An improved measurement of the flux distribution of the Ly-alpha forest in QSO absorption spectra: the effect of continuum fitting, metal contamination and noise properties

TL;DR

This study refines the measurement of the flux distribution of the Ly-alpha forest in QSO spectra by carefully accounting for continuum fitting, metal contamination, and noise, leading to more accurate insights into intergalactic hydrogen absorption.

Contribution

It provides an improved method for separating metal absorption from HI and offers more precise flux PDF measurements at various redshifts compared to previous studies.

Findings

Flux PDF measurements are systematically lower than previous estimates.

HI effective optical depth follows a single power law with redshift.

Noise has minimal impact on flux distribution in high-quality data.

Abstract

We have performed an extensive Voigt profile analysis of the neutral hydrogen (HI) and metal absorption present in a sample of 18 high resolution, high signal-to-noise QSO spectra observed with VLT/UVES. We use this analysis to separate the metal contribution from the HI absorption and present an improved measurement of the flux probability distribution function (PDF) due to HI absorption alone at <z> = 2.07, 2.52, and 2.94. The flux PDF is sensitive to the continuum fit in the normalised flux range 0.8 < F < 1.0 and to metal absorption at 0.2 < F < 0.8. Our new measurements of the flux PDF due to HI absorption alone are systematically lower at 0.2 < F < 0.8 by up to 30% compared to the widely used measurement of McDonald et al. (2000), based on a significantly smaller sample of Keck/HIRES data. This discrepancy is probably due to a combination of our improved removal of the metal absorption and cosmic variance, since variations in the flux PDF between different lines-of-sight are large. The HI effective optical depth at 1.7 < z < 4 is best fit with a single power law, (0.0023 +-0.0007) (1+z)^(3.65 +- 0.21), in good agreement with previous measurements from comparable data. As also found previously, the effect of noise on the flux distribution is not significant in high resolution, high signal-to-noise data.