APPLICATION OF GEOPHYSICAL METHODS FOR PLANNING

HAZARD MITIGATION IN MOUNTAIN AREAS

Surendra Raj Pant

Central Department of Geology, Tribhuvan University,
Kirtipur, Kathmandu, Nepal

INTRODUCTION

Landslides and glacier lake outburst floods are two main causes that
destroy properties and lives of mountain people. Investigation and time series observation
of landslides and glacier lakes by geophysical methods provide valuable information.
Before 1980s geophysical methods were mainly used for mineral and hydrocarbon
exploration. In late 1980s new range of geophysical methods and improved data
processing techniques in “traditional” geophysical methods has raised the standards and
effectiveness of the methods. These are high resolution data acquisition and processing in
seismic refraction, reflection, electrical resistivity and ground-penetrating radar. These
methods are able to produce detail tomographic image of shallow subsurface.

The application of geophysics in landslide and glacier is mainly concerned with

the study of material types and structures. The objectives of geophysics in landslide
investigation are to study material type, depth to water table and bedrock, and location of
possible slip surface. The bases for the application of geophysical methods in a landslides
and critical slopes are different material types with different physical properties and
differentiation of physical properties between the sliding and intact masses. The
objectives of geophysical methods in glacier study is to find out thickness of glacier,
location of seepage in the bottom of glacier lake and its dam, ice detection in lake bottom
and in its natural dam, and mapping of frozen and melting ground. Time series
observation by geophysical methods are usually carried out to monitor glacier and glacier
lake dam.

Geophysical methods are environmentally benign. There is no disturbance in

subsurface materials during an investigation. These methods are also cost effective: a
large area can be surveyed in a very short time with much detail subsurface information.

INVESTIGATION METHODS

Modern geophysical methods can be applied to solve a range of problems related

to geology, hydrogeology and environment. These geophysical methods can be applied to
study landslides and glaciers. Selection of any geophysical methods for landslides,
glaciers, glacier lake and natural dams largely depends on available budget, geophysical
instruments, geology of the site, political and social constraints.

1|P age
 Seismic Methods

In seismic methods a signal is generated at a time that is known as a seismic

waves. Seismic waves test the extent to which earth materials can be stretched or
squeezed somewhat as we can squeeze a sponge. These waves cause the particles of a
material to vibrate, which means that these particles are temporarily stretched out of
position as they move back and forth. The capacity of a material to be temporarily
deformed by passing seismic waves can be described by its properties of elasticity. These
physical properties can be used to distinguish different type of materials. As these waves
passes through subsurface media they will be reflected and refracted back to the surface
where the returning signals are detected. The elapsed time between the source being
triggered and the arrival of various waves is then used to determine the nature of the
subsurface layers.

There are two types of body waves that travel through the elastic medium. P-
waves, which are also known as longitudinal, primary, push or compressional waves and
S-waves, also known as transverse, secondary or shear wave. The particle motion for P-
wave is in the direction of the propagation and for S-wave is at right-angles to the
direction of propagation. The velocity of the propagation of seismic wave is given by the
expression

V= (K/ρ)1/2 (1)

where V is P- or S- wave velocity, K is the effective elastic parameter, and ρ is the

density of the medium. K is a function of Lame’s constants, λ and µ, which are related to
how a material responds to normal and shearing stresses. The effective elastic parameter
for the P-wave velocity is

K= λ + 2µ (2)

and for S-wave velocities, it is

K=µ (3)

Seismic Reflection Method

Since around 1980 the method has been widely used for mapping Quaternary deposits,
buried rock valleys and shallow faults, hydrogeological studies, etc. (Reynolds, 1997).
Seismic reflection method can be useful for the study of landslides. The method has the
capability to provide detail image of the subsurface of a landslide. The essence of the
seismic reflection technique is to measure the time taken for a seismic wave to travel
from a source down into the interface where it is reflected back to the surface and then
detected at a receiver placed at the surface (Figure 1). The receivers are called

2|P age
geophones. Geophones change ground motion into electrical energy. The output of the
geophones are further filtered, amplified and recorded.

