Estimation of Degree of Consolidation

for Vacuum Preloading Projects

J. Chu1 and S. W. Yan2
Downloaded from ascelibrary.org by Institute of Infrastructure, Tech, Research and Management on 08/04/16. Copyright ASCE. For personal use only; all rights reserved.

Abstract: The degree of consolidation is usually used as one of the criteria for assessing the effectiveness of soil improvement work
using the fill surcharge or vacuum preloading method. It is also often used as a design specification in a soil improvement contract. Degree
of consolidation is normally calculated using settlement data. However, as the effect of vacuum preloading is controlled largely by pore
water pressure changes, it is necessary to analyze the pore water pressure variations and to assess the degree of consolidation using pore
water pressures. In this paper, the problems involved in the estimation of degree of consolidation using settlement data are discussed. A
method to estimate the average degree of consolidation using pore water pressure data is suggested. Two case studies are presented to
examine the characteristics of the pore water pressure variation of soil under vacuum loading. The degree of consolidation achieved in
each of the two cases is assessed using pore water pressure data and compared with that estimated using settlement data. Factors affecting
the degree of consolidation assessment are discussed.
DOI: 10.1061/共ASCE兲1532-3641共2005兲5:2共158兲
CE Database subject headings: Soil consolidation; Pore water pressure; Soil improvement; Preloading.

Introduction
One of the commonly used soft soil improvement methods is
vacuum preloading. This method has been successfully used in a
number of countries for land reclamation and soil improvement
work 共Holtz 1975; Choa 1990; Jacob et al. 1994; Bergado et al.
1998, 2002; Chu et al. 2000兲. Sand drains and recently prefabri-
cated vertical drains 共PVDs兲 have often been used to distribute
vacuum load and discharge pore water. The principles and mecha-
nism of vacuum preloading have been discussed in the literature,
e.g., Kjellman 共1952兲 and Holtz 共1975兲. A vacuum load of 80 kPa
or greater can be maintained as long as it is required. Compared
with the fill surcharge method for an equivalent load, the vacuum
preloading method is cheaper and faster 共Chu et al. 2000兲.
A major difference between fill surcharge and vacuum pre-
loading is in the pore water pressure change. Under fill surcharge,
the excess pore water pressure will first build up from its initial
共normally hydrostatic兲 state by the same amount as the surcharge,
and then dissipate gradually, as shown in Fig. 1共a兲. On the other
hand, under vacuum pressure, the pore water pressure in the soil
will reduce from its initial 共normally hydrostatic兲 state by the
same amount as the applied vacuum pressure, as shown in Fig.
1共b兲. As the pore water pressure can reduce to a negative value,
the pore water pressure changes due to the vacuum load are more
complicated. This is particularly the case when a combined fill
1
School of Civil and Environmental Engineering, Nanyang
Technological Univ., 50 Nanyang Ave., Singapore 639798. E-mail:
cjchu@ntu.edu.sg.
2
Geotechnical Research Institute, Tianjin Univ., Tianjin, China.
Note. Discussion open until November 1, 2005. Separate discussions
must be submitted for individual papers. To extend the closing date by
one month, a written request must be filed with the ASCE Managing
Editor. The manuscript for this paper was submitted for review and pos-
sible publication on March 26, 2004; approved on October 25, 2004. This
paper is part of the International Journal of Geomechanics, Vol. 5, No.
2, June 1, 2005. ©ASCE, ISSN 1532-3641/2005/2-158–165/$25.00.
surcharge and vacuum load are applied. Therefore, for vacuum
preloading projects, the pore water pressure variation during con-
solidation should always be monitored.
In addition to pore water pressure, the ground settlement is
also usually monitored and used to calculate the degree of con-
solidation 共DOC兲. DOC is an important parameter in evaluating
the effectiveness of soil improvement. It is also often used as a
design specification in a soil improvement contract. DOC is nor-
mally calculated as the ratio of the current settlement to the ulti-
mate settlement. However, for a soil improvement project, the
ultimate settlement is unknown and has to be predicted. Several
methods are available for estimating the ultimate settlement.
Among them, Asaoka’s 共1978兲 and hyperbolic 共Sridharan and
Rao 1981兲 methods are commonly used. In Asaoka’s method, a
series of settlement data 共S1 , . . . , Si−1 , Si , Si+1 , . . . SN兲 which are ob-
served at constant time intervals are plotted in a Sn versus Sn−1
plot 共n = 1 , . . . , N兲. The ultimate settlement, Sult, is taken as the
intersecting point of the line with the 45° line 共Asaoka 1978兲, as
illustrated in Fig. 2. However, Sult obtained from Asaoka’s method
is affected by the time interval used. Matyas and Rothenburg
共1996兲, Bo et al. 共1999兲, and Goi 共2004兲 have shown that the
longer the time interval, the smaller the Sult predicted. In the hy-
perbolic method, settlement data are plotted as time/settlement
versus time curve 共Sridharan and Rao 1981兲. The Sult is estimated
as the inverse of the linear slope of the plot. However, Sult ob-
tained from this method is affected by the DOC achieved when
the last set of data was taken. The higher the DOC that the soil
has attained, the smaller the Sult obtained as observed by Matyas
and Rothenburg 共1996兲, Bo et al. 共1999兲, and Goi 共2004兲. Goi
共2004兲 also shows that the plot is not strictly a straight line and
thus, the value of the linear slope can be different if the slope is
taken from different sections along the curve. A smaller Sult is
obtained when the slope is taken from the end of the curve. The
uncertainties involved in the ultimate settlement calculation will
affect the estimation of the DOC. As a result, different DOCs are
obtained using different methods.
As an alternative, pore water pressure data can be used to

