Professional Documents
Culture Documents
Notes On Topic 2 - Slope Stability Notes On Topic 2 - Slope Stability
Notes On Topic 2 - Slope Stability Notes On Topic 2 - Slope Stability
2.1 Introduction
Landslides, slips, slumps, mudflows, rockfalls - these are just some of the
terms used to describe movements of soils and rocks under the influence of
gravity. These movements can at best be inconvenient, but in many parts of
the world slope instability is widely recognised as an ever-present danger
and the consequences can often be disastrous in terms of economic cost and
loss of life.
The topic of slope stability is covered in Knappett and Craig (KC), section
12.3. The main emphasis of KC, and also of the course, is to explain how
slope stability is analysed using limit equilibrium methods. Additional
information and case studies will be presented to highlight the main factors
which affect slope stability.
The main classes of slope failure - falls, slides and flows - are shown in the
figure below.
1. Falls are characterised by movement away from existing discontinuities,
such as joints and fissures, often assisted by water or ice pressure. Slides
are where the mass remains intact while sliding along a definite failure
surface. The two basic types of slides are;
2. Translational slides, which involve linear movement of soil blocks or a soil
layer lying near to the sloping surface. These movements are usually
fairly shallow and parallel to the surface.
3. Rotational slides, where the movement occurs along a curved shear
surface in such a way that the slipping mass slumps down near the top of
the slope and bulges up near the toe. These movements are
characteristic of homogeneous soft rocks or cohesive soils.
4. Flows, the slipping mass is internally disrupted and moves partially or
wholly as a fluid. Flows often occur in weak saturated soils when the
pore pressure has increased sufficiently to produce a general loss of
shear strength with no true shear surface developed.
When designing a slope which is to be built, you must identify the critical slip
surface i.e. the surface along which sliding is most likely to occur, and
determine the factor of safety against sliding along that surface.
The destabilising forces are caused by gravity i.e. the weight of the soil
W.
The resisting forces are due to the shear strength of the soil or rock i.e.
shear stress developed (or mobilised) along slip surface τd.
The factor of safety Fs is the ratio of the shear stress developed along the slip
surface to maintain stability τd to the available shear strength of the soil τf.
The Ultimate Limit State method adopted in EC7 uses partial factors on
actions and resistances, but the principles are similar.
If the actions and resistances are factored in accordance with EC7:DA1, then
a factor of safety Fs = 1.0 would be acceptable for design. Under EC7, this
number is sometimes called the Stability Ratio, but we will use the
traditional term Factor of Safety, as does Knappett and Craig (KC).
MR
Fs = (see KC, equation 12.15)
MA
For homogeneous soil conditions, charts have been produced which enable
the factor of safety (i.e. for the critical slip surface) to be found quickly.
For total stresses, cu & φu, use the Taylor and Janbu charts.
For effective stresses, c′ & φ′, use the Bishop and Morgenstern charts.
Groundwater
Changes in the groundwater conditions is the most important single factor in
triggering a slope failure, which usually occur during periods of heavy
rainfall.
Rising water table increases pore water pressures, reducing effective
stresses.
Seepage forces increase.
The unit weight of the soil increases as the degree of saturation
increases.
Saturation produces softening and swelling of clays.
Therefore, drainage is crucial to slope stability.
Toe removal
Quarrying or mining.
River or coastal erosion.
Surcharge loading
e.g. Fill placed for a new road.
Strength reduction
Produced by weathering.
Slow creep causes restructuring of the soil.
Slow processes eventually reach critical points.
During the course, reference will be made to case studies which illustrate
some of the factors described above (see references in section 2.11).
In KC 12.3 (p.473) you can read about the Holbeck Hall Hotel landslide,
Scarborough, which occurred in 1993, and which is well known in the UK.
Another well publicised series of landslides are those which have
affected the Scottish road network in recent years, particularly the A83
Rest-and-be-Thankful since 2004 (see Figure 2.1 below). Also shown
below are figures from the Scottish Road Landslide Study (2005) which
illustrates this type of debris flow.
Figure 2.1. Scottish debris flows landslides (Scottish Road Landslide Study, 2005)
This is a useful technique for analysing shallow translational slides where the
sliding surface is very long and parallel to the ground surface, as shown in
Figure 2.2.
