See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/268467940

Ground Improvement Case Histories and Advances in Practice

Conference Paper · January 2012

DOI: 10.3850/978-981-07-3559-3_103-0002

CITATIONS READS

4 1,547

2 authors:

Serge Varaksin Babak Hamidi

Menard SNC Independant Consultant
73 PUBLICATIONS 255 CITATIONS 32 PUBLICATIONS 137 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Serge Varaksin on 07 March 2023.

The user has requested enhancement of the downloaded file.

 GROUND IMPROVEMENT CASE HISTORIES AND
ADVANCES IN PRACTICE

Serge Varaksin1 and Babak Hamidi2

1 ISSMGE TC211, Paris, France. E-mail: [email protected]
2 GFWA, Perth, Australia. E-mail: [email protected]

Ground improvement methods have made considerable advances since today’s

commonly practiced techniques first began to develop in the 20th century; however
they continue to remain a mixture of empirical methods devised by the practition-
ers and analytical solutions developed by the academicians. Advances in ground
improvement have been through the invention and development of new meth-
ods, influence of equipment, quality control methods and monitoring of ground
improvement works, and innovations for treatment of particular soils. While analyt-
ical and empirical methods are still widely practiced and form the basis of many
calculations, especially for ground improvement technologies that do not require
introduction of material, commercially available geotechnical software have revo-
lutionised the concept of design, facilitated calculations and design methods and
have allowed ground improvement specialists to design concepts that were previ-
ously only attainable, if at all possible, through cumbersome and exhausting hand
calculations that required excessive amounts of time. However, these advance-
ments are highly dependent on the quality and availability of soil parameters, and
results may be erroneous if proper values are not used as input data. At the same
time the philosophy of quality control in ground improvement has also advanced,
and the application of performance specifications based on design criteria has
allowed the optimisation of ground improvement works. These advances shall be
highlighted in this paper through presentation of several case studies.
Keywords: Ground improvement, Soil improvement.

1. INTRODUCTION

Ground improvement, as we would know it by its modern definition, began to take the
form of a branch of geotechnical engineering in the mid 20th century, and was finally
realised as the 17th technical committee of ISSMGE many years ago. Since then, this com-
mittee that has now been renamed TC211 has continuously tried to expand the domain of
this knowledge by making information available to the geotechnical engineer through its
numerous technical meetings, seminars, conferences, publications and the latest State of
the Art Report (Chu, et al., 2009).
While it may not be immediately apparent, ground improvement methods have made
considerable advances since today’s commonly practiced techniques first began to develop
in the first half of the 20th century; however most techniques have gone through changes,

Proceedings of the International Conference on Ground Improvement and Ground Control

Edited by Buddhima Indraratna, Cholachat Rujikiatkamjorn and Jayan S. Vinod
Copyright © 2012 by Research Publishing Services. All rights reserved.
ISBN: 978-981-07-3559-3 :: doi:10.3850/978-981-07-3559-3 103-0002 209
210 Proceedings of the International Conference on Ground Improvement and Ground Control

mostly due to new ideas, advances and innovations in equipment and technological capa-
bilities and the emergence of newer technologies has provided the geotechnical engineer
with additional tools for optimising foundation design and treatment of particular soils.
Progress in ground improvement has not been limited only to construction methods;
the philosophy of quality control methods has also evolved into a robust procedure that
allows the optimisation of design by visualising the foundation system as a mass rather
than multiple layers of soil that have to meet a minimum test value individually (Hamidi
et al, 2011). Testing and monitoring devices have also been improved and live monitor-
ing of records that can be transmitted electronically to the engineer located thousands of
kilometres away has improved communications and minimised data interpretation time.
While analytical and empirical methods are still widely practiced and form the basis of
many calculations, design methods have been revolutionised, mostly as a result of quan-
tum leaps in computer technology and software that enable the engineer to model static
and dynamic loads, ground geometry, physical and mechanical properties, and interaction
with highly reliable accuracy.

