- Surya predicts solar flares up to two hours in advance, 16% better than current methods
- NASA made Surya model and code publicly available on HuggingFace and GitHub
- Surya development supported by NSF's NAIRR Pilot with partners like NVIDIA for computing power
NASA is raising the bar in solar science with the debut of the Surya Heliophysics Foundational Model, an artificial intelligence system built to decode the Sun's complex behavior.
Trained on nine years of continuous observations from NASA's Solar Dynamics Observatory (SDO), Surya is designed to analyse vast amounts of solar data, helping scientists better understand solar eruptions and predict space weather that threatens satellites, power grids, and communication systems.
Developed with IBM and other partners, Surya processes vast streams of solar data to provide insights that were once out of reach. The model already shows promise in one of heliophysics' toughest challenges: predicting solar flares.
Early tests demonstrate that Surya can generate visual forecasts up to two hours in advance, outperforming current methods by 16%.
NASA is making Surya openly available, the model on HuggingFace and the code on GitHub, to encourage scientists, educators, and innovators worldwide to build on its capabilities.
The foundation of Surya's success lies in the unique dataset from SDO. Since its launch in 2010, the spacecraft has provided an unbroken, high-resolution record of the Sun, capturing images every 12 seconds across multiple wavelengths along with detailed magnetic field measurements. This 15-year archive, spanning an entire solar cycle, gives Surya the depth and consistency needed to detect subtle patterns in solar activity that shorter datasets would miss.
NASA officials see Surya as a major step forward. "We are advancing data-driven science by embedding NASA's deep scientific expertise into cutting-edge AI models," said Kevin Murphy, chief science data officer at NASA Headquarters. "This model empowers broader understanding of how solar activity impacts critical systems and technologies that we all rely on here on Earth."
Surya's development was supported by the National Artificial Intelligence Research Resource (NAIRR) Pilot, a National Science Foundation initiative with industry partners such as NVIDIA, providing the computing power needed for advanced AI research.
As Joseph Westlake, director of NASA's Heliophysics Division, noted: "Just as we use meteorology to forecast Earth's weather, space weather forecasts predict the conditions and events in the space environment that can affect Earth and our technologies. Applying AI to data from our heliophysics missions is a vital step in increasing our space weather defense to protect astronauts and spacecraft, power grids and GPS, and many other systems that power our modern world."
