Application of geophysical methods in a dam project: Life cycle perspective and Taiwan experience

Journal of Applied Geophysics 158 (2018) 82–92

Contents lists available at ScienceDirect

Journal of Applied Geophysics
journal homepage: www.elsevier.com/locate/jappgeo

Application of geophysical methods in a dam project: Life cycle
perspective and Taiwan experience
Chun-Hun Lin a, Chih-Ping Lin b,⁎, Yin-Chun Hung c, Chih-Chung Chung d, Po-Lin Wu b, Hsin-Chan Liu b
a

Department of Marine Environment and Engineering, National Sun Yat-sen University, Kaohsiung, Taiwan
Department of Civil engineering, National Chiao Tung University, Hsinchu, Taiwan
Department of Urban Planning and Landscape, National Quemoy University, Kinmen, Taiwan
d
Department of Civil Engineering, National Central University, Zhongli, Taiwan
b
c

a r t i c l e

i n f o

Article history:
Received 28 March 2018
Received in revised form 27 July 2018
Accepted 27 July 2018
Available online 29 July 2018
Keywords:
Dam safety

Engineering geophysics
TDR
ERT
Surface wave
Seismic tomography

a b s t r a c t
There is a growing demand for using non-destructive geophysical techniques to internally image dam condition
and facilitate the early detection of anomalous phenomena. Near surface geophysical techniques have advanced
significantly in the last few decades, and can play a significant role in the siting, construction, and operational
safety and sustainable management of dams. Application of engineering geophysics in site characterization during feasibility investigation phase is already part of the standard of practice. This paper introduces newer applications of engineering geophysics during construction phase, dam safety assessment, and sustainable
management, including quality control of compacted soils, investigation of abnormal leakage in an earth dam,
evaluation of an aged concrete dam, geophysical health monitoring for a newly-constructed dam, and monitoring
of sediment transport for sediment management. The applications were presented with more emphases on the
needs of dam engineering and adapting appropriate geophysical methods to make assessment more effective
and consequential. The collage of these case studies is to broaden the view of how geophysical methods can be
applied to a dam project throughout a dam's life cycle and strengthen the linkage between geophysical surveillance and engineering significance at all stages.
© 2018 Elsevier B.V. All rights reserved.

1. Introduction
With growing population and higher demand for clean water, the
number of dams has increased considerably during the last century. In
addition to the number of dams, increased heights and larger reservoir
volume are common around the world. The purpose of a dam is to retain
water for societal benefits such as: flood control, irrigation, water
supply, energy generation, recreation, and pollution control. A great
percentage of dams are located near densely populated areas. Although
many benefits are gained from dams, the potential threats to public
safety and welfare cannot be ignored. The failures of Spain's Puentes
Dam in 1802, the U.S. Teton Dam in 1976, and Brazil's Germano mine

tailing dam in 2015 represent examples of the life threatening consequences resulting from unexpected or unrecognized dangers associated
with dams, as well as serve as a reminder of the importance of a robust
dam safety program. These high-profile failures resulted in stricter,
more prescriptive, regulatory procedures to better ensure safety during
the dam's service life. A dam project can be divided into three phases:
feasibility and planning (Phase I), construction (Phase II), and operation
(Phase III). For each phase, conceptual failure modes and risk
⁎ Corresponding author.
E-mail address: (C.-P. Lin).

/>0926-9851/© 2018 Elsevier B.V. All rights reserved.

assessment have been developed. Site investigation during feasibility
and planning study, quality control/assurance during construction,
monitoring programs and regular safety evaluation during operation
have been standardized to ensure public safety against risk of dam failure. Nonetheless, engineering geophysics can supplement these safeguards by enhancing the technical and economical effectiveness of the
resource management and safety throughout a dam's life cycle.
Application of engineering geophysics for Phase I site characterization was recognized as early as 1928, when I.B. Crosby and E.G.
Leonardon used electrical methods to map high-resistivity bedrock for
a proposed dam site (Burger et al., 2006). Since then, geophysical
methods have become part of the investigation program for potential
dam sites. Further growing of geophysical applications on dam mainly
focuses on the Phase III after the dam is completed. Typical dam safety
surveillance uses visual inspection, along with limited support from
geotechnical measurements. However, dams are massive structures
and their internal hydraulic conditions may require attention before
problems are detected by simple reconnaissance methods. Visual inspections do not provide information inside the dam, while the discrete
monitoring instruments provide engineering parameters with limited
spatial coverage of the dam. There is a growing demand for non-destructive geophysical techniques to internally image the dam for early
detection of anomalous phenomena and facilitating remedial actions

83

C.-H. Lin et al. / Journal of Applied Geophysics 158 (2018) 82–92

(Lee and Oh, 2018; Dai et al., 2017; Voronkov et al., 2004; Lum and
Sheffer, 2005). Nonetheless, the linkage between geophysical surveillance and engineering significance needs further strengthening. For
Phase II during construction phase and sustainable management of
Phase III, geophysical methods receive less attention. All three phases
in the life cycle of a dam are equally important, and geophysical
methods can play an equally important role in all three phases.
Near surface geophysical techniques, such as time domain reflectometry, travel time velocity tomography, electrical resistivity tomography, and multi-station analysis of surface wave, have advanced
significantly in the last couple of decades within the scientific community (Lin et al., 2015). Dam is a feat of engineering. Application and
adaptation of these methods on dams are of great interests to engineers.
A better understanding of the connections between geophysical results
and engineering significance related to dam safety and sustainability
can help engineers gain more useful information when employing
these technologies. This paper aims to broaden the view of how geophysical methods correct be applied to a dam project throughout a
dam's life cycle and strengthen the linkage between geophysical surveillance and engineering significance at all stages. A worldwide overview of geophysical applications on dams is first given. It is followed
by a collage of our studies in Taiwan to shed light on critical issues of future applications and creative developments from the perspective of
dam's life cycle. These studies are mostly newer applications of the engineering geophysical methods on dams, focusing mainly on construction and operation phases, including quality control of compacted
soils, investigation of leakage in an earth dam, evaluation of an aged
concrete dam, geophysical monitoring program in a newly-constructed
dam, and geophysical monitoring for reservoir sediment management.
2. Worldwide overview of geophysical applications on dams
Besides the site characterization in Phase I, the most frequent applications of geophysical methods in dam engineering are undoubtedly for
dam safety assessments. Many case studies from Asia, America, and
Europe are gathered and listed in Table 1. It is not meant to be comprehensive as many case studies were not published by dam owner's
choice. More cases were found in Asia simply because we have access

