Estimating the Feasibility of 21cm-Ly$\alpha$ Synergies using the Hydrogen Epoch of Reionization Array

TL;DR

This paper evaluates the potential for detecting the cross-correlation between 21cm and Ly$ ext{alpha}$ intensity maps during the Epoch of Reionization using next-generation experiments, considering observational limitations and foreground filtering.

Contribution

It provides detailed forecasts and simulations of the 21cm-Ly$ ext{alpha}$ cross-power spectrum detection feasibility with HERA, SPHEREx, and CDIM, accounting for experimental constraints.

Findings

Detection of SPHEREx-HERA cross-correlation is challenging and requires optimistic foreground filtering.

CDIM may detect the cross-power spectrum even with stricter foreground filtering.

Optimal detection occurs around redshift z ~ 8 in the models.

Abstract

Cross-correlating 21cm and Ly$\alpha$ intensity maps of the Epoch of Reionization (EoR) promises to be a powerful tool for exploring the properties of the first galaxies. Next-generation intensity mapping experiments such as the Hydrogen Epoch of Reionization Array (HERA) and SPHEREx will individually probe reionization through the power spectra of the 21cm and Ly$\alpha$ lines respectively, but will be limited by bright foregrounds and instrumental systematics. Cross-correlating these measurements could reduce systematics, potentially tightening constraints on the inferred astrophysical parameters. In this study, we present forecasts of cross-correlation taking into account the effects of exact uv-sampling and foreground filtering to estimate the feasibility of HERAxSPHEREx making a detection of the 21cm-Ly$\alpha$ cross-power spectrum. We also project the sensitivity of a cross-power spectrum between HERA and the proposed next-generation Cosmic Dawn Intensity Mapper. By isolating the sources of uncertainty, we explore the impacts of experimental limitations such as foreground filtering and Ly$\alpha$ thermal noise uncertainty have on making a detection of the cross-power spectrum. We then implement this strategy in a simulation of the cross-power spectrum and observational error to identify redshifts where fiducial 21cmFAST models predict the highest signal-to-noise detection ($z \sim 8$). We conclude that detection of the SPHEREx-HERA cross-correlation will require an optimistic level of 21cm foreground filtering, as well as deeper thermal noise integrations due to a lack of overlapping sensitive modes but for CDIM with its larger range of scales and lower noise forecast detection levels, may be possible even with stricter 21cm foreground filtering.