Kandilli Observatory digitizes 115 years of climate records using AI

Boğaziçi University is using AI to digitize 115 years of handwritten weather records from Istanbul's Kandilli Observatory. The dataset, collected since 1911, will support climate research and disaster planning once complete.

Categorized in: AI News Science and Research
Published on: Mar 30, 2026
Kandilli Observatory digitizes 115 years of climate records using AI

Turkish university digitizes 115 years of climate records with AI

Boğaziçi University's Kandilli Observatory and Earthquake Research Institute is converting more than a century of handwritten meteorological records into digital format using artificial intelligence. The project, launched in April 2023, aims to preserve one of Turkey's longest continuous climate datasets and make it available for research and public use.

The institute has collected temperature, precipitation, atmospheric pressure, humidity, wind, and sunshine duration measurements since 1911. Work is expected to finish by the end of this year.

Ayfer Serap Söğüt, the institute's meteorology laboratory chief, said the digitization effort will improve the accuracy of climate projections. Long-term records allow researchers to track climate change trends with greater precision than shorter datasets.

Atakan Çelebi, Istanbul director of the Turkish State Meteorological Service, called the dataset "Turkey's climate memory." Continuous observations spanning more than a century are rare; data collection at other Istanbul locations has been inconsistent over the same period.

Kandilli's uninterrupted record gives researchers a distinct advantage. Professor Özer Çinicioğlu, the institute's director, said these records belong not to a single institution but to society as a whole.

What the data will support

Once digitized, researchers will reanalyze the records using data analysis and machine learning methods. The effort will support research on climate change, atmospheric dynamics, and extreme weather events.

Experts at the project's public event emphasized that high-resolution long-term datasets are essential for understanding extreme weather and assessing disaster risk. The digitized records will inform decision-making in disaster management, agriculture, water management, and urban planning.


Get Daily AI News

Your membership also unlocks:

700+ AI Courses
700+ Certifications
Personalized AI Learning Plan
6500+ AI Tools (no Ads)
Daily AI News by job industry (no Ads)