Proceedings of the 1st International Conference on Civil Engineering for Sustainable Development

(ICCESD-2012), 2~3 March 2012, KUET, Khulna, Bangladesh

EVALUATION OF PEAK GROUND ACCELERATION USING CPT DATA FOR

LIQUEFACTION POTENTIAL

Ram Krishna Mazumder1*, Estiak Mahmud Murad2 and Mehedi Ahmed Ansary3
1
Research Engineer, Bangladesh Network Office for Urban Safety, Bangladesh University of Science and
Technology, Dhaka-1000, Bangladesh, email: [email protected]
2
Research Engineer, Bangladesh Network Office for Urban Safety, Bangladesh University of Science and
Technology, Dhaka-1000, Bangladesh, email: [email protected]
3
Professor, Department of Civil Engineering, Bangladesh University of Engineering and Technology (BUET),
Dhaka-1000, Bangladesh, Email: [email protected]

ABSTRACT
This paper presents liquefaction potential evaluation at six important locations of different development
projects. Bangladesh has been identified as moderate vulnerable country according to world seismic map.
Recent earthquakes points to the probability of facing a serious seismic hazard. Most of the private land
development projects of Dhaka city rested on a sand deposit. The ground water table was shallow for these
locations. Theoretically, the site was susceptible to liquefaction during ground-shaking. Infrastructure
construction by land fill has the potential risk for seismic hazard due to liquefaction of these reclaimed soils. To
identify liquefaction potential of such areas, in situ field test has been conducted for six important locations of
different development projects. The liquefaction analysis based upon force equilibrium concept cannot be
directly performed in this study, because there are no accelerometers in the hazard areas in any earthquake.
Therefore, the accelerations back-calculated by liquefaction analysis are recommended in this article. The
analysis was calculated using cone penetration test (CPT) data that collected from testing sites. The Peak
Ground Aaccelerations (PGA) from 0.21 g to 0.35 g will generate liquefactions about all depth, loose sand
layers in hazard areas.

Keywords: CPT, Earthquake, Liquefaction Potential, PGA, SPT.

1. INTRODUCTION

Bangladesh is seismically moderate country where Dhaka is its Capital. Dhaka City is centrally located in
Bangladesh, in the southern part of the district of Dhaka. It is situated between latitudes 24º40´ N to 24º54´ N
and longitudes 90º20´ E to 90º30´ E and bounded by Buriganga river in the south; the Balu and the Shitalakhya
rivers in the east; Tongi Khal in the north and the Turag river in the west. The city has developed on the higher
elevated Pleistocene terrace land or Order Alluvium of the central part of Bangladesh, otherwise referred to as
the Madhupur-Bhawal Garh Region, (Hossain, 2008). Present populations of this mega city are more than ten
millions.

The legal and administrative boundaries of the city nowadays have however been extended to a great extent. A
substantial portion of the adjoining low-lying areas have recently been brought under the structured zones of the
city due to the accelerated rate of the urban growth in Dhaka. To cope up with ever-increasing pressure Dhaka
has started going upwards, little scope for horizontal expansion due to topographical reasons. Dhaka is on the
verge of a change in its urban character, vertical growth taking the place of horizontal expansion. As a result,
most of the development projects have been undertaken by land filling of low-lying marshy areas.

To identify the susceptibility of liquefaction for those new development projects areas, liquefaction potential
analysis has been done using CPT and SPT data for six important locations. PGA value has been considered for
a maximum size of Earthquake M w 7.5. Figure 1illustrates procedure of CPT concerning liquefaction potential
evaluation. Proposed PGA map by Comprehensive Disaster Management Program (CDMP) has been used in
order to calculate liquefaction potential. This paper provides an update on PGA for liquefaction based on basic
CPT measurements of qc and fs. PGA map and locations of in-situ field test shown in Figure 2. Figure 3 shows
CPT test at Bramongoan.
 1st International Conference on Civil Engineering for Sustainable Development (ICCESD-2012)

Figure 1: Cone Penetrometer for evaluating site-specific soil liquefaction concerns and PGA map for Dhaka
(CDMP, 2009).

Figure 2: PGA map for Dhaka (left) and locations of field test.

Figure 3: Cone Penetration Test demonstration at BUET campus.

