Revolutionising non-invasive intelligence gathering

Revolutionising non-invasive intelligence gathering


Ground-penetrating radar, or GPR, is the
most rapidly developing non-destructive,
non-disruptive technology for accurately
measuring and geoloacting underground
infrastructure. By Geoff Zeiss

In the United States, an underground utility line is hit on
average every 60 seconds and damage to the national
economy owing to this is estimated to be in billions of
dollars. In most municipalities in North America, underground
utility lines, for years, have been put in the
ground not according to plan but wherever it has been easiest
and cheapest to build them. As a result, 2D as-builts of
underground infrastructure are notoriously unreliable. The
result is that in most municipalities the location of underground
utilities is poorly known. This and other application
areas have created demand for ways of detecting and identifying
underground structures, natural and man-made.

A number of technologies have been developed to help
measure, characterise and geolocate underground structures

including ground penetrating radar (GPR), seismic refraction;
magnetometry; electromagnetic terrain conductivity (EM);
very low frequency (VLF) profiling — electrical resistivity imaging;
borehole geophysical and video logging; crosshole seismic
testing; seismic tomography; and microgravity surveys.

Among these, ground-penetrating radar (GPR) is probably
the most rapidly developing field. The development of
GPR has occurred in geophysical science and technology
domain, and a wide range of scientific and engineering applications
are explored. From not even being mentioned in
geophysical texts 15 years ago, GPR has become the topic
of hundreds of research papers and special issues of journals
as a wide range of applications has made the technology a
valuable tool in the geophysical consulting and geotechnical
engineering industries.

While there is no established data on the size of the current
GPR market, the annual worldwide market for commercial
GPR solutions and software was less than $50 million
in 2005 but was growing between 10% and 15% annually,
found a Spar Point report in 2005. Going by that, the GPR
market would be anywhere between $100 and $200 million
currently. Furthermore, the study estimated that the market
for GPR associated services was 5 to 10 times the size of the
market for hardware and software solutions, which would
put the total market size at roughly a $1 billion annually.

How it works
GPR uses radio waves, which are radiated from a transmitter
that pulses a signal into the ground. In the most commonly
used type of GPR, the waves are reflected by underground
structures back to the surface where they are measured by a
receiver unit, amplified and digitised by a computer to record
the measurements. Reflection measurements can be a single
source and receiver combination or more sophisticated multi-
transmit/receive observation. Recent developments have
lead to a growing use of the transillumination mode whereby
the transmitter and receiver are at different locations.

Ground penetrating radar requires a geological environment
where radio waves can propagate a sufficient distance
through earth materials to be useful. GPR frequencies are predominantly
in the 1 to 10,000 MHz range. In general, electrical
conductivity determines the depth of exploration. For instance,
in case of clay with a lot of water content, radio signals will
only penetrate a few feet, whereas in highly resistive granite
formations signals can be transmitted through tens and even
hundreds of metres of rock and still be detected.

There are significant differences between the technologies
used for above ground laser scanning and subsurface
scanning with GPR and the interpretation of results. Laser
typically involves very short wavelengths, which
makes possible high resolution of the order of millimetres.
The media in which it operates is air, which is relatively
homogeneous and non-absorbing at the wavelengths used.
This means laser scanning can record accurately at distances
of up to hundreds of metres from the target, but requires
optical line-of-sight visibility. Measurements can be easily
ground-truthed by tape measure or total station. The result is
a point cloud that can be visualised and measured and compared
directly with the scanned object. In some cases feature
extraction techniques can be used to extract individual objects
to form a 3D model of the scanned object.

In contrast, the wavelengths used in GPR are longer,
sub-metre wavelengths to enable them to pass through soil
and rock which limits the achievable spatial resolution. The
medium is heterogeneous — it may be comprised of different
soil and rock types together with manmade materials. In the
United States, penetration depth for GPR ranges from about a
metre in concrete or asphalt to seven or eight meters in clay or
silt to 100 metres in limestone or granite. The sensor must be
relatively close to the object depending on the geology of the
ground and the strength of the transmitter.

GPR can provide detailed information about underground infrastructure more cost-effectively than
potholing and without disrupting traffic

Transmitters in Europe are permitted to be stronger than
in the US where the FCC severely limits transmitter power.
As with any type of radar, GPR signals must be interpreted.
It is difficult to ground truth what has been detected without
potholing. With GPR it is often necessary to use all other
available information, including existing engineering drawings,
above ground visible structures such as manholes and
storm drains, and historical records.

GPR emerged from polar ice radio
echo sounding in the late 1960s. The method
was initially used for engineering, soil
applications and in mining. Nowadays, the
technology is used for a broad range of applications
such as concrete inspection, underground
utility detecting, asphalt pavement
inspection, bridge deck and concrete
inspection, railroad ballast inspection, geological
fault detection and investigation,
tunnel scanning, archaeology, road inspection,
rebar detection and mapping, landmine
detection, snow scanning, borehole
inspection, pavement thickness and road
condition assessment to name just a few.

There are many firms providing GPR
services. In the construction sector for
example, there are many geophysical services
companies specialising in 3D subsurface
imaging and mapping, where accurate
subsurface information is needed before
construction excavation. This technology is capable of detecting
and identifying non-metallic as well as metallic utilities.
The deliverable is a GIS compatible map that provides the project
management team information about underground utilities
for pre-design that can significantly reduce the risk or project
delays and cut costs due to unexpected underground structures.

