Mapping concentrations of heavy metals in soils using Kriging and RBF Case study: Harris Township

Being informed of spatial distribution of heavy metals concentrations in soil for environmental pollution monitoring and maintaining the quality of the environment is necessary. This study aimed to map the spatial distribution of heavy metal contamination concentration of Manganese، Copper، Zinc and Iron in Harris agricultural soils located in the province of East Azerbaijan done. For this purpose، using systematic random sampling of 370 surface soil samples were collected at a depth of zero to 30 cm، and the concentrations of all these metals were measured. In order to modeling the spatial variation of heavy metal concentrations in the soil، the ordinary kriging and radial basis functions were applied in ArcGIS. Among the various models of ordinary kriging and radial basis functions، the lowest RMSE and MAE values and very close to zero MBE and the highest correlation coefficient R as a criterion for evaluating the best way to model the spatial distribution of heavy metals were considered. Both methods have accuracy، but based on proper the evaluation criteria for metals، Magnesium and Copper، kriging method with exponential model، for Zinc، kriging method with spherical model and for iron، RBF method with Multi-quadric function was chosen. The map after choosing the best model for each of the elements، drawing and taking into account the standard of the Institute of Soil and Water Conservation، re-classifying the elements in four classes deficiency، sufficiency، and infection. Analysis of the spatial distribution maps of heavy metal contamination showed that most of the land was sufficient range for Magnesium; for Copper was found that about 92 percent had a high class and about 8 percent of the study area was contaminated; contamination for zinc and iron wasn’t found and respectively، about 96 and 80 percent were deficient.