|
Subsurface Object Delineation at the SCWA Dune Road Wellfield,
Westhampton, Using GPR.
T. T. Fuller1*,
K. T. Goetz2+, R. G. Bova1
1 Suffolk
County Water Authority, Oakdale, NY *TFuller@scwa.com
2 Stony Brook University, Stony Brook,
NY +KGoetz@ic.sunysb.edu
Abstract
A
subsurface foundation was encountered at a shallow depth during the drilling
of a new public supply well in Westhampton Dunes, Westhampton. In order to delineate the extent of this
object, a tight, high-frequency grid using ground-penetrating radar (GPR) was
surveyed. After collection of data
near the surface extent of the foundation, the lateral limits of the
foundation were mapped. Additionally,
in an adjacent area believed to be clear of subsurface hazards, a smaller
grid showed the area to be relatively free of hazards, allowing the
successful installation of a well.
Introduction
The barrier
beach community known as the Village
of Westhampton Dunes is mostly
comprised of seasonal residences ranging from small cottages to multi-million
dollar homes all situated along a five-mile portion of Dune
Road (Fig. 1).
The vicinity is currently serviced by only one well located at the Dune
Road (South) well field. The area is frequently hampered by
extremely low water pressure, which adversely affects available fire
protection. Dune Road well no. 1
provides water for the entire barrier beach.
A second well was proposed to be constructed that would greatly
increase water reliability in the immediate area and improve pressure in the
mains, which in the past has reached critically low levels. The need for an additional water source in
this area is crucial and of the available solutions, the construction of a
new well is much more cost effective when compared to other alternatives.
Figure
1:
SCWA Service Areas and Location of Dune
Road Wellfield
Study Site
The Dune
Road North well field is located on the barrier
beach, in what is known as the Village
of Westhampton Dunes. Ground
surface elevation at the well site is approximately +9' (2.74 m) msl. The water table elevation is approximately
+5' (1.52 m) msl. Typical of
deep wells on barrier beaches, the Magothy potentiometric surface is above grade,
at an elevation of approximately +11’ (3.35 m) msl. According
to geophysical and lithologic logs of the subsurface taken during
construction of the Dune Road North
test boring in 2004, the Upper Glacial aquifer extends from the land surface
to a depth of approximately 95 feet (28.96 m) below grade. The Monmouth Greensand unit is present
beneath the upper Glacial aquifer and continues to a depth of approximately
230 feet (70.12 m) below grade. The
remainder of the subsurface geology is characteristic of the Magothy aquifer
consisting of fine to coarse sand strata, various clay and silty layers, and
gravel to a depth of approximately 715' (218 m) below grade, at which point
the test boring was terminated. For the
purposes of this study, which is concerned with only the first 5-10’ (2 m) of
subsurface, the sequence was represented as a medium gray sand.
B.
Site Characteristics
The Suffolk
County Water Authority (SCWA) proposed to construct a new production well
(no. 2) at its Dune Road North
well field, located on the north
side of Dune Road, approximately 110.38' (33.65 m) east of Dune Lane, Village of Westhampton Dunes.
The new well was designed be constructed to a depth of approximately 430
feet (131.1 m) below land surface with 375 feet of casing, 40 feet of
14" well screen and a capacity of 700 gpm. The well will be part of the
Moriches-Westhampton Low water supply system which serves the area south of
Sunrise Highway extending from Mastic to East Quogue. Of the 32 wells that currently make up the
system, 18 wells have been in service for more than 20 years (of which 13
wells have been in service for over 25 years). The SCWA realized that the complications
associated with an aging infrastructure may eventually comprise service to
the Moriches-Westhampton Low water supply system, specifically water pressure
in the mains. Construction of well no. 2 is part of the effort to revitalize
and stabilize the water supply system in an area currently experiencing rapid
growth and development.
During the
initial well construction phase that took place in September 2004, an anomaly
was encountered at approximately 3 feet (1 m) below grade. The remnant of a
concrete foundation was discovered around the wellfield construction site.
