Weak lensing power spectrum reconstruction by counting galaxies.-- II: Improving the ABS method with the shift parameter

TL;DR

This paper introduces an improved version of the ABS method for weak lensing power spectrum reconstruction, incorporating a shift parameter that enhances performance and accuracy, potentially enabling precise measurements in upcoming surveys.

Contribution

The paper presents a new ABS method with a shift parameter that significantly improves weak lensing power spectrum reconstruction accuracy over the original method.

Findings

The new ABS method outperforms the previous version in all tested scenarios.

It can potentially achieve 1% accuracy in lensing power spectrum at redshift ~1.

The method is promising for application in surveys like LSST and SKA.

Abstract

In paper I of this series (Yang et al. 2017, ApJ), we proposed an analytical method of blind separation ({\bf ABS}) to extract the cosmic magnification signal in galaxy number distribution and reconstruct the weak lensing power spectrum. Here we report a new version of the ABS method, with significantly improved performance. This version is characterized by a shift parameter $\mathcal{S}$, with the special case of $\mathcal{S}=0$ corresponding to the original ABS method. We have tested this new version, compared to the previous one, and confirmed its supreme performance in all investigated situations. Therefore it supersedes the previous version. The proof of concept studies presented in this paper demonstrate that it may enable surveys such as LSST and SKA to reconstruct the lensing power spectrum at $z\simeq 1$ with $1\%$ accuracy. We will test with more realistic simulations to verify its applicability in real data.