This Is AuburnElectronic Theses and Dissertations

The Role of Data Analytics in Accelerating Assessments of the Impacts of Per- and Polyfluoroalkyl Substances (PFAS) in Source and Drinking Water




Feltman, Meredith

Type of Degree

PhD Dissertation


Civil and Environmental Engineering


Per- and polyfluoroalkyl substances (PFAS) are a large, diverse, and growing family of persistent, mobile organofluorine compounds that are used for a wide variety of industrial and consumer applications, and that are ubiquitous in the environment. As the body of knowledge related to the impacts of PFAS on wildlife and human health increases, regulatory action is rapidly evolving. For example, in June 2022, the United States Environmental Protection Agency (USEPA) released new interim health advisories orders of magnitude lower than the previous advisory for two perfluoroalkyl substances: perfluorooctanoic acid (PFOA) and perfluorooctane sulfonic acid (PFOS). Given the complexity of PFAS issues and the costliness of PFAS data acquisition, this study investigates whether existing data are being fully exploited to address the pressing challenges related to PFAS occurrence in source and drinking water. As the understanding of environmental and human health impacts of PFAS continues to develop, and as substantial resources are allocated to acquiring PFAS occurrence data, accessible methods are needed to investigate the magnitude and extent of occurrence and exposure risks. In this study, a straightforward data analytics approach is developed using the most extensive publicly available national drinking water dataset, which features 221,831 PFAS concentration records, the USEPA’s third Unregulated Contaminant Monitoring Rule round (UCMR3) data, to provide an efficient methodology for informing research directions and stakeholder decision-making. Spatiotemporal analyses were performed at the finest resolution possible to demonstrate how the approach can be used to enhance the current understanding of nationwide PFAS occurrence, visualize exposure risks, and identify geographic regions of most significant concern as new data and risk assessments become available. The dataset was analyzed at multiple spatial scales (national, regional, state, individual water systems, and sample collection points). To the best of the author’s knowledge, the results reported herein are the first to 1) spatially analyze the results of the only currently existing nationwide database of PFAS in drinking water at the individual public water system (PWS) level, 2) evaluate nationwide PFAS exposure through drinking water using the USEPA’s 2022 interim lifetime health advisories for PFOA and PFOS, 3) evaluate the impact of database structure and varying data aggregation methods on analysis outcomes, and 4) demonstrate the utility of existing and pending UCMR data to investigate trends in PFAS occurrence in source water. The results will allow scientists, stakeholders, and regulatory agencies to identify geographic hotspots with the most significant levels of PFAS contamination; better understand nuances in database structure that affect the current understanding of exposure risks; and initiate efficient data management and analysis methods to more fully leverage large quantities of both currently available and future PFAS data.