Self-calibration of weak lensing systematic effects using combined two- and three-point statistics

TL;DR

This paper demonstrates that combining the weak lensing bispectrum with the power spectrum significantly improves control over systematic uncertainties in cosmological parameter estimation for Euclid-like surveys.

Contribution

It introduces a joint analysis of power spectrum and bispectrum that effectively mitigates key systematic effects and self-calibrates them, enhancing the accuracy of cosmological constraints.

Findings

Bispectrum reduces credible region areas for key parameters by up to 90%.

Joint analysis improves dark energy parameter constraints by nearly 20 times.

Including bispectrum self-calibrates systematics to required levels, reducing external calibration needs.

Abstract

We investigate the prospects for using the weak lensing bispectrum alongside the power spectrum to control systematic uncertainties in a Euclid-like survey. Three systematic effects are considered: the intrinsic alignment of galaxies, uncertainties in the means of tomographic redshift distributions, and multiplicative bias in the measurement of the shear signal. We find that the bispectrum is very effective in mitigating these systematic errors. Varying all three systematics simultaneously, a joint power spectrum and bispectrum analysis reduces the area of credible regions for the cosmological parameters $\Omega_\mathrm{m}$ and $\sigma_8$ by a factor of 90 and for the two parameters of a time-varying dark energy equation of state by a factor of almost 20, compared with the baseline approach of using the power spectrum alone and of imposing priors consistent with the accuracy requirements specified for Euclid. We also demonstrate that including the bispectrum self-calibrates all three systematic effects to the stringent levels required by the forthcoming generation of weak lensing surveys, thereby reducing the need for external calibration data.