Grantee Research Project Results
2023 Progress Report: Predicting and Communicating PFAS Exposure Risks from Rural Private Wells
EPA Grant Number: R840081Title: Predicting and Communicating PFAS Exposure Risks from Rural Private Wells
Investigators: MacDonald Gibson, Jacqueline , Salamova, Amina , Redmon, Jenny , Livanapatirana, Chamindu , de Bruin, Wandi Bruine
Institution: Emory University , North Carolina State University , University of Southern California , Research Triangle Institute
EPA Project Officer: Packard, Benjamin H
Project Period: September 1, 2020 through August 31, 2023 (Extended to August 31, 2025)
Project Period Covered by this Report: September 1, 2022 through August 31,2023
Project Amount: $1,584,420
RFA: National Priorities: Research on PFAS Impacts in Rural Communities and Agricultural Operations (2020) RFA Text | Recipients Lists
Research Category: Drinking Water , Water
Objective:
The main purpose of this project is to develop data and tools for managing PFAS risks to private well water. There are three main objectives: (1) build scalable models to predict PFAS exposure risks in rural private wells, (2) conduct a rural citizen-science PFAS monitoring campaign, and (3) create user-friendly maps and tools for PFAS risk communication and management.
Progress Summary:
Objective 1: Build Scalable Models of PFAS Exposure Risks in Rural Private Wells. In this reporting period, RTI completed mechanistic models (model construction and calibration for detailed groundwater flow and physics-based fate and transport models), and NCSU began training probabilistic machine-learned Bayesian Network (BN) models that integrate mechanistic model results. Mechanistic models included a 3-layer numerical groundwater fate-and-transport model, which the research team calibrated for steady state groundwater flow to represent PFAS fate and transport using synthesized available data in the North Carolina (NC) study area. Following the integrated model framework developed during the last reporting period, in this reporting period the NCSU research team has developed new BN models using outputs from the calibrated numerical groundwater model as additional inputs to our previously trained BN model. Results for model performance are being compared under different levels of mechanistic modeling effort: 1) no mechanistic model data (BN previously developed under Task B); 2) medium-level mechanistic modeling effort (groundwater flow model outputs only); and 3) high-level mechanistic modeling effort (groundwater flow and fate and transport model outputs). In this reporting period, the research team also finished the publication process for the submitted manuscript and submitted another manuscript focused on predicting temporal variation in PFAS exposure risks in well water at the Minnesota study site.
Objective 2: Conduct Rural PFAS Citizen-Science Monitoring Campaign. In this reporting period, the research team finalized report-back to 67 participants in Robeson County, North Carolina who were unable to receive results by email. The research team held an in-person workshop for these residents to explain test results, provide recommendations, and answer questions. This concludes test results report-back for the citizen-science campaign. Previously collected agricultural samples from Indiana University have been transferred to Emory University for PFAS analysis. In addition, we have identified six potential farms in Robeson County, NC for soil and runoff sampling and are working to finalize their enrollment.
Objective 3: Disseminate User-friendly Maps and Tools for PFAS Risk Management. During the reporting period, PI Jacqueline MacDonald Gibson co-organized a workshop including 72 participants from local health departments, as well as other public health experts, and water quality experts, to conduct crisis response table-top exercises for water contamination events involving PFAS. In addition, the research team finalized a computational approach to calculating cumulative risk from known/suspected PFAS sources based on nationally representative data sources curated to estimate PFAS risks in private well water. We also developed a first iteration of a nationwide PFAS risk mapping platform, which will be finalized and disseminated in the next reporting period. We also constructed a risk assessment model using Analytica Enterprise software to estimate the contribution of PFAS exposure from drinking water and food compared to other sources in rural areas relying on private wells. Preliminary results show the estimated mean contribution by geographical region and age group of the six pathways to total PFAS exposure (private wells, food, food contact materials, indoor dust, hand-to-mouth and air inhalation). The contribution of private wells to PFAS exposure in different age groups ranged from approximately 1-6 % in Monroe County, IN, to 17-54 % in Washington County, MN.
Future Activities:
Objective 1: Build Scalable Models of PFAS Exposure Risks in Rural Private Wells. Integrated BN and mechanistic model results will be finalized to predict GenX exceedance risk for the North Carolina (NC) study area. A conference presentation and a manuscript will be prepared to disseminate in 2024.
Objective 2: Conduct Rural PFAS Citizen-Science Monitoring Campaign. We will develop a manuscript for peer-review reporting the results of the citizen science sampling campaign. In addition, wastewater, runoff, and soil samples will be collected from multiple Robeson County farms potentially affected by the Chemours facility. All wastewater and agricultural samples will be analyzed for PFAS once the transfer of funds between institutions is complete.
Objective 3: Create User-Friendly Maps and Tools for PFAS Risk Communication and Management. In the next reporting period, we will further develop the risk assessment model to implement consumer products and include additional PFAS (PFBS, PFDA, PFHpA, PFHpS, PFHxA, PFNA, PFPeA, and neutral PFAS more common in air pollution). The risk assessment model will also undergo uncertainty and sensitivity analysis. We will fine-tune our mapping approach and generate nationwide maps of PFAS sources to disseminate in an interactive, online platform. We will also develop a peer-reviewed manuscript based on our data curation and risk calculation approach.
Journal Articles on this Report : 1 Displayed | Download in RIS Format
Other project views: | All 3 publications | 3 publications in selected types | All 3 journal articles |
---|
Type | Citation | ||
---|---|---|---|
|
Li R, Gibson JM. Predicting Groundwater PFOA Exposure Risks with Bayesian Networks:Empirical Impact of Data Preprocessing on Model Performance. Environmental Science & Technology 2023;57(46):18329-38. |
R840081 (2023) |
Exit |
Supplemental Keywords:
PFAS, Bayesian network, well water, drinking water, groundwater modeling, machine-learning, citizen-scienceRelevant Websites:
Progress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.