Investigation of Expansive Soil for Design of Light Residential Footings in Melbourne Aruna Nishantha Karunarathne Submitted for the Degree of Doctor of Philosophy, Ph.D. Faculty of Engineering and Industrial Sciences Swinburne University of Technology 2016
341
Embed
Investigation of expansive soil for design of light ...Investigation of Expansive Soil for Design of Light Residential Footings in Melbourne . Aruna Nishantha Karunarathne . Submitted
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Investigation of Expansive Soil for Design of
Light Residential Footings in Melbourne
Aruna Nishantha Karunarathne
Submitted for the Degree of Doctor of Philosophy, Ph.D.
Faculty of Engineering and Industrial Sciences
Swinburne University of Technology
2016
Abstract
Expansive soils are found in most Australian states. In fact, it is estimated that 20% of
surface soils in Australia are categorized as moderate to highly expansive clays.
Quaternary basalt clays generally exhibit expansive characteristics and are commonly
found in the Western part of Victoria. These expansive soils undergo heave and
settlement due to moisture changes. Such ground movements are capable of creating
differential movement in footings which results in cracks and damage to light structures.
Recently, it has been reported that more than 5000 houses experienced damage in
Victoria due to changes in soil moisture caused by extreme climate events. Since the
climate condition is the natural cause of soil moisture changes, it must be considered in
the design stage of relevant structures.
This study is part of a comprehensive research programme aimed at enhancing
knowledge about expansive soil behaviour and mitigating damages to residential
structures due to ground movement caused by climate influences. This particular
dissertation focused on estimating soil moisture changes in response to climate
conditions and its consequences. The Australian standard (AS2870) considers the effect
of climate on footing design in terms of Thornthwaite Moisture Index (TMI). It was
found in this study that there are several methods to calculate TMI, which produce
different values for a particular climate condition and hence result in different footing
designs. Furthermore, TMI depends mostly on rainfall and may not be adequate to
consider the variations in soil moisture condition particularly when the TMI values are
based on averaging of long periods.
Soil moisture changes and subsequent ground movement were monitored in a field site
established in Braybrook in Melbourne, which has typical basalt clay soils. The
collected data shows that soil moisture contents follow the rainfall pattern and a
subsequent ground movement, with seasonal variation. Field monitoring over a two year
period revealed that the changes in soil moisture were recorded mainly in top soils
which contributed to the ground movement.
A comprehensive laboratory investigation was performed to characterise the basic
properties of Braybrook soils. These properties suggested that the site has a consistent
profile of highly expansive clay. Further investigations were performed to obtain
specific expansive properties such as mineral composition, soil suction changes and
hydraulic conductivity. These results confirmed the presence of expansive clay soils in
Braybrook. Moreover, a series of shrink swell tests were performed using undisturbed
soils at various in situ moisture contents. These test results indicated the dependency of
shrink swell index on in situ moisture content, which is in contrast to the specifications
outlined in the Australian standard AS1289. The outcome of the laboratory
investigation led to publication of a comprehensive data set that benefits both
practitioners and researchers.
The soil properties collected from the Braybrook site were then used to develop a finite
element model to predict the soil moisture changes in response to climate conditions.
The climate data were obtained from a nearby weather station and the model was
validated against the monitoring data over a two-year period. The validated model was
used to investigate the soil moisture changes due to long-term climate conditions,
including extreme events such as the millennium drought and the subsequent above
average rainfall period in Melbourne. In addition, the soil moisture predictions taken
from this model were fed into another model, developed as part of this comprehensive
research programme, to obtain the ground movement. The results of ground movements
obtained from these models were then compared with the outcomes from the current
Australian standard (AS2870). The results from the model suggest that considerable
additional ground movement has occurred due to the millennium drought, which was
not captured by the AS2870. Furthermore, the model predictions were used to consider
the mound shapes underneath flexible cover slabs placed at different times of the recent
years. Finally, available climate predictions were used to examine the possible future
changes in soil moisture and ground movement.
This research provides a versatile prediction tool for soil moisture changes due to
climate conditions and, therefore, will greatly assist the footing design procedure given
in AS2870. The model can be used to observe future changes of soil moisture within the
design life of a structure using various climate prediction scenarios. Hence, it is an
invaluable tool for designing residential structures that can withstand different severities
of climate conditions as well as uphold the homeowner’s expectations.
Acknowledgement
I am deeply indebted to my principal supervisor, Professor Emad Gad for his endless
guidance, support and constructive criticisms. Working with Professor Emad is the most
pleasurable experience I had in my life. In addition to his assistance in my research
study, the discussions with Professor Emad enhanced my attitude towards success and
personal knowledge.
I also thankful to my second supervisor, Professor John Wilson, for his guidance and
support. I was very lucky to have the luxury of five supervisors including three co-
supervisors from various backgrounds. I wish to express heartfelt appreciation to co-
supervisors Dr. Mahdi Miri Disfani, Dr. Siva Sivanerupan and Dr. Pathmanathan
Rajeev. The guidance in laboratory work from Dr. Mahdi was really important. If I had
not received the invaluable assistance from Dr. Siva, I would not have been able to
initiate field monitoring in the very early stage of this research study. Dr. Rajeevs’s
assistance in finite element model development has been significant. I wish to express
sincere appreciation to all three co-supervisors for their enormous support.
I would like to acknowledge the guidance I had from Mr. Dominic Lopes. He has been
a mentor for our research group. The assistance through his vast knowledge in
expansive soil area has laid the sound foundation for my research study. I also thank Dr.
Robert Evans for giving me the opportunity to undertake tutoring in Geotechnical
Engineering subject, which enhanced my knowledge about footing design.
My warm thanks are due to the administrative and technical staff at the Department of
Civil and Construction Engineering of Swinburne University of Technology. Special
appreciation must be given to Senior Technical Officer, Alec Papanicolaou, in the
workshop for his help in modifications of testing devices and organising various
experimental setups. In addition, PC Support Officer Andrew Zammit in Information
Technology Services kindly provided assistance in various occasions.
I would like to acknowledge Australia Research Council (ARC), the main contributor of
this research project from the ARC Linkage Grant (LP100200306). I gratefully
acknowledge the financial and technical support provided by the collaborating
organizations, namely; Victorian Building Authority (VBA), Victorian Office of
Housing (OoH), Foundation and Footings Society of Victoria (FFSV), Association of
Consulting Structural Engineers Victoria (ACSEV) and Housing Engineering Design
and Research Association (HEDRA). I would like to recognise the invaluable feedback
from adversary panel members who represented the above-mentioned organizations. I
wish to express my deep and sincere gratitude to my colleagues of the research group;
Jenny Boyer and Deepti Wagle for their support during the study.
Finally, I owe my loving thanks to everyone in my family. Special thanks should be
given to my brother who has been looking after my parents while I was far from them.
This thesis would not have been possible without their love, encouragement and
understanding.
Declaration
I hereby declare that this thesis contains no material which has been accepted for the
award of any other degree or diploma in any university or institution. To the best of my
knowledge and belief, this thesis contains no material previously published or written
by another person, except when due reference is made in the text of the thesis.
Aruna Nishantha Karunarathne
August 2016
Preface
This dissertation is produced as part of a comprehensive research study aimed to assist
in mitigating damage to residential structures due to ground movement. The study
involved three PhD students and several resource persons directing at a number of
engagements of the research problem. The author of this dissertation has been
responsible for the geotechnical section of the study which indeed focused on
estimation of soil moisture and ground movement induced by climate conditions. This
dissertation is original, unpublished and independent work by the author, Aruna
Nishantha Karunarathne.
Following peer reviewed publications were produced from various outcomes of the
research and they are based on certain sections of thesis chapters.
Journal papers
KARUNARATHNE, A. M. A. N., SIVANERUPAN, S., GAD, E. F., DISFANI,
M. M., RAJEEV, P., WILSON, J. L. & LI, J. 2014. Field and laboratory
investigation of an expansive soil site in Melbourne. Australian Geomechanics,
49, 85-93.
KARUNARATHNE, A. M. A. N., GAD, E. F., DISFANI, M. M.,
SIVANERUPAN, S. & WILSON, J. L. 2016. Review of Calculation Procedures
of Thornthwaite Moisture Index and its Impact on Footing Design. Australian
Geomechanics, 51, 85-95.
FARDIPOUR, M., GAD, E., SIVAGNANASUNDRAM, S., RAJEEV, P.,
KARUNARATHNE, A. & WILSON, J. 2016. Interaction analysis of waffle
slabs supporting houses on expansive soil. Innovative Infrastructure Solutions,
1, 1-10.
Conference papers
KARUNARATHNE, A. M. A. N., GAD, E. F., SIVANERUPAN, S. &
WILSON, J. L. Review of Residential Footing Design on Expansive Soil in
Australia. In: SAMALI, B., ATTARD, M. M. & SONG, C., eds. 22nd
Australasian Conference on the Mechanics of Structures and Materials, 11-14
December 2012 Sydney, NSW. Taylor & Francis Group, 575-579
KARUNARATHNE, A. M. A. N., SIVANERUPAN, S., GAD, E. F., DISFANI,
M. M., WILSON, J. L. & LI, J. Field monitoring of seasonal ground movements
in expansive soils in Melbourne. In: KHALILI, N., RUSSELL, A. &
KHOSHGHALB, A., eds. 'UNSAT 2014', Unsaturated Soils: Research and
Applications - the 6th International Conference on Unsaturated Soils, 2-4July
2014 Sydney, NSW. Taylor & Francis Group, 1359-1365.
Moreover, recent concerns about climate change and its effects on footing design have
also been discussed in terms of the TMI (Austroads, 2004, Leao and Osman, 2013).
Table 3-1: Climate types together with their TMI limits (Thornthwaite, 1948)
Climate type TMI
A Perhumid 100 and above B4 Humid 80 to 100 B3 Humid 60 to 80 B2 Humid 40 to 60 B1 Humid 20 to 40 C2 Moist subhumid 0 to 20 C1 Dry subhumid -20 to 0 D Semiarid -40 to -20 E Arid -60 to -40
67
3.3.1 Calculation of TMI
Thornthwaite (1948) introduced the TMI in 1948 and later published a number of
papers (Thornthwaite, 1952, Thornthwaite and Mather, 1955, Thornthwaite and Mather,
1957) to provide a clearer understanding of the calculation. According to these
publications, the TMI calculation procedure can be expressed in the flow chart shown in
Figure 3-4. Precipitation and temperature are the main input parameters of the TMI.
Water balance is calculated using those inputs and this provides the surplus and deficit.
The surplus and deficit are then used to obtain Ih and Ia to calculate the TMI.
Figure 3-4: Flow chart of the TMI calculation
3.3.2 Definitions and Assumptions
TMI calculation is associated with various terms that have been defined and modified at
different times. However, certain terms are defined in alternative ways, which has
caused alterations in the calculation.
Precipitation (P), also known as rainfall, is the main soil moisture input.
Evapotranspiration causes soil moisture loss which transfers the water from soil to air
by evaporation and transpiration. Potential Evapotranspiration (PE) is defined as the
amount of water that would be evapotranspired under certain climate conditions given
an unlimited supply of water. However, the amount of water lost from soil in a
particular climate condition is always restrained by the water availability and is defined
as Actual Evapotranspiration (AE).
Precipitation (P)
Mean Maximum
Temperature
Mean Minimum
Temperature
Average
Temperature
Potential Evapotranspiration
(PE)
Water Balance
Surplus (R) and Deficit (D)TMI Aridity index (Ia) and Humidity index (Ih)
Field capacity and initial storage
Latitude of location
68
The TMI calculation generally considers the soil body as a tank, hence the soil moisture
storage (S) is defined as the amount of water held in the soil at any particular time. The
maximum amount of water the soil can hold is called the field capacity (Smax). Based on
the monthly values of P and AE, soil moisture storage changes when monthly P is not
equal to AE and this change in storage is denoted as ∆S.
A number of definitions exist for moisture surplus (R) or runoff. Thornthwaite (1948)
states that water surplus refers to seasonal additions to subsoil moisture and ground
water. Thornthwaite and Mather (1955) defined the moisture surplus as the precipitation
in excess of potential evapotranspiration which occurs when soil is at the field capacity.
Later, Mather (1978) gave a more descriptive explanation that “surplus is the excess
water available to percolate through the soil both as recharge to the ground water table
or as through flow. This amount is P-PE in the months of soil moisture is at the field
capacity. When the soil storage is not at its capacity, no surplus can exist.” These
definitions contain slight differences, which can cause inconsistencies in the water
balance calculation.
Moisture deficit (D) is defined as the additional water that would be necessary to
achieve potential evapotranspiration when the precipitation is not sufficient. Mather
(1978) stated that the deficit is the difference between the water demand in a particular
climate condition and the actual evapotranspiration losses which can be calculated as
PE-AE.
In addition to contrasting definitions, the TMI calculation also includes some
assumptions. To account for the difficulty of water extraction from its adsorbed state in
a soil, Thornthwaite (1948) assumed a surplus of 60% in one season will counteract a
deficiency of 100% in another. The TMI calculation assumes the excess precipitation
(P-PE) that comes after a deficiency period will entirely infiltrate into the soil and
recharge the soil moisture storage until it reaches the field capacity. The excessive P
received when the soil is at the field capacity is becomes a runoff. Therefore, surface
runoff which may occur when the soil is being recharged is not considered, which
frequently happen when the rain falls in high intensity. The Australian TMI map has
also been developed employing the same assumption (Aitchison and Richards, 1965).
69
The next important assumption is the value of field capacity. Chan and Mostyn (2008)
suggested that the Smax depends on the climate and proposed the use of 0 mm for dry, 50
mm for temperate and 100 mm for wet conditions. Even though Thornthwaite and Mather
(1957) specified different values for field capacities based on soil types, most
researchers (Aitchison and Richards, 1965, Jewell and Mitchell, 2009, Russam and
Coleman, 1961) have assumed a constant value over a number of different soil types
and 100 mm has been used commonly. Indeed, Aitchison and Richards (1965) assumed
Smax of 100 mm to obtain the TMI for more than 600 locations to develop the Australian
TMI map.
Another assumption of the TMI calculation relates to the behaviour of soil moisture
storage. Thornthwaite and Mather (Mather, 1978, Thornthwaite and Mather, 1955)
assumed that, when soil becomes dryer, the removal of water from the soil becomes
increasingly difficult and soil cannot extract the same amount of moisture from the
storage. Therefore, the soil moisture storage never becomes zero. In contrast, most of
the recent research (Barnett and Kingsland, 1999, Chan and Mostyn, 2008, Fityus et al.,
1998, Fox, 2000, Jewell and Mitchell, 2009, Lopes and Osman, 2010, McManus et al.,
2004, Mitchell, 2008) has assumed that the soil can provide the required additional
amount of moisture until the storage level becomes zero.
The initial storage (S0) needs to be assumed to perform the water balance in the TMI
calculation. This is the water storage available at the first month of the first year of TMI
calculation. When the calculation is performed over a longer period, the impact of this
assumed value becomes insignificant.
