Towards a unified description of the intergalactic medium at redshift z ≈ 2.5

We examine recent measurements of the z ≈ 2.5 intergalactic medium (IGM) which constrain the H I frequency distribution f(NHI) and the mean free path λ912mfp to ionizing radiation. We argue that line-blending and the clustering of strong absorption-line systems have led previous authors to systematically overestimate the effective Lyman limit opacity, yielding too small of a λ912mfp for the IGM. We further show that recently published measurements of f(NHI) at NHI≈1016cm−2 lie in strong disagreement, implying underestimated uncertainty from sample variance and/or systematics like line-saturation. Allowing for a larger uncertainty in the f(NHI) measurements, we provide a new cubic Hermite spline model for f(NHI) which reasonably satisfies all of the observational constraints under the assumption of randomly distributed absorption systems. We caution, however, that this formalism is invalid in the light of absorber clustering and use a toy model to estimate the effects. Future work must properly account for the non-Poissonian nature of the IGM.