Full text of DOI:10.1007/BF02909781

Vol. 44 No. 5
SCIENCE IN CHINA (Series D)
May 2001
Investigation of a landslide in the new site of Badong County
by integrated geophysical survey
LIU Song (刘 崧), ZHANG Shuangxi (张双喜) & LI Jiangfeng (李江风)
China University of Geosciences, Wuhan 430074, China
Correspondence should be addressed to Liu Song (email: wtb@cug.edu.cn)
Received September 30, 1999
Abstract An integrated geophysical survey which combines vertical seismic profile method, shal-
low reflection seismic method, electric sounding, soil temperature measurement and radioactive
gas measurement was used to investigate Zhaoshuling landslide in the new site of Badong County
and to assess the stability of the landslide. By rational use of these methods together with bore-
hole geological profile and other geological information, the spatial distribution of the landslide
body, the formations and structures within and without the landslide body were determined and the
stability of the landslide was also assessed, thus making great contribution to the successful and
rational investigation and assessment of the landslide.
Keywords: landslide, investigation of a new site, stability assessment, integrated geophysical survey, vertical seismic pro-
file.
The well-known Three-Gorge Project has started. The Badong County located on the south-
ern bank of Yangtze River is to be moved to a new site. Therefore investigation of the new site is
an important task. During the investigation, however, in the centrical part of it, a landslide named
Zhaoshuling landslide was discovered. Consequently, a correct assessment of the stability of the
landslide is of great importance. Therefore, in this area, a detailed geological engineering survey
which used traditional techniques, including drilling, trenching and tunneling, was carried out by
the Engineering Geological Party which is responsible for the investigation of the new site to in-
vestigate the formation and structures in the landslide slope and its water saturation, as well as the
properties and status of the soil and rocks comprising the slope. Owing to certain limitations of
these techniques, however, it was still difficult to investigate and assess all these problems satis-
factorily and economically, and a reasonable assessment of the landslide could not be finally made.
Therefore, an integrated geophysical survey was conducted to improve and to speed up the inves-
tigation and assessment.
1
Geological setting
The area studied is located on the southern bank of the Yangtze River, 6 km west of Badong
County. The area is made up of Middle Triassic rocks and Quaternary deposits, as is shown in the
geological map of the area (fig. 1).
The Middle Triassic rocks are mainly composed of the second and third members of Badong
group, which is named by F.V. Richthofen and is equivalent to Anisian. It is subdivided into four

No. 5
A LANDSLIDE IN THE NEW SITE OF BADONG COUNTY
427
members (table 1). There are two members in the study area. One is the second member (T2b2),
which is composed of purplish-red fine-grained argillaceous siltstone and silty mudstone
interbedded, the other is the third member
(T2b3), which is composed of whitish-grey
thin-bedded, medium-bedded and thick-bedded
argillaceous limestone and medium-bedded to
thick-bedded limestone interbedded, contain-
ing some marls. Quaternary deposits (Q) in-
clude mainly landsliding, collapsing and
flooding deposits.
Table 1 Subdivision of the Badong group in the upper
Yangtze area in China[1]
Stage
Member
T2b4
T2b3
T2b2
Anisian T12
Badong T2b
T2b1
The major structure in the area is charac-
terized by near-E-W-trending multiple folds, a
series of reversed faults as well as a great deal
of joints and fissure systems. Guandukou syn-
cline is a representative one among the fold
structures. On the plan, the axis of the syncline
strikes nearly E-W, its axial trace extending
along the southern bank of the Yangtze River
in the study area. On the section, the syncline
manifests itself, on the whole, as an isothick
symmetrical fold. On the two flanks of the
syncline, there exist many unsymmetrical sec-
Fig. 1. Geological map of the study area. 1, Q; 2, T2b3; 3,
T2b2; 4, axial trace of syncline; 5, axial trace of anticline; 6,
small fold hinge; 7, reversed fault deduced by geophysical
survey and its name; 8, normal fault; 9, reversed fracture
zone; 10, occurrence; 11, formation boundary; 12, landslide
boundary deduced by VSP, SRS and ES surveys.
ondary interlayer folds, which were developed mainly in soft formations in T2b3 containing inter-
bedded soft and hard rocks with different lithologies and formed by relative shear slide between
different rock formations with different lithologies. Following the E-W folds, developed were
near-E-W-trending reversed faults, bedding faults or bedding shear zones etc. Besides these E-W
direction major structures, there also exist near N-S direction gentle folds striking across the
southern flank of the syncline. Associated with these gentle folds, developed were
near-S-N-trending reversed faults, among which Sizibao fault is a representative one. The investi-
gation shows that the fault strikes nearly N-S and dips west at an angle of about 60° and that the
fault triangle plane is shown on Sizibao in the topography. In addition to the faults mentioned

