Predictions for BAO distance estimates from the cross-correlation of the Lyman-alpha forest and redshifted 21-cm emission

TL;DR

This study explores using the cross-correlation of the Lyman-alpha forest and 21-cm emission to improve measurements of baryon acoustic oscillations, showing it can significantly enhance distance estimate accuracy.

Contribution

It demonstrates that cross-correlation can outperform auto-correlation in measuring BAO distances, with specific sensitivity improvements for planned observations.

Findings

Cross-correlation improves BAO distance measurement sensitivity by up to 2.7 times.

Two years of observations with 2,000 antennas yield 1.7 times better sensitivity.

Cross-correlation can significantly enhance the accuracy of cosmological distance estimates.

Abstract

We investigate the possibility of using the cross-correlation of the Lyman-alpha forest and redshifted 21-cm emission to detect the baryon acoustic oscillation (BAO). The standard Fisher matrix formalism is used to determine the accuracy with which it will be possible to measure cosmological distances using this signal. Earlier predictions indicate that it will be possible to measure the dilation factor D_V with 1.9 % accuracy at z=2.5 from the BOSS Lyman-alpha forest auto-correlation. In this paper we investigate if it is possible to improve the accuracy using the cross-correlation. We use a simple parametrization of the Lyman-alpha forest survey which very loosely matches some properties of BOSS and predicts delta D_V/D_V = 2.0 % for the auto-correlation at z=2.5. For the redshifted 21-cm observations we consider individual antennas of size 2 m * 2 m distributed such that the baselines within 250 m are uniformly sampled. It is assumed that the observations span z=2 to 3 and covers the 10,000 deg^2 sky coverage of BOSS. We find that for 2 years of observation with an array of 2,000 antennas, the cross-correlation is 1.7 times more sensitive than the Lyman-alpha forest auto-correlation. The cross-correlation is 2.7 times more sensitive than the auto-correlation if we have 4,000 antennas and 4 years of observation. In conclusion, we find that it is possible to significantly increase the accuracy of the distance estimates by considering the cross-correlation signal.