Democracy of AI Numerical Weather Models: An Example of Global Forecasting with FourCastNetv2 Made by a University Research Lab Using GPU

2025-04-23

AI Summary

Academic research teams at universities often lack sufficient GPU resources to reproduce state-of-the-art AI-based weather forecasting models. Method: This study systematically investigates the feasibility of deploying and training FourCastNet variants under constrained computational budgets. We achieve efficient single-GPU (A100) inference for FourCastNetv2 and complete full-model distributed training of FourCastNet across 64 A100 GPUs in 16 hours, integrating ERA5 multi-variable data preprocessing, GPU memory optimization, and rigorous model validation. Contribution/Results: To our knowledge, this is the first end-to-end reproduction of an AI weather forecasting model under realistic university-level hardware constraints. We establish an open-source, reusable teaching–research integrated paradigm, accompanied by a fully documented codebase and tutorial materials. Our work significantly lowers the barrier to entry for AI-driven meteorological modeling, advancing both the democratization of numerical weather prediction and its integration into academic education.

Technology Category

• Machine Learning: Hardware-aware ML
• Computer Vision: Remote Sensing / Geospatial AI
• Humans and AI: AI for Accessibility

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Web Mining and Content Analysis: Large pretrained models with web data
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

This paper demonstrates the feasibility of democratizing AI-driven global weather forecasting models among university research groups by leveraging Graphics Processing Units (GPUs) and freely available AI models, such as NVIDIA's FourCastNetv2. FourCastNetv2 is an NVIDIA's advanced neural network for weather prediction and is trained on a 73-channel subset of the European Centre for Medium-Range Weather Forecasts (ECMWF) Reanalysis v5 (ERA5) dataset at single levels and different pressure levels. Although the training specifications for FourCastNetv2 are not released to the public, the training documentation of the model's first generation, FourCastNet, is available to all users. The training had 64 A100 GPUs and took 16 hours to complete. Although NVIDIA's models offer significant reductions in both time and cost compared to traditional Numerical Weather Prediction (NWP), reproducing published forecasting results presents ongoing challenges for resource-constrained university research groups with limited GPU availability. We demonstrate both (i) leveraging FourCastNetv2 to create predictions through the designated application programming interface (API) and (ii) utilizing NVIDIA hardware to train the original FourCastNet model. Further, this paper demonstrates the capabilities and limitations of NVIDIA A100's for resource-limited research groups in universities. We also explore data management, training efficiency, and model validation, highlighting the advantages and challenges of using limited high-performance computing resources. Consequently, this paper and its corresponding GitHub materials may serve as an initial guide for other university research groups and courses related to machine learning, climate science, and data science to develop research and education programs on AI weather forecasting, and hence help democratize the AI NWP in the digital economy.

Problem

• Democratizing AI-driven weather forecasting for university research groups
• Overcoming GPU resource limitations in training AI weather models
• Validating AI weather models with limited high-performance computing resources

Innovation

• Leveraging GPUs for AI weather forecasting
• Using FourCastNetv2 for global predictions
• Training AI models with limited GPU resources