428
SCIENCE IN CHINA (Series D)
Vol. 44
above, in the area also developed were conjugate joint systems, fracture zones and various joints,
as well as many structural indications such as gravitational creep-slippage, structure etc.
2
Choice and layout of geophysical surveys
To investigate and assess Zhaoshuling landslide, five geophysical methods including vertical
seismic profile (VSP), shallow reflection seismic (SRS), electric sounding (ES), soil temperature
(ST) and radioactive gas (RG) surveys, were
designed. The layout of these surveys is illus-
trated in fig. 2.
VSP survey was conducted in boreholes
Yun 19, Yun 21 and Yun 22, with energy
sources being located at points on lines ori-
ented in N-S or E-W directions. Four SRS
survey lines were placed in the vicinity of
borehole Yun 21 and eight ES survey lines, as
well as seven survey lines of ST and RG
methods were placed in the area in a way that
the distribution was approximately uniform
and some of them passed by the three bore-
holes as much as possible, so that these sur-
veys could cooperate well with each other.
The purpose of using VSP is to locate the
distribution of formations and sliding planes
as well as the elastodynamic moduli of rocks
in the vicinity of the three boreholes. The
method acquires the information of rocks and
Fig. 2. Layout of geophysical survey. ⊙, Yun 19: borehole
structures by analyzing down-going (direct)
and its number for VSP survey; S1-S1′, SRS survey line and
and up-going (reflected) wave fields.
Down-going waves, having stronger energy,
can reflect boundaries with significant
changes of elastic properties. Up-going waves,
its number; − ·−·, ES survey line; 1/E1, station num-
ber/line number for ES survey; − − −, ST and RG survey
line; 8/T1, station number/line number for ST and RG survey.
on the other hand, being weaker, can reflect boundaries with small changes of elastic properties.
Besides, transversal and converted waves (PS waves) can reflect nearly vertical fractures in two
directions, parallel to or vertical to the ray path. According to Gardner et al.[2], bulk density (ρ) can
be estimated empirically by the relationship
ρ = aVP1/ 4
,
where VP is the velocity of longitudinal wave and a = 0.31, when the unit of velocity is m/s. A
group of successive elastodynamic moduli of rocks in the vicinity of the borehole, such as, rigidity

