See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/257785932

On the application of GPR for locating underground utilities in urban areas

Article in Arabian Journal of Geosciences · September 2012

DOI: 10.1007/s12517-012-0588-4

CITATIONS READS

14 2,949

2 authors:

M. Rashed Mansour A. Al-Garni

Suez Canal University King Abdulaziz University
60 PUBLICATIONS 639 CITATIONS 64 PUBLICATIONS 504 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by M. Rashed on 03 March 2014.

The user has requested enhancement of the downloaded file.

On the application of GPR for locating
underground utilities in urban areas

Mohamed A. Rashed & Mansour A. Al-

Garni

Arabian Journal of Geosciences

ISSN 1866-7511
Volume 6
Number 9

Arab J Geosci (2013) 6:3505-3511

DOI 10.1007/s12517-012-0588-4

1 23
Your article is protected by copyright and
all rights are held exclusively by Saudi
Society for Geosciences. This e-offprint is
for personal use only and shall not be self-
archived in electronic repositories. If you wish
to self-archive your article, please use the
accepted manuscript version for posting on
your own website. You may further deposit
the accepted manuscript version in any
repository, provided it is only made publicly
available 12 months after official publication
or later and provided acknowledgement is
given to the original source of publication
and a link is inserted to the published article
on Springer's website. The link must be
accompanied by the following text: "The final
publication is available at link.springer.com”.

1 23
 Author's personal copy
Arab J Geosci (2013) 6:3505–3511
DOI 10.1007/s12517-012-0588-4

ORIGINAL PAPER

On the application of GPR for locating underground

utilities in urban areas
Mohamed A. Rashed & Mansour A. Al-Garni

Received: 10 January 2012 / Accepted: 25 April 2012 / Published online: 10 May 2012
# Saudi Society for Geosciences 2012

Abstract With the rapid growth of complex network of pipes. The nature, shape, size, and depth of these utilities
different types of underground utility under large cities, the vary significantly from one to another, ranging from more
need of a noninvasive technique capable of swiftly and pre- than 2-m-diameter concrete or asbestos sewage pipes to less
cisely detecting these utilities in such a noisy urban environ- than 2-cm-diameter fiber optic communication cables. Pre-
ment increases. Ground-penetrating radar (GPR) is considered cise location of underground utilities such as cables and
one of the most promising techniques in this field. This study pipes is of great importance because it can save money
presents the experience of GPR data acquisition, processing, and effort and reduces the risk of damaging the underground
and interpretation in three cities located along the coast of the utilities and hence causing environmental pollution and/or
Red Sea. These cities are Jeddah in Saudi Arabia and Sharm traffic inconvenience. However, it is a challenge to come up
El-Sheikh and Qusier in Egypt. Data acquisition parameters with a single geophysical technique capable of precisely
varied in the three cities based on site conditions, target locating such a wide range of underground utilities. Another
characteristics, and equipment availability. The processing challenge arises from the fact that a geophysical survey for
flows were kept simple to avoid introducing artifacts to the locating underground utilities is usually conducted in urban
collected data. The results show that despite the difference in areas infested with all kinds of noise. In the past two
site conditions and survey parameters among the three cities, decades, ground-penetrating radar (GPR) has emerged as a
with the exception of fiber optic cable, GPR technique is successful noninvasive technique for locating various sub-
capable of detecting different kinds of underground utilities surface utilities, including metallic pipes (Rashed and
and precisely determine the extension, diameter, and depth of Morsy 2012), nonmetallic pipes (Tong 1993), power cables
burial of these utilities. (Yelf 2007), and even water leaks from underground pipes
(Hunaidi et al. 2000). There are many advantages that made
Keywords GPR . Utilities . Urban . Pipe . Cable GPR a superior utility locator nowadays. GPR is a nonin-
vasive, nondestructive technique. A GPR survey can be
conducted in a short period of time without causing much
disturbance or inconvenience to the surrounding usual
Introduction activities. In particular cases, GPR can provide real-time
high-resolution images for the shallow subsurface.
Nowadays, our cities float over a huge network of dense, The present study introduces the experience of conduct-
extensive, and interfering utilities of different cables and ing GPR surveys for locating underground utilities in three
cities along the coast of the Red Sea which are: Sharm
M. A. Rashed (*) : M. A. Al-Garni
El-Sheikh and Qusier in Egypt and Jeddah in Saudi Arabia
Department of Geophysics, Faculty of Earth Sciences,
King Abdulaziz University, (Fig. 1). Sharm El-Sheikh City is a famous resort that
Jeddah, Saudi Arabia attracts millions of local and international tourists every
e-mail: [email protected] year. The city is located on the northern tip of the Red Sea
where the Gulf of Suez and Gulf of Aqaba meet open sea.
M. A. Rashed
Geology Department, Suez Canal University, Qusier City, on the other hand, is a small town on the
Ismailia, Egypt western coast of the Red Sea near the southern borders of
 Author's personal copy
3506 Arab J Geosci (2013) 6:3505–3511

