Researchers from NYU Abu Dhabi (NYUAD) have developed an AI model capable of predicting solar wind speeds up to four days ahead, offering significantly improved accuracy over existing methods. Their findings appear in The Astrophysical Journal Supplement Series.
Guided by Postdoctoral Associate Dattaraj Dhuri and Co-Principal Investigator Shravan Hanasoge at the Center for Space Science (CASS), the NYUAD team trained their AI using high-resolution ultraviolet (UV) images from NASA’s Solar Dynamics Observatory, alongside historical solar wind data.
Unlike popular AI language models that process text, this system analyzes solar images to detect patterns associated with changes in solar wind.
This approach has enhanced forecast accuracy by 45% compared to current models and by 20% over previous AI-based solutions.
“This marks a significant advancement in protecting satellites, navigation systems, and power grids essential to modern life,” stated Dhuri, the study’s lead author. “Through the integration of advanced AI and solar observations, we can provide early warnings to safeguard critical technologies on Earth and in space.”
