Effects of Hydrostatic Pressure on Limit State Design of Reinforced Soil

Wall Internal Stability

Tan, Yean-Chin & Khoo, Chee-Min

G&P Geotechnics Sdn Bhd, Wisma G&P, 39-5 Jalan 3/146, Bandar Tasik Selatan, Kuala Lumpur, Malaysia.
gnp-geo@gnpgroup.com.my

Abstract: Current design of reinforced soil walls in Malaysia generally does not consider action due to water (e.g. hydrostatic pressure,
seepage forces) because free-draining granular backfill material is used for the wall. However, numerous investigations of reinforced
soil wall failures revealed that low quality backfill materials coupled with a build up of pore pressure behind the wall are responsible
for the collapses. This paper demonstrates the importance of hydrostatic pressure on the internal stability of soil reinforcing systems.
From the analysis of a design example, it was observed that when hydrostatic pressure builds up, it increases the tensile forces in the
reinforcements and decreases the soil-reinforcement friction. In view of this, the effects of hydrostatic pressure on reinforced soil wall
internal stability should be carefully assessed in the design analysis.

1 INTRODUCTION failure must be considered, including internal and external stabil-

Reinforced soil (RS) walls have been used in a variety of applica-
tions since their first appearance in France in the early 1960s. Al-
though many reinforced soil walls have been safely constructed
and are performing well to date, there are many areas that need in
depth studies in order to better understand their mechanical be-
haviour under more aggressive and harsh environments. Thus,
the design methodology for reinforced soil structures is in a state
of transition. In addition, reinforced soil wall applications in
aquatic sites such as lakefronts, seawalls and storm water runoff
channels require additional specialised engineering considera-
tions as the current technology is normally used in dry condi-
tions.
Many reinforced soil wall failures are caused by poor drain-
age, leading to the building up of pore water (hydrostatic) pres-
sure, which in turn generates a destabilising force that has not
been considered in the standard design methodologies. As such,
in a high intensity rainfall region like Malaysia and Southeast
Asia, it is advisable to carefully assess the potential build up of
hydrostatic pressure in reinforced soil walls, particularly when
less permeable cohesive frictional fill materials are used in high
walls where water needs a longer time to discharge from the
wall.
This paper focuses on a study of hydrostatic pressure effects
on soil reinforcing systems which is generally not carried out in
the current limit state design computations. A design example is
carried out to demonstrate the effects of hydrostatic pressure on
the internal stability of walls. Findings of the study and recom-
mendations are presented and discussed.
2 CURRENT DESIGN APPROACHES AND PRACTICES
2.1 Limit State Design
The philosophy of limit state design is to ensure that none of the
ultimate limit states (i.e. collapse or major damage) or the ser-
viceability limit states (i.e. deformations in excess of acceptable
limits, other forms of distress or minor damage) occurs during
the service life of the reinforced soil wall. All potential modes of
ity. Ultimate limit states are deemed to be reached when, for a
specific mode of failure, the disturbing forces are equal to or ex-
ceed the restoring forces. The disturbing forces originate from
self weight, groundwater and live loads, whereas the restoring
forces are derived from the soil and the reinforcement. Practice in
soil reinforcement is to design against the ultimate limit state and
check for the serviceability limit state.
Current design approaches of reinforced soil walls (e.g. BS
8006, FHWA, FMOT) generally do not explicitly consider ac-
tions due to water in the backfill materials in the computation of
disturbing forces, as well as restoring forces, due to the use of
high permeability granular material as backfill. However, the
above assumption is not always safe and special consideration
should be given to amend the design equations where it is antici-
pated that building up of hydrostatic pore water pressures will af-
fect the strength of the fill materials due to reduction in effective
stress.
2.2 Backfill Materials
Generally, the selection of backfill material for reinforced soil
walls must satisfy the mechanical criterion of soil-reinforcement
friction. Most of the specifications indicate that the backfill mate-
rial used should be a frictional fill material or cohesive frictional
fill material free from organic or other deleterious material. The
majority of the walls constructed in Malaysia use high permeabil-
ity frictional fill materials (e.g. sand, quarry dust, crusher run).
Sometimes cohesive frictional fill material may be allowed in
certain sections of anchored earth systems with careful design.
For practical reasons, the criterion of friction is commonly re-
placed by criterion concerning the particle size distribution (grad-
ing) of the backfill materials. The predominant factor is the per-
centages in weight of fine particles. Having said that, fines
(material passing a 63 µm test sieve) of 15 % are usually permit-
ted in most of the specifications, thus they do not satisfy the re-
quirements of a good free draining material, as the gravity drain-
age is too slow according to the general guide presented by
Courtesy Moretrench American Corporation (Powers, 1981) (see
Fig. 1).
 Therefore, build up of hydrostatic pressure within the rein- 4 DESIGN EXAMPLE
forced soil mass can occur should backfill material of poor drain-
ing be used or the wall is very high with poor internal drainage Consider a simple reinforced soil wall to retain a 10m high fill as
thus causing water to take a longer time to discharge fully from an example. To conform to the design situation imposed in BS
the wall. 8002: Section 3.2.2.2, the retained surcharge will be 10kPa on
top and behind the reinforced fill. It will be assumed that the pre-
U.S. standard sieve numbers
existing groundwater table was high, so the phreatic surfaces in
both the short term, when water rises faster than discharge capac-
Wells and/or well- ity and in the long term will be taken at the various levels of
points
0.25Hwall, 0.5Hwall, 0.75Hwall and 1.0Hwall from the bottom of wall
Theoretical limit
of gravity drainage
Electro-
osmosis
at the retained side for parametric study. The objective of this
Percent finer by weight

Subaqueous excavation Gravity paper is to set out some simple design calculations considering
` or grout curtain
may be required
drainage
too slow hydrostatic pressure. Fig. 2 shows the design cross-section.
Gravity drainage 10kPa
Range may be extended
by using large sumps
with gravel filters
Strip 1
15% Z Strip 2
Strip 3
Strip 4
Grain size (mm) Strip 5
Reinforced
10m Mechanical
Coarse Me- Fine
Gravel sand dium sand Silt or clay Strip 6
d Height, Hwall Fill Strip 7
Strip 8
Fig. 1 Gradation of soil particles on gravity drainage (Powers, Strip 9
Strip 10
1981). Strip 11
Strip 12
Water
Strip 13 Height, Hwater

3 BEHAVIOUR OF SOIL-REINFORCEMENT L
INTERACTION DUE TO WATER Fig. 2 Design cross-section.

The actions of water, or more generally related to water, modify
the material characteristics as follows:-
(a) the phenomenon of uplift according to Archimedes’ Prin-
ciple
(b) the reduction in soil-reinforcement friction
Table 1 summarizes the engineering properties of the typical
soils adopted in this design example. The internal friction angle
at a density corresponding to the as-compacted state is approxi-
mately 36° and the apparent cohesion is disregarded as the cur-
rent design guideline requires. The wall is assumed to be situated
in a residual soil stratum with an internal friction angle of 30°
and cohesion of 5kPa.

