Developing a geospatial model to identify private well contamination risk in Alabama

Abstract

In the United States, many people rely on private wells, and a constellation of risk factors affect the nature and severity of well water pollution, including the capacity of the well user to effectively manage their water supply. Identifying well user communities at risk of contaminant exposure remains a complex and underexplored area of research. The primary objective of this research is to better understand the spatial distribution and correlation of risk factors associated with the potential contamination of private well water. We developed a framework to evaluate this “risk-scape” using an unsupervised multivariate clustering approach and spatial autoregressive models to evaluate three key risk factors - socio-economic vulnerability, flood risk, and anthropogenic activity – with well water dependence. Our findings show that approximately 15% of Alabama's communities with high well dependence also have a higher flood risk and a large minority group with population below poverty line while 29% of high well use communities are composed of a high percentage of agricultural land with a large child population. This framework highlights where policy intervention or targeted resource allocation should be focused to mitigate well contamination in these communities. The framework’s flexibility allows for application to any geographical area, offering a pathway for broad adoption.