Sensitivity of galaxy cluster dark energy constraints to halo modeling uncertainties

TL;DR

This study assesses how uncertainties in halo modeling affect dark energy constraints from galaxy cluster surveys, highlighting the importance of precise mass-observable relations and mass function modeling for accurate cosmological inferences.

Contribution

It provides a detailed sensitivity analysis of dark energy parameter constraints to halo model uncertainties across different survey configurations.

Findings

Higher mass limit surveys are more sensitive to mass-observable uncertainties.

Low mass limit surveys are more affected by mass function errors.

Current halo model uncertainties can degrade cosmological constraints by up to a factor of two.

Abstract

We perform a sensitivity study of dark energy constraints from galaxy cluster surveys to uncertainties in the halo mass function, bias and the mass-observable relation. For a set of idealized surveys, we evaluate cosmological constraints as priors on sixteen nuisance parameters in the halo modeling are varied. We find that surveys with a higher mass limit are more sensitive to mass-observable uncertainties while surveys with low mass limits that probe more of the mass function shape and evolution are more sensitive to mass function errors. We examine the correlations among nuisance and cosmological parameters. Mass function parameters are strongly positively (negatively) correlated with Omega_DE (w). For the mass-observable parameters, Omega_DE is most sensitive to the normalization and its redshift evolution while w is more sensitive to redshift evolution in the variance. While survey performance is limited mainly by mass-observable uncertainties, the current level of mass function error is responsible for up to a factor of two degradation in ideal cosmological constraints. For surveys that probe to low masses (10^13.5 h^-1 M_sun), even percent-level constraints on model nuisance parameters result in a degradation of ~ sqrt{2} (2) on Omega_DE (w) relative to perfect knowledge.