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

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Couairon, Guillaume, Lessig, Christian, Charantonis, Anastase, Monteleoni, Claire
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913415431192576
author Couairon, Guillaume
Lessig, Christian
Charantonis, Anastase
Monteleoni, Claire
author_facet Couairon, Guillaume
Lessig, Christian
Charantonis, Anastase
Monteleoni, Claire
contents 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 four of our models shows better RMSE scores than the IFS HRES and is competitive with the 1.4° 50-members NeuralGCM ensemble for one to three days ahead forecasting. Our code and models are publicly available at https://github.com/gcouairon/ArchesWeather.
format Preprint
id arxiv_https___arxiv_org_abs_2405_14527
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle ArchesWeather: An efficient AI weather forecasting model at 1.5° resolution
Couairon, Guillaume
Lessig, Christian
Charantonis, Anastase
Monteleoni, Claire
Machine Learning
Artificial Intelligence
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 four of our models shows better RMSE scores than the IFS HRES and is competitive with the 1.4° 50-members NeuralGCM ensemble for one to three days ahead forecasting. Our code and models are publicly available at https://github.com/gcouairon/ArchesWeather.
title ArchesWeather: An efficient AI weather forecasting model at 1.5° resolution
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2405.14527