Time

To

Distance (X)

Detector (D) Source (S) Detector (D)

h
Layer 1

o Reflector o'
Layer 2

Figure 1. Raypath diagram illustrating seismic reflection and its time-distance graph.

One of the principle reason for the increased use of shallow seismic reflection is
mainly improvement in equipment capabilities and availability of microcomputers for
processing. The equipment produced recently are portable and can be carried by one
person. New shallow seismic equipment can be used both for seismic reflection and
refraction, can record both P- and S-wave.

The new data acquisition technique applied in reflection method helps to enhance
signal/noise ratio. In this method more than one shot is used and reflections arising from
the same point on the interface will be detected at different geophones placed at the
surface (Figure 2). This common point of reflection is known as the common mid point
(CMP). Sometimes common depth point (CDP) or common reflection point (CRP) are
used synonym to common mid point. Nowadays almost all reflection data are acquired
and processed by this method. The waves reflected from the same point on the interface
will be staked. By staking several traces signals tends to be stronger whereas noise tends
to be reduced. The stacking is carried out after necessary corrections made on every trace
recorded. The usual way of presentation of seismic reflection results is a time section.

3|P age
 S1 S2 S3 S4 D1 D2 D3 D4

Reflector

Common mid-point

Figure 2. Principle of common mid point (CMP), S-source and D-detector.

Seismic Refraction Method

Snell’s law defines the relationship between the angle that the incident ray and the
angle that the transmitted (refracted) ray make with the normal to the boundary. At some
angle of incidence equal to critical angle of incidence the angle of refraction will be 900.
The necessary conditions for the waves to be critically refracted is that the velocity of
seismic waves in the lower media should be higher than the upper media. The waves that
have the angle of refraction 900 are called critically refracted waves or head waves. These
waves will travel along the interface with the velocity of lower media. While traveling
along the interface these waves will initiate vibration in the upper media. These vibrations
(seismic wave) eventually will reach to the surface and will be detected by geophones.
From here onwards by refracted waves we have to understand critically refracted waves.

The time-distance relationship for two layer case is given by equation

tx = x/V2 + 2h1 cos i/V1 (4)

The equation (4) is in the form of straight line. The second term in the right hand side of
the equation is the intercept of the refracted line with time axis. This is called the
intercept time. An example of the picking up of first arrival time of seismic wave and
their plot against geophone distance are presented in Figure 3. Similarly the time distance
relationship for multi-layered (n-layered) case can be calculated.

n −1
tx = x/Vn + ∑ 2 h j cos ij/ Vj (5)
j =1

4|P age
 Secondary arrivals

re fractor
Intercept time vals from
First arri

e
Time (t)

av
w
ct
re
di
m
ro
lf
va
rri
ta
rs
Fi

Offset distance (x)

Geophones
1 2 3 4 5 6 7 22 23 24
Source
direct wave

h
ic ic
V1
90 Critically refracted ray V2

Figure 3. Time-distance graph and direct and critically refracted ray paths in the ideal
two-layer case.

Refraction Data Acquisition

The processing and interpretational methods also demands on field operation. To

full fill this purpose three additional shots, off-end forward shot, off-end reverse shot and
mid-point shot, can be conducted. For the continuous coverage of subsurface along a long
seismic refraction profile, at least two geophone positions in the first profile are
overlapped by the next profile. For shallow seismic refraction first-break arrival time are
noted. These first-break arrival times are used for further processing.

After the completion of field work it is necessary to carry out detail checking of
travel time graphs. This is necessary to assure the quality of data and to select the
interpretational methods. This may help to optimize time and cost. During the selection of
interpretational methods it is strongly recommended to consider the geology, morphology
of the refractor, hydrogeology and probable thickness of the overburden of a landslide.

5|P age
Among several methods of data processing and interpretation only two methods are
considered to be useful for processing of data acquired in a landslide. They are method t0
(conventional method) and generalized reciprocal method (GRM).

