| Accurate characterization of soil and groundwater pollution at non-ferrous metal smelting sites is a prerequisite for efficient and low-cost remediation.However,the hidden nature of pollution generated by smelting activities,the highly discrete hydrogeological structure,the uncertainty of reaction processes between heavy metals and carriers,and the complexity of multi-phase transport of solids/liquids/gases in soil and water environments hinder the fine characterization of heavy metal pollution in soil and groundwater at smelting sites.This leads to a series of problems in site remediation projects,such as difficult delimitation of the pollution scope,unclear pollution transmission laws,ineffective prediction of diffusion trends,poor applicability of remediation strategies,and difficult coupling of remediation technologies.These issues seriously constrain the risk prevention and control and remediation of heavy metal pollution at non-ferrous metal smelting sites.In response to these challenges,this paper focuses on the problem of heavy metal pollution in soil and groundwater at smelting sites and proposes a systematic research approach,including grasping pollution characteristics,elucidating pollution causes,reconstructing pollution in 3D,clarifying pollution transmission,achieving trend prediction,and ultimately forming targeted and intelligent pollution prevention and control measures.The main achievements are as follows:(1)Based on the precise and multidimensional investigation of polluted sites,the spatial distribution and vertical migration of heavy metal pollution in soil on the smelting site were analyzed.It was found that Cd,Zn,As,Pb,and Hg were the main pollutants in the zinc smelting site.The pollution distribution pattern showed a high degree of heterogeneity at a macro level,and the main reasons were concealed pollution sources and complex geological structures.The analysis of spatial variability revealed that pollutants had a high degree of spatial autocorrelation,showing a short-distance radiation pattern with the pollution source as the center,characterized by "short-distance variable distribution" and "patchy aggregation." The analysis of vertical migration of pollution found a clear "vertical fault distribution" due to the effects of geological structures and changes in groundwater levels,with differences in pollutant content between the surface and deep layers reaching several orders of magnitude.(2)The influences of six factors at three levels,namely pollution transmission pathways,smelting production activities,and hydrogeological characteristics,on the formation of pollution patterns at the site were quantitatively analyzed.It was found that all six driving factors had statistical significance at the 0.05 level,indicating that the formation of heavy metal pollution at the site was the result of multiple factors acting together.Factors A(distance from potential pollution sources such as slag,q = 0.59)and D(soil permeability,q = 0.58)were identified as key driving factors for the formation of pollution distribution patterns."Leakage and waste storage" during the production process were identified as the key factors in pollution formation,and soil permeability was determined to be the dominant factor affecting distribution by influencing the migration and diffusion of pollutants.Hg pollution formation was dominated by factor F(distance from chimney,q = 0.69),and the secondary deposition of Hg-containing waste gas during the smelting process was identified as the key reason for Hg pollution formation.(3)The application of 3D modeling technology in the study of contaminated sites was extended,achieving the integrated analysis and 3D reconstruction of multi-source environmental information of the site.The modeling of geological structure reflects the differences in the geological conditions of the site and intuitively shows the three-dimensional interweaving and spatial distribution of the underground space,and identifies the mixed fill layer as the key restoration layer.The method of delineating pollution diffusion boundaries based on permeability differences effectively and accurately determines the boundaries of soil restoration,revealing that Cd pollution has a large radiation range.For the first time,a model of multi-metal composite pollution plume was established for the site,indicating that multi-metal composite pollution is dominant,accounting for 62.7%.The ecological hazard index model indicated that Cd and Hg,due to their high toxicity coefficients and high concentrations in extreme high concentration samples,are the most important ecological hazard contributing factors in the zinc smelting site,with contribution rates reaching 45.2% and 37.6%,respectively.(4)Based on large-scale and systematic long-term monitoring experiments,the migration and diffusion patterns of soil-groundwater pollution in the site are revealed from two aspects: the total amount and speciation of heavy metals.It is found that the migration of heavy metals in the vadose zone of the site is a multi-phase and multi-component seepage process,and the migration mode is dominated by vertical preferential flow in the unsaturated zone.After entering the saturated aquifer,heavy metals mostly exist in their ionic and dissolved forms in groundwater and are transported along with groundwater by convection or diffusion.Based on morphological analysis,the potential migration ability of heavy metals is ranked as Cd(52.6%)> Zn(32.2%)> Pb(22.5%)> As(3.7%).The exchangeable fraction of heavy metals has a strong positive correlation with their concentration in groundwater,and the exchangeable heavy metals in the soil are most likely to be released into groundwater,playing an important role in radial migration,which is the main reason for groundwater pollution formation.(5)A three-dimensional variably saturated solute transport numerical model of the soil-groundwater system is established to effectively predict the diffusion trend of the pollution in the smelting site.It is found that the risk of heavy metal pollution spreading outside the site is extremely high.The pollution can break through the geographic boundary of the site on the60 th day and affect downstream rivers on the 712 th day.Under the influence of the regional groundwater flow field,the Cd pollution plume continues to migrate for 953 m,and the simulated pollution range reaches1,971,940 m2.The numerical simulation results prove that the deployment of a zero-valent iron(ZVI)-bentonite-based permeable reactive barrier(PRB)is an effective way to reduce Cd pollution diffusion and prevent it from spreading to the river.It can delay the arrival time of the pollution plume to the river(225 days),reduce the diffusion range of the pollution plume(by 10.2%),and lower the concentration of the pollution plume entering the river(by 62%).However,due to the characteristics of a wide pollution range,high pollution concentration,and complex geographic boundary of the smelting site,there is still 31-42% of Cd pollution lateral transport occurring.Therefore,the correct remediation strategy should be to first remove or solidify the existing pollution sources and use measures such as PRB to block further pollution.(6)An intelligent selection system for classification,zoning,and remediation technologies for smelting site pollution based on machine learning algorithms was established.A case database of 215 domestic and foreign remediation cases for nonferrous smelting sites was constructed,covering five types and 23 projects.A pollution classification and zoning method for the site was proposed,which couples the total amount of heavy metals,bioavailability,and groundwater pollution levels.The clustering results of the Gaussian mixture model showed that the proportions of land without pollution,mildly polluted,moderately polluted,and heavily polluted were 45.02%,11.27%,32.43%,and 11.29%,respectively.Key factors were identified for determining the remediation strategy for nonferrous smelting sites using a neural network algorithm,including the area,pollutant types,groundwater pollution rate,soil pollution classification,pollution depth,groundwater depth,surrounding residential areas,and remediation project scale.The intelligent selection technology for remediation technology was verified using the scenario,and relying on the verification results,a comprehensive design of the remediation project was achieved.The coupling relationship of the remediation technology was discussed to support the implementation of the remediation project. |
PDF Full Text Request