Abstract:

In order to apply the new method in the evaluation of soil environmental quality, which can overcome the disadvantages of the national environmental quality standards (GB 15618-1995, China) and Nemerow comprehensive pollution index. The method is based on the BP neural network model combined GIS technology. The author analyzed the spatial distribution of soil heavy metal pollution in Baotou City of China. The concentrations of 8 heavy metals (As, Cd, Cr, Cu, Hg, Ni, Pb and Zn) were measured in 221 plough layer (0-20 cm) soil sampling in Baotou. The field investigation of 221 topsoil samples were statistically analyzed to show that the study areas were mainly Pb, Zn pollution. It was important to consider the study area geographical features. So the author made the learning samples of BP neural network based on the statistical results and data specificity in study areas. The BP soil environmental quality evaluation model was designed by using the pollution value. According to the evaluation results obtained by the Kriging interpolation method, the author drew Baotou soil environmental quality thematic charts, also the spatial characteristics of Baotou soil environment analysis was included. It was found that the enrichment of heavy metals in topsoil was very obvious in industrial areas and regions near both sides of the Kundulun River. The results indicated that, according to sampling statistics information, the method using BP neural network model could effectively establish special research area soil through non-linear mapping relation between heavy metal contaminate and environmental quality.