Method t0

This method of interpretation is the modification of Plus-minus method

originally proposed by Hagedoorn (1959). Method t0 helps to determine position of a
undulating refractor. The method is based on the following assumption: 1) the refracted
waves slides along the interface i.e. there is no effect of penetration in the refractor; 2)
radius of curvature of refractor is very high than the depth to the refractor; 3) refractor
velocity does not have any sharp change (Gurvich, 1972). As the thickness of the layers
above the refractor increases the real situation will be different.

After identifying first breaks from target refractor and phantoming, data are
subjected to calculation of target refractor velocity (Vb). For this purpose the difference
curve θ(x) is determined by the relation

θ(x) = t(x) - t’(x)+ T (6)

where t(x)- time reading at distance x, in forward phantomed curve. t’(x) - time reading at
distance x, in reverse phantomed curve. T- reciprocal time. The reciprocal time for the
forward shot and reverse shot should be within good agreement. By taking the inverse of
the slope of the curve θ(x), Vb is calculated

∆x
Vb = 2 (7)
∆θ
The change in the alignment of the points in the difference curve indicates the
change in the velocity of the target refractor. The curve t0 (x) and depth to the target
refractor h(x) for every x are calculated by using following relation

t0 (x) = t(x) + t’(x) - T (8)

h(x) = K t0 (x) (9)

Vav.Vb
where K = , Vav - average velocity of the overburden (above the
2 Vb 2 − Vav 2
target refractor)

The average velocity of the overburden is calculated by assuming it to be equal to

the effective velocities.

An example of the interpretation by method t0 is presented in Figure 4. t0 (x)

curve has the negative correlation with the refractor topography. θ(x) shows the

6|P age
possibility of several velocity changes along the target refractor. The depth calculation
shows that the overburden is very thick. In general as the thickness of the overburden
increases , the probability of layering within the overburden is very high. Furthermore
undulation of refractor topography may produce several fictitious velocity changes in the
target refractor. If this is the case how reliable are the interpretation? We will discuss in
more detail in case histories.

80 Time, ms
70
θ(x)
60
t‘(x) t(x)
50

40

30

20 t0(x)

10
surface
0

-10
871 m/s Layer 1
3100 m/s
-20 Layer 2
Depth, m

Figure 4. Time-distance graph and their interpretation by method t0, profile D-D’, Xiakou
Landslide, Sichuan.

GRM Processing And Interpretation

GRM method is considered to be superior than other conventional reciprocal

methods (method t0, ABC method, ITM, etc.) (Lankston, 1990; Palmer, 1981). The
generalized time-depth migration overcomes the errors related to dip, includes separation
of geophones and shot point delay times, migration and convergence correction into
single operation (Palmer, 1981). Spreadsheet interpretation of seismic refraction data are
more popular for plus-minus method (Fourie and Odgers, 1995). Although there are
commercially available software, the GRM processing can also be executed by using a
spreadsheet program such as EXCEL. The graphic presentation of EXCEL spreadsheet
are more useful for the selection of optimum parameters of the GRM processing.

The velocity analysis forms an initial part of the GRM processing. The velocity
analysis function tv is defined by the equation (Palmer, 1981)

tv = (tAY - tBX +T)/2 (10)

7|P age
where tAY and tBX are the arrival time for head waves from forward and reverse shots
that have been emerged from the same point in the refractor and arrived at locations Y
and X at the surface and T- the reciprocal time.

In the GRM processing the velocity analysis is conducted by using Palmer’s

minimum detail criteria (Palmer, 1981): the curve that exhibits the least amount of
irregularity.

The process of migration is important for both seismic refraction and reflection
data interpretation. Time-to-depth migration accomplishes in two stages. The first stage is
calculation and interpretation of time-depth function (Palmer, 1981)

tG = (tAY + tBX - (tAB + XY/Vn)/2 (11)

where, Vn is the refractor velocity determined from velocity function.

The time-depth analysis is conducted by using Palmer’s maximum detail criteria.

