S. Drozd, N. Kussul and A. Shelestov, "Satellite-Based Analysis of Forest Damage in Ukraine’s Protected Areas," 2025 IEEE 13th International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications (IDAACS), Gliwice, Poland, 2025, pp. 56-60, doi: 10.1109/IDAACS68557.2025.11322406.

Abstract: Ukraine’s protected forests face growing threats from military actions, illegal logging, and economic activities, requiring continuous monitoring. Current methods are often complex and resource-intensive, limiting scalability and frequent use. This study introduces an innovative satellite-based approach for rapid forest damage detection, providing a scalable, resource-efficient solution with minimal data and computational needs. We propose an anomaly detection method using the B4 spectral band in Sentinel-2 imagery, which has been experimentally identified as the most effective for detecting forest damage in Ukraine. Our findings show that the proposed method effectively distinguishes between undamaged and damaged forest areas. We identified significant forest destruction in protected areas of eastern, northern, and southern Ukraine, particularly in regions affected by active hostilities. However, the analysis also uncovered damaged forest areas in the Carpathian region and western Ukraine, suggesting that other factors such as economic activities and illegal logging may also contribute to forest degradation. Among the most affected protected areas are Siverskyi Donets River Valley (forest damage as of 2024 - 55.59 th. ha), Chornobyl Radiation and Ecological Biosphere Reserve (13.14 th. ha), Sviati Hory National Park (13.12 th. ha), Kreminski Lisy National Park (8.62 th. ha), and Oleshky Sands National Park (8.27 th. ha). Overall, by 2024, 249.71 th. ha of forest in protected areas had been damaged, accounting for approximately 5.17% of the total protected forest area in Ukraine.