2. ADVANCES IN IMPACT TECHNIQUES: DYNAMIC COMPACTION

Louis Menard invented and promoted dynamic compaction as early as 1969 but it was not
until 29 May 1970 that he officially patented his invention in France (Hamidi et al., 2009).
The concept of this technique is improving the mechanical properties of the soil by trans-
mitting high energy impacts to the soil by dropping a heavy pounder from a significant
height.
When feasible, dynamic compaction is probably the most favourite ground improve-
ment technique in granular soils as it is usually the most economical soil improvement
solution (Mitchell, 1981).
Depth of influence or improvement is the depth where there is limited or practically
insignificant amounts of improvement in the soil. Menard and Broise (1975) developed an
empirical equation in which the depth of influence, D, was a function of the square root
of the impact energy; i.e. the product of the pounder weight (in metric tons) by the drop
height (in metres). Later and based on further site experiences others introduced a coef-
ficient less than unity to the original equation and Varaksin (Chu et al., 2009) has further
refined the relationship by introducing drop type and energy function coefficients.
Hamidi et al. (2011a) have reviewed the equipment advances of dynamic compaction
rigs. Menard performed his first dynamic compaction projects using 80 kN pounders that
were dropped from 10 m. He was soon able to identify heavy duty cranes that were capa-
ble of efficiently lifting and dropping pounders weighing up to about 150 kN using a
single cable line. Menard then developed and manufactured his own rigs that were able to
lift 250 and more than 1,700 kN pounders. As much as these special rigs had their appli-
cations, they were specifically produced, their numbers were limited and they could not
be manufactured commercially or in great numbers. However, the introduction of a new
generation of cranes that are able to lift pounders using two single cable lines has now
increased lift capacity commercially to 250 kN. The introduction of these rigs was able
to increase pounder lift capacity; however it is still possible to improve the efficiency of
impact energy by dropping the pounder in free fall. Thus, the next major innovation in
dynamic compaction was the development of the Menard Accelerated Release System
 Ground Improvement Case Histories and Advances in Practice 211

Figure 1. Automatic pounder release and grab MARS technology.

(MARS) which is able to release the pounder from the lifting device as the pseudo free
fall commences. This is shown in Figure 1. Digital monitoring instruments are now able
to record the coordinates of the impact point, drop height, number of drops per point and
impact velocity. This enables the engineer to improve quality assurance and optimisation of
work parameters.

2.1. Case Study: King Abdulla University of Technology

King Abdulla University of Science and Technology (KAUST), located near the city of Jed-
dah in Saudi Arabia, is a unique project where the planning, design and construction were
undertaken simultaneously in order to handover the project during a period of less than
three years (Chu et al., 2009; Varaksin and Yee, 2009; Hamidi et al., 2010). As design and con-
struction were performed together the project entailed site preparation involving ground
improvement works for structures on non-defined locations. Due to the heterogeneous
ground conditions and the randomly placed structures, a concept for the works was for-
mulated was based on field observation method and re-confirmed by CPT and PMT.
The preliminary geotechnical investigation that was carried out rather sparingly indi-
cated that the ground was very heterogeneous loose or soft soils with rapid variations
within short distances. This investigation and further testing during the works indicated
that more than 2,600,000 m2 of the construction area was to be built on soil consisting of
up to 9 m of loose silty sand or soft sandy silt locally called sabkah. Groundwater was at
the depth of 1 m.
Although building and thus footing locations were not determined, it was anticipated
that KAUST would have structures with at most two to three storeys. Nevertheless these
loads were beyond the soil’s capacity, and the engineers realised that specific foundation
measures had to be implemented immediately to support the structural loads without hin-
dering the programme.
212 Proceedings of the International Conference on Ground Improvement and Ground Control

The design and construct ground improvement proposal that met the project manager’s
technical requirements, schedule and budget was based on the below design criteria:

• Footing location: Any place within the treatment area

• Maximum footing load: 1,500 kN
• Allowable bearing capacity: 200 kPa
• Maximum total settlement: 25 mm
• Maximum differential settlement between two adjacent footings: 1/500
• Liquefaction mitigation for an earthquake with PGA = 0.07g

As a concept of design it was also further accepted that footings would be 0.8 m below
final ground level, but in any case at least 2 m above sabkah level.
The construction method was a combination of dynamic compaction and dynamic
replacement with pounders weighing up to 21 tons. Dynamic replacement was used in
areas where the maximum depth of sabkah was 5 m. High energy dynamic replacement
was used when the sabkah layer’s depth was more than 5 m. In such cases, in addition to
the engineered fill required for reaching final ground level, a 3 m high surcharge was also
placed over the treatment area for 3 weeks.
After completion of ground treatment in some areas, changes in master plan necessitated
additional ground improvement to support a number of six storey buildings. Thus, addi-
tional dynamic surcharging was used to consolidate the deep sabkah layers by a combina-
tion of surcharging and vibration to re-introduce pore pressure in the soil-water system and
to accelerate settlement rates. In addition to the engineered fill required for reaching final
ground level a 3 m high surcharge was placed and dynamic compaction was performed
on top of it.
In the absence of sufficient preliminary geotechnical testing, the proper application and
the extent of the ground improvement works were based on the field observation method;
i.e. visual inspection of the pounder penetration per blow. It was observed that while
the first dynamic compaction pounder impact penetrated the ground by about 0.25 m,
the dynamic replacement pounder penetration was substantially more and in the range
of about 1 m. Also, performing dynamic compaction frequently resulted in the seepage
of groundwater to the surface, but this phenomenon was rarely encountered in dynamic
replacement areas. Ground heave due to pounding was not observed in dynamic com-
paction areas but was quite observable in dynamic replacement zones.
The allocated time frame for mobilization, execution and testing of ground improvement
works was set at 10 months. Ground improvement was carried out over a period of 8
months using a total of 13 rigs working two shifts per day. To the knowledge of the authors,
this is the world record for the number of dynamic compaction/dynamic replacement rigs
working simultaneously on a single project.
Menard Pressuremeter tests (PMT) were carried out to confirm that the bearing and
settlements had been achieved. CPT were also performed for verification of liquefaction
mitigation. Figure 2 shows the relation between the net pressuremeter limit pressure
(PL − Po ) and the energy per volume of improved soil. As expected, it can be observed
from this diagram that the improvement factor (I ) or the post to pre soil improvement
net limit pressure ratio (only the in situ soil without the dynamic replacement columns)
is higher and more efficient when the soil is more granular and fines content, FC, is less.
 Ground Improvement Case Histories and Advances in Practice 213

