ‘perfect’ spectrum (i.e., before it passes through the instrument), which we determine from simulations (LWC; Liske et al 2000). B contains the normalization and is given by…”