It is shown 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.
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.
P. Battaglia
13 papers
O. Vinyals
41 papers
Peter Wirnsberger
2 papers
Suman V. Ravuri
2 papers
Z. Eaton-Rosen
3 papers
Alexander Merose
3 papers
George Holland
2 papers
A. Pritzel
5 papers
Alvaro Sanchez-Gonzalez
7 papers
S. Mohamed
4 papers
Stephan Hoyer
2 papers
Meire Fortunato
5 papers
Rémi R. Lam
1 papers
M. Willson
1 papers
Ferran Alet
3 papers