【作者】 陈莹；

【导师】 吴丰昌； 潘波；

【作者基本信息】 昆明理工大学 ， 环境工程（专业学位）， 2023， 硕士

【摘要】 土壤重金属污染一直是土壤科学研究的热点问题,而合理地进行土壤重金属风险评估,则是治理土壤重金属污染的前提。土壤重金属污染评估过程中,离散检测数据的土壤重金属生态和健康风险评估往往受采样数据空间覆盖度有限、检测数据随机波动等问题影响,使评价结果具有较大的不确定性。相对而言,基于地方区域的土壤重金属概率统计以概率的形式解释污染等级的占比,能更有效指导风控方案的制定、污染的修复和场地的管控。本研究综合应用Monte-Carlo模拟统计手段和Arc GIS空间插值手段,结合Hakanson潜在生态风险指数法和健康风险评价模型,分别搭建Monte-Carlo-Hakanson概率评估模型和Monte-Carlo概率健康风险评估模型,对云南省内某小流域内农田土壤中的8种重金属(Cd、Hg、As、Pb、Cr、Cu、Ni和Zn)进行区域性土壤重金属生态和健康风险评估,并通过Arc GIS插值手段对区域内8种重金属的含量和污染概率大小进行插值。得到结论如下:(1)通过Monte-Carlo-Hakanson概率评估模型对研究区域进行概率生态风险评价,分别从污染系数C_f~i、风险指数E_r~i和环境风险综合指数RI三个指标判定区域内的污染严重程度,并通过敏感性分析来表征区域内各重金属污染程度的贡献占比。研究表明区域内金属Cd和Cu的污染最为严重,从概率层面的结果来看,金属Cd处于非常重污染等级的概率接近70%,处于极高风险的概率超过了50%。且8种金属中,只有金属Cd属于高风险等级,其敏感性占比达到了0.99。RI结果可知,场地为高生态风险的概率达到了40%,应引起有关部门的重视。值得注意的是,通过Monte-Carlo模拟后的金属Cu所处的污染等级与以均值作为评判标准得到的污染等级有所不同,可以认为,Monte-Carlo模拟能弥补样本数据不足带来的误差,增加生态风险评估的精确性。(2)基于8种重金属在研究区域内的污染情况,对这8种重金属的含量大小和污染系数C_f~i概率大小进行空间插值。比较普通克里金插值法和IDW逆距离加权插值法对各个重金属含量的插值结果,发现IDW逆距离加权插值法更适合该研究区域。因此,运用IDW逆距离加权插值法对各重金属污染系数C_f~i概率大小进行空间插值。比较含量插值结果与概率插值结果,发现概率插值结果更符合区域实际情况。且通过概率插值结果进行了简单的源识别:与历史遗留Cd危险废物密切相关的几种重金属为Cd、As、Pb、Cr、Cu、Ni和Zn。与生活垃圾处理厂密切相关的金属为Cd、Hg、As、Pb、Cu、Ni和Zn。与畜牧养殖场密切相关的金属为Hg。与历史遗留Cu矿区污染密切相关的金属为Cd、Hg、Pb、Cu和Ni。(3)通过Monte-Carlo健康风险评估模型对研究区内的儿童和成人两类人群进行概率健康风险评估,并在暴露评价阶段引入Monte-Carlo模拟,对暴露评价中两类人群、三种不同暴露方式的暴露量进行了比较。在风险表征中对非致癌和致癌风险作出了评估,得到各重金属产生非致癌和致癌风险的概率。在暴露评价过程发现,儿童的暴露量要远远高于成人。在风险表征阶段发现,每种金属对成人产生的非致癌风险都在1%以内,但Hg和Cr对儿童产生非致癌风险的概率却分别为10%和15%。所有金属对儿童的非致癌风险是成人的30倍,且对于儿童来说,Hg和Cr的危害较高。在产生致癌风险的几种金属中,金属Pb引起的风险需要重视,它对成人和儿童的致癌风险产生的概率都接近100%。在产生非致癌风险的金属中,Hg和Cd对成人的非致癌风险影响最大,Hg和Cr对儿童的非致癌风险影响最大。本研究将数理统计方法与空间插值方法结合,并应用于区域性风险评估,既在理论层面扩宽了概率方法的应用,实现了空间概率插值;又在实际应用层面为风险评估提供了新方法,同时Monte-Carlo模拟可一定程度上规避风险评估过程中的随机性,为评估提供更为准确的结果。更多还原

