UC Irvine

Per- and polyfluoroalkyl substances (PFAS) are synthetic chemicals widely used in industrial and consumer applications due to their thermal and chemical stability. Their persistence and mobility have caused widespread contamination of drinking water, soil, and biota, with PFAS exposure linked to immune dysfunction, reproductive disorders, and cancer. Their environmental distribution is shaped by complex factors including hydrology, geochemistry, and industrial activities. Traditional monitoring methods are costly, labor-intensive, and lack the scale required for effective management. To address these challenges, machine learning (ML) and artificial intelligence (AI) enhance the scalability and predictive power of PFAS monitoring and risk assessment. This dissertation applies big-data analytics with ML and AI models to improve PFAS data evaluation, prediction, and environmental risk analysis. It focuses on three key areas: (1) PFAS in California groundwater, (2) PFAS in wastewater treatment plant (WWTP) effluents and biosolids, and (3) nationwide PFAS evaluation across the U.S. As one of the first efforts to combine ML and big-data approaches for PFAS fate analysis, this work advances scalable, efficient strategies for long-term PFAS pollution control and risk mitigation.