arXiv：cs.LG· Guowen Li, Yang Liu, Mengxuan Chen, Jiashun Cheng, Haoyuan Liang, Shilei Cao, Yujie Wang, Juepeng Zheng, Fang Wang, Yerong Feng, Meng Jin, Hong Cheng, Yutong Lu, Haohuan Fu·· 13 小时前
基于扩散模型的天气预报中期降水生成框架
Diffusion Precipitation Generation with Medium-range Weather Forecasting
AI 导读
研究提出 Dynamics-Precipitation Decoupling 即插即用框架，用单独训练的扩散生成器将冻结天气模型的非降水预报映射为与观测对齐的日降水量。
正文
Authors:Guowen Li, Yang Liu, Mengxuan Chen, Jiashun Cheng, Haoyuan Liang, Shilei Cao, Yujie Wang, Juepeng Zheng, Fang Wang, Yerong Feng, Meng Jin, Hong Cheng, Yutong Lu, Haohuan Fu
Abstract:Accurate daily precipitation forecasts are important for weather-sensitive decisions. Although recent Weather Foundation Models (WFMs) have achieved significant progress, they are trained on reanalysis data, which exhibit regional and intensity-dependent biases. Since retraining WFMs on observation-based targets is costly, we study Dynamics-Precipitation Decoupling, a plug-and-play framework that uses a separately trained generator to map non-precipitation forecasts from a frozen weather model to observation-aligned daily precipitation. Its key challenges are to capture heterogeneous rainfall distributions and align daily accumulations with atmospheric signals that evolve across space and time. To address them, we train a diffusion generator conditional on atmospheric states and satellite-gauge estimates of precipitation. A multi-scale recurrent encoder summarizes atmospheric evolution, while cross-attention aligns atmospheric and precipitation features during denoising. Extensive evaluation of various backbones, including Pangu-Weather, GraphCast, and HRES, on global and regional forecasting shows improved daily precipitation accuracy and event detection, with better representation of localized accumulation patterns. Global ensemble evaluation shows lower Brier scores than ENS and training-from-scratch GenCast at all reported lead times, with lower CRPS in most comparisons. Remarkably, a heavy-rainfall case study using 2,288 stations in China further supports improved daily accumulation patterns and event detection against ground observations.
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.12976 [cs.LG] |
| (or arXiv:2610.12976v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.12976 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Guowen Li [view email]
[v1]
Fri, 9 Oct 2026 04:21:30 UTC (16,733 KB)
来源：arXiv：cs.LG · arxiv.org