3.1 Phenomenon of Soil Buoyancy Table 1 Engineering properties of the soils.

Material γb (kN/m3) c’ (kN/m2) φp’ (°)
The water that accumulates behind the wall exerts full hydro-
Granular Reinforced
static pressure but the earth effective overburden pressure is re- 20 0 36
fill for RS wall
duced because of the buoyancy effect of water on the weight of
Backfill behind wall
the submerged soil. The addition of disturbing pressure from hy- 18 5 30
(e.g. residual soil)
drostatic origin is directly transferred to the reinforcements.
Therefore, water can add excessive lateral pressure behind a wall
that can ultimately cause the wall to fail even at low heights if Limit equilibrium internal stability analyses were conducted
not properly designed. based on the BS 8006: 1995 design approach to check for rup-
ture, adherence (for reinforced earth systems) and pull-out (for
anchored earth systems) failures. The coherent gravity method
was adopted as the wall is reinforced by inextensible reinforce-
3.2 Variation of Soil-reinforcement Friction Capacity
ment (e.g. steel). For parametric study, an in-house computer
The law of friction mobilization relates to the normal stress ex- spreadsheet which can incorporate hydrostatic pressure in the
erted on the reinforcement. However, the normal stress exerted wall was used. This program was developed based on the BS
on the reinforcement is generally unknown and therefore an ap- 8006 design approach. These analyses were aimed at gaining in-
parent friction coefficient, f* has been introduced by the relation- sights into the effects of hydrostatic pressure on the internal sta-
ship: bility of reinforced soil structure.
Tmax The partial factors were applied to each component of load
f* = (1) based on load combination B in Table 17 of BS 8006, which is
σv'
the most critical. In this combination, the effects due to the
where Tmax is the maximum shear stress which can be mobilized weight of soil behind the wall structure are multiplied by 1.5.
on the reinforcement faces and σv’ is the average effective verti- However, the effects of hydrostatic thrust are not multiplied by a
cal stress resulting from the overburden stress above the consid- load factor (FMOT, 1980). This combination considers the
ered reinforcement. From this equation, it is clear that buoyancy maximum overturning loads together with minimum self mass of
effect on the soil will reduce the effective vertical stress, result- structure and superimposed traffic load. This combination nor-
ing in decreasing soil-reinforcement friction capacity. mally dictates the reinforcement requirement for pull-out resis-
tance and is normally the worst case for sliding along the base.
 For simplicity, a uniform distribution of identical reinforcing
elements length (L) of 0.7Hwall (i.e. 7m) was used throughout the 200mm
height of the wall. For reinforced earth systems, a typical rein-
forcement strip size of 50mm x 4mm available in the Malaysian
market is adopted for the analysis. Corrosion allowance of 200mm
0.5mm per face as recommended by BS 8006 is considered in φ16mm
addition to galvanizing of the steel reinforcement. Both smooth
reinforcement and highly adherent (H.A.) reinforcement were
considered (see Figure 3). 100mm

4mm Fig. 4 Anchor reinforcing element.

Elevation
5 RESULTS AND DISCUSSION
(a)

5.1 Tensile Force

Fig. 5 shows the variation of maximum acting tensile force, Tj to
50mm
be resisted by each layer of reinforcement due to the effect of
hydrostatic pressure. This force is mainly contributed by vertical
loading due to self weights plus any surcharge and bending mo-
Plan view ment caused by external loading acting on the wall based on the
4mm vertical stress estimated by assuming a uniform distribution ac-
cording to Meyerhof’s method and using a reduced width of the
Elevation
strip.
(b)
Fig. 3. (a) Smooth reinforcement (b) Highly adherent (H.A.) rein- 0 20 40 60 80 100 120 140 160 180 200 220 240 260
forcement. 0 0

1 Legend 1
The distribution of the shear stress is assumed constant along Without water
2 2
the reinforcement for simplification. In the case of smooth rein- Hwater = 0.25Hwall
Hwater = 0.5Hwall
forcement (as shown in Fig.3(a)), the coefficient f* is constant all 3
Hwater = 0.75Hwall
3
Depth , Z (m)

over the reinforced fill mass. However, in the case of highly ad-
herent reinforcement (as shown in Fig.3(b)), the effect of dila-
tancy leading to increased soil-reinforcement friction at shallow
depths is given by the following equations as a function of depth
z measured from the level of the mechanical height (FMOT,
4 Hwater = 1.0Hwall 4
5 5
6 6
7 7

1980): 8 8

9 9
For z ≤ zo = 6m 10 10

f* = fo* (1 – z / zo) + tan φ’ (z / zo) (2a) 0 20 40 60 80 100 120 140 160 180 200 220 240 260

Acting Tensile Force, Tj (kN)

