Limitations of simplified methods for estimating seismic settlements 12d

of 19
Limitations of Simplified Methods
for Estimating Seismic Settlements
by Robert Pyke Ph.D., G.E.
Updated October 18, 2020
Background
Simplified methods for evaluating both liquefaction and settlement under earthquake
loading have been widely used for some years without much comment on their
limitations, but now Boulanger et al. (2016) and Pyke and North (2019) have spelt out the
reasons that they are generally quite conservative. Pyke (2019) provided a case history
involving Lum Elementary School in Alameda CA, in which excessive conservatism led to
particularly adverse social impacts. Crawford et al. (2019) provided a case history
involving the River Island development in Lathrop CA where estimated seismic
settlements of up to 15 inches using the simplified methods of analysis built into the
computer program CLiq were reduced to at most several inches as a result of improved
site investigations, laboratory testing and analyses. On a number of other projects in
Northern California, estimates of seismic settlements on the order of 10 inches or more
made using CLiq or equivalent spreadsheets have been reduced to an inch or two as a
result of improved site investigations and use of the more detailed method of analysis
described by Pyke (2019), eliminating the need for ground improvement. The nonlinear
effective stress site response analyses employed in these evaluations have also reduced
seismic loadings on the planned structures, thus further reducing development costs.
Figure 1 – Error in Predicted Settlements of Saturated Sands in Christchurch

of 19
Another example of the over-conservatism of simplified methods in predicting seismic
settlements is given by Geyin and Maurer (2019) using data from Christchurch NZ and
shown in Figure 1. This data is for saturated sands in which there was widespread
liquefaction including ejection of sands and silts which should add to the observed
settlements, however, use of the method of Zhang et al. (2002), which relies on CPT
measurements, grossly over-predicts the observed settlements. Because they used a
straight-line regression to fit the data, the published paper states that settlements were
underpredicted when the settlements were very small, although this is a questionable
interpretation and has no practical significance. Note that for the larger estimated
settlements of 0.4 m or 1.3 feet, the error is typically 0.3 m or 0.99 feet, that is, a 75%
overprediction.
This note provides a new summary of the reasons that simplified analyses of seismic
settlements tend to be very conservative, warns against using “fudge factors” to reduce the
settlements calculated by simplified methods, and suggests how to conduct both more
reliable screening and final evaluations.
Reasons that Simplified Analyses of Both Liquefaction and Settlement Tend
to be Conservative
• Simplified methods are necessarily approximate and rely on multiple relationships
or correlations that have a limited range of applicability. It seems that very few of
the users of the spreadsheets or simple programs in which these methods are
embedded are actually aware of all the steps that are involved in a simplified
analysis. Review of these steps should cause responsible engineers to shake their
heads in amazement that anyone takes the results seriously!
• Use of the ground surface peak acceleration and earthquake magnitude to
represent the complex characteristics of earthquake ground motions is over-
simplistic and no substitute for using site-specific acceleration histories which
capture the particular source and site characteristics.
• The shear stresses computed using the standard formula in terms of peak ground
surface acceleration and standard depth reduction curves are generally too high.
There is a growing consensus that the occurrence of liquefaction can only be
understood by conducting nonlinear effective stress analyses in which excess pore
pressure development and dissipation is tracked. See Ntritsos et al. (2018),
Cubrinovski (2019), Hutabarat and Bray (2019), Kramer (2019) and Olson et al.
(2020).

of 19
• The failure to recognize that penetration resistance is a function of both strength
and compressibility under uni-directional loadings and that while some observed
trends might be similar to those observed as responses to cyclic loading, their
magnitudes might be quite different.
• Misclassification of the soil type in CPT interpretations – see for instance Pease
(2010) and the quote from Robertson and Wride (1998) below.
• The failure to account for thin layer effects, particularly when using CPT tip
resistance as an input, and the failure to recognize that both the SPT and CPT have
noses which detect the presence of softer, or stiffer, underlying layers.
• The failure to exclude the “transitions” in CPT data which are created by the cone
passing from sandy to clayey layers or back again. It is not uncommon to see
simplified analyses in which liquefaction and settlement is only reported in these
transitions.
• The failure to properly correct penetration resistance for soils with fines to
equivalent “clean sand” values in methods that rely on penetration resistance as a
proxy for behavior under cyclic loading. This factor is even worse if the simplified
method uses relative density as an input as discussed further below.
• The failure to properly account for ageing effects – which include over-
consolidation, pre-straining, improved packing and chemical bonding. See a
number of references included in the list below for more details.
• The failure to exclude lenses from the analysis. For the reasons explained by Pyke
(1995) and illustrated by Pyke and North (2019), lenses are much less susceptible
to liquefaction and settlement than through-going layers. Briefly, the cyclic shear
strains in a lens are controlled by the stiffness of the matrix in which it is
embedded and, to the extent that any excess pore pressures develop in the lens,
the lens simply becomes a soft inclusion with the same shear strains but lower
shear stresses than the surrounding matrix.
To the extent that simplified analyses are automated, by for instance reading
recorded CPT data directly into a program or spreadsheet, the problem is further
compounded because the engineer is not forced to think about some of these issues.
The essence of good geotechnical engineering is the take the depositional history of
the site into account and think about soil behavior, rather than to mindlessly do
calculations.

