Geoscience Reference
In-Depth Information
the slope (e.g., initial shear stress distribution; arrangement, thickness, and permeability
of strata), and the characteristics of the ground motions (e.g., amplitude and duration of
shaking). In adverse conditions, the process of void redistribution progressively reduces
the shear resistance of a loosening zone until it is less than the shear resistance required
for stability. The shear resistance in a loosening zone could be locally reduced to zero
if a water film forms, but the average shear resistance over a large area is unlikely to be
zero because water films can dissipate by piping into cracks that develop as the slope
deforms and the geologic interfaces where loosening develops are likely to be irregular
enough to preclude continuous films of water over large areas. The potential for void
redistribution to cause significant slope deformations decreases quickly with increasing
D
R
because a greater D
R
has the combined benefits of reducing the volume of water
expelledbycontractingzonesandincreasingthevolumeofwaterthatcanbeabsorbedby
dilatingzones.Simpleanalysismodels(e.g.,Malvicketal.,2006)andnonlinearcoupled
numerical analyses (e.g., Naesgaard et al., 2005) have illustrated the influence of various
factors under very simple conditions. The reliable application of such analysis methods
tothecomplexconditionsencounteredinthefield,however,arelesscertainandyettobe
adequately calibrated and fullyevaluated.
TheS
r
/
σ
vo
valuesinFigure1.5canalsobeinterpretedasrepresentingsmallfractionsof
thepre-earthquakedrainedstrength.Forexample,assumingtan
φ
≈
0
.
6forloosesands,
the S
r
/
σ
vo
values of 0.03 to 0.1 at
N
1
)
60cs
-
Sr
less than about 10 in Figure 1.5 are about
5 to 17% of the pre-earthquake drained strength (i.e., a strength loss of 83 to 95% due
to liquefaction). For situations where void redistribution is significant, it seems reason-
able to expect that the shear strength will drop to a small fraction of the pre-earthquake
drained shear strength because the impeded pore water seepage allows the soil to shear
atasustainedlowvalueof
(
σ
v
.Thus,theseS
r
/
σ
vo
ratiosarebelievedtoprovideabetter
representationofthepotentialeffectsofvoidredistributionthanareprovidedbythedirect
S
r
correlation in Figure 1.4, while recognizing that neither correlation fully accounts for
the numerous factors that influence void redistributionprocesses.
It is also useful to compare the S
r
/
σ
vo
values in Figure 1.5 to the values expected for
remolded, normally consolidated, slightly plastic silts that are at the transition between
soils that should be analyzed as sands versus soils that should be analyzed using proce-
dures appropriate for clays (Boulanger and Idriss, 2006). Such soils can be encountered
as lenses and layers in hydraulic fills or tailings embankments and have low penetration
resistances. The undrained shear strength ratio of young, normally consolidated, slightly
plastic silts can be about s
u
/
σ
vo
=
22 (Ladd, 1991) and have sensitivities in the range
of2to4,suchthattheremoldedshearstrengthratio,S
ur
/
σ
vo
,canbeabout0.05to0.11.
Remolded strength ratios in this range of values have been reported for slightly plastic
silts and sandy silts by Vasquez-Herrera et al. (1990), Castro and Troncoso (1989), and
Finn et al. (1991). These S
ur
/
σ
vo
values are comparable to the S
r
/
σ
vo
values for the
loose sands, siltysands, and nonplastic sandy siltsrepresented in Figure 1.5.
0
.
The lower design relationship in Figure 1.5 corresponds to conditions where the effects
of void redistribution could be significant. This would include sites with relatively thick
layersofliquefiablesoilsthatareoverlainbylower-permeabilitysoilsthatwouldimpede
