Assessing groundwater vulnerability using logistic regression

Determining the likelihood that groundwater contains elevated concentrations of contaminants can help water resource managers protect drinking water supplies. For example, this information is useful for selecting new sites for drinking water sources and designing more cost-effective monitoring strategies for existing sources. Groundwater vulnerability has typically been assessed using largely qualitative methods and expressed as relative measures of risk. In this study, a statistical approach was used to quantify the likelihood that a well contains an elevated concentration of nitrate or a detectable concentration of atrazine.

The occurrence of elevated nitrate concentrations or detectable concentrations of atrazine in groundwater was related to both natural and anthropogenic variables using logistic regression. The variables that best explain the occurrence of elevated nitrate concentrations were well depth, surficial geology, and the percentages of urban and agricultural land within a radius of 3.2 kilometers of a well. Well depth and roadside application of atrazine best explained the occurrence of detectable concentrations of atrazine. From these relations, multiple logistic regression models were developed which predict the probability that a well has an elevated nitrate concentration or a detectable concentration of atrazine.