2. METHODOLOGY

2
 1st International Conference on Civil Engineering for Sustainable Development (ICCESD-2012)

The CPT has major advantages over traditional methods of field site investigation such as drilling and sampling
since it is fast, repeatable and economical. In addition, it provides near continuous data and has a strong
theoretical background. These advantages have lead to a steady increase in the use and application of the CPT in
many places around the world. One of the major applications of the CPT has been the determination of soil
stratigraphy and the identification of soil type. This has typically been accomplished using charts that link cone
parameters to soil type (Robertson, 2009).

To evaluate liquefaction resistance of soils, two variables are required for evaluation. The level of ground
shaking from seismic loading is express in terms of Critical Stress Ratio (CSR) and the capacity of the soil to
resist liquefaction, expressed in terms of cyclic resistance ratio (CRR). The CSR can be estimated using seismic
ground hazard maps or alternatively evaluated more properly using site specific Peak Ground Acceleration
(PGA) data within relevant codes. Using conventional simplified procedure, the CSR can be calculated by using
peak ground surface acceleration (amax) at the site (Seed and Idriss, 1971). This simplified approach can be
summarized as follows:

CSReq=τave/ σ'v =0.65(amax/g) rd (σv/σ'v)

where τave is the average equivalent uniform shear stress generated by the earthquake (assumed to be 65% of the
maximum induced stress), amax is the peak horizontal ground surface acceleration, g is the gravitational
acceleration (g = 9.8 ms2), σv is total vertical overburden stress, σ' v is effective overburden stress at the same
depth and rd is shear stress reduction coefficient which is dependent on depth. By using recommendation of the
National Center for Earthquake Engineering workshop on soil liquefaction (Youd et al., 2001), r d can be
obtained with depth z (meters) as follows:

(130 -z)/131 if z<9.15 m

rd= (44 - z)/37 if 9.15m ≤ z ≤ 23m
(93 - z)/125 if 23m ≤ z ≤ 30m
0.5 if z >30m

The value of amax is taken from the appropriate design events for a given project (i.e., the 2% probability of
earthquake for a certain period of time, the maximum credible event for a known fault located a certain distance
from the site or a code based response spectrum).

Robertson (1990) stressed that the CPT-based charts were predictive of Soil Behaviour Type (SBT), since the
cone responds to the in-situ mechanical behavior of the soil and not directly to soil classification criteria based
on grain-size distribution and soil plasticity (e.g. Unified Soil Classification System, USCS). Grainsize
distribution and Atterberg Limits are measured on disturbed soil samples. Fortunately, soil classification criteria
based on grain-size distribution and plasticity often relate reasonably well to in-situ soil behaviour and hence,
there is often good agreement between USCS-based classification and CPT-based SBT (e.g. Molle, 2005).

The modified CPT soil behavior type index, ISBT is defined by Robertson and Wride using only CPT Q and F
data because porewater pressure are often near hydrostatic for loose firm clean sands (see figure 3). The
normalized SBT index Ic, is given by
Ic = √[(3.47-logQ)2 + (1.22+logF)2]
Where, Q = [(qc- σv)/Pa] (Pa/σ'v)n
F = [fs/(qc- σv)] 100

The value of Q is adjusted to qc1n for stress normalization. This requires iteration process as below:
qc1n = (qc/Pa2] (Pa/σ'v)0.5 if Ic <2.6
qc1n = Q if Ic >2.6
Again, Ic is calculated by Ic = √[(3.47-log qc1n)2 + (1.22+logF)2]
Now, if Ic> 2.6 then qc1n has to be recalculate,
qc1n = (qc/Pa2] (Pa/σ'v)0.75 if Ic >2.6
Finally, Ic = √[(3.47-log qc1n)2 + (1.22+logF)2]

Where, qc is measured cone tip resistance, f s is measured cone sleeve resistance, P a is a reference stress of 100
kPa (or 1 atm), Pa2 is reference stress in MPa and n = an exponent that depends on soil behaviour type. The
values of qc, fs, Pa, σv, and σ'v are all in the same units. The value of n ranges from 0.5 for clean sands to 1.0 for
clays (Olsen et al., 1997), and can be approximated through an iterative approach.