158 / INTERNATIONAL JOURNAL OF GEOMECHANICS © ASCE / JUNE 2005

Int. J. Geomech., 2005, 5(2): 158-165

Downloaded from ascelibrary.org by Institute of Infrastructure, Tech, Research and Management on 08/04/16. Copyright ASCE. For personal use only; all rights reserved.

Fig. 1. Schematic illustration of pore water pressure and effective stress changes under 共a兲 fill surcharge and 共b兲 vacuum load

assess the DOC. The pore water pressure dissipation ratio can be
calculated easily as the ratio between the amount of pore water
pressure dissipation to the initial pore water pressure, 关ui
− u共t兲兴 / ui, where ui = initial pore water pressure and u共t兲 = pore
water pressure at time t. However, this ratio indicates only the
DOC of a soil element, not the average DOC. To calculate the
average DOC, the pore water pressure distribution over the entire
soil depth needs to be established. In this paper, a method to
estimate the average degree of consolidation based on the pore
water pressure distribution is suggested. Two case studies are pre-
sented to examine the pore water pressure dissipation character-
istics under vacuum preloading. The degree of consolidation
achieved in each of the two cases is assessed using pore water
pressure data and compared with that estimated using settlement
data. Factors affecting the DOC assessment are discussed. Al-
though only vacuum preloading cases are discussed, the method
and conclusions derived from this study are generally applicable
to all soil improvement projects involving either vacuum or fill
surcharge preloading.

Degree of Consolidation Based on Pore Water

Pressure

Once the pore water pressures at different depths are measured

Fig. 2. Schematic illustration of Asaoka’s method during preloading, the initial and final pore water pressure distri-

