Forecasting Global Weather with Graph Neural Networks

TL;DR

This paper introduces a graph neural network-based method for global weather forecasting that learns to predict atmospheric states several days ahead, achieving performance comparable to traditional physical models.

Contribution

The authors develop a novel graph neural network approach for weather forecasting that improves upon previous data-driven models and aligns with operational physical models in accuracy.

Findings

Outperforms previous data-driven weather models on key metrics

Achieves accuracy comparable to operational physical models

Successfully integrates with live operational forecast data

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.