3
 1st International Conference on Civil Engineering for Sustainable Development (ICCESD-2012)

SBT zone
Soil Behavior Type
Robertson (1990)
1 Sensitive fine-grained
2 Clay - organic soil
3 Clays: clay to silty clay
4 Silt mixtures: clayey silt & silty clay
5 Sand mixtures: silty sand to sandy silt
6 Sands: clean sands to silty sands
7 Dense sand to gravelly sand
8 Stiff sand to clayey sand*
9 Stiff fine-grained*
* Overconsolidated or cemented

Figure 4: Updated non-normalized SBT chart based on dimensionless Cone Resistance, (q c/pa) and Friction
Ratio, Rf = (fs/qc-σv) 100%, showing contours of ISBT.

Kc is a correction factor for the apparent fines content and is empirically calculated from a modified CPT soil
classification index, Ic. Specifically Kc is evaluated from:

Kc = 1.0 for Ic ≤1.64

Kc = - 0.403Ic4+5.581Ic3-21.63Ic2+33.75Ic-17.88 for Ic >1.64

Normalized Cone Tip Resistance, (qc1n)cs= Kc(Pa/σ'v)n(qc/Pa).

Corrected Critical Stress Ratio Resisting Liquefaction (CSRL) was calculated using different correction factor.

CSRL = CSR km kα ks
Where,
km = Correction factor for earthquake magnitude other than 7.5 (1.00 for Mw 7.5)
kα = Correction factor for initial driving static shear (1.00 for no initial static shear)
ks = Correction factor for stress level larger than 96 kPa

The level of ground motion CSR and the adjacent tip resistance (q c1n) are compared with the CRR to determine
whether will or will not occur. CRR is calculated by the following equation for an earthquake moment
magnitude of 7.5 (Youd et al., 2001; Robertson and Wride, 1998):

CRR = [0.833 (qc1n)cs /1000] + 0.05 if (qc1n)cs <50

CRR = 93 [(qc1n)cs /1000]3 + 0.08 if 50 ≤ (qc1n)cs ≤ 160
The Cyclic Resistance Ratio (CRR) is the threshold for liquefaction and used to compare the available soil
resistance with level of ground shaking represented by the CSR. Therefore if the CSR value is higher than CRR,

4
 1st International Conference on Civil Engineering for Sustainable Development (ICCESD-2012)

the soil has a high likelihood of liquefaction. If the CSR falls beneath the CRR, the likelihood of liquefaction is
small. The CRR can be expressed using conventional approaches that give a binary decision (liquefaction or no
liquefaction), or alternatively, in terms of probabilistic curves of increasing risk of liquefaction.

Figure 5: Probabilistic cyclic resistance ratios (CRRs) for clean sands based on (upper) normalized cone tip
resistance and (lower) normalized shear wave velocity (after Juang and Jiang 2000).

A calculated factor of safety (F S) can be defined as FS= CRR/CSRL for a particular earthquake magnitude and a
set of data. In more recent evaluations, CRR curves of different probabilities of occurrence have been developed
from mapping functions (Chen and Juang, 2000; Juang and Jiang, 2000) to relate the safety factors F S to the
liquefaction probability PL. Based on database of 225 CPT-based cases reported by Juang and Jiang (2000) for
qc1n probability curves:
PL=1/ [1+ (FS/1.0)3.34]

3. RESULTS

From CPT site investigation report, tests are covered by 15 m to 23 m thick soil layers. The ground water table
at ground level considered as the worst condition. The peak horizontal ground acceleration value for the site has
been taken corresponding to zone factor according to CDMP. a max/g for Dhaka has been taken 0.21 to 0.35, the
extent to which liquefaction is expected for 7.5 magnitude earthquake. The saturated unit weight of the soil
taken 18 kN/m3 and unit weight of water is taken 9.8 kN/m3.

Figure 6 presents the measured CPT parameters in terms of cone resistance (q c), friction ratio (Rf) and pore
pressure (u2), as well as the profiles of non-normalized SBT index (ISBT or Ic) and the SBT descriptions, based on
the chart shown in Figure 4. The SBT zones are colour coded to aid in visual representation (Geologismiki,
CPeT-IT). The pore pressure sensor appears to have lost saturation when passing through the thin dense sand
layer up to 12 m, but regained saturation again at a depth of about 19 m.

