Аннотация:Water erosion is one of the most widespread degradation processes on the globe. There are many models for modeling this process. One of the most used is the RUSLE-2 model, a modified version of the Universal Soil Loss Equation (USLE) model. One of the important parameters of this model, the K-factor is the soil erodibility factor, which reflects the resistance of the soil to the manifestation of erosion processes. It represents the relationship of such soil parameters as soil texture, organic matter content, permeability, and soil structure of the soil. Within the framework of this work, the K-factor values were calculated for 289 points located in the steppe province of Russia (Volgograd region). The Empirical Bayesian Kriging regression prediction method was used to correlate spatial data such as soil maps and parent rock, digital elevation model (DEM), length and steepness factor of soil erosion (LS-factor), map of landscapes to develop a high-resolution K-factor map. Average number K-factor was estimated 0,573 t ha h ha−1 MJ−1 mm−1 with standard deviation 0,088 t ha h ha−1 MJ−1 mm−1.