Research map / Probabilistic and scientific modeling / Differential equations and PDEs
PhyOceanCast: Global Ocean Forecasting with Physics-Informed Diffusion
Qixiu Li, Xiang Zhu, Xiaoyong Li, Xiaolong Xu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use physics-informed diffusion (PhyOceanCast) instead of plain diffusion backbone DiffCast | diffusion forecaster · GLORYS12V1, 30-day lead | RMSE: 4.8514 -> 1.1109 | Improved |
| use PhyOceanCast instead of best deterministic baseline (Video Swin Transformer) | global ocean forecasting · GLORYS12V1, 3-day lead | RMSE: 0.5358 -> 0.4558 | Improved |
| use PhyOceanCast instead of 3D-Geoformer | global ocean forecasting · GLORYS12V1, 7-day lead | RMSE: 0.8863 -> 0.7205 | Improved |
| use PhyOceanCast instead of GraphCast | global ocean forecasting · GLORYS12V1, 30-day lead | RMSE: 1.7577 -> 1.1109 | Improved |
| gain over best baseline shrinks at longer lead time for SSIM and Pearson at 3 days | PhyOceanCast · GLORYS12V1, 3-day lead | Pearson: +0.05% | No clear effect |
| remove SGAN-MOC (add spherical graph attention coupling) (ablation) | PhyOceanCast · GLORYS12V1, 7-day lead | RMSE and SSIM | Improved |
| add PWTC wavelet and advection-diffusion module (ablation) | PhyOceanCast · GLORYS12V1, 7-day lead | RMSE and SSIM | Improved |
| use 10 ensemble members instead of 3 (ablation) | PhyOceanCast · GLORYS12V1, 7-day lead | RMSE and SSIM | Improved |
Closest papers by meaning
- OceanCBM: A Concept Bottleneck Model for Mechanistic Interpretability in Ocean Forecasting (NeurIPS 2026 · similarity 0.89)
- FuXi-Ocean: A Global Ocean Forecasting System with Sub-Daily Resolution (NeurIPS 2025 · similarity 0.87)
- Njord: A Probabilistic Graph Neural Network for Ensemble Ocean Forecasting (NeurIPS 2026 · similarity 0.87)
- ClimODE: Climate and Weather Forecasting with Physics-informed Neural ODEs (ICLR 2024 · similarity 0.86)
- Probablistic Emulation of a Global Climate Model with Spherical DYffusion (NeurIPS 2024 · similarity 0.86)
- OmniCast: A Masked Latent Diffusion Model for Weather Forecasting Across Time Scales (NeurIPS 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
