Probing early structure and model-independent neutrino mass with high-redshift CMB lensing mass maps

TL;DR

This paper proposes a model-insensitive method to isolate high-redshift mass maps from CMB lensing data, enabling precise, model-independent neutrino mass measurements and high-redshift structure growth studies.

Contribution

It introduces a novel low-redshift removal technique for CMB lensing maps, allowing high-redshift structure analysis and neutrino mass constraints that are robust against dark energy model assumptions.

Findings

High-$z$ mass maps can probe structure growth at $z>3.75$ with 2.3% accuracy.

Model-independent neutrino mass constraints are achievable with combined CMB-S4 and LSST data.

The method effectively nulls low-redshift contributions, reducing systematic biases.

Abstract

CMB lensing maps probe the mass distribution in projection out to high redshifts, but significant sensitivity to low-redshift structure remains. In this paper we discuss a method to remove the low-redshift contributions from CMB lensing mass maps by subtracting suitably scaled galaxy density maps, nulling the low redshift structure with a model-insensitive procedure that is similar to delensing. This results in a high-$z$-only mass map that can provide a probe of structure growth at uniquely high redshifts: if systematics can be controlled, we forecast that CMB-S4 lensing combined with a Rubin-LSST-like galaxy survey can probe the amplitude of structure at redshifts $z>3.75$ ($z>5$) to within $2.3\%$ ($3.3\%$). We then discuss other example applications of such high-$z$ CMB lensing maps. In standard analyses of CMB lensing, assuming the wrong dark energy model (or wrong model parametrization) can lead to biases in neutrino mass constraints. In contrast, we show with forecasts that a high-$z$ mass map constructed from CMB-S4 lensing and LSST galaxies can provide a nearly model-independent neutrino mass constraint, with only negligible sensitivity to the presence of non-standard dark energy models, irrespective of their parametrization.