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

net/publication/255585910

Behavior of Sheet Pile Walls at Deep Excavations in Soft Soils Overlying Hard
Rock in Stockholm

Article · January 2008

CITATIONS READS

2 1,057

5 authors, including:

Jianqin Ma Bo Berggren
Chang'an University Berggren Tech AB, Sweden
18 PUBLICATIONS   54 CITATIONS    8 PUBLICATIONS   16 CITATIONS   

SEE PROFILE SEE PROFILE

Håkan Stille Staffan Hintze

KTH Royal Institute of Technology KTH Royal Institute of Technology
81 PUBLICATIONS   1,376 CITATIONS    8 PUBLICATIONS   45 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

HYBRIT Research Program 1, WP 2.3 View project

In the business of building green View project

All content following this page was uploaded by Staffan Hintze on 04 September 2014.

The user has requested enhancement of the downloaded file.

 th
The 12 International Conference of
International Association for Computer Methods and Advances in Geomechanics (IACMAG)
1-6 October, 2008
Goa, India

Behavior of Sheet Pile Walls at Deep Excavations in Soft Soils

Overlying Hard Rock in Stockholm

J.Q. Ma
College of Highway, Chang'an University, Xi’an, China

B.S. Berggren, P.E. Bengtsson

Swedish Geotechnical Institute, Linköping, Sweden

H. Stille, S. Hintze
Royal Institute of Technology, Stockholm, Sweden

Keywords: Sheet pile wall behavior, Deep excavation, Soft soil, Bedrock, Stockholm

ABSTRACT: The deformation of retained soil is of importance for the design of a deep excavation in soft soil. The
influence of bedrock underlying the soils on the behavior of the retaining wall needs proper evaluation. This
contribution shows a case study on deep excavation in soft soil overlying bedrock at the South Link Infrastructure
Project in Stockholm. The behavior of sheet pile walls at the South Link shows that both lateral displacement and
settlement are larger than the magnitude from similar case histories. Exemplified by section 1/840N, the features
of lateral displacement increment are analyzed with monitoring and PLAXIS simulating results. The analysis
results indicate that the beneficial influence of bedrock may be overshadowed by the low strength of soils in which
excavation occurs.

1 Introduction
As society begins to demand greater and more efficient use of underground space, deep excavations are being
carried out to meet the surging need for infrastructure in big cites. During excavation, an in situ wall system is
often constructed to provide stability and to minimize movements of the adjacent ground. In order to ensure a
successful excavation work, the behaviours of the wall and the adjacent ground must be considered during the
design phase. Movements associated with excavations are related to a number of factors including: base stability,
soil type, soils responding to off-loading and pore-water pressure changes, wall type and its system stiffness,
construction procedures and workmanship. Any of these factors may control the overall movement of a supported
excavation (Boone, 2003). It is difficult to make a direct and quantitative analysis of ground movements
associated with excavation support since the total deformation is a complex interaction of the above factors. So, a
combination of analytical and empirical methods is used. Of this subject, Peck (1969), Mana and Clough (1981),
Clough and O’Rourke (1990), and Ou et al. (1993) provided well accepted empirical analysis diagrams on this
subject; Bjerrum et al. (1972), Hashash and Whittle (1996, 2002), Finno and Calvello (2005) gave examples for
numerical analysis. All studies indicate that the deformation feature of the wall-prop system depends not only on
the properties of the excavated soils, but also on the properties of the underlying layers. Mana and Clough (1981)
considered the influence of a firm layer underlying soft soil strata, where deep excavation was conducted, through
a parameter of “depth to an underlying firm layer”. Recently, studies carried out on excavations in multilayered
soil deposits overlying rock stratum (Wong et al.,1997; Yoo, 2001 and Long, 2001) showed that both the lateral
and vertical deformation are less than that in soil deposits without underlying bedrock. This implies that the
design practice, based on traditional empirical result, may be somewhat conservative. However, the studies also
show that the deformation features are of regional characteristic. There are great differences in deformation
magnitude between the cases histories, such as from Korea (Yoo, 2001), Singapore (Wong et al., 1997), Oslo
(Long, 2001) and Taipei (Ou et al., 1993). Therefore, whether the results of previous studies are applicable to in
situ walls in multilayered ground conditions of soils overlying hard rock stratum in the Stockholm region is of
interest. This study aims at understanding the behavior of in situ walls constructed in multilayered ground
conditions of residual soils overlying hard rocks. To meet this goal, measured data from the South Link 10 (SL10)
in Stockholm, are analyzed. Particular emphasis has been placed on the effects of relative excavation depth to
the thickness of the soft soil layer on lateral wall movements. A 2D finite element analysis on a test section is
used to provide insights into the influencing of bedrock on the performance of an anchor back-tied wall system.

