EarthMesh: A Multi-Scale Unstructured Mesh Generation Tool for Earth System Models

Variable-resolution meshes (VRMs) on unstructured grids direct resolution to regions where local processes need finer representation, but producing solver-ready meshes across Earth system model (ESM) components remains difficult. Existing tools typically focus on a single component, leaving users to write post-processing scripts for each solver. Here we present EarthMesh, a workflow for generating triangular and dual-polygonal meshes on the sphere and preparing them for atmospheric, oceanic, and land-surface models. EarthMesh starts with an icosahedral triangular mesh, refines cells using user-defined regions, physical criteria such as land cover types and leaf area index (LAI), or both, and reconstructs conforming transition rows while keeping vertex valence between 5 and 7. A modified spring-dynamics optimization reduces the number of obtuse triangles within transition areas. Model-specific drivers then produce the input files required by MPAS-A, FVCOM, and CoLM2024. In idealized MPAS-A baroclinic-wave and FVCOM geostrophic-eddy tests, the VRMs reproduce the fine-scale structures of uniform high-resolution runs within the refined regions, without spurious signals at the transition zones, while using about 8 % and 44–47 % of the high-resolution cell count, respectively. In a CoLM2024 runoff experiment over the eastern United States, refinement driven by LAI variability and land-cover diversity improves monthly runoff skill over 55 % of the domain (62 % for uniform refinement) with 28 % of the high-resolution cells. These results provide practical guidance on transition width, refinement criteria, and dual-mesh construction for different ESM components.

Authors

Institutions

Publication Details

Published
2026-11-19
DOI
https://doi.org/10.5194/egusphere-2026-6875
Primary Topic
Computational Geometry and Mesh Generation
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

EarthMesh: A Multi-Scale Unstructured Mesh Generation Tool for Earth System Models

L.C. Li, Qingchen Xu, Xingjie Lu, Wei Shangguan et al.
Computational Geometry and Mesh Generation
preprint

EarthMesh: A Multi-Scale Unstructured Mesh Generation Tool for Earth System Models

L.C. Li, Qingchen Xu, Xingjie Lu, Wei Shangguan, Xian‐Xiang Li, Shaofeng Liu, Huan Yuan, Shupeng Zhang, Yinhai Luo, Zhongwang Wei, Hanwen Fan, Nan Wei, Yongjiu Dai, Rui Zhang
preprint en

Abstract

Variable-resolution meshes (VRMs) on unstructured grids direct resolution to regions where local processes need finer representation, but producing solver-ready meshes across Earth system model (ESM) components remains difficult. Existing tools typically focus on a single component, leaving users to write post-processing scripts for each solver. Here we present EarthMesh, a workflow for generating triangular and dual-polygonal meshes on the sphere and preparing them for atmospheric, oceanic, and land-surface models. EarthMesh starts with an icosahedral triangular mesh, refines cells using user-defined regions, physical criteria such as land cover types and leaf area index (LAI), or both, and reconstructs conforming transition rows while keeping vertex valence between 5 and 7. A modified spring-dynamics optimization reduces the number of obtuse triangles within transition areas. Model-specific drivers then produce the input files required by MPAS-A, FVCOM, and CoLM2024. In idealized MPAS-A baroclinic-wave and FVCOM geostrophic-eddy tests, the VRMs reproduce the fine-scale structures of uniform high-resolution runs within the refined regions, without spurious signals at the transition zones, while using about 8 % and 44–47 % of the high-resolution cell count, respectively. In a CoLM2024 runoff experiment over the eastern United States, refinement driven by LAI variability and land-cover diversity improves monthly runoff skill over 55 % of the domain (62 % for uniform refinement) with 28 % of the high-resolution cells. These results provide practical guidance on transition width, refinement criteria, and dual-mesh construction for different ESM components.

Sun Yat-sen University (CN)
Computational Geometry and Mesh Generation
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon