Environmental Engineering Reference
In-Depth Information
N
k
normally varies from around 10 to 20, but may be higher for overconsolidated soils.
Hence the adoption of say N
k
15, as is often done, may introduce an inherent bias of up to
30% or more. These potential errors can and should be reduced by using high quality sam-
pling and laboratory testing to calibrate the in-situ test for the particular soil and conditions.
11.4
ESTIMATION OF PORE PRESSURES AND SELECTION OF
STRENGTHS FOR STEADY STATE, CONSTRUCTION AND
DRAWDOWN CONDITIONS
11.4.1
Steady state seepage condition
11.4.1.1
Steady state pore pressures
Steady state seepage conditions are usually assumed for the assessment of the long term
stability of the downstream slope of the dam. These are usually based on the reservoir being
at Full Supply Level.
In reality, it may take many years to reach steady state conditions in the dam core.
LeBihan and Leroueil (2000) calculate that for typical central core earth and rockfill
dams, full saturation will take a few years for a core saturated permeability of 10
6
m/sec,
several decades for 10
7
m/sec and centuries for 10
8
m/sec. The authors' experience
confirms that large dams may not have reached equilibrium pore pressures 20-30
years after construction, with construction pore pressures dominating in the lower
part of the dam and fully saturated conditions not established in the upper parts. It is
however good practice to design for the steady state conditions, making appropri-
ately conservative assumptions to estimate the pore pressures.
Pore pressures for the steady state seepage condition are estimated by calculating
the flownet for the embankment section either by graphical techniques or more com-
monly now by finite element methods. These techniques are described in detail in
other references, e.g. Cedergren (1967, 1972), Bromhead (1988) and Desai (1975)
and are not covered here.
The following issues are important when calculating the flownets for embankment
dams.
(a)
Zoning of the embankment
. Embankment zoning clearly has a vital role in determin-
ing the pore pressures in the embankment. This is discussed in Chapter 8.
(b)
Anisotropic permeability of embankment earthfill
. Earthfill in dam embankments is
compacted in layers and the action of rolling, possible drying and cracking of the sur-
face of each layer and greater compaction of the upper part of the layers compared to
the lower part will almost invariably lead to the horizontal permeability (k
H
) being
greater than the vertical permeability (k
V
). It would not be unusual for k
H
/k
V
15
and even as high as k
H
/k
V
100. This has a marked effect in pore pressures in an
embankment, particularly if no seepage control measures such as a vertical drain are
incorporated in the design.
Figure 11.12
shows pore pressures for a zoned earthfill
dam for k
H
/k
V
1, 16, 50 and 100 in the core. The permeability of the downstream
zone is 20 times the vertical permeability of the core. It can be seen that pore pressures
on the downstream slope are affected greatly by the permeability ratio. The affect on
pore pressures of using k
H
/k
V
1 are also evident in
Figure 11.13
.
(c)
Foundation permeability
. The permeability of the foundation (compared to that of the
embankment) has an important effect on the seepage flownet. No dam foundation is
“impermeable” and in the majority of cases, even a rock foundation will have a per-
meability greater than that of compacted earthfill, e.g. most rocks have a permeabil-
ity of between 1 and 100 lugeons (10
7
to 10
5
m/sec) compared to compacted clay
earthfill with a k
V
15 and
10
7
to 10
10
m/sec. Soils in a dam foundation (which will usually
