Procopio Nicholas A, Atherholt Thomas B, Goodrow Sandra M, Lester Lori A
Division of Science, Research and Environmental Health, New Jersey Department of Environmental Protection, 428 East State St., Trenton, NJ, 08625-0420.
Ground Water. 2017 Sep;55(5):722-735. doi: 10.1111/gwat.12518. Epub 2017 Mar 29.
The influence of precipitation on coliform bacteria detection rates in domestic wells was investigated using data collected through the New Jersey Private Well Testing Act. Measured precipitation data from the National Weather Service (NWS) monitoring stations was compared to estimated data from the Multisensor Precipitation Estimate (MPE) in order to determine which source of data to include in the analyses. A strong concordance existed between these two precipitations datasets; therefore, MPE data was utilized as it is geographically more specific to individual wells. Statewide, 10 days of cumulative precipitation prior to testing was found to be an optimal period influencing the likelihood of coliform detections in wells. A logistic regression model was developed to predict the likelihood of coliform occurrence in wells from 10 days of cumulative precipitation data and other predictive variables including geology, season, coliform bacteria analysis method, pH, and nitrate concentration. Total coliform (TC) and fecal coliform or Escherichia coli (FC/EC) were detected more frequently when the preceding 10 days of cumulative precipitation exceeded 34.5 and 54 mm, respectively. Furthermore, the likelihood of coliform detection was highest in wells located in the bedrock region, during summer and autumn, analyzed with the enzyme substrate method, with pH between 5 and 6.99, and (for FC/EC but not TC) nitrate greater than 10 mg/L. Thus, the likelihood of coliform presence in domestic wells can be predicted from readily available environmental factors including timing and magnitude of precipitation, offering outreach opportunities and potential changes to coliform testing recommendations.
利用通过《新泽西州私人水井检测法案》收集的数据，研究了降水对家庭水井中大肠菌群检测率的影响。将国家气象局（NWS）监测站的实测降水数据与多传感器降水估计（MPE）的估计数据进行比较，以确定在分析中使用哪一个数据源。这两个降水数据集之间存在很强的一致性；因此，采用MPE数据，因为它在地理上对各个水井更具针对性。在全州范围内，发现检测前10天的累计降水量是影响水井中大肠菌群检测可能性的最佳时期。建立了一个逻辑回归模型，根据10天的累计降水数据以及其他预测变量（包括地质、季节、大肠菌群分析方法、pH值和硝酸盐浓度）来预测水井中大肠菌群出现的可能性。当之前10天的累计降水量分别超过34.5毫米和54毫米时，总大肠菌群（TC）和粪大肠菌群或大肠杆菌（FC/EC）的检测频率更高。此外，在基岩地区的水井中，在夏季和秋季，采用酶底物法进行分析，pH值在5至6.99之间，并且（对于FC/EC而非TC）硝酸盐大于10毫克/升时，大肠菌群检测的可能性最高。因此，可以根据包括降水时间和降水量在内的现成环境因素来预测家庭水井中大肠菌群存在的可能性，这为开展宣传工作以及对大肠菌群检测建议进行潜在调整提供了机会。
