Labite Herve E, Cummins Enda
UCD School of Biosystems Engineering, College of Engineering and Architecture, University College Dublin, Belfield, Dublin 4, Ireland,
Environ Monit Assess. 2015 Mar;187(3):91. doi: 10.1007/s10661-015-4325-9. Epub 2015 Feb 7.
During this study, a groundwater screening tool was developed to assess the temporal risk of groundwater contamination from the use of pesticides. It is based on a source, vector, target approach. The method utilised in this study uses a semi-quantitative probabilistic risk assessment where the input parameters were classified and assigned a relative score from 1 to 5 (i.e. 1 = no risk and 5 = high risk). The model was parameterised by using national data and calibrated with 2 years of national pesticide groundwater monitoring data. After calibration, two specific sites were selected for model validation. Based on the presence of the source, vector and target, the evaluation indicated that the temporal risk is site specific (i.e. May to December for the country model, June to September for the Oak Park site and September for the Castledockrell site). A sensitivity analysis performed on the national scale revealed that the groundwater vulnerability category (gv), the clay content (cc%), the persistence of pesticides in soil (DT50) and the rainfall represented by wet day (wd) were the most important parameters that affected model predictions (correlation coefficients of 0.54, -0.39, 0.35 and 0.31, respectively), highlighting the importance of soil hydrogeological conditions, soil type and rainfall in influencing water model predictions. The model developed can help to identify the temporal risk from pesticides to groundwater and guide regulators in highlighting at-risk periods, therefore allowing more focused monitoring programmes.
在本研究中,开发了一种地下水筛选工具,以评估使用农药导致地下水污染的时间风险。它基于源、媒介、目标方法。本研究中使用的方法采用半定量概率风险评估,其中输入参数被分类并赋予1至5的相对分数(即1 =无风险,5 =高风险)。该模型通过使用国家数据进行参数化,并使用两年的国家农药地下水监测数据进行校准。校准后,选择了两个特定地点进行模型验证。基于源、媒介和目标的存在,评估表明时间风险是特定地点的(即国家模型为5月至12月,橡树公园地点为6月至9月,卡斯尔多克雷尔地点为9月)。在全国范围内进行的敏感性分析表明,地下水脆弱性类别(gv)、粘土含量(cc%)、农药在土壤中的持久性(DT50)以及以湿日(wd)表示的降雨量是影响模型预测的最重要参数(相关系数分别为0.54、-0.39、0.35和0.31),突出了土壤水文地质条件、土壤类型和降雨对水模型预测的重要性。所开发的模型有助于识别农药对地下水的时间风险,并指导监管机构突出风险期,从而实现更有针对性的监测计划。