The time-depth curve for the optimum XY value gives the shape of the refractor. One of
the main advantage of GRM processing is calculation of average velocity above the target
refractor without defining all layers. The expression for the calculation of average
velocity is

V = Vn'2 XY/(XY+2tGVn' ))1/2 . (12)

The total thickness of all the layers above the refractors can be computed by the
following equation

n −1

∑Z
j =1
jG = tG V /cos i (13)

where
i =sin-1 ( V / Vn' ).

Electrical Resistivity Methods

Electrical resistivity is a fundamental and diagnostic physical properties that can

be determined by a wide variety of techniques, including electromagnetic induction. At
present wide application of electrical resistivity methods is due primarily to the
availability of micro computers for data processing. The methods has got broad sphere of
application in engineering and environmental studies, groundwater, glaciology,
archaeology, and mineral and petroleum exploration.

In comparison to other physical properties the range of variation of electrical

resistivity is very high. It gives the possibility to differentiate material type in a landslide

8|P age
than other geophysical methods (Novitskii, 1974). The resistivity contrast between sliding
mass and intact mass allows mapping sliding mass, slip surface, estimation of intensity
of fracture in bedrock and forecasting landslides (Brodovoi and Nikitin, 1984).

Let us consider a semi-infinite solid with uniform resistivity ρ. Assume that the
current I is introduced into the ground through the electrode A and B and the potential
difference ∆V is picked up between electrodes M and N. The resistivity is calculated by

∆V 1
ρ = 2π
I 1 1 1 1
− − +
r1 r2 r3 r4
∆V
ρ =K (14)
I

where K is called geometrical coefficient. If the resistivity is measured for

inhomogeneous and anisotropic ground, the resistivity thus calculated is called apparent
resistivity and is denoted by ρa.

A M N B
r1 r2
r3 r4

Figure 5. Arrangement of current electrodes (A and B) and potential electrodes (M and N)

The responses of the current on geological formations, such as clay, silt, sand,
gravel, boulders, and bedrock, are different. The bedrock if it is weathered and fractured
and saturated with water will have low electrical resistivity than the fresh rocks. Different
material type mentioned above usually will have different degree of porosity and
permeability. For porous formations electrical resistivity depends on the amount of water
present and the salinity of the water. Thus an electrical resistivity of a saturated rock or
sediment is a function of porosity, electrical resistivity of saturating fluid, the resistivity
of solid rock or sediment, the surface conduction of rock or sediment and the turtuosity of
fluid and electrical path. Surface conduction is a form of ionic transport which takes place
at solid-liquid interface by means of exchange mechanisms.

An empirical relationship for the dependence of resistivity of saturated rocks on

the porosity and fluid resistivity ρw is given by Archie’s Law:

9|P age
 ρ = a ϕ-mS-nρw (15)

where S is the fraction of pores that are fluid filled, n=2, and a and m are constant:

0.6<a<2.5; 1.3<m<2.5

ρ = F S-n ρw (16)

F is called formation factor.

There are mainly three procedures of resistivity measurement. Application of any

procedures depends upon problems to be solved and the stage of the investigation. These
are electrical profiling, vertical electrical sounding and 2-D profiling. 2-D profiling is also
called electrical resistivity tomography(ERT) and/or subsurface imaging (SSI).

Data acquisition

Electrical Profiling (EP): In profiling inter-electrode distances are kept fixed, and the
array of the electrodes is moved laterally along a profile, the apparent resistivities being
plotted at the mid points. This is called continuous profiling or simply profiling. In
profiling search depth will remain same throughout the profile. Profiling is used to find
out lateral variation in resistivity. This method is useful in mapping loose and sliding
mass from intact mass in the slope, fracture zones in rocks, contact between ice and
saturated material, or steeply dipping contacts between different type of earth materials.

Vertical Electrical Sounding (VES):VES is conducted to find out resistivity variation

with depth. This is done by increasing the distance between outer current electrodes or by
increasing relative distance between current and potential electrodes. The center of the
measurement is kept fixed. The curve plotted in a bi-logarithmic paper electrode spacing
versus apparent resistivity. This is known as sounding curve which infers variation of
electrical resistivity with depth.

