Mapping Spatially Varying Additive Biases in Cosmic Shear Data

TL;DR

This paper introduces a new method to map and identify spatially varying additive biases in cosmic shear data using autocorrelation discrepancy maps, validated with simulations and applied to real survey data.

Contribution

The paper presents a novel approach leveraging isotropy and anisotropy properties to detect and map additive biases in cosmic shear measurements.

Findings

Detected spatially varying additive biases of up to 0.002 in DES Year 1 data.

Validated the method with simulations showing accurate bias feature recovery.

Provides a tool for bias validation and modeling in cosmic shear analyses.

Abstract

In this paper we address the challenge of extracting maps of spatially varying unknown additive biases from cosmic shear data. This is done by exploiting the isotropy of the cosmic shear field, and the anisotropy of a typical additive bias field, using an autocorrelation discrepancy map; which identifies significant non-Gaussian components of the map. We test this approach using simulations and find that the autocorrelation discrepancy map produces spatially varying features that are indicative of the additive bias field both in amplitude and spatial variation. We then apply this to the Dark Energy Survey Year 1 data, and find evidence for spatially varying additive biases of at most 0.002 on large-scales. The method can be used to empirically inform modelling of the spatially varying additive bias field in any cosmological parameter inference, and can act as a validation test for cosmic shear surveys.