Spatial analysis of provision the population of Kharkiv with civil defense facilities

Abstract

The article analyzed the territorial aspect of the distribution and capacity characteristics of civil defense structures in the city of Kharkiv, comparing it with the population size. In the first stage, population density was calculated for specific areas within the maximum permissible radius of the nearest available shelter. The spatial characteristics of the location (level of clustering) and capacity of the defense structures were examined for these same areas. Finally, the population size was correlated with the capacity of the civil defense structures, and spatial clustering of the identified areas was conducted based on this attribute. The geostatistical method of spatial interpolation was used to determine population density in specific areas and fill gaps in the primary data. The application of this method required the following sequential procedures: transforming the primary data according to a normal distribution, constructing a semivariogram model of the transformed variables, aggregating the model into a surface, and defining target polygons. During the calculation of the territorial provision of various types of shelters in the city of Kharkiv, the following sources were utilized: a layer of point objects from the interactive map of open data on the Kharkiv Geoportal (to determine the locations and types of protective structures), and information from the website of the non-commercial enterprise «Emergency Medical Care and Disaster Medicine Center» of the Kharkiv Regional Council (to update information on the capacity of storage facilities and anti-radiation shelters). The Global Moran’s Index and Local Moran’s Index are statistical methods used to assess spatial autocorrelation, which is the degree of clustering or spatial pattern in a variable across a defined area. In this case, they were employed to evaluate the spatial autocorrelation of the capacity of civil defense structures in specific parcels in Kharkiv. The Global Moran’s Index indicated a high level of clustering of areas based on this attribute. Using the Local Moran’s Index, parcels were classified into five object classes: High-High cluster (HH), Low-Low cluster (LL), a high-value outlier surrounded by low-values (HL), a low-value outlier surrounded by high-values (LH), and areas without cluster or outliers (non-significant). The scientific novelty of the article, compared to related studies on a similar topic, lies in the utilization and transformation of a hexagonal grid of population density distribution in the city of Kharkiv, in accordance with the research requirements. In the conclusions, based on the results of cluster analysis and through the adaptation of the tectological principle of the weakest link to the realities of the present, a comprehensive sequential strategy for addressing the shortcomings of territorial provision of civil defense structures in ensuring the population of Kharkiv was proposed.