Forecasting constraints on the mean free path of ionizing photons at $z \geq 5.4$ from the Lyman-$\alpha$ forest flux auto-correlation function

TL;DR

This paper forecasts how measurements of the Lyman-alpha forest flux auto-correlation can constrain the mean free path of ionizing photons at high redshifts, revealing rapid evolution and informing reionization history.

Contribution

It introduces a modeling approach using mock data to forecast constraints on the mean free path from flux auto-correlation at z ≥ 5.4, highlighting the potential precision and challenges.

Findings

Ideal data can recover mean free path with ~25% uncertainty.

High-resolution data can reduce errors by ~40%.

Auto-correlation distribution is highly non-Gaussian at high z.

Abstract

Fluctuations in Lyman-$\alpha$ (Ly$\alpha$) forest transmission towards high-$z$ quasars are partially sourced from spatial fluctuations in the ultraviolet background (UVB), the level of which are set by the mean free path of ionizing photons ($\lambda_{\text{mfp}}$). The auto-correlation function of Ly$\alpha$ forest flux characterizes the strength and scale of transmission fluctuations and, as we show, is thus sensitive to $\lambda_{\text{mfp}}$. Recent measurements at $z \sim 6$ suggest a rapid evolution of $\lambda_{\text{mfp}}$ at $z>5.0$ which would leave a signature in the evolution of the auto-correlation function. For this forecast, we model mock Ly$\alpha$ forest data with properties similar to the XQR-30 extended data set at $5.4 \leq z \leq 6.0$. At each $z$ we investigate 100 mock data sets and an ideal case where mock data matches model values of the auto-correlation function. For ideal data with $\lambda_{\text{mfp}}=9.0$ cMpc at $z=6.0$, we recover $\lambda_{\text{mfp}}=12^{+6}_{-3}$ cMpc. This precision is comparable to direct measurements of $\lambda_{\text{mfp}}$ from the stacking of quasar spectra beyond the Lyman limit. Hypothetical high-resolution data leads to a $\sim40\%$ reduction in the error bars over all $z$. The distribution of mock values of the auto-correlation function in this work is highly non-Gaussian for high-$z$, which should caution work with other statistics of the high-$z$ Ly$\alpha$ forest against making this assumption. We use a rigorous statistical method to pass an inference test, however future work on non-Gaussian methods will enable higher precision measurements.