How does an AI Weather Model Learn to Forecast Extreme Weather?
Fuente:
arXiv
Guardado en:
| Autores principales: | , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866918141074866176 |
|---|---|
| author | Baiman, Rebecca Barnes, Elizabeth A. Mahesh, Ankur |
| author_facet | Baiman, Rebecca Barnes, Elizabeth A. Mahesh, Ankur |
| contents | In a warming climate with more frequent severe weather, artificial intelligence (AI) weather models have the potential to provide cheaper, faster, and more accurate forecasts of high-impact weather events. To realize this potential, there is a need for more research on how models learn extreme events and how that learning might be improved. We investigate how a spherical Fourier neural operator model (SFNO) learns extreme weather by saving every checkpoint throughout training and analyzing a collection of 9 extreme weather events including heatwaves, atmospheric rivers, and tropical cyclones. The SFNO learns heatwaves similarly to other weather days, but we find evidence that the model learns information about atmospheric river and tropical cyclone forecasts that it loses later in training. We propose a possible training strategy to improve the forecasting of extreme events by retaining information from earlier training checkpoints, and provide initial evidence of its utility. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_10639 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | How does an AI Weather Model Learn to Forecast Extreme Weather? Baiman, Rebecca Barnes, Elizabeth A. Mahesh, Ankur Atmospheric and Oceanic Physics In a warming climate with more frequent severe weather, artificial intelligence (AI) weather models have the potential to provide cheaper, faster, and more accurate forecasts of high-impact weather events. To realize this potential, there is a need for more research on how models learn extreme events and how that learning might be improved. We investigate how a spherical Fourier neural operator model (SFNO) learns extreme weather by saving every checkpoint throughout training and analyzing a collection of 9 extreme weather events including heatwaves, atmospheric rivers, and tropical cyclones. The SFNO learns heatwaves similarly to other weather days, but we find evidence that the model learns information about atmospheric river and tropical cyclone forecasts that it loses later in training. We propose a possible training strategy to improve the forecasting of extreme events by retaining information from earlier training checkpoints, and provide initial evidence of its utility. |
| title | How does an AI Weather Model Learn to Forecast Extreme Weather? |
| topic | Atmospheric and Oceanic Physics |
| url | https://arxiv.org/abs/2509.10639 |