Egypt (Fig. 1). The city has small population and is going The main purpose of the GPR survey in Qusier City was
under a lot of development in recent years. Jeddah City, to investigate the depth variations of water table in the study
however, is a megacity located on the Saudi coast of the Red area. However, a secondary purpose of the survey was to
Sea (Fig. 1). The city is considered as the second largest city detect a buried network of steel pipes in the area. The data
in Saudi Arabia after the capital Riyadh and is inhabitant by were collected using SIR-3000 GSSI system and 100-MHz
more than 5 million people. The only common factor among antenna similar to those used in the Sharm El-Sheikh survey.
the three GPR surveys conducted in the three cities is their Due to the unavailability of survey wheel, the data were
capabilities to locate underground utilities buried at shallow collected in time mode and marks were entered manually
depths. every 5 meters during the data collection. Because there are
no reflected signals could be seen beyond 250 ns, the range
was set to 300 ns with transmit rate of 100 KHz. Sampling
Data acquisition rate of 1024 samples per trace, 24 scans per second, and 36
scans per meter were used in collecting the data in this
In Sharm El-Sheikh City, the GPR survey was conducted to survey. Gain function was set automatically by restarting
investigate the source of hydrocarbon contamination disaster the antenna at each survey site. Low-cut filer of 25 MHz and
that affects the water of Sharm El-Maya Bay (Morsy 2011). high-cut filter of 300 MHz were applied while collecting the
Previous magnetic and gravity investigations on the area GPR data (Table 1).
showed the presence of underground storage facilities and The GPR survey, conducted in Jeddah City, aims to
buried pipes (Morsy et al. 2012). The GPR survey was con- investigate the capability of GPR to detect and estimate
ducted using SIR-3000 GSSI system equipped with a single the depth and dimensions of various underground utilities
100-MHz transceiver antenna. In an early stage of the GPR including reinforced concrete pipes, electricity cables, fiber
survey, several field tests were conducted in different parts of optic cables, and PVC tubes. The survey was conducted
the study area to estimate the optimum field acquisition adjacent to the Jeddah-Al-Lith main road, on both sides of
parameters that would be appropriate for it. Data were collect- a trench dug perpendicular to the road. The trench is 7 m
ed in time mode and marks were entered manually every 5 m long and 1.5 m wide and exposes all utilities buried adjacent
during the data collection. The range was set to 400 ns with to the road. The data were collected using SIR-3000 GSSI
transmit rate of 100 KHz. Sampling rate of 2048 samples per system and a 400-MHz antenna with transmit rate of
trace, 16 scans per second, and 40 scans per meter were found 25 KHz. GSSI cart was used to collect the data in distance
adequate for this particular survey. An automatic gain function mode with time range of 50 ns. Sampling rate of 512
was selected automatically for each survey site and no filters samples per trace, 50 scans per meter, and 100 scans per
were used during data collection in order to allow for the second were used to conduct the survey. Low-cut filter of
widest range of frequencies to be recorded. This arrangement 100 MHz and high-cut filter of 800 MHz were used to
gave a wide range of choices during data processing. Table 1 collect the data (Table 1).
summarizes the acquisition parameters used for collecting the Variation on the parameters used to collect the GPR data
GPR data in the study area. in the three cities are due to the availability of equipment,
site conditions, and the priori information about the targeted
objects. For example, the data were collected in time mode