For z > zo
f* = tan φ’ (2b)
where fo* = Soil-reinforcement friction coefficient at top of
wall mechanical height (Figure 2.21, FMOT)
z = Depth from top of wall
φ’ = Effective angle of friction for backfill materials
While for anchored earth systems, a typical reinforced block
200mm x 200mm x 100mm thick with a reinforcement bar of
T16 is adopted to check for pull-out failure as shown in Figure 4.
All the reinforcement strips should be connected to the concrete
panel at 0.5m centre to centre spacing horizontally and 0.75m
centre to centre vertically in order to satisfy the initial internal
stability.
Fig. 5 Variation of maximum acting tensile force.
In the event of hydrostatic pressure existing, the effective
overburden stress will reduce due to the buoyant weight of the
soils. However, the increase in total horizontal stress due to hy-
drostatic pressure is more significant as compared to the decrease
in overburden stress. Moreover, the external acting force due to
water behind the reinforced fill will increase the eccentricity of
resultant vertical load at each level of the wall and subsequently
increases the repartition vertical stress exerted on reinforcement.
In this context, the efficiency of wall block size due to hy-
drostatic pressure was studied for typical wall block sizes ranging
from L = 0.4Hwall to 1.0Hwall. The sensitivity of increase in acting
tensile force for each layer of reinforcement with various water
levels is presented in Fig 6. It should be noted that the impact of
the above phenomenon is far more notable in the bottom levels of
reinforcement with increasing water level. In addition, the incre-
ment of tensile force is fairly consistent for wall block size, L ≥
0.7Hwall, and water level, Hwater ≤ 0.5Hwall. Nonetheless, it is criti-
cal for wall block size, L < 0.7Hwall, in which wall toppling or
overturning would be more likely with the presence of water.
 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 heights up to 20m except low wall of less than 6m height, which
100 100 overestimated the additional acting tensile force.
90
Hwater = 0.25Hwall
90
This chart can assist in preliminary assessment of disturbing
Legend tensile force due to hydrostatic pressure. Careful assessment is
Increase in Percentage (%)

80 80
Strip 1
Strip 2
required for safe design of reinforced soil walls. Generally, adop-
70
Strip 3
70 tion of wall block size L < 0.7Hwall should be avoided in view of
60 Strip 4 60 potential overturning problems.
Strip 5
50 Strip 6 50
Strip 7 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
40 40
Strip 8 500 500
30 Strip 9 30
Strip 10 450 Legend 450

Average addition of acting force (%)

20 Strip 11 20 L = 0.4Hwall
Strip 12 400 L = 0.6Hwall 400
10 Strip 13
10
350 L = 0.7Hwall 350
0 0 L = 0.8Hwall
300 L = 1.0Hwall 300
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2
200 200 250 250
Hwater = 0.5Hwall 200 200
180 180
Legend
Increase in Percentage (%)

160 160 150 150

Strip 1
Strip 2
140 140 100 100
Strip 3
120 Strip 4 120 50 50
Strip 5
100 Strip 6 100 0 0
Strip 7
80 80 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
Strip 8
60 Strip 9 60 Hwater / Hwall

40
Strip 10
Strip 11 40
Fig. 7 Average percentage of addition in maximum acting tensile
Strip 12 force.
20 Strip 13
20

0 0
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 5.2 Adherence Capacity of the Reinforcement
300 300
Hwater = 0.75Hwall
For limit state design, adherence should be checked beyond the
275 275
line of maximum tension and compared with the relevant rein-
250 Legend 250
forcement tension at each of these points. Obviously, the buoyant
Increase in Percentage (%)

Strip 1
225 225
Strip 2 soil under water will reduce the normal stress induced to the rein-
200 Strip 3 200
Strip 4 forcement and subsequently reduce the soil-reinforcement fric-
175 175
Strip 5 tion and adherence capacity. The reduction of adherence capacity
150 Strip 6 150
Strip 7
for both smooth reinforcement and highly adherent reinforcement
125 125
Strip 8 are shown in Fig. 8 and 9 respectively.
100 100
75
Strip 9
Strip 10 75
It is observed that the reduction in magnitude is relatively
50
Strip 11
50
constant for each layer of reinforcement for the lower portion of
25
Strip 12
25
a wall. Further analyses were performed for various wall heights
Strip 13
0 0
ranging from 5m to 20m and the average percentage of reduction
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2
in maximum adherence capacity (last bottom reinforcement) due
to water is presented in Fig. 10.
500 500
Hwater = 1.0Hwall
450 450 0 10 20 30 40 50 60 70 80 90 100
Legend
Increase in Percentage (%)

