A new non-parametric method to infer galaxy cluster masses from weak lensing

TL;DR

This paper presents a fast, non-parametric method to determine the 3D mass profiles of galaxy clusters from weak lensing data, assuming spherical symmetry at larger radii, and demonstrates its effectiveness with real data.

Contribution

The paper introduces a novel, efficient non-parametric approach for inferring galaxy cluster mass profiles from weak lensing, applicable even with non-symmetric inner regions.

Findings

Method produces mass profiles consistent with literature.

Implementation runs in milliseconds per cluster.

Successfully applied to real KiDS DR4 data.

Abstract

We introduce a new, non-parametric method to infer deprojected 3D mass profiles $M(r)$ of galaxy clusters from weak gravitational lensing observations. The method assumes spherical symmetry and a moderately small convergence, $\kappa \lesssim 1$. The assumption of spherical symmetry is an important restriction, which is, however, quite common in practice, for example in methods that fit lensing data to an NFW profile. Moreover, with a mild assumption on the probability distributions of the source redshifts, our method relies on spherical symmetry only at radii larger than the radius $r$ at which the mass $M$ is inferred. That is, the method may be useful even for clusters with a non-symmetric inner region, since it correctly estimates the enclosed mass beyond the radius where spherical symmetry is restored. We discuss how to correct, statistically and approximately, for miscentering given that the probability distribution of miscentering offsets is known. We provide an efficient implementation in Julia code that runs in a few milliseconds per galaxy cluster. We explicitly demonstrate the method by using data from KiDS DR4 to infer mass profiles for two example clusters, Abell 1835 and Abell 2744, finding results consistent with existing literature.