Comparison analysis of stochastic simulation for the spatial distribution of heavy metal pollution of soil

Semivariograms and its parameters have a significant impact on Kriging interpolation and stochastic simulation. This paper used GS+ and ArcGIS to research the heavy metal of Yichang citrus orchard, which used spherical model, Gaussian model, exponential model and linear model to fit the experimental semivariograms and applied these models in the Sequential Gaussian Simulation. The stochastic simulation results of different models show that the average of Hg is over the national standard of soil quality, so there are some pollution. The best model is exponential model which can keep semivariograms isomorphic before and after the simulation, which is different from the model selected by the largest coefficient of determination, so during stochastic simulation, the appropriate model should be selected by the comparison of the simulation results of various model.