Abstract

Contributed Talk - Splinter LargeScale

Thursday, 10 September 2026, 15:30   (MW-2235)

Evolution Mapping of the Nonlinear Matter Bispectrum

Rishabh Ranjan (1), Alexander Eggemeier (1), Cristiano Porciani (1), Ariel G Sanchez (2)
(1) University of Bonn, (2) MPE Garching

The matter bispectrum is a key ingredient for predicting the bispectra of galaxies and cosmic shear---yet modelling or emulating its full cosmology and redshift dependence in the nonlinear regime is challenging. In this talk, I show how the evolution mapping framework can significantly reduce this complexity, charting a path towards its accurate emulation. The framework sorts cosmological parameters into shape parameters, which set the shape of the linear matter power spectrum, and evolution parameters, which affect only its amplitude. The degeneracies between evolution parameters and redshift can then be encoded into a single quantity, σ_12, the RMS of linear density fluctuations smoothed on a scale of 12 Mpc. Using the Aletheia suite of N-body simulations, I demonstrate that this approach compresses the full parameter and redshift dependence of the matter bispectrum across a wide range of cosmologies, even in the nonlinear regime. When additionally including a parameter that characterises deviations from the integrated Einstein–de Sitter growth history, it is possible to reach better than 1% accuracy up to k ~ 1 Mpc^{-1} across many different triangle configurations. I close by presenting first steps towards practical application, constructing Gaussian process emulators for the individual components of the bispectrum within this framework.