【Abstract】 Soil heavy metal pollution has always been a hot issue in soil science research,while a reasonable soil heavy metal risk assessment is a premise of controlling soil heavy metal pollution.In the process of soil heavy metal pollution assessment,the ecological and health risk assessment of soil heavy metals is often hindered by the limited spatial coverage of the sampling area and fluctuation of the detection data,which resulted in high uncertainty of the evaluation results.The analysis on the statistical characteristics of soil heavy metal distribution will enable a determination of pollution levels based on probability,which can guide the establishment of effective strategies for risk control,pollution remediation,and site management.In this study,Monte-Carlo-Hakanson probability evaluation model and Monte-Carlo probabilistic health risk assessment model were constructed by combining Monte-Carlo simulation statistical method and Arc GIS spatial interpolation method,combined with Hakanson potential ecological risk index method and health risk evaluation model,respectively.A regional ecological and health risk assessment of eight heavy metals(Cd,Hg,As,Pb,Cr,Cu,Ni and Zn)in agriculture soils in a small watershed in Yunnan Province was carried out,and the content and pollution probability of eight heavy metals in the region were interpolated by Arc GIS interpolation.The conclusions are as follows:(1)The probabilistic ecological risk assessment of the study area was carried out by Monte-Carlo-Hakanson probability evaluation model,and severity of the contamination in the area was determined from the three indicators of pollution coefficient C_f~i,risk index E_r~i and environmental risk comprehensive index RI,and the contribution of each heavy metal pollution degree in the area was characterized by sensitivity analysis.The results have shown that metal Cd and Cu are the most polluted in the region,and from the probabilistic results,the probability of metal Cd being in a very heavy pollution level is close to 70%,and the probability of being at a very high risk is more than 50%.Among the eight metals,only metal Cd belongs to the high-risk level,and its sensitivity ratio reaches 0.99.RI results show that the probability of high ecological risk of the site has reached 40%,which should attract the attention of relevant departments.It is worth noting that the pollution level of metal Cu after Monte-Carlo simulation is different from the pollution level obtained by the mean value,and it can be considered that Monte-Carlo simulation can make up for the error caused by insufficient sample data and increase the accuracy of ecological risk assessment.(2)Based on the pollution of eight heavy metals in the study area,the content of these eight heavy metals and the probability of pollution coefficient C_f~i were spatially interpolated.Comparing the interpolation results of the ordinary kriging interpolation method and the IDW inverse distance weighted interpolation method for each heavy metal content,and it was found that IDW inverse distance weighted interpolation method was more suitable for this study area.Therefore,the IDW inverse distance weighted interpolation method was used to spatially interpolate the probability of C_f~i of each heavy metal pollution coefficient.Comparing the content interpolation results with the probability interpolation results,it was found that the probability interpolation results were more in line with the actual situation of the region.And the simple source identification was carried out through the probabilistic interpolation results:Cd,As,Pb,Cr,Cu,Ni and Zn are closely related to historical Cd hazardous wastes.Cd,Hg,As,Pb,Cu Ni and Zn are closely related to domestic waste treatment plants.Hg is closely related to livestock farms.Cd,Hg,Pb Cu and Ni are closely related to contamination in historic Cu mining areas.(3)The Monte-Carlo health risk assessment model was used to conduct probabilistic health risk assessment for children and adults in the study area,and Monte-Carlo simulation was introduced in the exposure evaluation stage to compare the exposure of the two groups and three different exposure methods in the exposure evaluation.Non-carcinogenic and carcinogenic risks were assessed in the risk characterisation,and probabilities of non-carcinogenic and carcinogenic risks were obtained for each heavy metal.During the exposure assessment process,it was found that children were exposed at much higher levels than adults.During the risk characterization stage,it was found that the non-carcinogenic risk of each metal in adults was within 1%,but the probability of Hg and Cr having non-carcinogenic risk in children was 10%and 15%,respectively.All metals are 30 times more likely to be non-carcinogenic in children than adults,and Hg and Cr are more harmful in children.Among several metals that pose carcinogenic risk,the risk caused by metal Pb requires attention,and its probability of carcinogenic risk in adults and children is close to 100%.Among the metals that pose non-carcinogenic risk,Hg and Cd have the greatest impact on non-carcinogenic risk in adults,and Hg and Cr have the greatest impact on non-carcinogenic risk in children.In this study,the mathematical statistical method and the spatial interpolation method are combined and applied to regional risk assessment,which not only broadens the application of probability method at the theoretical level,but also realizes spatial probability interpolation.It also provides a new method for risk assessment at the practical application level,while Monte-Carlo simulation can avoid the randomness in the risk assessment process to a certain extent and provide more accurate results.更多还原