to some project reports that are not published in journal or conference
papers. Nevertheless, Table 1 shows the application trend and major

problems to which geophysical methods can be applied. Among these
cases, it can be seen that abnormal seepages in earth dams draw the
most attention of geophysical groups. Internal erosion is the top safety
concern in earth dams and abnormal seepage is the observable symptom as a result of it. However, depending on the source and seepage
path, not all the abnormal seepages are resulted from internal erosion.
Electrical methods such as electrical profiling, electrical resistivity tomography (ERT) and self potential (SP) method have been recognized
as water-sensitive technologies and used to investigate the spatial distribution of wetted area and possible flowing paths (Song et al., 2005;
Rozycki et al., 2006; Cho and Yeom, 2007; Panthulu et al., 2001;
Sjödahl et al., 2005; Taiwan Power Company, 2009; Al-Fares, 2011;
Engemoen et al., 2011; Moore et al., 2011; Karastathis and Karmis,
2012; Ikard et al., 2014; Lin et al., 2013; Mooney et al., 2014; Loperte
et al., 2015; Camarero and Moreira, 2017; Dai et al., 2017; Yılmaz and
Köksoy, 2017; Sentenac et al., 2018).
The other major application is investigation of the cracks or voids in
dams. The cracks or voids in dams create preferential flowing paths susceptible to further erosion evolution. Ground penetrating radar (GPR)
and seismic tomography (ST) are popular technologies for such purpose. If the voids or cracks are close to the surface, GPR may be an effective tool to quickly map their locations and depths (Xu et al., 2010; Li
and Ma, 2011). On the other hand, if the voids or cracks are too deep,
ST is a good alternative (Kepler et al., 2000). It is difficult to detect
small voids or cracks by seismic methods. The concept of applying ST
here is not to directly locate them, but to search for low velocity anomalies caused by the diffraction around the voids or cracks. The diffraction
would increase the ray path and hence reduce the estimated velocity. In
Table 1, more applications of ST can be found in dealing with the
strength of concrete dams (Hsieh et al., 2012; WRA, 2012) or seepage
in dams (Karastathis and Karmis, 2012; Dai et al., 2017).
Comparing to the application of engineering geophysics for site
characterization in Phase I, the more complex conditions in dams,
such as trapezoidal topography, zoned layers, and different targeting

problems, inspire more creative and advanced applications. For example, Cho and Yeom (2007) proposed a method named crossline resistivity tomography to investigate possible flowing path in a horizontal plan
showing spatial distribution from upstream to downstream; Moore et
al. (2011) applied a trial and error inversion of SP to study possible vertical flowing path of seepage. Furthermore, pushing geophysical

Table 1
World wild case studies of applying geophysical methods in dam safety assessments.
Area

Name of the dam

Aim of investigation

Applied geophysical method

Reference

Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
America

America
America
America
America
Europe
Europe
Europe
Europe
Europe
Europe
Europe

Shuishe Dam
Hsinshan Dam
Wushantou Dam
Shigang Dam
Xishi Dam
Sandong Dam
Unkonwn Dam in Korea
Afamia B Dam
Som-Kamla-Amba Dam
Nanshui Dam
Sanqingting Dam
S Dam
Akdeğirmen Dam
Dana Lake Dam
Amistad Dam
Barker Dam
Avon Dam
Cordeirópolis Dam

Mornos Dam
CHB Dam
EI Tejo Dam
Hällby Dam
IJKdijk test Dam
Monte Cotugno Dam
Vitineves Dam

Abnormal seepage in downstream shell
Abnormal seepage in downstream shell
Slips of slope in downstream shell
Strength of concretes after chichi earthquake
Aging concrete
Abnormal seepage in dam abutment
Abnormal seepage in downstream shell
Abnormal seepage in dam foundation
Abnormal seepage in downstream shell
Voids inside dam body
Cracks in dam body
Seepage in dam foundation (grout curtain)
Abnormal seepage in downstream shell
Abnormal seepage in dam body
Abnormal seepage in dam foundation
Cracks in dam body
Abnormal seepage in downstream toe
Abnormal seepage in dam body
Abnormal seepage in dam body
Abnormal seepage in dam body
Abnormal seepage in dam body
Abnormal seepage in dam body

Internal erosion in dam body
Abnormal seepage in dam body
Abnormal seepage in dam body

2D & 3D ERT
Time-lapse 2D ERT
MASW
Seismic tomography
Seismic tomography
2D ERT; SP
2D ERT
EM; Electrical profiling; 2D ERT
SP; Electrical profiling
GPR
GPR
Crosshole ERT; Seismic Tomography
2D ERT
SP; electrical profiling
2D ERT
Seismic Tomography
SP; 2D ERT
2D ERT
2D ERT; Seismic Tomography
SP; 2D ERT
SP
2D ERT monitoring
SP; Acoustic emission
2D ERT monitoring
EM; 2D ERT; SP