3922
2 Project description
SL10 comprises a 460 m long underground structure including 40 meters of rock tunnel, a cut and cover concrete
tunnel and tunnel ramps (Fig. 1). The project was open to use on schedule in 2004.

2.1 Geotechnical conditions

The geotechnical conditions can be simply expressed as soils overlying bedrock (Fig. 1). The soils consist of a fill
deposit overlying a sequence of glacial deposits, mainly composed of soft clay, with about 1 m thick dry crust in
top layer. There occurs thin layer of silt and sand in the soft clay. At the bottom of the soils, there is granular soil
or dense moraine about 1 m thick. In some areas the granular soil is missing. The bedrock is gneiss with a
surface layer mainly unweathered but of varying elevation within the working area (Fig. 1). The maximum depth of
soil strata is about 25 meters in the central part of the working area. Groundwater level is around 2 below surface.
The undrained shear strength of the clay is around 16 kPa, with an increase of 2 kPa/m below the level of +10 m
in the local elevation system. The ground surface level is around +15 m (Fig. 1). The overconsolidation ratio
(OCR) of the clay decreases from approximately 1.4 at surface to 1.0 at a level of +6 m and below (Hintze 2002).

2.2 Sheet pile walls and anchors

Deep excavations were supported by sheet pile walls with back-tied anchors. The structures of the sheet pile
walls are commonly used in Scandinavia. Anchors, with an inclination angle of 30º, 35º or 45º respectively, are
keyed into bedrock. The toe of sheet pile is grouted to prevent groundwater leakage to reduce settlements
(Hintze et al., 2000). The number of anchor varies with the length of the wall since the excavation depth and the
thickness of soils are changeable along the alignment of tunnel and ramp. The depth of excavations varies from
less than 3.0 m to 16 m and 1 to 8 levels of anchors are accordingly installed. The vertical and horizontal intervals
between anchors vary from 1.5 m to 3.0 m and from around 1.0 m to more than 3.0 m, respectively.

2.3 Excavation procedures

Construction activities at the working site are presented as wall installation, excavation and support. The sheet
pile walls are first installed. The wall installation is followed by grouting at the toe of the sheet piles walls.
Excavations inside the sheet pile walls are proceeded in stages, which are alternated by the installation and
preloading of anchors. Excavation stages I through VI and five anchor installations in an alternate way in section
1/840N are exemplified in figure 2.

+30 Soil surface

Relative elevation (m)

+25 Fill +15.0

South Link 10 Excavation I
Dry crust +13.5 +13.0
+20 Groundwater +10.5 Anchor 01 1 +11.5 Excavation II
Tunnel structure Soil surface 1: +10.0
level +9.0 Excavation III
+15 +8.0 02
+7.5
Rock surface +6.5 Excavation IV
+10 +5.5 03
Soft clay +5.0
+4.4 Excavation V
+3.4 04 +2.9
+5 1.0 m
05 Excavation VI
0 Central line of the tunnel
-2.8
-5
Rock surface Granular soil
1/600 1/700 1/800 1/900 1/1000 1/1100 1/1200 1/1300 Grouted part
Bedrock of gnesis

Figure1. Longitudinal section of the SL10 showing Figure 2. Excavation and anchor installation in
structures and geotechnical layers section 1/840N

2.4 Monitoring instruments

The site instrumentations include 11 inclinometers and 11 settlement points (Ma et al., 2006a). The inclinometers
are mostly 2.0 m from the sheet pile wall and the settlement points are 1.0 m from the wall. Inclinometers, with
their toes keyed into bedrock, are vertically installed. The interval of the sensors in an inclinometer is 2.0 m and
its toe is considered totally fixed in displacement calculation. In general, a complete set of instrument readings
was obtained every day when excavation occurred around the test sections.