Geolocating underground utilities and
highway design

In 2010, the National Cooperative Highway Research Program
(NCHRP) of the United States published a report Utility
Location and Highway Design: A Synthesis of Highway
Practice that summarised research about how highway designers
incorporate the location of underground and above
ground utilities into their designs. The literature review identified
the issues influencing the decision to keep utilities in
place or to relocate them.

Historically, transportation designers ignored utilities
during design. If the utilities conflicted with the design,
they are relocated. As a result utilities are routinely relocated,
often incurring huge unwarranted expenses. An alternative
approach is to design the highway in a way that
avoids the utilities so that the existing utilities remain in
place. The challenge is that accurate data about the location
of underground utilities is generally lacking. Between
the extremes of relocating all the utilities and designing the
highway to leave utilities in place is a workable compromise
that meets the highway construction scope and mission,
while minimising impacts to utility facilities. With
this optimal solution, substantial savings in utility relocation
costs and impacts, as well as overall savings to the
project budget and schedule can be realised.

The survey results indicated that the state DOTs would
like to get utilities involved as early as possible in the construction
process. The most important reason is to determine
as early as possible which utilities potentially will be affected
and where they are located. The literature survey showed
that there is a general consensus that accurate and comprehensive
utility location data helps make better decisions and

reduces the risk of unforeseen problems with utilities emerging
during the construction phase.

Geolocating underground infrastructure
Traditionally, the way to accurately geolocate underground
facilities relies on potholing or digging a hole to expose
the infrastructure. This is not only expensive, but it is also
disruptive. Excavating in roadways requires diverting traffic
and also runs the risk of potentially hitting other utility
infrastructure. As a result there is an accelerating interest in
non-disruptive technologies.

GPR technology has been in use for nearly three decades,
but it has required a substantial amount of time to develop a sufficient
level of understanding of the method and its benefits to
be appreciated by the broad user community. Over the last few
years the range of applications has expanded greatly and an important
application area is mapping underground infrastructure.

Today, there is a better understanding of the geological
context in which GPR is effective. There is also a much
better understanding of the physical properties of soils and
manmade materials which determine the penetration and reflection
of radio waves. Radar systems with higher power and
digital data recording capabilities have been developed. Computer
processing power has increased dramatically enabling
enhanced digital data processing and 3D visualisation that
were not possible until a few years ago. With the development
of affordable 3D visualisation, GPR processing is becoming
widespread, resulting in major changes in the state-of-thepractice.
In favourable geologic conditions, GPR can provide
detailed information about underground infrastructure more
cost-effectively than potholing and without disrupting traffic.

There is an expanding effort worldwide to accurately record
the geolocation of underground infrastructure. Among
the leading projects worldwide is a 10-year effort in Lombardy,
which includes Milan, to map all underground infrastructure
using ground penetrating radar.

The American Society of Civil Engineers (ASCE) has developed
a standard ASCE Standard 38-02 ‘Standard Guideline
for the Collection and Depiction of Existing Subsurface
Utility Data’ for classifying the reliability of information
about underground infrastructure:

      • Quality Level D (QL D) Review of Existing Records &




      • Quality Level C (QL C) Surveying & plotting above


      ground (surficial) features and connecting points


      • Quality Level B (QL B) Surface geophysical methods to


      map subsurface utilities


      • Quality Level A (QL A) Non-destructive excavation to expose


    & survey subsurface utilities, typically by potholing

The classification is based on level of reliability of the
location about subsurface facilities depending on the means
by which the subsurface information was obtained.

Using GPR to geolocate underground
infrastructure in Lombardy/Milan

A pilot project was carried out on the site of the Expo

Milano 2015 event in Milan. The total project area is
about 230,000 sq metre. All underground infrastructure
including electric power, water, sewers, gas, district
heating, streetlighting, and telecommunication, were
mapped both from historical records and using GPR. A
data model for underground infrastructure was developed
for the different types of underground networks
based on the Italian DigitPA and the INSPIRE-US utility
standards. Most of the data is 2D, but some 3D
data has been recorded and used to demonstrate 3D

Comparison of the historical records with the results
captured by GPR revealed significant discrepancies in
the historic record including thousands of meters of
unknown infrastructure. For known infrastructure, the
average error in geolocation was about 30%, but much
larger errors of up to 100% were also recorded. The
conclusion is that even in Europe the record of underground
infrastructure can be highly unreliable. That the
exercise identifies underground infrastructure that had
been previously unknown to the municipalities provides
some financial motivation for municipalities because
they tax utilities based on the total infrastructure the
utilities maintain within city limits. The data has been
made available on the Web via Open Geospatial Consortium
(OGC) standard protocols and formats (WMS,

An economic analysis of the data has been carried
out and the estimated return on investment is about €16
($21.8) for every euro invested in improving the reliability
of information of underground infrastructure. For
comparison, the ROI in the United States has been estimated
to range from $3 to $21 for every dollar invested.
Other benefits include improved safety for workers and
the public and fewer traffic disruptions.

Several critical factors enable a project like this to
be successful. A clear legal framework is absolutely
essential. In addition, it is necessary to ensure that all
stakeholders are involved. In the case of Lombardy this
means EU, national, regional, provincial, and municipal
governments. But absolutely critical to the success of
the project was non-destructive, non-disruptive technology
for accurately detecting underground infrastructure.
As a result of the successful pilot project, it was made
mandatory for all municipalities in Lombardy, which
includes about 1500 towns, to map their underground
infrastructure using GPR.

Geoff Zeiss,
Editor-Building & Energy, Geospatial