Dark Energy from the log-transformed convergence field

TL;DR

This paper investigates how a log transform of the convergence field affects the precision of dark energy parameter measurements, finding benefits in noise-free cases but limited advantages when shape noise is present, and highlighting the importance of pixel size.

Contribution

It demonstrates the impact of log-transforming convergence fields on dark energy constraints using simulations, emphasizing the role of shape noise and pixel size.

Findings

Log transform improves information content in noise-free scenarios.

Shape noise reduces the benefits of the log transform.

Larger pixel sizes enhance the effectiveness of the log transformation.

Abstract

A logarithmic transform of the convergence field improves `the information content', ie., the overall precision associated with the measurement of the amplitude of the convergence power spectrum by improving the covariance matrix properties. The translation of this improvement in the information content to that in cosmological parameters, such as those associated with dark energy, requires knowing the sensitivity of the log-transformed field to those cosmological parameters. In this paper we use N-body simulations with ray tracing to generate convergence fields at multiple source redshifts as a function of cosmology. The gain in information associated with the log-transformed field does lead to tighter constraints on dark energy parameters, but only if shape noise is neglected. The presence of shape noise quickly diminishes the advantage of the log mapping, more quickly than we would expect based on the information content. With or without shape noise, using a larger pixel size allows for a more efficient log-transformation.