{
  "abstract": "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.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2611.02460",
  "author_total": 7,
  "authors": [
    {
      "name": "Ruiqi Shu"
    },
    {
      "name": "Guangliang Liu"
    },
    {
      "name": "Ruijian Gou"
    },
    {
      "name": "Tianrun Gao"
    },
    {
      "name": "Haomin Wen"
    },
    {
      "name": "Hao Li"
    },
    {
      "name": "Xiaomeng Huang"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/dya6pyjgd7ptf.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2611.02460",
  "fulltext_url": null,
  "id": "dya6pyjgd7ptf",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2611.02460",
  "language": null,
  "licence": "CC-BY-4.0",
  "licence_label": "CC BY 4.0",
  "licence_source": "article",
  "licence_url": "http://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": "CC0-1.0",
  "native_subjects": [
    [
      "arxiv",
      "physics.ao-ph"
    ]
  ],
  "openalex": null,
  "pdf_url": "https://arxiv.org/pdf/2611.02460",
  "pmcid": null,
  "pmid": null,
  "published": "2026-11-02",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/dya6pyjgd7ptf.ris",
  "sources": [
    {
      "first_seen": "2026-11-03T12:20:59Z",
      "native_id": "2611.02460",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2611.02460"
    }
  ],
  "subjects": [
    {
      "field": "earth",
      "origin": "arxiv:physics.ao-ph",
      "subfield": "atmospheric"
    }
  ],
  "title": "New Wine in an Old Bottle -- NeuralPOM: A Hybrid Ocean Forecast Model with Differentiable Dynamics and Staged Learning",
  "title_text": "New Wine in an Old Bottle -- NeuralPOM: A Hybrid Ocean Forecast Model with Differentiable Dynamics and Staged Learning",
  "type": "preprint",
  "updated_at": "2026-11-03T12:20:59Z",
  "url": "https://research.draconitic.com/paper/dya6pyjgd7ptf",
  "venue": "arXiv",
  "versions": []
}