400 Strip 1 400 0 0

Strip 2
350 350 1 Legend 1
Strip 3
Strip 4 Without water
300 300 2 2
Strip 5 Hwater = 0.25Hwall
250 Strip 6 250 3 Hwater = 0.5Hwall 3
Strip 7 Hwater = 0.75Hwall
Depth , Z (m)

200 200 4 4
Strip 8 Hwater = 1.0Hwall
150 Strip 9 150 5 5
Strip 10
100 Strip 11 100
6 6
Strip 12
50 Strip 13
50
7 7
0 0
8 8
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2
9 9
L / Hwall
Fig. 6. Variation of maximum acting tensile force for each layer 10 10
of reinforcement. 0 10 20 30 40 50 60 70 80 90 100

Adherence Capacity, Tj (kN)

Fig. 7 summarizes the average addition in percentage of Fig. 8 Variation of adherence capacity for smooth reinforcement.
maximum acting tensile force of last bottom reinforcement due to
hydrostatic pressure. This chart should be applicable to all wall
 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 10 20 30 40 50 60 70 80 90 100 110 120 130 140

0 0 0 0

1 Legend 1 1 Legend 1
Without water Without water
2 2 2 Hwater = 0.25Hwall 2
Hwater = 0.25Hwall
3 Hwater = 0.5Hwall 3 3 Hwater = 0.5Hwall 3
Hwater = 0.75Hwall Hwater = 0.75Hwall
Depth , Z (m)

Depth , Z (m)
4 Hwater = 1.0Hwall 4 4 Hwater = 1.00Hwall 4

5 5 5 5

6 6 6 6

7 7 7 7

8 8 8 8

9 9 9 9

10 10 10 10

10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 10 20 30 40 50 60 70 80 90 100 110 120 130 140

Adherence Capacity, Tj (kN) Pull-out Capacity, Tj (kN)

Fig. 9 Variation of adherence capacity for highly adherent (H.A.) Fig. 11 Variation of pull-out capacity.
reinforcement.
5.4 Summary of Results of Analyses
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

0 0 Fig. 12 shows the initial resisting capacities for a typical 10m re-
Average reduction in resisting capacity (%)

5 Legend 5 inforced soil wall design while Figures 13 to 16 show the inade-
10
5m high wall
10
quacy in rupture resistance, adherence and pull-out capacities
10m high wall
15m high wall taking into consideration the varying hydrostatic pressure.
15 15
20m high wall
20 20
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150
25 25 0 0

30 30 1 Legend 1
Acting Tensile Force
35 35 2 Rupture Capacity (50mm x 4mm) 2
Adherence Capacity (Smooth)
40 40 3 3
Adherence Capacity (H.A.)
Depth , Z (m)

45 45 Pull-out Capacity
4 4

50 50 5 5
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
6 6
Hwater / Hwall
7 7
Fig. 10 Average percentage of reduction in resisting adherence or
8 8
pull-out capacity.
9 9

Again, it is noted that the average percentage of reduction in 10 10

adherence capacity is constant for wall heights of 6m to 20m. It 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150
is slightly variable for wall heights less than 6m due to the varia- Force (kN)
tion in coefficient of active earth pressure from Ko at the top of Fig. 12 Initial resisting capacities without water.
the wall reducing linearly with depth to a value of Ka at depth of
6m below the top of the wall as adopted in the coherent gravity 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150
method. As revealed from this chart, it could be expected that a 0 0
reduction of adherence capacity due to hydrostatic pressure can 1 Legend 1
be as much as 50%. Acting Tensile Force
Rupture Capacity (50mm x 4mm)
2 2
Adherence Capacity (Smooth)
3 Adherence Capacity (H.A.) 3
5.3 Pull-out Capacity of the Anchor
Depth , Z (m)

