刘泽宇,赵胜豪,廖朋辉,钱瑜.某工业场地砷污染的加密布点优化与精度对比分析[J].环境科学学报,2020,40(9):3415-3421
某工业场地砷污染的加密布点优化与精度对比分析
- Optimization and accuracy comparative analysis of sampling methods in an Arsenic pollution site survey
- 基金项目：国家重点研发计划项目（No.2016YFC0207603）
- 刘泽宇
- 1. 南京大学污染控制与资源化研究国家重点实验室, 南京 210023;2. 南京大学环境学院, 南京 210023
- 赵胜豪
- 南京国环科技股份有限公司, 南京 210042
- 廖朋辉
- 南京国环科技股份有限公司, 南京 210042
- 钱瑜
- 1. 南京大学污染控制与资源化研究国家重点实验室, 南京 210023;2. 南京大学环境学院, 南京 210023
- 摘要：场地污染调查是开展场地风险评估和修复的基础，随着我国颁布建设用地土壤风险管控标准等文件，提出风险筛选值和管制值的分级管控体系，我国土壤污染调查工作对于精确掌握污染物浓度梯度及空间分布有了更高的要求，因而如何构建高效采样布点方法变得尤为重要.本研究以南京市某砷工业污染场地为例，开展了优化布点方法与调查精度的研究.首先，在初查数据的基础上，运用指示克里格的方法预测土壤污染概率，参考城市土壤背景值开展阈值设定，提取加密布点区域，之后结合场地历史生产资料与自然信息布设加密点位，最终将插值预测污染结果与非加密、均匀加密、真实情况结果进行对比，验证方法精度.结果表明：本方法符合国际中基于设计抽样以达到场地调查简单、经济、客观的理念，适用于数据存在明显空间相关性的场地；该方法获得的污染区域面积占比为8.32%，与8.57%的现实情况较为接近，但普通法得到的污染面积比相对未加密情况不升反降，可能的原因是在非污染区域增设点位会造成阈值边缘区域在模拟时往清洁区域偏移，污染区域被误判为清洁，反而造成结果偏差；在目前我国环境管理对超阈范围重点关注的背景下，该方法既能够节约布点成本，又精确化预测场地污染超标范围，为后续的场地修复和治理工作提供帮助.
- Abstract：Contaminated site survey, as the basis of risk assessment and remediation, should be accurate in describing pollutant concentration and spatial distribution because of the stringent Soil Environmental Quality Risk Control Standard in China, which stipulated a multilevel risk control system. Therefore, it's important to study on sampling methods with both high efficiency and high accuracy. In the case of an Arsenic polluted industrial site in Nanjing, indicator kriging method was used to predict pollution probability of the whole site according to the first-round regular investigation with 82 sampling points. Then high pollution area was identified and 20 more sampling points were located in these area. Finally, Arsenic distribution was simulated in four situations including regular sampling, high pollution probability sampling (20 more calculated points), common increased sampling (20 more random points), and fully increased sampling (100 more points), and the investigation accuracy was evaluated. The results show that sampling based on contamination probability forecast is in accordance with the principal of simple, economical and objective, and is an optimized method for sites with large spatial difference of pollution. The proportion of Polluted areas calculated by this method is 8.32%, close to 8.57% obtained from the fully increased sampling, which could be considered as the real situation. While the proportion of common increased sampling is even lower than the regular situation with only first-round sampling, indicating that more sampling points in unpolluted area could make miscarriage and deviation. In conclusion, the sampling method based on kriging interpolation can guarantee the survey accuracy in limited cost, and provide support for contaminated site remediation scientifically.
