Risk Zoning Of Heavy Metals In Soils Of Mountainous Areas

Heavy metal pollution has become a particularly prominent problem of cultivated land pollution.According to the spatial distribution and influencing factors of soil heavy metal pollution,selecting social and ecological factors for regional soil heavy metal risk zoning is the key to regional soil heavy metal risk management and control.In this study,the risk classification of heavy metals in mountain soils was established based on the analysis of factors affecting heavy metals,spatial mapping and risk assessment.Based on the establishment of risk zoning of heavy metals in mountain soils,taking Hezhang as an example,this paper uses statistical analysis,spatial analysis and neural network to demonstrate and analyze the risk zoning.The main conclusions are as follows:(1)Framework for Risk Zoning of Heavy Metals in Soils in Mountain AreasBased on the characteristics of mountainous areas,this paper chooses social and ecological factors as the systematic framework to study the risk classification and zoning of heavy metals in mountainous soils,and obtains the framework method for the risk zoning of heavy metals in mountainous soils,which is based on the analysis of influencing factors of heavy metals in soils,the prediction and mapping of heavy metals in mountainous soils,the environmental risk assessment and the social and ecological factors.(2)Influencing factors of cadmium content distribution in agricultural soils in the study areaBased on the risk zoning framework of heavy metals in mountainous soils,a case study of cadmium in Hezhang soils was conducted.The results showed that the average Cd content in the study area was 2.15mg/kg,which was significantly higher than the background value(0.659mg/kg),and its coefficient of variation was 72.6,showing a moderate coefficient of variation.The contribution rate of each influencing factor to the distribution of Cd content in the study area was obtained by using the geographic detector method.The factors influencing the distribution of Cd content from large to small were Precipitation > Organic Matter > Elevation > Road Distance > Population Spatial Distribution > Water Distance > pH > Soil Type > Geological Type > Land Use Type..The results were basically consistent with Pearson's correlation analysis.Interactive detection shows that the interaction between organic matter and precipitation has a greater impact on the distribution of cadmium content in soil,and the interaction is a non-linear synergistic effect.Through the analysis of the distribution characteristics of Cd content under different environmental factors,it was found that the contents of heavy metals in soils varied greatly under different altitudes,organic matter,pH,soil types and geological types.(3)Spatial Distribution Characteristics of Cadmium in Agricultural Soils in the Study AreaThe spatial trend analysis of heavy metal cadmium in soil showed that Cd in the study area had a second-order trend in the east-west direction and a uniform distribution in the East-West direction,and a first-order trend in the North-South direction,and Cd increased from south to north.By introducing auxiliary variables into residual Kriging of BP neural network and Kriging interpolation of multiple linear regression,cross validation shows that residual Kriging interpolation of BP neural network has better spatial prediction ability for soil cadmium distribution,while ordinary Kriging and multiple regression residual Kriging can only roughly estimate the distribution of soil cadmium in the study area.In the interpolation simulation results of RBPNN,the high and low value patches are more dispersed,which highlights the fluctuation of the numerical value,and can more objectively express the local heterogeneity of the sudden spatial distribution of heavy metal cadmium with respect to the original sample data.The spatial distribution map of soil cadmium content showed that the distribution of heavy metal cadmium was low in the northern part of the study area,and high in the southern part,especially in the southwestern part.Based on the spatial distribution map of heavy metal pollution in soil,the potential ecological risk of cadmium in the study area was analyzed and characterized by potential ecological risk assessment method.(4)Study on Soil Risk Regionalization in Research AreasUsing self-organizing mapping network model,seventeen different social-ecological factors were clustered at the township level.According to the clustering results and the characteristics of each cluster,the study area was divided into six clusters.Among them,the southwest and North regions of the study area had the highest risk level of heavy metals,followed by the eastern region and the north-central region,while the north and northeast regions had the lowest risk level of heavy metals.