Taiwan Power Company (2009)
Lin et al. (2013)
Taiwan Chia-Nan Irrigation Association (2006)
Hsieh et al. (2012)
Water Resource Agency (2012)
Song et al. (2005)
Cho and Yeom (2007)
Al-Fares (2011)
Panthulu et al. (2001)
Xu et al. (2010)
Li and Ma (2011)
Dai et al. (2017)
Yılmaz and Köksoy (2017)
Moore et al. (2011)
Engemoen et al. (2011)
Kepler et al. (2000)
Ikard et al. (2014)
Camarero and Moreira (2017)
Karastathis and Karmis (2012)
Rozycki et al. (2006)
Rozycki et al. (2006)
Sjödahl et al. (2005)
Mooney et al. (2014)
Loperte et al. (2015)
Sentenac et al. (2018)


84

C.-H. Lin et al. / Journal of Applied Geophysics 158 (2018) 82–92

methods from investigation to monitoring is another great effort to further bring geophysics into dam engineering, be it active or passive
methods (Sjödahl et al., 2005; Lin et al., 2013; Mooney et al., 2014;
Loperte et al., 2015). All the cases in Table 1 are related to dam safety assessment of Phase III. For Phase II during construction phase and sustainable management of Phase III, geophysical methods receive less
attention. Moreover, there is still a lot of room for further strengthening
the linkage between geophysical surveillance and engineering significance. In the following, a few of our recent studies in Taiwan are introduced to shed light on critical issues of future applications and creative
developments from the perspective of dam's life cycle.
3. Geophysics in construction phase: quality control of compacted
soils
The compacted earthen dam is the most common type of dam
worldwide. Quality control and quality assurance of field compaction
relies mainly on measuring the dry density and gravimetric water content of the compacted soil. The well accepted procedure utilizes the
oven-dry method [ASTM D2216, 2011] for water content measurements and the sand-cone method [ASTM D1556, 2011] for total density
measurement. Dry density is then calculated from the water content
and total density, making the entire procedure highly time consuming.
The nuclear gauge later became available and more commonly used
for making such measurements, because it is rapid and thus does not
delay the construction activities. However, due to its regulatory restrictions and concerns over the safety and overhead of using a device with a
nuclear source, there has been increased efforts to find possible alternatives to the nuclear gauge for compaction quality control.
The compactness of a compacted soil is characterized by void ratio,
which is equivalent to dry density considering specific gravity of soil
grains is almost a constant. The moisture content affects the compaction
efficiency and soil structure, which has important implication on
hydaulic conductivity. In quality control of compacted soils, the goal is
to be able to quickly measure water content and dry density (or void
ratio) simultaneously. Geophysical properties such as dielectric constant, electrical conductivity, shear wave velocity, and thermal conductivity are all related to the composition of compacted soils (Rathje et al.,
2006). Among those, only the dielectric constant has strong relationship
with volumetric water content of soils that is relatively independent of
soil types. At least one more measurement is required to simultaneouly
determine the two parameters (gravimetric water content and dry density). Other parameters such as shear wave velocity, electrical conductivity has good correlation with void ratio or dry density. However,

this correlation is affected by water content and soil type, making quantitative estimation difficult (Rathje et al., 2006; Lin et al., 2012).
Dielectric constants of soils in the field can be measured by time domain reflectometry (TDR) technique (Topp et al., 1980; Lin, 1999). It is
based on transmitting an electromagnetic pulse through a leading coaxial cable to a sensing waveguide and recording reflections of the transmission due to changes in characteristic impedance along the sensing
waveguide. The sensing probe is designed such that there is an apparent
impedance mismatch at the start and end of the probe. TDR probes for
laboratory and field measurements of compacted soils are shown in
Fig. 1. The round-trip travel time of the EM pulse in the sensing waveguide of known length is determined from the arrival times of the two
reflections. Propagation velocity of the EM pulse can then be calculated
that determines the dielectric permittivity of the material surrounding
the probe. The dielectric constant of a soil is dominantly affected by its
volumetric water content. For immediate application, TDR is used to accelerate current standard sand-cone method, which is usded to determine “total” density of top soils. A new method named S-TDR method
was proposed. Combining the total density from sand-cone method
and volumetric water content from TDR, the dry density and gravimetric water content can be measured from the S-TDR method. Fig. 2 shows
a typical result of field tesings at the construction site of Hushan earth

Fig. 1. TDR probes and illustration of their associated electrical potentential distribution:
(a) for measurements in compaction mold and (b) for field measurements.

dam in central Taiwan. The testings were performed on the compacted
silty sand during the construction of dam shell. The result shows that
the gravimetric water content and dry density obtained by the S-TDR
method are both within 1% of the the standard conventional method
(oven dry method for water content and sand-cone method for density). The difference is smaller than the expected variation of the standard conventional method, supporting the S-TDR mehtod as a quick
alternative to the conventional method for compaction quality control.
Seismic surface wave testing has become a convenient method for
measuring shear wave velocity profile non-destructively (Xia et al.,
1999). Although shear wave velocity of a soil is affected not only by
dry density, but also by water content (i.e., matric suction) and soil
type (Cho and Santamarina, 2001), it can be used to quickly scan the
compacted area for potential problematic spots of insufficient compaction for further quantitative testing by the S-TDR method. Each compaction lift is typically 30 cm, accouting for the effective depth of

compaction energy. A mini surface-wave testing was experimented
for obtaining shear wave velocity within top 30 cm of the compacted
soil. Successful results were obainted by a small cone-shape hammer
as the impact source and a short geophone spread consisting of 12
4.5-Hz geophones with 5 cm interval, as shown in Fig. 3(a) and (b).
The phase velocities of Rayleigh wave in the frequency range between
500 Hz and 1200 Hz were obtained by the dispersion analysis, as
shown in Fig. 3(c). Fig. 3(d) shows the corresponding wavelengths
between 10 cm and 25 cm are within the targeted compaction lift.
Shear wave velocity of the compaction lift can be directly estimated
from the averaged Rayleigh wave velocity without inversion since it
was relatively uniform. The mini surface-wave testing was shown to
be a convenient method for scanning the lateral variability of shear
wave velocity for each compaction lift. Research into broadband
dieletric spectroscopy and multi-physical data fusion is currently pursuing an approach for determining water content and dry density
simultaneouly and fully non-destructively. Significant progress has
been made for dielectric spectroscopy using practical TDR probes for
both laboratory and field measurements (Lin et al., 2018). Dielectric
spectra and shear wave velocities as a function of soil physical
properites are under investigation.


