An artificial intelligence team used 700 million photos of the aurora borealis to teach artificial intelligence to better understand them. The goal is to try to teach the machine to predict geomagnetic storms.
Machine algorithms in the study of auroras
The aurora borealis is known as a stunning light show in the night sky, but it being caused by solar flares can signal the approach of geomagnetic storms that can interrupt vital communications and security infrastructure on Earth. Using artificial intelligence, researchers at the University of New Hampshire have classified and marked the largest database of aurora borealis images in history, which could help scientists better understand and predict destructive geomagnetic storms.
A study recently published in the Journal of Geophysical Research: Machine Learning and Computation developed artificial intelligence and machine learning tools that were able to successfully identify and classify more than 706 million images of auroral phenomena in NASA’s Time History of Events and Macroscale Interactions during Substorms (THEMIS) database collected by two twin spacecraft studying the space environment around Earth. THEMIS provides images of the night sky every three seconds from sunset to sunrise from 23 different stations across North America.
“The massive dataset is a valuable resource that can help researchers understand how the solar wind interacts with the Earth’s magnetosphere, the protective bubble that shields us from charged particles streaming from the sun,” said Jeremiah Johnson, associate professor of applied engineering and sciences and lead author of the study. “But until now, its huge size limited how effectively we can use that data.”
Sorting sky images
Researchers created a new algorithm to sort THEMIS all-sky images (ASI) from 2008 through 2022 and efficiently annotate them into six different categories – arc, diffuse, discrete, cloudy, moon and clear/no aurora – making it easier to filter, sort, and find valuable information.
A marked database can help provide a deeper understanding of auroral dynamics, but at the most basic level, scientists have sought to organize the THEMIS image database so that the vast amount of historical data it contains can be used more efficiently by researchers and provide a large enough sample for future studies.
According to phys.org
