EPT-2 Technical Report

2025-07-13

AI Summary

Accurate and efficient subseasonal forecasting of key meteorological variables—such as 10-m/100-m wind speed, 2-m temperature, and surface solar radiation—remains challenging in energy applications due to insufficient accuracy and high computational cost over 0–240-hour lead times. Method: We propose EPT-2, a physics-informed geophysical foundation model based on the Transformer architecture, and its ensemble variant EPT-2e. EPT-2 incorporates physically motivated positional encoding, multi-scale spatiotemporal modeling, and perturbation-based ensemble strategies to enable efficient probabilistic forecasting. Contribution/Results: Trained and validated on multi-source observational and reanalysis datasets, EPT-2e achieves state-of-the-art (SOTA) performance across all target variables—outperforming both ECMWF’s high-resolution deterministic forecast and ensemble mean. It reduces inference latency by 3× and GPU memory usage by 40%. Deployed operationally on app.jua.ai, EPT-2e supports real-time, production-grade energy forecasting.

Technology Category

• Machine Learning: Efficient ML / Green AI
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Intelligent Robots: State Estimation

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

We present EPT-2, the latest iteration in our Earth Physics Transformer (EPT) family of foundation AI models for Earth system forecasting. EPT-2 delivers substantial improvements over its predecessor, EPT-1.5, and sets a new state of the art in predicting energy-relevant variables-including 10m and 100m wind speed, 2m temperature, and surface solar radiation-across the full 0-240h forecast horizon. It consistently outperforms leading AI weather models such as Microsoft Aurora, as well as the operational numerical forecast system IFS HRES from the European Centre for Medium-Range Weather Forecasts (ECMWF). In parallel, we introduce a perturbation-based ensemble model of EPT-2 for probabilistic forecasting, called EPT-2e. Remarkably, EPT-2e significantly surpasses the ECMWF ENS mean-long considered the gold standard for medium- to longrange forecasting-while operating at a fraction of the computational cost. EPT models, as well as third-party forecasts, are accessible via the app.jua.ai platform.

Problem

• Improves Earth system forecasting accuracy
• Outperforms leading AI and numerical weather models
• Enables probabilistic forecasting with lower computational cost

Innovation

• Earth Physics Transformer AI model
• Perturbation-based ensemble forecasting
• Computationally efficient probabilistic forecasting