Burdens of Per- And Polyfluoroalkyl Substances (PFAS) In U.S. Public Water Systems Open Access
Lee, Madison (Spring 2021)
Objective: Identify relationships between the detection of per- and polyfluoroalkyl substances (PFAS) in public water systems and county-level sociodemographic characteristics across the United States.
Methods: Drinking water and concentrations of six per- and polyfluoroalkyl substances were obtained from the United States Environmental Protection Agency’s (U.S. EPA) Third Unregulated Contaminant Monitoring Rule (UCMR3), and county-level characteristics were obtained from the 2010 U.S. Census. The data from these two sources were used to construct classification trees to identify predictors of PFAS detection.
Results: The detection frequency of the six PFAS ranged between 0.05% and 1.00%. With these low detection frequencies, only PFHxS, PFOA, and PFOS data produced classification trees. The main predictors for the PFHxS, PFOA, and PFOS included different measures of household income, facility water type, population size, and residential mobility. Surface water as the facility water type was a common split among all three of the contaminants.
Conclusions: The classification trees were a novel approach to identifying disparities in the detection of PFAS in drinking water; however, the low detection frequencies from the 2012 – 2015 data limited potential subgroups of importance. The same approach should be used in future PFAS drinking water data to better predict which demographic characteristics predict PFAS burden.
Table of Contents
TABLE OF CONTENTS
MATERIALS AND METHODS
PUBLIC HEALTH IMPLICATIONS
TABLES AND FIGURES
About this Master's Thesis
|Committee Chair / Thesis Advisor|
|Burdens of Per- And Polyfluoroalkyl Substances (PFAS) In U.S. Public Water Systems ()||2021-05-02 17:13:30 -0400||