No. 5
A LANDSLIDE IN THE NEW SITE OF BADONG COUNTY
429
modulus (G), Young’s modulus (E), bulk modulus (K), shear modulus (μ ) and Posson’s ratio (σ )
etc., can be derived from ρ and the velocities of shear and longitudinal waves[3—5]. It is therefore
possible to locate these geological features such as weathered zones, sliding planes, formations
and fractures etc. properly according to VSP data, thus providing foundations for the stability as-
sessment of landslides.
SRS method, which locates sliding zones and boundaries between formations according to
inphase axes that can be traced continuously and locates faults based on dislocation, stop or skew
of inphase axes, was used to determine the distribution of formations, sliding planes and structures
in the four SRS profiles. In the SRS survey, the offset is 40 m.
ES method can be used to locate the depth to sliding planes and the boundaries between dif-
ferent formations with different conductivities[6—8]. Schlumberger sounding was used, in combi-
nation with the above two methods, to locate the spatial distribution of the landslide body as well
as the formation and structures in and out of it.
ST measurement, which measured temperatures in soils at a depth of 1 m, was used to inves-
tigate the character of seepage flow of groundwater and the degree of water saturation in the land-
slide body so as to provide useful information on assessment of the landslide[6, 8]. When the
groundwater is seeping mainly through the sliding plane, if the survey is conducted in a scorching
summer, the centre part of the landslide body will be characterized by higher temperatures and its
boundary where groundwater seeps out will be characterized by lower temperatures, and if the
survey is conducted in a cold winter, the centre part of the landslide body will be characterized by
lower temperatures and its boundary where groundwater seeps out will be characterized by higher
temperatures. These characteristics are the signs showing an unstable landslide.
The use of RG measurement which measured the radioactive intensity of 210Po in soil sam-
ples was also to provide information for assessing the stability of the landslide. The principle is
that in the stage that rocks are squeezed and creeped, the concentration of radioactive gases in the
air of soil will be enhanced, because the rocks under it are being destroyed[8].
3
Data interpretation
3.1 Determination of landslide body, formations and structures
3.1.1 VSP method. Shown in fig. 3 are the profiles of up-going, down-going and shear waves
in borehole Yun 19 when energy source is located at a point 10 m west of the borehole. It can be
seen that at 19, 45, 63 m in the up-going wave profile, there exist strong reflected events and at 45
m there exist converted events. The inphase axes of these up-going events are disjointed, which
indicates that at these zones rocks are stratified but that there exist fractures among these forma-
tions. In the down-going profile, longitudinal waves are complete, while converted waves incom-
plete, showing the existence of vertical fractures oriented in E-W direction because the source was
located at a point on the line oriented in E-W direction. At the depth of 22—28, 45—62 m, the-
down-going converted events are relatively distinct, which implies that there exist no more frac-

430
SCIENCE IN CHINA (Series D)
Vol. 44
Fig. 3. Up-going wave profile (a), down-going wave profile (b) and shear wave profile (c) in borehole Yun 19.
tures. Beneath 62 m, the inphase axes of down-going events are curved and divergent, which
shows that rocks are broken. According to the regional geology, the vertical fractures in E-W di-
rection are controlled by layers.
A comprehensive comparison of elastodynamic modulus diagrams in the borehole and the
borehole geological section is illustrated in fig. 4. Each elastodynamic modulus diagram is nor-
Fig. 4. Normalized electrodynamic modulus diagrams and borehole geological section in borehole Yun 19. VS, Velocities of S
waves; VP, velocities of P waves. 1, Q; 2, argillaceous limestone, limestone interbedded; 3, argillaceous siltstone, sandstone in-
terbedded.

No. 5
A LANDSLIDE IN THE NEW SITE OF BADONG COUNTY
431
malized with its maximal value. It is easy to see that in all indices, the sensitivity of Posson’s ratio
is the highest. According to the gradual change of all indices, we can deduce that the depth of
weathered zone may reach 20—40 m deep. At the depth of about 10 m and 20 m, all the parame-
ters have apparent changes, which implies that both the two places are probably boundaries of
formations. At the depth of 65 m, all the parameters also have apparent changes, especially Pos-
son’s ratio increases rapidly, which indicates that rocks there are broken. These changes are more
distinct and detailed than those in the profiles of up-going, down-going and shear events. Based on
all the above interpretations, and the borehole geological section we can deduce that at the depth
of about 10 m and 20 m are the boundary of T2b3 and T2b2 and the sliding plane respectively, and
at the depth of 65 m, T2b2 and T2b3 are faulted contact. The fault, which is a reversed one, is
named F1.
The VSP data in borehole Yun 21 and Yun 22 are also interpreted in the same manner. It is
deduced that in borehole Yun 21, the weathered zone is about 25 m thick and the sliding plane is
situated at the depth of about 53 m, while in borehole Yun 22, no sliding plane or the boundary
between T2b2 and T2b3 is discovered, and under the taluvium all rocks are limestone in which the
development of cracks is also determined by VSP survey.
3.1.2 SRS method. The SRS profile of survey line S3-S3′is shown in fig. 5(a). It can be seen
that above 20 m, there is no effective signal
of seismic events, which indicates that the
zone is a weathered one and that there exists
no significant difference of elastic properties
in rocks. At the depth of 30 m, there exists a
reflecting boundary, combined with the inter-
pretation of the VSP data in Borehole Yun 19,
it can be deduced as the sliding plane. At the
depth of 40 m on the side of S3 and at the
depth of 100 m on the side of S3′, there both
exist events whose inphase axes can be traced
continuously. Compared with the interpreta-
tion of the VSP data in borehole Yun 19, both
locations can be deduced as the boundaries
between T2b2 and T2b3. The boundary is
obviously shown in all the four SRS profiles
and is chosen to be the standard boundary in
interpretation. In the middle of the profile,
Fig. 5. SRS profile (a) and deduced lithologic and
obvious dislocation and stop phenomena can
structural geological section (b) for survey line S3-S3′..
be easily seen and the existence of a fault can

