Improving the methodology of on-farm land management in the direction of transition from the formation of work sites to the formation of management zones for the specific requirements of the agricultural producer upon implementation of precision farming is extremely important for the agricultural sector of the Belarusian economy. The article presents the results of applying the methods of geostatistical and multifactor geoinformation analysis for the formation of management zones within the limits of land use of RUE “Uchkhoz BGSHA” (Republic of Belarus, Mogilev region, Gorky district). The total area of the surveyed territory is 83420.1 hectares. The nature of the spatial distribution of data on the content of humus, mobile phosphorus and potassium in the soil as well as pH level was estimated using the tools of the Spatial Statistics module of ArcGIS version 10.5. The presence of reliable clustering of data on soil parameters was established, since the value of the global Moran index I ranged from 0.197827 to 0.360388, and the z-score in all cases exceeded 2.58. The universal kriging method turned out to be the most suitable for modeling the spatial distribution of soil pH data, while the empirical Bayesian kriging method is the most acceptable when modeling the spatial distribution of the content of humus, phosphorus, and potassium in the soil. The method of principal components and the simple summation of rasters using a calculator proved to be suitable for identifying management zones by a set of soil parameters (the discrepancy with the actual area was 16.56 and 16.24 ha, respectively).
Read full abstract