of 19
Particular Problems with Estimating Seismic Settlements
• For saturated sands, perhaps the biggest problem is that simplified methods for
analysis of liquefaction over-predict the number of layers in the profile that will
actually reach initial liquefaction and trigger larger settlements. Most simplified
methods for estimating the settlement of saturated sands that occurs upon
dissipation of excess pore pressures generated during earthquake shaking rely on
Ishihara and Yoshimine (1992) which suggests that large settlements, on the
order of 2 to 5%, can occur in layers that have reached complete liquefaction
(defined as 100% excess pore pressure or a single amplitude shear strain on the
order of 3%). While these numbers were based on laboratory tests of clean sands
and they do not necessarily apply to naturally occurring sands, it does seem that
they provide a useful upper bound for naturally occurring sands that actually
liquefy. Settlement resulting from ejecta, such as sand boils, is not addressed
explicitly in these procedures but the Ishihara and Yoshimine values likely cover
any contribution to total settlement from ejecta. In any case, simplified methods
for analysis of liquefaction tend to over-predict the number of layers in the profile
that will actually reach initial liquefaction, as illustrated for instance by Crawford
et al. (2019), so that use of Ishihara and Yoshimine will likely be conservative to
very conservative.
• Also, Ishihara and Yoshimine’s data is presented in terms of relative density and
this can lead to enormous errors if the relative density is blindly interpreted from
CPT data, because the standard relationships for relative density in terms of CPT
tip resistance make no correction whatsoever for fines. An example of the
inadequacy of this lack of correction of CPT data for fines content when
interpreting relative density is provided in Figure 2.
Figure 2 – Interpreted Relative Density as a Function of Ic
0
10
20
30
40
50
60
70
80
90
100
0 0.5 1 1.5 2 2.5 3
RelativeDensity(%)
Ic

of 19
This data was obtained in a single SCPT sounding in an alluvial fan deposit that
showed measured shear wave velocities in excess of 700 feet per second even for
the more silty or clayey sands and shear wave velocities more like 1,100 ft/sec in
clean sands, and appears to have apparent relative densities on the order of 80
percent of more. However, if the interpreted relative densities are plotted against
the parameter Ic, which is an indicator of fines content, it may be seen that the
interpreted relative densities are more an indication of fines content than they
are of relative density. That the silty sands in this deposit actually have relative
densities as low as 30 percent is not credible.
• For saturated sands there is also a further problem caused by use of Ishihara and
Yoshimine (1992), namely that they develop a relationship between settlement on
reconsolidation and the factor of safety against liquefaction using the maximum
shear strain as an intermediary. That makes little sense to most engineers and
the resulting settlements for factors of safety against liquefaction greater than
unity tend to be somewhat greater than those obtained using the procedure
described by Pyke (2019) which is summarized below. And again, if the
evaluation of liquefaction is conservative for any layer, the settlements on
reconsolidation will be overpredicted.
• For sands above the water table there are multiple reasons for excessive
conservatism including the fact that the laboratory data that Tokimatsu and Seed
(1987) relied on was inappropriate and overly conservative. Tokimatsu and Seed
used data from Silver and Seed (1971) and Lee and Albaisa (1974). But Lee and
Albaisa’s data was obtained from triaxial tests, which are inappropriate, and
Silver and Seed used Crystal Silica No. 20 sand, a particularly angular sand,
which, notwithstanding the findings of Duku et al (2008) might be assumed to
show greater settlements than the sands encountered in practice. Pradel (1998)
then relied on the data from Silver and Seed and the Tokimatsu and Seed paper.
In the case of the method of Robertson and Shao (2010) for estimating seismic
settlements from CPT data, which is based on Pradel (1998) and is included in the
computer program CLiq, the interpreted N160cs values may be too low for a variety
of reasons including the presence of clayey fines and thin layer effects. Robertson
and Shao use the correction to tip resistance as a function of the parameter Ic that
was proposed by Robertson and Wride (1998), but the intelligent qualifications
contained in the earlier paper are not necessarily kept in mind by users: “The
proposed correction factor, Kc, is approximate, since the CPT responds to
many factors, such as soil plasticity, fines content, mineralogy, soil sensitivity,
and stress history. However, for small projects or for initial screening on larger
projects, the above correlation provides a useful guide. Caution must be taken

