StormInsight: Hierarchical Environmental Forcing and Vertical Coupling for Convective Systems Evolution

Jun Chen, Yan Fang, Minghui Qiu, Yueran Qiu, Lin Chen, Shuxin Zhong, Yu Zhang, Kaishun Wu
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:14145-14169, 2026.

Abstract

Nowcasting forms the first line of defense against rapidly evolving weather hazards, where even minutes of delay can lead to severe societal impacts. However, existing systems predominantly extrapolate 2D radar reflectivity, which struggles under rapid intensification regimes. We introduce StormInsight, a multi-scale modeling framework that enables coherent reconstruction of the three-dimensional evolution of convective systems while explicitly conditioning on the ambient environment. StormInsight integrates multi-source observations—including radar, satellite, and station—with reanalysis fields through two components: (i) Storm Evolution Encoder that explicitly disentangles convective system state form vertical thermodynamic coupling and large-scale environmental forcing; (ii) Convective System Decoder that predicts future echos by adaptively aggregating cross-layer interactions conditioned on evolving environmental conditions. To support comprehensive evaluation, we build a new benchmark StormBench that integrates observational and reanalysis data across regions. On this benchmark, StormInsight consistently achieves the best performance, reducing MAE by 12.4% and improving the mCSI by 34.0%.

Cite this Paper

BibTeX
@InProceedings{pmlr-v306-chen26aa, title = {{S}torm{I}nsight: Hierarchical Environmental Forcing and Vertical Coupling for Convective Systems Evolution}, author = {Chen, Jun and Fang, Yan and Qiu, Minghui and Qiu, Yueran and Chen, Lin and Zhong, Shuxin and Zhang, Yu and Wu, Kaishun}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {14145--14169}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/chen26aa/chen26aa.pdf}, url = {https://proceedings.mlr.press/v306/chen26aa.html}, abstract = {Nowcasting forms the first line of defense against rapidly evolving weather hazards, where even minutes of delay can lead to severe societal impacts. However, existing systems predominantly extrapolate 2D radar reflectivity, which struggles under rapid intensification regimes. We introduce StormInsight, a multi-scale modeling framework that enables coherent reconstruction of the three-dimensional evolution of convective systems while explicitly conditioning on the ambient environment. StormInsight integrates multi-source observations—including radar, satellite, and station—with reanalysis fields through two components: (i) Storm Evolution Encoder that explicitly disentangles convective system state form vertical thermodynamic coupling and large-scale environmental forcing; (ii) Convective System Decoder that predicts future echos by adaptively aggregating cross-layer interactions conditioned on evolving environmental conditions. To support comprehensive evaluation, we build a new benchmark StormBench that integrates observational and reanalysis data across regions. On this benchmark, StormInsight consistently achieves the best performance, reducing MAE by 12.4% and improving the mCSI by 34.0%.} }
Endnote
%0 Conference Paper %T StormInsight: Hierarchical Environmental Forcing and Vertical Coupling for Convective Systems Evolution %A Jun Chen %A Yan Fang %A Minghui Qiu %A Yueran Qiu %A Lin Chen %A Shuxin Zhong %A Yu Zhang %A Kaishun Wu %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-chen26aa %I PMLR %P 14145--14169 %U https://proceedings.mlr.press/v306/chen26aa.html %V 306 %X Nowcasting forms the first line of defense against rapidly evolving weather hazards, where even minutes of delay can lead to severe societal impacts. However, existing systems predominantly extrapolate 2D radar reflectivity, which struggles under rapid intensification regimes. We introduce StormInsight, a multi-scale modeling framework that enables coherent reconstruction of the three-dimensional evolution of convective systems while explicitly conditioning on the ambient environment. StormInsight integrates multi-source observations—including radar, satellite, and station—with reanalysis fields through two components: (i) Storm Evolution Encoder that explicitly disentangles convective system state form vertical thermodynamic coupling and large-scale environmental forcing; (ii) Convective System Decoder that predicts future echos by adaptively aggregating cross-layer interactions conditioned on evolving environmental conditions. To support comprehensive evaluation, we build a new benchmark StormBench that integrates observational and reanalysis data across regions. On this benchmark, StormInsight consistently achieves the best performance, reducing MAE by 12.4% and improving the mCSI by 34.0%.
APA
Chen, J., Fang, Y., Qiu, M., Qiu, Y., Chen, L., Zhong, S., Zhang, Y. & Wu, K.. (2026). StormInsight: Hierarchical Environmental Forcing and Vertical Coupling for Convective Systems Evolution. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:14145-14169 Available from https://proceedings.mlr.press/v306/chen26aa.html.

Related Material