GIS-based assessment of fire impact on the landscapes of the Kherson Region

Purpose. To develop a methodology for monitoring landscape changes caused by wildfires using satellite data, the Python programming language, and geographic information systems (QGIS), based on a case study of the Kherson Region. The research focuses on the spatial localization of fire hotspots, analysis of land cover transformation dynamics, and identification of the most vulnerable ecosystems.

Methodology. The study employed remote sensing methods to identify and spatiotemporally detect thermal anomalies and land cover changes, while geographic information system (GIS) techniques were used for the integration of vector and raster data, spatial overlay, and land-use classification. Mathematical and statistical methods, including the normalization of fire intensity indicators relative to the area of administrative districts and the analysis of their temporal dynamics, were also applied. To ensure reproducibility of calculations and to optimize analytical procedures, computer modeling methods were used, based on the Python programming language and SQL queries.

Findings. An automated algorithm for spatial interpretation of wildfire activity was developed, incorporating classification by land cover categories. A significant increase in wildfire frequency was recorded for the period 2021–2024, especially between 2022 and 2024. Forests, wetlands, and urbanized areas were identified as the most affected. A series of fire density maps was generated across administrative districts and land use categories. Spatial analysis confirmed a correlation between military operations and the intensification of fires across different landscapes.

Originality. The study presents a novel methodology for automated wildfire monitoring that integrates open satellite data sources (FIRMS, ESA), GIS tools (QGIS), the Python language, and spatial normalization techniques. For the first time, a region-specific algorithm has been proposed which assesses dynamic changes in land cover caused by wildfires, taking into account land use categories, administrative boundaries, and fire density. The methodology is applicable for regional environmental zoning and further systemic research.

Practical value. The results can be used for monitoring the environmental consequences of military actions, planning post-conflict recovery strategies, implementing conservation measures, identifying priority areas for demining, and assessing risks to human safety. The methodology is scalable, adaptive, and can be used for research of other regions.