重金属污染场地土壤风险筛选值关键影响因子研究——以砷为例

合理确定重金属土壤筛选值是污染场地风险识别和调查评估的基础,过松或过严的标准都会增加风险评估的不确定性,甚至可能会导致风险管控措施失效或修复资金浪费,确定重金属土壤筛选值的关键影响因子是合理确定重金属土壤筛选值的前提。选择工业污染场地中检出频率高、毒性大的砷作为研究对象,通过假设工商业用地下的暴露情景,根据《污染场地风险评估技术导则》(HJ25.3—2014)得到砷筛选值计算公式,利用基于蒙特卡罗模拟的Crystal Ball模型计算公式中各参数对结果的敏感性和贡献率来确定关键影响因子,通过调研国内外砷筛选值现状,从计算模型、毒性评估及关键影响因子等方面对引起各国砷筛选值差异的原因进行了分析探讨。结果表明:工商业用地下土壤砷筛选值取值为0.84～175.7 mg·kg~(-1),均值为21.4 mg·kg~(-1),95%的置信上限为24.19 mg·kg~(-1);风险可接受水平、每日土壤摄入量(IR)、暴露频率(EF)和暴露周期(ED)对砷筛选值的贡献率依次为41.3%、-27.3%、-16.3%和-12.7%,其余因子的贡献率均小于1%;关键影响因子按贡献率绝对值从大到小分别为风险可接受水平、每日土壤摄入量(IR)、暴露频率(EF)、暴露周期(ED)。计算模型和毒性评估存在差异是国内外土壤砷筛选值差异的基础原因,当计算模型和毒性评估差异不大时,关键影响因子才是决定性因素,其影响程度与模拟计算的结果一致。建议我国完善筛选值计算模型,对砷的毒性效应和符合我国暴露人群特征的关键影响因子展开深入研究。

Study on Key Factors of Soil Screening Levels of Heavy Metal Contaminated Sites: An Example of Arsenic

Reasonable determination of heavy metal soil screening level (SSL) is the basis for contaminated sites risk identification and assessment. Imprecise standard will increase the uncertainty of risk assessment of heavy metal contaminated sites, and may even result in the failure of site risk management or over-remediation. How to get more reasonable SSL of heavy metal has arised a lot of awareness by stakeholders. Determining the key factors is an important issue for the determination of heavy metal SSL. In order to better understand the SSL of heavy metal, arsenic (As) which is more toxic and frequently detected in contaminated sites was selected as typical representative, and a comprehensive study of the key factors affecting the As SSL was conducted. By assuming the exposure scenario of industrial and commercial land use, a formula for calculating the As SSL was derived according to the "Technical Guidelines for Risk Assessment of Contaminated Sites (HJ25.3-2014)". The Crystal Ball model based on Monte Carlo simulation analyzing the sensitivity and contribution rate of the parameters for the As SSL was employed to determine the key factors of the formula. The results showed that the As SSL was from 0.84 to 175.7 mg·kg-1, the mean was 21.4 mg·kg-1, and the 95% confidence limit was 24.19 mg·kg-1. The contribution rate of acceptable risk levels, soil ingestion rate (IR), exposure frequency (EF) and exposure duration (ED) to As SSL determination was 41.3%, -27.3%, -16.3% and -12.7% respectively, while other factors were less than 1%. Key factors ranged by absolute value from the large to the small were: acceptable risk levels, soil ingestion rate (IR), exposure frequency (EF), exposure duration (ED). The main factors resulted in the differences of As SSL were the model of derivation and toxicity assessment. Suggestion was promoted that China should improve the model of derivation and study on the toxicity of As and the key factors in the future.