Cosmological parameters from the joint analysis of Density Split and Second Order Statistics: an Emulator based on the Halo Occupation Distribution

TL;DR

This paper introduces a simulation-based emulator for predicting density split and second-order shear and clustering statistics, improving modeling of non-linear scales for cosmological parameter estimation.

Contribution

It develops a novel emulator trained on AbacusSummit simulations to accurately model complex non-linear statistics, enhancing cosmological analysis capabilities.

Findings

Emulator achieves high accuracy in modeling non-linear scales.

Joint analysis improves constraints on galaxy-halo connection.

Density split statistics provide strong cosmological parameter constraints.

Abstract

In this work, we develop a simulation-based model to predict the density split (DSS) and second-order shear and clustering statistics. A simulation-based model has the potential to model highly non-linear scales where current DSS models fail. To build this model, we use the $\texttt{AbacusSummit}$ N-body simulation suite from which we measure all necessary statistics and train an emulator based on $\texttt{CosmoPower}$. In that context, we discuss possible improvements for future emulators to make the measurement less noisy and biased, resulting in more accurate and precise model predictions. Regarding the emulator's accuracy, we find that the most important aspect is the average of the summary statistics over multiple-shot noise realizations of the foreground galaxies. However, these results probably depend on the chosen number density of the foreground galaxies. Regarding the parameter forecast based on preliminary LOWZxUNIONS data, we find that DSS has more constraining power to derive cosmological parameters and that the joint analysis with second-order statistics is particularly useful for extracting parameters of the galaxy-halo connection.