DURHAM, N.H. -- (January 9, 2025) -- Researchers at the University of New Hampshire have utilized artificial intelligence to categorize the largest database of aurora borealis images, aiding in the understanding and forecasting of geomagnetic storms that can disrupt communications on Earth.
The study, published in the Journal of Geophysical Research, involved the development of AI and machine learning tools to classify over 706 million images from NASA's THEMIS dataset, collected by twin spacecraft monitoring the Earth's space environment. The THEMIS project captures images every three seconds from 23 locations across North America.
Jeremiah Johnson, associate professor of applied engineering and sciences and lead author, emphasized the dataset's significance, stating, "The massive data set is a valuable resource that can help researchers understand how the solar wind interacts with the Earth's magnetosphere. However, its size has previously limited effective usage."
The team implemented a new algorithm to annotate images from 2008 to 2022 into six categories: arc, diffuse, discrete, cloudy, moon, and clear/no aurora. This organization allows for easier filtering and retrieval of information.
Johnson noted, "The labeled database could reveal further insight into auroral dynamics, but our primary goal was to organize the THEMIS database for more effective research use and to provide a substantial sample for future studies."
Co-authors include Amy Keesee, Doğacan Su Öztürk, Donald Hampton, Matthew Blandin, and Hyunju Connor from NASA Goddard Space Flight Center. The research received funding from NASA's heliophysics division and the National Science Foundation.