of 19
in applying the relationship to sands that plot in the region defined by 1.64 <
Ic < 2.36 and F < 0.5% so as not to confuse very loose clean sands with sands
containing fines”. If N160cs values are underestimated, this leads to two
compounding errors: (1) the low-strain shear modulus is underestimated so that
the computed cyclic shear strains are too great; and (2) the settlement as a
function of a given cyclic shear strain is also too great.
Doubling of Estimated Settlements?
The estimated settlements for non-saturated sands are then commonly doubled citing
Pyke, Seed and Chan (1975) or Seed, Pyke and Martin (1978), but the concept of adding
the settlements caused by two orthogonal components of shaking applies only to more
accurate calculations. It is not necessary when the calculation is very approximate and
conservative in the first place. Likewise, the effect of vertical motions can generally be
ignored. In addition to the conservatism involved in estimating the cyclic stresses or
strains using simplified methods, because the volume change data used by Ishihara and
Yoshimine and Tokimatsu and Seed, and hence Pradel, was obtained on clean, washed
and screened sands freshly placed in a laboratory test apparatus – what Professor
Jamoilkowski calls “baby sands” – the use of this data for naturally occurring sands,
which may show some effects of fines content, overconsolidation, pre-straining and
other ageing phenomena, can be thought of as cancelling out the need to increase the
calculated settlements in order to account for multi-directional shaking. Additionally,
the fact that the rate of settlement tends to saturate with increasing accumulated
settlement as discussed below, means that although accounting for the second
component of motion increases the rate at which settlements or latent settlement
accumulate, it does not have as much effect on the maximum settlement. For saturated
sands, methods that rely on Ishihara and Yoshimine (1992) should most certainly not be
doubled because the calculated settlements are controlled by the occurrence of
liquefaction or the factor of safety against liquefaction and the effect of multi-directional
shaking should already be taken into account.
Use of Fudge Factors
The estimated settlements are, however, sometimes reduced using a “depth weighting
factor” suggested by Cetin et al. (2009). This is basically an arbitrary fudge factor and
should not be used. That said, the Cetin et al. weighting factor does make at least some
sense in that the errors in the simplified methods likely increase with depth, but it is
much better to eliminate or account for those errors. Another method for reducing the
calculated settlements is to call on the paper by Robertson (2016) and use CPT data to

of 19
distinguish between sand and clay-like and dilative and contractive soils. However, this
is a gross misuse of the excellent Robertson paper, which is directed to a proposed new
classification scheme for soils which might supplement or replace the standard ASTM
classification scheme which has a number of shortcomings. However, the argument that
soils which are shown to be dilative under this classification scheme should be excluded
from the analysis of seismic settlement is just wrong and misses a basic fact about soil
behavior under cyclic loading which is that even soils which are dilative under
monotonic loading can decrease in volume and, if they are saturated, generate excess
pore pressures under cyclic loading. The late Professors Harry Seed and Ken Lee were
awarded ASCE’s Norman medal for first documenting this finding and it has been
confirmed repeatedly since then. To be sure, as the relative density or Robertson’s CD
number (as defined in his paper) increases, volume changes caused by cyclic loading will
decrease, but they will not suddenly fall to zero.
So, not doubling the settlements estimated by simplified methods is legitimate but use
of either of these fudge factors is not.
Summary of Main Points
Some of the major points made above are listed in the following text box.
• The failure to exclude materials with clayey fines
• The failure to exclude the “transitions” in CPT data.
• The failure to properly correct penetration resistance for fines content to
equivalent “clean sand” values and to account for the effect of the presence of
fines on settlement due to compaction
• The failure to exclude lenses from the analysis
• Overprediction of the number of layers that might liquefy
• Unnecessary doubling of calculated settlements

of 19
A Better Approach
Pyke (2019) and Pyke and North (2019) describe an improved method for evaluating
liquefaction and settlement which uses bi-directional, nonlinear effective stress site
response analyses as embodied in the computer program TESS2. Data on the settlement
of dry Monterey No. 0 sand caused by cyclic loadings obtained from Pyke (1973) is built
into the program but the user can apply a multiplier to this data or specify site-specific
data should that be available. In lieu of acquiring site specific data, users can refer to
Ramadan (2007), Duku et al. (2008) and Yee et al. (2014) for data on other sands. Note
that Yee et al (2014) suggest that compaction caused by cyclic shearing is reduced when
even the low plasticity fines content exceeds 10 percent. For saturated sands, the
settlement on reconsolidation from Ishihara and Yoshimine (1992) is built into the
program for use when excess pore pressure in any layer reaches 100%, but again, the user
can specify site-specific data should that be available. Otherwise the latent settlements,
that is the settlements which are only seen on dissipation of excess pore pressures, of
saturated sands are based on Martin et al. (1975) and Seed et al. (1978), who assumed
that, short of the development of 100% excess pore pressure, the settlement on
dissipation of excess pore pressures in a saturated sand is the same as the settlement that
would occur under the same loading in a non-saturated sand.
The key to this improved method is that you need to know the history of cyclic shear
strains in each layer in order to make a reasonably accurate estimate of the likely
settlement in a given earthquake. That in turn requires the selection of appropriate
acceleration histories to use as input motions but that task has been made much easier
by the development of the PEER and other earthquake ground motion databases. It is in
fact astonishing that the development of modern computers which has made much
more precise and useful calculations of the response of structures to earthquakes, has
been sidelined in geotechnical engineering in favor of simplified methods that basically
could be done by hand. That was understandable in 1970 when the first simplified
procedure for the evaluation of liquefaction potential was published because only a
handful of engineers could conduct site response analyses, but it is hard to understand
today. If the simplified methods forced the user to conduct better site investigations and
more carefully analyze the data, that would be an argument in their favor, but the
opposite is true. In practice the simplified methods tend to be more automated and the
user is not required to study the data carefully.
The background and details of the calculation performed by TESS2 can be summarized
as follows: Pyke (1973) conducted numerous cyclic simple shear and shaking table tests
to evaluate the settlement behavior of dry sand. If the results were reduced and plotted
in the traditional way in terms of the applied cyclic stress ratio the results that were
obtained can be plotted as shown in Figure 3.

of 19
Figure 3 – Summary of Results from Pyke (1973)
From Uni-Directional Shaking Table Tests
This form of presentation of the data shows the relative effects of relative density, the
average cyclic stress ratio, and the number of uniform cycles at a glance. It is not used in
TESS2 but may be convenient for simple “back of the envelope” calculations.
For use in a more complete and accurate calculation, the data needs to be reduced and
presented in terms of cyclic shear strains, rather than cyclic shear stresses, as shown in
the subsequent three figures which show data from more or less constant strain cyclic
simple shear tests. For a relatively constant cyclic shear strain, the settlement decreased
and the stiffness increased with an increasing number of cycles, as may be seen in
Figure 4. Both of these factors should be addressed in any attempt to perform more
accurate calculations. Although this was intended to be a constant cyclic strain test, the
cyclic shear strains were not constant because of compliance in the test apparatus and
this needs to be accounted for in the data reduction. Following the suggestion of
Geoffrey Martin, as published in Martin et al. (1975), the data on settlement was
reduced as shown in Figure 5 in which the settlement per cycle is shown as a function of
the cyclic shear strain and the accumulated settlement.

of 19
Figure 4 – Typical Results of Cyclic Simple Shear Test on Monterey #0 Sand
Figure 5 - Settlement per Cycle for Monterey #0 Sand

of 19
The accumulated settlement turns out to serve as a very good measure of the effects of
the strain history to that point. It was also found that when the data was reduced in this
way, the settlement per cycle was largely independent of confining pressure, confirming
that behavior under cyclic loading is more fundamentally controlled by the cyclic shear
strain, rather than the cyclic shear stress or stress ratio.
The data on the secant shear modulus can also be reduced in a similar fashion as shown
in Figure 6. The increase in the secant shear modulus for this dry sand is quite marked,
and Pyke (1973) suggests that it still applies when there is multi-directional shaking. As
pointed out by Vucetic and Mortezaie (2015), this effect can also be seen in undrained
cyclic tests on saturated sands although in that case it is quickly overwhelmed by the
decrease in stiffness that accompanies the development of excess pore pressures.
Figure 6 – Hardening of Monterey #0 Sand
Figures showing the effect of horizontal shaking with two orthogonal components and of
vertical shaking are not shown in this note, but they may be found in Pyke (1973) and
Pyke et al. (1974). The latter publication may be found online at
https://peer.berkeley.edu/ucbeerc-report-series. The principal finding was the
settlements caused by horizontal shaking with two orthogonal was approximately equal
to the sum of the settlements caused by horizontal shaking with each component acting
alone. In TESS2, the stiffnesses and the settlements or latent settlements are calculated
for each half cycle. The contributions to settlements or latent settlements and excess
pore pressures from the two horizontal components, which are run simultaneously, are
then added. For soils below the water table, if the excess pore pressure ratio in any layer
reaches 100%, the latent settlement for that layer jumps to the Ishihara and Yoshimine

