GraphCast: Learning skillful medium-range global weather forecasting

TL;DR

GraphCast is a machine learning model that provides fast, accurate medium-range global weather forecasts by directly learning from historical data, outperforming traditional methods on many verification targets.

Contribution

It introduces GraphCast, a novel ML-based weather forecasting method trained on reanalysis data, capable of high-resolution predictions over 10 days globally in under a minute.

Findings

Outperforms operational deterministic systems on 90% of verification targets.

Supports better prediction of severe weather events like cyclones and extreme temperatures.

Provides accurate, efficient forecasts at high spatial and temporal resolution.

Abstract

Global medium-range weather forecasting is critical to decision-making across many social and economic domains. Traditional numerical weather prediction uses increased compute resources to improve forecast accuracy, but cannot directly use historical weather data to improve the underlying model. We introduce a machine learning-based method called "GraphCast", which can be trained directly from reanalysis data. It predicts hundreds of weather variables, over 10 days at 0.25 degree resolution globally, in under one minute. We show that GraphCast significantly outperforms the most accurate operational deterministic systems on 90% of 1380 verification targets, and its forecasts support better severe event prediction, including tropical cyclones, atmospheric rivers, and extreme temperatures. GraphCast is a key advance in accurate and efficient weather forecasting, and helps realize the promise of machine learning for modeling complex dynamical systems.