3923
3 Features of movement
It is a common way to measure the maximum lateral displacement as a value in relation to the excavation depth
(H) since Peck’s well-known article (Peck, 1969), and so is the maximum settlement. The data from SL10 are
shown in figure 3. Figure 3 shows that the maximum lateral displacement (δh,m) is between 0.4%H and 1.7%H,
and is in tendency of around 1%H. The maximum settlement (δv,m) is between 0.6%H and 2.7%H, and is in
tendency of around 1.5%H. The magnitude of the maximum lateral displacement is about 0.4 to 0.8 times that of
the maximum settlement. This is different from the general pattern, which shows that the maximum lateral and
vertical movements are of the same order of magnitude (e.g. Goldberg et al., 1976; Mana and Clough, 1981;
Clough and O’Rourke,1990; Ou, 1993; Hsieh and Ou, 1998; Long, 2001).

To compare to another database, which was categorized into five sets in Long (2001). Here only four sets are
quoted as follows: (1) Propped walls, with the thickness of the soft soil layer (h) less than 0.6H of the excavation
depth (shown as: h<0.6H in figure 3(a)); (2) Propped walls, with the thickness of the soft soil layer (h) larger than
0.6H of the excavation depth and stiff soil at dredged level (shown as: h>0.6H, stiff soil at dredge in figure 3(a));
(3) Propped walls, with the thickness of soft soil layer (h) more than 0.6H of excavation depth and soft soil at
dredge level (shown as: h>H in figure 3); (4) Low factor of safety against base heave (shown as: Low FOS
against base heave in figure 3).

440 1000

400 SL data
900
SL data
h>H
360
h>H 800
Low FOS against base heave
Maximum settlement δv ,m (mm)

320

Maximum settlement δv,m (mm)

h<0.6H
700 h>0.6H,stiff soil at dredge
280 h>0.6H,stiff soil at dredge
600 δh,m/δv,m=0.8
240 Low FOS against base
heave
δh,m/δv,m=0.4
δh,m/δv,m=0.8 500
200 δh,m/δv,m=2.0
δh,m/δv,m=0.4 400
160
δh,m/δv,m=2.0
300
120

80 200

40 100

0 0
0 50 100 150 200 250 300 350 400 0 50 100 150 200 250 300 350 400
Maximum lateral deformation δ h,m (mm) Maximum lateral deformation δ h,m (mm)

(a) (b)
Figure 3. Maximum lateral displacement versus settlement

Figure 3(a) shows that both lateral and vertical movements at SL10 are far larger than those of the data from
propped walls with h<0.6H. Both the displacement from h>0.6H with stiff soil at dredged level and that from
propped walls with h<0.6H are generally less than those from SL10. Figure 3(b) shows that the magnitude of h>H
covers almost all the data from SL10, though a point with a vertical displacement of 1000 mm seems abnormal.
This indicates that the deformation of SL10 is, to some extent, similar to the data of Long (2001) relating to the
situation of “Propped wall h>0.6H-soft soil at dredged level”. The data from the set of low FOS against base
heave is almost in the same range as the data from SL10, which lies between the lines of δh,m/δv,m = 0.8 and
δh,m/δv,m = 0.4. Regarding the geological situation, the data from the SL10 are more comparable with the data set
of h>H than the other sets.

4 Analysis on the feature of lateral displacement of section 1/840N

4.1 Monitoring results

4.1.1 Increment at excavation stages

In section 1/840N, monitoring instruments was put into usage after the anchor 01 was preloaded. For the
excavation stages II through VI, the increment of lateral displacement from their preceding excavation stage is
shown in figure 4. Figures 4 shows that most of the lateral displacement increment takes place in the excavation
II, III and IV, and that there occurs quick increase in lateral displacement at the end of the excavation III.
However, in excavation IV, the increment occurs mainly in the lower part of the test section, about 5.0 m below
the excavation bottom. The increment pattern changes from excavation III to IV. This corresponds to the
decrease tendency of the elevation of the maximum lateral displacement points with excavation depth increasing.