of 19
value. Because of this it is important to use an effective stress analysis in which the
excess pore pressures are redistributed and dissipated as appropriate. The point of
performing the calculations this way is that the shear strain history and the peak excess
pore pressures make a difference, thus the character and duration of the input motions
also make a difference. Although this calculation is too onerous to perform by hand or
even in a spreadsheet, it provides a more accurate calculation and, if run with a suitable
number of input motions, shows the sensitivity of the computed settlements to the
random nature of earthquake ground motions.
Although, short of reaching 100 percent relative density, there is no hard limit on the
total amount of settlement that can occur. It can be seen from Figures 4 and 5 that
additional settlements will be small once the accumulated settlement reaches a value on
the order of 0.5 percent under uni-directional loading. Since the settlements caused by
each component of motion are additive, once the accumulated settlement reaches about
0.5% of the layer thickness, additional settlements caused by motion in either direction
will also be small. In other words, accounting for the second component of motion
increases the rate at which settlements or latent settlement accumulate but does not
have as much effect on the maximum settlement.
As previously noted, while only the data for Monterey No. 0 sand is built into TESS2,
users can refer to various references for data on other sands and either use that data or
apply a multiplier as necessary. As discussed by Pyke (2021), vertical motions are usually
neglected for saturated sands but maybe have a small effect on non-saturated sands.
There is no way to include this in the analysis directly, but the effect of vertical motions
can also be included in the multiplier as appropriate. Note also that the data presented
above is on a baby sand deposited by dropping it through the air. Pyke (1973) was one of
the first studies to explore the effect of the method of sample preparation on settlement
and liquefaction under cyclic loading and includes data on the effect of overconsolidation
and the application of an initial static shear strain, both of which reduce the settlement
per cycle. The upshot of this is that the processes involved in deposition and subsequent
ageing in the field are likely to decrease the settlement per cycle, perhaps significantly,
from those seen in laboratory tests and that Pyke’s data on Monterey No. 0 sand likely
provides an upper bound on expected settlements in the field.
Other Contributions to Settlement
As suggested by Macedo and Bray (2018), settlements of buildings caused by
liquefaction can be categorized as ejecta-induced, shear-induced, or volumetric-induced.
Most of the literature, and indeed this note, focuses on settlement due to volume
change. As noted above, conservative evaluations of settlement due to volume change
likely account for the impact of ejecta as well. What Macedo and Bray refer to as shear-

of 19
induced settlements might be better called “distortion”, but regardless of the
terminology, this mechanism has led to some of the most dramatic failures due to
liquefaction such as the overturning of the apartment buildings at Kawagishi-cho in the
1964 Niigata earthquake. However, prediction of such movements is largely of academic
interest. From the practical point of view the lesson to be learned from these failures is
to not put buildings with shallow (or no) foundations on loose sandy or silty soils,
especially when there is a high water table. And it is doubtful that such displacements
can be predicted with sufficient accuracy to take them into account in performance-
based engineering design.
In addition to the distortion caused by the superposition of cyclic shear strains on top of
initial shear strains, the behavior of sands under a building is also impacted by the
weight of the building, which increases the shear modulus or stiffness of the sand and
thus decreases the cyclic shear strains, and the mass of the building which increases the
inertia forces and hence the shear stresses and strains in the sand. Both these effects can
now be modeled in TESS2 analyses and limited experience suggests that the second of
these two factors will normally dominate over the first. However, while including the
weight and the mass of the building in the analysis tends to increase the computed
settlements or latent settlements, it can have a much greater effect on reducing the
foundation input motions that are applied in a structural analysis of the building.
Note also that one of the things that makes back-calculation of the settlements observed
in case histories very difficult, is that vertical displacements, or settlement, might be
associated with lateral spreading or landsliding. Such settlements are very difficult to
predict so it is important that these issues also be addressed in design and eliminated as
possible contributors to settlement. If liquefaction, lateral spreading and landsliding are
accounted for in design, earthquake-induced settlements are not particularly dramatic
or important, which is why academic studies of this issue have been relatively limited
and the subject has only become a big issue as a result of inappropriate or incorrect
application of simplified methods of analysis.
Recommended Practice
No responsible engineer should use any of the simplified methods for evaluating
liquefaction or settlement unless they are familiar with each step in the procedure, the
limits of applicability of that step and whether the site in question fits within the limits of
the overall applicability of the method. Even then, simplified methods for estimating
seismic settlement should at best be used only for screening evaluations. If a screening
evaluation indicates settlements that are not of practical concern, nothing further need be
done, but if larger settlements are obtained it should not be assumed that ground

