Forecasting Global Weather with Graph Neural Networks

Abstract

We present a data-driven approach for forecasting global weather using graph neural networks. The system learns to step forward the current 3D atmospheric state by six hours, and multiple steps are chained together to produce skillful forecasts going out several days into the future. The underlying model is trained on reanalysis data from ERA5 or forecast data from GFS. Test performance on metrics such as Z500 (geopotential height) and T850 (temperature) improves upon previous data-driven approaches and is comparable to operational, full-resolution, physical models from GFS and ECMWF, at least when evaluated on 1-degree scales and when using reanalysis initial conditions. We also show results from connecting this data-driven model to live, operational forecasts from GFS.

Cited by in corpus (9)

• The rise of data-driven weather forecasting
• SwinVRNN: A Data-Driven Ensemble Forecasting Model via Learned Distribution Perturbation
• A Machine Learning Tutorial for Operational Meteorology, Part II: Neural Networks and Deep Learning
• Global Extreme Heat Forecasting Using Neural Weather Models
• A Practical Probabilistic Benchmark for AI Weather Models
• Proper Scoring Rules for Multivariate Probabilistic Forecasts based on Aggregation and Transformation
• Leveraging data-driven weather models for improving numerical weather prediction skill through large-scale spectral nudging
• Community Research Earth Digital Intelligence Twin (CREDIT)
• Benchmarking atmospheric circulation variability in an AI emulator, ACE2, and a hybrid model, NeuralGCM