Spectral Characterization of Post-Fire Landscape Changes in Combat Zones Using Multitemporal Satellite Imagery

Combat-related landscape fires require spatially explicit assessment because active-fire detections, spectral burn severity, thermal anomalies, and vegetation recovery are represented at different spatial and temporal scales. This study develops an integrated geographic information system and remote sensing workflow for the Borodianka Territorial Community, Kyiv Oblast, using Sentinel-2 surface-reflectance imagery, Sentinel-3 land-surface-temperature products, and VIIRS active-fire detections, with meteorological and atmospheric reanalysis used as contextual evidence. The normalized burn ratio, differenced normalized burn ratio, and normalized difference vegetation index were calculated, and burn severity was interpreted using a four-class burned-area reflectance classification. VIIRS clusters identified the location and southwestward displacement of active burning, whereas Sentinel-2 raster analysis delimited a broader fire-affected envelope of 600-700 ha. Moderate to very high spectral damage occupied 280-370 ha, and mean vegetation greenness decreased by 35%. The strongest dNBR responses formed spatially coherent northern and central clusters that corresponded to the active-fire corridor. One year later, the share of critically altered NBR pixels had declined substantially, while higher NBR and NDVI classes expanded, indicating pronounced spectral recovery. The results demonstrate that multi-temporal spectral indices and GIS-based classification provide a reproducible framework for mapping fire extent, differentiating severity zones, and monitoring post-fire recovery where field access is constrained.