College of Environmental Natural Resource Sciences, Zhejiang University, Hangzhou, 310058, China; Zhejiang Provincial Key Laboratory of Agricultural Resources and Environment, Hangzhou, 310058, China.
Department of Land Resource Management, School of Tourism and Urban Management, Jiangxi University of Finance and Economics, Nanchang, 330013, China.
Chemosphere. 2022 Feb;289:133182. doi: 10.1016/j.chemosphere.2021.133182. Epub 2021 Dec 6.
The integrated analysis of the distribution characteristics, health risks, and source identification of heavy metals is crucial for formulating prevention and control strategies for soil contamination. In this study, the area around an abandoned electronic waste dismantling center in China was selected as the research area. The probabilistic health risks caused by heavy metals were evaluated by the Monte Carlo simulation. Random forest, partial least squares regression, and generalized linear models were utilized to predict heavy metal distributions and identify the potential driving factors affecting heavy metal accumulation in soil. The relationships of spatial variation between the heavy metal contents and environmental variables were further visualized. The results revealed that cadmium (Cd) and copper (Cu) were the primary soil pollutants in the study area and caused high ecological risks. The probabilistic health risk assessment indicated that the non-carcinogenic and carcinogenic risks for all populations were acceptable. However, children are more susceptible to heavy metal soil contamination than adults. The sensitivity analyses indicated that the total contents of soil heavy metals and soil ingestion rate were the dominant factors affecting human health. The random forest model, with R values of 0.41, 0.65, 0.57, 0.71, and 0.58 for Cd, Cu, Ni, Zn, and Pb, respectively, predicted the heavy metal concentrations better than the other two models. The distance to the nearest industrial enterprise, industrial output, and agricultural chemical input were the main factors affecting Cd, Cu, Zn, and Pb accumulations in the soil, and soil pH and soil parent material were the primary factors influencing Ni accumulation in the soil. The visualization results of the geographically weighted regression model showed a significant relationship between soil heavy metal contents and industrial activity level. This study could be utilized as a reference for policymakers to formulate prevention and control strategies for heavy metal pollution in agricultural areas.
对重金属的分布特征、健康风险和来源进行综合分析，对于制定土壤污染防治策略至关重要。本研究选择中国一个废弃电子垃圾拆解中心周围的区域作为研究区域。通过蒙特卡罗模拟评估重金属引起的概率健康风险。利用随机森林、偏最小二乘回归和广义线性模型预测重金属分布，并识别影响土壤重金属积累的潜在驱动因素。进一步可视化重金属含量与环境变量之间的空间变化关系。结果表明，镉（Cd）和铜（Cu）是研究区域的主要土壤污染物，造成了高生态风险。概率健康风险评估表明，所有人群的非致癌和致癌风险均在可接受范围内。然而，儿童比成年人更容易受到重金属土壤污染的影响。敏感性分析表明，土壤重金属总量和土壤摄入率是影响人体健康的主要因素。随机森林模型对 Cd、Cu、Ni、Zn 和 Pb 的 R 值分别为 0.41、0.65、0.57、0.71 和 0.58，对重金属浓度的预测优于另外两个模型。距离最近的工业企业、工业产值和农业化学品投入是影响 Cd、Cu、Zn 和 Pb 在土壤中积累的主要因素，土壤 pH 值和土壤母质是影响 Ni 在土壤中积累的主要因素。地理加权回归模型的可视化结果表明，土壤重金属含量与工业活动水平之间存在显著关系。本研究可为政策制定者制定农业区重金属污染防治策略提供参考。