5
 1st International Conference on Civil Engineering for Sustainable Development (ICCESD-2012)

Figure 6: Example CPT Profile to illustrate non-normalized SBT and SBT Index, ISBT.

Figure 7 presents the normalized CPT parameters as defined and updated by by Robertson (2009), as well as the
profiles of normalized SBT index (Ic) and the SBTn descriptions, base on Robertson SBT zone 1990. Figures 6
and 7 show that there is little difference between the soil behavior type interpretation for this example profile.

Figure 7: Example CPT Profile to illustrate normalized SBT and SBT Index, I c.

Figure 8 shows the CPT data plotted on both the updated non-normalized SBT chart and the normalized SBT n
(Robertson, 1990) chart. Each layer is represented by different colour data points to aid identification. Figure 8
shows that the CPT data are more closely clustered on the normalized charts, as expected, but that both charts
provide similar interpretation of soil behaviour type. Note that the “normally consolidation” region suggested by
Robertson (1990) can only been shown on the normalized SBT chart, since a similar region is only valid on the
non-normalized chart when the effective overburden stress is close 1 atmosphere (i.e. ~100kPa).

6
 1st International Conference on Civil Engineering for Sustainable Development (ICCESD-2012)

Figure 8 Comparison between (a) updated non-normalized SBT and (b) normalized SBTn (Robertson, 1990) for
the CPT profile shown in Figure 6

Figure 9: Seieve analysis for a Site

In order to identify the soil constitution, Grain size analysis have been performed for collected soil samples
where CPT test was performed. Figure 9 shows result of seieve test performed in the laboratory for site S1.

7
 1st International Conference on Civil Engineering for Sustainable Development (ICCESD-2012)

Figure 10: Factor of Safety for Six Locations

4. CONCLUSION

REFERENCES

Geologismiki Geotechnical Software, CPeT-IT v 1.6, 2009, CPT interpretation software,

http://www.geologismiki.gr/Products/CPeT-IT.html
Hossain, S., 2008, Rapid Urban Growth and Poverty in Dhaka City Bangladesh e-Journal of Sociology. V 5(1).
Molle, J., 2005. The accuracy of the interpretation of CPT-based soil classification methods in soft soils. MSc
Thesis, Section for Engineering Geology, Department of Applied Earth Sciences, Delft University of
Technology, Report No. 242, Report AES/IG/05-25, December.
Olsen, R.S. Cyclic liquefaction based on the cone penetration test. Proceedings, NCEER Workshop on
Evaluation of Liquefaction Resistance of Soils, National Center for Earthquake Engineering Research, State
University of New York at Buffalo, 1997; 225- 276.

8
 1st International Conference on Civil Engineering for Sustainable Development (ICCESD-2012)

P.K. Robertson, and R.G. Campanella, "Liquefaction potential of sands using the CPT", Journal of Geotechnical
Engineering, Vol. 111 No. 3, pp. 384-403, 1985.
Robertson, P.K., Campanella, R.G., Gillespie, D., and Greig, J., 1986. Use of Piezometer Cone data. In-Situ’86
Use of In-situ testing in Geotechnical Engineering, GSP 6 , ASCE, Reston, VA, Specialty Publication, pp
1263-1280.
Robertson, P.K., 1990. Soil classification using the cone penetration test. Canadian Geotechnical Journal, 27(1):
151-158.
Seed, H.B. and I.M. Idriss, “Simplified Procedure for Evaluating Soil Liquefaction Potential,” Journal of the
Soil Mechanics and Foundations Division, Vol. 97, No. SM 9, 1971, pp. 1249–1273.
Robertson, P.K., 2009. CPT interpretation – a unified approach, Canadian Geotechnical Journal, 46: 1-19
Youd, T.L., et al., “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, Vol. 127, No. 10, 2001, pp. 817–833.
Juang, C.H. and T. Jiang, “Assessing Probabilistic Methods for Liquefaction Potential Evaluation,” Soil
Dynamics and Liquefaction (GSP 107), American Society of Civil Engineers, Reston, Va., 2000, pp. 148–
162.
Juang, C.H., C.J. Chen, and T. Jiang, “Probabilistic Framework for Liquefaction Potential by Shear Wave
Velocity,” Journal of Geotechnical and Geoenvironmental Engineering, Vol. 127, No. 8, 2001, pp. 670–678.