FourCastNet: A Global Data-driven High-resolution Weather Model using Adaptive Fourier Neural Operators

TL;DR

FourCastNet is a fast, high-resolution, data-driven weather forecasting model that matches or exceeds traditional NWP models in accuracy for certain variables and enables rapid large-ensemble predictions.

Contribution

The paper introduces FourCastNet, a novel neural network-based weather model that achieves high-resolution global forecasts with unprecedented speed and accuracy, especially for complex variables.

Findings

Matches ECMWF IFS accuracy for large-scale variables at short lead times.

Outperforms IFS in predicting precipitation and fine-scale structures.

Generates week-long forecasts in less than 2 seconds.

Abstract

FourCastNet, short for Fourier Forecasting Neural Network, is a global data-driven weather forecasting model that provides accurate short to medium-range global predictions at $0.25^{\circ}$ resolution. FourCastNet accurately forecasts high-resolution, fast-timescale variables such as the surface wind speed, precipitation, and atmospheric water vapor. It has important implications for planning wind energy resources, predicting extreme weather events such as tropical cyclones, extra-tropical cyclones, and atmospheric rivers. FourCastNet matches the forecasting accuracy of the ECMWF Integrated Forecasting System (IFS), a state-of-the-art Numerical Weather Prediction (NWP) model, at short lead times for large-scale variables, while outperforming IFS for variables with complex fine-scale structure, including precipitation. FourCastNet generates a week-long forecast in less than 2 seconds, orders of magnitude faster than IFS. The speed of FourCastNet enables the creation of rapid and inexpensive large-ensemble forecasts with thousands of ensemble-members for improving probabilistic forecasting. We discuss how data-driven deep learning models such as FourCastNet are a valuable addition to the meteorology toolkit to aid and augment NWP models.