The use of geostatistical methods to analyze the transition feasibility to the differential application of agrochemicals technologies

In recent years, the tasks of precision farming, including the technology of differential application of agrochemicals, which significantly save resources, increase yield and product quality, while reducing environmental impact, have been particularly relevant and promising. However, for each specific agricultural territory, an assessment of the prospects for the transition to such technologies is necessary; it will not always be justified. The paper proposes a method for solving this problem, based on the use of variogram analysis. The idea is based on a geostatistical model of territory heterogeneity, where the spatial variability of the studied parameter $Z$ is represented as the sum of three components: $m$ — macrocomponent reflecting global spatial variations of a parameter caused, for example, by landscape features; $s$ — mesocomponent that describes the variability of the parameter within the scale of the agricultural field; $\varepsilon$ — microcomponent characterizing random micro-scale variability of the parameter. It is assumed that for the effectiveness of the transition to the differential application of agrochemicals in a specific agricultural territory, it is necessary that the in-field variation of agroecological indicators make a significant contribution to the overall picture of field heterogeneity. The paper also presents a computational experiment demonstrating the efficiency of the proposed method, using aerial photographs, GIS programs, and the programming language R.