Efficient AI for Weather Forecasting

Overview

Extreme weather events have caused severe damage and loss of life in recent decades. Traditional numerical weather prediction, while accurate, is computationally intensive—requiring supercomputers that consume massive amounts of energy.

In contrast, energy-efficient AI offers a transformative alternative. This course explores how modern AI models can drastically reduce energy consumption and CO₂ emissions while improving the accuracy and accessibility of weather forecasting.

Through practical examples, we demonstrate how techniques like LoRA fine-tuning, diffusion models, and GNSS-based sensing can enhance forecasting capabilities on both large and personal devices. By the end of this course, learners will understand how efficient AI methods enable sustainable, high-precision weather prediction for the future.

Syllabus

• Introduction to Weather Forecasting
Overview of traditional numerical weather prediction
Energy consumption and environmental impact of supercomputers
Introduction to AI in weather forecasting
• Efficient AI Technologies
Understanding AI energy consumption and CO₂ emissions
Comparison of traditional and AI-based methods for energy efficiency
• AI Techniques for Weather Prediction
Introduction to LoRA (Low-Rank Adaptation) Fine-Tuning
Overview of Diffusion Models
Leveraging GNSS (Global Navigation Satellite System) for weather data sensing
• Energy-Efficient Model Training
Strategies for reducing computational costs in AI training
Case studies in low-power model deployment
• AI Model Evaluation and Improvement
Techniques for enhancing model accuracy
Balancing precision with energy efficiency
Case study: Optimizing AI models for personal devices
• Practical Applications and Case Studies
Real-world examples of AI in weather forecasting
Analysis of efficiency gains and accuracy improvements
• Tools and Software
Overview of open-source tools for AI weather modeling
Hands-on session with AI frameworks and platforms
• Ethical and Environmental Considerations
Discussion on the environmental impact of AI technologies
Strategies for sustainable AI development and deployment
• Future Directions in AI for Weather Forecasting
Emerging technologies and research areas
The role of AI in future climate challenges and opportunities
• Final Project and Assessment
Applying learned concepts to develop an energy-efficient AI weather model
Evaluation of model performance and efficiency improvements

---

Taught by

PD Dr. Haojin Yang, Weixing Wang, Jona Otholt, Zi Yang, Gregor Nickel, Dr. Zhitong Xiong, Constantin Le Cleï, Dr. Peng Yuan, and Dr. Liangjing Zhang

---

Subjects

Artificial Intelligence