Scientists Unlock Solar Data Analysis Potential with AI

Scientists Unlock Solar Data Analysis Potential with AI

A team of astronomers and computer scientists at the University of Hawaiʻi Institute for Astronomy (IfA) has made a groundbreaking discovery in solar astronomy, combining cutting-edge technology to analyze data from the largest ground-based solar telescope in the world.

As part of the 'SPIn4D' project, recently published in the Astrophysical Journal, researchers aimed to develop advanced deep learning models that rapidly process and analyze vast amounts of data received from the US National Science Foundation (NSF) Daniel K Inouye Solar Telescope atop Haleakalā, Maui.

According to Dr. Kai Yang, lead author of the study and a postdoctoral researcher at IfA, large solar storms can "pose risks to satellites, radio communications, and power grids." The goal behind this project is to unlock the full potential of observations from the Inouye Solar Telescope, which may lead to significant breakthroughs in speed, accuracy, and scope of solar data analysis.

Using state-of-the-art simulations and machine learning, researchers were able to create an extensive dataset of 120 terabytes mimicking Inouye Solar Telescope observations with extremely high resolutions. This has enabled astronomers to visualize the Sun's atmosphere in near real-time, rather than waiting hours for the same accuracy.

The Inouye Solar Telescope, operated by the NSF National Solar Observatory (NSO), is the most powerful solar telescope worldwide. Its instruments are designed to measure the magnetic field of the Sun through polarized light. Scientists from NSO and High Altitude Observatory (HAO) use deep neural networks to estimate the physical properties of the solar photosphere.

"We have developed a machine learning model that provides fast approximations to expensive computations," said Dr. Peter Sadowski, an associate professor at UH Mānoa information and computer sciences department. "This technology will revolutionize our ability to analyze solar data, enabling us to better understand the birthplace of large solar storms."