INTERNATIONAL JOURNAL OF GEOMECHANICS © ASCE / JUNE 2005 / 159

Int. J. Geomech., 2005, 5(2): 158-165

Downloaded from ascelibrary.org by Institute of Infrastructure, Tech, Research and Management on 08/04/16. Copyright ASCE. For personal use only; all rights reserved.
Fig. 3. Schematic illustration of pore water pressure distributions
versus depth under both combined surcharge and vacuum load
butions with depth can be plotted 共Chu et al. 2000兲. For general-
ity, a combined fill surcharge and vacuum load case is considered.
The typical pore water pressure distribution profiles for a com-
bined loading case are shown schematically in Fig. 3. Using this
profile, the average DOC, Uavg, can be calculated as
冕
关ut共z兲 − us共z兲兴dz
共1兲
冕
Uavg = 1 −
关u0共z兲 − us共z兲兴dz
where
us共z兲 = ␥wz − s 共kPa兲
In Eq. 共1兲, u0共z兲 = initial pore water pressure at depth z; ut共z兲
= pore water pressure at depth z at time t; us共z兲 = suction line, z
= depth; ␥w = unit weight of water; and s = suction applied. The
Fig. 5. Elevation view of instrumentation in Case I
160 / INTERNATIONAL JOURNAL OF GEOMECHANICS © ASCE / JUNE 2005
Int. J. Geomech., 2005, 5(2): 158-165
Fig. 4. Project site and plan view of instrumentation for Case I
value of s is normally assumed to be 80 kPa. The integral in the
numerator in Eq. 共1兲 is the area between the curve ut共z兲 and the
suction line us共z兲, and the integral in the denominator the area
between the curve u0共z兲 and the suction line us共z兲.
The method shown in Eq. 共1兲 has the following advantages
over the method using settlement data: 共1兲 The DOC calculated
using Eq. 共1兲 relies only on field pore water pressure data,
whereas when calculating the DOC using settlement data, the
ultimate settlement has to be predicted; 共2兲 Not only the final
DOC, but also the DOC at any time can be calculated using Eq.
共1兲, as ut共z兲 represents the pore water pressure at any time, t; and
共3兲. For consolidation involving multiple layers, Eq. 共1兲 can be
applied to any single layer to calculate the DOC achieved in a
particular layer. In this case, the upper and lower limits of the
integrals in Eq. 共1兲 are set to be the top and bottom of that soil
layer. However, it is not easy to calculate the DOC for each layer
for multilayer soils using settlement, as the settlement of each
Downloaded from ascelibrary.org by Institute of Infrastructure, Tech, Research and Management on 08/04/16. Copyright ASCE. For personal use only; all rights reserved.

Fig. 6. Settlement measured at different depths plotted against

duration of vacuum loading

layer may not be monitored directly and the ultimate settlement of

each layer has to be predicted, too.

Case Studies

Two case studies are presented to examine the characteristics of

pore water pressure variation in vacuum preloading projects and Fig. 8. Pore water pressure distributions at different durations for
to assess the DOC for these cases using pore water pressure data. Section I

Case I: An Oil Storage Station

intervals during the entire consolidation process. Owing to page
An oil storage station was constructed in 1996 near the coast of
limit, only the data collected for Section I are presented.
Tainjin, China, on a site that was recently reclaimed using clay
Under vacuum load, the settlement and pore water pressure
slurry dredged from the seabed. The site for the oil storage station
reductions measured at different depths at Section I are plotted
is shown in Fig. 4. It covered a total area of approximately
against duration in Figs. 6 and 7, respectively. It can be seen from
50,000 m2. For the purpose of soil improvement, the site was
Fig. 6 that the ground had settled nearly 900 mm toward the end
divided into two sections: Section I of 30,000 m2 and Section II
of consolidation. Fig. 7 shows that at the end of vacuum loading,
of 20,000 m2, as shown in Fig. 4. The soil profile consisted of a
the pore water pressures had reduced by 70– 140 kPa. Using the
6-m-thick very soft consolidating slurry clay layer and a 16 m
data presented in Fig. 7, the initial and final pore water pressure
soft silty clay layer overlying a stiff sandy silt layer. The soil in
distributions together with the hydrostatic pore water pressure
both layers had high water content and was very soft. Vacuum
line and the suction line are plotted versus depth in Fig. 8. The
preloading was adopted to improve the two layers of soft clay.
initial pore water pressures were greater than the hydrostatic pore
For more information on the soil properties and the soil improve-
water pressure as the subsoil was still undergoing consolidation.
ment procedure, see Chu et al. 共2000兲. A vacuum pressure of
The pore water pressure distributions at 30, 60, and 90 days are
80 kPa was applied continuously for 125– 145 days.
also plotted in Fig. 8 to illustrate the pore water pressure dissipa-
A field instrumentation and monitoring program was imple-
tion process. It can be seen that the final pore water pressure
mented in this project. Instruments including surface settlement
distribution is very close to the suction line. This is indicative that
plates, deep settlement gauges and pore water transducers were
the vacuum preloading method was very effective throughout the
used as shown in Figs. 4 and 5. Data were recorded at frequent
entire soft clay. Applying Eq. 共1兲 to Fig. 8, the DOC calculated at
the end of the project was 95%.

