Phase-space structures I: A comparison of 6D density estimators

TL;DR

This paper compares various 6D phase-space density estimators, highlighting the advantages of local adaptive metric methods over traditional SPH and Delaunay tessellation, and discusses their effectiveness in identifying structures within halo profiles.

Contribution

It introduces and evaluates local adaptive metric methods for 6D phase-space density estimation, demonstrating their superiority over existing techniques like SPH and Delaunay tessellation.

Findings

Local adaptive metric methods outperform SPH in phase-space estimation.

Delaunay tessellation requires a global scaling and is sensitive to anisotropies.

The proxy Q is a rough approximation and misses detailed structures.

Abstract

This paper reviews and analyses methods used to identify neighbours in 6D space and estimate the corresponding phase-space density. It compares SPH methods to 6D Delaunay tessellation on statical and dynamical realisation of single halo profiles, paying attention to the unknown scaling, S_G, used to relate the spatial dimensions to the velocity dimensions. The methods with local adaptive metric provide the best phase-space estimators. They make use of a Shannon entropy criterion combined with a binary tree partitioning and with SPH interpolation using 10-40 neighbours. Local scaling implemented by such methods, which enforces local isotropy of the distribution function, can vary by about one order of magnitude in different regions within the system. It presents a bimodal distribution, in which one component is dominated by the main part of the halo and the other one is dominated by the substructures. While potentially better than SPH techniques, since it yields an optimal estimate of the local softening volume (and the local number of neighbours required to perform the interpolation), the Delaunay tessellation in fact poorly estimates the phase-space distribution function. Indeed, it requires, the choice of a global scaling S_G. We propose two methods to estimate S_G that yield a good global compromise. However, the Delaunay interpolation still remains quite sensitive to local anisotropies in the distribution. We also compare 6D phase-space density estimation with the proxy, Q=rho/sigma^3, where rho is the local density and 3 sigma^2 is the local 3D velocity dispersion. We show that Q only corresponds to a rough approximation of the true phase-space density, and is not able to capture all the details of the distribution in phase-space, ignoring, in particular, filamentation and tidal streams.