Hyperspectral Inversion Of Heavy Metals In Tailings And Its Application Simulation Of Biochar Remediation Effect

China is rich in metal mineral resources and has a wide range of types.However,with the mining,heavy metal pollution in mining areas and surrounding areas is becoming more and more serious,and tailings is an important source of pollution.The tailings pond is a place to store waste and sewage left after ore separation.In order to study the heavy metal pollution in tailings pond,a vanadium-titanium magnetite tailings pond in Chengde was taken as the research object.The tailings samples were collected by five-point method,and the spectral information was collected and the heavy metal content was determined,and then the inversion model was established.The remediation of vanadium titanium magnetite tailings with different doses of biochar was further studied,and the tailings after remediation were analyzed,and the inversion model of heavy metal content was established.The results show that :1)After differential transformation,the hyperspectral of vanadium-titanium magnetite tailings improves its correlation with the contents of Cu,Zn,Cr,Cd and Pb.In addition,the accuracy of the inversion model has also been significantly improved,but the overall accuracy is low.Only the Cu and Zn partial least squares models have high accuracy.The highest accuracy is the Zn spectral first-order differential transformation partial least squares model,and the modeling accuracy and prediction accuracy are 0.69 and 0.58,respectively.2)By applying biochar to repair vanadium-titanium magnetite tailings,it was found that 1.5 % biochar had the best inhibitory effect on DTPA-extracted Cu and Zn.5 %biochar had the best passivation effect on Cu activity extracted by BCR method,and 1.5 %biochar had the best passivation effect on Zn activity extracted by BCR method.3)Through the hyperspectral inversion modeling of tailings after biochar remediation,it is found that different forms of Cu and Zn inversion models have high accuracy.The accuracy of the modeling set of all models is generally above 0.7.The accuracy of the prediction set of the partial least squares model is generally above 0.6,and the highest accuracy is above 0.9.The hyperspectral inversion modeling research on the heavy metal content of original tailings and tailings after biochar restoration provides a method reference for large-scale prediction of heavy metal content of tailings,which is of great significance to the ecological restoration monitoring of mining areas.Figure 20;Table 21;Reference 64...