Assessment of military impacts on protected areas of Ukraine using Sentinel-1 and machine learning

doi:10.26565/1992-4224-2026-45-13

L. A. Horoshkovа National university of “Kyiv-Mohyla academy”, 2, Skovorody, Str., Kyiv, 04070, Ukraine https://orcid.org/0000-0002-7142-4308
O. I. Menshov Taras Shevchenko National University of Kyiv, 60, Volodymyrska, Str., Kyiv, 01033, Ukraine https://orcid.org/0000-0001-7280-8453
D. V. Maslov National university of “Kyiv-Mohyla academy”, 2, Skovorody, Str., Kyiv 04070, Ukraine https://orcid.org/0009-0009-7397-8329

DOI: https://doi.org/10.26565/1992-4224-2026-45-13

Keywords: Sentinel-1, SAR, Forest disturbance, affected forests, Remote sensing, Machine learnin, Radar Vegetation Index

Abstract

Purpose. To assess military impacts on protected areas of Ukraine using Sentinel-1 Radar Vegetation Index (RVI) data and machine learning methods in order to identify spatial and temporal patterns of vegetation disturbance and ecosystem transformation under wartime conditions

Methods. Spatial and temporal changes are analyzed using remote sensing techniques combined with machine learning methods, including unsupervised classification algorithms to detect patterns of vegetation disturbance and ecosystem transformation. Additionally, comparative analysis and time-series analysis are applied to assess the impact of military activities on forest ecosystems under wartime conditions.

Results. This study assesses the impact of military activity on forest ecosystems in eastern Ukraine using Sentinel-1 SAR data, the Radar Vegetation Index (RVI), baseline-relative change analysis, and unsupervised machine learning. The primary objective was to detect, quantify, and characterize war-related forest disturbance in the Serebrianskyi Botanical Reserve which is directly exposed to active military operations and to understand the extent, severity, and temporal dynamics of that damage relative to a pre-conflict baseline. A conflict-free control site, Homilsha Forests National Nature Park, was used to distinguish military-driven change from background ecological variability. The study addresses whether Sentinel-1 RVI, VV, and VH backscatter can capture the spatial patterns and progressive development of military-induced forest disturbance over the period 2020–2025. Sentinel-1 data were processed in Google Earth Engine and restricted to forest pixels using a land-cover mask. Annual summer composites were generated for each year, and a pre-conflict baseline (2020–2021) was used to quantify post-disturbance change. The analysis encompassed annual RVI trend assessment, rule-based damage classification, K-means clustering, and detection of isolated forest anomalies. After 2022, Serebrianskyi ROI showed a marked RVI decline from stable values in 2020–2022, with changed forest pixels in 2025, including severely disturbed pixels increasing. Homilsha ROI remained stable, and no deterioration trend. Machine-learning results were consistent.

Conclusions. SAR methods have proven to be effective for remote monitoring with limited field access, although derived categories of damage should be interpreted as remote sensing indicators and not as field validated categories of damage.

Downloads

Download data is not yet available.

Author Biographies

L. A. Horoshkovа, National university of “Kyiv-Mohyla academy”, 2, Skovorody, Str., Kyiv, 04070, Ukraine

DSc (Economy), Prof., Professor of the Department of Ecology

O. I. Menshov , Taras Shevchenko National University of Kyiv, 60, Volodymyrska, Str., Kyiv, 01033, Ukraine

DSc (Geology), Senior Researcher of the Department of Geoinformatics

D. V. Maslov , National university of “Kyiv-Mohyla academy”, 2, Skovorody, Str., Kyiv 04070, Ukraine

PhD Student of the Department of Ecology

References

Maksymenko, N. V., Voronin, V. O., & Burchenko, S. V. (2023). Remote Monitoring of the Impact of Military Actions on Forest Landscapes of the Kharkiv Region. Man and Environment. Issues of Neoe-cology, (40), 20–32. https://doi.org/10.26565/1992-4224-2023-40-02

Chornohor, L. F., Nekos, A. N., Titenko, H. V., & Chornohor, L. L. (2022). Mathematical Models for Assessing the Environmental Consequences of the Pyrogenic Factor Impact on Forest Ecosystems. Visnyk of V. N. Karazin Kharkiv National University, Series “Ecology”, (27), 51–62. https://doi.org/10.26565/1992-4259-2022-27-04

Hanson, T. (2018). Biodiversity conservation and armed conflict: A warfare ecology perspective. Annals of the New York Academy of Sciences, 1429(1), 50–65. https://doi.org/10.1111/nyas.13689

Lawrence, M. J., Stemberger, H. L. J., Zolderdo, A. J., Struthers, D. P., & Cooke, S. J. (2015). The effects of modern war and military activities on biodiversity and the environment. Environmental Reviews, 23(4), 443–460. https://doi.org/10.1139/er-2015-0039

McNeely, J. A. (2003). Conserving forest biodiversity in times of violent conflict. Oryx, 37(2), 142–152. https://doi.org/10.1017/S0030605303000334

Mahreen, K. (2022). The environmental impacts of war and conflict. Institute of Development Studies. https://opendocs.ids.ac.uk/articles/report/The_Environmental_Impacts_of_War_and_Conflict/26427280

