Mathematical model for predicting the yield of apple trees on rootstock 62-396

Traditionally, forecasting apple tree yields was carried out on the basis of statistics of already obtained long-term yield dynamics or based on generative formations formed on the tree. The purpose of the work is to identify the most significant soil parameters for the formation of the apple tree varieties yield on the dwarf rootstock 62-396 and to build a mathematical model for predicting fruit yield in the conditions of the Central Chernozem region of the Russian Federation. Field research was carried out in 2004-2022 in industrial apple tree plantations at the age of full fruiting in the Lipetsk (two farms) and Tambov (eight farms) regions. Seven horticultural soil types have been studied. Based on statistical analysis, a mathematical model of apple tree productivity was obtained in the form of an analytical dependence on three factors. A geometric interpretation of the regression model is given using the Mathcad symbolic mathematics system in the form of response surfaces and the corresponding level lines. Practical recommendations are given for choosing rational parameters within selected intervals of factor variation.