MO-4506_14908_CA_O_20230418_Barrier sampling

April 18, 2023

Mr. James P. McDorman, P.G.
Environmental Compliance Coordinator
Charlotte Douglas International Airport
5601 Wilkinson Boulevard
Charlotte, North Carolina 28208

Subject: Groundwater Modeling Results
Proposed Hydraulic Cut-off Barrier
Former U.S. Airways Maintenance Hangar
5020 Hangar Road
NCDEQ UST Section Incident No. 14908
Charlotte, North Carolina
WSP Project: 6228-21-0061
Dear Mr. McDorman:
WSP USA Environment & Infrastructure Inc. (WSP) is pleased to submit this letter summarizing the results of
groundwater flow modeling performed for the above-referenced property in Charlotte, North Carolina (Site).
Included in this report is a summary of our understanding of the project background, a brief description of the
groundwater flow modeling objectives and development, and a summary of the modeling results. Model details
are provided an Attachment.
Project Background
U.S. Airways, Inc. (U.S. Airways) previously operated an aircraft jet fuel defueling/refueling system at the U.S.
Airways Maintenance Hangar located at 5020 Hangar Road in Charlotte, North Carolina. The site is located in the
southwest portion of the Charlotte/Douglas International Airport property (Tax Parcel ID 115-211-05). American
Airlines (formerly U.S. Airways) currently leases the maintenance hangar facility from the City of Charlotte and
performs aircraft maintenance at the property. A 1995 release of petroleum occurred at the hangar location. In
December 2018, CLT personnel observed evidence of petroleum seepage from the bank of Coffey Creek
approximately 800 feet north of contaminant source area.
To control the migration of free phase product at the Site and prevent discharge to Coffey Creek, a hydraulic cut-
off barrier (barrier)approximately 400 feet in length is proposed to be installed along the upgradient side of the
creek between monitoring wells MW-8 and MW-5, as depicted on page 3 of the Attachment. A Corrective Action
Design report (dated October 20, 2022) for the barrier was prepared for and submitted to the Charlotte Douglas
International Airport and the North Carolina Department of Environmental Quality (NCDEQ), Division of Waste
Management (DWM), Underground Storage Tank (UST) Section.

WSP USA Environment & Infrastructure Solutions Inc. 2801 Yorkmont Road, Suite 100 Charlotte, North Carolina 28208 NC Geology License: C-247
T: 704-357-8600

Groundwater Modeling Results April 18, 2023
Former U.S. Airways Maintenance Hangar
WSP Project: 6228-21-0061
The UST Section issued a letter approving the design (dated November 30, 2022). To support the design of the
barrier wall, a numerical groundwater flow model was developed by WSP.
Modeling Objectives
The barrier is designed as a near-impermeable wall to prevent migration of free phase petroleum product to
Coffee Creek. As an indirect result, the barrier will also prevent the flow groundwater through the wall, which
could potentially cause mounding of the water table upgradient of the barrier. However, the barrier is designed as
a hanging structure, straddling the water table. The purpose of this is to allow groundwater to flow beneath the
barrier, minimizing potential mounding effects relative to the groundwater table upgradient of the wall. The
purpose for the groundwater flow model is to evaluate the effect the barrier may have on the groundwater table
at the site and to confirm that there will not be significant mounding upgradient of the wall.
Model Development
The groundwater flow model was developed using a combination Groundwater Vistas 8 (GV) graphical user
interface and Environmental System Research Institute ArcMap 10.8.1. The fundamental groundwater flow model
code used is MODFLOW-USG Transport version 1.8.0 (Panday, 2021)1, which is an enhanced version of the
MODFLOW-USG (Panday, Langevin, Niswonger, Ibaraki, & Hughes, 2013)2 and MODFLOW (McDonald & Harbaugh,
1988)3 finite difference model codes. The unstructured grid approach used in MODFLOW-USG Transport is
designed to better solve problems involving refinement of the grid resolution in areas of interest using nested grid
methods as opposed to traditional telescopic mesh refinement (TMR) used in MODFLOW, in which the added
resolution must be extended to the edges of the model grid. The unstructured grid approach also allows for
pinching of model layers, so the model grid layers may more explicitly represent observed geologic layering.
Details related to model development, including the finite element grid, model layering, boundary conditions, and
hydraulic properties are provided in Attachment, pages 3-14.
The model was calibrated to site-specific water level targets using the Parameter Estimation tool, PEST++ Version
5.0.11. Calibration parameters included hydraulic conductivity, drain conductance, GHG conductance, and
recharge. Because hydraulic conductivity data for the Site is limited, pilot points were used to allow for
heterogeneity across the model domain within each model layer and improvement of the calibration. Calibration
results are provided in Attachment, pages 15-20.

1 Panday, S., 2021. USG-Transport Version 1.8.0: The Block-Centered Transport Process for MODFLOW-USG, GSI Environmental,
August 2021. http://www.gsi-net.com/en/software/free-software/USG-Transport.html
2 Panday, S., Langevin, C.D., Niswonger, R.G., Ibaraki, M., and Hughes, J.D., 2013. MODFLOW-USG version 1: An unstructured
grid version of MODFLOW for simulating groundwater flow and tightly coupled processes using a control volume finite-
difference formulation: U.S. Geological Survey Techniques and Methods, book 6, chap. A45.
3 McDonald, M. G., & Harbaugh, A. W., 1988. Chapter A1, A Modular Three-Dimensional Finite-Difference Ground-Water Flow
Model. In Techniques of Water-Resources Investigations of the United States Geological Survey (p. 586). Washington: United
States Geological Survey Printing Office.
Groundwater Modeling Results April 18, 2023
Former U.S. Airways Maintenance Hangar
WSP Project: 6228-21-0061

Barrier Simulation Results
The barrier was added to Layer 3 of the of the calibrated model and the model was rerun. The barrier was
specified as 1.5 feet thick with a hydraulic conductivity of 5.0E-08 centimeters/second (1.4E-04 feet/day). Results
are presented in Attachment, pages 21 and 22. As shown by the residuals presented on page 22 of Attachment,
which are calculated as the simulated water level minus the observed water level at target locations, the model
predicts water levels upgradient of the barrier to rise less than a half foot in most areas, with a maximum increase
of around 0.61 feet immediately adjacent to the wall. No flooding is predicted to occur due to the addition of the
barrier.
Because the primary goal of this modeling exercise is to demonstrate that the barrier will not cause an
unacceptable rise in the groundwater table at the site, an additional sensitivity simulation was performed, where
recharge within the permeable areas of the Site were doubled, increasing it from 4.38 inches/year to 8.76
inches/year. The results of this simulation are provided in the Attachment, pages 23-25. As shown on page 24 of
Attachment, residuals at target locations rose between 0.18 feet beneath the hanger to 1.37 feet near the ditch
without the barrier in place and rose 0.21 feet beneath the hanger and 2.15 feet near the ditch with the barrier in
place. Groundwater levels within Layer 1 are presented on page 25 of Attachment. As shown, there are two small
areas where the model predicts surface flooding to occur. However, this flooding is predicted with or without the
barrier and the degree of flooding did not change with the addition of the barrier. Thus, it is concluded that
recharge at the Site is most likely lower than what is simulated in this sensitivity simulation, because flood is not
generally observed at the site in these areas. It is also concluded that the installation of the barrier will not cause a
significant rise in the water table at the site based on the model simulations performed.
It should be noted that if there are utilities within two feet of current maximum observed water levels near or just
upgradient of the proposed barrier location, then further evaluation should be performed to determine if those
would be affected by a rise in the water table.
We appreciate your selection of WSP for this project and look forward to assisting you further on this project. If
you have questions, please contact us.
Sincerely,
WSP USA Environment & Infrastructure Inc.

Samuel Best Robert C. Foster, L.G.
Senior Water Resource Engineer Associate Geologist
Registered, NC #1335
Cc: Mr. Dan Bowser, L.G., NCDEQ, UST Section
Enclosure
Hanger Barrier Wall Groundwater Modeling Results April 18, 2023
Former U.S. Airways Maintenance Hangar
WSP Project: 6228-21-0061

ATTACHMENT

Groundwater Modeling Overview and Results

Corrective Action Design Support, Charlotte Douglas International Airport
Groundwater Flow Modeling Overview and Results
Groundwater Flow Model
Development
3 April 18, 2023
Model Domain
Site Location
4 April 18, 2023
Finite Difference Grid
10x10 spacing
5 April 18, 2023
Boundary Conditions
Linearly
interpolated
between MW-3 and ditch
Constant between MW-3 and eastern boundary of
the domain
•General Head Boundaries (GHBs)
•Heads set to average surface water elevation (646.6 feet) of the Catawba River
located approximate 15,000 feet from the Site, as determined from LiDAR data for
the area provided by the State of North
Caroline•Hydraulic conductivity based on lithologic
unit•Saturated thickness set to layer thickness
•Constant Head Boundaries (CHB)
•Based on MW-3 groundwater elevation and estimated surface water elevation
within the ditch at the bottom left corner of domain
•Drain Boundaries (DRNs)
•Heads set to bottom of ditch•Hydraulic conductivity based on lithologic
unit•Hydraulic Flow Barriers (HFBs)
•Hydraulic barrier wall added to layer 3
(post-model calibration)•Wall alignment and top and bottom
elevations of layer 3 along the wall alignment consistent with design
•Hydraulic conductivity = 1.4E-04 ft/d (5.0E-
08 cm/s)•Thickness = 1.5 feet
6 April 18, 2023
GHB Head
~15,000 feet
Average Head
= 646.6 feet
7 April 18, 2023
Layer 1 (Discontinuous)
•Fill Unit
•Top defined by current topography
•Bottom defined by historical grading
plan
•Pinched out where <0.5-ft thick
Layer 2 (Discontinuous)
•Native overburden
•Top defined by historical grading plan
•Bottom set top of Saprolite unit defined
based on Site boring logs
•Pinched out where <0.5-ft thick
Layer 3 (Continuous)
•Saprolite Unit
•Top set to top of Saprolite unit defined
based on Site boring logs
•Bottom set to bottom of proposed barrier
wall
Layer 4 (Continuous)
•Saprolite Unit
•Set to bottom of proposed barrier wall
•Bottom set to constant elevation of 640-ft
msl
Model Layering
8 April 18, 2023
Historical vs. Current Ground Surface Elevations
Historical Current
9 April 18, 2023
•Historical surface was higher than current grade in
some areas
•Top of Overburden = Historical surface elevation,
except where historical elevation > current grade, in
which the top of overburden is set to set to 0.1 feet
below the current elevation
Top of Overburden
10 April 18, 2023
•Based on Site boring logs
•As shown, the creek is in direct contact with the
Saprolite unit
Top of Saprolite
11 April 18, 2023
Fill and Overburden Extents and Thickness
12 April 18, 2023
Model Cross-Section
L1
L2
L3
L4
L1
L2
L2
L3
L4
Northwest
Southeast
Row 1Row 180Coffey
Creek
640-ft msl
700-ft msl
690-ft msl
640-ft msl
1,800-ft msl
13 April 18, 2023
Well K (ft/min)K (ft/d)Screened Interval (ft-bgs)
MW-3 1.97E-05 0.028 4-14
RW-6 2.00E-03 2.880 10-20 or 25
PMW-11 5.91E-05 0.085 10-20
Measured Hydraulic Conductivity (K)
•Because transmissivity is unknow, the range of K used for model
calibration was 0.01 to 100 ft/d for the overburden and saprolite
units
•K for the Fill Unit set to 0.01 ft/d
14 April 18, 2023
Recharge
•Pervious area
•0.001 ft/day (or 4.38 in/yr)
•Impervious area
•0.000 ft/day (or 0.00 in/yr)
Model Calibration
15
16 April 18, 2023
•Maximum observed water levels
Calibration Targets
17 April 18, 2023
Initial sensitivity analysis
•K zones by model layer
•Range: 0.001 to 50 ft/d
•GHB conductance
•Range: 1.0 to 1,000 ft2/d
•DRN conductance
•Range: 1.0 to 1,000 ft2/d
•RCH
•0.001 ft/day to 0.002 ft/day (4.38 in/yr to 8.76 in/yr)
Automated calibration with PEST
•Kx pilot points + GHB and DRN conductance
Calibration Process
18 April 18, 2023
Calibration Results
Residuals (Simulated Head –Observed Head)
19 April 18, 2023
Calibrated Hydraulic Conductivity
Layers 1 and 2
Layer 1 Layer 2
L1 Kx (ft/d)
Mean 0.01
Min 0.01
Max 0.01
L2 Kx (ft/d)
Mean 1.35
Min 0.85
Max 12.6
20 April 18, 2023
Calibrated Hydraulic Conductivity
Layers 3 and 4
L3 Kx (ft/d)
Mean 1.77
Min 0.02
Max 38.3
L4 Kx (ft/d)
Mean 1.99
Min 0.01
Max 40.1
Layer 3 Layer 4
Barrier Wall Simulation
Results
22 April 18, 2023
Layer 3
Barrier Wall Simulation Results
Layer 4
Wall
•Simulated water levels at target locations
rose by 0.01 to 0.68 feet
Sensitivity
Analyses
24 April 18, 2023
L4 -With Wall
Sensitivity Analysis: 2X Recharge
L4 -Without Wall
•Without the wall, the water table rose by 0.18 to 1.37 feet at target locations as the result of doubling recharge in pervious areas
•With the wall in place, the water table rose by 0.21 to 2.15 feet at target locations as a result of doubling recharge in pervious areas
Layers 4
25 April 18, 2023
Sensitivity Analysis: 2X Recharge
L1 -Without Wall L1 -Without Wall
•Without the wall, the model predicts minor flooding in two locations as a result of doubling recharge in pervious areas
•With the wall in place, the model does not predict any additional flooding as a result of doubling recharge in pervious areas
Layer 1