Fig. 7. Pore water pressure reductions plotted against duration of

vacuum loading Fig. 9. Project site and plan view of instrumentation for Case II

INTERNATIONAL JOURNAL OF GEOMECHANICS © ASCE / JUNE 2005 / 161

Int. J. Geomech., 2005, 5(2): 158-165

Downloaded from ascelibrary.org by Institute of Infrastructure, Tech, Research and Management on 08/04/16. Copyright ASCE. For personal use only; all rights reserved.

Fig. 10. Applied vacuum pressure and ground settlement measured with duration

Fig. 11. Settlement measured at different depths plotted against duration of vacuum loading at: 共a兲 Section I and 共b兲 Section II

162 / INTERNATIONAL JOURNAL OF GEOMECHANICS © ASCE / JUNE 2005

Int. J. Geomech., 2005, 5(2): 158-165

Downloaded from ascelibrary.org by Institute of Infrastructure, Tech, Research and Management on 08/04/16. Copyright ASCE. For personal use only; all rights reserved.

Fig. 12. Pore water pressure reductions measured at different depth plotted against duration of vacuum loading at: 共a兲 Section I and 共b兲
Section II

Case II: Road on Soft Ground
A section of a road leading to a container terminal at Tianjin Port,
China, had to be constructed on a 20-m-thick soft clay layer. The
top 5 – 6 m of the clay layer was reclaimed recently using clay
slurry dredged from seabed. The rest 14– 15 m was original sea-
bed clay. The vacuum preloading method was adopted in this
project for soil improvement.
The section of the road to be improved is schematically shown
in Fig. 9. It was 364.5 m long and 51 m wide. For the conve-
nience of construction, the site was divided into two sections. The
soil properties and the soil improvement procedure were similar
to those in the first case. A vacuum pressure of 80 kPa was ap-
plied continuously for 90 days. The ground settlement induced by
the vacuum preloading is shown in Fig. 10.
The settlement measured at different depths are plotted against
duration in Figs. 11共a and b兲 for Sections I and II, respectively.
The pore water pressure reductions are plotted in Figs. 12共a and
b兲 for Sections I and II, respectively. Using the data shown in Fig.
12, the pore water pressure distributions with depth at durations
of 30, 60, and 90 days can be plotted in Figs. 13共a and b兲 for
Sections I and II, respectively. The initial pore water pressure
profile, u0共z兲, and the suction line, us共z兲, are also plotted in Fig.
13. The initial pore water pressures were greater than the hydro-
static pore water pressure as the subsoil was still under consoli-
dation. Applying Eq. 共1兲 to Fig. 13, the DOC calculated at
90 days were 73% for Section I and 75% for Section II.

INTERNATIONAL JOURNAL OF GEOMECHANICS © ASCE / JUNE 2005 / 163

Int. J. Geomech., 2005, 5(2): 158-165

 Comparison

The DOC was also estimated based on the monitored settlement

data for the two cases. Asaoka’s method was adopted for both
cases to calculate the ultimate ground settlement, Sult. Using the
Sult calculated, the DOCs at durations of 30, 60, and 90 days and
at the end of the project are calculated. Using the pore water
pressure distribution curves shown in Figs. 8 and 13, the DOCs at
Downloaded from ascelibrary.org by Institute of Infrastructure, Tech, Research and Management on 08/04/16. Copyright ASCE. For personal use only; all rights reserved.