Because of the high costs of excavating the entire site, the extent of the
foundation was not initially determined. A steel pick was first used to
estimate the perimeter of the underground foundation. The endpoints were flagged
and eventually plotted onto Arcview GIS, in order the get a spatial
representation of the foundation, GPR was considered to be an ideal option to
further delineate the foundation, because of the low costs and accuracy in
determining anomalies at shallow depths.
Basics of GPR
Ground-penetrating radar is a
geophysical investigation tool which involves the subsurface emission of
electromagnetic radiation in the radio spectrum. The system generally consists of a pair of
antennae which operate on a central frequency (Fig. 2), in this case moved in
tandem in a cart along the area to be surveyed. The transmitting antenna puts forth a short
pulse of radiation, which travels outward in a hemispheric path into the ground. A portion of this radiation is reflected
off of subsurface areas (whether individual objects, layers, or general
regions) of differing electromagnetic characteristics, and can return upward
to the receiving antenna. Computer
software is able to record this information and display it as a 2-dimensional
figure called a radargram (Fig. 3): depth is calculated as half the time it
takes for the wave to travel to the reflective area and back to the surface
(provided the subsurface velocity can be constrained: the present study was
found to have a velocity of 15-16 cm/ns [5 ft/ns], typical for dune sand)
while the horizontal distance covered in the course of the survey is recorded
by moving the antennae a prescribed distance between measurements or, in the
present study, via odometer (Knight,
2001; Fig. 2).
Fig. 2: GPR Basics: A short
electro-magnetic wavelet is emitted by the transmitter. Portions of this wavelet travel directly
through the air and along the surface (air and ground waves). The subsurface portion may be reflected and
recorded by the receiving antenna, and the resulting signal is displayed
graphically on a computer. After Hubbard et al.,
2002.
Survey Methodology
A series of short radar surveys were conducted on October 13, 2004. Two preliminary GPR lines were gathered in
the vicinity of the foundations, the first towards the east, the second
diagonally towards the SE, using 500 MHz (fairly high frequency and
resolution; shallow penetration) antennae to discern whether the foundations
were visible on GPR. In addition, a
south-running line was gathered in an area immediately to the west, where no
remnants of the foundation were visible above ground. The results of these tests were good,
leading to a second pair of test lines which repeated the first two with 800
MHz antennae (higher frequency and better resolution, but less
penetration). However, these lines
appeared noisy and difficult to interpret; thus it was decided to carry out
the main survey in the vicinity of the foundation using only the 500 MHz
antennae.
The main grid around the area of the foundation was
constructed to make the maximum amount of use out of the limited area of the
potential well field site. It
consisted of a 10 m by 15 m (32.8’ x 49.2’) rectangular grid oriented
parallel to Dune Road. A total of 20 east-trending lines were run,
evenly spaced every half meter (the final line was deemed too close to a
surface hazard to run) (Fig. 4B).
Samples were taken automatically every 5 cm (2”), as registered by the
odometer attached to the cart. To avoid
aliasing (incorrectly interpreting the signal frequency due to inadequate
sampling), the sampling rate was set to 6408 MHz, and the signal was repeated
(stacked) 16 times at each location to improve the strength of the signal
with respect to random noise.
Fig. 3: 500 MHz radargram of the
foundation at the Dune Road well field
with vertical exaggeration removed.
The foundation measures about 7 m (23’) long and is about 2 m (6.5’)
deep in this radargram. The vertical scale
is radar wave travel time (which has been converted to depth), and the
horizontal is west-east (left-right) distance.
Following the
completion of the first grid, it was decided to explore the adjacent area to
the west, which appeared free of surface obstacles other than the clean fill
which had been laid down for the drill rigs.
This portion of the site could potentially be employed as an
alternative location for the wells.
Due to the importance of locating subsurface hazards in this area, the
higher-resolution 800 MHz system was utilized. A 3 m by 10 m grid (9.84’ x 32.8’) of
north-trending lines was established, once again using half-meter spacing
between each of seven lines (Fig. 4B).
