29223

ArchesWeather: An efficient AI weather forecasting model at 1.5° resolution

Guillaume Couairon, Christian Lessig, Anastase Charantonis, Claire Monteleoni
Inria, France
arXiv:2405.14527 [cs.LG], (23 May 2024)

@misc{couairon2024archesweather,

   title={ArchesWeather: An efficient AI weather forecasting model at 1.5{deg} resolution},

   author={Guillaume Couairon and Christian Lessig and Anastase Charantonis and Claire Monteleoni},

   year={2024},

   eprint={2405.14527},

   archivePrefix={arXiv},

   primaryClass={cs.LG}

}

Download Download (PDF)   View View   Source Source   

786

views

One of the guiding principles for designing AI-based weather forecasting systems is to embed physical constraints as inductive priors in the neural network architecture. A popular prior is locality, where the atmospheric data is processed with local neural interactions, like 3D convolutions or 3D local attention windows as in Pangu-Weather. On the other hand, some works have shown great success in weather forecasting without this locality principle, at the cost of a much higher parameter count. In this paper, we show that the 3D local processing in Pangu-Weather is computationally sub-optimal. We design ArchesWeather, a transformer model that combines 2D attention with a column-wise attention-based feature interaction module, and demonstrate that this design improves forecasting skill. ArchesWeather is trained at 1.5° resolution and 24h lead time, with a training budget of a few GPU-days and a lower inference cost than competing methods. An ensemble of two of our best models shows competitive RMSE scores with the IFS HRES and outperforms the 1.4° 50-members NeuralGCM ensemble for one day ahead forecasting. Code and models will be made publicly available.
No votes yet.
Please wait...

Recent source codes

* * *

* * *

HGPU group © 2010-2024 hgpu.org

All rights belong to the respective authors

Contact us: