Grantee Research Project Results
2021 Progress Report: Predicting Drinking Water Contamination from Extreme Weather to Reduce Early Life Contaminant Exposures
EPA Grant Number: R840181Title: Predicting Drinking Water Contamination from Extreme Weather to Reduce Early Life Contaminant Exposures
Investigators: Hochard, Jacob
Current Investigators: Hochard, Jacob , Clark, Kayla , Collier, David , Curtis, Scott , Etheridge, Randall , Kruse, Jamie , Peralta, Ariane
Institution: University of Wyoming
Current Institution: University of Wyoming , East Carolina University , The Citadel
EPA Project Officer: Aja, Hayley
Project Period: December 1, 2020 through November 30, 2023 (Extended to November 30, 2024)
Project Period Covered by this Report: December 1, 2020 through November 30,2021
Project Amount: $799,952
RFA: Contaminated Sites, Natural Disasters, Changing Environmental Conditions and Vulnerable Communities: Research to Build Resilience (2019) RFA Text | Recipients Lists
Research Category: Drinking Water , Endocrine Disruptors , Human Health , Safer Chemicals , Sustainable and Healthy Communities , Children's Health
Objective:
Researchers are working on a multidisciplinary (atmospheric science, economics, ecological engineering, design, pediatrics, microbiology, soil ecology) approach to (1) predict groundwater contamination that leads to human exposures, (2) based on predictive models, engage with county health offices to notify at-risk households with a newborn and (3) assess the impact of the risk messenger on risk mitigation choices. Investigators hypothesize: (1a) chemical concentrations and coliform bacteria in wells relates to proximity to contaminated sites during precipitation events, (1b) bacterial contamination depends on source and seasonality, (1c) homes with aging wells are more vulnerable to contaminants, (3) risk communication from an ECU pediatrician promotes households’ risk mitigation behaviors.
Progress Summary:
The Year 1 progress were focused on designing communication interventions for pilot testing and constructing a statewide econometric and machine learning model to predict groundwater contamination events. To achieve this, funds were expended on a PhD student, research scientist, faculty summer salary support and computing power. While the final design for communication messaging intervention will continue to be refined, a pilot design has been completed. The project team has focused on and completed the geospatial and statewide synthesizing of North Carolina surface contamination sources (e.g., coal ash ponds, poultry and swine confined feeding operations, landfills, etc.) and external stressors (e.g., rainfall intensities, temperature, soil qualities, flooding events, topography, local drainage basins, etc.). An initial predictive model is under construction and reveals strong predictive power between specific surface sources and downstream private well arsenic and bacteria contamination rates. The predictive power is particularly strong during hot spells and intensive rainfall – i.e., Hurricanes Irene and Matthew. The team has also made progress setting up the hydrological model, SWAT+, for the Cape Fear River basin (Figure 1) during the first reporting period. This includes collecting data from all the United States Geological Survey and North Carolina Department of Environmental Quality monitoring stations in the watershed (Figure 2).
Figure 1:Map of the Cape Fear River basin with animal feeding operation locations.
Figure 2: Map of the Cape Fear River basin with observation station locations.
Inputs to the model included a digital elevation model (DEM), land use maps, soil maps, and weather data. With these inputs, the model setup utility divides the larger watershed into sub watersheds and even smaller hydrologic response units (HRUs). The research team has put substantial effort into balancing accuracy and resolution with model run time. At the end of this reporting period, running the model for the calibration period from 2003 to 2010 takes approximately 12 hours on a desktop computer. A comparison of the model output to the observed data shows the model underestimates peak discharge and overestimates discharge during low flow periods. This indicates that the model is simulating more storage, either in surface reservoirs or through groundwater, than is occurring in the watershed. This can be adjusted through many model parameters such as increasing the curve number for multiple land uses. Investigators are currently in the process of calibrating the model to flow/discharge and nitrogen concentrations at multiple observation locations in the watershed (Figure 2).
Future Activities:
The project team has an annual and in-person team meeting scheduled in Greenville, North Carolina March 15th, 2022, that will be attended by all PIs and our stakeholders from the NC Department of Health and Human Services, NC Department of Environmental Quality, and various non-profit organizations. In year 2 the team will be finalizing the communication design and transition to predictive model testing and collecting water samples from at-risk households. The predictive model will continue to be refined and integrated with outputs from our hydrological and atmospheric modeling teams. The team will also launch its dedicated web platform in Year 2 as investigators transition from modeling to private well water sample collection and analysis.
Journal Articles on this Report : 2 Displayed | Download in RIS Format
Other project views: | All 3 publications | 3 publications in selected types | All 3 journal articles |
---|
Type | Citation | ||
---|---|---|---|
|
Hochard J, Li Y, Abashidze N. Associations of hurricane exposure and forecasting with impaired birth outcomes. NATURE COMMUNICATIONS 2022;12(1):6746 |
R840181 (2021) |
Exit |
|
Hochard J, Sbashidze N, Bawa R, Ethridge R, Li Y, Peralta A, Sims C, Vogel T. Air temperature spikes increase bacteria presence in drinking water wells downstream of hog lagoons. Science of the Total Environment 2023;867, doi: 10.1016/j.scitotenv.2023.161426. |
R840181 (2021) R836942 (Final) |
Exit |
Supplemental Keywords:
Community-based risk management, averting behavior, predictive modelingProgress 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.