432
SCIENCE IN CHINA (Series D)
Vol. 44
be deduced. The fault dips towards east approximately and is just the reversed fault F1 deduced by
the interpretation of the VSP data in borehole Yun 19. In formations T2b2 and T2b3, there are a lot
of inphase axes which can be traced continuously though the continuity is not so good, indicating
that both formations are not unitary. Based on the above interpretation, the deduced lithologic and
structural geologic section is plotted (fig. 5(b)). The deduced lithologic and structural geologic
sections for the other three survey lines are also obtained according to the interpretation of their
SRS profiles. After interpretation of the SRS profile for line S1-S1′, not only the sliding plane and
the boundary between T2b2 and T2b3, but also a reversed fault located on the side of S1′are de-
duced (cf. fig. 1). The fault dips towards west approximately and is named F2.
3.1.3 ES method. Electric sounding curves were interpreted by using the reflectance profile
technique. By interpretation of them, the imitative resistivity for each layer, which is approxi-
mately but not really equal to the intrinsic one, was acquired. The contour sections of the imitative
resistivities for the eight ES survey lines were interpreted based on the interpreted results of VSP
data and compared with the interpreted results of SRS data. In the interpretation, the contour sec-
tions for E4, E3 and E1 survey lines passed by the above mentioned three boreholes were inter-
preted first and then those for the other survey lines.
Shown in fig. 6(a) is the contour section of the imitative resistivities for survey line E4, in
which borehole Yun 19 is at No.9 station. Compared with the interpretation of VSP in the borehole,
it can be seen that the resistivities for weathered limestone is less than 100 Ωm, that for
Fig. 6. Contour section of imitative resistivities (in Ωm) (a) and deduced geological section (b) for survey line E4. 1, Limestone;
2, argillaceous limestone; 3, purplish-red mudstone; 4, soil and fine debris; 5, soil and coarse debris; 6, sliding mass; 7, sliding
between layers; 8, sliding plane.

No. 5
A LANDSLIDE IN THE NEW SITE OF BADONG COUNTY
433
sandstone is about 100—200 Ωm, the sliding plane corresponds to the low resistivity zone with
values less than 100 Ωm and the reversed fault F1 corresponds to the low resistivity zone with
values less than 150 Ωm. By tracing the low resistivity zone, the distribution of the sliding plane is
delineated. The low resistivity zone corresponding to the reversed fault F1 inclines to the east and
the resistivities at its two sides are quite different, the distribution of the fault in the section is de-
termined accordingly. Under stations 15, 16, another reversed fault inclining to the west is also
determined using the same approach, which is the reversed fault F2 determined by interpretation of
SRS profile S1S1′. Besides, the boundaries between taluvium, T2b3 argillaceous limestone, T2b3
limestone and T2b2 sandstone are also located according to the values of imitative resistivities (fig.
6(b)).
The other contour sections of imitative resistivities were also interpreted in the same manner.
After interpretation, the configuration of the landslide body, the distribution of formations and
structures in and out of the landslide body are determined and a three-dimensional geological map
is then plotted (fig. 7).
Fig. 7. Three-dimensional geological map of Zhaoshuling Landslide. 1, Q; 2, argillaceous limestone, limestone interbedded; 3,
argillaceous siltstone, sandstone interbedded; 4, borehole and its number; 5, formation boundary; 6, reversed fault deduced by
integrated geophysical survey; 7, limit of landslide; 8, simply-built highway.
3.1.4 ST measurement. After altitude correction and substraction of the normal temperature,
which is the temperature at a base point in the central part of the landslide, measured ST data were
used to draw the contour map of temperature anomalies (fig. 8). In the study area, soil tempera-
tures are basically steady, with changes being less than about 1.5℃. The fact implies that the

