// NASA BREAKING NEWS — SPAZIO & SCIENZA
NASA’s COFFIES Uses AI to Predict Storm-Causing Active Regions on Sun
As humanity looks to the Moon and stars for future exploration, predicting space weather — conditions in space primarily driven by the Sun — is more important than ever.
Now, a team of astrophysicists and data scientists with NASA’s COFFIES (Consequence Of Fields and Flows in the Interior and Exterior of the Sun) has developed a novel machine-learning model capable of predicting the emergence of active regions on the Sun up to 12 hours before they appear.
The Sun is constantly churning. Intense concentrations of localized magnetic fields can suddenly break through the solar surface, forming sunspots. Space weather forecasters then collectively number and track sunspots since they are visible manifestations of active regions, which serve as the main engines behind severe space weather events such as solar flares and coronal mass ejections. These eruptions send waves of high-energy radiation and charged particles across space, creating storms that can threaten astronauts, disable satellites, and disrupt radio communications on Earth.
By bridging expertise across different scientific institutions, COFFIES, a NASA DRIVE (Diversify, Realize, Integrate, Venture, Educate) Science Center, brought together a team of researchers from New Jersey Institute of Technology (NJIT), Princeton University, and NASA’s Ames Research Center in California’s Silicon Valley. The team turned to advanced artificial intelligence architectures — which dictate how data is processed and used to produce reliable predictions or actions — to capture subtle, time-based pattern changes on the solar surface before an active region took shape. By analyzing data captured by the agency's Solar Dynamics Observatory and using NASA Ames' supercomputing resources, this new approach, published in the Journal of Geophysical Research: Machine Learning and Computation, looks at fluctuations in acoustic waves caused by sunspot regions when the regions form beneath the solar surface and begin the journey upward to emerge on the surface.
“We cannot directly see the magnetic structure while it is still rising through the solar interior. Instead, we must look for indirect effects — very small changes in the magnetic field and in the pattern of acoustic waves continually traveling through the Sun,” said Alexander Kosovichev, a COFFIES co-investigator at NJIT. “The developed technique identifies precursors associated with an emerging active region in slight changes of the Sun’s acoustic power — more like a slight change in rhythm within a very noisy orchestra.”
To view this video please enable JavaScript, and consider upgrading to a web browser that
supports HTML5 video
To develop current operational forecasts, the National Oceanic and Atmospheric Administration’s Space Weather Prediction Center and the United States Air Force monitor active regions that are already visible on the Sun to analyze the regions’ characteristics and estimate the probability of solar flares.
The COFFIES team aims to revolutionize this process. The AI model the team developed a specialized early detection system to handle very long sequences of data — called sliding-window transformer architecture — to use observations to find tiny reductions in the Sun’s acoustic activity and magnetic field, signals that scientists struggled to capture until now. These reductions form patterns that the AI model uses to predict active regions several hours before they become visible on the solar surface. Instead of looking at all activity on the solar surface at once, like earlier deep learning approaches have done, this new model moves a fixed-size "viewing window" across a long timeline of the Sun’s activity to focus on recent data while remembering overall patterns. This method allows forecasters the ability to predict approximate locations of emerging sunspots, rather than relying on counting already visible sunspots.
This promising AI architecture shows how deep machine learning can contribute to heliop