SPATIAL AND TEMPORAL DYNAMICS OF RAINFALL EROSIVITY FACTOR IN POLISSYA AND FOREST-STEPPE OF UKRAINE

The article presents the results of research on the study of spatial and temporal variability of the precipitation erosion factor in the period from 1960 to 2023 within the administrative regions of Polissya and Forest-Steppe of Ukraine. MEM-spatial variables were able to explain 80.8% of the variability of the precipitation erosion factor. The ANOVA revealed that 8 canonical axes, which were extracted after the RDA analysis, were statistically significant. The canonical axes represent different spatial patterns of variability of the precipitation erosion factor. The contribution of spatial MEM variables to the explanation of the canonical axes is different, which allows us to identify the hierarchical structure of variability of the main spatial patterns of precipitation in the region. The canonical axes denoting the main spatial patterns of precipitation erosion variability were correlated with soil properties and land cover types. The temporal AEM predictors 4, 17, 25, 29, 32, 39, 44, and 61 were able to statistically significantly predict temporal patterns of precipitation variability within the study area. These temporal predictors were able to explain 25.9% of the variation in the total matrix of precipitation erosion coefficients. The highest explanatory power of the AEM predictors was found for the southern and southeastern regions, and the lowest for the western regions. The forecast for administrative regions was made for the period up to 2060. The spatial and temporal dynamics of the precipitation erosion factor has a complex hierarchical structure, which can be represented as a set of spatial and temporal patterns with a specific ratio of components of different scale levels. In the spatial context, the patterns are a superposition of processes of broad-, medium-, and detailed-scale levels. The combination of these levels, the nature of spatial variability, and the correlation with soil and landscape indicators allows us to formulate hypotheses about the relevant processes that generate spatial patterns of precipitation erosion factors. Obviously, there are three groups of factors that cause natural variability in precipitation erosion. The first group includes factors of geographical nature, the second includes factors caused by soil cover heterogeneity, and the third includes factors caused by landscape cover heterogeneity. The latter also includes factors that are the result of anthropogenic transformation of landscapes, primarily through agricultural activities. The factors of geographical origin are represented by large-scale patterns, soil factors are represented mainly by medium-scale patterns and to some extent by detailed-scale patterns, and landscape factors are represented mainly by detailed-scale patterns and to a lesser extent by medium-scale patterns.