3924
In excavation V and VI, the increment from preceding excavation is small or even negative. In general, the
maximum increment from preceding excavation decreases from excavation II through VI.

4.1.2 Increment with excavation depth

The depth of the maximum lateral displacement increment point in various excavation stages seems increasing
with excavation depth as shown by the correlation line in figure 5. The maximum increment point is about 3 m
below the bottom of excavation in a stage. However, there is no simple linear increasing correlation between
them. From excavation II to III, there is no increase of the depth of the maximum lateral increment point, while
from excavation III to IV, there occurs a significant increase. In excavation V, the increment of the depth of the
maximum lateral increment point is almost the same as the increment of excavation depth. In excavation III and
VI, there is no increment of the depth of the maximum lateral increment point.

4.2 Analysis with PLAXIS

4.2.1 Model description

A 2D Mohr-Coulomb model is used to represent the real three-dimensional problem and the excavation process
is simplified as a symmetrical problem in PLAXIS analysis. The model includes a representative of the five distinct
geotechnical layers, sheet pile wall and anchor system, construction stages and distribution loading on the
surface. The dimension of the model is 101.5 m wide and 25 m high. A 15-node triangular element is chosen to
get an accurate calculation. The total elements in the model are 1472. Standard fixities are for boundary load
condition. Except wall installation, 6 soil excavations, 5 anchor preloading stages and 7 consolidation analysis
stages are presented in the model. Anchor installation is included in its preceding excavation stage respectively,
while the preloading of an anchor is simulated as a separated stage. The wall installation is assumed
“wished in place”, but the shear strength of the retained soil can only be 67% mobilized along the soil-
wall interface.

Excavation depth (m)

15 Excavation II
13 3 5 7 9 11 13
Exacation II (A)
Depth of the maximum lateral increment

3
11
Excavation III 5
9
point from surface (m)

Excavation III (A) 7

7
Elevation (m)

y=x
5 Excavation IV 9

3 Excavation IV (A) 11 y = 1.1157x + 3.1186 Excavation IV

2
1 Excavation V 13 R = 0.8885
-1 15
Excavation V (A)
-3 Excavation II
17
Excavation VI
-25 -10 5 20 35
19 Excavation VI
Increment from preceding excavation (mm) Excavation VI (A)

Figure 4. Increment of lateral displacement from Figure 5. Correlation between the depth of the
preceding excavation maximum increment point and excavation depth

All five distinct geotechnical layers are simplified as horizontal in occurrence. From +15 m to +14 m there is a fill
of loose fine soil, which is underlain by 1.0 thick dry crust. The soft clay is 13.8 m thick. The surface of the
bedrock is at -2.8 m. Groundwater level is 2.0 m below the surface. The water level is updated according to the
monitoring results during excavations. Length of sheet pile wall is 17.5 m long and its toe is totally fixed into rock.
6 2
Its normal stiffness and flexural rigidity are 5×10 kN/m and 151259 kNm /m, respectively. The lateral coefficient
value of k0 is given from 1.38 to 1.0 from surface to the elevation of +6 m. With the landscape and temporary
transportation outside the excavation considered, surface load is given as varying from 30kN/m2 to 10kN/m2 on
the left side of the excavation. Anchors incline at an angle of 45°. The maximum force of anchor rod is 1.0×1015
8
kN and the normal stiffness of the geogrid is 1.0×10 kN/m. The preloads of the anchors are given either
according to monitoring results or deduced from design values with a ratio analogy from the monitored anchors
(Ma et al., 2006b). In the numerical analysis model, the bedrock is considered to be an elastic material extending
to large depth. The friction angle, cohesion and Young's modulus of the bedrock are given as 45º, 200 kN/m2 and
6 2
2×10 kN/m , respectively. The toe of the sheet pile wall is totally fixed into the bedrock.

