用机器学习提升12小时云量预报精度，融合卫星观测与数值天气模型。
From Nowcasting to Forecasting: Adapting a Reanalysis-Trained
- 先用再分析数据学云演变规律，再通过条件流匹配适配卫星云图。
- 相比前代模型，1-12小时均方误差降低10%，3-6小时空间一致性超越前代。
- 适合需要高精度短期云量预测的光伏调度与气象预报场景。
准确的云量预报对温度预测、辐射估算和太阳能发电运营至关重要。短时预报方法可在前几小时保持观测到的云分布，但随云系生成、消散和形变，其精度下降。更长预报需考虑大气演变，但业务数值天气预报（NWP）在初始化时可能无法准确反映卫星观测的云状态。我们提出CloudCast v2，一种基于观测初始条件的12小时云量预测机器学习模型。模型首先在Copernicus欧洲区域再分析数据（Ridal2024）上训练以学习云演变动力学，再通过条件流匹配（Lipman2023）方法，将噪声转化为依赖于观测云场和NWP输入的云量预测。CloudCast v2在1-12小时范围内相较前代模型CloudCast v1（Partio2025）均方误差降低10%。在不同云量等级下，约3-6小时后，其分数技能得分（邻域空间一致性指标）超过v1。结果表明，基于观测初始化的机器学习预报可突破常规1-3小时短临预报范围，同时保留卫星云图的空间细节。
原文摘要 · Abstract (English)
Accurate cloud-cover forecasts are important for temperature prediction, radiation forecasting, and solar-power operations. Short-range forecasting methods can preserve observed cloud placement during the first forecast hours, but their skill decreases when cloud fields evolve through formation, dissipation and deformation. Longer lead times require accounting for atmospheric evolution, but operational numerical weather prediction (NWP) forecasts may not accurately represent the satellite-observed cloud state at initialization. We develop CloudCast v2, a machine-learning model for 12-hour cloud-cover forecasting from observation-based initial conditions. The model is first trained on the Copernicus European Regional Reanalysis (Ridal2024) to learn cloud-evolution dynamics, and is then adapted to satellite-derived cloud fields using conditional flow matching (Lipman2023), a generative method that transforms noise into cloud-cover forecasts conditioned on the observed initial cloud fields and NWP inputs. CloudCast v2 reduces mean absolute error by 10% relative to its predecessor, CloudCast v1 (Partio2025), over the 1-12 h range. It also overtakes CloudCast v1 in fractions skill score, a neighborhood-based measure of spatial agreement, after approximately 3-6 h, depending on the cloudiness category. These results show that observation-initialized machine-learning forecasts can extend beyond the usual 1-3-hour nowcasting range while retaining spatial detail from satellite cloud fields.
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。