Water abstraction is a key aspect when performing drinking water risk assessment within the scope of Water Safety Plan (WPS). In this context, evaluating aquifer vulnerability could be not exhaustive. Care must be taken to consider the well structure and dynamic behavior that can determine different water quality and vulnerability than those of the tapped aquifer. Here, a methodological approach to assess well vulnerability to natural and anthropogenic pollution is proposed. A data-driven hydrogeochemical conceptual modeling was performed through cluster analysis, supporting the identification of the main features determining well vulnerability. Well vulnerability was then assessed through a semi-quantitative method based on a well-specific analysis of the tapped aquifer type, water table depth, vadose zone permeability, natural redox conditions, and qualitative classification of the tapped groundwater age, considering evidence of mixing processes induced by the well dynamic behavior. Each feature was categorized into several classes, and a score was associated with each class according to the vulnerability induced by that specific condition. The method was applied to a dataset of ∼200 wells, in northern Italy. Each well was classified, and its vulnerability was assessed. The method was validated by analyzing the distribution of wells affected by natural or anthropogenic pollution in each of the vulnerability classes. Results of the chi-square test showed that the distribution of the polluted wells was associated with vulnerability classes, which is evidence of the robustness and reliability of the proposed method. This method was developed exploiting only hydrogeological information and raw water quality data, largely available to water managers from routine monitoring, which makes this approach widely applicable and cost-effective.

在水安全计划 (WPS) 范围内进行饮用水风险评估时，取水是一个关键方面。在这种情况下，评估含水层脆弱性可能并不详尽。必须仔细考虑井的结构和动态行为，这些井的结构和动态行为可以确定与自来水含水层不同的水质和脆弱性。在这里，提出了一种评估井对自然和人为污染的脆弱性的方法。通过聚类分析进行了数据驱动的水文地球化学概念建模，支持识别确定井易损性的主要特征。然后通过半定量方法评估井的脆弱性，该方法基于对已抽出的含水层类型、地下水位深度、渗流层渗透率、自然氧化还原条件、考虑到井动态行为引起的混合过程的证据，对开采的地下水年龄进行定性分类。每个功能都分为几个类别，并且根据该特定条件引起的脆弱性与每个类别相关联的分数。该方法应用于意大利北部约 200 口井的数据集。每口井都进行了分类，并评估了其脆弱性。该方法通过分析每个脆弱性类别中受自然或人为污染影响的井的分布情况进行了验证。卡方检验结果表明，污染井的分布与脆弱性等级相关，这证明了所提出方法的稳健性和可靠性。