Based on the above definitions and the stated assumptions, TMI has been defined by an
aridity index (Ia) and humidity index (Ih). Ia is a relationship between moisture deficit
and water necessity (Equation 3-1) whereas Ih is a relationship between moisture
surplus and water necessity (Equation 3-2).
Ia = 100 ×D
PE ……...…………………………………………...…… Equation 3-1
70
Ih = 100 ×R
PE ……...…………………………………………...…… Equation 3-2
Thornthwaite (1948) suggests that, PE be calculated using Equation 3-3 for each month
and then the monthly summation is taken as annual PE. The unit of D, R and PE is
centimetres.
PE = 1.6 × (10 × t
I)
a
.………...…………………………………...…… Equation 3-3
‘t’ is the average temperature in a particular month and ‘I’ is annual heat index in a
particular year which is taken as the summation of monthly heat index values (i)
calculated using Equation 3-4. Parameter ‘a’ is calculated using Equation 3-5.
i = (0.2 × t)1.514 .……...……...…………………………………...…… Equation 3-4
The calculated ys is higher than 75 mm which is the lower limit of class E (Table 2-4)
and hence the site was classified in the Extremely reactive category.
4.5 DEVELOPMENT OF MAIN EXPANSIVE SOIL PARAMETERS
Models developed to describe the volume change behaviour of expansive soil due to
moisture changes commonly use suction as a main soil characteristic (Fredlund and
Rahardjo, 1993, Fredlund and Vu, 2003, Mitchell and Avalle, 1984). This is because
suction represents the stress state of the soil (Fredlund and Rahardjo, 1993). The
relationship between suction and moisture content is one of the two main constitutive
114
parameters (Likos, 2000). The amount of water that can be held by the soil pores at
certain suctions varies with the soil type (Likos, 2000). Water can be trapped in soil
pores and this amount can also change with the compaction of the soil (Dingman,
2002). Hence, the relationship of suction and water content (SWCC) is an important
parameter, which is essential to study expansive soil behaviour.
The moisture content of soil varies with time due to changes in climate conditions. This
phenomenon associates with the permeability of the soil at different moisture contents.
The hydraulic conductivity allows water to flow easily through the porous media and
therefore depends on many soil parameters including pore size, soil grain size and the
amount of water held in the pores (Fratta et al., 2007). It can therefore also be correlated
to the soil suction. This relationship is called the hydraulic conductivity function.
The following sections describe the details of the development of SWCC and the
hydraulic conductivity function of Braybrook soil.
4.5.1 Soil Water Characteristic Curve (SWCC)
The negative pore water pressure created due to water content in soil pores is called
matric suction (Fredlund and Rahardjo, 1993). The variation of matric suction with
water content is expressed by SWCC. Therefore, matric suction and the corresponding
moisture content should be measured to obtain the coordinates of SWCC. The moisture
measurements are straight forward and can be performed simultaneously with the
suction measurements. The moisture content of the soil can be expressed in various
forms. Therefore, SWCC function can be plotted using volumetric water content,
gravimetric water content or degree of saturation. However, the constitutive model
explanations of volume change in expansive soil are associated with volumetric basis
(Fredlund and Rahardjo, 1993, Fredlund and Vu, 2003, Mitchell and Avalle, 1984) and
therefore Braybrook SWCCs have been developed based on both the matric suction and
the volumetric water content.
4.5.1.1 Soil suction
A variety of equipment and different techniques can be used to measure soil suction.
Each of the techniques has its own measuring range and different equilibration time to
produce readings. Table 4-8 shows approximate equilibration time and the measuring
rage of different instruments and techniques. None of the available equipment can
115
measure the full range of suction in SWCC function and therefore the development of
SWCC must be performed in different stages with different equipment.
Table 4-8: Approximate measurement ranges and times for equilibration in measurement and control of soil suction (Murray and Sivakumar, 2010)
Instrument Suction component measured
Typical measurement range (kPa)
Equilibration time
Pressure plate Matric 0-1,500 Several hours to days Tensiometers and suction probes Matric 0-1,500 Several minutes Thermal conductivity sensors Matric 1-1,500 Several hours to days
Electrical conductivity sensors Matric 50-1,500 Several hours to weeks
Filter paper contact Matric 0-10,000 or greater 2-57 days
Thermocouple psychrometers Total 100-8,000 Several minutes to several hours
Transistor psychrometers Total 100-70,000 About 1 hour Chilled mirror psychrometer Total 1-60,000 3-10 minutes
Filter paper non-contact Total 1,000-10,000 or
2-14 days greater
Electrical conductivity of pore water extracted using pore fluid squeezer
Osmotic entire range —
Suction control
Negative (or Hanging) water column technique Matric
0-30 or greater with multiple columns or vacuum control
Several hours to days
Axis translation technique Matric 0-1,500 Several hours to days Osmotic technique Matric 0-10,000 up to 2 months Vapour equilibrium technique Total 4,000-600,000 1-2 months
During the development of Braybrook SWCCs, the high suctions (dry soils) were
measured using WP4C (Decagon, 2012). WP4C uses the chilled mirror psychometric
technique and hence it measures the total suction of the soil. The low suctions (wet
soils) were measured from Hyprop (UMS, 2013). The Hyprop uses the tensiometer
technique which measures the matric suction. However, there is a gap between the
measurable suction ranges of these equipment and that gap was filled using
conventional filter paper measurements. Both total suction and the matric suction can be
measured using the filter paper test and, as a result, the osmotic suction can be
116
calculated. The osmotic suction, which is created by salt concentration of the soil,
appears almost as a constant with the moisture content (Fredlund and Rahardjo, 1993).
The filter paper test was therefore used in this study to obtain the osmotic suction and
then to convert the WP4C readings into matric suction.
4.5.1.2 Hyprop measurements
Hyprop (UMS, 2013) is an instrument specifically developed to produce SWCC of soil
in wet conditions. This instrument uses the tensiometer technique to measure the suction
of soil. The tensiometers consist of a high air entry ceramic tip connected to one end of
a hollow shaft carrying de-gassed water and the other end of the shaft is connected to a
pressure transducer. When the porous tip is inserted into the soil sample, the water is
drawn from the tensiometer due to the suction difference until the stress of the water
inside the tensiometer is equal to the suction of the soil (Murray and Sivakumar, 2010).
The soluble salts that create the osmotic suction can transfer freely through the porous
tip (Murray and Sivakumar, 2010). Therefore, pressure transducer records only the
matric suction. The Hyprop arrangement can take continuous weight measurements of
the soil and can measure the suction in addition to the corresponding moisture content,
which are essentially the coordinates of SWCC.
The initialization of the instrument takes few hours and some experience to set it up
properly. The measurements begin from the saturated stage and continue until the
sample is dried out. The moisture evaporates from the top surface of the sample during
the test. The clay soils take a longer time than sandy soils to reach the dry stage so the
total measurement time can vary from a few hours to few weeks depending on the soil
type.
The Hyprop device uses undisturbed soil samples of 80 mm diameter and 50 mm
height, as shown in Figure 4-11. The samples cannot be taken out of normal tube
sampling, which has a 50 mm diameter. Indeed, 80 mm is an uncommon tube size and it
is difficult to push this size of a ring into soil using a vehicle mounted rig to collect
undisturbed soil at deeper depths. Therefore, the Hyprop samples were collected from a
pre-excavated pit in the Braybrook site shown in Figure 4-12. An eight tonne excavator
was used to dig a 3 m deep pit in a 2x3 m area. The pit was excavated as steps of 0.5 m,
117
as shown in Figure 4-12, to allow access at roughly 0.5 m depths for sample collection.
This pit also helped to visually examine the soil profile in Braybrook site (Figure 4-6).
Figure 4-11: Hyprop sample in the ring
Figure 4-12: Excavation of a pit in Braybrook site to collect Hyprop samples
118
Undisturbed soils were collected into sampling rings using the apparatus that is
provided with the Hyprop devices, as shown in Figure 4-13. The sampling rings were
pushed into soils by hammering and removed by clearing the surrounding soil. The
samples were trimmed at the site, as shown in Figure 4-11, and labelled with the depths
before being taken to the laboratory.
Figure 4-13: Excavation of undisturbed samples using Hyprop sampling device
The samples were saturated under a surcharge pressure corresponding to their depth.
The soil bulk density was about 2 gcm3. For example, soil samples at 0.5 m depth were
saturated under 10 kPa pressure and two porous plates were placed at the top and
bottom to prevent the samples dispersing in water. Next, the samples were submerged in
distilled water (Figure 4-14) and the swelling of the soil was also monitored during
saturation. The samples were left submerged for at least two weeks to confirm that full
saturation was achieved.
119
Figure 4-14: Hyprop samples saturation under a surcharge
Once the saturation process was completed, the Hyprop apparatus was initialized to start
the test. This process involves refilling the tensiometers and the Hyprop sensor unit. The
refilling kit provided with the Hyprop device was used. Figure 4-15 shows the initial
steps of the refilling process. Distilled and de-gassed water must be used for refilling.
De-gassing can be performed using syringes with spacer snaps and once the water is de-
gassed, the tensiometers should be filled without trapping air bubbles. The tensiometers
must be kept in an upright position as shown in Figure 4-15(a). The bottom syringe is
filled with de-gassed water and all the air bubbles must be removed before connecting
to the ceramic tip of the tensiometer. The top syringe is half filled with de-gassed water
and suction is applied by locking the spacer snaps, as shown in Figure 4-15(a). The de-
gassed water from the bottom syringe is forced to travel to the top syringe through the
ceramic tip and the shaft of the tensiometer. This process removes all the air entrapped
in the tensiometers and at least 2 hours is required for this process. The Hyprop consists
of two tensiometers and both of them can be refilled at the same time using four
syringes. The Hyprop sensor unit must also be filled with de-gassed water using the
acrylic attachment and a syringe as shown in Figure 4-15(b).
120
Figure 4-15: Refilling of de-gassed water; a) into tensiometer, b) into Hyprop sensor unit
The refilled tensiometers are then attached to the sensor unit while monitoring the
pressure developed at the pressure sensors. The “tensioVIEW” software, which comes
with the Hyprop equipment, was used to monitor the pressure during the initialization
process. The Hyprop sensor unit needs to be connected to a computer before the
tensiometers attach to the provided slots in the sensor unit. The pressure reading of the
sensor must be carefully monitored and, as described in the manual, should not be
allowed to exceed 100 kPa during the tensiometer attaching process (UMS, 2013). O-
rings are pushed over each of the tensiometer shafts to prevent the entering of dirt
during the test and a silicon gasket is also placed through tensiometers to avoid contact
between the sensor base and the sample as shown in Figure 4-16.
The Hyprop device comes with a balance that can be connected to a computer. The
weight measurements can be recorded using “tensioVIEW” software. Therefore, the
Hyprop sensor unit and the balance were connected to a laptop computer to record the
suction and weight during the test period.
a) b)
121
The saturated soil sample is then prepared to insert two tensiometers. The tensiometers
are in two different heights, as shown in Figure 4-16. This is to measure the suction at
top and bottom of the sample. The top tensiometer is 50 mm long and penetrates about
37.5 mm into the soil samples. The tip of the top tensiometer is located approximately
12 mm below the top surface of the soil. The bottom tensiometer is 25 mm long and
penetrates about 12.5 mm into the soil. An augur with an adapter is provided with the
Hyprop device to make two holes in the soil sample similar to the heights of the two
tensiometers to be penetrated into the soil.
Figure 4-16: Suction measuring unit of Hyprop device(UMS, 2013)
Figure 4-17 shows the sample, with two holes drilled at the bottom surface, ready to
insert the tensiometers. The holes must be filled with de-gassed distilled water before
inserting the tensiometers. The sensor unit with tensiometers is attached to the soil
sample using two fastener clips at either side (Figure 4-18). The sensor is then
connected to the laptop computer and the arrangement is placed on top of the balance
(Figure 4-19). The readings can be recorded in pre-specified time period. The readings
have been recorded in 1 minute intervals during the first few hours of the test and
extended to 10 minutes later.
122
Figure 4-17: The auger adapter and the sample with two holes drilled in the bottom surface
Figure 4-18: The soil sample attached to the Hyprop sensor unit
The moisture evaporates during the test period from the surface of the sample, which
leads to changes in suction and the weight of the soil. The test was conducted inside an
environmental chamber to smoothen the evaporation process and to avoid the unwanted
interruptions such as turbulent wind over the samples surface, disturbance to the
balance, etc. The humidity and the temperature were set to 60% and 20 0C inside the
chamber.
123
Figure 4-19: The Hyprop test is running inside the environmental chamber
The measurements continued for 5 to 7 days. The moisture evaporation from the top
surface crated cracks, as shown in Figure 4-20, which propagated towards the bottom.
Once a developed crack reaches the tensiometer, the porous tip is exposed to the air. At
this point, the tensiometer fails and starts reading an untrue value for the soil suction.
The Hyprop test can be stopped at this point. The dried out sample is shown in Figure
4-20. During the drying process of the Hyprop test, the clay samples tend to stick to the
tensiometer shaft and therefore, it is difficult to remove the samples immediately after
the test. This can be overcome by placing the sample in a water bath after the test
without letting water get into the cable outlet of the sensor unit. Leaving the sample in a
water bath for a few hours will moisturize the samples and facilitate removal from the
sensor unit. The entire soil sample needs to be collected including soils attached to
124
sample ring and the tensiometer shafts. Then, the sample is placed in an oven at 105 0C
to determine the oven dry weight.
Figure 4-20: The soil sample at the end of the Hyprop test
UMS (2013) has provided another software called “HYPROP” to analyse the data
collected from tensioVIEW to develop the SWCC. This software requires the data file
from tensioVIEW and the oven dry weight of the soil sample. The “HYPROP” software
uses the top and bottom tensiometer readings and calculates the representative matric
suction of the soil at a particular time. It uses the initial volume of the soil (the volume
of the sampling ring) and the provided oven dry soil weight to calculate the volumetric
moisture content of the soil. However, for expansive soil, the calculation procedure of
volumetric moisture content of the soil is incorrect in the drying stage. This is because
the expansive soils undergo a significant volume change and cracking during the drying
process hence, the volume of the soil at a particular time cannot be considered as the
volume of the sampling ring. This problem has been overcome using the relationship
between volumetric moisture content and the gravimetric moisture content for the
Braybrook soil. The instant weight of the soil at each record has been extracted from
“HYPROP” software and then the gravimetric moisture contents were calculated using
the oven dry soil weight. Next, they were converted into volumetric moisture contents
using the relationship shown in Figure 4-21. This relationship has been developed using
125
undisturbed samples collected from Braybrook site at different depths. The first point
has been assumed such that, at the zero percent volumetric moisture content,
gravimetric moisture content is also zero. The relationship has an R2 value of 0.99.
