Analysis of Building Destruction as a Consequence of Military Actions Based on Multi-Temporal Remote Sensing Data – A Case Study in Moschum and Mariupol Cities, Ukraine

Abstract

The article presents a methodology for analyzing the destruction of buildings caused by Russian military aggression against Ukraine based on multi-temporal remote sensing (RS) data. The study aims to assess the scale of destruction using satellite imagery, geoinformation technologies, and automated data processing algorithms. The research consists of two analytical case studies. The first case examines the destruction of buildings in Moshchun (Kyiv region) by comparing satellite images taken before and after the full-scale invasion. To improve the accuracy of damage assessment, the analysis results were verified using aerial imagery obtained after the de-occupation of the area. It was determined that the number of destroyed or damaged buildings in the village due to hostilities amounted to 663, of which 590 were completely destroyed. The second case focuses on determining the extent of building destruction in Mariupol by comparing multi-temporal satellite images with vector building layers that reflect the state of urban development before the onset of hostilities. The analysis was conducted using image processing algorithms and geospatial analysis implemented in the Python environment. It was established that during the full-scale invasion, 8,349 buildings were destroyed, including 2,013 multi-story buildings, of which 934 were completely demolished.