Beyond the 3rd moment: A practical study of using lensing convergence CDFs for cosmology with DES Y3

TL;DR

This study demonstrates that the cumulative distribution function (CDF) of galaxy lensing convergence provides a modestly improved summary statistic over moments, capturing more information for cosmological analysis, with practical considerations for observational systematics.

Contribution

It introduces the use of CDFs at multiple scales as an effective summary statistic for lensing convergence, and evaluates their practical application with DES Y3 data, including systematic effects.

Findings

CDFs outperform moments in constraining power.

Systematics like PSF and shear approximation have minimal impact.

Noise correlations require careful modeling for accurate cosmology.

Abstract

Widefield surveys of the sky probe many clustered scalar fields -- such as galaxy counts, lensing potential, gas pressure, etc. -- that are sensitive to different cosmological and astrophysical processes. Our ability to constrain such processes from these fields depends crucially on the statistics chosen to summarize the field. In this work, we explore the cumulative distribution function (CDF) at multiple scales as a summary of the galaxy lensing convergence field. Using a suite of N-body lightcone simulations, we show the CDFs' constraining power is modestly better than that of the 2nd and 3rd moments of the field, as they approximately capture the information from all moments of the field in a concise data vector. We then study the practical aspects of applying the CDFs to observational data, using the first three years of the Dark Energy Survey (DES Y3) data as an example, and compute the impact of different systematics on the CDFs. The contributions from the point spread function are 2-3 orders of magnitude below the cosmological signal, while those from reduced shear approximation contribute $\lesssim 1\%$ to the signal. Source clustering effects and baryon imprints contribute $1-10\%$. Enforcing scale cuts to limit systematics-driven biases in parameter constraints degrades these constraints a noticeable amount, and this degradation is similar for the CDFs and the moments. We also detect correlations between the observed convergence field and the shape noise field at $13\sigma$. We find that the non-Gaussian correlations in the noise field must be modeled accurately to use the CDFs, or other statistics sensitive to all moments, as a rigorous cosmology tool.