This type of analysis is often called an infinite slope. KC (p.483) derives two
separate equations for the factor of safety, for effective stresses (eq. 12.26b)
and total stresses (eq. 12.26c). However, this is a rather complicated
approach. A single equation is often presented which can be used for both
total and effective stresses, and this will be derived in class. The starting
point for the derivation is the figure shown below.
If the water table is at an elevation hw above the slip surface, the pore water
pressure u can be expressed as;
u = γwhwcos2β
If you compare this equation with the two equations in KC, you can show
that they are identical if you make the correct substitutions for the shear
strength parameters.
For total stresses, set c′ = cu , and φ′ = φu = 0.
For effective stresses, set c′ = 0 , and φ′ = φ′.
It can be seen that for effective stresses, when c′ = 0 (e.g. for sands), the
above expression becomes;
tan φ ′ γ w hw
Fs = (1 − ru ) where; ru =
tan β γH
For slopes with no ground water pressures (ru = 0), the limiting slope
angle will be φ′.
When the water table is at ground level (ru ≅ 0.5), the factor of safety will
be approximately halved (tanβ = 0.5tanφ).
ru is known as the pore pressure ratio, and it is a useful way of
generalising the water table position when you are designing a slope,
because you don’t always know where the water table will be located. It
is also used in the Bishop and Morgenstern charts.
Rotation slips, either circular or non-circular, are common in clay soils. Recall
our previous discussion on infinite slopes. When the value of Hcr (H for Fs = 1)
approaches the height of the slope, the slope is considered to be finite.
The methods for analysing a finite slope are shown below. Note that when
you are designing a new slope, you need to assume the location and shape
of the potential slip surface.
However, we can use our knowledge of slope stability to help locate the
position of the critical slip surface.
Occurs where there is a relatively weak layer in the upper part of the
slope.
2. Toe failure
The most common location when the slope is relatively steep or where
soil beneath the toe is strong.
3. Base failure
This is the simplest type of circular slip analysis, and was the first to be
carried out, and also led to the development of Taylor’s chart. It is relatively
simple because the soil shear strength parameters are cu and φu, and the
stability of the slope is not affected by the water table position.
Furthermore, for the short term (i.e. undrained) stability of slopes in
saturated clays, φu = 0, which simplifies the analysis even further.
The φu = 0 method
Note that these coefficients can only be applied to slopes in saturated clay
under undrained conditions i.e. short term stability (cu , φu).
Taylor’s chart is shown in Figure 2.4. See also KC Figure 12.9, which shows an
extended version for soils with cu increasing linearly with depth (the chart
has been extended and improved many times over the years).
For a slope of height H, the stability number Ns for the slip surface along
which the factor of safety is a minimum is;
cu
Ns =
FsγH
You can also see that the results from the chart are consistent with our
discussion in section 2.6 concerning the position of the critical slip surface.
For β > 53° and φu > 3° , the critical slip circle is always a toe circle i.e. for
soils with a relatively high shear strength.
For β < 53° and φu < 3° , the critical slip circle may be a toe, slope or base
circle depending on the depth to the firm stratum i.e. base circles are
more likely for soils with a relatively low shear strength.
This allows us to take moments about the slip circle centre o for each
slice and then sum for each slice.
However, you can see that there are additional forces acting on each
slice (E1 and E2, X1 and X2) which are known as the inter-slice forces.
Therefore the problem is statically indeterminate and assumptions must
be made regarding the inter-slice forces to obtain a solution.
General solution
Taking moments about the centre of the slip circle gives the following
solution.
Fellenius Solution
The solution assumes that the resultant of the inter-slice forces is zero
for each slice, to give the following solution.
You can see that the factor of safety Fs appears on both sides of the
equation. However, it can be solved using a trial and error procedure.
The method is laborious to solve by hand, but it is ideal for computer
programs.
The pore water pressure can be related to the total vertical pressure by
the dimensionless pore pressure ratio, ru.
u
ru =
γh
Therefore in Bishop’s equation, (W – ub) becomes W(1 – ru)
As mentioned previously, using ru is a good way of generalising the water
table position, and was used by Bishop to develop a series of charts for
effective stress slope stability analysis (these are not covered in the
course).
Bishop and Morgenstern’s charts (1960) are discussed further in KC
p.481.
