Reconciling large-scale Lyman-$α$ correlations with the SCRIPT Semi-numerical Model

Abstract

Recent analyses of high-redshift Lyman-$α$ forest observations have revealed strong correlations on scales exceeding 200 cMpc at redshift z = 6. Reproducing these large-scale correlations has proven challenging for current large-volume reionization simulations. In this work, we investigate these large-scale correlations using mock spectra generated from the extended SCRIPT semi-numerical reionization model. We find that while the fiducial model ensemble systematically predicts smaller correlation lengths than those inferred from the 67 sightlines in the extended XQR-30 sample, a small fraction of individual mock realizations can naturally reproduce the observed signal. Using a delete-2 jackknife analysis, we demonstrate that the observed large-scale correlation length is disproportionately driven by a rare pair of highly transmissive sightlines associated with high-redshift transmission spikes. By inserting two such highly transmissive sightlines into our mock realizations, the fraction of models consistent with the observed redshift evolution and correlation length increases significantly from 17.5% to 74.1%. Furthermore, we show that spatial fluctuations in the ionizing mean free path remain an essential physical ingredient for reproducing the observed correlation structure. Our results suggest that the unexpectedly large Lyman-$α$ correlations can be reconciled with existing reionization models when accounting for cosmic variance and the outsized statistical impact of rare, highly transmissive sightlines.