2-D Electrical Profiling: 2-D electrical profiling measures both vertical and lateral
change in electrical resistivity. This can be considered as the combination of both
profiling and sounding. By increasing the distance between current and potential
electrodes effect of the sounding can be obtained and by shifting current electrode in the
next station effect of the conventional profiling can be obtained. The method is useful for
preparation of electrical resistivity tomography (ERT) or image of the subsurface of a
landslide and glacier lake dam.

Interpretation

Qualitative interpretation is the first stage of the interpretation of electrical

resistivity data. All conventional electrical profiling data undergo this type of

10 | P a g e
interpretation only. The end products of electrical profiling data are apparent resistivity
maps and profiles.

The mathematical analysis for quantitative interpretation of resistivity results is

highly developed for VES data for depth probing. Recent development in finite element
processing technique applied in electrical resistivity methods made it possible for the
inversion of 2-D profiling data for the preparation of electrical resistivity images and 2-D
geological profiles.

Filtration Potential Method

Filtration potential are related to the movement of groundwater in porous rocks.

They are found everywhere in nature and can be observed by method of self potential.
The method can be used for mapping surface materials of a landslide and to find out
prominent groundwater movement direction in shallow depth of a landslide. The method
has been successfully used for locating seepage area in the bottom of the reservoir. Time
series observation of filtration potential can be useful for monitoring condition of earth
dam.

Since filtration potential method is natural source method: only two receiving
potential electrodes are necessary for measurement. Receiving electrodes are made of
porous pot that contains a metal immersed in its concentrated salt solution. For the
acquisition of data in a landslide one of the potential electrode is kept fixed outside the
landslide and other electrode is moved to measure along a profile. The end product are
presented as maps or profiles. The recharge area will be indicated as negative and
discharge as positive potential areas. The area with predominant movement of
groundwater will be indicated by high intensity potential areas than others.

There are other methods that can be applied for the investigation of landslides.
Among these are ground conductivity electromagnetic and ground penetrating radar
(GPR) methods. All these methods may be useful for shallow depth investigation, usually
few meters to few tens of meters.

EXAMPLES

Investigation Of Xiakou Landslide, Yaan City Area, Sichuan, China

Xiakou Landslide was investigated by seismic refraction and electrical resistivity

methods. The objectives of seismic refraction was to find out depth to bedrock. Where as
electrical resistivity was used to know about material type in the overburden.

VES curves were interpreted both qualitatively and quantitatively. The curves
indicate that the water table is at shallow depth (1-3 m). VES curves indicate that a very
low resistivity layer that overlies the bedrock is found every where. One of the

11 | P a g e
representative sounding curve along with its quantitative interpretation is presented in
Figure 6.

ρa Ohm.m
1000

S u rfa ce la y e r

B e d ro c k
100
B o u ld e rs m ixe d
w ith fin e m a te ria ls

D e p th to C la y d o m in a n t
w a te r ta b le m a te ria l

10

1 10 100 1000
E L E C T R O D E S P A C IN G A B /2 (m )

Figure 6. Representative sounding curve with its curve matching results, Xiakou
Landslide, Sichuan, China.

The result of quantitative interpretation of VES curves were presented as

geoelectric sections. Geoelectric section G-G’(figure 6) is presented as an example. The
profile is across the slope of the landslide. Northern part of the profile G-G’ reveals four
electric layer and reduces to three electric layer in the south. The first layer is very loose
surface layer above water table, second layer as a cumulative effect from different layers
of clay and boulders or clay and boulder mixed together and the third layer above the
bedrock is due to clay dominant material. The fourth electric layer effect is from bedrock.

12 | P a g e
 North
ES 5
ES24
820 450 ES14 ES26 South
400
325 180
Elevation (m)

85 100
800 800

Elevation (m)
230 28
? 40
780 35 150 780
150
190
760 300 760

Scale Legend
0 20 40 60 m

Bouldrs mixed with

Rock
fine materials

ES14 Electrical sounding

Predominantly clay
190 Electrical resistivity
in Ohm.m
Figure 7. Geoelectric section G-G’, Xiakou Landslide, Sichuan, China.

