Environmental Engineering Reference
In-Depth Information
The authors have on occasions resorted to, or have seen others resort to, reliance on
three-dimensional effects to show that a factor of safety is adequate. In most cases this has
been done simply by averaging the factors of safety of sections taken adjacent to each
other. Usually, if the situation is so marginally stable, it is in reality in need of some reme-
dial action to improve the factor of safety.
11.2.4
Shear strength of partially saturated soils
In recent years, there has been growing interest in the originally proposed effective stress
approach for the determination of the shear strength of unsaturated soils (e.g. Fleureau
et al., 1995; Oberg and Sallfors, 1995; Bolzon et al., 1996; Khalili and Khabbaz, 1996).
In this approach the shear strength is determined on the basis of the effective strength
parameters c
and
and a single stress variable defined as:
ss
(
u)
(uu)
′
(11.3)
a
a
w
is the effective stress and
is the effective stress parameter, which has a value
of 1 for saturated soils and a value of 0 for dry soils. The advantage of the effective stress
approach is that the change in the shear strength with changes in total stress, pore water
pressure and pore air pressure is related to a single stress variable. As a result a complete
characterisation of the soil strength requires matching of a single stress history rather than
two or three independent stress variables. Furthermore, the approach requires very lim-
ited testing of soils in an unsaturated state. A major difficulty of the effective stress
approach has been the determination of
.
Khalili and Khabbaz (1998) used experimental data to demonstrate that
in which
may be
determined from the using the equation
0.55
(
uu
uu
)
a
w
x
(11.4)
(
)
a
wb
where (
u
a
u
w
)
b
is the air entry value and
uu
uu
a
w
is termed the suction ratio.
(11.5)
(
)
a
wb
The air entry value, also called the bubbling pressure, corresponds to the matrix suction
above which air penetrates into the soil pores and can be measured using the filter paper tech-
nique (ASTM, 1992; Swarbrick, 1992) or pressure plate technique (ASTM, 1968, 1992).
This approach was further assessed by Geiser (2000) who demonstrated an even
stronger correlation between
and the suction ratio than Khalili and Khabbaz (1998) had
with their data.
In practice most stability analyses carried out ignore the effects of partial saturation, but
in so doing, the factor of safety, particularly for the upper part of the dam, may be signifi-
cantly underestimated. Hence either an approach such as that of Khalili and Khabbaz
(1998) should be used to estimate the strength above the full supply level (or above the
phreatic surface) or alternatively undrained strength estimated for existing dams by in-situ
testing. However allowance should also be made for potential cracking or softening in areas
affected by cracking as discussed in Section 11.3.2.