434
SCIENCE IN CHINA (Series D)
Vol. 44
status of groundwater seepage flow in soil has no significant difference within the area, since soil
temperatures are mainly controlled by the amount of seeping discharge and the temperature of
water in the soil. Anomaly I is approximately coincident with the range of the landslide body de-
duced by VSP, SRS and ES surveys, and yet it is of low temperature. As the ST survey was carried
out in a scorching summer, the existence of low temperature anomaly in the limit of the landslide
body indicates that the groundwater does not seep mainly along the sliding plane, but in taluvium,
and that the taluvium in the landslide has certain capacity of gathering groundwater seepage flows.
Fig. 8. Contour map of soil temperature anomalies (in
℃) in study area. 1, Positive contour; 2, negative con-
tour; 3, contour with zero value; 4, borehole and its
number; 5, landslide boundary deduced by VSP, SRS, ES
surveys.
Fig. 9. Contour map of radioactive intensity of 210Po (in
pulse number/hour) in study area. 1, Positive contour; 2,
borehole and its number; 3, landslide boundary deduced by
VSP, SRS, ES, surveys; 4, reversed fault deduced by inte-
grated geophysical survey.
3.1.5 RG measurement. The contour map of radioa ctive I ntensity of 210Po (pulse number per
hour) is shown in fig. 9. It can be seen from the figure that in the range of the landslide body, the
intensity is less than in the surrounding area. The fact indicates that in the landslide body, there
exist no deformation zones or fractures caused by apparent creep in rocks. The higher intensity of
210Po of anomaly II on the east boundary of the landslide is probably related to the reversed fault

No. 5
A LANDSLIDE IN THE NEW SITE OF BADONG COUNTY
435
F2. In the southern part of the area the increase in the radioactive intensity of 210Po is probably
related to the lithology of formations.
3.2 Assessment of the landslide
The integrated geophysical data interpretation shows that the landslide has the following
character:
(a) The landslide body is about 1000 m long in N-S direction and 350 m wide in E-W direc-
tion, with an average thickness of about 35 m.
(b) The configuration of two flanks of the landslide is related to and partly controlled by the
two near-N-S trending reversed faults F1 and F2.
(c) The sliding plane is controlled by lithologies and occurrences of formations.
(d) The sliding plane is mainly developed in argillaceous limestone and distributed along the
boundary of T2b3 and T2b2 and partly across formations, suggesting that the landslide is one that
has slided along the bedding plane of bedrocks and still preserves the original sequence.
There were two reasons for the formation of the landslide. One was the boundary conditions
and the numerous weakness surfaces provided by early stage tectonic movements. The other was
special narrow valley landform in the Three-Gorge Reservior area which was the essential condi-
tions for producing gravity sliding. In the slip process of the landslide, as the overlain rocks lost
stability and began to slip, the underlying rocks close to and between reversed faults F1 and F2
were drawn, thus forming a main deformable zone in deep formations, i.e. the main slipping zone.
However, the following data show that after undergoing a long-period geological action of
millions of years, the ancient landslide is now still in a relatively stable stage:
(a) Though the area is a typical negative landform, the formation lain on the main sliding
plane is thin, owing to the surficial geological actions in a long period. As to the geological situa-
tion at present, because of the unloading, the load on the sliding plane is greatly reduced, which is
favorable to the stability of the landslide.
(b) In the zone of altitude 200—200 m, the main sliding plane is gently dipping so that the
gravitational gliding effect is small. Besides, the landslide body is pressed on both flanks by sur-
rounding rocks, owing to the existence of reversed faults F1 and F2. The distance between the two
faults decreases from south to north, namely it decreases as the altitude diminishes, so that the
clamping effect is large and plays an important role in the stabilization of the landslide body.
(c) Both interpretation of VSP and RG surveys show that the landslide is not in an apparent
creep and press stage.
(d) ST survey shows that the groundwater seepage in the landslide flows mainly in the shal-
low part of the landslide body, not along the main sliding plane, which is helpful to the stability of
the landslide.
To sum up, we consider that the landslide is now still in the stable stage and is an ancient
landslide only.