All seismic profiles were interpreted by method t0. The results for some of the
profiles interpreted by method t0 were not equitable to bedrock geology. This is the case
for the profiles where the overburden is thick. The difference was much more
pronounced in the profiles where thick or more clay was expected. The velocity of
seismic waves in bedrock as deduced by method t0 is extremely affected by bedrock
topography. Due to the bedrock topography several fictitious velocity changes were
found.

To overcome this situation a more sophisticated method of data processing and

interpretation was felt to be necessary. This is generalized reciprocal method (GRM). All
seismic profiles were processed by GRM. Final migrated section of GRM processing for
one of the profile is presented in Figure 8. A suite of arcs are given and the interface
between the target refractor and overburden is tangent to the arcs. The results of GRM
processing also prove that the conventional method of interpretation can be justified only
for shallow depth investigation (less than 10-15 m).

Overburden

Bedrock

13 | P a g e
 Figure 8. Final migrated depth section, Profile B-B’, Xiakou Landslide, Sichuan,
China.

Electrical Resistivity Tomography (ERT) of Landslides at Km 104, Arniko Highway

2-D electrical profiles were conducted at the landslides below road level to check
rock at shallow depth for anchoring of sliding materials. The site was investigated by
three profiles. One of the electrical tomography is presented in Figure 9. This
tomographic image is able to show more detail subsurface information than conventional
1-D methods.

Figure 9. Electrical resistivity tomography of a landslide. The rocks and the materials
above watertable (or above slip surface ) are indicated by high electrical resistivity zones
and saturated and fine materials by low electrical resistivity zones.

Electrical Tomography of Thulagi Glacier Lake Dam, West Nepal

Electrical imaging at Thulagi glacier lake was used to delineate buried glacier ice
and permafrost zones within the dam containing the lake. Thulagi Glacier Dam was
studied by seven 2-D profiles. Data were processed using a 2-D inversion program and by

14 | P a g e
finite-difference forward modeling. Interpretation of the processed electrical images
indicates that the method can be used to detect the presence of buried ice and of
permafrost by virtue of their very high electrical resistivities (>50,000 Ohm.m and
>10,000 Ohm.m, respectively). One of the longitudinal section is presented in Figure 10.
The resistivity of permafrost is strongly dependent on the material particle size: the finer
the particle size the lower is the value of electrical resistivity. Water-saturated glacial
sediments have values of electrical resistivity less than 3,000 Ohm.m. The ERT method
has been demonstrated to be useful in the mapping of buried glacier ice and permafrost as
well as in the differentiation of other geological materials present.

Figure 10. Electrical tomography of Thulagi Glacier Lake Dam, West Nepal.

REFERENCES

Brodovoe, V. V. and Nikitin, A. A., 1984. Complexirovania metodov razvedochnoi

geofiziki. Moskva "Nedra", 385 p. (in Russian).

Fourie, C.J.S. and Odgers, A.T.R., 1995. Spreadsheet interpretation of seismic refraction
data. Pergamon, Computers & Geosciences, Vol.21, No.2, pp. 273-277.

Gurvich, I.I., 1972. Seismic Prospecting. Mir Publishers, Moscow, 463 p.Hagedoorn,
J.G., 1959. The plus-minus method of interpreting seismic refraction sections.
Geophysical Prospecting , 7(2), pp.158-182.

15 | P a g e
Lankston, R.W., 1990. High-resolution refraction data acquisition and interpretation. In:
Ward, S.H. (ed.), Geotechnical and environmental geophysics. Vol. I: Review and
Tutorial. Tulsa: Society of Exploration Geophysicists, pp. 45-73.

Novitskii, G. P., 1974. Complexirovannia geofizicheskikh metodov razvedki. Izdatelstvo

"Nedra", Leningradskoe otdelenie, 256 p. (in Russian).

Palmer, D., 1981. An introduction to generalized reciprocal method of seismic refraction

interpretation: Geophysics, 46, pp.1508-1518.

Reynolds, J.M., 1997. An Introduction to applied and environmental geophysics. John

Wiley & Sons, 796 p.

16 | P a g e