Table 1 GPR data acquisition parameters in the three cities

Sharm El-Sheikh Qusier Jeddah

GPR system SIR-3000 SIR-3000 SIR-3000

Antenna frequency 100 MHz 100 MHz 400 MHz
Time range 400 ns 300 ns 50 ns
Transmit rate 100 kHz 100 kHz 25 kHz
Collecting mode Time Time Distance
Samples/trace 2048 1024 512
Scans/meter 40 36 50
Scans/second 16 24 100
Filters Off 25–300 MHz 100–800 MHz
Fig. 1 Location map of the three cities where GPR surveys were Gain Automatic Automatic Automatic
conducted
 Author's personal copy
Arab J Geosci (2013) 6:3505–3511 3507

because there was no cart or survey wheel was available in reinforced concrete storage tank. These two profiles
Sharm El-Sheikh or Qusier surveys. A 400-MHz antenna showed straight dipping reflections resulting from the
was used in Jeddah City survey because the trench shows walk-away effect of the vertical walls of the tank. The
that the depth to the utilities ranges from 0.5 to 2 meters velocity of the upper layer could be calculated from the
below the ground surface. slope of these dipping reflections using the following
equation:
dt
Data processing V ¼ ð1Þ
dx
The main purpose of GPR data processing is to transform Where V is the layer velocity and dt is the time difference
the raw GPR data collected in field into realistic GPR and dx is the distance difference along the slope of the
sections, displaying high-resolution images of the subsur- reflector. The calculated velocity value (0.11 m/ns) was then
face. The data collected in the three cities were processed used to calculate the relative dielectric constant using the
using the RADAN 6.6 commercial software provided by following equation:
Geophysical Survey Systems, Inc. (GSSI) (2001–2011). The  c 2
processing routines applied to the collected GPR datasets Er ¼ ð2Þ
v
were kept simple in order to avoid introducing artifacts into
the processed data. Where Er is the relative dielectric constant, c is the speed
In Sharm El-Sheikh City data, the processing initiated of light of 30 cm/ns, and v is the velocity of electromagnetic
with editing the file header followed by horizontal distance energy passing through the material.
normalization in order to compensate for inconsistencies in The estimated dielectric constant (7.5) was then edited
the antenna towing speed. The horizontal scale was then into the file header to produce a maximum depth of 15 m for
adjusted using stacking adjacent scans followed by routine all GPR sections in the area. However, no significant fea-
wide bandpass filtering. The surface position adjustment tures could be seen beyond 10 m depth. So, only those 10 m
was interactively edited in the header to align the ground are displayed.
surface with the first reflection on the GPR record using The GPR data collected in Qusier City were pro-
antenna lifting field tests conducted during the field survey. cessed with a processing routine very similar to the
A linear five-point gain function was applied to the data to one that was applied to Sharm El-Sheikh City Data.
compensate for signal amplitude attenuation with depth. The The only difference was that velocity information was
data were then filtered using finite impulse response (FIR) obtained using wide-angle refraction reflection (WARR)
filter and treated with background removal function. Decon- method. Two 100-MHz antennas were used to collect
volution was then applied to the data in order to completely GPR data using a bistatic antenna configuration in the
remove ringing and multiples. Finally, a low-pass filter was static stack mode. The transmitter antenna was kept at a
applied to the data to remove high-frequency, snow-like fixed location and the receiver antenna is towed away at
noise. Table 2 shows a summary of the processing flow increasing offsets. The collected WARR data provided
applied to the collected GPR data. precise velocity information of the subsurface in Qusier
Two of the GPR profiles collected in Sharm El- City survey area.
Sheikh City were acquired over a vertical cylindrical The GPR data collected in Jeddah City were treated
differently because of the presence of the trench and
because of the different field parameters that were used
to collect the GPR data in this area. A GPR profile was
Table 2 Sharm
El-Sheikh GPR data Header editing collected close and parallel to the trench. The depth to
processing sequence Distance normalization the exposed pipes was measured and used to calculate
Horizontal scale adjustment the dielectric constant that was then edited to the file
Vertical scale adjustment header. Distance normalization and horizontal and verti-
Bandpass filtering cal scales adjustment were then applied to the collected
Surface position adjustment data. Surface position adjustment was then applied using
Automatic gain control antenna lifting field tests followed by 50–600-MHz
FIR filtering bandpass filtering. The data were then subjected to
Background removal
background removal and deconvolution functions to
Deconvolution
remove ringing and multiples. Finally, the data were
Low-pass filtering
filtered using high-cut frequency filter to remove high-
frequency noise.
 Author's personal copy
3508 Arab J Geosci (2013) 6:3505–3511

