A new cosmic shear function: Optimised E-/B-mode decomposition on a finite interval

TL;DR

This paper introduces an optimized E-/B-mode decomposition method for cosmic shear analysis on a finite interval, significantly improving signal-to-noise and cosmological parameter estimation over previous techniques.

Contribution

It presents a new shear statistic optimized for cosmological applications, outperforming existing methods like the ring statistic and aperture-mass dispersion in S/N and FoM.

Findings

S/N increased by a factor of three on certain scales

FoM improved by a factor of two compared to previous methods

Method adaptable to various survey parameters and observational setups

Abstract

The decomposition of the cosmic shear field into E- and B-mode is an important diagnostic in weak gravitational lensing. However, commonly used techniques to perform this separation suffer from mode-mixing on very small or very large scales. We introduce a new E-/B-mode decomposition of the cosmic shear two-point correlation on a finite interval. This new statistic is optimised for cosmological applications, by maximising the signal-to-noise ratio (S/N) and a figure of merit (FoM) based on the Fisher matrix of the cosmological parameters Omega_m and sigma_8. We improve both S/N and FoM results substantially with respect to the recently introduced ring statistic, which also provides E-/B-mode separation on a finite angular range. The S/N (FoM) is larger by a factor of three (two) on angular scales between 1 and 220 arc minutes. In addition, it yields better results than for the aperture-mass dispersion <M_ap>^2, with improvements of 20% (10%) for S/N (FoM). Our results depend on the survey parameters, most importantly on the covariance of the two-point shear correlation function. Although we assume parameters according to the CFHTLS-Wide survey, our method and optimisation scheme can be applied easily to any given survey settings and observing parameters. Arbitrary quantities, with respect to which the E-/B-mode filter is optimised, can be defined, therefore generalising the aim and context of the new shear statistic.