![\"AI](\"https:\/\/static.toiimg.com\/thumb\/msid-132152416,imgsize-164259,width-400,height-225,resizemode-4\/ai-image.jpg\")
<\/div>\n<\/div>\n<\/div>\n<\/section>\n<\/div><\/div>\n<\/div>\nBENGALURU: Artificial Intelligence has helped scientists unlock a century of the Sun\u2019s history by analysing hand-drawn observations from the Kodaikanal Solar Observatory, turning fragile paper records into a valuable digital archive that could improve understanding of long-term solar activity and future space weather.In a study published in The Astrophysical Journal, researchers led by Dibya Kirti Mishra of the Aryabhatta Research Institute of Observational Sciences (Aries) along with the collaborators from Isro\u2019s Indian Institute of Space Science and Technology (IIST), Southwest Research Institute, Boulder, USA and Indian Institute of Astrophysics (IIA), used machine learning to analyse daily solar drawings made at the Kodaikanal Solar Observatory (KoSO) between 1916 and 2007.The work demonstrates how historical observations, once considered difficult to analyse systematically, can now be transformed into reliable scientific data. Aries and IIA are autonomous institutes of the department of science and technology (DST).For more than a century, scientists have studied the Sun\u2019s magnetic activity, which waxes and wanes in roughly 11-year cycles. These cycles drive sunspots, solar flares and eruptions that can interfere with satellites, navigation systems, radio communications and power grids on Earth. While modern telescopes provide precise digital observations, records from earlier decades are often incomplete or inconsistent, making it difficult to understand how the Sun has behaved over long periods.KoSO, one of the world\u2019s oldest solar observatories, has maintained an exceptional archive of daily \u201csuncharts\u201d from 1904 to 2022. \u201cBefore digital imaging became commonplace, astronomers carefully sketched features such as sunspots, plages, filaments and prominences onto standard grids. Although scientifically valuable, differences in drawing styles, ageing paper and varying scan quality have made these records challenging to analyse using conventional methods,\u201d DST said.To overcome these hurdles, researchers employed a supervised machine learning model known as \u201cU-Net\u201d. The system first identified the Sun\u2019s disc in each scanned drawing, accurately determining its centre, size and orientation. It then detected and mapped bright magnetic regions known as plages across nine solar cycles spanning 1916 to 2007.Plages are bright patches in the Sun\u2019s atmosphere associated with strong magnetic fields. Because they closely reflect the Sun\u2019s magnetic activity, scientists regard them as one of the most reliable indicators of long-term changes. Extracting them from historical records helps bridge the gap between early observations and the continuous measurements available in the space age.The AI-generated data allowed the team to construct a classic \u201cbutterfly diagram\u201d, which shows how solar activity migrates from higher latitudes towards the equator during each solar cycle. \u201cThey also found that the plage areas identified from the drawings closely matched measurements obtained from KoSO’s Ca II K full-disc images, confirming that the century-old sketches provide dependable scientific information,\u201d DST said.Researchers say such long-term records are essential for comparing the strength and structure of different solar cycles, reconstructing past changes in the Sun’s magnetic influence and improving models of long-term space weather.<\/div>\n