the same durations are also calculated. The DOCs calculated

using the two different methods are compared in Table 1. The
comparison shows that the DOCs estimated at different durations
using the settlement data are generally greater than those using
the pore water pressure data. It is also reported in other studies
共Holtz and Broms 1972; Hansbo 1997; Bo et al. 1999; Bergado et
al. 2002兲 that the DOC estimated using settlement is higher than
that using pore water pressure. The difference can be partially
explained by the way settlements or pore water pressures were
measured. Owing to budget constraints, only a limited number of
instruments were used in both cases. The settlement gauges were
installed only at the center locations where the maximum settle-
ment might take place. As a result, the ground settlement mea-
sured would be greater than the average ground settlement. Fur-
ther, the settlement measurements can be affected by secondary
consolidation. In Case II, as the width of the embankment is
limited compared to the thickness of the compressible soil layer,
the settlement measured may also include immediate settlement.
All the above factors lead to a gross overestimation of the settle-
ment. Thus, the DOC tends to be overpredicted using settlement
data. On the other hand, pore water pressure transducers were
installed at the middle point between the drains, which was the
furthest point away from the drain. Further away from the drain,
the higher the pore water pressure, the highest pore water pres-
sures were likely measured. Consequently, the DOC estimated
using the pore water pressures was likely to be under predicted.
Further, during the compression and rearrangement of the soil
structure, the excess pore water pressure may maintain at higher
levels similar to the observations of Holtz and Broms 共1972兲 and
Bo et al. 共1999兲. This explains why the DOC estimated using pore
water pressure data are smaller than that using settlement data.
However, when most of the pore water pressures have dissipated,
that is, when more than 85% of the DOC has been achieved, the
changes in both settlement and pore water pressure may become
smaller. In this case, the effect of uncertainties in the settlement or
pore water pressure measurements on the estimation of the DOC
tends to be less significant. Thus the differences between the
DOC measured using settlement data and that using pore water
pressure data become smaller. This explains why a relatively good
agreement has been achieved in Case I at a duration of 90 days
and at the end of the project.
It should be mentioned that both methods can only be as reli-
able as the field monitoring data. Any uncertainties involved in
the field pore water pressure and settlement measurements will
inevitably affect the DOC estimation. Further, when pore water
pressures are only measured at a limited number of points, the
spatial pore water pressure distribution cannot be constructed. In
this case, the pore water pressure distribution profile established
for one section has to be assumed to be the same as that at other
sections. This may not be the case, although with the use of PVDs
the pore water pressure distributions tend to even out. For the
DOC estimated based on settlement, there is one more source of
uncertainty, that is, the uncertainties involved in the ultimate
Fig. 13. Pore water pressure distributions against depth at different
settlement prediction. In view of the various uncertainties in-
durations at: 共a兲 Section I and 共b兲 Section II
volved in the DOC calculation, it is recommended to estimate the

