Cosmic density field reconstruction from Ly-alpha forest data

TL;DR

This paper introduces a fast, non-parametric method to reconstruct the cosmic density field from high-redshift quasar Ly-alpha forest data by leveraging flux probability distributions and flux-density relations.

Contribution

The authors develop a novel inversion technique that reconstructs the matter density field from Ly-alpha forest spectra without relying on physical models of the intergalactic medium.

Findings

Achieves less than 12% relative error in over 50% of pixels with only 10 spectra

Successfully reconstructs the density field in over 90% of the line of sight

Method is validated using synthetic spectra with known density distributions

Abstract

We present a novel, fast method to recover the density field through the statistics of the transmitted flux in high redshift quasar absorption spectra. The proposed technique requires the computation of the probability distribution function of the transmitted flux (P_F) in the Ly-alpha forest region and, as a sole assumption, the knowledge of the probability distribution function of the matter density field (P_Delta). We show that the probability density conservation of the flux and matter density unveils a flux-density (F-Delta) relation which can be used to invert the Ly-alpha forest without any assumption on the physical properties of the intergalactic medium. We test our inversion method at z=3 through the following steps: [i] simulation of a sample of synthetic spectra for which P_Delta is known; [ii] computation of P_F; [iii] inversion of the Ly-alpha forest through the F-Delta relation. Our technique, when applied to only 10 observed spectra characterized by a signal-to noise ratio S/N >= 100 provides an exquisite (relative error epsilon_Delta <~ 12 % in >~ 50 % of the pixels) reconstruction of the density field in >~ 90 % of the line of sight. We finally discuss strengths and limitations of the method.