| With the development of green sustainable development concept and popularization,the attention to environmental quality is getting higher and higher,the research on soil quality which is closely related to human life,production and resource utilization has become one of the hot spots.The comprehensive evaluation of soil heavy metal pollution can provide important basis for soil management and protection.However,the existing mature and widely used evaluation models are mainly based on the content of heavy metal elements,there is a lack of comprehensive consideration of the actual terrain features of the evaluation area,therefore,a dynamic and predictive soil quality evaluation is not possible.So based on KPCA and GIS,this paper constructs a new comprehensive evaluation model of soil heavy metal pollution,specific methods and recognition are as follows.On the basis of systematic research on the evaluation object and evaluation method of soil heavy metal pollution,for the above problems,firstly,we set up an evaluation factor set including heavy metal element content,toxicity characteristics,topographic elevation and gradient;the element content is a composite index based on KPCA extraction,and the element toxicity is characterized by an integrated potential ecological hazard index,and the elevation and gradient of topography are based on the elevation of GIS extraction.Secondly,the analytic hierarchy process is used for the judgment matrix constructed by the expert to establish the weight vector.Then,based on the fuzzy comprehensive evaluation method,a soil heavy metal pollution degree evaluation model including heavy metal element content,toxicity,elevation and slope is constructed.Finally,based on statistical analysis of soil heavy metal elements in Yijiawan area,a comprehensive evaluation of pollution degree was carried out using the constructed model.The results show that there are heavy metal pollution in the Yijiawan area,the single element Cd pollution is the most serious,the spatial element diversity is obvious,and there are many sources of pollution.KPCA based on gaussian radial basis kernel function is more suitable for comprehensive information extraction of heavy metal elements in Yijiawan area,evaluation model based on this not only follow the premise of that element content is the main control factors of soil heavy metal pollution degree,and reflects the potential toxicity,elevation and slope on heavy metal pollution degree of influence.For the Yijiawan area,the evaluation model makes an accurate assessment of the area affected by the element content,at the same time,the predictive evaluation of local area is also made,for example,the soil in the flat areas downstream of Xiangjiang river may be seriously polluted.The result of this prediction is of great significance to the soil management and protection in Yijiawan area. |