However, the accuracy of Bishop’s method has not prevented many other
solutions being proposed, as you will see when you use the slope stability
software later in the course.
Non-circular slips
Notwithstanding the comment made above, there is also a need to be able
to analyse non-circular slip surfaces. This is due to the complexity of ground
conditions in many cases producing a non-homogeneous ground profile,
particularly when you excavate to form a cutting. Again, many solutions have
been proposed. These are based on the solutions by Morgenstern and Price
(1967) and Janbu (1973), which are method of slices capable of analysing slip
surfaces of any shape. For example, the shape shown below.
Some of the effects which need to be taken into account when carrying out
hand calculations are outlined below. Remember that the computer will
carry out these procedures automatically.
For hand calculation, calculate the soil weight (of each slice) using the
effective unit weight γ′ below groundwater level and the bulk unit weight
γ above groundwater level.
In this situation, the factor of safety increases as the water level rises,
and decreases as the water level falls.
The changes in pore water pressure and factor of safety over time for slopes
in clay soils are shown in Figure 2.7 for a cutting (excavation) and an
embankment.
Figure 2.7. Pore pressure and factor of safety changes over time for (a) a cutting, (b) an
embankment (KC Figure 12.14).
Cuttings (excavations)
During excavation of the cutting, the pore water pressure change ∆u at
any point P is negative.
After the end of construction tc, pore water will flow towards the slope
and the water table will draw down, and the pore water pressure will
gradually increase to a steady value, uf.
The time taken for the change in pore water pressures to dissipate
depends on the permeability of the soil.
Only when the uf condition is reached will conditions be fully drained and
an effective stress analysis will be appropriate.
Since the porewater pressure increases over time from the end of
construction, there will be a reduction in the effective stresses and hence
the shear strength over time.
Therefore the factor of safety will be lower in the long term and an
effective stress analysis should be used with shear strength parameters
c′, φ′.
Embankments
During construction, the pore water pressure change ∆u at any point P is
positive.
As the embankment is built, the pore water pressure in the soil will
increase. If the construction period is relatively short, no significant
dissipation will occur during the construction phase.
After the end of construction tc, pore water will flow away from the
embankment and the pore water pressure will gradually decrease to a
steady value, uf.
Therefore the factor of safety will be lower at the end of construction
and a total stress analysis should be used with shear strength parameters
cu, φu.
Embankment Dams
An embankment dam is an earth structure which usually consists of a clay
core of low permeability which is supported by shoulders composed of other
suitable materials such as rockfill. It is often used in preference to a concrete
The critical slip surface is the slip surface which gives the lowest factor of
safety. This is the slip surface along which the slope is most likely to fail and
therefore it needs to be found. You may need to analyse hundreds of
potential slip surfaces, so it is best and more usual to use computer
software.
The computer software packages available all work along similar principles:
1. The slope geometry is drawn, or inputted as a set of coordinates.
2. A set of soil properties are ascribed to each soil layer or zone.
3. The water table or pore water pressure distribution is inputted.
4. A solution method is chosen (e.g. Bishop Simplified, Morgenstern and
Price).
5. An optimisation procedure, or routine, is chosen to search for the critical
slip surface.
6. The routine is implemented and the lowest factor of safety is reported,
plus other information if requested.
Figure 2.8. Slope stability analysis using grid and radius (from GeoStudio – Slope/w).
Textbook
Knappett, J.A. & Craig, R.F. (2012). Craig’s Soil Mechanics, 8th Edn., Spon
Press, ISBN 0-415-32703-2. [Recommended textbook]
Case Studies
Balasingam, M., & Sithampara, P. (2008). Teton dam, USA: uncovering the
crucial aspect of its failure, Proceedings of ICE, Civil Engineering, 161, pp.
35-40.
Barnyard, J.K., Coxon, R.E. & Johnston, T.A. (1993). Carsington reservoir –
reconstruction of the dam, Proceedings of ICE, Civil Engineering, 92, pp.
106-115.
Bromhead, E.N. (1997). The treatment of landslides. Proceedings of ICE,
Geotechnical Engineering, 125, pp. 85-96.
Winter, M.G. et al. (2005). Scottish Road Network Landslides Study, Scottish
Executive. [Download at www.scotland.gov.uk]