Figure 2. The relationship between net limit pressure and improvement energy.

Furthermore, it can be demonstrated that even though the soil soils have improved param-
eters after treatment, nevertheless the criteria would not have been satisfied without the
introduction of the granular dynamic replacement columns.

2.2. Case Study: Marine Works in Southeast Asia

To the knowledge of the authors the deepest marine dynamic compaction or replacement
project has been carried out 30 m below seawater level in Southeast Asia (Hamidi et al.,
2010a).
The original seabed of the site was composed of very stiff clay with shear strength about
250 kPa and SPT blow counts exceeding 50. However dredging works and reducing the
seabed level to about −30 m CD (Chart Datum) disturbed the upper 1 to 1.5 m of the
superficial clay layer, and post dredging geotechnical tests indicated that the clay’s shear
strength after backfilling with sand had dropped to approximately 80 kPa. Further testing
at later stages using the PMT revealed that that the shear strength had even further reduced
from this value to only 16 kPa.
Thus the sand was removed and it was stipulated that the soft soil had to be treated. Due
to the nature of the soft soil and its thickness, ground improvement by dynamic replace-
ment was envisaged as a possible solution. Prior to commencement of dynamic replace-
ment 1.8 m of rock fill was placed over the soft seabed.
The 38.5 ton pounder used in this project was specially designed double sided pounder
to minimise water resistance during the pounder’s descent in the sea. As shown in
Figure 3, one side of the pounder had a smaller area and was designed shear into the soil
for dynamic replacement penetration. The other side of the pounder had a larger area, and
was intended for compacting the granular rock fill by dynamic compaction. Each dynamic
replacement print location was subject to 30 blows. Furthermore, 3 to 6 blows were applied
as the ironing phase over the entire treatment area.
The drop height of the pounder was set at 5 m above seabed. Digital records of the rig’s
winch speed during the works indicate that the maximum drop speeds achieved were in
the range of 430 m/min. This speed is the equivalent to a free fall with a drop height of
2.6 m and verifies that significant drop heights would not have increased the impact energy.
The pounder penetration into the ground was measured for every blow. It was observed
that while the pounder penetrated the ground at a more pronounced rate during the first
214 Proceedings of the International Conference on Ground Improvement and Ground Control

Figure 3. Specially designed double purpose pounder.

four blows, the penetration rate rapidly decreased to the point where it appears that practi-
cally no penetration was achieved after the 15th blow. During the first 15 blows, the amount
of pounder penetration was variable from 1.1 to 1.7 m.
The total ground settlement was measured by echo sounding and the survey showed
that the top of the blanket had dropped by 38 cm as a result of the ground improvement
works.
Divers were also sent to visually scrutinize the impact results. Based on the larger
amount of crushed rock at the impact locations they were able to confirm that the col-
umn diameters were 2.4 m. This figure is equal to the diagonal length of the pounder’s
base.
Due to the large water depth and open sea working conditions PMT were carried out by
using 100 mm guide tubes followed by the 60 mm PMT tube. During testing visual obser-
vation on the return of drilling fluid was recorded. When there was no return of drilling
fluid, it indicated that the test was carried out in free-draining rock material whereas test-
ing in impervious clay was indicated by the return of the drilling fluid. In this project
2 PMT were carried out prior to dynamic replacement and six were carried out after treat-
ment. It was observed that the limit pressure, PL , of the soft clay layer was less than 0.1 MPa
which indicates the very low strength of the material. The comparison of PMT that were
done before and after ground improvement in the almost same location indicates that while
the rock fill has been driven into the soft clay, its Menard Modulus, E M , and PL values have
also increased respectively by 118% and 132%.

