Optimization Software Tool of Variogram Model for Spatial Estimation in Soil

Many existing variogram theoretical models which include nonlinear model, linear model and hole effect model had been existed. However, in the process of estimating the spatial distribution of soil heavy metals, the selection of interpolation method, variogram fitting and parameter setting all depended on human subjective experience, which greatly affects the accuracy and reliability of spatial estimation. Based on the mixed programming of Java and python, this study realized the optimization software tool of variogram model for heavy metals spatial estimation variogram in soil, which integrated DS evidence theory and deep learning algorithm, and realized the optimal selection of six mainstream spatial interpolation algorithms, optimal fitting of variogram and automatic parameter setting. The application had showed that the system is simple and easy to use, which could avoid the interference of human experience in the process of spatial estimation, improves the accuracy and accuracy of spatial estimation, and could provide good tool support for researchers in the field of spatial prediction of soil element content.