Figure 4-21: Relationship between volumetric and gravimetric moisture consents in Braybrook soil
Figure 4-22 shows a portion of SWCC obtained using Hyprop devices as described
above. Appendix-A provides the necessary calculations related to the Hyprop test.
Figure 4-22: A portion of typical SWCC developed using Hyprop
0 5 10 15 20 25 30 35 400
10
20
30
40
50
60V
olum
etric
Moi
stur
e C
onte
nt (%
)
Gravimetric Moisture Content (%)
y = -0.014x2 + 1.9168xR2= 0.99
1 10 1000.35
0.40
0.45
0.50
0.55
Volu
met
ric M
oist
ure
Con
tent
Matric Suction (kPa)
Hyprop measurements
126
4.5.1.3 WP4C measurements
WP4C (Decagon, 2012) measures the total suction of the soil using the chilled-mirror
psychometric technique. The suction readings can be observed in “MPa” units and the
conventional “pF” units. As shown in Figure 4-23, the WP4C consists of a temperature
controller, temperature sensor, mirror and a photo detector cell. The sample must be
placed in a standard ring of 37.5 mm diameter and about 8 mm in height. The sample
ring is made out of plastic or metal. The sample is placed on the drawer and pushed into
the chamber. When the switch is turned to the “READ” position, the sample is raised
and the chamber is sealed to begin the measurement process. Therefore, the sampling
cup must not entirely fill with the samples to prevent the contamination of the sensor
while sealing the chamber.
Figure 4-23: Schematic of chilled-mirror dew-point device (after Leong et al. (2003) )
When the switch is turned to the “READ” position, the equilibration process begins
which makes the relative humidity of the air above the soil equal to the relative
humidity of the air in the soil pore spaces (Murray and Sivakumar, 2010). This takes a
certain amount of time depending on the soil type and the moisture content of the soil.
Once the equilibrium is achieved, the mirror is cooled by carefully controlling the
temperature using a Peltier current (Murray and Sivakumar, 2010). When the
temperature of the mirror reduces, the vapour starts to condensate on the mirror at a
certain point, called the dew point, and it can be recognized by the photo detector cell
because of the difference of the reflection from the mirror (Leong et al., 2003). By using
dew point temperature and the controlling temperature, the relative humidity is
127
calculated and then it can be correlated to the total suction using Equation 4-1 (Bulut et
Since, the radioactive material can be harmful for the health, the amount of neutron
absorbed to the persons involved in the measuring was carefully monitored. An
authorization certificate and training were required to use the neutron probe and users
wore a monitoring badge at all times. The badge was independently checked every three
months for neutron absorption.
5.2.3 Ground movement monitoring
5.2.3.1 Installation of magnetic extensometers
The installation procedure of the extensometer is a challenging task. After the spider
magnets are attached to the collapsible pipe at the desired spacing, the magnet legs are
folded and held together with temporary ties, as shown in Figure 5-5. Magnetic
extensometers are installed into pre-drilled boreholes. The diameters of the bore holes
are limited by the size of the folded legs of the magnets and the capability of them to
penetrate to the soil. If the borehole diameter is too large then the magnet legs cannot
penetrate the surrounding soil. The spider magnets used in this study were capable of
150
penetrating soil in a borehole up to 150 mm diameter. However, the compressed legs of
the magnets were able to push into 100 mm diameter borehole safely. The temporary
ties must be removed after installation, which will release the legs and allow them to
penetrate the soil. All the temporary ties are held by a steel code which is parallel to a
collapsible pipe and attached to the bottom using a glue tape. The steel cord is removed
after the installation and the magnet legs are allowed to unfold. They enter the soil with
the sudden releasing of the ties. Figure 5-6 shows the installation of one of the magnetic
extensometers in the Braybrook site. The prearranged spacing between the spider
magnets could change during the installation due to the flexibility of the collapsible pipe
and the pressure applied to insert into the borehole.
Figure 5-5: Releasing mechanism of the spider magnet in extensometer
151
Figure 5-6: Magnetic extensometer installation at the Braybrook site
The extensometers were covered using plastic caps as shown in Figure 5-7. Since the
measuring probe reads the distance between the spider magnets, a reference point at the
surface is required to convert the readings into layer depths. A concrete paver was
placed and kept undisturbed next to the extensometer (see Figure 5-7). The depths of
spider magnets were measured with respect to the level of this paver and its level was
considered as the ground level at that location. As demonstrated in Figure 5-7, a spirit
level was used to transfer the top level of the paver to the top of the extensometer
conduit pipe to obtain the measurements.
Figure 5-7: Ground surface movement measurements using extensometer
152
5.3 SITE LAYOUT
Figure 5-8 shows the monitoring locations in the Braybrook site. Three extensometers
are located on the site marked as E1, E2 and E3. All the extensometers were located in
the intact area of the blocks and away from the trees. The datum of the E1 extensometer
was placed 5 m below the ground surface while E2 and E3 had datum magnets at 4 m
depth. The access tubes for neutron probes were located next to each extensometer to
monitor the moisture variations along with the soil layer movements. Concrete paving
blocks of measuring 300x300 mm in plain area and 50 mm thick were placed on the
ground surface at locations marked as TP and CP. The purpose of the paving blocks was
to measure the movement of the ground surface which was done using a surveying
level. Moisture changes have been monitored close to each TP paver to monitor the
effect of trees on soil moisture. However, the measurements related to the tree are not in
the scope of this thesis. The E1 extensometer was located in the first block where the
tree is situated, while E2 and E3 extensometers were located in third block away from
the tree. CP paving blocks around E2 and E3 extensometers were placed in a grid of 2
m x 5 m. The purpose of those paving blocks was to check the deviation of ground
surface movement around extensometers.
Figure 5-8: Monitoring Plan of the Braybrook Site
TP 1 to 14 : Paving blocks around the treeCP 1 to 10 : Paving blocks away from the tree siteE 1 to 3 : Magnetic extensometersTN 6 : Neutron probe access tube close to E1CN 1 : Neutron probe access tube close to E2CN 2 : Neutron probe access tube close to E3
Location marks are not to scale
TN6
CN1
CN2
153
5.4 INVESTIGATION OF FIELD MONITORING DATA
The extensometers and the neutron probe access tubes were installed at two different
times. The extensometer E1 was installed in October 2012 and the other two were
installed in March 2013. The neutron probe access tubes which were inserted next to
extensometers are considered in this analysis. Neutron probe access tube denoted as
“TN6” was located next to the E1 extensometer. The neutron probe access tubes, CN1
and CN2 were located next to E2 and E3 extensometers, respectively. The data
collected from those locations are described in the following section.
5.4.1 Soil moisture profiles with time
The CN1 and CN2 were located close to each other and hence a similar moisture
variation was expected. However, slight differences in the clay content of surface soils
at two locations were observed. The CN2 location has a sandy top layer about 200 mm
and then gradually changed to silty clay and clay soil. Apart from the CN2 location,
such a different soil profile was not observed in other areas of the Braybrook site.
Hence, it could be a site disturbance from its previous usage. However, the area used to
perform the ongoing monitoring had been part of the back yard as shown in Google map
photo taken in 2010 (Figure 4-3). Some potholes surrounded by clay soil chunks were
observed near the CN1 location. There was also a slope difference on the ground.
Figures 5-9 and 5-10 show the moisture variation at CN1 and CN2 locations.
154
Figure 5-9: Volumetric moisture content profiles at CN1 location
Figure 5-10: Volumetric moisture content profiles at the CN2 location
The neutron probe calibration equation gives the volumetric moisture content and it also
has been used in the finite element modelling. Therefore, volumetric moisture content
has received more interest than gravimetric moisture content. The volumetric moisture
40 Daily Rainfall (mm) Average Neutron Probe Measurements (CN1 and CN2) at 0.35 m Model prediction with Vegitation included Model prediction with Vegitation excluded
Date
Dai
ly R
ainf
all (
mm
)
0.0
0.1
0.2
0.3
0.4
0.5
0.6
Vol
umet
ric m
oist
ure
cont
ent
211
allowed in the validated model. But these two scenarios can be considered in model
applications to observe the effect of local conditions.
Figure 6-35: Effect of ponding condition on soil moisture at Braybrook site
6.11 SUMMARY
This chapter describes the development of a finite element model to predict soil
moisture changes in response to climate conditions. Vadose/w package, available in
GeoSlope software, was used to develop the finite element model. A 1D soil column
was modelled using the material properties of Braybrook soils described in Chapter 4.
The model requires specifying SWCC, hydraulic conductivity, thermal conductivity and
the specific heat capacity of the soil. A sensitivity analysis suggested that the thermal
properties had a minimal impact on soil moisture changes. Therefore, thermal
conductivity and specific heat capacity obtained from literature related to a site close to
Braybrook were deemed appropriate for this research. The soil properties were defined
at different depths of the soil column to represent actual soil profile. The depth of the
soil column is governed by the location of bedrock. In Braybrook site, bedrock was not
hit at 4.5 m, therefore a 6 m deep soil column with bottommost meter (5m to 6m) of
bedrock was modelled. Soil properties of Braybrook soils were measured up to 2.5 m
depth. Braybrook soils have similar SWCCs below surface layer and therefore, the
measured SWCC at 2.0 m is also used for the depths below 2 m.. Since hydraulic
properties of soil gradually vary with the bulk density and the soil properties and are
consistent in Braybrook site, a gradual variation of saturated hydraulic conductivities of
40 Daily Rainfall (mm) Average Neutron Probe Measurements (CN1 and CN2) at 0.35 m Model prediction with Ponding not allowed Model prediction with Ponding allowed
Date
Dai
ly R
ainf
all (
mm
)
0.0
0.1
0.2
0.3
0.4
0.5
0.6
Vol
umet
ric m
oist
ure
cont
ent
212
the soil between 2 and 5 m were considered where bulk density is governed by the
surcharge. These saturated hydraulic conductivities were used to develop hydraulic
conductivity functions using the Fredlund et al. (1994) method.
The model requires specifying boundary conditions to represent the moisture flow
conditions. The climate boundary was specified at the top surface. Vadose/w software
uses daily inputs of rainfall, evaporation, relative humidity, wind and temperature to
define the climate boundary. These climate components were collected from a weather
station close to the Braybrook site. The no-flow boundary was defined at the bottom of
bedrock to prevent the moisture flow from bedrock.
This model was analysed for the two year period of field monitoring. The model
predicts soil moisture and suction profiles at different times. These predictions were
compared with the neutron moisture measurements described in Chapter 5. The results
showed a good agreement between soil moisture predictions and monitored data and
hence the model was considered as validated.
Next, the 1D model was extended to a 2D model to determine lateral moisture
movements. The only additional parameter required in the 2D model is hydraulic
conductivity in lateral direction. During the mass excavation in Braybrook site,
homogeneous soil chunks were found where no layer changes could be observed.
Hence, it was assumed the both lateral vertical hydraulic conductivities are the same in
Braybrook soils. The 2D model was developed to incorporate impervious cover in order
to observe the soil moisture beneath cover slabs due to the climate influences on
adjacent open ground.
The sensitivity analysis of the input parameters revealed that soil moisture changes are
mostly affected by SWCC of soils followed by the hydraulic conductivity. Rainfall is
the most sensitive climate component in changing soil moistures. Furthermore,
evaporation and relative humidly have significant impacts on soil moisture contents.
The responses of the models to vegetation effects and ponding conditions were also
considered. Grass covers prevent the evaporation and increase the water available for
the infiltration. Therefore, both grass covers and pooling effects significantly increase
the soil moisture. These sensitive input parameters must be given more attention in
obtaining soil moisture using prediction models. This model was then used to determine
213
the soil moisture changes due to long-term climate conditions, which is described in the
next chapter.
214
7. MODEL APPLICATIONS
7.1 OVERVIEW ON MODEL PREDICTIONS OF SOIL MOISTURES
The validated model described in the previous chapter was used to investigate the soil
moisture changes in response to various long-term climatic conditions. A number of
models were analysed to consider different site conditions and climate scenarios.
Following notations were used in this chapter to denote these different models.
VB1 1D Vadose/w model, Braybrook site, 6 m deep, period 1945-2015
(Section 7.1.1.1)
VB2 1D Vadose/w model, Braybrook site, 3 m deep, period 1945-2015
(Section 7.2.5)
FLAC 1D soil column developed in FLAC-3D software which uses moisture
predictions from Vadose/w models
VB3 1D Vadose/w model, Braybrook site, 6 m deep, short term wet condition,
period 1945-2018 (Section 7.3)
VB4 1D Vadose/w model, Braybrook site, 6 m deep, short term dry condition,
period 1945-2018 (Section 7.3)
VB5_A 1D Vadose/w model, Braybrook site, 6 m deep, typical average climate,
period 50 years (Section 7.4)
VB5_M 1D Vadose/w model, Braybrook site, 6 m deep, modified climate, period
50 years (Section 7.4)
2DVB1 2D Vadose/w model, Braybrook site, 6 m deep, period 1945-2015
(Section 7.5)
2DVB2 2D Vadose/w model, Braybrook site, 6 m deep, with flexible cover,
period 1983-1992 (Section 7.6.1)
2DVB_I 2D Vadose/w model, Braybrook site, 6 m deep, without flexible cover,
period 1945-2015 (Section 7.6.1)
215
2DVB3 2D Vadose/w model, Braybrook site, 6 m deep, with flexible cover,
period 1992-2010 (Section 7.6.2)
2DVB4_S 2D Vadose/w model, Braybrook site, 6 m deep, with flexible cover, soil
dipping towards slab edge, period 1992-2010 (Section 7.7.1)
VF1 1D Vadose/w model, Fawkner site, 3 m deep, period 1945-2015 (Section
7.1.1.1)
VF2 1D Vadose/w model, Fawkner site, 3 m deep, short term wet condition,
period 1945-2018 (Section 7.3)
VF3 1D Vadose/w model, Fawkner site, 3 m deep, short term dry condition,
period 1945-2018 (Section 7.3)
VF4_A 1D Vadose/w model, Fawkner site, 3 m deep, typical average climate,
period 50 years (Section 7.4)
VF4_M 1D Vadose/w model, Fawkner site, 3 m deep, modified climate, period
50 years (Section 7.4)
7.1.1 Prediction of soil moistures
The models described in this chapter were analysed using climate data collected from
Essendon airport weather station. Figure 7-1 shows the variation of annual rainfall from
Essendon airport weather station from 1945 to 2014. Isolated dry and wet years were
recorded pre-1995; however, overall, the rainfalls prior to 1995 can be considered
average conditions. However, a significant reduction of annual rainfall occurred during
the millennium drought (1996-2009). The drought-breaking rainfalls in 2010 and 2011
are clearly noticeable and created back-to-back extreme events. Since 2012, a gradual
reduction of annual rainfall was observed. Interestingly, 2014 was recorded as the
warmest year for Victoria (BoM, 2015b), which suggests that climate conditions are
heading towards another dry period (Hannam, 2015).
