MW-Nowcast: Six-hour ensemble nowcasting of extreme precipitation
Abstract
Extending reliable nowcasting of extreme precipitation could provide critical additional time for warnings and emergency response during high-impact events such as flash floods. Radar-based generative machine-learning models have enabled skilful hyperlocal precipitation nowcasting, but accurate prediction of intense precipitation remains confined to the first few hours. Because storm-scale structure is predictable for longer than individual cells, a natural strategy is to predict that structure while generatively modelling only the uncertain local growth, decay, reorganisation and initiation of storms. Here we present Microsoft Weather Nowcast (MW-Nowcast), a six-hour ensemble radar nowcasting model that jointly learns a deterministic predictor to capture organised precipitation structure shared across ensemble members, and a generator to produce diverse local residuals around this shared prediction. Across independent test data from the United States, Europe and China, MW-Nowcast achieves higher detection skill than leading methods for heavy and extreme precipitation throughout the 6 h horizon. For the most intense rainfall, MW-Nowcast doubles the available warning time across all three regions, delivering 6 h forecasts with skill previously limited to 3 h for the leading generative baseline. A cost-loss decision analysis shows that MW-Nowcast retains substantial value for a broad range of applications even at 4-6 h, where alternative methods offer little benefit. These additional hours can give forecasters and emergency managers the time to warn and act before extreme rainfall strikes, helping to protect lives and property.
Related
- Tail Skill and Ensemble-Configuration Sensitivity in Six-Hour Precipitation Postprocessing
- IRENE: A Convolutional GRU Ensemble Model for Radar Precipitation Nowcasting over Italy
- Precipitation nowcasting model based on multi-source data fusion and efficient additive attention
- Diffusion-Based Refinement for Kilometer-Scale Probabilistic Precipitation Nowcasting
- Wet-Biased Data Pipelines Mislead the Training and Evaluation of Generative Precipitation Nowcasting
BibTeX
@misc{wang2026mwnowcast,
title = {{MW-Nowcast: Six-hour ensemble nowcasting of extreme precipitation}},
author = {Ning Wang and Zuliang Fang and Weixin Jin and Zhongjian Lv and Shuang Qin and Pengcheng Zhao and Siqi Xiang and Jiang Bian and Haoyi Xiong and Nan Guan and Bin Zhang and Liangjie Zhang and Denvy Deng and Qi Zhang and Matt Corey and Jitu Keshri and Sridhar Iyer and Hongyu Sun and Kit Thambiratnam and Jonathan Weyn and Richard E. Turner and Haiyu Dong},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.34836},
eprint = {2609.34836},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.34836},
}
RIS
TY - UNPB
TI - MW-Nowcast: Six-hour ensemble nowcasting of extreme precipitation
AU - Ning Wang
AU - Zuliang Fang
AU - Weixin Jin
AU - Zhongjian Lv
AU - Shuang Qin
AU - Pengcheng Zhao
AU - Siqi Xiang
AU - Jiang Bian
AU - Haoyi Xiong
AU - Nan Guan
AU - Bin Zhang
AU - Liangjie Zhang
AU - Denvy Deng
AU - Qi Zhang
AU - Matt Corey
AU - Jitu Keshri
AU - Sridhar Iyer
AU - Hongyu Sun
AU - Kit Thambiratnam
AU - Jonathan Weyn
AU - Richard E. Turner
AU - Haiyu Dong
PY - 2026
DA - 2026/09/28
PB - arXiv
DO - 10.48550/arxiv.2609.34836
UR - https://arxiv.org/abs/2609.34836
AB - Extending reliable nowcasting of extreme precipitation could provide critical additional time for warnings and emergency response during high-impact events such as flash floods. Radar-based generative machine-learning models have enabled skilful hyperlocal precipitation nowcasting, but accurate prediction of intense precipitation remains confined to the first few hours. Because storm-scale structure is predictable for longer than individual cells, a natural strategy is to predict that structure while generatively modelling only the uncertain local growth, decay, reorganisation and initiation of storms. Here we present Microsoft Weather Nowcast (MW-Nowcast), a six-hour ensemble radar nowcasting model that jointly learns a deterministic predictor to capture organised precipitation structure shared across ensemble members, and a generator to produce diverse local residuals around this shared prediction. Across independent test data from the United States, Europe and China, MW-Nowcast achieves higher detection skill than leading methods for heavy and extreme precipitation throughout the 6 h horizon. For the most intense rainfall, MW-Nowcast doubles the available warning time across all three regions, delivering 6 h forecasts with skill previously limited to 3 h for the leading generative baseline. A cost-loss decision analysis shows that MW-Nowcast retains substantial value for a broad range of applications even at 4-6 h, where alternative methods offer little benefit. These additional hours can give forecasters and emergency managers the time to warn and act before extreme rainfall strikes, helping to protect lives and property.
ER -