Monitoring land-cover dynamics is important for understanding spatial transformations of territories affected by rapid environmental and socio-economic changes. The Makariv Territorial Hromada in Kyiv Oblast was selected to assess land-cover dynamics during 2021-2025 using annual 10 m land-use/land-cover data derived from Sentinel-2 imagery. The study aimed to quantify changes in the main land-cover classes, identify dominant interclass transitions, and assess their temporal frequency. Annual raster datasets for 2021-2025 were harmonized to a common projection, extent, spatial resolution, and pixel grid. The areas and shares of seven land-cover classes were calculated each year. Post-classification comparison was used to generate transition matrices for consecutive years and for the direct 2021-2025 comparison. A change-frequency map was produced to determine how many times each pixel changed class during the study period. Crops and Trees dominated the land-cover structure throughout 2021-2025. The largest transitions occurred between Trees, Crops, and Rangeland. The recurrent Crops–Rangeland transitions may reflect changes in agricultural land use, including temporary non-use and subsequent return to cultivation, whereas Trees–Rangeland transitions may partly reflect classification uncertainty. A direct comparison between 2021 and 2025 showed that 91.6% of the territory retained the same land-cover class. The five-year frequency analysis showed that 85.1% of the territory did not change class in any annual interval, while only 2.1% changed three or four times. The results indicate overall land-cover stability combined with localized and recurrent interclass variability, highlighting the need for cautious interpretation of automatically classified annual land-cover products.
Brown, C. F., Brumby, S. P., Guzder-Williams, B., Birch, T., Brooks Hyde, S., Mazzariello, J., Czerwinski, W., Pasquarella, V. J., Haertel, R., Ilyushchenko, S., Schwehr, K., Weisse, M., Stolle, F., Hanson, C., Guinan, O., Moore, R., & Tait, A. M. (2022). Dynamic World, near real-time global 10 m land use land cover mapping. Scientific Data, 9, 251. https://doi.org/10.1038/s41597-022-01307-4
Karra, K., Kontgis, C., Statman-Weil, Z., Mazzariello, J. C., Mathis, M., & Brumby, S. P. (2021). Global land use/land cover with Sentinel 2 and deep learning. In 2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS (pp. 4704–4707). IEEE. https://doi.org/10.1109/IGARSS47720.2021.9553499
Kurach, T., Korenets, O., Lytvynenko, N., & Pidlisetska, I. (2023). Degradation of land covers in the Kharkiv region due to military actions. Bulletin of Taras Shevchenko National University of Kyiv. Military-Special Sciences, 4(56), 50–56. https://doi.org/10.17721/1728-2217.2023.56.50-56
Kussul, N., Shelestov, A., Yailymov, B., Yailymova, H., Lemoine, G., & Deininger, K. (2025). Assessment of war-induced agricultural land use changes in Ukraine using machine learning applied to Sentinel satellite data. International Journal of Applied Earth Observation and Geoinformation, 140, 104551. https://doi.org/10.1016/j.jag.2025.104551
Myroniuk, V., Weinreich, A., von Dosky, V., Melnychenko, V., Shamrai, A., Matsala, M., Gregory, M. J., Bell, D. M., & Davis, R. (2024). Nationwide remote sensing framework for forest resource assessment in war-affected Ukraine. Forest Ecology and Management, 569, 122156. https://doi.org/10.1016/j.foreco.2024.122156
Pyvovar, P. V., Topolnytskyi, P. P., Skydan, O. V., & Yanchevskii, S. L. (2023). GIS-based land-use/land cover change analysis: A case study of Zhytomyr Region, Ukraine. Space Science and Technology, 29(4), 24–42. https://doi.org/10.15407/knit2023.04.024
Skakun, S., Justice, C. O., Kussul, N., Shelestov, A., & Lavreniuk, M. (2019). Satellite data reveal cropland losses in South-Eastern Ukraine under military conflict. Frontiers in Earth Science, 7, Article 305. https://doi.org/10.3389/feart.2019.00305
Xie, G., & Niculescu, S. (2021). Mapping and monitoring of land cover/land use (LCLU) changes in the Crozon Peninsula (Brittany, France) from 2007 to 2018 by machine learning algorithms (support vector machine, random forest, and convolutional neural network) and by post-classification comparison (PCC). Remote Sensing, 13(19), 3899. https://doi.org/10.3390/rs13193899