Constraining Dynamical Dark Energy with Tomographic Galaxy and CMB Weak Lensing Through Higher-Order Statistics

Abstract

Tomographic weak gravitational lensing probes cosmic structure growth and geometry, making it a powerful probe of dynamical dark energy. We forecast information beyond the lensing power spectrum from non-Gaussian statistics of tomographic galaxy and CMB lensing convergence maps. Using the $\texttt{DDELenS}$ suite of ray-traced $N$-body simulations, we build Gaussian process emulators for power spectra, Minkowski functionals, moments, peak and minima counts, and the wavelet scattering transform, varying $Ω_m$, $σ_8$, $w_0$, and $w_a$. We consider Rubin-like galaxy lensing and Simons Observatory-like CMB lensing over $15{,}000\,\mathrm{deg}^2$, and validate the analysis using a withheld fiducial cosmology, analytical power-spectrum forecasts, and Gaussian random-field null tests. As an initial step, we neglect baryonic effects and intrinsic alignments and approximate galaxy sources as discrete redshift planes. Using all source redshifts at $z_s=\{0.5, 1, 1.5, 2, 2.5, 1100\}$, combining all five non-Gaussian statistics reduces uncertainties on $w_0$ and $w_a$ by $24.3\%$ and $28.0\%$, and on $Ω_m$ and $σ_8$ by $41.9\%$ and $45.8\%$, relative to the power-spectrum-only baseline. The resulting $w_0$ and $w_a$ uncertainties are $34.7\%$ and $48.7\%$ smaller than current DESI+DES constraints. If the DESI+DES best-fit values remain unchanged, this precision would increase the significance of the departure from $Λ$CDM from $3.0σ$ to $5.2σ$, compared with $3.8σ$ from the power spectrum alone, assuming a Gaussian posterior and unchanged $w_0$--$w_a$ correlation. Minkowski functionals and moments provide comparable constraining power, while combining them with the other summaries gives the strongest constraints. Broad galaxy-lensing redshift coverage is particularly important for dynamical dark energy, while CMB lensing provides complementary information.

Abstract: arXiv descriptive metadata (CC0). Thank you to arXiv for use of its open access interoperability. This service was not reviewed or approved by, nor does it necessarily express or reflect the policies or opinions of, arXiv.