Extreme weather amplified by climate change is causing increasingly devastating impacts across the globe. The current use of physics-based numerical weather prediction (NWP) limits accuracy due to high computational cost and strict time-to-solution limits. We report that a data-driven deep learning Earth system emulator, FourCastNet, can predict global weather and generate medium-range forecasts five orders-of-magnitude faster than NWP while approaching state-of-the-art accuracy. FourCast-Net is optimized and scales efficiently on three supercomputing systems: Selene, Perlmutter, and JUWELS Booster up to 3,808 NVIDIA A100 GPUs, attaining 140.8 petaFLOPS in mixed precision (11.9%of peak at that scale). The time-to-solution for training FourCastNet measured on JUWELS Booster on 3,072GPUs is 67.4minutes, resulting in an 80,000times faster time-to-solution relative to state-of-the-art NWP, in inference. FourCastNet produces accurate instantaneous weather predictions for a week in advance, enables enormous ensembles that better capture weather extremes, and supports higher global forecast resolutions.
@article{arxiv.2208.05419,
title = {FourCastNet: Accelerating Global High-Resolution Weather Forecasting using Adaptive Fourier Neural Operators},
author = {Thorsten Kurth and Shashank Subramanian and Peter Harrington and Jaideep Pathak and Morteza Mardani and David Hall and Andrea Miele and Karthik Kashinath and Animashree Anandkumar},
journal= {arXiv preprint arXiv:2208.05419},
year = {2022}
}