**Executive Summary**
This report outlines a study where Artificial Intelligence was used to analyze 100 years of hand-drawn solar records (1916–2007) from the Kodaikanal Solar Observatory (KoSO). The research, published in *The Astrophysical Journal*, aimed to track magnetically active regions on the Sun to understand long-term solar activity cycles. By converting historical "suncharts" into machine-readable data, the study provides a consistent record to help assess space weather risks that affect modern technology.
**Key Points / Main Content**
* **Study Overview and Objectives**
* Researchers led by Dibya Kirti Mishra (ARIES) used AI to scan and analyze hand-drawn Sun records from 1916 to 2007.
* The project utilized KoSO’s unique collection of daily "suncharts" (1904–2022), which include sunspots, plages, filaments, and prominences.
* The goal was to create a long-term view of solar magnetic activity shifts to overcome the inconsistencies of older, incomplete observations.
* **Technical Methodology**
* The study employed a supervised machine learning approach known as U-Net.
* The AI model automatically pinpointed the Sun's disk, center, size, and tilt in scanned drawings to ensure accurate placement of features.
* The model identified and traced "plages"—magnetically active patches—across nine solar cycles (Cycles 15–23).
* **Scientific Findings**
* The researchers created a "butterfly diagram" showing how plage activity and magnetic influence shift with latitude and solar-cycle phases.
* Extracted data from hand-drawn suncharts matched well with results from modern Ca II K full-disk observations.
* The study proved that machine learning can successfully turn uneven historical archives into clean, consistent datasets that traditional methods cannot produce.
* **Significance of Solar Monitoring**
* Plages are considered a reliable "fingerprint" of the Sun’s magnetism.
* Understanding solar cycles is critical as eruptions and flares can disrupt satellites, navigation systems, and power grids on Earth.
**Impact Analysis**
**Researchers and Scientists**
**Impact**
They now have access to a consistent, 100-year dataset that connects historical records with modern space-age measurements. This allows for better comparisons of solar cycle strength and structure.
**Action Required**
Scientists should utilize this machine-readable data to fill gaps in existing solar series and improve reconstructions of the Sun’s past energy output and magnetic changes.
**Technology and Infrastructure Providers (Satellites, Navigation, Power Grids)**
**Impact**
Improved understanding of long-term solar activity cycles leads to better assessment of space weather risks that can disrupt critical technical infrastructure.
**Action Required**
Providers should integrate these long-term space weather risk insights into the planning and protection of global technology and power systems.
**Society**
**Impact**
The research contributes to a broader understanding of how the Sun affects Earth, potentially leading to more resilient technology and better-informed risk management for space weather events.
**Action Required**
No direct action is required from the general public, though the study serves as a basis for enhanced global preparedness against space weather disruptions.
Key Entities Referenced
Kodaikanal Solar Observatory (KoSO): The facility whose century-long hand-drawn solar records and 'suncharts' served as the primary data source for the AI-driven solar activity study.
Aryabhatta Research Institute of Observational Sciences (ARIES): The autonomous research institute under the Department of Science and Technology that led the study to digitize and analyze historical solar data.
Department of Science and Technology (DST): The nodal government department overseeing the research institutes and initiatives involved in this solar physics and machine learning project.
U-Net: The specific supervised machine learning architecture employed to automatically detect and trace solar features from digitized historical archives.
Dibya Kirti Mishra: The lead researcher from ARIES who headed the collaborative scientific team responsible for the study.
Ministry of Science & Technology
AI studies 100 years of Sun images to track
bright solar regions from the Kodaikanal Solar
Observatory
प्रव तथ: 01 JUL 2026 4:02PM by PIB Delhi
Artificial Intelligence has been used to trace the shift in magnetically active patches on the Sun from 1916
to 2007 by scanning 100 years of hand-drawn Sun records from the Kodaikanal Solar Observatory
(KoSO). This could give a much longer view of how solar activity changes over time.
For more than a hundred years, scientists have been trying to understand how the Sun’s magnetic activity
rises and falls in rhythmic cycles. These cycles affect sunspots, flares, and eruptions, which can disrupt
satellites, navigation, and power on Earth. However, older observations are often incomplete and
inconsistent, making long-term study difficult. That’s why historical records are very valuable.
In a new study, researchers led by Dibya Kirti Mishra from the Aryabhatta Research Institute of
Observational Sciences (ARIES), an autonomous institute under Department of Science and Technology
(DST), Govt. of India, along with the collaborators from Indian Institute of Space Science and
Technology, Thiruvananthapuram; Southwest Research Institute, Boulder, USA; Indian Institute of
Astrophysics (IIA), Bangalore shows that 100 years of hand-drawn Sun records from the Kodaikanal Solar
Observatory (KoSO) can be turned into useful data using modern machine learning techniques. The
observatory has a unique collection of observations, including daily ‘suncharts’ from 1904 to 2022, where
features like sunspots, plages, filaments, and prominences were carefully drawn on a standard grid.Fig: Time–latitude “butterfly” map of solar plage area (in μDF) measured in 1° latitude bands for each
day. (a) Plage distribution derived in this work using U-Net–based detection on KoSO suncharts. (b)
Butterfly diagram from KoSO Ca II K plage detections (B. K. Jha et al. 2024). (c) A combined record that
merges both datasets, using the Ca II K results to fill gaps in the sunchart series. Colors show plage area,
revealing how magnetic activity shifts with latitude and solar-cycle phase from 1916–2007 (Solar Cycles
15–23).
Before digital tools, scientists relied on careful drawings to record what they saw. KoSO’s suncharts are
valuable because they show solar activity over many cycles and include different features marked in
specific ways. However, differences in drawing styles, paper aging, and scan quality make it difficult to
create a clean and consistent dataset using traditional methods.
To address the problem of messy, hand-drawn historical records, the research published in the
Astrophysical Journal used a supervised machine learning approach (U-Net) in two main steps. First, the
model automatically found the Sun’s disk in each scanned drawing, pinpointing the center, size, and tilt, so
every feature could be placed in the correct location on the Sun. Next, it identified and traced plages
(butterfly, magnetically active patches on the Sun) across drawings covering nine solar cycles from 1916
to 2007. This is important because plages are a reliable “fingerprint” of the Sun’s magnetism, and
extracting them from old archives helps scientists connect today’s space-age measurements with what the
Sun was doing decades earlier.
By turning drawings into machine-readable data, the researchers led by Dibya Kirti Mishra were able to
track how plage activity shifts over time, creating a ‘butterfly diagram’ that shows the solar cycle. They
also found that the plage areas from these drawings match well with those derived from KoSO’s Ca II K
full-disk observations, proving that the suncharts can help fill gaps and improve long-term solar data
Long-term, consistent records of the Sun’s magnetic activity are crucial because they let scientists
compare how different solar cycles vary in strength and structure, improve reconstructions of how the
Sun’s energy output and magnetic influence have changed in the past, and helps society better understand
long-term space weather risks that can affect technology on Earth. The study shows that old, uneven
historical records can be improved using machine learning to create consistent data over many decades,
something traditional methods struggle to do.
Publication link: https://doi.org/10.3847/1538-4365/ae381e
*****
NKR/FT/NM
(रलीज़ आईडी: 2279849) आगंतुक पटल : 825
इस वज्ञ को इन भाषाओ ंम पढ़: Urdu , ही