85

12

(a)

+1%

11
10
9

-1%

8
8

9

10

11

12

Oven dry water content, %

Measured dry density, g/cm3

Measured water content, %

C.-H. Lin et al. / Journal of Applied Geophysics 158 (2018) 82–92

2.1

(b)

2.05

2

+1%

1.95

-1%

1.9
1.9

1.95

2

2.05

2.1

Dry density from conventional
method, g/cm3

Fig. 2. (a) Comparison of S-TDR measured water contents in the field with oven dry water content; (b) Comparison of S-TDR measured dry densities in the field with dry densities from
sand cone method.

4. Geophysics in operation phase: dam safety and management
4.1. Investigation of dam safety problems
Uncontrolled seepage is one of the most concerned problem in earth
dams. Zoned drains are fundamental elements of earthen dams
designed to control seepage through the embankment. However, preferential flow paths may develop inside the dam that initiate abnormal

seepage pathways. Inappropriate treatment to an abnormal seepage
may evolve into piping (i.e., internal erosion) of the embankment material, ultimately causing dam failure. Several successful cases in applying
electrical resistivity tomography (ERT) and self-potential method for
seepage investigations have been reported (e.g., Oh et al., 2003;
Sjödahl et al., 2005; Kim et al., 2007; Bièvre et al., 2017 among others).

Most case studies show single temporal and spatial snapshot measurements with the ERT results used simply to qualitatively support a
known situation, or provide an untested hypothesis for potential causes
or scenarios. Interpreting the results of ERT data in their correct context
can be challenging, because earth resistivity is affected by many
hydrophysical properties, including water content (or saturation),
porosity, soil composition, and cementation (Lin et al., 2012). Although
resistivity anomalies can be detected if they are of significant size and
contrast relative to the background, it is often inconclusive regarding
the engineering significance of these anomalies. Furthermore, the topography and zonation of different materials may complicate the ERT
survey and interpretation. The resistivity of neighboring zone of different material and the topology change on the two sides of the survey
line may cause 3D effects on the 2D inverted resistivity section right

Fig. 3. (a) Receivers, (b) mini source, (c) seismogram in time-space domain, (d) dispersion image in frequency-velocity domain (white line showing the dispersion curve), and (e)
dispersion curve in terms of wavelength vs. phase velocity for a mini surface wave testing on compacted soils.


86

C.-H. Lin et al. / Journal of Applied Geophysics 158 (2018) 82–92

beneath the line. Therefore, this subsurface imaging tool should be used
with caution acknowledging above factors.
Fig. 4(a) shows an example of ERT surveys to investigate abnormal
seepage in Hsinshan earth dam (Lin et al., 2013), whose main cross section is shown in Fig. 4(b). The original dam has a sloping core with the

crest at EL. 75 m. The original shell is composed of clayey sand with
low permeability close to the core. Therefore, the original dam is essentially a homogeneous dam with toe drain. With increasing demand in
water supply, it was raised by adding a core tipping further downstream
and a vertical drain was added aside the new core and on top of the original downstream face. A downstream shell was added to stabilize the
new structure. As the water level was raised over the old crest (EL. 75
m), water was found seeping out the downstream face at several
spots, as indicated Fig. 4(b). From the result of ERT surveys (Line A on
the dam crest and Line B on the downstream face along an access
road) in Fig. 4(a), two low resistivity zones were identified which are
likely related to the underground pathways of abnormal seepage. To
further understand the possible mechanism, it is necessary to integrate
geophysical results with geotechnical data.
According to the groundwater monitoring, the estimated phreatic
line is still much lower than the identified low resistivity zones and
the abnormal leakage spots on the downstream face. Therefore, the
steady-state seepage through the dam is not directly responsible for
the abnormal seepage. The results of leakage monitoring reveal high
correlation between the precipitation and flow rate of abnormal leakage. From these observations, it was hypothesized that some dirty
layer (with lower hydraulic conductivity) may have existed and trapped
the rainfall infiltration to cause the perched low resistivity zones. The
perched water migrates horizontally along the confining boundary to
an exit point on the downstream face. The one-time ERT surveys
could not fully support the hypothesis since the resistivity values are affected not only by soil moisture but also by soil types. The ERT investigation could be augmented by time-lapse measurements to provide the
variation of subsurface resistivity in direct and unique response to
change in soil moisture, whose relationship with reservoir water level
and precipitation can be examined.
The ERT survey at Line B was repeatedly conducted once a month for
one year. Qualitatively, the inverted resistivity profiles at different times