Gaynor, K. M., Fiorella, K. J., Gregory, G. H., Kurz, D. J., Seto, K. L., Withey, L. S., & Brashares, J. S. (2016). War and wildlife: Linking armed conflict to conservation. Frontiers in Ecology and the Envi-ronment, 14(10), 533–542. https://doi.org/10.1002/fee.1433

Ordway, E. M. (2015). Political shifts and changing forests: Effects of armed conflict on forest conserva-tion in Rwanda. Global Ecology and Conservation, 3, 448–460. https://doi.org/10.1016/j.gecco.2015.01.013

Daiyoub, A., Gelabert, P., Saura-Mas, S., & Vega-García, C. (2023). War and deforestation: Using remote sensing and machine learning to identify the war-induced deforestation in Syria 2010–2019. Land, 12(8), 1509. https://doi.org/10.3390/land12081509

Butsic, V., Baumann, M., Shortland, A., Walker, S., & Kuemmerle, T. (2015). Conservation and conflict in the Democratic Republic of Congo: The impacts of warfare, mining, and protected areas on defor-estation. Biological Conservation, 191, 266–273. https://doi.org/10.1016/j.biocon.2015.06.037

Kaplan, G., Rashid, T., Gasparović, M., & others. (2022). Monitoring war-generated environmental se-curity using remote sensing: A review. Land Degradation & Development. https://doi.org/10.1002/ldr.4249

Shevchuk, S. A., Vyshnevskyi, V. I., & Bilous, O. P. (2022). The use of remote sensing data for investi-gation of environmental consequences of Russia–Ukraine war. Journal of Landscape Ecology, 15(3), 36–53. https://doi.org/10.2478/jlecol-2022-0017

Sticher, V., Wegner, J. D., & Pfeifle, B. (2023). Toward the remote monitoring of armed conflicts. PNAS Nexus, 2(6), pgad181. https://academic.oup.com/pnasnexus/article/2/6/pgad181/7185602

Mandal, D., Kumar, V., Ratha, D., Dey, S., Bhattacharya, A., Lopez-Sanchez, J. M., McNairn, H., & Rao, Y. S. (2020). Dual polarimetric radar vegetation index for crop growth monitoring using Sentinel-1 SAR data. Remote Sensing of Environment, 247, 111954. https://doi.org/10.1016/j.rse.2020.111954

Plank, S. (2014). Rapid damage assessment by means of multi-temporal SAR—A comprehensive re-view and outlook to Sentinel-1. Remote Sensing, 6(6), 4870–4906. https://doi.org/10.3390/rs6064870

Vreugdenhil, M., Wagner, W., Bauer-Marschallinger, B., Pfeil, I., Teubner, I., Rüdiger, C., & Strauss, P. (2018). Sensitivity of Sentinel-1 backscatter to vegetation dynamics: An Austrian case study. Remote Sensing, 10(9), 1396. https://doi.org/10.3390/rs10091396

Vreugdenhil, M., Navacchi, C., Bauer-Marschallinger, B., Hahn, S., Steele-Dunne, S., Pfeil, I., & Wag-ner, W. (2020). Sentinel-1 cross ratio and vegetation optical depth: A comparison over Europe. Remote Sensing, 12(20), 3404. https://doi.org/10.3390/rs12203404

De Luca, G., Silva, J. M. N., Di Fazio, S., & Modica, G. (2022). Integrated use of Sentinel-1 and Senti-nel-2 data and open-source machine learning algorithms for land cover mapping in a Mediterranean re-gion. European Journal of Remote Sensing, 55(1), 52–70. https://doi.org/10.1080/22797254.2021.2018667

Hirschmugl, M., Deutscher, J., Sobe, C., Bouvet, A., Mermoz, S., & Schardt, M. (2020). Use of SAR and optical time series for tropical forest disturbance mapping. Remote Sensing, 12(4), 727. https://doi.org/10.3390/rs12040727

Saim, A. A., & Aly, M. H. (2025). Fusion-based approaches and machine learning algorithms for forest monitoring: A systematic review. Wild, 2(1), 7. https://www.mdpi.com/3042-4526/2/1/7

Gatti, R. C., Lobos, R. B. C., Torresani, M., & others. (2025). An early warning system based on ma-chine learning detects huge forest loss in Ukraine during the war. Global Ecology and Conservation. https://www.sciencedirect.com/science/article/pii/S2351989425000289

Mullissa, A., Reiche, J., & Herold, M. (2023). Deep learning and automatic reference label harvesting for Sentinel-1 SAR-based rapid tropical dry forest disturbance mapping. Remote Sensing of Environ-ment. https://www.sciencedirect.com/science/article/pii/S0034425723003504

Xu, C., Ding, Y., Zheng, X., Wang, Y., Zhang, R., Zhang, H., & others. (2022). Comparison of machine learning and feature selection methods for maize biomass estimation using Sentinel-1 SAR, Sentinel-2 vegetation indices, and biophysical variables. Remote Sensing, 14(16), 4083. https://doi.org/10.3390/rs14164083