Accurate typhoon intensity forecasts using a non-iterative spatiotemporal transformer model
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866915514412957696 |
|---|---|
| author | Qu, Hongyu Xu, Hongxiong Dong, Lin Xiang, Chunyi Nie, Gaozhen |
| author_facet | Qu, Hongyu Xu, Hongxiong Dong, Lin Xiang, Chunyi Nie, Gaozhen |
| contents | Accurate forecasting of tropical cyclone (TC) intensity - particularly during periods of rapid intensification and rapid weakening - remains a challenge for operational meteorology, with high-stakes implications for disaster preparedness and infrastructure resilience. Recent advances in machine learning have yielded notable progress in TC prediction; however, most existing systems provide forecasts that degrade rapidly in extreme regimes and lack long-range consistency. Here we introduce TIFNet, a transformer-based forecasting model that generates non-iterative, 5-day intensity trajectories by integrating high-resolution global forecasts with a historical-evolution fusion mechanism. Trained on reanalysis data and fine-tuned with operational data, TIFNet consistently outperforms operational numerical models across all forecast horizons, delivering robust improvements across weak, strong, and super typhoon categories. In rapid intensity change regimes - long regarded as the most difficult to forecast - TIFNet reduces forecast error by 29-43% relative to current operational baselines. These results represent a substantial advance in artificial-intelligence-based TC intensity forecasting, especially under extreme conditions where traditional models consistently underperform. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_21349 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Accurate typhoon intensity forecasts using a non-iterative spatiotemporal transformer model Qu, Hongyu Xu, Hongxiong Dong, Lin Xiang, Chunyi Nie, Gaozhen Atmospheric and Oceanic Physics Machine Learning 68T07, 86A10 I.2.6; I.5.1; I.5.4 Accurate forecasting of tropical cyclone (TC) intensity - particularly during periods of rapid intensification and rapid weakening - remains a challenge for operational meteorology, with high-stakes implications for disaster preparedness and infrastructure resilience. Recent advances in machine learning have yielded notable progress in TC prediction; however, most existing systems provide forecasts that degrade rapidly in extreme regimes and lack long-range consistency. Here we introduce TIFNet, a transformer-based forecasting model that generates non-iterative, 5-day intensity trajectories by integrating high-resolution global forecasts with a historical-evolution fusion mechanism. Trained on reanalysis data and fine-tuned with operational data, TIFNet consistently outperforms operational numerical models across all forecast horizons, delivering robust improvements across weak, strong, and super typhoon categories. In rapid intensity change regimes - long regarded as the most difficult to forecast - TIFNet reduces forecast error by 29-43% relative to current operational baselines. These results represent a substantial advance in artificial-intelligence-based TC intensity forecasting, especially under extreme conditions where traditional models consistently underperform. |
| title | Accurate typhoon intensity forecasts using a non-iterative spatiotemporal transformer model |
| topic | Atmospheric and Oceanic Physics Machine Learning 68T07, 86A10 I.2.6; I.5.1; I.5.4 |
| url | https://arxiv.org/abs/2509.21349 |