436
SCIENCE IN CHINA (Series D)
Vol. 44
4
Summary
The application of integrated geophysical survey to investigation and assessment of
Zhaoshuling landslide is successful. The study provides extremely useful information for the com-
plete and successful investigation and assessment of the landslide:
(a) The spatial configuration of Zhaoshuling ancient landslide body, and the spatial distribu-
tion of the formations and structures in and out of it is well determined by integrated use of VSP,
SRS and ER surveys.
(b) The evidence that the ancient landslide is now still in a stable stage is provided by inte-
grated use of VSP, SRS, ES, ST and RG surveys in combination with available geological data.
Based on the data of the detailed geological engineering survey and the integrated geophysi-
cal survey and a thorough study of it, the conclusion that the landslide is now in a stable stage and
the area above 300 m altitude will be still stable even if the water in the Three-Gorge Reservoir is
fully stored and can be properly used as common sites for civil construction is made by the Engi-
neering Geological Party responsible for the investigation of the new site. The civil construction in
the area is already in progress and the reversed faults F1 and F2 have been confirmed during exca-
vation, proving the application and data interpretation of integrated geophysical surveys to be
correct.
Acknowledgements The authors are indebted to the chief engineer of Bureau of Geotechnique of Changjiang Water
Resources Commission, Cui Zhengquan and the academician of the Chinese Academy of Sciences Xu Houze for their support
and help. We would like to express our gratitude to Cao Xianyu, Zhang Chijun, Zhou Dianyou for their helpful suggestions con-
cerning the study. We wish to acknowledge the contributions of Xu Ganglin, Xu Jianhua and Liu Yonghua who performed or
attended the field work. The project is financially supported by Bureau of Geotechnique of Changjing Water Resources Commis-
sion and the study is organized by Hubei Geophysical Society.
References
1. Wu, Y. L., Zhu, Z. F., Wang, J. L. et al., Lithofacies-Paleogeography and Environmental Control of Sedimentary Deposits
of the Early and Middle Triassic in the Upper Yangtze Area (in Chinese), Chongqing: Chongqing Publishing House, 1989,
1—2.
2. Gardner, G. H. F., Gardner, L. W., Gregory, A. R., Formation velocity and density——the diagnostic basics for strati-
graphic traps, Geophysics, 1974, 39: 770.
3. Omnes, G., Logs from P and S vertical seismic profiles, Jour. Petroleum Technology, 1980, 32: 1843.
4. Zhu, G. M., Vertical Seismic Profile (in Chinese), Beijing: The Oil Industry Publishing House, 1988, 343—380.
5. Sheriff, R. E., Encyclopedia Dictionary of Exploration Geophysics, Tulsa, Oklahoma: Soc of Expl. Geophysicists Press,
1991, 99—100.
6. Bogoslovsky, V. A., Ogilvy, A. A., Geophysical methods for the investigation of landslides, Geophysics, 1977, 42: 562.
7. Cummings, D., Use of seismic refraction and electrical resistivity surveys in landslide investigations, Bulletin of the Asso-
ciation of Engineering Geologists, 1988, 25: 459.
8. Ogilvy, A. A., The Foundation of Engineering Geophysics (in Russian), Moscow: Nedra, 1990, 343—378.