do not show significant change and the results do not seem to provide

additional information. However, more quantitative interpretation can
be made by correlating the resistivity variation with its influencing factors (i.e. the reservoir water level and precipitation). The Zone 1 (with
low resistivity) and Zone 2 (with high resistivity) marked in Fig. 4a
were considered to represent an abnormal soaked zone and a normal
permeable zone, respectively. The time variation of average resistivity
in these two areas is plotted in Fig. 5 to show its relationship with the
reservoir water level and precipitation in terms of two-week accumulated rainfall prior to each ERT measurement. The reservoir water
level was relatively stable during the monitored period. While there
was a significant variation of resistivity value in the high resistivity
zone in response to remarkable precipitation variation, the resistivity
value in the low resistivity zone remained relatively constant. The former is a normal behavior in a homogenous permeable shell, where rainfall infiltration seeps through the shell and drains to the filter beneath,
causing the resistivity to decrease during the infiltration and then
increase as the seepage drains out. The latter further supports the
hypothesis that the low resistivity zones are nearly saturated areas
with perched water. This example demonstrated that time-lapse ERT,
together with monitored precipitation and water level, can provide additional strong information if the relationship between resistivity and
hydrological factors is quantitatively analyzed.
Geophysical methods can be utilized to evaluate concrete dam as
well. Old concrete dams face different type of problems, as illustrated
by an investigation conducted in northern Taiwan. It is a concrete gravity dam with more than 90 years of service. Schmidt hammer and uniaxial strength tests performed on cored samples from the downstream
face indicated the strength of surface concrete is below regulatory
limits. The condition inside the massive dam body is unknown. Seismic
tomography (Lehmann, 2007) testing was used to assess the internal
strength of the concrete dam. Five P-wave travel-time tomography sections were conducted as shown in Fig. 6. Impact sources by rubber mallet were generated on the downstream face, and the generated waves
were received by 28 Hz hydrophones attached to the upstream side.
Both seismic source and receivers were spaced at 1 m interval. L1-L3
are vertical cross sections, whereas H1 and H2 are horizontal ones
slightly inclined to the downstream. Fig. 7 shows that most P-wave

(a)

Line A
Zone 1

Line B

Zone 2

(b)

Fig. 4. (a) ERT results of Line A at dam crest and Line B on downstream face; (b) The low resistivity zones from ERT and hydraulic heads from piezomters on the dam cross section near
abnormal leakage spots 2 (EL. 46 m) and 4 (EL. 62 m).


C.-H. Lin et al. / Journal of Applied Geophysics 158 (2018) 82–92

87

Fig. 5. Time-lapse resistivity in relation with reservoir water level and precipitation in (a) the low resistivity zone (Zone 1 in Fig. 4a) and (b) the high resistivity zone (Zone 2 in Fig. 4a).

velocities inside the dam body were between 3.0–4.2 km/s. Low velocity
spots (below 3.0 km/s, which is ranked as in poor condition according to
Whitehurst, 1951), concentrated mainly on the downstream face. No
obvious weak zone extended significantly into the dam interior. This
is a successful example showing geophysical exploration can play consequential role in the dam safety management.
4.2. Geophysical monitoring of dam health
The application of geophysical methods for dam safety can be
extended from a single investigation survey to a regular monitoring
program. As shown in the previous case study, time-lapse geophysical
measurements are appealing for process monitoring of the dam

behavior. Construction of a large embankment dam for the Hu-Shan
Reservoir in Taiwan was completed in 2016, providing a unique opportunity for geophysical monitoring of the initial water filling phase of the
reservoir as a baseline for future performance. The reservoir consists of
three zoned earth dams with 614.5 m, 393 m, and 648 m in length, respectively, and a maximum height of 75 m. A geophysical monitoring
program was devised for the new dam that includes electrical resistivity
tomography (ERT), self potential (SP), and multichannel analysis of surface wave (MASW), as shown in Fig. 8. The dominant potential failure
mode for an earth dam is seepage-related problems, justifying the use
of ERT and SP. MASW was used to measure the dynamic property (i.e.
shear-wave velocity) for the analysis of dynamic response and evaluating the strength condition of the stabilizing downstream shell. Since the

Fig. 6. (a) Field configuration of seismic tomography field testing at a concrete dam; (b) Cross-sectional schematic of source and receiver layout for L1~L3.


88

C.-H. Lin et al. / Journal of Applied Geophysics 158 (2018) 82–92

Fig. 7. Fence diagram of the seismic tomography results.

purpose of the geophysical monitoring is to evaluate potential anomalies in any location of the dam, the whole extent of the dam in the longitudinal direction are covered by 2D ERT survey lines both on the crest
(to cover mainly the core) and downstream shell. Only one transverse
survey line at the deepest section is planned to provide cross-sectional
information on seepage behavior during water filling. Other transverse
investigation may be warranted should any anomalous spots on the longitudinal section be found. 2D surveys assume the ground condition
perpendicular to the survey line is homogeneous. This assumption is
apparently violated when conducting 2D ERT surveys in the longitudinal direction of the dam due to variation of topology and filled materials
in the transverse direction. The ability of 2D ERT to detect seepage
anomalies under the influence of 3D effects has been investigated.
Dams are complex 3D structures. Even the transverse survey line does

not conform to the 2D condition due to abrupt elevation change of the
valley. 3D forward simulations were conducted for detailed planning
of the survey and evaluating potential problems and resolution limitations of 2D ERT investigation on embankments. The results show that

the effect of change in reservoir water level can be so pronounced that
the seepage anomaly is masked. In order to constrain the effect of
change in reservoir water level, we suggest ERT monitoring and time
lapse analysis be performed under similar reservoir water level and
environmental conditions (i.e., temperature and water salinity). The
first stage of water filling began in May of 2016. Initial ERT measurements were collected before water filling as a baseline for future timelapse analyses. The monitoring program provides a rare opportunity
to make geophysical observation of the seepage process in a dam.
Before impoundment, the ERT and MASW surveys on the downstream shell were conducted under different weather conditions. Of
particular interest is the shear wave velocity variation after rainfall infiltration. The largest rainfall event at the dam during the measurement
period was during Typhoon Megi in 2016. Fig. 9 shows the initial
shear wave velocity profile along L2 and the change of shear wave velocity after 200 mm of rainfall from the typhoon. The decrease in
shear wave velocity due to rainfall infiltration is rather significant and
can reach over 40% comparing to the background values measured

Fig. 8. The layout of geophysical monitoring program in the Hu-Shan Reservoir.


