FourCastNet: Accelerating Global High-Resolution Weather Forecasting Using Adaptive Fourier Neural Operators

By Thorsten Kurth, Shashank Subramanian, P. Harrington, Jaideep Pathak, Morteza Mardani, D. Hall, A. Miele, Karthik Kashinath, Anima Anandkumar. Platform for Advanced Scientific Computing Conference, 2022

407 citations recorded by the source corpus.

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 and resolution 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. FourCastNet 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 3,072 GPUs is 67.4 minutes, resulting in an 80,000 times faster time-to-solution relative to state-of-the-art NWP, in inference. FourCastNet produces accurate instantaneous weather predictions for a week in advance and enables enormous ensembles that could be used to improve predictions of rare weather extremes.

Subfields: Atmospheric Science, Global and Planetary Change

View Paper: https://www.semanticscholar.org/paper/286aafc2453a3faeae539e939273938b690fa64a

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Atmospheric Science

1. Shakir Mohamed
2. Paris Perdikaris
3. Stephan Hoyer
4. G. Em Karniadakis
5. Steven L. Brunton
6. I. Pisso

In Global and Planetary Change

1. Shakir Mohamed
2. Wanli Ouyang
3. Paris Perdikaris
4. Stephan Hoyer
5. Maizar Raissi
6. I. Pisso

Citations in the corpus

• A foundation model for the Earth system (2025, positive): “The analysis of other large-scale global machine learning models 2,3,38 has been limited to comparisons of tracks, with recent comparisons indicating that performance lags behind that of the official operational forecasts 39 .”
• Neural operators for accelerating scientific simulations and design (2023, neutral): “In addition to learning metrics, the benchmarks also need to incorporate hardware performance metrics that are typically used in HPC studies [93] and explore the benefits of low-precision learning [94].”
• The operational medium-range deterministic weather forecasting can be extended beyond a 10-day lead time (2023, positive): “Different from existing data-driven weather forecasting methods 11,12,15 – 17 , FengWu solves the problem of medium-range forecasting from a multi-modal and multi-task perspective 25,26 .”
• GraphCast: Learning skillful medium-range global weather forecasting (2022, methodology): “Performance of GraphCast compared to top ML forecasting models Of the three most recent ML-based forecasting models in the past year, Keisler (2022)’s model, FCN (Kurth et al., 2022; Pathak et al., 2022), and Pangu-Weather (Bi et al.”