Geoscience Reference
In-Depth Information
0
O F.E. Method
0.2
“Shear Slice”
(range for all data)
0.4
y/h
0.6
Average of
all data
0.8
1.0
0
0.2
0.4
0.6
0.8
1.0
k
max /
PGA/g
Fig. 14.6. Estimatingseismic coefficient as afunction of the peak acceleration at the
crest and thedepth of sliding(Makdisi and Seed, 1978)
in estimating the crest
PGA
is high. Even with advanced analyses, estimating the crest
PGA
is difficult, and the need to perform any level of dynamic analysis to estimate the
crest
PGA
conflicts with the intent of a simplified method that should not require more
advanced analysis.
Moreover,theboundsshownontheMakdisiandSeed(1978)plotof
k
max
/
h
(Figure14.6)arenottrueupperorlowerbounds.Stiffearthstructuresundergoingground
motionswithmeanperiodsnearthedegradedperiodoftheearthstructurecanhave
k
max
values exceeding 50% of the crest
PGA
for the base sliding case (i.e.,
y
PGA
vs.
y
/
/
=
.
0), and
flexibleearthstructuresundergoinggroundmotionswithlowmeanperiodscanhave
k
max
values lessthan 20% ofthe crest
PGA
for basesliding.
h
1
When typically used in practice, the final step is to estimate seismic displacement as a
function of the ratio of
k
y
/
k
max
and earthquake magnitude. Again the range shown in
Figure 14.7 does not constitute the true upper and lower bounds of the possible seismic
displacement, as only a limited number of earth structures were analyzed with a very
limitednumberofinputgroundmotions.AsrecommendedbyMakdisiandSeed(1978):
“
It must be noted that the design curves presented are based on averages of a range of
results that exhibit some degree of scatter and are derived from a limited number of cases.
These curves should be updated and refined as analytical results for more embankments
are obtained.
” Similar to how the Seed and Idriss (1971) simplified liquefaction trigger-
ing procedure was updated through Seed et al. (1985) and then Youd et al. (2001), it is
time toupdate and move beyond the Makdisi and Seed (1978) design curves.