3. ADVANCES IN VIBRATORY TECHNIQUES: VIBROCOMPACTION

Vibro Compaction is a deep compaction technique that was invented in the mid 1930s in
Germany for treating sandy soils. In this technique an electric or hydraulic vibrating unit
called a vibroflot or vibroprobe penetrates the ground, and densifies the loose sands.
Although the appearance of vibroflots have not changed much during the past seven
decades and most equipment would seem very similar to the untrained eye, today
 Ground Improvement Case Histories and Advances in Practice 215

specialist ground improvement companies manufacture vibroprobes with different capa-

bilities. Vibration frequencies are now closer to the soil’s natural frequency and the power
range of the plant allow specific uses of each machine.

3.1. Case Study: Peribonka Hydro Electric Dam

Peribonka Dam is a 80 m high earth dam that has been constructed on 100 m of allu-
vial deposits of Peribonka River in Quebec, Canada (Lauzon et al., 2006). The geotechnical
investigation indicated that the alluvial soils that were mainly sand to gravelly sand, and
also gravel with some sand, cobbles and occasional small boulders were loose and did not
meet the geotechnical engineer’s requirements down to a maximum depth of 35 m from
the riverbed. Backfilling the riverbed with loose sand and gravel further thickened the
thickness of the loose layers to a maximum depth of 52 m.
Although it was envisaged to perform vibrocompaction for densifying the loose granular
soils prior to the placement of the backfill, and compacted bands under the truck paths
were loosened up to allow the penetration of vibroflots, testing at later stages demonstrated
that the soil had nevertheless become dense.
Performing vibrocompaction became an evident challenge in this project due to the pres-
ence of large size granular material, presence of dense layers of soil before final treatment
depth and the requirement to treat a maximum depth of 52 m that to the knowledge of
Lauzon et al., was the world record for treating in-situ soils using the vibrocompaction
technique. The challenge becoming more concerning when it was observed that conven-
tional vibroflots were meeting refusal before the target depth in about 60% of the site, and
production was progressing significantly slower than expected.
Consequently, a second specialist geotechnical contractor who had manufactured the
powerful V48 vibroflot. This 26 MN vibroprobe with 175 kW of power had an amplitude
of up to 48 mm and a frequency of 1200 to 1500 Hz. This machine demonstrated its power
when it proved to be able to efficiently penetrate all layers down to the maximum specified
depth of 52 metres.

4. ADVANCES IN CONSOLIDATION TECHNIQUES: VACUUM CONSOLIDATION

The idea of vacuum consolidation was proposed 60 years ago (Kjellman, 1952); however
practical use of this technology is more recent. This technique provides an effective alterna-
tive to surcharging and preloading soils whereas instead of increasing the effective stress
in the soil mass by increasing total stress by means of a conventional surcharge, in vacuum
consolidation the soil preloads the soil by reducing the pore pressure while maintaining a
constant total stress (Cognon, 1994).
In the Menard Vacuum System (Cognon, 1991) vertical drains are installed to the
required depth, and horizontal drains are placed in longitudinal and latitudinal directions
in the sand blanket. Impermeable isolation trenches or cut-off walls are placed around
the preload area and an impermeable membrane is used to seal off the ground surface.
Vacuum pumps are connected to the discharge module extending from the trenches, and
vacuum pressure is applied to the ground. The uniqueness of this system is the dewatering
below the membrane which permanently keeps a gas phase between the membrane and
the lowered water level (Chu et al., 2009).
216 Proceedings of the International Conference on Ground Improvement and Ground Control