4.2.2 Comparison between numerical and monitoring results

Numerical analysis results are checked with corresponding monitoring values. For stages 4 through 12, the
numerical analysis results are compared to monitored results in figure 6. Figure 6 shows that the magnitude of
the monitoring and analysis results can be generally considered similar. For the excavations II through VI, the
increment of the lateral displacement from preceding excavation is compared in figure 4. Figures 4 and 6 show
that most of the lateral displacement occurs in excavation II through IV. Figure 7 shows the comparison between
the elevations of the maximum lateral displacement points of the analysis and monitoring results. The difference

3925
between the analysis and monitoring results is not significant except for stage 3, at which anchor 01 is loaded in
the analysis model.

A- St age 4 15 A- St age 5
15
M- St age 4 M- St age 5
12
12
A- St age 6
A- St age 7

El evat i on( m)
El evat i on( m)

M- St age 6 9
9
M- St age 7
A- St age 8
6 6
A- St age 9
M- St age 8

A- St age 10 3 M- St age 9
3

M- St age 10 A- St age 11
0 0
0 50 100 A- St age 12 0 50 100 M- St age 11
Di spl acement ( mm) -3 Di spl acement ( mm)
-3 M- St age 12

(a) (b)

Figure 6. Comparison between numerical analysis (A-) and monitoring (M-) lateral displacements

Figures 4 and 6 also show that the difference between the numerical analysis and monitoring results is less in the
lower part of the test section than in the middle and upper parts. This tendency implies that the influence of the
bedrock on the lateral displacement is significant where excavation approaches the bedrock. On the other hand,
the difference between the numerical analysis and monitoring results is larger in the upper part of the test section
than in the middle and upper parts indicates the influence of the bedrock may be overshadowed by other factors,
such as surface load and anchor preloads.

5 Discussion on the factors of controlling lateral displacement

5.1 Effect of bedrock

The studies on deep excavations in soft clay overlying stiff soil layer or bedrock show that stiff layers (stiff soil or
rock) could have a dominating influence on the displacement nature of a propped wall system (Mana and Clough,
1981; Wong et al., 1997; Yoo, 2001 and Long, 2001). Yoo (2001) made a conclusion regarding deep excavation
in soft soil overlying bedrock in Korea, which is, to some extent, similar to SL10 in terms of geological setting. The
data from Yoo (2001), which includes different wall types, i.e. diaphragm walls, sheet pile walls or H-pile walls,
soil cement walls, and cast-in-place walls, is shown in figure 8 in terms of excavation depth versus δh,m/H, where
δh,m is the maximum lateral displacement and H is depth of excavation. Data relevant to sheet pile walls or H-pile
walls from Long (2001) are also shown in figure 8. Figure 8 shows that all of the data from Korea (Yoo, 2001) are
below the ratio 0.6 of δh,m /H, while the data from SL10 is above this ratio. This implies the displacement feature
of the walls in the Stockholm area is far different from that of Korea.

Figure 8 shows that the maximum lateral displacement magnitude of SL10 is similar to the data from Long (2001)
in stead of the data from Yoo (2001). The similarity between the displacement feature of SL10 and that of the
propped walls with h>H (Fig. 3) shows that the relative thickness of the soft soil layer may have a key influence
on the deformation of the sheet pile wall-anchor system during excavation. Data from SL10 and Long (2001) are
plotted in figure 9 in terms of H/h versus the ratio of maximum lateral deformation to H. Figure 9 shows that the
normalized maximum lateral deformation increases with decreasing H/h. This means that the maximum lateral
deformation magnitude increases with the formation level more above the bedrock or stiff layer.

Figures 8 and 9 together indicate that the influence of the bedrock may be overshadowed by a low ratio of H/h,
provided the deep excavation formation level does not reach the bedrock or the ratio of H/h is less than 1.0 as
shown in SL10. In order to check this deduction, the normalized lateral displacement with excavation stage in
section 1/840N is shown in figure 10 in terms of the ratio of excavation depth to soil thickness. Figure 10 shows
that the normalized lateral displacement increases from excavation II to III, and then decreases almost in a linear
manner with increasing excavation depth. This indicates that the influence of the bedrock on the lateral
displacement increases with the excavation bottom getting nearer to bedrock.

