Bayesian inference of dark matter voids in galaxy surveys

TL;DR

This paper uses Bayesian inference with the BORG algorithm to reconstruct the initial and present dark matter density fields from galaxy survey data, enabling more accurate void identification and reducing uncertainties in large-scale structure analysis.

Contribution

It introduces a method to infer physical dark matter density fields from galaxy surveys and generates improved void catalogs by leveraging these reconstructions.

Findings

Reconstructed initial density fields at early universe epoch.

Generated constrained realizations of current dark matter distribution.

Produced void catalogs with significantly reduced statistical uncertainties.

Abstract

We apply the BORG algorithm to the Sloan Digital Sky Survey Data Release 7 main sample galaxies. The method results in the physical inference of the initial density field at a scale factor $a~=~10^{-3}$, evolving gravitationally to the observed density field at a scale factor $a~=~1$, and provides an accurate quantification of corresponding uncertainties. Building upon these results, we generate a set of constrained realizations of the present large-scale dark matter distribution. As a physical illustration, we apply a void identification algorithm to them. In this fashion, we access voids defined by the inferred dark matter field, not by galaxies, greatly alleviating the issues due to the sparsity and bias of tracers. In addition, the use of full-scale physical density fields yields a drastic reduction of statistical uncertainty in void catalogs. These new catalogs are enhanced data sets for cross-correlation with other cosmological probes.