4.1. Case Study: Port of Brisbane Paddock S3A

Port of Brisbane is the third largest container port in Australia, and the vacuum consoli-
dation project of this port’s Paddock S3A has been the subject of several papers (Berthier
et al, 2009; Ameratunga et al., 2010; Indraratna et al., 2011).
Port of Brisbane is located in the lower reaches of the Brisbane River on reclaimed land
at the mouth of the river. Land reclamation has been in progress since the early 1980s and
the current Port location, Fisherman Islands, is almost entirely constructed in the adjacent
Moreton Bay. In 1999, the port authority commenced the expansion programme of a 235ha
reclamation site from sub-tidal area.
The subsurface conditions in the this reclamation was significantly different from the
previously developed areas at Fisherman Islands whereas the normally consolidated clay
was much deeper and up to about 25 m thick in Paddock S3A. Reclamation was carried
out using channel maintenance and berth dredging materials resulting in approximately 7
to 9 m of mud being deposited on the original seabed and being capped off with approxi-
mately 1 to 5 m of sand.
The water contents of the soil layers were at or beyond their liquid limits and field vane
tests indicated that the undrained shear strength of the dredged mud and the in-situ clays
varied from 5 kPa to 60 kPa, the compression index varied from 0.1 to 1, the coefficient of
consolidation in vertical direction, Cv , was approximately the same as that in the horizontal
direction, Ch , for the totally remoulded dredged mud layer and Cv /Ch was about 2 in the
in-situ clay layer.
Consolidation period was predicted to be more than 50 years if surcharging was the only
treatment employed, and settlements were estimated to be in the range of 2 to 3 m. With
the intention of accelerating consolidation and mitigating instability of the fill boundary
the port authority awarded a design and construct contract to a specialist ground improve-
ment contractor for implementing vacuum consolidation in Paddock S3A with an area of
15,000 m2 .
Initial design criteria stipulated a residual settlement of less than 250 mm over a period
of 20 years under a uniformly distributed load of 15 kPa applied at final surface level.
Design criteria were later revised by reducing maximum settlement by 60% to 150 mm
under an increased load (by 67%) of 25 kPa.
Time dependent analysis of the primary consolidation was performed with consider-
ation of the paddock’s reclamation history, and the timing of the vacuum system and
surcharging. Vacuum pressure was assumed to be 70 kPa applied on the top of the ground.
Secondary consolidation and effect of pre-aging by over surcharging were also consid-
ered. Stability analyses of the southern and western boundaries was done using numerical
modelling.
The apparent cohesion in the sand blanket and the strengthening of the clay realised by
the vacuum process provided a safety factor of more than the minimum requirement of 1.3
to allow the safe surcharging of adjacent areas that were being treated by wick drains only.
Construction works were carried out during a period of 4 months. Circular drains that
due to more efficient pressure depression assist the vacuum pressure propagation more
effectively than band drains were installed in a square grid of 1.2 m to a maximum depth
of 35 m. As shown in Figure 4, using a long arm excavator, a 15 m deep soil bentonite cut-off
wall was also installed and extended 1.5 m into the in-situ clay layer to provide boundary
 Ground Improvement Case Histories and Advances in Practice 217

Figure 4. Construction of 15 m deep soil bentonite wall using a long arm excavator.

sealing due to the presence of permeable layers within the reclaimed material. This was
the first application of vacuum consolidation with a deep cut-off wall in Australia.
With consideration of a 70 kPa pressure realised by vacuum consolidation, additional
surcharge was provided by placement of only 2.2 m of surcharge. Later, the surcharge
height was increased by 0.5 over a limited area with the deepest clays to satisfy the new
design criteria.
Vacuum efficiency was measured using pressure gauges that were monitored on a daily
basis during the first three months and then once every other week. Three instrument sta-
tions, comprising vibrating wire piezometers in the ground and in selected, vertical trans-
mission pipes, extensometers and deep settlement plates were also installed in the vacuum
area. Additionally, two inclinometers were installed on the external boundary of the treat-
ment zone.
Within less than one week from commencement of vacuum consolidation, gauges indi-
cated an average depressurization of more than 70 kPa. Piezometers installed at different
depths in the vertical drains also showed a rate of decrease of the pore pressure in the same
range. Results also showed that vacuum pressure consistently reached depth with limited
head loss.
2.2 m of surcharge was placed two weeks later. Inclinometers did not show signifi-
cant movement during the surcharge placement on the wick drain area. Comparison of
maximum horizontal displacement to vertical displacement in the vacuum zone and the
wick drain only areas with similar ground conditions and applied pressures show that
the ratio was about 2 to 3 times more in the wick drain only areas and indicates that lat-
eral movements are well controlled by isotropic consolidation when vacuum pressure is
implemented.
The final back analyses of the monitoring shows that the degree of consolidation
achieved by the time the vacuum system was discontinued after 18 months varied from
92% to 97%.

5. ADVANCES IN TECHNIQUES WITH INCLUSIONS: JET GROUTING

Jet grouting is a method involving drilling down with a small diameter rod system, inject-
ing high pressure fluids while rotating and withdrawing the rod to erode soil and to mix
218 Proceedings of the International Conference on Ground Improvement and Ground Control

it with cement grout to form a rigid circular cementituous column. Jet grouting installa-
tion methods include single, double and triple fluid injection systems. In the single fluid
method neat cement grout that is injected through a small nozzle at high pressure is used
to break up the soil and to mix with it. In the double fluid method the cement grout is
aided by a concentric cone of compressed air which shrouds the grout injection. The air
reduces the friction loss and allows the cement grout to travel further from the injection
point thereby producing greater column diameters. In the triple fluid method water and
air break up the soil and cause partial substitution of the finer soil particles. Cement grout
is injected independently through a nozzle located beneath the air or water nozzle.
Jet grouting is a multipurpose geotechnical method and can be used for soil improve-
ment, construction of impermeable boundaries, and solidification of ground, and has
advanced in line with the improvements in drilling methods, pumps, and grout batching
facilities.

