Forecasts on primordial non-Gaussianity from 21 cm intensity mapping experiments

Abstract

We forecast the ability of dedicated 21 cm intensity mapping experiments to constraint primordial non-Gaussianity using power spectrum and bispectrum. We model the signal by including the non-linear biasing expansion through a generalized halo model approach. We consider the importance of foreground filtering scale and of the foreground wedge. We find that the current generation intensity mapping experiments like CHIME do not posses sufficient sensitivity to be competitive with the existing limits. On the other hand, upcoming experiments like HIRAX can improve the current constraints and the proposed PUMA experiment can substantially improve them, reaching sensitivities below σ(f NL ) < 5 for equilateral and orthogonal configurations and σ(f NL ) < 1 for the local shape, if good foreground control is achieved.

Primordial information content of Rayleigh anisotropies

“…This is convincingly shown in Fig. 5 where we compare our projected Rayleigh constraints on primordial non-Gaussianity to predictions for proposed and upcoming photometric, spectroscopic, radio galaxy surveys as well as 21 cm experiments [30][31][32]. While future large-scale structure surveys will be able to significantly improve upon CMB constraints for local type non-Gaussianities, improving beyond current constraints on equilateral and orthogonal type non-Gaussianities is challenging and Rayleigh constraints are competitive with futuristic 21 cm experiments for these shapes.…”

“…A comparison of expected primordial non-Gaussianity constraints from CMB measurements (both with and without including Rayleigh measurements) and from a range of up-coming large-scale structure surveys (LSS). We report the predictions for LSS surveys from Castorina et al [30], Karagiannis et al [31,32]; the radio-continuum survey, the spectroscopic survey and the photometric survey are based on SKA-like, DESI-like and LSST-like surveys respectively. We see that the future CMB constraints, when Rayleigh measurements are included, provide competitive constraints to these upcoming LSS surveys.…”

“…While future large-scale structure surveys will be able to significantly improve upon CMB constraints for local type non-Gaussianities, improving beyond current constraints on equilateral and orthogonal type non-Gaussianities is challenging and Rayleigh constraints are competitive with futuristic 21 cm experiments for these shapes. Recent work by Watkinson et al [95] suggests that 21 cm bispectrum measurements required to obtain these constraints could be even more challenging than was considered in Karagiannis et al [32], because mixing between the foregrounds and the signal suggest there is no bispectrum equivalent of the power-spectrum's foreground wedge. We stress that further work is required to fully assess the impact of this, particularly as Watkinson et al [95] studied the epoch of reionization, rather than the lower redshift regime z ∼ 0.3-6 relevant to PUMA-like experiments.…”

“…This obscuration, combined with uncertainties in the details of small scale physics, such as baryonic processes, and observational challenges, such as redshift uncertainties and foreground removal, mean that it is challenging to improve significantly beyond current CMB measurements [65][66][67][68][69][70]. For example, Karagiannis et al [32] find that future generation 21 cm experiments, such as the recently proposed PUMA experiment [30], will only improve by a factor of 2 over current CMB constraints, all while practically mapping out all linear modes up to redshift 6.…”

“…Likewise, large galaxy and 21 cm surveys can provide powerful constraints on some types of primordial non-Gaussianity [24,[27][28][29][30]. Unfortunately, for many types of non-Gaussianity it is challenging to constrain these with large-scale structure surveys due to signal contamination from the non-Gaussian evolution of gravity and structure formation, as well as observational limitations and computationally expensive estimators [31][32][33].…”