of 19
improvement is required. If a screening analysis indicates seismic settlements that are of
practical concern, then analysis of the kind described above should be performed in order
to refine the estimate and determine whether or not ground improvement is necessary.
However, a “simplified analysis” is not necessarily required even as a screening
evaluation. The widespread belief that “one has to show a calculation” tends not to
promote better geotechnical engineering practice but rather worse practice. A good
screening analysis should emphasize common-sense and experience.
A better screening methodology:
The first step in any screening analysis should be evaluation of the regional geology and
seismicity and answering the question “is there any evidence of earthquake-induced
liquefaction and settlement of similar soils in a similar tectonic environment? See Pyke
(1995, 2003, 2015) and Semple (2013). While there are rare instances of liquefaction
being reported in Pleistocene age sands, the vast majority of well-documented case
histories have occurred in geologically recent materials and man-made fills,
particularly hydraulically placed fills. Thus, for a start all soil layers that have “clayey”
descriptors should be tested as necessary to confirm that description and along with
soils that are older than several thousand years should be excluded from any
quantitative analysis of liquefaction or settlement. Sand “layers” seen in individual
borings or soundings that are shown to be discontinuous by adjacent borings or
sounding should also be excluded.
Then, if there are any remaining non-saturated sand or silt layers that are less than
medium dense to dense, on the basis of Pyke (1973) and the argument laid out above
on use of baby sand data offsetting the effects of multi-directional shaking a very
simple check can be made ignoring the possible effects of vertical motions and
assuming that the settlement caused by even strong shaking is unlikely to exceed 0.5%
of the layer thickness, even for a non-saturated sand with a relative density as low as
40%. A more refined estimate can be made using the data shown in Figure 5 which
suggests that if the relative density of the sand is greater than 60%, the 0.5% number
might be cut in half, and if the relative density of the sand is greater than 80%, the
number might be cut in half again. Ishihara and Yoshimine (1992) in their Figure 10
suggest that these three relative densities correspond to normalized SPT N values for
clean sand of 6, 14 and 25 and normalized cone tip resistances in kg/cm2 of 45, 80 and
147. Do not forget that these are for clean sands and the corrections for fines need to
be made if the percent passing the No. 200 sieve exceeds, say, 12% or the Ic from a CPT
test exceeds 1.64. For saturated sands that have relative densities of less than say 60%
and are likely to liquefy, the estimated settlements, still using Ishihara and Yoshimine,
jump to 4.5%, 2.8% and 1.7% for the same three relative densities.

of 19
This common-sense or “back of the envelope” calculation likely constitutes a more
reliable screening evaluation than any of the published simplified methods and it forces
the engineer to study the data and think, rather than just plugging numbers into a
computer program or spreadsheet.
But whether the screening analysis involves simplified analyses or just common-sense
and experience, if unacceptably large settlements are obtained and an improved analysis
is called for, or, if you are dealing with a critical or high value facility on which an accurate
analysis is required even on the first pass, the first step in conducting an improved
analysis is likely an improved site investigation and careful study of the data that is
obtained (unless this has already been done as part of an initial assessment). An adequate
site investigation will generally include measurement of shear wave velocities, advancing
borings and obtaining samples in addition to pushing CPTs, and hydrometer tests and
plasticity index tests to learn the character of any fines. It should be kept in mind that
actual sand layers or lenses are usually offset from the depths indicated by CPTs since the
CPT is measuring the properties ahead of the cone, but good practice is to first push CPTs
and then to follow-up with borings and SPT measurements and sampling in any sand
layers or lenses. The fraction of the sample passing the No.200 sieve should then be
determined for each separate material found in the tip and the barrel of the SPT sampler
and hydrometer tests and plasticity index tests then should be performed on samples with
more than say 30 percent passing the No. 200 sieve. Additional borings or CPTs should
be advanced as necessary to confirm that sand layers are not continuous if this is
suggested by an initial or preliminary investigation.
Conclusion
Existing simplified methods for evaluation of earthquake-induced liquefaction and
settlement are not simple at all. They contain multiple steps often involving complex
formulae which are nonetheless gross approximations. If used blindly without
understanding of the fact that they don’t capture the complexities of earthquake ground
motions, real soil deposits and soil behavior under cyclic loading, they are better
characterized as dumb, rather than simple. The “back of the envelope” method
suggested above is likely a better screening tool than existing spreadsheets and simple
computer programs, but on critical or high value projects responsible engineers should
conduct improved site investigations and analyses from the start.

