ClimODE: Climate and Weather Forecasting with Physics-informed Neural ODEs
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910411772657664 |
|---|---|
| author | Verma, Yogesh Heinonen, Markus Garg, Vikas |
| author_facet | Verma, Yogesh Heinonen, Markus Garg, Vikas |
| contents | Climate and weather prediction traditionally relies on complex numerical simulations of atmospheric physics. Deep learning approaches, such as transformers, have recently challenged the simulation paradigm with complex network forecasts. However, they often act as data-driven black-box models that neglect the underlying physics and lack uncertainty quantification. We address these limitations with ClimODE, a spatiotemporal continuous-time process that implements a key principle of advection from statistical mechanics, namely, weather changes due to a spatial movement of quantities over time. ClimODE models precise weather evolution with value-conserving dynamics, learning global weather transport as a neural flow, which also enables estimating the uncertainty in predictions. Our approach outperforms existing data-driven methods in global and regional forecasting with an order of magnitude smaller parameterization, establishing a new state of the art. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_10024 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | ClimODE: Climate and Weather Forecasting with Physics-informed Neural ODEs Verma, Yogesh Heinonen, Markus Garg, Vikas Artificial Intelligence Emerging Technologies Machine Learning Atmospheric and Oceanic Physics Climate and weather prediction traditionally relies on complex numerical simulations of atmospheric physics. Deep learning approaches, such as transformers, have recently challenged the simulation paradigm with complex network forecasts. However, they often act as data-driven black-box models that neglect the underlying physics and lack uncertainty quantification. We address these limitations with ClimODE, a spatiotemporal continuous-time process that implements a key principle of advection from statistical mechanics, namely, weather changes due to a spatial movement of quantities over time. ClimODE models precise weather evolution with value-conserving dynamics, learning global weather transport as a neural flow, which also enables estimating the uncertainty in predictions. Our approach outperforms existing data-driven methods in global and regional forecasting with an order of magnitude smaller parameterization, establishing a new state of the art. |
| title | ClimODE: Climate and Weather Forecasting with Physics-informed Neural ODEs |
| topic | Artificial Intelligence Emerging Technologies Machine Learning Atmospheric and Oceanic Physics |
| url | https://arxiv.org/abs/2404.10024 |