Results and discussion (Morsy et al. 2012). The GPR profile, shown in Fig. 2a,
shows a group of interesting features.
The interpretation of the GPR data is usually quite difficult, The first feature is the reinforced concrete tank near the
especially, in the absence of information about the subsur- northwestern edge of the section between distance marks 1
face either from drilling data or other geophysical data. and 7 m. The top of this tank lies at slightly more than 1 m
However, the GPR data collected in the three cities during below the ground surface. The strong reflections caused by
this study are either collected near a trench or close to an the top of the tank are clear and also the dipping reflections,
exposure, which aided the interpretation process and helped on both sides of the tank, due to the walk-away effect from
verifying the final results. the vertical walls of the tank (Fig. 2a).
The GPR survey in Sharm El-Sheikh City was conducted The section also shows two metallic pipes near the south-
to image subsurface artificial structures that may contributed eastern end of the section at distances of 14.5 and 17.5 m.
to the disastrous contamination in the area. Previous mag- These two pipes could be tracked on several GPR profiles
netic and gravity surveys delineated several anomalous collected in the area (Fig. 2b). The diameter of the buried
zones and the GPR data were collected over these zones pipes can be determined using the generalized Hough

Fig. 2 A GPR profile across

underground utilities in Sharm
El-Sheikh (a), its interpretation
(b), and a photograph of the
excavated tank and pipes (c)
 Author's personal copy
Arab J Geosci (2013) 6:3505–3511 3509

