HClimRep-Ocean: A Global Ocean Emulator on an Unstructured Mesh
Abstract
Machine-learning (ML) emulators for atmospheric processes have advanced rapidly in recent years, transforming weather forecasting. Although early ML ocean forecasting models now exist, they remain less developed than their atmospheric counterparts. Unlike the atmosphere, much of the ocean's kinetic energy resides in mesoscale eddies whose characteristic spatial scales are approximately an order of magnitude smaller than those of comparable atmospheric features. Moreover, complex coastlines, narrow straits, and ice-covered seas make boundary representation a central challenge that atmospheric models do not face. Consequently, numerical ocean simulations commonly use locally refined or even completely unstructured meshes. However, their data-driven counterparts have so far been built around latitude-longitude grids. We present HClimRep-Ocean, an ocean emulator that operates directly on the native unstructured mesh of FESOM2. The emulator is trained on a 209-year AWI-CM3 control integration and is run without atmospheric forcing, receiving the atmospheric state only at initialisation time, which isolates the predictability carried by the ocean state itself. Skill is strongly field-dependent: for currents, HClimRep-Ocean outperforms every reference at 30 day forecast, whereas for temperature and salinity a damped-anomaly persistence forecast remains the more accurate estimator. This behaviour is physically interpretable: current variability is largely geostrophic and internally generated, whereas sea-surface temperature and salinity fluctuations are driven by atmospheric forcing through weather state. Evaluated independently on the OceanBench benchmark, a reanalysis-trained variant of HClimRep-Ocean achieves the lowest RMSE against GLORYS reanalysis among all assessed systems, confirming the competitiveness of the native-mesh approach.
Related
- Ocean-Aware Sea-ice Emulator for Future Hybrid Coupled Prediction Systems
- DeepMelt-GL v1: a neural network emulator of sub-shelf melt rates for the unrepresented regions of ice-shelf cavities in ocean models
- An unstructured-grid, nonhydrostatic, GVC ocean model Part I: Model description and application of vertical hybrid coordinates to internal solitary waves
- FloeNet: A Mass‐Conserving Global Sea Ice Emulator That Generalizes Across Climates
- Extreme events and statistics in an AI emulator of a global storm-resolving model
BibTeX
@misc{nowak2026hclimrepocean,
title = {{HClimRep-Ocean: A Global Ocean Emulator on an Unstructured Mesh}},
author = {Kacper Nowak and Aleksei Koldunov and Nikolay Koldunov and Savvas Melidonis and Ankit Patnala and Simon Grasse and Julius Polz and Christian Lessig and Martin Schultz and Thomas Jung},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.28601},
eprint = {2609.28601},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.28601},
}
RIS
TY - UNPB
TI - HClimRep-Ocean: A Global Ocean Emulator on an Unstructured Mesh
AU - Kacper Nowak
AU - Aleksei Koldunov
AU - Nikolay Koldunov
AU - Savvas Melidonis
AU - Ankit Patnala
AU - Simon Grasse
AU - Julius Polz
AU - Christian Lessig
AU - Martin Schultz
AU - Thomas Jung
PY - 2026
DA - 2026/09/28
PB - arXiv
DO - 10.48550/arxiv.2609.28601
UR - https://arxiv.org/abs/2609.28601
AB - Machine-learning (ML) emulators for atmospheric processes have advanced rapidly in recent years, transforming weather forecasting. Although early ML ocean forecasting models now exist, they remain less developed than their atmospheric counterparts. Unlike the atmosphere, much of the ocean's kinetic energy resides in mesoscale eddies whose characteristic spatial scales are approximately an order of magnitude smaller than those of comparable atmospheric features. Moreover, complex coastlines, narrow straits, and ice-covered seas make boundary representation a central challenge that atmospheric models do not face. Consequently, numerical ocean simulations commonly use locally refined or even completely unstructured meshes. However, their data-driven counterparts have so far been built around latitude-longitude grids. We present HClimRep-Ocean, an ocean emulator that operates directly on the native unstructured mesh of FESOM2. The emulator is trained on a 209-year AWI-CM3 control integration and is run without atmospheric forcing, receiving the atmospheric state only at initialisation time, which isolates the predictability carried by the ocean state itself. Skill is strongly field-dependent: for currents, HClimRep-Ocean outperforms every reference at 30 day forecast, whereas for temperature and salinity a damped-anomaly persistence forecast remains the more accurate estimator. This behaviour is physically interpretable: current variability is largely geostrophic and internally generated, whereas sea-surface temperature and salinity fluctuations are driven by atmospheric forcing through weather state. Evaluated independently on the OceanBench benchmark, a reanalysis-trained variant of HClimRep-Ocean achieves the lowest RMSE against GLORYS reanalysis among all assessed systems, confirming the competitiveness of the native-mesh approach.
ER -