It is noticed that large increment of lateral displacement takes place in excavation II and III, as shown in figure 4
and 6 in section 1/840N. In order to indicate the influence of bedrock on the lateral displacement of the sheet pile

3926
wall retained soil system in the test section, the ratio of the depth from excavation bottom to bedrock (D) and
excavation depth (H) is given in figure 11. In general, the maximum lateral displacement increment from
preceding excavation decreases with the ratio of D/H decreasing. Figure 11 shows that the increment in
excavation V from preceding excavation IV is small and so does in excavation VI. In excavation IV, D is 9.3 m,
and the ratio of D/H is 1.09 (>1.0). At the excavation V, D is 7.2 m and the ratio of D/H is 0.71 (<1.0). The
increment of lateral displacement takes place mainly on the condition of the ratio of D/H lager than 1.0. The
influence of the bedrock becomes significant when D/H is less than 1.0.

14 6 Data from Long (2001)

Elevation of the maximum
displacement point (m)

12 SL data

Maxium Lateral displacement

5
10 Data from Yoo (2001)
4
8

δh,m /H(%)
3
6
4 Monitoring result 2
2 Analysis result 1
0 0
2 3 4 5 6 7 8 9 10 11 12 0 10 20 30 40
Stage
Excavation Depth H (m)

Figure 7. Comparison of the elevation of the

maximum lateral displacement points in stages Figure 8. Variation of δ h ,m /H with Excavation Depth

1.20
0.80

1.00 Data from Long(2001)

0.70
SL10 data Excavation
0.80
H: Excavation depth; 0.60 Excavation V
h: Soil layer thickness
H/h

0.60
H/h

0.50

0.40
0.40

0.20 H: Excavation depth;

0.30
h: Soil layer thickness
Excavation II
0.00
0.20
0.00 1.00 2.00 3.00 4.00 5.00
0.30 0.50 0.70 0.90 1.10 1.30
Maximum lateral displacement δhmax/Excavation depth H (%) Maximum lateral displacement/Excavation depth (%)

Figure 9. H/h versus normalized maximum lateral Figure 10. Variation of lateral displacement with the ratio
displacement of excavation depth to soil thickness in section 1/840N

5.2 Effect of System Stiffness

The effect of system stiffness for a propped retaining wall system on the maximum lateral deformation has been a
subject for deep excavation project in soft soil. Since Rowe (1952) studied this problem with structural
engineering principles, different definitions for wall system stiffness were developed by others, such as Goldberg
et al. (1976), Mana and Clough (1981) and Clough and O’Rourke (1990). For a sheet pile wall-anchor system, the
4
definition of system stiffness as ks = EI/(γw h avg ) is well accepted, where γw is the unit weight of water and havg is
the average vertical spacing of the anchors, E is the Young’s modulus of the wall and I is the structural internal
moment of inertia per unit length of the wall. Yoo (2001) shows that there is a boundary line and most of the data
are plotted into the area below this line (Fig. 12). The boundary line follows an exponential decay function and is
described as δh,m /H (%)= 0.5exp(-0.007ks). The maximum lateral deformation of a sheet pile wall system
decrease with its system stiffness ks increasing.

Figure 12 demonstrates the variation of maximum lateral wall movements δh,m /H (%) with the system stiffness ks
for the measured data from SL10 together with the data from Yoo (2001). It is evident that the data from SL10 is
far above the boundary line δh,m/H(%) = 0.5exp(-0.007ks) and that the deformation magnitude of SL10 is different
from the Korea cases. There is no clear relationship, which shows that the relative lateral deformation δh,m/H (%)
varies strongly depending on the system stiffness ks.

Further inspection of Figure 12, reveals that the scatter in the data for a given system stiffness is greater for the
data from SL10 than for the data from Korea cases. This implies that the lateral wall movements in SL10 are
more prone to site conditions such as ground water, retained soil stiffness, and workmanship. A possible
explanation for the lack of dependency on system stiffness is that, once sufficient stiffness is available, movement
is determined by the magnitude of excavation base heave, preloading (Ma et al., 2006b) and also if these soils

3927
have significant capacity to ‘‘arch’’ or ‘‘self support’’ (Long, 2001).

