Metraipolis › archive › dc_440bc202e891
New Wine in an Old Bottle -- NeuralPOM: A Hybrid Ocean Forecast Model with Differentiable Dynamics and Staged Learning — arXiv:2611.02460
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.02460
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-03T11:50:37.226Z
- SHA-256 of the archived text
49a63f8aac8cc4327ee8054edacb8c32c01a45cd2bf01cea7004e888331c29b1- Stable URL of this page
https://metraipolis.com/documento/dc_440bc202e891
How to cite
New Wine in an Old Bottle -- NeuralPOM: A Hybrid Ocean Forecast Model with Differentiable Dynamics and Staged Learning — arXiv:2611.02460 — arXiv (en), https://arxiv.org/abs/2611.02460 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_440bc202e891 (Metraipolis archive dc_440bc202e891, sha256 49a63f8aac8cc4327ee8054edacb8c32c01a45cd2bf01cea7004e888331c29b1)
BibTeX
@misc{dc440bc202e891,
title = {New Wine in an Old Bottle -- NeuralPOM: A Hybrid Ocean Forecast Model with Differentiable Dynamics and Staged Learning — arXiv:2611.02460},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.02460},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_440bc202e891; sha256 49a63f8aac8cc4327ee8054edacb8c32c01a45cd2bf01cea7004e888331c29b1},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.02460. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Coarse-grid ocean circulation models rely on empirical subgrid parameterizations that struggle to capture mesoscale eddies, whereas purely data-driven neural forecasters accumulate trajectory drift, violate tracer conservation, and damp small-scale variability over extended integrations. Here we present NeuralPOM, a hybrid ocean forecast model that couples differentiable explicit mode-split and semi-implicit free-surface implementations of the Princeton Ocean Model (POM) to a conservation-constrained neural corrector. The three-dimensional primitive-equation solver advances the ocean state, while the neural network periodically updates the physical fields at high-frequency (1 h) or low-frequency (1 day) intervals under volume, heat, and salt conservation constraints. Using a fine-grid reference dataset exceeding 100,000 model days in a wind-driven double-gyre basin, we introduce a three-stage training procedure comprising offline single-step pretraining, multi-step differentiable training through the ocean solver, and multi-day closed-loop rollout training. Over 100-day forecasts, NeuralPOM reduces errors across all prognostic fields relative to coarse-grid physical closures and direct neural forecasters, preserving mesoscale kinetic energy while improving eddy phase alignment. Controlled ablations show that closed-loop training suppresses multi-day drift, whereas backpropagating gradients through the differentiable solver stabilizes high-frequency coupling. Finally, because high-frequency tracer updates can introduce density anomalies via the equation of state that excite spurious baroclinic pressure gradients, we introduce Density-Normal Damping (DND) to attenuate the density-modifying (diapycnal) tracer component while retaining the density-compensated (isopycnal) correction. Tema degli esploratori: terra, clima ed energia (elenco arXiv physics.ao-ph, nuovi lavori del giorno). Categorie arXiv: Atmospheric and Oceanic Physics (physics.ao-ph). Pagina del lavoro: https://arxiv.org/abs/2611.02460 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-11-03T11:11:37.284Z dalla pagina https://arxiv.org/list/physics.ao-ph/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).