A roadmap to cosmological parameter analysis with third-order shear statistics IV. Analytic cross-covariance between second- and third-order aperture masses

Abstract

Weak gravitational lensing is a powerful probe of cosmology, with second-order shear statistics commonly used to constrain parameters such as the matter density $\Omega_\mathrm{m}$ and the clustering amplitude $S_8$. However, parameter degeneracies remain and can be reduced by including higher-order statistics such as the third-order aperture mass. To jointly analyse second- and third-order statistics, an accurate model of their cross-covariance is essential. We derive and validate a non-tomographic analytical model for the cross-covariance between second- and third-order aperture mass statistics. Analytical models are computationally efficient and enable cosmological parameter inference across a range of models, in contrast to numerical covariances derived from simulations or resampling, which are costly or biased. Our derivation is based on real-space estimators of the aperture mass. Substituting the Halofit power spectrum, BiHalofit bispectrum, and a halo-model tetraspectrum, we validate the model against numerical covariances from the $N$-body Scinet LIghtCone Simulations (SLICS) using shear catalogues and convergence maps. We perform a Markov chain Monte Carlo analysis with both analytical and numerical covariances for several filter scale combinations. The cross-covariance separates into three terms governed by the power spectrum, bispectrum, and tetraspectrum, with the latter dominating. The analytical model qualitatively reproduces simulations, though differences arise from modelling approximations and numerical evaluation. Analytical contours are systematically tighter, with a combined figure of merit 72% of the numerical case, rising to 80% when small-scale information is excluded. This work completes the analytical covariance framework for second- and third-order aperture mass statistics, enabling joint parameter inference without large simulation suites.