Forecast of anthropogenic fire danger for vegetation according to satellite images of the south of the Russian Far East

In this study, a spatial – temporal analysis of separate vegetation fires monitoring data from 2013 to 2023 was carried out on the basis of MODIS satellite images and geographical data from the Open Street Map resource for the subjects in the Russian Far East south. Open thematic maps of the study area are represented by vector layers of settlements and highways. The map spatial objects are subdivided into 12 categories, according to the number of population in the localities (hamlet, village, town) and the type of highways (unclassified, tertiary, secondary, primary, trunk). To create the deterministic probabilistic methodology for predicting the vegetation fires occurrence, the author has proposed an algorithm for determining the anthropogenic load during specified time periods at a fire season in forest and non-forest areas located at a distance up to 12 km near highways and up to 9 km from settlements. The decomposition of multilinear highway objects in the form of a regular network with a resolution of 0.025 degrees in the WSG84 cartographic projection is performed by the author. It is defined the settlements and highway sections in need of forest protection measures – for federal and municipal organizations of the Jewish Autonomous region to perform them. The longest sections of roads characterizes by a high fire danger in April and October are situated in Leninsky and Oktyabrsky districts; the highest fire danger is recorded near the settlement of Yekaterino-Nikolskoye. The methodology verification was made on the example of the fire season in 2023. It was revealed a satisfactory dependence of vegetation fires occurrence in the selected sections with a length of up to 9 km.