C.-H. Lin et al. / Journal of Applied Geophysics 158 (2018) 82–92

89

Fig. 9. (a) Shear wave velocity profile along L2 of Hu-Shan Reservoir and (b) the change of shear wave velocity after over 200 mm of rainfall after Typhoon Megi in 2016.

during the dry season. This has engineering significance in the context
of dynamic response analysis. The shear wave velocities used for the
analysis were obtained during construction and after the dam completion before impoundment. It is expected that shear wave velocity of

the upstream shell will decrease due to impoundment. Although the
downstream shell is protected against seepage by the vertical drain,
its shear wave velocity will also decrease due to rainfall infiltration.
This effect should be taken into account in the dynamic analysis to
avoid underestimation of deformation during earthquake. Geophysical
methods can monitor the physical properties of a dam non-destructively. Not only can the results be used for safety inspection, it can also
provide more realistic parameter values for related dam analyses.
4.3. Geophysical monitoring for sedimentation management
While dam safety is the number one issue of dam management, a
reservoir's sustainability relies on maintaining the storage capacity.
Unfortunately, erosion and landslides in many watersheds are aggravated due to geological weathering and climate change. Sedimentation
is becoming a serious problem in sustainable reservoir management
worldwide. Various actions are being taken to reduce the sedimentation

rate in reservoirs, including watershed management to reduce incoming sediment yield, constructing bypass structures or low-level outlets
for sediment pass-through to minimize sediment deposition, and
removal of sediment from reservoirs by dredging. Of all these measures,
sediment sluicing or density current venting through low-level outlets
is most cost-effective when hydrological conditions apply. As illustrated
in Fig. 10, turbidity currents develop when water with a high sediment
load enters a reservoir and plunges to the bottom, travelling through the
original channel until settling near the dam in what is called a “muddy
pool” (Morris and Fan, 1998). Density current venting involves the discharge of turbid sediment-laden water from a low-level outlet while
surface waters remain clear or unchanged. Management of these currents can drastically reduce sediment build-up at the base of a dam.
However, density current venting is seldom-used because density currents form only under certain hydrological conditions and the venting
operation relies on surveying of a density current. The monitoring of
density current is where engineering geophysics can play a critical
role in sediment management of reservoirs.
Commercial instrumentation for suspended sediment concentration
(SSC) monitoring is limited by particle size dependency and measurement range. A new technique based on time domain reflectometry

Fig. 10. Main questions defined for the SSC monitoring program in a reservoir.


90

C.-H. Lin et al. / Journal of Applied Geophysics 158 (2018) 82–92

(b)

Reflection coefficient, ρ

(a)

0.2
0
-0.2
-0.4
-0.6
0

(c)
1

x 10

10

20

30

40

30

40

-3

ρ'

0
-1
-2
-3
0

Pulse 1
10

Pulse 2
20
Travel time, ns

Fig. 11. (a) Illustration of a TDR pulsing system; (b) Typical step-pulse waveform of the
new coaxial TDR SSC probe; (c) the corresponding derivative of the waveform.

was developed to monitor SSC with the same basic principle of TDR
water content measurement (Chung and Lin, 2011). As shown in Fig.

11(a) and (b), the travel time between step-pulse reflections from the
start and end of a sensing waveguide is related to the dielectric constant
of turbid water, which is a function of SSC. However, due to the common
range of SSC in density currents, the required accuracy of SSC measurement is at least an order higher than that of water content

measurement. Travel time analysis in the time domain could not yield
satisfactory results. To determine the round-trip travel time of the EM
wave in the sensing waveguide with high precision, a novel algorithm
was developed with a concept borrowed from the surface wave dispersion analysis (Lin et al., 2017). By taking the derivative of the step-pulse
waveform, the impulse waveform is obtained as shown in Fig. 11(c).
The impulse reflections from the start and end of the sensing waveguide
are then extracted. They can be treated as two propagating waveforms
of two receivers spaced at a distance twice the probe length. Applying
Fourier transforms and calculating the phase shift between the two “receivers”, the frequency-dependent phase velocity can be determined. At
frequency higher than 100 MHz, the phase velocity was found independent of electrical conductivity and uniquely related to SSC.
An extensive SSC monitoring program for sediment management
was recently implemented in the Shihmen reservoir, Taiwan (Wu et
al., 2016). The Shihmen reservoir is one of the three major reservoirs
in Taiwan that are facing serious threat of sedimentation. By 2013, it
has lost nearly 30% of its total storage capacity. The monitoring program
was initiated to understand the mechanism of sediment transport in the
reservoir for planning remediation measures against sedimentation,
and later expanded to provide all the information needed for sediment
management, including total sediment income, total sediment
discharge, the formation and characteristics of density currents, and
evolution of muddy pool, as illustrated in Fig. 10. Among all the monitoring stations, the most challenging are those that are mounted on
floating platforms in the reservoir for capturing the behavior of density
current. Each monitoring station on the float consists of 8 SSC sensing
waveguides (or probes) at different water depths to survey the SSC profile. The sensing waveguides are pulsed every 30 min by a single TDR
device on the floating platform through a multiplexer. The TDR device

and data acquisition system are powered by a solar panel and two batteries that last more than 5 days without recharge.
Fig. 12 gives an example of density current monitored by such a system during Typhoon Trami, 2013. The variation of SSC in ppm with

Fig. 12. SSC profile with time at several float stations during Typhoon Trami, 2013.