transform (Windsor et al. 2005). The principle of this depth and to estimate the depth to water table in the study area.
estimation method is based on the fact that the time of flight, However, another target of the survey was to map a network
as the radar antenna is scanned along a line perpendicular to of underground pipes buried in study area.
the pipe, shows an arc. The shape and position of this arc Figure 3 is a 25-m long part of one of the GPR profile
depend on the position of the center of the pipe along the collected in Qusier City. The GPR section does not show
scan, the depth of the pipe center, the pipe radius, and the any faults or fractures but it clearly shows the water table as
velocity of the surrounding medium. This relationship is a strong, continuous, horizontal reflection that extends from
given by: north to south through the entire section at slightly more
than 3 m below the ground surface (Fig. 3b). The section
1
Z0 ¼ ½ðR0 þ V0 t1 =2Þ2  ðy1  y0 Þ2  2 ð3Þ also shows a resonating hyperbola at distance of 17 m. The
hyperbola may be caused by a metallic pipe located at depth
Where Z0 is the depth to the center of the pipe, R0 is the of 2.2 m below the ground surface. Moreover, the saturated
radius of the pipe, V0 is the velocity of the medium, y0 zone below water table is clearly distinguished from the
is the position of the top of the hyperbola on the unsaturated zone above the water table by the difference in
distance axis, y1 is the position on one wing of the GPR amplitude, continuity, and character.
hyperbola, and t1 is the flight time corresponding to The GPR survey in Jeddah City was conducted next to
position y1 (Windsor et al. 2005). the southern Jeddah-Al-Lith road where many utilities are
The generalized Hough transform was applied to the two buried at depths ranging from 0.5 to 2 m below the ground
pipes and R0 was found to be 12.1 and 12.8 cm. This surface. GPR lines were carried out perpendicular to the
indicates that the diameter of the two pipes is approximately road and parallel to a 7-m long trench dug in the area.
25 cm. Excavations conducted in the area revealed an up- Figure 4 shows one of the GPR profiles collected in the
right cylindrical reinforced concrete tank in the location area (Fig. 4a), its interpretation (Fig. 4b), and a photograph
pinpointed by the GPR survey. The excavation also revealed of the detected utilities exposed in the trench.
two steel pipes of diameter 25 cm sticking out of the con- This GPR section shows clearly a single strong hyperbola
crete tank and go to the underground (Fig. 2c). at distance of 4.5 m with its top at depth of slightly more
The prime target of the GPR survey conducted in Qusier than 1 m below the ground surface (Fig. 4a). This hyperbola
City was to confirm the presence of subsurface faults and is caused by the concrete water supply pipe shown in
fractures zones depicted from the analysis satellite images Fig. 4c. The smaller and yet stronger hyperbola, located at

Fig. 3 A GPR profile across an

buried pipe in Qusier (a) and its
interpretation (b)
 Author's personal copy
3510 Arab J Geosci (2013) 6:3505–3511

Fig. 4 A GPR profile across

underground utilities in Jeddah
(a), its interpretation (b), and a
photograph of the trench
exposing utilities (c)

distance of 3.2 m and at depth of 0.7 m (Fig. 4a), is believed The GPR section does not only show the utilities buried
to be caused by the electric power cable shown in Fig. 4c. in the underground along the profile, but also shows a zone
The GPR section shows yet another smaller and weaker of disturbed soil in the area where the utilities are buried.
hyperbola at distance of 9.5 m that is believed to be caused The zone is marked with the dashed black line (Fig. 4b) and
by an unidentified object. is well distinguished from the undisturbed zone. The dis-
The diameter of the water supply pipe was calculated turbed zone is characterized by its relatively lower ampli-
using the same principle of generalized Hough transform tude than its surrounding zone, its rough discontinuous
mentioned in earlier section (Windsor et al. 2005). The reflections, and by the strong reflector descending from near
calculated diameter was exactly 1.2 m which coincides well the 10-m mark on the eastern side of the profile. The outline
with the real diameter of the pipe measured in the trench. of the disturbed soil zone interpreted on the GPR section
The diameter of the power cable could not be estimated due coincides well with the soil conditions seen along the trench
to the resonating hyperbola which prevents precise estima- in the survey site.
tion of its depth in addition to the small size of the The GPR survey in Jeddah failed to see neither the fiber
hyperbola. optic communication cable nor its metallic warning tape.
 Author's personal copy
Arab J Geosci (2013) 6:3505–3511 3511

