Loading...
Estimating historical concentrations of per- and polyfluoroalkyl substances with groundwater flow and transport models and iterative ensemble smoothing analysis
Pedraza, Isabella A.
Pedraza, Isabella A.
Citations
Altmetric:
Editor
Date
Date Issued
2025
Date Submitted
Keywords
Collections
Files
Research Projects
Organizational Units
Journal Issue
Embargo Expires
Abstract
Per- and polyfluoroalkyl substances (PFASs) are a class of contaminants of considerable concern due to their toxicity and persistence in the environment. The accurate prediction of PFAS transport in the subsurface remains a significant challenge in numerical modeling, due to uncertainty surrounding the processes that govern PFAS transport such as mass loading mechanisms from the vadose zone to groundwater. This research aims to address this uncertainty by exploring the simulated controls on PFAS fate and transport in the vadose zone, using the numerical model HYDRUS, and integrating simulated PFAS mass-loading to the water table into groundwater flow and transport models developed with MODFLOW 6 to investigate PFAS migration in groundwater. We use data from and downgradient of the Peterson Space Force Base, situated southeast of Colorado Springs in El Paso County, Colorado, USA. At Peterson, fire-fighting training activities using aqueous film-forming foams (AFFF) have resulted in PFAS contamination of the Fountain Aquifer groundwater, a vital source for the region’s drinking water, primarily with perfluorooctane sulfonic acid (PFOS), perfluorooctanoic acid (PFOA), and perfluorohexane sulfonate (PFHxS). To investigate this contamination, we built a HYDRUS-1D model for a 50-year period simulating variably saturated flow and PFAS fate and transport with a 30-year contamination period from 1970 to 2000, and a 20-year post-contamination period from 2000 to 2020. The purpose of this model is to create a simple yet representative simulation of variably saturated PFAS transport in the vadose zone at this site to gain a general understanding of PFAS loading to groundwater in the absence of clear records on AFFF-contamination events at the site (frequency, volume, mass, application area, etc.). This model is not intended to be a precise, predictive tool but rather a conceptual-deterministic model used to simulate an initial input of PFAS mass loading into the groundwater model. The resulting concentration profiles at the water table served as solute concentration boundary conditions for a MODFLOW 6-based groundwater flow and transport model. We calibrated hydraulic conductivities and recharge multipliers using observed head data and developed a solute-transport model for PFOS, PFOA, and PFHxS. Transport parameters—including bulk density, porosity, immobile porosity, distribution coefficient (Kd), and mass loading—were estimated and refined using Iterative Ensemble Smoothing (IES), a data assimilation approach that quantifies uncertainty and improves model fit. This integrated modeling framework successfully reproduced key spatial and temporal patterns of PFAS concentrations in downgradient municipal well fields and provided insight into the primary controls on PFAS fate in the subsurface. Although the model tended to underpredict observed concentrations, especially for PFOA and PFHxS, it offers a valuable approach for assessing PFAS exposure through groundwater pathways. The trend of under prediction for PFOA and PFHxS may be attributed to PFAS precursors whose transformations from precursor to PFOA or PFHxS were not captured in this model. By addressing vadose-zone transport, PFAS loading to the water table, and subsurface transport uncertainties, this work contributes to improving the understanding of historical PFAS contamination so that the impact of contamination on humans can be better quantified.
Associated Publications
Rights
Copyright of the original work is retained by the author.