5.1. Case Study: Sydney International Airport Runway End Safety Area

The Runway End Safety Area (RESA) is an extension to one of Sydney International
Airport’s runways and has been built in a location that intersects with a heritage listed
sewer, a diverted airport perimeter road and a motorway tunnel. In order to construct
RESA a 90 × 90 m2 concrete platform bridges over the mentioned facilities. In addition
to the bridge construction, the perimeter road has been lowered and relocated to allow a
minimum clearance of at least 5 m under the proposed aircraft bridge. (Hamidi et al, 2012).
Ground surface was from approximately +3.5 to +1 m RL. The fill layer was composed
of a variation of loose to dense sand and silty sand extending down to approximately
−1.5 m RL. This layer was followed by soft to firm clay or very loose silty sand down to
about −8 m RL. qc of this layer was less than 1 MPa and sleeve friction ratio was generally
about 4%. The fine layer was underlain by 2 m of dense sand that extended to −10 m RL.
qc of this layer ranged from 25 to 35 MPa. The soil from −10 to −16 m RL was stiff or very
stiff clay with qc frequently being about 3 to 4 MPa and friction ratio being as high as 8%.
The soil in between the mentioned layer and bedrock at about -20 m RL was dense sand
with qc in the range of 25 to 35 MPa, and finally bedrock was medium to high strength
sandstone.
Groundwater level was at approximately RL +1.0 m and fluctuated with sea tides. This
level could have risen during periods of heavy rainfall due to substantial infiltration into
the exposed soils.
In order to span the bridge over the sewer, the motorway tunnel and the realigned air-
port perimeter road it was necessary to identify and implement geotechnical solutions for:

• Working in dry conditions.

• Excavation under the sewer with sufficient head room for positioning construction
equipment.
• Providing bearing for the formwork and temporary false work props of the new
support structure that would replace the existing sewer piles.
• Providing a sound working platform to build the underpass bottom slab.
• Ensuring full lateral support to the sides of the multiple excavation levels.
Horizontal movement had to be limited to 10 mm in normal working conditions
 Ground Improvement Case Histories and Advances in Practice 219

and 11 mm for accidental failure of the dewatering system and the rise of ground-
water level outside the cut-off area by 2 m.
• Providing the temporary support wall and foundation system of the new pumping
station and drainage lines.
• Stabilising the soft clay ground on the north and south sides of the peripheral wall
to facilitate construction of the underpass bottom slab.

Jet grouting was proposed as the solution to all of the above issues and a design and
construct contract was awarded to a specialist geotechnical contractor who had proposed
the solution.
A complex system of jet grouted columns were designed with different diameters, spac-
ing, and lengths to satisfy the various project requirements. As manual calculation meth-
ods for the numerous ground and load conditions were not possible during the allocated
time, sophisticated numerical analyses using the latest geotechnical software were carried
out to verify that stresses, ground deformations and global stabilities would remain within
acceptable limits during the different phases of construction. Multiple groundwater lev-
els were assumed for different situations. Structural verification of the peripheral wall jet
grouting columns was carried out for bending moment and shear modes, and consequently
steel bars were used for reinforcing the columns of the peripheral cut-off wall.
In addition to laboratory grout mixing and testing and prior to the commencement of jet
grouting production, site trials were performed to confirm that expected design parameters
were met.
Double fluid jet grouting system, sometimes preceded by pre-cutting, was used to install
a total of 1,628 columns. Replacement ratios ranged from 41% to 75%. Column diameters
were 1 to a 2.5 m, and column lengths were 2 to 13 m. Total drilling and column lengths
were respectively 13,761 and 6,368 m, and total jet grouting volume was 5,134 m3 .
As shown in Figure 5, Installing and recording of column positions was done using an
advanced GPS system that was able to provide the spatial coordinates of each point, and
drilling and jetting were monitored by a digital recording system. During drilling, depth,

Figure 5. Application of advanced GPS systems in jet grouting.

220 Proceedings of the International Conference on Ground Improvement and Ground Control

advancement speed, rotary speed and thrust pressure were recorded. Grout pressure, grout
flow, grout volume, stationary time, air pressure, air flow, uplift speed and rotation speed
were monitored during the jetting phase. The start and stop time for each columns was
also recorded. Furthermore, the grout density and viscosity were measured during each
working shift.
Works were carried out in two shifts using 3 rigs, one of which was a mini rig that was
able to operate in areas with low overheads. A discontinuation of jet grouting operations
was included in the schedule to allow for the first excavation, construction of sewer frame
support and cutting the redundant piles.

