Scientists at the University of Graz have made a breakthrough in solar physics by using artificial intelligence to simulate the magnetic field in the Sun’s upper atmosphere in semi-real time. This study, published in Nature AstronomyIt holds great promise for improving our understanding of the Sun’s behavior and its effect on space weather.

The Sun’s magnetic field is the main driver of space weather, which can damage critical infrastructure such as electricity, aeronautics and our space technology. The main source of severe space weather events are solar active regions, the regions around sunspots where strong magnetic fields emanate from the sun’s surface. Current observational capabilities only allow the measurement of the magnetic field at the Sun’s surface, but energy accumulation and release occurs higher in the solar atmosphere, in the solar corona.

Using the power of physical neural networks, the team combined observational data with a force-free model of the magnetic field, providing a comprehensive understanding of the relationship between the observed phenomenon and the fundamental physics governing solar activity. This advanced method marks an important milestone in solar physics and opens up new possibilities for numerical simulations of the Sun.

The researchers modeled the evolution of the observed solar active region and demonstrated the ability to perform force-free simulations of the magnetic field in real time. Impressively, this process required less than 12 hours of computation time to simulate a series of five-day observations. This unprecedented speed allows scientists to perform real-time analysis and prediction of solar activity, improving our ability to predict space weather events.

The team also studied the time evolution of free magnetic energy in the coronal volume associated with solar flares on the Sun, such as coronal mass ejections – large clouds of plasma ejected from the Sun’s atmosphere at speeds of 100-3500 km/s. . Comparison with extreme ultraviolet observations confirmed the reliability and accuracy of the methodology. More importantly, the results revealed a significant reduction in free magnetic energy in both space and time, and this is directly related to the observed solar flares.

Lead researcher Robert Jarolim said: “Our use of AI in this context is a transformative step forward. Using AI techniques for computational modeling allows us to better combine observational data and has great potential to further enhance our modeling capabilities.”

“Computational speed holds great promise for improving space weather forecasting and expanding our knowledge of the Sun’s behavior,” says Skoltech Associate Professor Tetiana Podlachykova.

This study, led by scientists from the University of Graz and Skoltech, is a significant advance in the field of solar physics. Using the power of artificial intelligence and physical neural networks, they have succeeded in real-time modeling of the solar coronal magnetic field and revolutionized our ability to understand solar activity. Source