CHORUS IV: Mapping the Spatially Inhomogeneous Cosmic Reionization with Subaru HSC

TL;DR

This study maps the inhomogeneous distribution of cosmic reionization at z~6.6 using Subaru HSC data, revealing significant spatial variation in neutral hydrogen and its impact on galaxy populations, though current data limits definitive conclusions.

Contribution

First to simultaneously map LAEs and LBGs over a large area at z~6.6, estimating neutral fraction variations and their relation to galaxy properties through numerical simulations.

Findings

Average neutral fraction x_HI < 0.4, consistent with previous studies.

Spatial variation in LAE/LBG ratio up to a factor of three within the survey area.

LAEs in high-density regions tend to have higher equivalent widths, suggesting a link to neutral fraction.

Abstract

The spatial inhomogeneity is one of the important features for understanding the reionization process; however, it has not yet been fully quantified. To map this inhomogeneous distribution, we simultaneously detect Ly$\alpha$ emitters (LAEs) and Lyman break galaxies (LBGs) at $z \sim 6.6$ from the Subaru/Hyper Suprime-Cam (HSC) large-area ($\sim1.5\,\mathrm{ deg}^2 = 34000\,\mathrm{cMpc}^2$) deep survey. We estimate the neutral fraction, $x_\mathrm{HI}$, from the observed number density ratio of LAEs to LBGs, $n(\mathrm{LAE})/n(\mathrm{LBG})$ based on numerical radiative transfer simulation, in which model galaxies are selected to satisfy the observed selection function. While the average $x_\mathrm{HI}$ within the field of view is found to be $x_\mathrm{HI} < 0.4$, which is consistent with previous studies, the variation of $n(\mathrm{LAE})/n(\mathrm{LBG})$ within the field of view for each $140\,\mathrm{pMpc}^2$ is found to be as large as a factor of three. This may suggest a spatially inhomogeneous topology of reionization, but it also leaves open the possibility that the variation is based on the inherent large-scale structure of the galaxy distribution. Based on the simulations, it may be difficult to distinguish between the two from the current survey. We also find that LAEs in the high LAE density region are more populate high $\mathrm{EW}_0$, supporting that the observed $n(\mathrm{LAE})/n(\mathrm{LBG})$ is more or less driven by the neutral fraction, though the statistical significance is not high.