216
Figure 7-1: Annual rainfall recorded in Essendon airport weather station
The Vadose/w model developed in this study was used to investigate the soil moisture
changes due to these different conditions, including different climate predictions.
7.1.1.1 Sites considered in the long term model predictions
Soil moisture changes depend not only on the climatic conditions, but also on the soil
profile. AS2870 defines sites with clay soil deeper than 3 m as deep-seated moisture
sites. Braybrook has a clay soil profile deeper than 4.5 m; hence, it is one of the deep-
seated moisture sites. Therefore, a 6 m deep soil column was considered in Vadose/w
modelling for Braybrook site. The 1D Vadose/w model analysed for Braybrook to
consider the 1945-2015 period is denoted as VB1 hereafter.
The required soil parameters were found in Rajeev et al. (2012) for Fawkner, which is
another expansive soil site in Melbourne. This site is a nature strip located beside a
road, and it was used to investigate the effect of expansive soil on buried gas pipeline.
In contrast to Braybrook, Fawkner has a clay soil profile of up to only 2 m. Table 7-1
shows the details of the Fawkner soil which has a shallow profile. There are MH silts in
the top 300 mm. Brown coloured ashes were observed in the top layers, which appears
the signs of filled material. A high plasticity clay layer continues for only about 1.7 m
below the top layer before meeting the bedrock. Table 7-2 lists the soil properties of the
Fawkner site. Atterberg limits in the Fawkner site are lower than those of Braybrook,
which suggests that Fawkner soils are less reactive than Braybrook soils. This site was
not classified based on AS2870, because the Iss values are not available for Fawkner
soils. However, hydraulic and thermal properties are available in Rajeev et al. (2012)
Hs and ΔU were selected from the standard for the corresponding climate zone. Table
7-5 lists ys calculations for 25-year periods.
Table 7-5: Estimation of ys based on AS2870 (2011) for Braybrook
Period Hs (m)
ΔU (pF)
ys (mm)
Site class
1950-1974 1.8 1.2 66 H2 1960-1984 2.3 1.2 83 E 1970-1994 2.3 1.2 83 E 1980-2004 2.3 1.2 83 E 1990-2014 3.0 1.2 107 E
According to Table 7-5, the Braybrook site classification changed from H2 to E during
25-year periods. The influence of drought reduced the TMI value in the last two 25-year
blocks. Since Hs increased from 2.3 m to 3.0 m when changing the climate zone from 3
to 4, the estimated ys was increased.
The 25-year blocks were considered in the Vadose/w model predictions. In contrast to
the standard specifications, the model predictions have shown that the ΔU has changed
as a result of climate conditions. Hs of Braybrook is higher than that of Fawkner, but
both sites have shown no changes in Hs during the 25-year periods (Figure 7-12).
231
Figure 7-12: Changes in characteristic suction profiles within 25 year periods; a) Braybrook -VB1 model and b) Fawkner-VF1 model
Hs and ΔU were taken from the Vadose/w model, which is shown in Figure 7-12(a). The
crack depth is considered as 0.75 of Hs. Measured Iss values were considered along the
depth of the Braybrook soil profile. Table 7-6 summarises the ys estimation for
Braybrook using different methods. The calculations and model predictions shown in
Table 7-6 are broadly consistent with the AS2870 method calculations given in Table
7-5. The effect of the millennium drought can be observed in the last 25-year block in
1990-2014. Table 7-7 shows that ys of the Fawkner site has also increased due to the
inclusion of the drought period in 1990-2014, but the magnitude of increment is lower
than that of Braybrook. This is possibly due to the less reactivity and the shallower soil
profile in Fawkner. However, these values suggest that the impacts of extreme events
can possibly be captured using the AS2870 method by reducing the average period of
TMI.
1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5
-3.0
-2.7
-2.4
-2.1
-1.8
-1.5
-1.2
-0.9
-0.6
-0.3
0.0
Log (Matric Suction)
Dep
th (m
)
1950-1974 Characteristic min 1950-1974 Characteristic max 1960-1984 Characteristic min 1960-1984 Characteristic max 1970-1994 Characteristic min 1970-1994 Characteristic max 1980-2004 Characteristic min 1980-2004 Characteristic max 1990-2014 Characteristic min 1990-2014 Characteristic max
25 yr periods
a) b)
1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5
-6
-5
-4
-3
-2
-1
0
Log (Matric Suction)
Dep
th (m
)
1950-1974 Characteristic min 1950-1974 Characteristic max 1960-1984 Characteristic min 1960-1984 Characteristic max 1970-1994 Characteristic min 1970-1994 Characteristic max 1980-2004 Characteristic min 1980-2004 Characteristic max 1990-2014 Characteristic min 1990-2014 Characteristic max
25 yr periods
-0.3
232
Table 7-6: Estimation of ys for 25 year periods – Braybrook site
Period Vadose + AS2870 Method Vadose + FLAC
Method Hs (m)
ΔU (pF)
ys (mm)
Site class
ys (mm)
Site class
1950-1974 3.0 1.2 107 E 70 H2 1960-1984 3.0 1.2 107 E 70 H2 1970-1994 3.0 1.3 116 E 85 E 1980-2004 3.0 1.3 116 E 85 E 1990-2014 3.0 1.3 116 E 101 E
Table 7-7: Estimation of ys for 25 year periods – Fawkner site
Period Vadose + FLAC
Method ys (mm) Site class
1950-1974 38 M 1960-1984 32 M 1970-1994 37 M 1980-2004 37 M 1990-2014 44 H1
7.2.5 Effects of the depth of bedrock on ground movement
The bedrock depth is frequently observed to be within 2 m to 4 m in the Western
suburbs of Melbourne. However, in the Braybrook site, the depth to the bedrock is
greater. The model was therefore developed up to 6 m depth. The measured soil
properties were available down to only about 2.5 m depth. In this research, a second,
modified model was created using the same soil properties, assuming that the bedrock is
located at 3 m depth. This model is denoted as VB2 hereafter. The VB2 model can be
considered as a general case for typical basaltic soil sites in the Western suburbs. The
VB2 model was developed with a 4 m soil column and the bottommost meter was
considered as the bedrock. The SWCCs of soils were considered similar to the VB1
model. The Ksat of soils between 1.8 m and 3 m were considered to be decreasing
gradually from the measured value at 1.8 m to the Ksat of bedrock. Then, the hydraulic
conductivity functions were developed accordingly as explained in Chapter 4.
Figure 7-13 shows the characteristic suction profiles obtained from the VB2 model. The
surface suction values highly depend on extreme climate events; hence, they are similar
233
to the results of the actual model with bedrock at 5 m. However, the shallower bedrock
caused a reduction in the depth of suction change. The VB2 model predicts that the Hs
is approximately 2.7 m for the pre-drought period. The suction profiles in this period
can be considered as typical shapes similar to the VB1 model results. The millennium
drought moved the dry suction profile further towards the dry side and the wettest
suction profile recorded after drought-breaking rainfalls was within the typical suction
profile. Figure 7-13 also shows the idealised triangular shapes drawn to calculate ys
using the AS2870 guideline.
Figure 7-13: Changes in characteristic suction profiles -VB2 model; a) Extreme profiles in three periods b) Idealized triangles for AS2870 calculations
Figure 7-14 shows the variation of ground movement obtained for the Braybrook site
using the VB2 and FLAC models. The fluctuations of the ground due to seasonal
movements in this figure are similar to the predictions of the VB1 model shown in
Figure 7-9. The seasonal ground movement was about 40 mm during the pre-drought
period and reduced to 25 mm during the millennium drought. Figure 7-14 shows that
there was a total ground movement (peak to peak) of 85 mm prior to the millennium
drought. Approximately 16 mm of additional ground movement occurred due to the
millennium drought.
2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5
-4.0
-3.5
-3.0
-2.5
-2.0
-1.5
-1.0
-0.5
0.02.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5
-4.0
-3.5
-3.0
-2.5
-2.0
-1.5
-1.0
-0.5
0.0
Log (Matric Suction)
Dep
th (m
)
1945-1995 Characteristic min 1945-1995 Characteristic max 1996-2010 Characteristic max
a) b)
-0.3
Log (Matric Suction)
Dep
th (m
)
1945-1995 Characteristic min 1945-1995 Characteristic max 1996-2010 Characteristic max 2011-2012 Characteristic min
234
Figure 7-14: Ground movement prediction from VB2 and FLAC model
Table 7-8 shows a summary of estimation of ys using the VB2 model predictions and
the AS2870 method. The standard recommends similar Hs and ΔU values, even for the
modified case, because they are only dependent on climate conditions. Therefore, the ys
calculation for this modified case is essentially similar to the actual model (section
7.2.3) and produced no changes due to the inclusion of the millennium drought.
Although the Hs of the VB2 model is less than that of the VB1 model, the ys calculation
based on the Vadose and AS2870 method show similar ys values. The additional ground
movement of 16 mm due to millennium drought caused an increase of about 15% in the
typical ys. Similarly, in the Vadose and FLAC model, an additional 16 mm caused an
increase to the ys value of about 19%. According to the model predictions for 6 m and 3
m deep bedrock soil profiles, there was a 15% to 20 % increase in ground movement
due to the millennium drought.
Even though the depth of the bedrock is reduced in this analysis to consider the effect of
depth of bedrock, it is not considered to be less than the Hs limit of 3 m observed in the
VB1 model. It is only reduced to represent the typical depth of Bedrock in Melbourne
area. Hence, in this case, Hs is not affected by bedrock and the ground movement was
60 from VB3 model - Predicted wet climate: 2015 - 2018 from VB4 model - Predicted dry climate: 2015 - 2018 from VB1 model - Actual climate: 1945 - 2014
25 from VF2 model - Predicted dry climate: 2015 - 2018 from VF3 model - Predicted wet climate: 2015 - 2018 from VF1 model - Actual climate: 1945 - 2014
Date
Gro
und
mov
emen
t (m
m)
POND_OFF
240
7.4 LONG TERM CLIMATE PREDICTIONS
Climate predictions for Australia suggest that the ongoing drying trend will continue
into the future (Austroads, 2004, Smith et al., 2009, Hughes, 2003). Most Australian
cities will experience a reduction in rainfall and an increase in temperature. According
to the average of several scenarios, there would be a reduction in annual rainfall by
about 14% and an increase in temperature by 2.5 0C within the typical 100-year period.
These predictions were incorporated in the 1D models developed in Vadose/w software
to observe possible future changes in soil moisture content.
In this case, the climate data set period of 1987-1992 from Essendon airport was chosen
and considered as an average weather condition. This 5-year period was replicated over
50 years to create a typical average climate data set. Then, the Austroads (2004) climate
predictions were applied to this data set. Minimum and maximum temperatures were
modified to show a gradual increase of 1.25 0C at the end of the 50 years. A gradual
reduction of rainfall was also applied to the average data set, such that there is a total
reduction of 7% by the end of the 50-year period. 1D models developed in Vadose/w
software were analysed for these climate data sets for both the Braybrook and Fawkner
soil properties. The Braybrook models analysed using average and modified climate
data are denoted as VB5_A and VB5_M respectively. Similarly, the Fawkner models
are denoted as VF4_A and VF4_M.
Figures 7-19 and 7-20 show changes in ground movement due to predicted climate
changes. They clearly indicate that there would be an additional settlement in ground
surface due to predicted conditions compared to the average typical condition. The
Braybrook site indicated 12 mm of additional settlement, whereas Fawkner indicated 6
mm. The settlement in Braybrook appears to be severe compared to Fawkner, which is
due to the shallower bedrock and less reactive soil in Fawkner.
241
Figure 7-19: Effect of long-term climate predictions on ground movement – Braybrook
Figure 7-20: Effect of long-term climate predictions on ground movement – Fawkner
FLAC model predictions Vadose model predictions+ AS2870 method
Gro
und
mov
emen
t (m
m)
Distance (m)
6 m long cover slab Open area exposed to climate effects300 mm
Soil
Intial condition
Axis of symmetry
Extent of cover
255
7.6.3 Comparison of mound shape predictions
Based on the 2DVB2 and 2DVB3 model results described in the previous two sections,
important parameters for mound shapes can be identified. The ground movement
immediately under the edge of the slab (ym) and the movement in open ground (ys) can
be compared in each mound shape. The ratios between these two parameters are
proposed in Walsh’s and Mitchell’s methods as listed in AS2870 and shown in Table
7-9. Table 7-13 shows those ratios obtained from the modelling results presented in the
previous two sections. For the edge heave condition in 1983-1992 in the Braybrook site,
ym/ys factors are 0.7 and 0.8. For the edge settlement condition in 1992-2010 in the
Braybrook site, ym/ys factors are 0.6 and 0.8. These values are in line with the proposed
0.7 factor in Mitchell’s method. By using more analyses of different sites, these
relationships, including ‘e’ distance, can be established.
Table 7-13: Mound shape parameters obtained from models
Method
Edge heave (1983-1992)
Edge settlement (1992-2010)
ys (m)
ym (m) ym/ys ys
(m) ym (m) ym/ys
Vadose + AS2870 72 52 0.7 89 51 0.6
Vadose + FLAC 61 46 0.8 79 60 0.8
7.7 EFFECT OF ABNORMAL MOISTURE CONDITIONS
Climate conditions can significantly affect moisture conditions around and underneath
slabs, as described in the above section. These influences can be exaggerated due to
improper maintenance of the area adjacent to footings, which includes non-attendance
to broken water pipes. The author observed house construction sites and visited some
damaged houses during the study period. In most cases, the soils around the footing
were not given adequate attention during construction. In some cases, rainwater
downpipes were not connected to the drain lines, which created significant amounts of
water next to the slab edges during the construction period. Moreover, some sites have
open areas with slopes towards the footing. Some new houses had additions after
constructions, such as pavements with slopes towards the houses. As a result, runoff
water flowed towards the slab, which increased the amount of water available for
256
infiltration. In this study, the modelling of a sloping ground was used to investigate soil
moisture changes.
7.7.1 Soil slope towards the cover slab
In this case, a mild slope of 1:50 towards the cover slab was introduced over the 5 m
length of open area in the 2D Vadose/w model. The model was analysed from 1945 to
2012 to observe the suction changes in different climate conditions. This model will be
denoted as the 2DVB4_S model hereafter. Since the effect of the slope of open ground
and consequent water collection next to the slab was required in the 2DVB4_S model, a
ponding effect available in Vadose/w software was used. In this case, the model allows
water to travel along the dip and collect at the lowest point of the surface. Hence, more
water is available for infiltration at the edge of the slab. The 2DVB4_S model results
were compared with the 2DVB1 model, which had no slope in any direction and
pooling on the surface was also not allowed. Figure 7-32 shows the suction variation at
300 mm depth observed in the 2DVB1 and 2DVB4_S models.