C.-H. Lin et al. / Journal of Applied Geophysics 158 (2018) 82–92

depth and time was obtained at each monitoring station. When data
from all stations were assembled, a quasi 4D presentation of SSC distribution in the reservoir were generated, from which a great deal of realtime sediment information can be drawn. The inflow to the Shihmen
reservoir induced by Typhoon Trami was not particularly large. The
upstream monitoring stations on the right hand side show the increase
in SSC at shallow depth in the early stage of the storm. Later on, the formation and plunge of density current were observed. The density current migrated downstream and banked up near the dam. In the
curved channel, the roll up of density current on the outside of bend
was also observed. The accumulated muddy water downstream settled
slowly until the opening of the sluice tunnel that rapidly vented out the
muddy pool. The detailed field observation of the development and
venting of density current was unprecedented. This sediment monitoring system is valuable for effective sediment management. It is currently
realized by the TDR technique, but other more efficient waterborne geophysical survey may be possible.
5. Conclusions
Near surface geophysical techniques have advanced significantly
during the past few decades, including time domain reflectometry, electrical resistivity tomography, seismic travel-time tomography, and
multi-station analysis of surface wave. Safety and management issues
in the context of dam's sustainability, and how engineering geophysics
can play an important role in the decision-making process are corroborated by the case histories described in this paper. The collage of these
case studies is to broaden the view of how geophysical methods can
be applied to a dam project throughout a dam's life cycle and strengthen
the linkage between geophysical surveillance and engineering significance at all stages. These case studies include quality control of
compacted soils, identification of abnormal seepage pathways in an
earth dam, quality evaluation of an aged concrete dam, geophysical

health monitoring program for a newly-constructed dam, and monitoring of sediment transport for sedimentation management. In current
practice, geophysical results generally provide qualitative information.
More quantitative engineering interpretation and process monitoring
were proposed in these case studies at various stages of dam engineering. These case studies also demonstrate that, while promoting more
uses of geophysical methods on dam, it is equally important to bring
in the knowledge of dam engineering to make them optimally effective
and consequential.
Acknowledgement
Funding for this research was provided by the Water Resources
Agency and Ministry of Science and Technology, Taiwan (NSC
100-2622-E-009-007).
References
Al-Fares, W., 2011. Contribution of the geophysical methods in characterizing the water
leakage in Afamia B dam, Syria. J. Appl. Geophys. 75, 464–471.
ASTM D1556, 2011. Standard Test Method for Density and Unit Weight of Soil in Place by
the Sand-Cone Method, Annual Book of ASTM Standards. ASTM International, West
Conshohocken, PA.
ASTM D2216, 2011. Standard Test Methods for Laboratory Determination of Water
(Moisture) Content of Soil and Rock by Mass, Annual Book of ASTM Standards.
ASTM International, West Conshohocken, PA.
Bièvre, G., Lacroix, P., Oxarango, L., Goutaland, D., Monnote, G., Fargier, Y., 2017. Integration of geotechnical and geophysical techniques for the characterization of a small
earth-filled canal dyke and the localization of water leakage. J. Appl. Geophys. 139,
1–15.
Burger, H.R., Sheehan, A.F., Jones, C.H., 2006. Introduction to Applied Geophysics: Exploring the Shallow Subsurface. W.W. Norton & Company (600p).
Camarero, P.L., Moreira, C.A., 2017. Geophysical investigation of earth dam using the electrical tomography resistivity technique. Rem Revista Escola Minas 70 (1), 47–52.
Cho, G.C., Santamarina, J.C., 2001. Unsaturated particulate materials—particle level studies. J. Geotech. Geoenviron. 127 (1), 84–96.
Cho, I., Yeom, J.-Y., 2007. Crossline resistivity tomography for the delineation of anomalous seepage pathways in an embankment dam. Geophysics 72, G31–G38.

91

Chung, C.-C., Lin, C.-P., 2011. High concentration suspended sediment measurement using
time domain reflectometry. J. Hydrol. 401, 134–144.
Dai, Q., Lin, F., Wang, X., Feng, D., Bayless, R.C., 2017. Detection of concrete dam leakage
using an integrated geophysical technique based on flow-field fitting method.
J. Appl. Geophys. 140, 168–176.
Engemoen, W., Mead, R., Hernandez, L., 2011. Evaluating the risks of an internal erosion
failure at Amidtad dam. Proceeding of 31st Annual USSD Conference, San Diego,
California, pp. 527–543.
Hsieh, S.-H., Hsieh, C.-H., Hu, C.-H., Huang, J.-K., Lin, C.-M., 2012. Application of elastic
wave tomography for dam safety. Symposium on the Application of Geophysics to
Engineering and Environmental Problems, 25th Annual Meeting SAGEEP 2012,
Tucson, Arizona, USA.
Ikard, S.J., Revil, A., Schmutz, M., Karaoulis, M., Jardani, A., Mooney, M., 2014. Characterization of focused seepage through an earthfill dam using geoelectrical methods.
Groundwater 52 (6), 952–965.
Karastathis, V., Karmis, P., 2012. Investigation of seepage and settlement problems at the
Mornos earth dam, Greece, by geophysical methods. Symposium on the Application
of Geophysics to Engineering and Environmental Problems, 25th Annual Meeting
SAGEEP 2012, Tucson, Arizona, USA.
Kepler, F.W., Bond, L.J., Frangopol, D.M., 2000. Improved assessment of mass concrete
dams using acoustic travel time tomography. Part II-application. Construct. Build.
Mater. 14, 147–156.
Kim, J.H., Yi, M.J., Song, Y., Seol, S.J., Kim, K.S., 2007. Application of geophysical methods to
the safety analysis of an earth dam. J. Environ. Eng. Geophys. 12, 221–235.
Lee, B., Oh, S., 2018. Modified electrical survey for effective leakage detection at concrete
hydraulic facilities. J. Appl. Geophys. 149, 114–130.
Lehmann, B., 2007. Seismic Traveltime Tomography for Engineering and Exploration
Applications. (273 p). EAGE Publications.
Li, H., Ma, H., 2011. Application of ground penetrating radar in dam body detection. Proc.
Eng. 26, 1820–1826.
Lin, C.-P., 1999. Time Domain Reflectometry for Soil Properties. Purdue University (Ph.D.