of 19
References
Andrus, D.R. and Stokoe, K.H., II “Liquefaction Resistance of Soils from Shear-Wave
Velocity,” Journal of Geotechnical and Geoenvironmental Engineering, ASCE, Vol. 126,
No. 11, 2000
Andrus, R.D., Hayati, H. and Mohanan, N.P., “Correcting Liquefaction Resistance for
Aged Sands Using Measured to Estimated Velocity Ratio”, Journal of Geotechnical and
Geoenvironmental Engineering, ASCE, Vol 135, No. 6, 2009
Arango, I., Lewis, M.R. and Kramer, C., “Updated Liquefaction Potential Analysis
Eliminates Foundation Retrofitting at Two Critical Structures”, Soil Dynamics and
Earthquake Engineering, Vol 20, pp 17-25, 2000
Boulanger, R.W., et al., “Evaluating Liquefaction and Lateral Spreading in Interbedded
Sand, Silt and Clay Deposits Using the Cone Penetrometer”, Geotechnical and
Geophysical Site Characterization 5, Australian Geomechanics Society, Sydney,
Australia, 2016
Cetin, K.O., et al., “Probabilistic Model for the Assessment of Cyclically Induced
Reconsolidation (Volumetric) Settlements”, Journal of Geotechnical and
Geoenvironmental Engineering, ASCE, Vol 135, No. 3, 2009
Crawford, C., Tootle, J., Pyke, R. and Reimer, M., “Comparison of simplified and more
refined analyses of seismic settlements”, Proc. 7th International Conference on
Earthquake Geotechnical Engineering, Rome, June 2019
Cubrinovski, M., Keynote Lecture 09, “Key aspects in the engineering assessment of soil
liquefaction”, Proc. 7th International Conference on Earthquake Geotechnical
Engineering, Rome, June 2019
Duku, P.M., et al., “Volumetric Strains of Clean Sands Subject to Cyclic Loads”, Journal
of Geotechnical and Geoenvironmental Engineering, ASCE, Vol. 134, No. 8, 2008
Gayin, M., and Maurer, B.W., “An Analysis of liquefaction-induced Free Field Ground
Settlement Using 1000+ Case Histories”, ASCE Geotechnical Special Publication 308,
2019
Hayati, H. and Andrus R.D., “Updated Liquefaction Resistance Correction Factors for
Aged Sands, Journal of Geotechnical and Geoenvironmental Engineering, ASCE, Vol
135, No 11, pp 16831692, 2009
Hayati, H., Andrus R.D., Gassman, S.L., Hasek, M., Camp, W.M., and Talwani, P.,

of 19
“Characterizing the Liquefaction Resistance of Aged Soils”, Geotechnical Earthquake
Engineering and Soil Dynamics IV Congress, ASCE GSP 181, 2008
Hutabarat, D., and Bray, J.D., “Effective stress analysis of liquefiable site in
Christchurch to discern the characteristics of sediment ejecta”, Proc. 7th International
Conference on Earthquake Geotechnical Engineering, Rome, June 2019
Ishihara, K., and Yoshimine, M., “Evaluation of Settlements in Sand Deposits Following
Liquefaction During Earthquakes”, Soils and Foundations, Vol.32, No.1, pp.173-188,
March 1992
Kramer, S., Keynote Lecture 08, “The use of numerical analysis in the interpretation of
liquefaction case histories”, Proc. 7th International Conference on Earthquake
Geotechnical Engineering, Rome, June 2019
Leon, E., Gassman, S.L., and Talwani, P., “Accounting for Soil Aging when Assessing
Liquefaction Potential”, Journal of Geotechnical and Geoenvironmental Engineering,
ASCE, Vol. 132, No.3, pp 363-377, 2006
Lee, K.L., and Albaisa, A., “Earthquake-induced settlements in saturated sands”,
Journal of the Soil Mechanics and Foundations Division, ASCE, 100 (4), 1974
Lewis, M.R., et al., “Site Characterization Philosophy and Liquefaction Evaluation of
Aged Sands”, Geotechnical Earthquake Engineering and Soil Dynamics IV Congress,
ASCE GSP 181, 2008
Macedo, J., and Bray, J.D., “Key Trends in Liquefaction-Induced Building Settlement”,
Journal of the Geotechnical and GeoEnvironmental Division, ASCE, Vo. 144, No. 11,
2018
Martin, G.R., Seed, H.B., and Finn, W.D.L., “Fundamentals of Liquefaction under Cyclic
Loading”, Journal of the Geotechnical Engineering Division, ASCE, Vol.101, No. GT5,
May 1975
Mitchell, J.K., “Practical Problems from Surprising Soil Behavior - the Twentieth
Terzaghi Lecture”, Journal of Geotechnical Engineering, ASCE, Vol. 112, No. 3, pp. 255–
289, 1986
Ntritsos, N., et al., “Evaluation of Liquefaction Case Histories from the 2010-2011
Canterbury Earthquakes Using Advanced Effective Stress Analysis”, Geotechnical
Earthquake Engineering and Soil Dynamics V ASCE Geotechnical Special Publication
290, 2018
Olson, S.M., et al. “Nonlinear Site Response Analysis with Pore-Water Pressure

