Measuring primordial non-Gaussianity with weak-lensing surveys

TL;DR

Future weak lensing surveys like Euclid and LSST can constrain primordial non-Gaussianity and other cosmological parameters, especially when combining multiple probes, but neglecting non-Gaussianity biases results.

Contribution

This study evaluates the potential of weak lensing surveys to measure primordial non-Gaussianity and highlights the importance of combining multiple probes for accurate parameter estimation.

Findings

Unmarginalized constraint on f_NL is about 20.

Combined probes reduce uncertainties to Delta f_NL ~ 50.

Neglecting f_NL biases sigma_8 and w_0 estimates.

Abstract

We study the ability of future weak lensing (WL) surveys to constrain primordial non-Gaussianity of the local type. We use a large ensemble of simulated WL maps with survey specifications relevant to Euclid and LSST. The simulations assume Cold Dark Matter cosmologies that vary certain parameters around fiducial values: the non-Gaussianity parameter f_NL, the matter density parameter Omega_m, the amplitude of the matter power spectrum sigma_8, the spectral index of the primordial power spectrum n_s, and the dark-energy equation-of-state parameter w_0. We assess the sensitivity of the cosmic shear correlation functions, the third-order aperture mass statistics, and the abundance of shear peaks to these parameters. We find that each of the considered probes provides unmarginalized constraints of Delta f_NL ~ 20 on f_NL. Marginalized constraints from any individual WL probe are much weaker due to strong correlations between parameters. However, the parameter errors can be substantially reduced by combining information from different WL probes. Combining all WL probes yields the following marginal (68% confidence level) uncertainties: Delta f_NL ~ 50, Delta Omega_m ~ 0.002, Delta sigma_8 ~ 0.004, Delta n_s ~ 0.007, and Delta w_0 ~ 0.03. We examine the bias induced by neglecting f_NL on the constraints on the other parameters. We find sigma_8 and w_0 to be the most affected. Moreover, neglecting non-Gaussianity leads to a severe underestimation of the uncertainties in the other cosmological parameters.