Simulation of Regional Groundwater Flow and Advective Transport of Per- and Polyfluoroalkyl Substances, Joint Base McGuire-Dix-Lakehurst and Vicinity, New Jersey, 2018
Links
- Document: Report (7.96 MB pdf) , HTML , XML
- Plates:
- Plate 1 (212 MB pdf) - Forward particle tracks from aqueous film-forming foam source areas 1 to 15 and reverse particle tracks from per- and polyfluoroalkyl substances reconnaissance areas 4 and 14, Joint Base McGuire-Dix-Lakehurst and vicinity, New Jersey, 2018
- Plate 2 (200 MB pdf) - Forward particle tracks from aqueous film-forming foam source areas 16 to 21 and reverse particle tracks from per- and polyfluoroalkyl substances reconnaissance areas 16 to 19, Joint Base McGuire-Dix-Lakehurst and vicinity, New Jersey, 2018
- Data Release: USGS data release - MODFLOW6 and MODPATH7 used to simulate regional groundwater flow and advective transport of per- and polyfluoroalkyl substances, Joint Base McGuire-Dix-Lakehurst and vicinity, New Jersey, 2018
- Download citation as: RIS | Dublin Core
Abstract
A three-dimensional numerical model of groundwater flow was developed and calibrated for the unconsolidated New Jersey Coastal Plain aquifers underlying Joint Base McGuire-Dix-Lakehurst (JBMDL) and vicinity, New Jersey, to evaluate groundwater flow pathways of per- and polyfluoroalkyl substances (PFAS) contamination associated with use of aqueous film forming foam (AFFF) at the base. The regional subsurface flow model spans an area of approximately 518 square miles around JBMDL and is based on a previously developed hydrogeologic framework of the area. Steady-state flow in the unconsolidated aquifers was simulated using the MODFLOW 6 groundwater flow model, which is able to account for hydrostratigraphic pinchouts and discontinuities in the Coastal Plain aquifers underlying JBMDL. To account for local patterns of fluid flow driving advective subsurface migration of PFAS, the grid was refined using quadtree meshes spanning 21 areas where historical AFFF use was identified, five off-site reconnaissance areas identified by AFCEC as areas in which the occurrence of PFAS is most likely to pose a potential danger to local drinking water supplies, and along streams that behave as drains in the base-flow-dominated Coastal Plain.
Following grid refinement, four physical processes known to govern subsurface flow were introduced to the model. These included effective precipitation recharge, discharge to streams and stream-connected wetlands, regional inflows and outflows along the model bottom, and withdrawals from wells, each of which were incorporated into the model as either external or internal boundary conditions. To account for effective precipitation recharge, a specified-flow boundary was assigned along the top of the model. Similarly, regional flows predicted using the modified U.S Geological Survey’s New Jersey Coastal Plain Regional Aquifer System Analysis model were treated as specified-flow boundary conditions along the bottom of the model. Base-flow losses were treated as drains along streams delineated using a 10-foot LiDAR dataset. Drains were also assigned to cells falling within stream-connected National Hydrologic Database wetlands. Finally, well-pumpage data mined from the New Jersey Water Transfer database were added to the model to account for extraction of groundwater through pumping from industrial-supply and drinking-water-supply wells. Along model edges established at groundwater divides, where the net flux of water across the boundary is equal to zero, natural no-flow boundary conditions were imposed.
The refined flow model was calibrated using the parameter-estimation (PEST) program, which adjusts model parameters by performing a gradient search over the sum-of-squared-error objective function until the parameter set that produces simulated water levels and base flows most closely matches 544 water levels and 20 estimated base flows and closely adheres to initial parameter estimates. Based on the analysis of calibration residuals, the model did not appear to be affected by significant model structural error.
The MODPATH particle-tracking algorithm was used to estimate advective transport paths of PFAS in the vicinity of JBMDL. Forward tracking was used to determine paths of PFAS away from AFFF source areas to streams, wetlands, pumping wells, and geographic areas that PFAS may contaminate. Additionally, reverse tracking was used to determine particle pathlines away from off-site PFAS reconnaissance areas, or areas within which all sources of PFAS might be advectively transported into subsurface drinking-water supplies, to locations at land surface that may indicate a source of PFAS.
The coupled and calibrated groundwater flow and particle-tracking transport model provide valuable tools for predicting the relative extent of PFAS contamination from onsite legacy source areas. The calibrated model also provides measures of water-level and base-flow observation influence that can help guide future data-collection efforts related to groundwater and surface water sampling for PFAS.
Suggested Citation
Fiore, A.R., and Colarullo, S.J., 2023, Simulation of regional groundwater flow and advective transport of per- and polyfluoroalkyl substances, Joint Base McGuire-Dix-Lakehurst and vicinity, New Jersey, 2018: U.S. Geological Survey Open-File Report 2022–1112, 41 p., 2 pls., https://doi.org/10.3133/ofr20221112.
ISSN: 2331-1258 (online)
Study Area
Table of Contents
- Acknowledgments
- Abstract
- Introduction
- Description of Study Area
- Data Sources
- Simulation of Regional Groundwater Flow
- Model Calibration
- Regional Groundwater Flow Paths and Advective Transport of Per- and Polyfluoroalkyl Substances
- Limitations of the Regional Model
- Summary
- References Cited
- Appendix 1. Description of Model Layers and Their Thicknesses
- Appendix 2. Approach for Assigning Weights to Calibration Observations
| Publication type | Report |
|---|---|
| Publication Subtype | USGS Numbered Series |
| Title | Simulation of regional groundwater flow and advective transport of per- and polyfluoroalkyl substances, Joint Base McGuire-Dix-Lakehurst and vicinity, New Jersey, 2018 |
| Series title | Open-File Report |
| Series number | 2022-1112 |
| DOI | 10.3133/ofr20221112 |
| Year Published | 2023 |
| Language | English |
| Publisher | U.S. Geological Survey |
| Publisher location | Reston, VA |
| Contributing office(s) | New Jersey Water Science Center |
| Description | Report: ix, 41 p.; 2 Plates: 35.00 x 45.00 inches and 45.00 x 30.00 inches; Data Release |
| Country | United States |
| State | New Jersey |
| Online Only (Y/N) | Y |
| Additional Online Files (Y/N) | Y |
| Google Analytic Metrics | Metrics page |