Figure 7-32: Comparison of lateral moisture movement at 300 mm depth in with and without slope condition
0 1 2 3 4 5 6 7 8 9 10 11
2.6
2.8
3.0
3.2
3.4
3.6
3.8
4.0
4.2
4.4
4.6
4.8
5.0
0.45 pF
Soil
1
}Distance (m)
Log
(Mat
ric s
uctio
n)
1945-1995 Characteristic min 1945-1995 Characteristic max 1996-2010 Characteristic max 2011-2012 Characteristic min 1945-1995 Characteristic min 1945-1995 Characteristic max 1996-2010 Characteristic max 2011-2012 Characteristic min
Open area exposed to climate effects
300 mm
Soil
}from 2DVB1model
}from 2DVB4_Smodel
6 m long cover slab
50Axis of symmetry
257
Since rainwater can flow towards the cover, the moisture content of the soils next to
slab edge is higher in the 2DVB4_S model than in the 2DVB1 model. Specifically, the
amount of water available to move laterally beneath the cover slab is substantially
increased. This phenomenon reaches significant proportions in 2011 and 2012, the two
years of above average rainfalls after the millennium drought. During this wet period,
suction at the edge of the slab is 0.45 pF lower in the 2DVB4_S model than in the
2DVB1 model, which emphasises the influence of water flow, and collects near the slab
edge. The decrease in moisture content during the drought period was also lower in the
2DVB4_S model predictions than in the 2DVB1 model. This is because the additional
amount of water available at the slab edge moves into the soils beneath and decreases
the deficit. Taken together, the model results suggest that soil moisture changes beneath
covers can be significantly increased due to slopes towards the footing.
7.8 SUMMARY
This chapter describes the applications of the finite element models developed in this
study to predict soil moisture changes. The models were used to determine soil moisture
changes due to several long-term climate scenarios and site conditions. In addition to
the Braybrook site, Fawkner, another reactive soil site in Melbourne, was also
considered. Fawkner has a shallower soil profile and less reactive soils than Braybrook.
Each analysed model was denoted based on the considered climate data set and the site
conditions. The model predictions were then used to obtain ground movements using
three different methods - the AS2870 method, the Vadose + AS2870 method and the
Vadose + FLAC method.
The long-term climate data were taken from Essendon airport weather station for the
analysis period of 1900 to 2015. The initial conditions were not known for these periods
and hence the initial conditions corresponding to calibration (given in Chapter 6) were
used. However, the model results for the first 30 to 40 years were ignored to avoid the
effects of the initial condition. The model results considered the period commencing 50
years before the millennium drought. This period was divided into three sections based
on the millennium drought: pre-drought (1945-1995), millennium drought (1996-2010)
and the wet period after the drought (2011-2012). The condition in the 1945-1995
period was considered as the average climate condition during a lifespan of a structure.
258
ΔU and Hs values were obtained from the models’ results for different periods. There
was a very high suction change observed at the ground surface, which is due to direct
interaction with climate conditions. Nevertheless, during the construction of footings,
the top soil is normally removed and the slab would typically be founded below the
natural ground surface. Therefore, in this calculation, ΔU is considered at 0.3 m depth,
below the ground surface.
The predictions of VB1 (the Braybrook model for 1945-2012) suggest that between
1945 and 1995, ΔU and Hs for Braybrook were 1.3 pF and 3.0 m respectively.
Similarly, VF1 (the Fawkner model for 1945-2012) suggests that Fawkner had ΔU and
Hs of 1.5 pF and 1.8 m respectively. Deep-seated moisture change was observed in the
VB1 model as Braybrook has a deep soil profile. However, because of the shallower
soil profile in Fawkner, the VF1 model showed a smaller Hs value than VB1 for the
Fawkner site. Since the moisture movements continue at a deeper depth in Braybrook,
the fluctuations and changes near the surface (at 0.3 m depth) were less than those of
Fawkner. Both the VB1 and VF1 models predicted that these sites experienced an
additional increase of suction of 0.2 pF due to the moisture deficit created by the
millennium drought. Ground movement predictions of the FLAC model using the VB1
and VF1 model results showed variations in seasonal ground moments. These results
suggest that the millennium drought was the worst drought recorded during the past 70
years and highlight the settlement trend during the 1996-2009 period. According to the
ground movement calculations from these model results, the millennium drought
created a 15% to 20% increase in the ground movement experienced in the pre-drought
period. This increment was not captured by the AS2870 method, which depends on
average TMI over the calculated period. However, when the TMI was calculated in 25-
year blocks instead of over the total period, the impact of the drought was reflected in
the ground movements.
The VB2 model was analysed to consider the effect of the depth of bedrock. Here, the
bedrock of Braybrook was assumed to be at 3 m, which represents the typical condition
of a basaltic soil site. This model resulted in a lesser Hs value (2.7 m) but the ΔU was
the same as in VB1, indicating the influence of bedrock on the depth of Hs. ΔU
primarily depends on soil properties and extreme climate conditions. However, the VB2
259
model also predicted a 15% to 20% increase in ground movement due to the millennium
drought.
The Vadose/w software considers the runoff when the rainfall is distributed over 24
hours of the day provided that the surface layer is saturated. The daily rainfall is treated
in Vadose/w models such that it is a sinusoidal distribution over 24 hours and hence,
even if the daily intensity is high, the hourly intensity can be low. However, in most
rainy days, rainfall is not evenly spread throughout but falls at high intensity during
certain parts of the day, and most of such rainwater runs off, allowing only a small
amount to infiltrate the soil. Therefore, in such intense rainy days, a further correction is
added to the data before they input to the model. A surface runoff correction of 60%
(typical value for urban residential areas) was considered in all the models discussed
here. Further models were analysed to determine the influence of site drainage
conditions on ground movement. In this case, 30% and zero runoff corrections were
considered. A lower surface runoff is associated with greater availability of water for
infiltration. Hence, poor drainage conditions will create greater ground movements.
Compared to the 60% runoff considered in VB1, 30% and zero runoff increase the
maximum observed ground movement by 20% and 45%, respectively.
The 1D models were also used to determine the ground movement due to predicted
climate conditions. In this case, 4 years of wet and dry climate conditions were
forecasted for 2015-2018 using past climate data. These climate data sets were used to
analyse 1D models from 1945 to 2018. The model results from a forecasted wet climate
condition suggest that the ongoing settling trend will terminate due to wet conditions
and the ground conditions will be similar to those experienced in 2011. However, 4
years of wet climate conditions is not sufficient to allow the grounds to fully recover
from the moisture deficit caused by the millennium drought period. However, in actual
condition, cracks allow rainwater to move deep into the soil and accelerate the wetting
process. Therefore, in actual condition, a further recovery could have been observed by
such 4 years of above average rainfall. The model results from a forecasted dry climate
condition showed that a dry period will drag the soils back to a drought condition at the
end of 2018. The changes in ground movement in the Fawkner site due to these
different climate conditions are smaller than in Braybrook. This is due to the shallower
and less reactive soil profile in Fawkner.
260
Moreover, long-term climate predictions were also considered using 1D models.
Reported climate predictions suggest an ongoing drying with an increase in temperature
and a decrease in rainfall. These changes were applied to a data set of typical average
climate conditions over 50 years. Model predictions suggest that both Braybrook and
Fawkner soils will experience a moisture deficit due to an ongoing drying compared to
the typical average condition. At the end of the 50-year period, Braybrook will
experience an additional 12 mm settlement, whereas Fawkner will experience half of
that.
The 2D model developed in Vadose/w software was used to determine the soil moisture
changes in the lateral direction. A perfectly flexible cover was introduced into the 2D
model to investigate the soil moisture changes beneath the cover due to climate
influences on an adjacent open area. The results of this model (2DVB1) were also
useful in identifying changes in ‘e’ distance due to different climate conditions. The ‘e’
distances obtained from these models were compared with those of the AS2870 method.
The model predictions suggested that the millennium drought has created an increase of
15% to 20% in ‘e’ distances. Since the AS2870 method produced the same ys for the
1945-1995 and 1945-2009 periods, the change in ‘e’ distance was not captured using
this method.
Changes in soil mound shapes beneath cover slabs were also investigated using the 2D
model results. To observe edge heave and edge settlement conditions, two periods of the
most extreme wet and dry climate conditions were identified: 1983-1992 and 1992-2010
respectively. Then, two different models were analysed for climate conditions in these
periods. The moisture predictions of these models were used in a one-dimensional soil
column in the FLAC model to determine the ground movement at several discrete
points along the distance from the axis of symmetry. ys values were also calculated
based on the AS2870 method using characteristic suction profiles at these discrete
points. Considering mound profiles in both edge heave and edge settlement conditions,
the calculations based on the AS2870 method are in strong agreement with the FLAC
model predictions. However, the FLAC model predictions showed a steep slope in
mound profiles near the edge of the slab. This is possibly because the one-dimensional
soil column in the FLAC model does not consider the shear effects between adjacent
soils. The shear interaction in adjacent soils will reduce sudden high deflection of
261
discrete points and hence produce a smooth mound profile. A full 2D model in FLAC
will overcome this issue; however, development of the 2D model in FLAC is yet to be
completed and is beyond the scope of this thesis.
Further analyses in the 2D models in Vadose/w software were performed to observe the
effects of abnormal moisture sources on soil moisture changes. A cover slab was
modelled with adjacent soils dipping towards it. The sloping ground creates runoff
towards the slab edge and increases the water available for infiltration. This model
(2DVB4_S) was analysed for 1945 to 2012 and then compared with the control model
(2DVB1). The predictions indicated that the sloping ground significantly increased the
soil moisture at the edge of the slab. In fact, during the wet period in 2010 and 2011, the
characteristic suction at the slab edge was 0.45 pF lower in the sloping ground case
compared to the control model.
262
8. CONCLUSIONS AND FUTURE WORK
8.1 OVERVIEW OF THE STUDY
Prior investigations have suggested that the Victorian climate has changed over the last
few decades. In addition, back-to-back extreme climate events were observed in the
recent past and more of these events are expected in the future. These changes affect the
soil moisture conditions and have a severe impact on volume changes in expansive
soils. Recent media reports and anecdotal evidence suggest that a large number of
houses were damaged as a result of footing movements which may have been caused by
climate related soil moisture changes. Hence, it is timely that the consideration of
climate influence and the procedure of estimating ground movement in the standard of
residential footing design are reviewed.
The procedure outlined in the AS2870 for calculating characteristic ground movement
was investigated, with a particular emphasis on climate influences. A field site was
established to monitor the expansive soil behaviour. A site in Braybrook, which is in the
Western suburbs of Melbourne, was selected for field monitoring. Braybrook has a
consistent profile of extremely reactive basaltic clays. The purpose of the field site was
to collect a comprehensive dataset of soil properties and monitor soil moisture changes
and the subsequent ground movement over several seasons. A one dimensional finite
element model using Vadose/w software was developed to investigate the soil moisture
changes in response to climate conditions. The finite element model was based on the
measured expansive soil properties of Braybrook site and the climate data collected
from a nearby weather station to define boundary condition of the model. The model
was validated against the soil moisture data collected from the Braybrook site
monitoring. The model was then extended to a two dimensional model to observe the
moisture changes in soil beneath cover slabs due to climate influences on adjacent open
ground. These models were used to analyse the soil moisture changes due to long-term
climate conditions including recent extreme events. Furthermore, the soil moisture
changes in response to various climate scenarios were examined. The soil moisture
predictions were used as the input to another model developed by another researcher
using FLAC3D as a part of this comprehensive research study. The FLAC3D model
was developed to predict the ground movements due to the changes in soil moisture
263
conditions. The development of the FLAC3D model is out of the scope of this PhD
thesis however, the soil moisture and ground movement predictions are discussed in this
thesis. The model predictions of moisture and suction changes and ground movements
due to different climate conditions were compared with the estimations based on the
AS2870 method.
8.2 SUMMARY OF CONCLUSIONS
8.2.1 Ground movement, climate changes and TMI
The estimation of ground movement in footing design is dependent on factors affecting
the volume change behaviour of expansive soils, such as degree of reactivity of the soil
and amount of moisture change. The moisture changes in soil are created by several
sources including climate condition and manmade causes.
The method of estimating ground movement in the AS2870 provides a simplified
approach to obtain ys value of an undeveloped site for the purpose of site classification.
This method uses Iss to account for the degree of reactivity of soils. AS2870 considers
soil moisture changes by means of idealized suction profile which is defined using ∆U
and Hs. The standard procedure allows for correlation of Hs with the climate condition
of the area using the TMI. The depth of bedrock and water table affect the Hs. The
Standard provides a single ΔU value (1.2 pF) for all parts of Australia, irrespective of
the climate or site condition.
There are certain weaknesses in AS2870 in estimating ground movement one of which
is the lack of definition of the basis of TMI assumed in the Standard. Several methods
are available to calculate the TMI and each method can produce different values. Four
methods for calculating TMI were presented in Chapter 3 and their results compared,
Since each TMI calculation method produces different values, the correlation between
TMI and Hs generates different values for the same climate condition. This can result in
different footing designs. It was found that the method noted as “Method 1” produces
the closest results to those in AS2870.
In addition, the use of different averaging periods for TMI also produces different
values. Specifically, the more years used in the average calculation, the lesser the
sensitivity to isolated extreme weather events, such as droughts. The long-term average
264
TMI is appropriate to reflect long-term trends. However, for the residential footing
design, this averaging period must be considered together with more soil specific
parameters. Furthermore, it was found that the TMI largely depends on rainfall and,
hence, linear correlations were observed between annual rainfall and TMI variations in
most of cities in Victoria. The influence of the other climate components, such as
evaporation and relative humidity are not critical in determining soil moisture changes.
Furthermore, the climate zone map given in the Standard was developed based on
climate data from 1940-1960. However, in Australia, there has been a change in the
climate over the last few decades. The TMI trend of the last 50 years is clearly showing
the ongoing drying. If the same trend continues, it means that the TMI will also reduce.
The drying trend is forecasted by climate models such as that by CSIRO. Irrespective of
the TMI calculation method, it appears that there is an ongoing drying in the Victorian
climate. The modifications to AS2870 in the 2011 edition captured the changes of TMI
due to the drying effect experienced in the last 25 years. However, this may not be
sufficient to capture ongoing changes predicted for the future. Moreover, the values of
Hs and ΔU have not been updated in the Standard to reflect the recent changes and
possible future changes.
The findings presented in Chapter 3 also highlight an issue in estimating the degree of
reactivity of soil in terms of Iss. The shrink swell test, which is used to determine Iss, is
recommended in AS2870 to assess soil reactivity. The corresponding standard of the
test (AS1289) specifies Iss as a constant for a given soil type. In this research, soils
collected from different sites at different times of the year were tested and a number of
Iss values were obtained for the same soil. The results indicated that the Iss increases
with increasing in situ moisture. Even though the experimental data is limited, this
suggests the dependency of Iss on in situ moisture content, which is in contrast to the
assumptions outlined in the AS1289.7.1.1. Specifically, it can result in different footing
designs for a particular site, when the soils are tested at different times of the year.
Hence, Iss values should be reported with the initial moisture content of the soil and the
results considered in relation to the site climatic conditions.