Maximum lateral displacement δ h,m /H (%)

40 Data from Yoo (2001)
Maximum increment from
preceding excavation (mm
Boundary line(Yoo 2001)
1.6 SL10 data
30
Excavation II
1.2
20 Excavation IV
0.8
10 Excavation V
0.4
0
0.0 1.0 2.0 3.0 4.0 5.0 0
10 100 1000 10000
Depth to bedrock (D)/Depth of excavation (H)
System stiffness (k s )

Figure 11. Variation of maximum lateral

displacement increment from preceding excavation Figure 12. Variation of δh,m/H with stiffness of support
with ratio of D/H in section 1/840N system ks

The trend in the data in figure 8 seems to be decreasing lateral movement with increasing excavation depth (H),
suggesting a greater degree of control on the lateral displacement by soft soil conditions at dredge level. Figure
12 shows that the system stiffness should play a key role influencing the feature of sheet pile wall system, once
certain wall stiffness is conditioned. For section 1/840N, the system stiffness increases generally and the lateral
displacement increment from preceding excavation stage decreases from excavation II through VI. Since the
larger lateral displacement increment in excavation II and III (Fig.s 4 and 11) and the largest lateral displacement
takes place in excavation III, it may be an effective way to lessen the lateral displacement through decreasing the
distance of the three anchor levels.

For a wall with less stiffness the maximum wall bending moment will be effectively reduced (Potts and Day,
1990). They showed that if the increased movements, associated with more flexible walls, can be accommodated
or reduced by extra propping, such walls can provide economic and viable solutions.

5.3 The influence of soil strength

It is well accepted that the deformation feature of the wall-prop system varies depending heavily on the properties
of the soil layer. The comparison (Ma et al., 2006a) shows that the soil strength of the cases from Yoo (2001) is
higher than that of both SL10 and the cases from Long (2001). Compared to the cases from Yoo (2001), the
larger lateral deformation in SL10 should, at least, partially be due to the lower soil strength. The comparable
lateral displacement and soil strength between the case of h>H from Long (2001) and SL10 support this
deduction. More recently, Moormann (2004) concluded that ground conditions were one of the identified most
influential parameters. The maximum lateral and vertical movements due to deep excavation in cohesive soils of
very soft to soft consistency, with undrained shear strength cu<75kN/m2, may be more than 1% of H, the
excavation depth, and the ratio of the maximum surface settlement to the maximum lateral displacement may be
as high as 2.0. The magnitude of the movement in SL10 falls in this varying range.

6 Conclusion
(1) The lateral wall deformations for sheet pile walls in SL10 are larger than for walls, which are also keyed into
rock but with formation level in rock layer, while they are similar to walls, which are embedded into a stiff soil layer
but with deep excavation totally in soft soils.

(2) The magnitude of the lateral displacement in SL10 varies in a wide range from 0.4 %H to 1.7%H (excavation
depth), which is about 40% to 80% the magnitude of the maximum surface settlement. It is therefore suggested
that the beneficial effect of the presence of rock stratum should be carefully considered for estimating lateral
movements of the walls in soft soils overlying bedrock, especially where the soft soil strength is as low as that in
SL10.

(3) The feature of the lateral displacement from section 1/840N shows that the lateral displacement mainly takes
place before the end of the excavation IV, and the increment of the lateral displacement from its preceding
excavation stage is uneven through excavation procedures. The increment of lateral displacement in excavation II
and III are far larger than that in other stages, though the increment in excavation I is not completely measured.

3928
 (4) Of the parameters effecting the lateral deformation of the walls, the properties of the retained soil and the ratio
of excavation depth (H) to the total thickness of the soft soil layers (h) are the most prominent factors if the
retaining capacity of the wall is enough. The influence of the bedrock is shadowed by low strength of t
he soils, where depth of the excavation bottom to bedrock (D) is larger than the excavation depth (H), i.e,
D/H>1.0.

(5) There is no clear evidence that the lateral displacement can be effectively reduced merely through increasing
the wall system stiffness ks (Clough and O’Rourke 1989) in SL10. However, the lateral displacement increasing
feature in section 1/840N implies that it is may be an effective approach of decreasing the vertical space of the
anchor levels in the upper part of a section.