Pull-out Capacity
4 4

The pull-out capacity of anchor reinforcing elements to sat- 5 5

isfy local stability is determined from the shaft or loop resistance 6 6

developed by friction beyond the potential failure plane and bear- 7 7
ing resistance at each layer of anchors (Cl. 6.6.4.2.3 BS 8006).
8 8
The effect of hydrostatic pressure is shown in Figure 11. Inadequate
Again, it is observed that the reduction in magnitude is rela- 9 Capacity 9

tively constant for each layer of reinforcement. This is mainly 10 10

due to the constant increment in vertical applied pressure at each 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150

layer of anchors. It was also found that the average percentage of Force (kN)
reduction in maximum pull-out capacity due to water for various Fig. 13. Effects of hydrostatic pressure (Hwater = 0.25Hwall).
wall heights is the same as Figure 10.
 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 From these charts, generally, it was observed that slight in-
0 0 adequacy of rupture capacity at the last bottom reinforcement
1 Legend 1
with the water level (Hwater) approaching 0.25Hwall. On top of
Acting Tensile Force this, inadequacy of adherence capacity (for smooth reinforce-
2 Rupture Capacity (50mm x 4mm) 2
Adherence Capacity (Smooth) ment) starts at about Z = 7.5m when the water level is 0.5Hwall.
3 Adherence Capacity (H.A.) 3
At the water level of 0.75Hwall, both adherence capacity (for both
Depth , Z (m)

4
Pull-out Capacity
4 smooth and highly adherent reinforcement) and pull-out capacity
5 5 are inadequate after Z > 3m and worse if the water level is equal
6 6
to the wall height.
7 7

8
Inadequate 8 6 CONCLUSION
9 Capacity 9

10 10 Based on the design example above, it can be seen that hydro-

0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150
static pressure on a wall can result in multiple negative effects.
Force (kN)
Generally, when hydrostatic pressure builds up, it will increase
Fig. 14 Effects of hydrostatic pressure (Hwater = 0.5Hwall). the tensile force in the reinforcements and decrease the soil-
reinforcement friction capacity. The additional acting tensile
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 force is relatively constant for wall block size, L ≥ 0.7Hwall due to
0 0 the resisting moment of block self weight in preventing overturn-
1 Legend 1
ing, but the acting tensile force increases drastically for wall
Acting Tensile Force block size, L < 0.7Hwall. The reduction in resisting capacity of ei-
2 Rupture Capacity (50mm x 4mm) 2
Adherence Capacity (Smooth) ther adherence or pull-out capacity is in proportion to Hwater/Hwall
3 Adherence Capacity (H.A.) 3 ratio. Therefore, the effects of hydrostatic pressure on reinforced
Depth , Z (m)

4
Pull-out Capacity
4 soil wall internal stability should be carefully assessed during de-
5 5 sign to avoid failure.
6 6

7 7

8
Inadequate
8 REFERENCES
9 Capacity 9

10 10 British Standards Institution. 1994. BS 8002 Code of Practice for

0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 Earth Retaining Structures. BSI, London.
Force (kN) British Standards Institution. 1995. BS 8006 Code of Practice for
Fig. 15. Effects of hydrostatic pressure (Hwater = 0.75Hwall). Strengthened/reinforced Soils and Other Fills. BSI, London.
Federal Highway Administration (FHWA). 2001. Mechanically
0 20 40 60 80 100 120 140 160 180 200 220 240 260 Stabilized Earth Walls And Reinforced Soil Slopes Design &
0 0
Construction Guidelines. Publication No. FHWA-NHI-00-
1 Legend
Acting Tensile Force
1 043. Washington, D.C.
2 Rupture Capacity (50mm x 4mm) 2 French Ministry of Transport (FMOT). 1980. Reinforced Earth
Adherence Capacity (Smooth)
3 Adherence Capacity (H.A.) 3 Structures: Recommendations and Rules of the Art (English
translation). French Ministry of Transport.
Depth , Z (m)

Pull-out Capacity
4 4

5 5 Powers, J. Patrick. 1981. Construction Dewatering: A Guide to

6 6
Theory & Practice. John Wiley & Sons.
7 7

8 8
Inadequate
9 Capacity 9

10 10

0 20 40 60 80 100 120 140 160 180 200 220 240 260

Force (kN)
Fig. 16. Effects of hydrostatic pressure (Hwater = 1.0Hwall).