Self-Calibration Technique for 3-point Intrinsic Alignment Correlations in Weak Lensing Surveys

TL;DR

This paper extends a self-calibration method to reduce intrinsic alignment contamination in weak lensing surveys, specifically addressing the bispectrum and GGI correlations, achieving significant contamination reduction.

Contribution

It introduces an expansion of the self-calibration technique to the bispectrum, enabling estimation and removal of GGI contamination in weak lensing data.

Findings

Potential to reduce IA contamination by a factor of 5-10

Effective for non-adjacent redshift bins, reducing contamination to percent level

Allows for isolation of 2-point and 3-point IA signals

Abstract

The intrinsic alignment (IA) of galaxies has been shown to be a significant barrier to precision cosmic shear measurements. (Zhang, 2010, ApJ, 720, 1090) proposed a self-calibration technique for the power spectrum to calculate the induced gravitational shear-galaxy intrinsic ellipticity correlation (GI) in weak lensing surveys with photo-z measurements which is expected to reduce the IA contamination by at least a factor of 10 for currently proposed surveys. We confirm this using an independent analysis and propose an expansion to the self-calibration technique for the bispectrum in order to calculate the dominant IA gravitational shear-gravitational shear-intrinsic ellipticity correlation (GGI) contamination. We first establish an estimator to extract the galaxy density-density-intrinsic ellipticity (ggI) correlation from the galaxy ellipticity-density-density measurement for a photo-z galaxy sample. We then develop a relation between the GGI and ggI bispectra, which allows for the estimation and removal of the GGI correlation from the cosmic shear signal. We explore the performance of these two methods, compare to other possible sources of error, and show that the GGI self-calibration technique can potentially reduce the IA contamination by up to a factor of 5-10 for all but a few bin choices, thus reducing the contamination to the percent level. The self-calibration is less accurate for adjacent bins, but still allows for a factor of three reduction in the IA contamination. The self-calibration thus promises to be an efficient technique to isolate both the 2-point and 3-point intrinsic alignment signals from weak lensing measurements.