265
8.2.2 Characterization of typical basaltic clay in Western Melbourne
As part of the research study described in this thesis, a field site was established to
collect the required data related to expansive soils in Melbourne. A comprehensive data
set was developed for the typical basaltic clay soils as found in Braybrook, located in
West Melbourne.
The data set consists of basic soil properties and some specific expansive soil
parameters. The basic properties include Atterberg limits, linear shrinkage, specific
gravity and clay content. The plastic limit of Braybrook clay varies from 20 to 25 %
while the liquid limit varies from 70 to 80%. Based on these results, the Braybrook soil
was categorised as CH type in the Unified soil classification. The linear shrinkage
varies by approximately 20% and the clay content of soils below 0.5 m is about 45%.
The clay content of the surface layer is less than that of deep soils, which reflects the
presence of silts and organic matter in the top soil. All of these properties are consistent
in the soils below the surface layer. The bedrock was not encountered in the Braybrook
site, even though the boreholes were cored down to 4.5 m.
In addition to the basic soil properties, some specific properties were investigated to
classify the site and examine the behaviour of expansive soils. The shrink swell index
was calculated to vary from 4 to 6% and, based on this index, the Braybrook was
classified as an extremely reactive site. However, X-Ray Diffraction tests revealed that
there was more than 50% of Quartz in the mineral composition of Braybrook clay.
However, Braybrook soils have less than 10% of sand. Hence, this Quartz content could
represent fine sediments, which have clay size and silt size minerals. Importantly, there
is more than 30% of Montmorillonite in the mineral composition in Braybrook clay,
which is the primary cause of the expansiveness.
In addition to the clay mineralogy, the hydraulic conductivity and SWCC functions
were developed for Braybrook soils to use in the prediction models. Matric suction and
volumetric moisture contents were employed which provided the coordinates of the
SWCC. Hyprop, WP4C and filter papers were used to measure the suctions values at
different moisture levels. Osmotic suction obtained from the filter paper method was
used to convert total suctions into matric suction measured from WP4C. A correlation
was developed between volumetric and gravimetric moisture contents of Braybrook
266
soils. This correlation was used to obtain volumetric moistures from measured
gravimetric moistures corresponding to WP4C and filter paper suctions. Since the
Braybrook site has a constant soil profile throughout the depth, the SWCC functions are
assumed to be similar in deep soils. However, the presence of organic matter caused a
different SWCC for the surface soils.
The saturated hydraulic conductivity of surface soils was found in the range of 10-7 m/s.
The hydraulic conductivities were reduced in deeper soils due to a higher clay content
and density. The saturated hydraulic conductivities of deeper soils were in the range of
10-9 m/s. Hydraulic conductivity functions of unsaturated soils were developed using
these values and SWCCs.
This unique data set is beneficial for both practitioners and researchers. It provides the
characteristics of typical basaltic clay found in West Melbourne, which are useful in site
classification and modelling.
8.2.3 Field monitoring of expansive soil behaviour
The field monitoring of a typical expansive soil site was performed in this study to
collect the data required to calibrate and validate the prediction models. The soil
moisture changes were monitored using neutron probe technique and the corresponding
ground movements were monitored using magnetic extensometers. Paving blocks were
placed at several locations to monitor the ground movement using a surveying level.
The neutron probe reads the number of neutrons that react with soil moisture during
measurements. Therefore, in this study, a calibration equation was developed between
neutron counts and volumetric moisture content. The calibration equation, which has
0.86 coefficient of determination, provides the corresponding volumetric moisture
content to the neutron count measurements. The field monitoring continued regularly
from April 10th, 2013 to March 25th, 2015. Seasonal moisture variations were observed
at three locations over the two-year period. All three locations showed similar moisture
profiles at deeper depths. However, there were some differences in near surface
measurements possibly due to local effects such as slope differences, potholes and grass
cover changes. The soil moisture changes were observed up to 1.25 m over the
monitoring period. The most significant changes were observed in the top 0.75 m soils.
Moisture contents of near surface soils followed the rainfall pattern with a certain time
267
lag. The seasonal heave and settlements were observed in three magmatic extensometers
located next to the neutron probe access tubes. Similar to the moisture changes, most of
the movements occurred in the top soil layers. The maximum seasonal ground
movement was in the range of 40-50 mm.
8.2.4 Finite element modelling approach of expansive soil and climate
interaction
The expansive soil properties and the data collected from regular monitoring were used
to develop finite element model using Vadose/w to predict soil moisture changes due to
climate conditions. The model requires specifying SWCC, hydraulic conductivity,
thermal conductivity and the specific heat capacity of the soil. The thermal properties
had a minimal impact on soil moisture changes, as suggested by sensitivity analysis.
Therefore, thermal conductivity and specific heat capacity were obtained from a site
close to Braybrook described in the literature. The soil properties were defined to a
depth of 6 m deep soil column to represent the soil profile. The climate boundary was
specified at the top surface and it includes daily inputs of rainfall, evaporation, relative
humidity, wind and temperature. These climate components were collected from
Essendon airport - a weather station close to the Braybrook site. The no-flow boundary
was defined at the bottom to simulate the effect of bedrock.
This model was analysed for the two-year period of field monitoring from 2013 to 2015.
The model predicts soil moisture and suction profiles as daily outputs. These predictions
were compared with the neutron moisture measurements. The results showed a good
agreement between soil moisture predictions and field measurements and hence the
model was considered as validated.
In order to study the soil moisture beneath a flexible cover, the 1D model was extended
to a 2D model. The observations of homogeneous soil in Braybrook soil suggest that the
lateral hydraulic conductivity, which is the only additional parameter required in the 2D
model, is similar to the hydraulic conductivity in the vertical direction. The results at
open area exposed to climate effects in the 2D model showed same moisture changes as
1D model for the period of field monitoring. However, the results of mound shapes
obtained from 2D models depend on the assumptions of lateral hydraulic conductivity.
The sensitivity analysis of the input parameters revealed that soil moisture changes are
268
mostly affected by the SWCC of soils followed by the hydraulic conductivity. A 20%
change in SWCC and hydraulic conductivity resulted in 22% and 2% change in soil
moisture, respectively, at 0.3 m below the surface. Rainfall is the most critical climate
component in changing soil moistures. The responses of the models to vegetation
effects and pooling situations were also considered. Grass covers prevent the
evaporation and increase the water available for the infiltration. Therefore, both grass
covers and pooling effects significantly increase the soil moisture.
The effect of cracks was not included in these models. The presence of cracks can
significantly change the moisture content of soils even at deeper depths. The cracks
allow rainwater to infiltrate and hence increase the moisture content of deep soils.
Therefore, in actual conditions during the rainy days of summer, the moisture may
increase abruptly. This is not captured in the models described in this thesis.
8.2.5 Prediction of ground movement due to several site conditions and
climate scenarios
The finite element models were used to investigate the soil moisture, suction and
ground movement due to long term climate conditions. In addition to the Braybrook
site, Fawkner - another reactive soil site in Melbourne - was considered. Fawkner has a
shallower soil profile and less reactive soils compared to Braybrook. The long-term
climate data taken from Essendon airport weather station were considered in the
analyses as Braybrook and Fawkner sites are close to each other. The model results
were considered for a period starting 50 years before the severe millennium drought.
The total period was divided in three based on the millennium drought; pre drought
(1945-1995), millennium drought (1996-2010) and the wet period after the drought
(2011-2012).
The predictions of the models developed in Vadose/w software were used to obtain ΔU
and Hs values in different periods. The corresponding ground movements were
calculated using the AS2870 method. In addition, the model results were fed into a one-
dimensional soil column developed in FLAC-3D software to predict ground movement.
The predictions suggest that during 1945-1995, ΔU and Hs of Braybrook were 1.3 pF
and 3.0 m, respectively. Deep-seated moisture change was observed in Braybrook
because of the deeper soil profile. The 1D model was modified to determine these
269
parameters if Braybrook had a shallower soil profile with bedrock at 3 m, which is the
general condition of basaltic soil sites in West Melbourne. However, the modified
model results suggested only a small reduction of Hs (2.7 m). ΔU appears to be a
dependent on climate conditions and properties of top soils.
Even though the climate conditions are the same in both Fawkner and Braybrook, the
Fawkner site had ΔU of 1.5 pF and Hs of 1.8m during 1945-1995. This is due to the
shallower soil profile in Fawkner. Further, the hydraulic conductivities of Fawkner soils
below the top layer are lower compared to Braybrook. Therefore, moisture flow is
restricted through deeper soils, which creates higher fluctuations in top soils and hence
higher ∆U.
Both Braybrook and Fawkner sites showed an additional increase of ΔU of 0.2 pF at the
end of the millennium drought that resulted in additional ground movement. The FLAC
model showed that the millennium drought produced the greatest ground movement
(settlement) during the past 70 years. In fact, the model results suggest that the
millennium drought created a 15-20% increase in the ground movement compared to
the pre-drought period. AS2870 provides a method to calculate ys for an undeveloped
site experiencing normal climate condition. Therefore, isolated extreme climate
conditions may not be captured in the AS2870 approach. Consequently, the impact of
the millennium drought would not have been captured in its estimation of the surface
movement, which depends on average TMI over a long period. However, when the TMI
was calculated in 25 year blocks instead of 50 year period, the impact of the drought
was reflected in the calculated ground movements.
A surface runoff correction of 60% (typical value for urban residential area) was used in
the prediction models. However, the effect of different site drainage condition on soil
moisture and ground movement was also investigated. A lower surface runoff results in
more water available for infiltration. Hence, poor drainage condition will create higher
ground movements. Compared to the 60% runoff correction considered, 30% and zero
runoff corrections will increase the maximum observed ground movement by 20 and
45% respectively.
Further, the validated model was used to determine the ground movement due to future
climate scenarios. In this case, 4 years of wet and dry climate conditions were
270
artificially forecasted for the period 2015-2018 using past climate data. These climate
data sets were used to analyse the 1D model from 1945 to 2018. The model results in
both Braybrook and Fawkner suggested that the ongoing settling trend would terminate
and the ground will reach a similar condition to that experienced in 2011. However, 4
years of wet climate condition is not sufficient to allow the grounds to completely
recover from the moisture deficit caused by millennium drought. However, in actual
condition, cracks allow rainwater to move deep into the soil, and accelerate the wetting
process. Therefore, in actual condition, a further recovery could have been observed by
such four years of above average rainfall.
Long-term climate predictions were also considered using the 1D models. A set of
climate data representing typical average climate condition was considered. It was
modified to integrate the ongoing drying trend reported by climate predictions
(Austroads, 2004). Models predictions suggest that both Braybrook and Fawkner soils
would experience a moisture deficit due to ongoing drying compared to the typical
average condition. At the end of 50 year period, Braybrook would experience an
additional 12 mm settlement whereas Fawkner would experience half of that.
The 2D model developed in Vadose/w software was used to determine the soil moisture
changes in the lateral direction. A perfectly flexible cover was introduced in the 2D
model to investigate the soil moisture changes beneath the cover due to climate
influences on an adjacent open area. The 2D model was used to determine the changes
in soil mound shapes beneath a flexible cover. To observe the highest edge heave and
edge settlement conditions, two periods of the worst wet and dry climate condition were
identified;1983-1992 and 1992-2010, respectively. Then, two different models were
analysed for climate conditions in these periods. The soil moisture predictions were
used in the FLAC model to determine the ground movement at several discrete
locations measured from the axis of symmetry. The ys values were also calculated based
on the AS2870 method using characteristic suction profiles at these discrete locations.
Considering mound profiles in both edge heave and edge settlement conditions, the
calculations based on AS2870 method are in strong agreement with the FLAC model
predictions. The model results suggest that this approach is effective in predicting
mound shapes for slabs placed at different times (i.e., slabs constructed during wet or
dry periods).
271
Additional analyses using the 2D models were performed to observe the effects of
abnormal moisture sources on soil moisture changes. The condition of a cover slab with
adjacent surface slope towards it was modelled. The sloping ground creates a runoff
towards the slab edge and increases the water available for infiltration. The predictions
indicated that this case resulted in a significantly wet condition in soils at the edge of
the slab. In fact, during the wet period in 2010 and 2011, the characteristic suction at the
slab edge is 0.45 pF lower in sloping ground case compared to the control model with
no slopes. .
In conclusion, the models developed in this study provide a comprehensive and versatile
approach to investigate soil moisture and ground movement due to different climate and
manmade conditions. They can be used not only to obtain ΔU and Hs but also to
determine variation in ground movement and mound shapes. Consequently, correlations
of the ground movement induced by various climate scenarios can be established using
further investigations of different soil types. These models will therefore greatly assist
the development of design tools for the footings of light structures.
8.3 RECOMMENDATIONS FOR FUTURE WORK
Even though, the research described in this thesis is a part of a comprehensive group
research programme, there were some limitations in the study, which opens various
paths for future research.
The models developed in this study were validated only for the Braybrook site for a
period of 2 years. However, more analysis is required, including different sites with
various soils and site conditions. Further analysis is essential for the provision of
generalized conclusions in expansive soil behaviour in response to the changes in
climate conditions.
The Vadose/w model developed in this study considers a crack free soil profile.
However, most of the clay soils experience cracking during dry weather conditions. In
fact, there were more than 1 m deep cracks observed in Braybrook site during the
summer. This condition can result in sudden increments of the moisture contents at deep
soils due to infiltration of runoff water through the cracks. It can affect the depth of soil
moisture change. Moreover, the cracks can reduce the vertical soil movement due to
272
availability of space to volume changes in the lateral direction. The cracking of soils
depends on various factors, including the climate, soil type and the local effects. Hence
it is difficult to generalize the way of crack propagation. However, since the cracks can
impact Hs and ys values, adopting the cracking behaviour in models could be effective
in designing residential structures on expansive soils.
In addition, the differential equations used in the Vadose/w model are limited only for
the “No snow” condition in climate. Hence, modifications must be employed
accordingly to use this model in analysing soil moisture changes in alpine areas.
Furthermore, the Vadose/w model only considers the effect of vegetation in terms of
grass cover. In most cases, grass covers reduce the amount of solar radiation available
on soil surface. This reduces the evaporation resulting in an increment in soil moisture.
Importantly, the effect of tree roots is different from grass covers. Tree roots absorb
moisture from deep soils and reduce the soil moistures. Therefore, the soil experiences
additional deficits and settlement due to trees. This phenomenon has been a common
cause for differential settlements in houses with trees in the influential zone. The
influence of trees must be considered in reference to a structure which is affected by
various factors, including the soil type, distance from the structure, type of tree canopy
and distribution of roots. The model developed in this study can be extended in future
research to include the influence of trees.
In addition to the potential advancements in the Vadose/w model, the development of
the 2D FLAC model is also an area for future research. The development of the 2D
FLAC modal began as part of this comprehensive research program but is yet to be
finalized. The 2D FLAC model overcomes the issues of neglecting shear effects of
adjacent soils in the 1D FLAC model when estimating ground movement profiles. The
2D FLAC model will predict the ground movement underneath cover slabs and,
therefore, will be highly useful in estimating mound shapes of slabs placed at different
times and exposed to different climate scenarios.
