Discrete Gaussian Vector Fields On Meshes

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Gillan, Michael, Siegert, Stefan, Youngman, Ben
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911079194427392
author Gillan, Michael
Siegert, Stefan
Youngman, Ben
author_facet Gillan, Michael
Siegert, Stefan
Youngman, Ben
contents Though the underlying fields associated with vector-valued environmental data are continuous, observations themselves are discrete. For example, climate models typically output grid-based representations of wind fields or ocean currents, and these are often downscaled to a discrete set of points. By treating the area of interest as a two-dimensional manifold that can be represented as a triangular mesh and embedded in Euclidean space, this work shows that discrete intrinsic Gaussian processes for vector-valued data can be developed from discrete differential operators defined with respect to a mesh. These Gaussian processes account for the geometry and curvature of the manifold whilst also providing a flexible and practical formulation that can be readily applied to any two-dimensional mesh. We show that these models can capture harmonic flows, incorporate boundary conditions, and model non-stationary data. Finally, we apply these models to downscaling stationary and non-stationary gridded wind data on the globe, and to inference of ocean currents from sparse observations in bounded domains.
format Preprint
id arxiv_https___arxiv_org_abs_2507_20024
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Discrete Gaussian Vector Fields On Meshes
Gillan, Michael
Siegert, Stefan
Youngman, Ben
Methodology
Statistics Theory
Machine Learning
Though the underlying fields associated with vector-valued environmental data are continuous, observations themselves are discrete. For example, climate models typically output grid-based representations of wind fields or ocean currents, and these are often downscaled to a discrete set of points. By treating the area of interest as a two-dimensional manifold that can be represented as a triangular mesh and embedded in Euclidean space, this work shows that discrete intrinsic Gaussian processes for vector-valued data can be developed from discrete differential operators defined with respect to a mesh. These Gaussian processes account for the geometry and curvature of the manifold whilst also providing a flexible and practical formulation that can be readily applied to any two-dimensional mesh. We show that these models can capture harmonic flows, incorporate boundary conditions, and model non-stationary data. Finally, we apply these models to downscaling stationary and non-stationary gridded wind data on the globe, and to inference of ocean currents from sparse observations in bounded domains.
title Discrete Gaussian Vector Fields On Meshes
topic Methodology
Statistics Theory
Machine Learning
url https://arxiv.org/abs/2507.20024