AI Digitises Century-Old Solar Records from Kodaikanal Solar Observatory

Key facts

• Researchers led by Dibya Kirti Mishra from the Aryabhatta Research Institute of Observational Sciences (ARIES), an autonomous institute under the Department of Science and Technology (DST), used artificial intelligence to process hand-drawn solar observations from the Kodaikanal Solar Observatory.
• The Kodaikanal Solar Observatory maintains an archive of daily suncharts from 1904 to 2022, recording solar features including sunspots, filaments, prominences, and plages.
• The team utilised a supervised machine learning model known as U-Net to identify the Sun's disk and outline plages, which are bright magnetically active regions.
• The AI model traced magnetic regions across nine solar cycles between 1916 and 2007 to create a butterfly diagram, a visual map showing the latitude shift of magnetic activity.
• The study, involving the Indian Institute of Space Science and Technology (IIST), the Indian Institute of Astrophysics (IIA), and the Southwest Research Institute (SwRI), aims to enhance space weather research to protect satellites, navigation systems, communication networks, and power infrastructure.