Apart from the possible upgrades to the models, the results of the model described in
this thesis can be utilized into a standard procedure. Since this model can produce soil
moisture and ground movement due to different climate conditions, they can be
categorized into a rationalized form. For example, the climate conditions could be
273
classified based on the severity of the extreme events and the prediction scenarios. The
severities can be categorized based on return periods. Then, the influence of those
climate events on footing design parameters can be specified for use in design
guidelines, for example changes in ΔU, Hs and ys based on the severity of extreme
events. This will allow footing designers to obtain relevant parameters based on the
expectable climate conditions, quality and financial feasibility of the construction rather
than depending solely on the past climate conditions. For, example; if a home owner
desires a house which can withstand severe climate events, the model can be employed
to obtain ground movements for climate conditions with appropriate extreme events.
However, such designs may require higher construction cost. Moreover, such a
procedure requires model predictions from a number of expansive soil sites.
274
9. REFERENCES
ACA. 2012. A Current Affair Channel 9 [Online]. http://aca.ninemsn.com.au/investigations/8424660/homes-cracking-under-pressure [Accessed 01-03- 2012].
AITCHISON, G. D. 1965. Moisture Equilibria and Moisture Change in Soils Beneath Covered Areas. Statement of the Review Panel, Engineering Concepts of Moisture Equilibria and Moisture Change in Soils. Butterworths, Sydney.
AITCHISON, G. D. & HOLMES, J. W. 1953. Aspects of Swelling in the Soil Profile. Australian Journal of applied Science, 4, 244.
AITCHISON, G. D. & RICHARDS, B. G. 1965. A Broad scale Study of Moisture Conditions in Pavement Subgrades throughout Australia. Part 1 - 4. Butterworths, Sydney.
AL-RAWAS, A. A. & GOOSEN, M. F. A. 2006. Expansive Soils: Recent Advances in Characterization and Treatment, Taylor & Francis.
ALONSO, E. E., GENS, A. & JOSA, A. 1990. A constitutive model for partially saturated soils. Geotechnique, 40, 405 –430.
ALONSO, E. E., VAUNAT, J. & GENS, A. 1999. Modelling the mechanical behaviour of expansive clays. Engineering Geology, 54, 173–183.
ALTMEYER, W. T. 1955. Discussion of Engineering properties of expansive clays. Proceedins ASCE, 81(separate No. 658).
AS1289.3.1.1 2009. Methods for testing soils for engineering purposes. Soil classification tests - Determination of the liquid limit of a soil - four point Casagrande method. Sydney, Australia.: Standards Association of Australia.
AS1289.3.2.1 2009. Methods for testing soils for engineering purposes. Soil classification tests - Determination of the plastic limit of a soil - Standard method. Sydney, Australia.: Standards Association of Australia.
AS1289.3.4.1 2008. Methods for testing soils for engineering purposes. Soil classification tests - Determination of the linear shrinkage of a soil - Standard method. Sydney, Australia.: Standards Association of Australia.
AS1289.7.1.1 2003. Methods for testing soils for engineering purposes. Determination of the shrinkage index of a soil; shrink swell index. Sydney, Australia.: Standards Association of Australia.
AS2870 1986. Residential slabs and footings. Sydney, Australia.: Standards Associatiion of Australia.
AS2870 1996. Residential slabs and footings - Construction. Sydney, Australia.: Standards Associatiion of Australia.
AS2870 2011. Residential slabs and footings. Sydney, Australia: Standards Association of Australia.
ASTM-D422 2007. Standard Test Method for Particle-Size Analysis of Soils. D422 - 63. ASTM International.
ASTM-D854 2010. Standard Test Methods for Specific Gravity of Soil Solids by Water Pycnometer. D854. ASTM International.
ASTM-D2487 2011. Standard Practice for Classification of Soils for Engineering Purposes (Unified Soil Classification System). D854. ASTM International.
ASTM-D4943 2008. Standard Test Method for Shrinkage Factors of Soils by the Wax Method. D4943. ASTM International.
ASTM-D5084 2003. Standard Test Methods for Measurement of Hydraulic Conductivity of Saturated Porous Materials Using a Flexible Wall Permeameter. D5084 - 03. ASTM International.
ASTM-D5298 2003. Standard Test Method for Measurement of Soil Potential (Suction) Using Filter Paper. D5298. ASTM International.
ASTM-D6836 2008. Standard Test Methods for Determination of the Soil Water Chararcteristic Curve for Desorption Using a Hanging Column, Pressure Extractor, Chilled Mirror Hygrometer, and/or Centrifuge. D6836. ASTM International.
ATWELL, B. J., KRIEDEMANN, P. E. & TURNBULL, C. G. N. 1999. Plants in Action - Adaptation in Nature, Performance in Cultivation [Online]. Macmillan Education Australia Pty Ltd, Melbourne, Australia: http://plantsinaction.science.uq.edu.au/edition1/. [Accessed 25-02- 2015].
AUSTROADS 2004. Impact of Climate Change on Road Infrastructure. Sydney, Australia.
BANDYOPADHYAY, S. S. 1981. Prediction of Swelling Potential for Natural Soils. Journal of the Geotechnical Engineering Division, 107, 658-661.
BARNETT, I. C. & KINGSLAND, R. I. 1999. Assignment of AS2870 Soil Suction Change Profile Parameters to TMI Derived Climate Zones For NSW. Australian Geomechanics, 34, 25-31.
BARRY-MACAULAY, D., BOUAZZA, A. & SINGH, R. 2011. Study of Thermal Properties of a Basaltic Clay. Geo-Frontiers 2011.
BARRY-MACAULAY, D., BOUAZZA, A., SINGH, R. M., WANG, B. & RANJITH, P. G. 2013. Thermal conductivity of soils and rocks from the Melbourne (Australia) region. Engineering Geology, 164, 131-138.
BELCHER, D. J., SACK, H. S. & CUYKENDALL, T. R. 1950. The measurement of soil moisture and density by neutron and gamma-ray scattering, Indianapolis, Civil Aeronautics Administration, Technical Development and Evaluation Center.
BENSON, C. H. & TRAST, J. M. 1995. Hydraulic Conductivity of Thirteen Compacted Clays. Clays and Clay Minerals, 43, 669-681.
BIDDLE, G. 2001. Tree Root Damage to Buildings. Expansive Clay Soils and Vegetative Influence on Shallow Foundations.
BIDDLE, P. G. 1983. Patterns of soil drying and moisture deficit in the vicinity of trees on clay soils. Géotechnique, 33, 107-126.
BIDDLE, P. G. 1998. Tree root damage to buildings, Wantage, UK, Willowmead Publishing Ltd.
BOM 2012. Special Climate Statement 38. National Climate Centre: Bureau of Meteorology, Australia.
BOM. 2015a. Average annual, seasonal and monthly rainfall [Online]. http://www.bom.gov.au/jsp/ncc/climate_averages/rainfall/index.jsp?period=an&area=vc#maps. [Accessed 01-09- 2015].
BOM. 2015b. Victoria in 2014: Another very warm year with very dry conditions in the west [Online]. Bureau of Meteorology http://www.bom.gov.au/climate/current/annual/vic/summary.shtml. [Accessed 20-07- 2015].
BRIAUD, J.-L. 2013. Laboratory Tests. Geotechnical Engineering. John Wiley & Sons, Inc.
BULUT, R. 2001. Finite Element Method Analysis of Slabs on Elastic Half Space Expansive Soil Foundations. PhD, Texas A & M University.
BULUT, R., LYTTON, R. L. & WRAY, W. K. 2001. Soil Suction Measurements by Filter Paper. In: VIPULANANDAN, C., ADDISON, M. B. & HANSEN, M. (eds.) Expansive Clay Soils and Vegetative Influence on Shallow Foundations. Houston, Texas, United States: ASCE Geotechnical Special Publication.
BURNETT, A. 1995. A quantitative X-ray diffraction technique for analyzing sedimentary rocks and soils. Journal of testing and evaluation, 23, 111-118.
CAMERON, D. A. 1977. Parameters for the design of residential slabs and strip footing. MEng, Victoria Institutes of Colleges (Swinburne College of Technology).
CAMERON, D. A. 1989. Tests for reactivity and prediction of ground movement, Canberra, AUSTRALIE, Institution of Engineers.
CAMERON, D. A. 2001. The extent of soil desiccation near trees in a semi-arid environment. Geotechnical & Geological Engineering, 19, 357-370.
CAMERON, D. A. & WALSH, P. F. 1984. The prediction of moisture induced foundation movements using the instability index. Australian Geomechanics, 8, 5-11.
CHAN, D., KODIKARA, J., GOULD, S., RANJITH, P., CHOI, X. S. K. & DAVIS, P. Data analysis and laboratory investigation of the behaviour of pipes buried in reactive clay. Proceedings of 10th Australia New Zealand Conference on Geomechanics - Common Ground 2007, 21st-24th October 2007 Brisbane, Australia. 206-211.
CHAN, D., RAJEEV, P., GALLAGE, C. & KODIKARA, J. Regional field measurement of soil moisture content with neutron probe. Proceedings of The Seventeenth Southeast Asian Geotechnical Conference, 2010. Taiwan Geotechnical Society/Southeast Asian Geotechnical Society, 92-95.
CHAN, D. C. C. 2014. A study of pipe-soil-climate interaction of buried water and gas pipes. PhD, Monash University.
CHAN, I. & MOSTYN, G. 2008. Climate Factors for AS2870 for the Metropolitan Sydney Area. Australian Geomechanics, 43, 17-28.
CHANASYK, D. S. & NAETH, M. A. 1996. Field measurement of soil moisture using neutron probes. Canadian Journal of Soil Science, 76, 317-323.
CHAO, K. C. 2007. Design Principles for Foundations on Expansive Soils. Ph.D, Colorado State University.
CHEN, F. H. 1988. Foundations on expansive soils, Elsevier. CLARKE, D. & SMETHURST, J. A. 2010. Effects of climate change on cycles of
wetting and drying in engineered clay slopes in England. Quarterly Journal of Engineering Geology and Hydrogeology, 43, 473-486.
CORBITT, R. A. 1999. Standard handbook of environmental engineering, New York McGraw-Hill.
COVAR, A. P. & LYTTON, R. L. 2001. Estimating Soil Swelling Behavior Using Soil Classification Properties. Expansive Clay Soils and Vegetative Influence on Shallow Foundations.
CRICHTON, A. J. 1974. Towards a Rational and Echonomical Design Method for Residential Slabs. MEng, Swinburne College of Technology, Melbourne.
CRONEY, D. & COLEMAN, J. D. 1961. Pore Pressure and Suction in Soils. Conference on Pore Pressure and Suction in Soils. Butterworths, London.
277
CUTLER, D. F. & RICHARDSON, I. B. K. 1981. Tree roots and buildings, London, Construction Press.
D.G.FREDLUND 1975. Engineering Properties of Regina Clay. Seminar on Shallow Foundations on Expansive Clays. Regina, Saskatchewan, Canada.
DAHLHAUS, P. G. & O'ROUKE, M. The Newer Volcanics. In: PECK, W. A., NEILSON, J. L. & OLDS, R. J., eds. Engineering Geology of Melbourne: Proceedings of the Seminar on Engineering Geology of Melbourne, 16 September 1992 Melbourne, Victoria, Australia. A.A. Balkema.
DARCY, H. Les Fontaines Publiques de la Ville de Dijon. 1856 Dalmont, Paris. 590-594.
DAS, B. M. 1998. Principles of Geotechnical Engineering, Boston, Mass, PWS Publishing.
DECAGON 2010. WP4C Dewpoint PotentiaMeter Manual. Operator’s Manual - Version 2. Pullman WA, USA: Decagon Devices, Inc.
DELANEY, M., ALLMAN, M. & SLOAN, S. A Network of Field Sites for Reactive Soil Monitoring in the Newcastle-Hunter Region. Proceedings of the 7th. ANZ Conference in Geomechanics., 1996 Adelaide. 381-387.
DELANEY, M. G., LI, J. & FITYUS, S. G. 2005. Field Monitoring of Expansive Soil Behaviour in the Newcastle-Hunter Region. Australian Geomechanics, 40, 3-14.
DINGMAN, S. L. 2002. Physical Hydrology, Prentice Hall. DRISCOLL, R. 1984. A Review of British Experience of Expansive Clay Problems.
Fifth International Conference on Expansive Soils Adelaide, South Australia. EDLEFSEN, N. & ANDERSON, A. 1943. Thermodynamics of soil moisture.
Hilgardia, 15, 31-298. EVETT, S. R., HOWELL, T. A., STEINER, J. L. & CRESAP, J. L. 1993.
Evapotranspiration by soil water balance using TDR and neutron scattering. Management of Irrigation and Drainage Systems, Irrigation and Drainage Div./ASCE, 914-921.
FATAHI, B. 2007. Modelling of Influence of Matric Suction Induced by Native Vegetation on Sub-soil Improvement. PhD, University of Wollongong.
FITYUS, S., CAMERON, D. A. & WALSH, P. F. 2005. The Shrink-swell Test. Geotechnical Testing Journal, 28.
FITYUS, S. G., SMITH, D. W. & ALLMAN, M. A. 2004. Expansive Soil Test Site Near Newcastle. Journal of Geotechnical and Geoenvironmental Engineering, 130, 686-695.
FITYUS, S. G., WALSH, P. F. & KLEEMAN, P. W. 1998. The influence of climate as expressed by the Thornthwaite index on the design depth of moisture change of clay soils in the Hunter Valley. Conference on geotechnical engineering and engineering geology in the Hunter Valley. Springwood, Australia.
FOX, E. 2000. A Climate Based Design Depth of Moisture Change Map of Queensland and the Use of Such Maps to Classify Sites Under AS2870-1996. Australian Geomechanics, 35, 53-60.
FRATTA, D., AGUETTANT, J. & ROUSSEL-SMITH, L. 2007. Introduction to Soil Mechanics Laboratory Testing, New York, Taylor & Francis Group.
FREDLUND, D. G. 2000. The 1999 R.M. Hardy Lecture: The Implementation of Unsaturated Soil Mechanics into Geotechnical Engineering. Canadian Geotechnical Journal, 37, 963 – 986.
FREDLUND, D. G. & RAHARDJO, H. 1993. Soil Mechanics for Unsaturated Soils, John Wiley & Sons.
FREDLUND, D. G., RAHARDJO, H. & FREDLUND, M. D. 2012. Unsaturated Soil Mechanics in Engineering Practice, Hoboken, New Jersey, John Wiley and Sons.
FREDLUND, D. G. & VU, H. Q. 2003. Numerical Modelling of Swelling and Shrinking soils around slabs-on-ground. Proceeding of Post-Tensioning Institute Annual Technical Conference. Huntington Beach, CA, USA.
