The Sun, though 93 million miles away, has a direct influence on modern life. Solar flares and coronal mass ejections can disable satellites, disrupt flight operations, cause blackouts, and endanger astronauts. With global reliance on space infrastructure rising, accurate forecasting of solar weather has become essential.

Industry studies estimate that a severe solar storm could expose the global economy to $2.4 trillion in losses over five years. Recent solar disturbances have already damaged satellites, forced aircraft to reroute, and impaired GPS services. The risks extend from spacecraft hardware to agriculture, which depends on reliable GPS systems for production.

"Think of this as a weather forecast for space," said Juan Bernabe-Moreno, Director of IBM Research Europe, UK and Ireland. "Just as we work to prepare for hazardous weather events, we need to do the same for solar storms. Surya gives us unprecedented capability to anticipate what's coming and is not just a technological achievement, but a critical step toward protecting our technological civilization from the star that sustains us."

Unlike traditional prediction methods that rely on limited solar surface views, Surya leverages the largest curated heliophysics dataset. The model has already shown a 16 percent gain in solar flare classification accuracy and, for the first time, provides high resolution forecasts pinpointing where flares will appear up to two hours in advance.

NASA contributed nine years of high resolution solar imagery for training, requiring IBM to develop specialized architectures to process datasets ten times larger than those used in conventional AI. This enables Surya to resolve solar features at scales previously inaccessible in large model training.

"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. "By developing a foundation model trained on NASA's heliophysics data, we're making it easier to analyze the complexities of the Sun's behavior with unprecedented speed and precision. This model empowers broader understanding of how solar activity impacts critical systems and technologies that we all rely on here on Earth."

Surya joins IBM and NASA's Prithvi family of AI models, which includes earlier releases for climate and geospatial analysis. By publishing Surya on Hugging Face, the organizations aim to democratize access for global researchers, enabling new applications in solar storm prediction and broader space science.

Research Report:Interplanetary Causes and Impacts of the 2024 May Superstorm on the Geosphere: An Overview

Related Links
IBM Research
Solar Science News at SpaceDaily