Thesis, 229 p).
Lin, C.-H., Lin, C.-P., Drnevich, V.P., 2012. TDR method for compaction quality control:
multi evaluation and sources of error. Geotech. Test. J. 35 (5), 817–826.
Lin, C.-P., Hung, Y.-C., Yu, Z.-H., Wu, P.-L., 2013. Investigation of abnormal seepages in an
earth dam using resistivity tomography. J. GeoEng. 8, 61–70.
Lin, C.-P., Lin, C.-H., Wu, P.-L., Liu, H.-C., 2015. Applications and challenges of near surface
geophysics in geotechnical engineering. Chin. J. Geophys. 58, 2664–2680.
Lin, C.-P., Ngui, Y.-J., Lin, C.-H., 2017. A novel TDR signal processing technique for measuring apparent dielectric Spectrum. Meas. Sci. Technol. 28, 015501.
Lin, C.-P., Ngui, Y.-J., Lin, C.-H., 2018. Multiple reflection analysis of TDR signal for complex
dielectric spectroscopy. IEEE Transactions on Instrumentation and Measurement (in
print).
Loperte, A., Soldovieri, F., Lapenna, V., 2015. Monte Cotugno dam monitoring by the electrical resistivity tomography. IEEE J. Sel. Top. Appl. Earth Observ. Remote Sens. 8 (11),
5346–5351.
Lum, K.Y., Sheffer, M.R., 2005. Dam safety: review of geophysical methods to detect seepage and internal Erosion in embankment dams. Hydro-Review 29. http://www.
hydroworld.com/articles/hr/print/volume-29/issue-2/articles/dam-safety-review.
html.
Mooney, M.A., Parekh, M.L., Lowry, B., Rittgers, J., Grasmick, J., Koelewijn, A.R., Revil, A.,
Zhou, W., 2014. Design and Implementation of Geophysical Monitoring and Remote
Sensing During a Full-Scale Embankment Internal Erosion Test, 2014 ASCE
GeoCongress, Atlanta, USA.
Moore, J.R., Boleve, A., Sanders, J.W., Glaser, S.D., 2011. Self-potential investigation of moraine dam seepage. J. Appl. Geophys. 74, 277–286.
Morris, G.L., Fan, J., 1998. Reservoir Sedimentation Handbook: Design and Management of
Dams, Reservoirs, and Watersheds for Sustainable Use. McGraw-Hill Book Co., New
York, New York.
Oh, Y.C., Jeong, H.S., Lee, Y.K., Shon, H., 2003. Safety Evaluation of Rock-Fill Dam by Seismic
(MASW) and Resistivity Method, Proceedings of the 16th Annual Symposium on
the Application of Geophysics to Engineering and Environmental Problems.
pp. 1377–1386.
Panthulu, T., Krishnaiah, C., Shirke, J., 2001. Detection of seepage paths in earth dams
using self-potential and electrical resistivity methods. Eng. Geol. 59, 281–295.

Rathje, E.M., Wright, S.G., Stokoe II, K.H., Adams, A., Tobin, R., Salem, M., 2006. Evaluation
of Non-Nuclear Methods for Compaction Control, Rep. No.: FHWA/TX-06/0-4835-1.
Center for Transportation Research, The University of Texas at Austin, Texas Department of Transportation />pdf.
Rozycki, A., Ruiz Fonticiella, J.M., Cuadra, A., 2006. Detection and evaluation of horizontal
fractures in earth dams using the self-potential method. Eng. Geol. 82, 145–153.
Sentenac, P., Benes, V., Keenan, H., 2018. Reservoir assessment using non-invasive geophysical techniques. Environ. Earth Sci. 77, 293.
Sjödahl, P., Dahlin, T., Johansson, S., 2005. Using resistivity measurements for dam safety
evaluation at Enemossen tailings dam in southern Sweden. Environ. Geol. 49,
267–273.
Song, S.H., Song, Y., Kwon, B.D., 2005. Application of hydrogeological and geophysical
methods to delineate leakage pathways in an earth fill dam. Explor. Geophys. 36,
92–96.
Taiwan Chia-Nan Irrigation Association, 2006. The Plan of Remediation Program of
Wushantou Reservoir (1st Year) (In Chinese).
Taiwan Power Company, 2009. Technical Report of the Third Safety Assessment of Sun
Moon Lake Reservoir (In Chinese).
Topp, G.C., Davis, J.L., Annan, A.P., 1980. Electromagnetic determination of soil water content: measurement in coaxial transmission lines. Water Resour. Res. 16, 574–582.


92

C.-H. Lin et al. / Journal of Applied Geophysics 158 (2018) 82–92

Voronkov, O.K., Kagan, A.A., Krivonogova, N.F., Glagovsky, V.B., Prokopovich, V.S., 2004.
Geophysical methods and identification of embankment dam parameters: In
Proceedings of the 2nd International Conference on Geotechnical and Geophysical
Site Characterization, 593–599.
Water Resource Agency, 2012. Technical Report of Application of Geophysical methods in
Dam Safety Evaluation (In Chinese).
Whitehurst, E.A., 1951. Soniscope tests concrete structures. J. Am. Concr. Inst. 47, 433–444.

Wu, I.-L., Lin, C.-P., Chung, C.-C., 2016. Technical Report on Sediment Transport Monitoring Facility Maintenance and Flood Event Observations in Shihmen Reservoir. Water
Resource Agency, Taiwan (In Chinese).

Xia, J., Miller, R.D., Park, C.B., 1999. Estimation of near surface shear wave velocity by
inversion of Rayleigh waves. Geophysics 64, 691–700.
Xu, X., Zeng, Q., Jin Wu, D.L., Wu, X., Shen, J., 2010. GPR detection of several common subsurface voids inside dikes and dams. Eng. Geol. 111, 31–42.
Yılmaz, S., Köksoy, M., 2017. Application of electrical resistivity imaging and self-potential
methods to determine leakage pathways in an earth fill dam. Pamukkale University
Journal of Engineering Sciences 23 (6), 799–803.