Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2511.11090 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911412753793024 |
|---|---|
| author | Harris, Levi Chen, Tianlong |
| author_facet | Harris, Levi Chen, Tianlong |
| contents | Meteorological agencies around the world rely on real-time flood guidance to issue life-saving advisories and warnings. For decades traditional numerical weather prediction (NWP) models have been state-of-the-art for precipitation forecasting. However, physically-parameterized models suffer from a few core limitations: first, solving PDEs to resolve atmospheric dynamics is computationally demanding, and second, these methods degrade in performance at nowcasting timescales (i.e., 0-4 hour lead-times). Motivated by these shortcomings, recent work proposes AI-weather prediction (AI-WP) alternatives that learn to emulate analysis data with neural networks. While these data-driven approaches have enjoyed enormous success across diverse spatial and temporal resolutions, applications of video-understanding architectures for weather forecasting remain underexplored. To address these gaps, we propose \textbf{SaTformer}: a video transformer built on full space-time attention that skillfully forecasts extreme precipitation from satellite radiances. Along with our novel architecture, we introduce techniques to tame long-tailed precipitation datasets. Namely, we reformulate precipitation regression into a classification problem, and employ a class-weighted loss to address label imbalances. Our model scored \textbf{first place} on the NeurIPS Weather4Cast 2025 ``Cumulative Rainfall'' challenge. Code and model weights are available: \texttt{\href{github.com/leharris3/w4c-25}{github.com/leharris3/satformer}} |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_11090 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | A Space-Time Transformer for Precipitation Nowcasting Harris, Levi Chen, Tianlong Computer Vision and Pattern Recognition Meteorological agencies around the world rely on real-time flood guidance to issue life-saving advisories and warnings. For decades traditional numerical weather prediction (NWP) models have been state-of-the-art for precipitation forecasting. However, physically-parameterized models suffer from a few core limitations: first, solving PDEs to resolve atmospheric dynamics is computationally demanding, and second, these methods degrade in performance at nowcasting timescales (i.e., 0-4 hour lead-times). Motivated by these shortcomings, recent work proposes AI-weather prediction (AI-WP) alternatives that learn to emulate analysis data with neural networks. While these data-driven approaches have enjoyed enormous success across diverse spatial and temporal resolutions, applications of video-understanding architectures for weather forecasting remain underexplored. To address these gaps, we propose \textbf{SaTformer}: a video transformer built on full space-time attention that skillfully forecasts extreme precipitation from satellite radiances. Along with our novel architecture, we introduce techniques to tame long-tailed precipitation datasets. Namely, we reformulate precipitation regression into a classification problem, and employ a class-weighted loss to address label imbalances. Our model scored \textbf{first place} on the NeurIPS Weather4Cast 2025 ``Cumulative Rainfall'' challenge. Code and model weights are available: \texttt{\href{github.com/leharris3/w4c-25}{github.com/leharris3/satformer}} |
| title | A Space-Time Transformer for Precipitation Nowcasting |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2511.11090 |