The Influence of Galaxy Formation Physics on Weak Lensing Tests of General Relativity

TL;DR

Upcoming weak lensing surveys can test general relativity effectively, but must account for baryonic physics uncertainties; internal calibration methods improve constraints significantly over traditional approaches.

Contribution

The paper demonstrates that internal calibration of halo structures can mitigate baryonic uncertainties, enhancing the ability to test GR with future weak lensing data.

Findings

Baryonic physics can bias GR tests if neglected.

Parametrized models can correct for baryonic effects.

Internal calibration improves constraints by up to 70%."

Abstract

Forthcoming projects such as the DES, a JDEM, and LSST, aim to measure weak lensing shear correlations with unprecedented accuracy. Weak lensing observables are sensitive to both the distance-redshift relation and the growth of structure in the Universe. If the cause of accelerated cosmic expansion is dark energy within general relativity (GR), both cosmic distances and structure growth are governed by the properties of dark energy. Consequently, one may use lensing to check for this consistency and test GR. After reviewing the phenomenology of such tests, we address one major challenge to such a program. The evolution of the baryonic component of the Universe is highly uncertain and can influence lensing observables, manifesting as modified structure growth for a fixed cosmic distance scale. Using two proposed methods, we show that one could be led to reject the null hypothesis of GR when it is the true theory if this uncertainty in baryonic processes is neglected. Recent simulations suggest that we can correct for baryonic effects using a parametrized model in which the halo mass-concentration relation is modified. The correction renders biases small compared to statistical uncertainties. We study the ability of future weak lensing surveys to constrain the internal structures of halos and test the null hypothesis of GR simultaneously. Compared to nulling information from small-scales to mitigate sensitivity to baryonic physics, this internal calibration program should provide limits on deviations from GR that are several times more constraining. Specifically, we find that limits on general relativity in the case of internal calibration are degraded by only ~30% or less compared to the case of perfect knowledge of nonlinear structure.