General Requirements on Matter Power Spectrum Predictions for Cosmology with Weak Lensing Tomography

TL;DR

This paper analyzes the calibration requirements for matter power spectrum predictions in future weak lensing surveys, emphasizing the importance of combining galaxy clustering data to improve dark energy constraints and reduce calibration demands.

Contribution

It provides a new analysis of calibration requirements considering photo-z uncertainty and demonstrates the benefits of joint weak lensing and galaxy clustering analyses.

Findings

Photo-z uncertainty tightens power spectrum prediction requirements.

Joint analysis with galaxy clustering improves constraints by ~20%.

Clustering data relaxes photo-z calibration demands.

Abstract

Forthcoming projects such as DES, LSST, WFIRST, and Euclid aim to measure weak lensing shear correlations with unprecedented precision, constraining the dark energy equation of state at the percent level. Reliance on photometrically-determined redshifts constitutes a major source of uncertainty for these surveys. Additionally, interpreting the weak lensing signal requires a detailed understanding of the nonlinear physics of gravitational collapse. We present a new analysis of the stringent calibration requirements for weak lensing analyses of future imaging surveys that addresses both photo-z uncertainty and errors in the calibration of the matter power spectrum. We find that when photo-z uncertainty is taken into account the requirements on the level of precision in the prediction for the matter power spectrum are more stringent than previously thought. Including degree-scale galaxy clustering statistics in a joint analysis with weak lensing not only strengthens the survey's constraining power by ~20%, but can also have a profound impact on the calibration demands, decreasing the degradation in dark energy constraints with matter power spectrum uncertainty by a factor of 2-5. Similarly, using galaxy clustering information significantly relaxes the demands on photo-z calibration. We compare these calibration requirements to the contemporary state-of-the-art in photometric redshift estimation and predictions of the power spectrum and suggest strategies to utilize forthcoming data optimally.