of 19
Generation for Liquefaction Triggering Evaluation”, Journal of the Geotechnical and
GeoEnvironmental Division, ASCE, Vo. 146, No. 2, 2020
Pease, J.W., “Misclassification in CPT liquefaction evaluation”, 2nd Int. Symposium 0n
Cone penetration Testing, Huntington Beach CA, May 2010.
Pradel, D., “Procedure to Evaluate Earthquake-Induced Settlements in Dry Sandy Soils”,
Journal of the Geotechnical and GeoEnvironmental Division, ASCE, Vo. 124, No. 4,
1998
Pyke, R., "Settlement and Liquefaction of Sands Under Multi-Directional Loading,"
Ph.D. Thesis, University of California, Berkeley, 1973
Pyke, R., "Practical Aspects of the Evaluation of Liquefaction Potential", Earthquake
Geotechnical Engineering, Ishihara (ed.), Balkema, 1995
Pyke, R., Discussion of “Liquefaction Resistance of Soils: Summary Report From the
1996 NCEER and 1998 NCEER/NSF Workshops on Evaluation of liquefaction
Resistance of Soils”, Journal of Geotechnical and Geoenvironmental Engineering, ASCE,
Vol. 129, No.3, pp 283-284, 2003
Pyke, R., “Evaluating the Potential for Earthquake-Induced Liquefaction in Practice”, 6th
International Conference on Earthquake Geotechnical Engineering, Christchurch, New
Zealand, November 2015
Pyke, R., “Improved analyses of earthquake-induced liquefaction and settlement”, Proc
7th International Conference on Earthquake Geotechnical Engineering, Rome, June
2019
Pyke, R, “Lessons learned from the observed seismic settlement at the Jensen Filtration
Plant in the San Fernando Earthquake”, paper submitted to ASCE Lifelines Conference
2021
Pyke, R., and North, J., “Shortcomings of simplified analyses of earthquake-induced
liquefaction and settlement”, Proc. 7th International Conference on Earthquake
Geotechnical Engineering, Rome, June 2019
Pyke, R., Chan, C.K., and Seed, H.B., "Settlement and Liquefaction of Sands Under
Multi-Directional Shaking," Report No. EERC 74-2, University of California, Berkeley,
1974
Pyke, R., Seed, H.B., and Chan, C.K., “Settlement of sands under multidirectional
shaking.” Journal of Geotechnical Engineering, ASCE 101(4), 379-398, 1975
Ramadan, J.I., “Settlement of Dry Cohesionless Soil Deposits Under Earthquake
Induced Loading”, Ph.D. Dissertation, University of Southern California, December

of 19
2007. Available online at https://144g4i11tdrn3j2cti9hn458-wpengine.netdna-
ssl.com/wp-content/uploads/2017/02/Dissertation_Ramadan2007.pdf
Robertson, P.K., “Cone Penetration Test (CPT) Based Soil Behavior Type (SBT)
Classification System – An Update”, Canadian Geotechnical Journal, Vol. 53, No.12,
2016
Robertson, P.K. and Shao, L., “Estimation of Seismic Compression in Dry Soils Using
the CPT”, 5th Int. Conf. on Recent Advances in Geotechnical Earthquake Engineering
and Soil Dynamics. San Diego, 2010
Robertson, P.K., and Wride, C.E., “Evaluating cyclic liquefaction potential using the
cone penetration test.” Canadian Geotechnical Journal, 35(3), 1998
Seed, H.B., Pyke, R., and Martin, G.R., "Effect of Multi-directional Shaking on Pore
Pressure Development in Sands," Journal of the Geotechnical Engineering Division,
ASCE, Vol. 104, No. GT1, January 1978
Semple, R., “Problems with Liquefaction Criteria and Their Application in Australia”,
Australian Geomechanics, Vol. 48, No. 3, pp 15-48, September 2013
Silver, M.L., and Seed, H.B., “Volume changes in sand due to cyclic loading”, Journal of
Geotechnical Engineering, ASCE, 97(9), 1971
Tokimatsu, K., and Seed, H.B., “Evaluation of settlements in sands due to earthquake
shaking”, Journal of Geotechnical Engineering, ASCE, 113(8), 1987
Vucetic, M., and Mortezaie, A., “Cyclic secant shear modulus versus pore water pressure in
sands at small cyclic strains”, Soil Dynamics and Earthquake Engineering, Volume 70,
2015
Zhang, G., et al., “Estimating Liquefaction-Induced Ground Settlements from CPT for
Level Ground”, Canadian Geotechnical Journal, Volume 39, pp.1168-1180, 2002
Yee, E., Duku, P.M., and Stewart, J.P., “Cyclic Volumetric Strain Behavior of Sands with
Fines of Low Plasticity”, Journal of Geotechnical and Geoenvironmental Engineering,
ASCE, Vol. 144, No.4, 2014

Limitations of simplified methods for estimating seismic settlements 12d

More Related Content

What's hot (20)

Similar to Limitations of simplified methods for estimating seismic settlements 12d (20)

Recently uploaded (20)

Limitations of simplified methods for estimating seismic settlements 12d