The first line was run along the western edge of the previous grid,
and the survey continued westward. The
sampling rate was set to 10282 MHz, samples were collected every 4.1 cm
(1.6”), and samples were stacked 32 times to further limit the amount of
noise influencing the sensitive antennae.
Finally, a single 18 m (59’)-long 800 MHz line was run along the
western edge of the well field property for purely exploratory purposes using
the same settings as the 800 MHz grid.
Subsequently
the GPR data were processed at Stony
Brook University
using RAMAC GroundVision and Sandmeier Software’s REFLEX processing
software. Steps were taken to remove
consistent background and system noise, limit the frequencies recorded by the
receiver to those similar to the transmitter frequency, and restore the loss
of energy due to natural attenuation of signal with depth (gaining).
Results
The 500 MHz survey in the area near the concrete blocks
clearly delineated a subsurface object within two meters of the surface (Fig.
3, 4A). The foundation was easily
visible even without the application of a gain, and the approximate location
of the boundaries of this object matched well with the estimated boundaries
discovered during the initial drilling process, as demarcated by flags. It appeared to extend from about the 4.5 m
line (northern boundary) to the 9 m line, and from about 8.5 m (western
boundary) to the end of each line (Fig. 3, 4A). A three-dimensional cube was constructed by
plotting several radargrams in a grid-like format, from which the extent of
the foundation could be recognized.
This illustrates the powerful mapping capabilities of GPR when used in
concert with other geological techniques.
The 800 MHz survey had slight difficulties penetrating
through the clean fill which covered the majority of the area, but was still able
to locate buried objects outside of the study area which are thought to
represent pipes beneath Dune Road
and other features not of interest to the current project. However, no significant objects within the
system’s resolution (about 5 cm [2”] ) were observed, allowing the drilling
to be undertaken in this alternate area (Fig. 4A).
 
Fig. 4: Results and details of GPR
survey near Westhampton. A (left):
Aerial photo of survey area, near the intersection of Dune Road and Dune Lane,
Westhampton Dunes. The SCWA property
line (study area) is the orange boundary line. Superimposed is the outline of the
foundation (as imaged by GPR) which continues into the adjacent parcel (as
marked by jagged lines to the right).
Figure links to larger version. B (right): Close-up of highlighted area in
A showing area of GPR grids, including orientation of lines (red). Also shown is the location of a flagged
section of the foundation for ground-truthing.
Conclusion
The use of GPR, in conjunction with field observations,
provided a quick, inexpensive, and fairly accurate means of delineating a
subsurface hazard to hydrologic well drilling. It also revealed the lack of such hazards
beneath an alternate adjacent site, allowing drilling to continue. Such surveys can also be undertaken as an
exploratory tool prior to the initiation of drilling at prospective sites,
possibly saving thousands of dollars and significant amounts of lost
man-hours and delay time in creating vital access to groundwater
resources.
Acknowledgments
KTG
would like to thank DMD for technical assistance in planning the survey and a
review of the manuscript.
References
Hubbard, S., Grote, K., and Rubin, Y., 2002, Mapping the
volumetric soil water content of a California vineyard using high frequency
GPR ground wave data, The Leading Edge, vol. 21 no. 6. p. 552-559.
Knight, R., 2001, Ground
penetrating radar for environmental applications, Annual Review of Earth and
Planetary Sciences, vol. 29. p.
229-255.
Long Island
Source Water Assessment Program, NYSDOH, Albany,
NY 2003
McClymonds, N.E., and Franke,
O.L., 1972. Water Transmitting Properties of Aquifers on Long Island, New
York, USGS Professional Paper 627-E, Washington, DC
New York State Department of
Environmental Conservation, Undated, Development of a Procedure for
Determining the Zone of Capture of Wells Penetrating the upper Glacial
Aquifer on Long Island, New York.
USEPA, 1994, Handbook: Ground
Water and Wellhead Protection, Office of Research and Development, Office of
Water, Washington D.C.
|
