Identification Of Soil Heavy Metal Spatial Distribution And Sources Based On Geostatistics And Data Mining

Soil plays an important role in the whole ecosystem,and it is also the place where most of pollutants are eventually detained.Among the many pollutants,the pollution of the heavy metal elements to the soil is particularly prominent.Because of the special properties of heavy metals,it causes great harm to soil,and it is difficult to remediate.Therefore,understanding the status of heavy metal pollution and exploring its sources of pollution are prerequisites for the prevention and environmental governance.In China,the Yangtze River Delta is one of the most soil polluted areas with development of economy.This paper selects south of the Yangtze River Delta as the study area,using statistical analysis,spatial analysis,source apportionment and other technical means to explore spatial variation and pollution sources of Cr,Pb,Cd,Hg,As,Cu,Zn,Ni.The main research contents and results are as follows:(1)Statistical analysis of heavy metals in soil and pollution assessmentStatistical analysis showed that except Cr,the average of heavy metals content in the study area was higher than that in the soil background value.Heavy metal Cr-Ni,Cr-As,Pb-Hg,Zn-Cu have high correlation.In different environmental factors,except for river distance,soil parent material,soil type,organic matter,pH,fertilizer application rate,land use type,traffic density,population density and number of industrial enterprises all have different degrees of influence on heavy metal content in soil.The evaluation results show that Cr,Pb,As,Cu and Zn have Less pollution,polluted sites accounted for 0.73%,0.05%,0%,3.71%,1.37%;Cd and Ni have moderate pollution,polluted sites accounted for 6.54%,7.51%;pollution degree of Hg is high,polluted sites accounted for 30.72%.(2)Spatial analysis of heavy metals in soilSemi variance analysis showed that coefficients of nugget of Pb,Cd,As,Hg and As in the is less than 25%,with a strong spatial correlation,the other three kinds of heavy metals are greater than 25%but less than 75%,with a moderate spatial correlation.Trend surface analysis showed that Cd,As,Cu,Zn and Ni had no obvious difference in East and west directions,and the other three heavy metals have obvious trend difference.The spatial interpolation analysis showed that the spatial distribution of Cr,As,Cu and Ni are similar,mainly distribute in the urban center and the coastal area.The distribution of heavy metal Cd and Zn contents was scattered,mainly distributed in the central and northern areas and some parts in southern.(3)Source analysis of heavy metals in soilPrincipal component analysis shows that Cr,As and Ni are the first principal component,controlled by natural sources;Cd,Cu and Zn are second main components,the non-point pollution source control;Pb and Hg are the third principal components,caused of industrial sources.Finite mixture distribution model show that the model does not recognize the sub distribution of heavy metal Cd,Zn and Cu,namely the main pollution sources have no obvious difference;Cr,Pb,As and Ni have two distributions,namely there are two kinds of main pollution sources in the study area;Hg has three sub distribution,namely there are three main sources of pollution.Random forest analysis showed that natural factors such as soil parent material,and surface pollution sources such as fertilizer and population gathering were the main factors for heavy metal accumulation in the study area.Besides,industrial pollution had a great influence on the enrichment of Hg.(4)Soil heavy metals probability of pollutionUsing the indicator Kriging method,the pollution probability of Cr,Pb,As are less than 0.1 in most of the areas;the pollution probability Cd and Cu,Zn and Ni are more than 0.5 in some parts;the pollution probability of Hg is higher,even more than 0.7 in parts area based on the standard of soil environmental quality standard.According to the threshold of finite mixed distribution function,Cr,Pb,As and Ni are high probability in most areas of the study area,indicating that human activities have begun to affect the accumulation of heavy metals in these place.The multivariate indicator Kriging interpolation method combined with the toxic coefficient show that the probability of heavy metal accumulation in the most areas of the study area is less harmful to the environment.The conclusion of this study can preliminarily reveal the heavy metal pollution status,spatial distribution and main pollution sources of soil in the study area,so as to provide effective support for subsequent heavy metal pollution prevention and governance in soil.