Straightening the Density-Displacement Relation with a Logarithmic Transform

TL;DR

This paper demonstrates that using a logarithmic density variable improves the estimation of the Lagrangian displacement field in cosmological simulations, especially at nonlinear scales and low redshifts.

Contribution

It introduces and tests a logarithmic density transformation for better relation with displacement divergence, outperforming linear methods in certain regimes.

Findings

Logarithmic density relation is tighter at low redshifts.

Grid-based methods outperform tessellation in measuring fields.

Logarithmic transformation improves nonlinear scale estimations.

Abstract

We investigate the use of a logarithmic density variable in estimating the Lagrangian displacement field, motivated by the success of a logarithmic transformation in restoring information to the matter power spectrum. The logarithmic relation is an extension of the linear relation, motivated by the continuity equation, in which the density field is assumed to be proportional to the divergence of the displacement field; we compare the linear and logarithmic relations by measuring both of these fields directly in a cosmological N-body simulation. The relative success of the logarithmic and linear relations depends on the scale at which the density field is smoothed. Thus we explore several ways of measuring the density field, including Cloud-In-Cell smoothing, adaptive smoothing, and the (scale-independent) Delaunay tessellation, and we use both a Fourier space and a geometrical tessellation approach to measuring the divergence. We find that the relation between the divergence of the displacement field and the density is significantly tighter with a logarithmic density variable, especially at low redshifts and for very small (~2 Mpc/h) smoothing scales. We find that the grid-based methods are more reliable than the tessellation-based method of calculating both the density and the divergence fields, though in both cases the logarithmic relation works better in the appropriate regime, which corresponds to nonlinear scales for the grid-based methods and low densities for the tessellation-based method.