7 References
Bjerrum L., Clausen C.J. and Duncan J.M. 1972. Earth pressures on flexible structures (a state-of-the-art report), Proc.5th
European Conf. on Soil Mechanics and Foundation Engineering, 2, Madrid, Spain.
Boone S. J. 2003. Design of deep excavations in urban environments, A thesis submitted in conformity with the requirements
for the degree of Doctor of Philosophy, Graduate Department of Civil Engineering, University of Toronto.
Clough G. W. and O'Rourke T. D. 1990. Construction induced movements of in situ walls. Proceedings on Conf. on Design and
Performance of Earth Retaining Structures, ASCE, Geotechnical Special Publication No. 25, 439-470.
Finno R. J. and Calvello M. 2005. Supported Excavations: the Observational Method and Inverse Modeling, J. Geotech. and
Geoenvir. Engrg., Vol.131, 826-836.
Goldberg D. T. Jaworski W. E. and Gordon M. D. 1976. Lateral support systems and underpinning, construction methods.
Rep.FHWA-RD-75-128, 129 and 130, Federal Highway Administration, Washington, D.C.
Hashash Y. M. A. and Whittle A. J. 1996. Ground movement prediction for deep excavations in soft clay, J. Geotech. Engrg.,
122(6), 474-486.
Hashash Y. M. A. and Whittle A. J. 2002. Mechanisms of load transfer and arching for braced excavations in clay, J. Geotech.
and Geoenvir. Engrg., ASCE, 128(3), 187-197.
Hintze S. 2002. Prediction and impact on nearby structures during deep excavation for the Southern Link Road Construction.
Proc. of 2nd International Conference on Soil Structure Interaction in Urban Civil Engineering, Zurich, Switzerland.
Hintze S., Ekenberg M. and Holmberg G. 2000. Southern Link Road Construction: Foundation and Temporary Constructions.
Proc. 16th IABSE. Switzerland.
Hsieh P.-G.and Ou C.Y. 1998. Shape of ground surface settlement profiles caused by excavation, Can. Geotech. J., Ottawa,
35: 1004–1017.
Long M. 2001. Database for Retaining Wall and Ground Movements due to Deep Excavations, J. Geotech. and Geoenvir.
Engrg., Vol.127, 203-224.
Ma J.Q., Berggren B.S., Bengtsson P.E., Stille H. and Hintze S. 2006a. Apparent earth pressure of soft soils overlying hard
bedrock at South Link in Stockholm. Proc. 4th international conference on soft soil engineering, Vancouver, Canada, 4-6
October 2006, 299-307.
Ma J.Q., Berggren B.S., Bengtsson P.E., Stille H. and Hintze S. 2006b. Back analysis on a deep excavation in Stockholm with
finite element method. Proc. 6th European Conference on Numerical Methods in Geotechnical Engineering, 423-429.
Mana A. I. and Clough G. W. 1981. Prediction of movements for braced cuts in clay, J. Geotech. Engrg. Div., ASCE, 107(6),
759–777.
Moormann C. 2004. Analysis of Wall and Ground Movements due to Deep Excavations in Soft Soil Based on a New Worldwide
Database, Soils and Foundations, Vol.44 (1), 87-98.
Ou C. Y., Hsieh P. O. and Chiou D. C. 1993. Characteristics of ground surface settlement during excavation, Can. Geotech. J.,
Ottawa, 30(5), 758–767.
Peck R. B. 1969. Deep excavations and tunneling in soft ground. Proc., 7th Int. Conf. on Soil Mech. and Found. Engrg., State-
of-the-Art Rep., State-of-the-Art Vol., 225–290.
Potts D.M. and Day R.A. 1990. The use of sheet pile retaining walls for deep excavations in stiff clay. Proc. Inst. Civ. Engineers,
Part 1, Vol 88, 899-927.
Wong I.H., Poh T.Y.and Chuah H.L. 1997. Performance of Excavations for Depressed Expressway in Singapore, J. Geotech.
and Geoenvir. Engrg., Vol.123, 617-625.
Yoo C.S. 2001. Behavior of braced and anchored walls in soils overlying rock, J. Geotech. and Geoenvir. Engrg., Vol. 127, 225-
233.

3929

View publication stats