PREDICTIVE STATISTICAL MODELS OF SURFACE MELT WATER RUNOFF ON A SLOPE IN THE AZOV ZONE ROSTOV REGION

Purpose: to conduct a statistical analysis of the relationship between a number of natural factors and snow evaporation, melt water surface runoff during the snowmelt period on a slope with ordinary chernozem in Rostov region for various agricultural backgrounds (winter wheat and fallow) for the period 2008–2022. Materials and methods. Methodological recommendations for accounting for surface runoff were used. Statistical data analysis was performed using the MathCAD and Statistica packages. The significance of regression equations was determined using the Fisher criterion. Results. The use of statistical analysis to process the results of multifactorial studies made it possible to identify the influence of individual natural and anthropogenic factors on erosion processes and their interaction. It has been determined that the value of surface runoff on winter wheat crops depends most of all on the amount of precipitation (correlation coefficient r = 0.72...0.80), more with moldboard tillage, and on stubble plowing – on water reserves in snow (r = 0.40...0.45), also more with moldboard tillage. Conclusions. The equations of the relationship between factorial and resultant indicators of various types are presented, allowing to increase the accuracy of forecasting the surface melt water runoff value in crop rotations depending on the influencing factors with different methods of primary soil tillage, within the limits of the obtained experimental data use, which makes it possible to use methods for regulating negative erosion processes timely. The obtained regularities, reflecting the influence of soil cultivation methods and agricultural background on the value of water balance elements during snowmelt, can be used to predict the melt water surface runoff value and application of anti-erosion agrotechnical measures.