Exploring Beyond {\Lambda}CDM with the Weak Lensing Power Spectrum and Bispectrum

TL;DR

This paper uses Fisher matrix forecasts to show that including the weak lensing bispectrum alongside the power spectrum significantly improves constraints on beyond-$ mf \Lambda$CDM models, highlighting the importance of higher-order statistics.

Contribution

It provides the first detailed forecast of how the weak lensing bispectrum enhances constraints on dynamical dark energy, interacting dark energy, and $f(R)$ gravity models.

Findings

Bispectrum tightens constraints on dark energy parameters by up to 60%.

$f(R)$ models are most sensitive to systematics, especially in bispectrum.

Higher order weak lensing statistics are essential for maximizing scientific return.

Abstract

In this work, we present Fisher matrix forecast of the tomographic weak lensing power spectrum and bispectrum for three physically distinct types of models of beyond-$\Lambda$CDM: the CPL parametrisation of dynamical dark energy, interacting dark energy (IDE) with a dark sector energy-momentum exchange, and Hu-Sawicki models of $f(R)$ gravity. We find that for all three models, including the bispectrum significantly tightens the Fisher constraints: the bispectrum reduces the marginalised $1\sigma$ error on the CPL equation of state parameter from $\sigma(w_0) = 0.2511$ (power spectrum only) to $\sigma(w_0) = 0.1557$, on the IDE coupling from $\sigma(\alpha) = 2.6895$ to $\sigma(\alpha) = 0.2944$, and on the scalaron amplitude from $\sigma(\ln|f_{R0}|) = 2.236$ to $\sigma(\ln|f_{R0}|) = 2.237$ after full marginalisation over nuisance parameters e.g., photo-z error $\sigma_z$ and intrinsic alignment amplitude $\mathcal{A}_{\rm IA}$. We find that $f(R)$ models are the most sensitive to systematics and especially in bispectrum. The results also demonstrates the importance of higher order weak lensing statistics as a practical necessity to maximise the scientific return of Stage IV surveys.