6. CONTROLLED MODULUS COLUMNS

Controlled Modulus Columns (CMC) technology was developed by Menard in 1994

(Chu et al., 2009). These rigid inclusions are installed into loose or soft ground using a
specially designed auger that displaces the soil laterally with practically no spoil or vibra-
tion to form a cylindrical cavity. The auger is screwed into the soil to the required depth
which, in granular soils, increases the density and thus the mechanical parameters of the
surrounding soil. During the auger extraction process grout is pumped through the hollow
auger. This results in a column with a diameter that is usually 250 to 450 mm. The strength
of the columns can be controlled by varying the strength of the grout.
Unlike stone columns whose stability relies on the horizontal retainment of the soil or
deep soil mixing where column strength is dependent on the in-situ soil’s properties, CMC
do not rely on external parameters for lateral stability nor are column strengths affected by
the surrounding soil. Thus this method can reduce settlements more efficiently compared
to other techniques in which inclusions are installed in the soil.

6.1. Case Study: Oil Tanks in Marrero

As part of the development programme of an oil terminal located on the banks of the
Mississippi River in New Orleans five steel tanks, each 12.8 m high are being constructed
(Buschmeier et al., 2012). Three tanks have diameters of 39.6 m and the other two tanks’
diameters are 45.7 m. The tanks’ will exert a maximum pressure of 130 kPa to the ground,
and elevating the ground level to tank level will impose an additional 16 kPa of pressure.
Below a superficial fill layer of approximately 0.15 to 1.2 m soft to medium stiff silty clays
with some trace of organic matter and localized sand pockets extend down to a depth of
4 to 6 m. This layer is followed by very soft clay with silt and sand which reach depths
of 20 to 24 m. A thin sand layer has also been identified at an approximate depth of 21 m.
Medium stiff to stiff clay with fine sand pockets and shell fragments are observed to depths
of up to 32 m followed by stiff to very stiff silty to sandy clays over a very dense layer of
silty sands at depth of about 34 m. Groundwater level is quite high and at less than 1 m
below ground level.
Project specification stipulated that the tanks’ maximum and central settlements be
limited to respectively 200 and 100 mm and additionally 50% of settlements allowed by
American Petroleum Institute three years after hydrotesting. However these criteria were
not satisfied due to the poor ground conditions and consequently a ground improvement
contract was awarded to a specialist geotechnical contractor.
 Ground Improvement Case Histories and Advances in Practice 221

Figure 6. Installation of 34 m long CMC.

As the soil investigation indicated a reduction in compressibility at the depth of about

21 m at the level of the thin sand layer it was decided to install CMC with a higher replace-
ment ratio down to the depth of the sand layer and a with a reduced replacement ratio
down to the maximum treatment depth at about 34 m.
Due to the variations in the site’s soil profile it was necessary to design each tank indi-
vidually. Analyses included three dimensional digital modelling of a quarter of a tank,
three dimensional modelling of a thin slice of the tank and manual calculations of rafts on
floating piles.
CMC diameters for columns installed to approximately 21 m and 34 m were respectively
318 and 470 mm (see Figure 6 for installation). Installation depth was variable for each tank
due to variations in the site’s soil profile.
To confirm design, a large scale zone load test was carried out on site on an area of 13.7 by
13.7 m2 loaded to the maximum design load. Vibrating wire piezometers, vibrating wire
rebar strain gages, multi depth settlement gages, horizontal extensometer, inclinometers
and settlement plates were used to measure ground behaviour for a period of slightly less
than three months. Maximum settlement measured at the top of the load transfer platform
was 107 mm for a settlement plate located in between two columns, and minimum settle-
ment was 64 mm for a settlement plate placed directly on top of a column at the bottom
of the load transfer platform level. Maximum differential settlement was measured to be
43 mm between a CMC and the grid centre.
Further numerical modelling was carried out to compare actual measurements with cal-
culated figures. Numerical analysis suggested a maximum settlement of 97 mm during the
trial period which was in good agreement with actual measurements.

7. CONCLUSION

Ground improvement has been consistently advancing in all aspects. These advances
are sometimes due to innovations and introduction of new soil improvement techniques,
and at other times the result of modifications to existing techniques or the improvement of
equipment. Introduction of digital systems into geotechnical works in general and
into ground improvement specifically has facilitated data logging, management and
222 Proceedings of the International Conference on Ground Improvement and Ground Control

interpretation. Digital technology has also advanced ground improvement design meth-
ods and allowed the analyses of special cases and conditions that would have been very
difficult to model using manual methods. However, this is highly dependent on the accu-
racy and availability of soil parameters that will be used as input data. Advances have not
only been limited to construction and design methods and tools, but have also been the
result changes to the concept of ground improvement criteria.