The detectability of a target depends on the its depth, size, In Jeddah City, the GPR survey was conducted next a
orientation, composition, and surrounding medium proper- dug trench to test the detectability of different underground
ties, in addition to the frequency of the GPR antenna used in utilities on GPR images. The survey was successful in
the survey. However, the failure to detect the fiber optic detecting concrete water pipes, PVC pipe housing commu-
cable in Jeddah may be attributed to the composition and the nication cable, and power cable but it failed to detect a fiber
diameter of the fiber optic cable. The composition of the optic cable buried directly in the ground. The calculated
cable that creates very minor contrast with the surrounding diameters for the detected utilities, using the generalized
sand and the diameter of the cable (about 1 cm) make the Hough transform, were coincident with the real diameters
cable completely invisible on the GPR section. However, as measured for the exposed utilities in the trench.
the cable can be tracked by the distinguished disturbed soil Regardless of the difference in the surrounding environ-
zone around it. ment, the level of cultural noise, or the nature of the targeted
Moreover, the GPR survey could be used successfully to utility, GPR technique works properly in most of the cases.
track and to detect another fiber optic cable inside a 10-cm It is worth mentioning, however, that the GPR method is
diameter PVC tube buried at depth of 0.5 m in the study highly site dependent and that the selected field parameters
area. The cable itself was not detected but the PVC pipe can have a great influence on the quality of the collected GPR
be easily seen and tracked from one GPR profile to the next. data. Accordingly, field tests before starting data acquisition
are essential in order to setup the appropriate acquisition
parameters before collecting the GPR data. The rapid devel-
Summary and conclusions opment of GPR hardware and software capabilities and the
continuing reduction of their prices are expected to enhance
Ground-penetrating radar is evolving as a swift, efficient, the GPR imaging technique significantly in the next few
noninvasive geophysical investigation tool for locating years. GPR investigation technique is a strong candidate to
underground utilities in urban areas. The reasons for its become the prime utilities locator in the near future.
increase use in this field are its high precision, cost effec-
tiveness, and that it can provide real-time, high-resolution
References
images of the shallow subsurface. This study represents the
experience of conducting GPR surveys in three different
cities on the coast of the Red Sea, namely Sharm El- Hunaidi O, Chu W, Wang A, Guan W (2000) Detecting leaks in plastic
pipes. J Am Water Work Assoc 92:82–94
Sheikh, Qusier, and Jeddah.
Morsy M (2011) Geo-environmental studies on Sharm El-Maya Area,
In Sharm El-sheikh City, the GPR survey succeeded in Sharm El-Sheikh, Southern Sinai, Egypt. M.Sc. Thesis, Suez
detecting three upright cylindrical reinforced concrete tanks Canal University: 114p
and a sophisticated network of steel pipes connecting them. Morsy M, Rashed M, El-Masry N, Soliman F (2012) Application of
potential field methods to inspect the source of hydrocarbon
These pipes are buried at depths ranging between 1 and 2 m
contamination in Sharm El-Maya Bay area, Sharm El-Sheikh,
below the ground surface. The steel pipes are quite distin- Egypt. Journal of Earth Sciences, King Abdulaziz University (in
guished by its resonance. The diameters of these pipes were press)
calculated using generalized Hough transform. Excavations RADAN (Version 6.6) “Software for Processing GPR Data” (2001–
2011) Salem, 03079 New Hampshire, USA: Geophysical Survey
conducted later in Sharm El-Sheikh City exposed steel pipes Systems, Inc. (GSSI)
sticking out of the tanks and showed an excellent agreement Rashed M, Morsy M (2012) Combined magnetic, gravity, and GPR
with the calculated diameter (Fig. 2). surveys revealing the source of hydrocarbon contamination in
The survey conducted in Qusier City aims to tracking Sharm El-Sheikh area, south Sinai, Egypt. Near Surface Geophys-
ics (In press)
water table and confirming the presence of fractures and
Tong L (1993) Application of ground penetrating radar to locate
faults in the study area. The GPR survey, however, suc- underground pipes. Terr Atmos Ocean Sci 4:171–178
ceeded in detecting a metallic pipe buried at depth of about Windsor C, Capineri L, Falorni P (2005) The estimation of buried pipe
2 m below the ground surface. In addition, water table has diameters by generalized Hough Transform of radar data. Prog-
ress in Electromagnetics Research Symposium, Hangzhou, China:
been clearly identified where difference in GPR amplitude
345–349
and characteristics between the zones below and above Yelf R (2007) Application of ground penetrating radar to civil and
water table are well recognized. geotechnical engineering. Electromagn Phenom 7:102–117

View publication stats