用图神经网络实现高分辨率海域中短期海洋预报，速度快且精度高。
Regional Ocean Forecasting with Hierarchical Graph Neural Networks
- 基于图结构建模复杂海区网格，融合区域气象强迫数据。
- 在地中海高分辨率场景下，预报性能接近官方数值模型。
- 适合需要快速海洋预测的科研与环保机构使用。
精确的海洋预报系统对理解海洋动力学至关重要，有助于环境管理与气候适应策略。传统数值求解器虽有效但计算成本高、耗时长。机器学习近期已革新天气预报，提供快速节能的替代方案。我们提出 SeaCast，一种用于高分辨率、中短期海洋预报的神经网络。该方法采用图结构框架，有效处理海洋网格的复杂几何特征，并集成针对区域海洋情境定制的外部强迫数据。通过在地中海高分辨率场景下的实验验证，使用来自欧洲哥白尼海洋服务的运营数值模型，以及数值和数据驱动的气象强迫数据，证明了其有效性。
原文摘要 · Abstract (English)
Accurate ocean forecasting systems are vital for understanding marine dynamics, which play a crucial role in environmental management and climate adaptation strategies. Traditional numerical solvers, while effective, are computationally expensive and time-consuming. Recent advancements in machine learning have revolutionized weather forecasting, offering fast and energy-efficient alternatives. Building on these advancements, we introduce SeaCast, a neural network designed for high-resolution, medium-range ocean forecasting. SeaCast employs a graph-based framework to effectively handle the complex geometry of ocean grids and integrates external forcing data tailored to the regional ocean context. Our approach is validated through experiments at a high spatial resolution using the operational numerical model of the Mediterranean Sea provided by the Copernicus Marine Service, along with both numerical and data-driven atmospheric forcings.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。