REFERENCES

1. Ameratunga, J., Boyle, P. and Berthier, D. (2010). Ground Improvement in Port of Brisbane (PoB)
Clay. ISSMGE Bulletin, 4(2) : 28–54.
2. Berthier, D., Boyle, P., Ameratunga, J. and Vincent, P. (2009). A Successful Trial of Vacuum
Consolidation at the Port of Brisbane. Coasts and Ports 2009 Conf., Wellington, 16–18 September.
3. Buschmeier, B., Masse, F., Swift, S. and Walker, M. (2012). Full Scale Instrumented Load Test
for Support of Oil Tanks on Deep Soft Clay Deposits in Louisiana using Controlled Modulus
Columns. Intl. Symp. on Ground Improvement (IS-GI) Brussels 2012, Brussels, 31 May - 1 June:
Accepted.
4. Chu, J., Varaksin, S., Klotz, U. and Mengé, P. (2009). State of the Art Report: Construction
Processes. 17th Intl. Conf. on Soil Mech. and Geotech. Engrg.: TC17 meeting ground improvement,
Alexandria, Egypt, 7 October 2009, 130.
5. Cognon, J. (1991) Vacuum Consolidation. Revue Française Géotechnique, 57(October): 37–47.
6. Cognon, J. M., Juran, I. and Thevanayagam, S. (1994) Vacuum Consolidation Technology - Prin-
ciples and Field Experience. Vertical and Horizontal Deformations of Foundations and Embankments:
ASCE GSP No. 40, 2, College Station, Texas, 16-18 June: 1237–1248.
7. Hamidi, B., Nikraz, H. and Varaksin, S. (2009). A Review on Impact Oriented Ground Improve-
ment Techniques. Australian Geomechanics J., 44(2): 17–24.
8. Hamidi, B., Nikraz, H. and Varaksin, S. (2011). Ground Improvement Acceptance Criteria. 14th
Asian Reg. Conf. on Soil Mech. and Geotech. Engrg., Hong Kong, 23–27 May: Paper No. 404.
9. Hamidi, B., Nikraz, H. and Varaksin, S. (2011a). Advances in Dynamic Compaction. Indian
Geotech. Conf. IGC2011, Kochi, India, 15–17 December: Paper No. H–146.
10. Hamidi, B., Varaksin, S. and Nikraz, H. (2010). Implementation of Optimized Ground improve-
ment techniques for a Giga Project. GeoShanghai 2010 Conf., ASCE GSP No 207: Ground Improve-
ment and Geosynthetics, Shanghai, 3–5 June: 87–92.
11. Hamidi, B., Yee, K., Varaksin, S., Nikraz, H. and Wong, L. T. (2010a). Ground Improvement in
Deep Waters Using Dynamic Replacement. ISOPE 2010, Beijing, 20-26 June: 848–853.
12. Hamidi, B., Yates, M., Berthier, D. and Krzeminski, M. (2012). Overcoming Challenges with
Jet Grouting at Sydney International Airport. 4th Intl. Conf. on Grouting and Deep Mixing, New
Orleans, Louisiana, 15–18 February.
13. Indraratna, B., Rujikiatkamjorn, C., Geng, X., Ameratunga, J. and Boyle, P. (2011). Performance
and Prediction of Vacuum Combined Surcharge Consolidation at Port of Brisbane. Coastal and
Marine Geotechnics: Foundations for Trade, Sydney, 12 October: 45–60.
14. Kjellman, W. (1952). Consolidation of Clay Soil by Means of Atmospheric Pressure. Conf. on Soil
Stabilization, Cambridge, Massachusetts: 258–263.
15. Lauzon, M., Gagné, B., Rattue, A. D., Bigras, A. and Hammamji, Y. (2006). Vibrocompaction of the
Foundation Soils of the Peribonka Hydro-Electric Dam. Canadian Dam Association 2006 Annual
Conf., Quebec City, 30 September - 5 October.
16. Menard, L. and Broise, Y. (1975). Theoretical and Practical Aspects of Dynamic Compaction.
Geotechnique, 25(3): 3–18.
17. Mitchell, J. K. (1981). Soil Improvement State-of-the-Art Report. 10th Intl. Conf. on Soil Mech. and
Found. Engrg., 4, Stockholm: 509–565.
18. Varaksin, S., Yee, K. and Wong, L. T. (2009). Formulation of a Concept and Realistic Soil
Parameters for the Foundations of a Randomly Located Structures on a Mega Scaled Project.
Intl. Symp. on Ground Improvement Technologies and Case Studies, Singapore.

View publication stats