[Paper Review] Hydrogen reionisation ends by $z=5.3$: Lyman-$\alpha$ optical depth measured by the XQR-30 sample

Sarah E. I. Bosman, Frederick B. Davies+8 more|arXiv (Cornell University)|Aug 10, 2021

Radio Astronomy Observations and Technology6 references7 citations

TL;DR

This study uses high signal-to-noise VLT/X-Shooter spectra from the XQR-30 sample to measure Lyman-alpha optical depth at z > 5.5, finding that homogeneous UVB models fail to reproduce observed transmission scatter at z ≥ 5.4. The results indicate hydrogen reionisation ended by z = 5.3, with persistent IGM inhomogeneities in ionisation, neutral fraction, or temperature, supporting a late, patchy end to reionisation.

ABSTRACT

The presence of excess scatter in the Ly-$\alpha$ forest at $z\sim 5.5$, together with the existence of sporadic extended opaque Gunn-Peterson troughs, has started to provide robust evidence for a late end of hydrogen reionisation. However, low data quality and systematic uncertainties complicate the use of Ly-$\alpha$ transmission as a precision probe of reionisation's end stages. In this paper, we assemble a sample of 67 quasar sightlines at $z>5.5$ with high signal-to-noise ratios of $>10$ per $\leq 15$ km s$^{-1}$ spectral pixel, relying largely on the new XQR-30 quasar sample. XQR-30 is a large program on VLT/X-Shooter which obtained deep (SNR $>20$ per pixel) spectra of 30 quasars at $z>5.7$. We carefully account for systematics in continuum reconstruction, instrumentation, and contamination by damped Ly-$\alpha$ systems. We present improved measurements of the mean Ly-$\alpha$ transmission over $4.9<z<6.1$. Using all known systematics in a forward modelling analysis, we find excellent agreement between the observed Ly-$\alpha$ transmission distributions and the homogeneous-UVB simulations Sherwood and Nyx up to $z\leq5.2$ ($<1 \sigma$), and mild tension ($\sim 2.5 \sigma$) at $z=5.3$. Homogeneous UVB models are ruled out by excess Ly-$\alpha$ transmission scatter at $z\geq5.4$ with high confidence ($>3.5 \sigma$). Our results indicate that reionisation-related fluctuations, whether in the UVB, residual neutral hydrogen fraction, and/or IGM temperature, persist in the intergalactic medium until at least $z=5.3$ ($t=1.1$ Gyr after the Big Bang). This is further evidence for a late end to reionisation.

Motivation & Objective

• To precisely measure Lyman-alpha optical depth at z > 5.5 using high signal-to-noise quasar spectra.
• To test whether homogeneous UVB models (Sherwood, Nyx) can reproduce observed Lyman-alpha transmission at high redshift.
• To assess the role of systematics in continuum reconstruction, instrumentation, and damped Lyman-alpha systems in shaping transmission measurements.
• To determine the redshift at which hydrogen reionisation ended by analyzing scatter and transmission profiles.

Proposed method

• Constructed a sample of 67 high signal-to-noise quasar sightlines at z > 5.5, primarily from the XQR-30 survey (SNR > 10 per ≤15 km s⁻¹ pixel).
• Used deep X-Shooter spectra (SNR > 20 per pixel) from 30 quasars at z > 5.7 to improve signal-to-noise and reduce noise-induced scatter.
• Performed forward modelling of Lyman-alpha transmission using homogeneous UVB simulations (Sherwood, Nyx), incorporating all known systematics in continuum reconstruction, instrument response, and DLA contamination.
• Quantified tension between observed transmission distributions and model predictions using statistical significance (σ levels), focusing on scatter and mean transmission at z ≈ 5.3–5.8.
• Accounted for cosmic variance and line-of-sight differences by comparing multiple sightlines at fixed redshift.
• Applied rigorous error propagation and systematics correction to ensure robustness of optical depth measurements.

Experimental results

Research questions

• RQ1 At what redshift does hydrogen reionisation end, based on Lyman-alpha transmission and scatter?
• RQ2 Do homogeneous UVB models accurately reproduce the observed Lyman-alpha transmission at z ≈ 5.3–5.8?
• RQ3 What is the significance of excess Lyman-alpha transmission scatter at z ≥ 5.4, and what does it imply about IGM inhomogeneities?
• RQ4 How do systematics in continuum fitting, DLA contamination, and instrumental response affect high-redshift Lyman-alpha measurements?
• RQ5 Can the observed transmission profiles at z > 5.5 be explained by a late, patchy reionisation scenario?

Key findings

• The mean Lyman-alpha optical depth is consistent with homogeneous UVB models (Sherwood, Nyx) at z ≤ 5.2 at the <1σ level.
• At z = 5.3, there is mild tension (~2.5σ) between observations and homogeneous UVB models.
• Excess Lyman-alpha transmission scatter at z ≥ 5.4 is ruled out by homogeneous UVB models at >3.5σ significance.
• The persistence of inhomogeneous IGM conditions—such as fluctuations in UV background, neutral hydrogen fraction, or temperature—indicates reionisation ended by z = 5.3.
• The results support a late, patchy end to hydrogen reionisation, with residual neutral regions surviving until at least z = 5.3 (1.1 Gyr after the Big Bang).
• The XQR-30 sample provides the most precise high-redshift Lyman-alpha transmission measurements to date, enabling robust constraints on reionisation end stages.