164 / INTERNATIONAL JOURNAL OF GEOMECHANICS © ASCE / JUNE 2005

Int. J. Geomech., 2005, 5(2): 158-165

 Table 1. Comparison between Degree of Consolidation Calculated Using Pore Water Pressure and that Using Settlement
Based on settlement data
Degree of
consolidation 30 60 90
Case I 45% 80% 88%
共Section I兲
Case II 54% 77% 86%
共Section I兲
Downloaded from ascelibrary.org by Institute of Infrastructure, Tech, Research and Management on 08/04/16. Copyright ASCE. For personal use only; all rights reserved.
Case II 49% 66% 83%
共Section II兲
DOC using both settlement and pore water pressure data. Even if
the DOC is to be calculated using settlement data, the pore water
pressure distribution profile provides a way to visualize whether
the pore water pressure dissipation is consistent with the DOC
calculated based on settlement. If the differences between the two
measurements are too large and cannot be explained, the results
should then be examined before they are accepted. For contract-
ing purpose, it will be necessary to specify clearly whether the
DOC should be evaluated based on settlement or pore water pres-
sure or both to avoid future dispute. If the specifications regarding
the DOC are clearly spelled out, contractors may feel the field
instrumentation and monitoring tasks more meaningful. This may
in turn improve the quality of the field monitoring data.
Conclusions
Several aspects related to the assessment of the DOC for vacuum
preloading projects are discussed in this paper. Based on the
methods reviewed and the case studies presented, the following
conclusions can be made.
1. The DOC is often used to assess the effectiveness of soil
improvement work using vacuum preloading. As settlement
or pore water pressure data are required to assess the DOC,
field instrumentation and monitoring are essential for
vacuum preloading projects.
2. Settlement data are often used to predict the ultimate settle-
ment and calculate the DOC. However, the DOC estimated
in this way is affected by the methods used for predicting the
ultimate settlement.
3. Using the monitored pore water pressure data, the pore water
pressure distribution versus depth profiles can be plotted for
the initial, final, and any intermediate states. The DOC can
be estimated based on the pore water pressure profiles using
the method suggested in this paper.
4. The applicability of the suggested method to different types
of vacuum preloading projects is shown using the two case
studies. It is observed from the case studies that the DOC
estimated using settlement data is generally greater than that
using pore water pressure data. This can be partially ex-
plained by the fact that when only limited instruments can be
used, settlement and pore water pressure gauges will be in-
stalled only at the locations where the maximum settlement
and pore water pressure will be likely to occur. As a result,
the DOC tends to be overestimated when settlement data are
used and underestimated when pore water pressure data are
used.
5. For contracting purpose, it is necessary to specify the method
used to calculate the DOC and indicate clearly whether the
INTERNATIONAL JOURNAL OF GEOMECHANICS © ASCE / JUNE 2005 / 165
Int. J. Geomech., 2005, 5(2): 158-165
Based on pore water pressure data
End 30 60 90 End
95% 38% 73% 84% 92%
90% 54% 69% 73% —
90% 54% 66% 75% —
DOC is to be estimated using settlement or pore water pres-
sure data. It is recommended to work out the DOC based on
both methods. Even if the DOC is to be calculated using
settlement data, the pore water pressure distribution profile
provides a way to visualize whether the pore water pressure
dissipation agrees with the DOC calculated based on settle-
ment.
References
Asaoka, A. 共1978兲. “Observational procedure of settlement prediction.”
Soils Found., 18共4兲, 87–101.
Bergado, D. T., Balasubramaniam, A. S., Fannin, R. J., and Holtz, R. D.
共2002兲. “Prefabricated vertical drains 共PVDs兲 in soft Bangkok clay: A
case study of the new Bangkok International Airport project.” Can.
Geotech. J. 39共2兲, 304–315.
Bergado, D. T., Chai, J. C., Miura, N., and Balasubramaniam, A. S.
共1998兲. “PVD improvement of soft Bangkok clay with combined
vacuum and reduced sand embankment preloading.” Geot. Eng., J.
Southeast Asian Geot. Soc., 29共1兲, 95–122.
Bo, M. W., Chu, J., and Choa, V. 共1999兲. “Factors affecting the assess-
ment of degree of consolidation.” Proc., 5th Int. Symp. on Field Mea-
surements in Geomechanics, Singapore, Balkema, Rotterdam, The
Netherlands, 481–486.
Choa, V. 共1990兲. “Soil improvement works at Tianjin East Pier project.”
Proc., 10th Southeast Asian Geotechnical Conf., Taipei, 1, 47–52.
Chu, J., Yan, S. W., and Yang, H. 共2000兲. “Soil improvement by the
vacuum preloading method for an oil storage station.” Geotechnique,
50共6兲, 625–632.
Goi, M. H. 共2004兲. “Use of stabilised sewage sludge for land reclama-
tion.” MEng Thesis, Nanyang Technological Univ., Singapore.
Hansbo, S. 共1997兲. “Practical aspects of vertical drain design.” Proc.,
14th Int. Conf. or Soil Mechanics and Foundation Engineering, Ham-
burg, Germany, 3, 1749–1752.
Holtz, R. D. 共1975兲. “Preloading by vacuum: Current prospects.” Trans-
portation Research Record 548, Transportation Research Board,
Washington, D.C., 26–79.
Holtz, R. D., and Broms, B. B. 共1972兲. “Long term loading tests at Ska-
Edeby, Sweden.” Proc., Speciality Conf. on the Performance of Earth
and Earth-Supported Structures, Lafayette, Ind., 1, 435–464.
Jacob, A., Thevanayagam, S., and Kavazanjian, E. 共1994兲. “Vacuum as-
sisted consolidation of a hydraulic landfill.” Geotechnical Special
Publication No. 40, ASCE, Reston, Va., 1249–1261.
Kjellman, W. 共1952兲. “Consolidation of clayey soils by atmospheric pres-
sure.” Proc., Conf. on Soil Stabilization, Massachusetts Institute of
Technology, Boston, 258–263.
Matyas, E., and Rothenburg, L. 共1996兲. “Estimation of total settlement of
embankments by field measurements.” Can. Geotech. J., 33, 834–
841.
Sridharan, A., and Rao, S. 共1981兲. “Rectangular hyperbola fitting method
for one dimensional consolidation.” Geotech. Test. J., 4共4兲, 161–168.