Scientists at NYU Abu Dhabi (NYUAD) have developed a groundbreaking AI model capable of forecasting solar wind speeds up to four days in advance — a significant improvement over current forecasting techniques. Published in The Astrophysical Journal Supplement Series, the study details how the team led by Postdoctoral Associate Dattaraj Dhuri and Co-Principal Investigator Shravan Hanasoge at the Center for Space Science (CASS) trained their AI system using high-resolution ultraviolet images from NASA’s Solar Dynamics Observatory, combined with extensive historical solar wind data. Unlike popular AI language models that analyze text, this model focuses on analyzing images of the Sun to identify patterns linked to solar wind fluctuations, enabling a highly accurate early warning system for space weather events.
The AI model achieves a 45% improvement in forecast accuracy compared to existing operational models and outperforms previous AI-based methods by 20%. Dhuri emphasizes the model’s potential to safeguard vital infrastructure: “This is a major step forward in protecting satellites, navigation systems, and power grids that underpin modern life.” By providing earlier and more precise warnings of solar wind changes, the innovation promises to enhance the resilience of both Earth-based and space-based technologies. As reliance on satellites and space infrastructure grows globally, this advancement marks a pivotal development in astrophysics and AI-driven space weather prediction.