FREDLUND, D. G. & XING, A. 1994. Equations for the soil-water characteristic curve. Canadian Geotechnical Journal, 31, 521-532.
FREDLUND, D. G., XING, A. & HUANG, S. 1994. Predicting the permeability function for unsaturated soils using the soil-water characteristic curve. Canadian Geotechnical Journal, 31, 533-546.
FU, X., SHAO, M., LU, D. & WANG, H. 2011. Soil water characteristic curve measurement without bulk density changes and its implications in the estimation of soil hydraulic properties. Geoderma, 167–168, 1-8.
GALLERIES. 2014. The clay mineral group [Online]. http://www.galleries.com/clays_group. [Accessed 03-11- 2014].
GIKAS, V. & SAKELLARIOU, M. 2008. Settlement analysis of the Mornos earth dam (Greece): Evidence from numerical modeling and geodetic monitoring. Engineering Structures, 30, 3074-3081.
GOLAIT, Y. S. & WAKHARE, A. S. Unit Swell Potential Concept for Expansive Soils and its Simple Evaluation. Proceedings 8th Australia New Zealand Conference on Geomechanics, 1999 Barton. Australian Geomechanics Society, 171-177.
GOLDBERG, R. N. 1981. Evaluated activity and osmotic coefficients for aqueous solutions: thirty-six uni-bivalent electrolytes. Journal of Physical and Chemical Reference Data, 10, 671-764.
GOLDBERG, R. N. & NUTTALL, R. L. 1978. Evaluated activity and osmotic coefficients for aqueous solutions: The alkaline earth metal halides. Journal of Physical and Chemical Reference Data, 7, 263-310.
GRAY, C. W. & ALLBROOK, R. 2002. Relationships between shrinkage indices and soil properties in some New Zealand soils. Geoderma, 108, 287-299.
GUGGENHEIM, S. & MARTIN, R. T. 1995. Definition of clay and clay mineral. Journal report of the AIPEA nomenclature and CMS nomenclature committees. Clays and Clay Minerals.
HAMER, W. J. & WU, Y.-C. 1972. Osmotic Coefficients and Mean Activity Coefficients of Uni-univalent Electrolytes in Water at 25[degree]C. Journal of Physical and Chemical Reference Data, 1, 1047-1100.
HAMILTON, J. J. 1969. EFFECTS OF ENVIRONMENT ON THE PERFORMANCE OF SHALLOW FOUNDATIONS. Canadian Geotechnical Journal, 6, 65-80.
HANNAM, P. 2015. World headed for an El Nino and it could be a big one, scientists say [Online]. http://www.smh.com.au/environment/weather/world-headed-for-an-el-nino-and-it-could-be-a-big-one-scientists-say-20150507-ggw8bo.html. [Accessed 22-05- 2015].
HAZELTON, P. A. & MURPHY, B. W. 2007. Interpreting Soil Test Results: What Do All the Numbers Mean?, CSIRO Publishing.
HMA. 2013. Model 4700: Magnetic Extensometer [Online]. http://www.geotechsystems.com.au/products/s4000/4700.php. [Accessed 20-09- 2012].
HMA. 2014. Model 4000: Borehole Rod Extensometers [Online]. http://www.geotechsystems.com.au/products/s4000/4000.php. [Accessed 23-04- 2015].
HOLLAND, J. E. & LAWRANCE, C. E. 1980. Seasonal Heave of Austrahan Clay Soils. Proceedigns of 4th Intemational Conference on Expansive Soils. Denver, CO.
HOLLAND, J. E., PITT, W. G., LAWRANCE, C. E. & CIMINO, D. J. 1980. The Behavior and Design of Housing Slabs on Expansive Soils. Proceedigns of 4th Intemational Conference on Expansive Soils. Denver, CO.
HOLLAND., J. E. 1978. Residential Slab Research - Recent Findings, Swinburne College Press, Melbourne.
HOLLAND., J. E., CAMERON., D. A. & WASHUSEN., J. A. 1975. Residential Raft Slabs, Swinburne College of Technology, Hawthorn, Victoria.
HOLTZ, W. G. 1959. Expansive clays-properties and problems. Quarterly of the Colorado School of Mine, 54, 89-125.
HOLTZ, W. G. & GIBBS, H. J. 1956. Engineering properties of expansive clays. Transactions ASCE, 121, 641-677.
HUGHES, L. 2003. Climate change and Australia: Trends, projections and impacts. Austral Ecology, 28, 423–443.
HUNG, V. Q. 2002. Uncoupled and coupled solutions of volume change problems in expansive soils. PhD, University of Saskatchewan.
HUPET, F. & VANCLOOSTER, M. 2002. Intraseasonal dynamics of soil moisture variability within a small agricultural maize cropped field. Journal of Hydrology, 261, 86-101.
JENNINGS, J. E., FIRTH, R. A., RALPH, T. K. & NAGAR, N. 1973. An improved method for predicting heave using the oedometer test. Proceedings of the 3rd International Conference on Expansive Soils. Haifa, Israel.
JENNINGS, J. E. B. & KNIGHT, K. 1957. The Prediction of Total Heave From the Double Oedometer Test. Proc. Symposium on Expansive Clays. Johannesburg: South African Institute Civil Engineers.
JEWELL, S. A. & MITCHELL, P. W. 2009. The Thornthwaite Moisture Index and Seasonal Soil Movement in Adelaide. Australian Geomechanics Journal, 41, 59-68.
JONES, E. W., KOH, Y., TIVER, B. & WONG, M. 2009. Modelling the behaviour of unsaturated, saline clay for geotechnical design. School of Civil Environment and Mining Engineering,: University of Adelaide, Australia.
KARUNARATHNE, A. M. A. N., GAD, E. F., SIVANERUPAN, S. & WILSON, J. L. Review of Residential Footing Design on Expansive Soil in Australia. In: SAMALI, B., ATTARD, M. M. & SONG, C., eds. 22nd Australasian Conference on the Mechanics of Structures and Materials, 11-14 December 2012 Sydney, NSW. Taylor & Francis Group, 575-579.
KARUNARATHNE, A. M. A. N., GAD, E. F., SIVANERUPAN, S. & WILSON, J. L. 2013. Review of Residential Footing Design on Expansive Soil in Australia. In: SAMALI, B., ATTARD, M. M. & SONG, C. (eds.) 22nd Australasian
Conference on the Mechanics of Structures and Materials. Sydney: Taylor & Francis.
KARUNARATHNE, A. M. A. N., SIVANERUPAN, S., GAD, E. F., DISFANI, M. M., RAJEEV, P., WILSON, J. L. & LI, J. 2014. Field and laboratory investigation of an expansive soil site in Melbourne. Australian Geomechanics, 49, 85-93.
KAY, N. 1990. Use of the liquid limit for characterisation of expansive soils. Civil Engineering Transactions, Institution of Engineers, Australia., 32, 151-156.
KEIM, B. D. 2010. THE LASTING SCIENTIFIC IMPACT OF THE THORNTHWAITE WATER–BALANCE MODEL. Geographical Review, 100, 295-300.
KODIKARA, J., RAJEEV, P., CHAN, D. & GALLAGE, C. 2013. Soil moisture monitoring at the field scale using neutron probe. Canadian Geotechnical Journal, 51, 332-345.
KOVÁCS, G. 1981. Seepage Hydraulics. Developments in Water Science 10. Elsevier Science Publishers: Amsterdam.
KRAHN, J. & FREDLUND, D. G. 1972. On toatal, matric and osmotic suction. Soil Science, 114, 339-348.
LANG, A. 1967. Osmotic coefficients and water potentials of sodium chloride solutions from 0 to 40 0C. Australian Journal of Chemistry, 20, 2017-2023.
LEAO, S. & OSMAN, N. Y. TMI for urban resilience: Measuring and mapping long-term climate change effects on soil moisture. In: ROWLEY, S., ONG, R. & MARKKANEN, S., eds. 7th Australasian Housing Researchers’ Conference, 2013 Fremantle, WA, Australia
LEONG, E., HE, L. & RAHARDJO, H. 2002. Factors Affecting the Filter Paper Method for Total and Matric Suction Measurements. Geotechnical Testing Journal, 25, 322-333.
LEONG, E. C., TRIPATHY, S. & RAHARDJO., R. 2003. Total suction measurement of unsaturated soils with a device using the chilled-mirror dew-point technique. Geotechnique, 53, 173 –182
LI, J., SMITH, D., FITYUS, S. & SHENG, D. 2003a. Numerical Analysis of Neutron Moisture Probe Measurements. International Journal of Geomechanics, 3, 11-20.
LI, J., SMITH, D. W. & FITYUS, S. G. 2003b. The effect of a gap between the access tube and the soil during neutron probe measurements. Soil Research, 41, 151-164.
LI, J. & SUN, X. 2015. Evaluation of Changes of the Thornthwaite Moisture Index in Victoria. Australian Geomechanics, 50, 39-49.
LIKOS, W. J. 2000. Total suction-moisture content characteristics for expansive soils. Ph. D, Colorado School of Mines.
LIKOS, W. J. 2004. Measurement of crystalline swelling in expansive clay. Geotechnical Testing Journal, 27, 540-546.
LIU, G., NG, C. & WANG, Z. 2005. Observed Performance of a Deep Multistrutted Excavation in Shanghai Soft Clays. Journal of Geotechnical and Geoenvironmental Engineering, 131, 1004-1013.
LOPES, D., MCMANUS, K. J. & OSMAN, N. 2003. The effect of Thornthwaite Moisture Index changes on ground movements in expansive soils in Victoria, Australia. In: EFFERSON, I. & FROST, M. (eds.) Proceedings of the International Conference on Problematic Soils. Nottingham, England: CI-Premier
281
LOPES, D. & OSMAN, N. Y. 2010. Changes Of Thornthwaite's Total Moisture Indices In Victoria From 1948-2007 And The Effect On Seasonal Foundation Movements. Australian Geomechanics, 45, 37-48.
LOWE, P. R. 1977. An approximating polynomial for the computation of saturation vapour pressure. Journal of Applied Meteorology, 16, 100-103.
LU, N. & GRIFFITHS, D. 2004. Profiles of Steady-State Suction Stress in Unsaturated Soils. Journal of Geotechnical and Geoenvironmental Engineering, 130, 1063-1076.
LU, N. & KAYA, M. 2014. Power Law for Elastic Moduli of Unsaturated Soil. Journal of Geotechnical and Geoenvironmental Engineering, 140, 46-56.
LYTTON, R., AUBENY, C. & BULUT, R. 2004. Design Procedure for Pavements on Expansive Soils. FHWA/TX-05/0-4518-1 Vol. 1. Texas Transportation Institute: The Texas A&M University System.
LYTTON, R. L. 1970a. Analysis for Design of Foundations on Expansive Clay. Geomechanics Journal, 29-37.
LYTTON, R. L. 1970b. Design Criteria for Residential Slabs and Grillage Rafts on Reactive Clay. Melbourne, Australia: Division of Applied Mechanics, CSIRO.
LYTTON, R. L. & WOODBURN, J. A. 1985. Design and Performance of Mat Foundations on Expansive Clay. Golden Jubilee of the International Society for Soil Mechanics and Foundation Engineering: Commemorative Volume. Institution of Engineers, Australia.
MANN, A. 2003. The identification of road sections in Victoria displaying roughness caused by expansive soils. MSc, Swinburne University of Technology.
MAPS. 2015. Energy and Earth Resources - Earth resources online store [Online]. http://dpistore.efirst.com.au/categories.asp?cID=4. [Accessed 09-02- 2015].
MATHER, J. R. 1974. Climatology: fundamentals and applications, McGraw-Hill, USA.
MATHER, J. R. 1978. The climatic water budget in environmental analysis, Lexington, US, Lexington Books.
MATYAS, E. L. & RADHAKRISHNA, H. S. 1968. Volume Change Characteristics of Partially Saturated Soils. Geotechnique, 18, 432 –448.
MCKEEN, R. G. & JOHNSON, L. D. 1990. Climate-Controlled Soil Design Parameters for Mat Foundations. Journal of Geotechnical Engineering, 116, 1073-1094.
MCMANUS, K., LOPES, D. & OSMAN, N. Y. 2004. The Effect of Thronthwaite Moisture Index Changes on Ground Movement Predictions in Australian Soils. 9th Australia-New Zealand Conference on Geomechanics. Auckland.
MITCHELL, P. 1984a. A Simple Method of Design of Shallow Footings on Expansive Soil In: INSTITUTION OF ENGINEERS, A. A. G. S. S. A. G. (ed.) Fifth International Conference on Expansive Soils. Adelaide, South Australia: Barton, A.C.T. : The Institution.
MITCHELL, P. W. 1979. The Structural Analysis of Footings on Expansive Soil. Newton: Kenneth W.G. Smith & Associates.
MITCHELL, P. W. The Concepts Defining the Rate of Swell of Expansive Soils. 4th International Conference on Expansive Soils, 1980 Denver. 106-116.
MITCHELL, P. W. 1984b. The Design of Shallow Footings on Expansive Soil. University of Adelaide.
MITCHELL, P. W. 2008. Footing Design for Residential Type Structures in Arid Climates. Australian Geomechanics Journal, 43, 51-68.
MITCHELL, P. W. & AVALLE, D. L. 1984. A Technique to Predict Expansive Soil Movements. Fifth International Conference on Expansive Soils Adelaide, South Australia.
MORRIS, P. O. & GRAY, W. J. 1976. Moisture conditions under roads in the Australian environment. In: GRAY, W. J., NATIONAL ASSOCIATION OF AUSTRALIAN STATE ROAD AUTHORITIES, C., MAINTENANCE PRACTICES, C. & SYMPOSIUM ON SUB-SOIL, D. (eds.) Research report ARR ; no. 69. Vermont South, Vic: Australian Road Research Board.
MURPHY, B. F. & TIMBAL, B. 2008. A review of recent climate variability and climate change in southeastern Australia. International Journal of Climatology, 28, 859-879.
MURRAY, E. J. & SIVAKUMAR, V. 2010. Unsaturated Soils: A Fundamental Interpretation of Soil Behaviour, John Wiley & Sons.
NELSON, J. D., CHAO, K. C., OVERTON, D. D. & NELSON, E. J. 2015. Nature of Expansive Soils. Foundation Engineering for Expansive Soils. John Wiley & Sons, Inc.
NELSON, J. D. & MILLER, D. J. 1992. Expansive soils: problems and practice in foundation and pavement engineering, John Wiley & Sons Inc.
NELSON, J. D., OVERTON, D. D. & DURKEE, D. B. Depth of Wetting and the Active Zone. Shallow Foundation and Soil Properties Committee Sessions at ASCE Civil Engineering, 2001 Houston, Texas, United States. American Society of Civil Engineers, 95-109.
NELSON, J. D., OVERTON, D. O. & CHAO, K. C. 2003. Design of Foundations for Light Structures on Expansive Soils. California Geotechnical Engineers Association Annual Conference. Carmel, California.
NG, C. W. W. & MENZIES, B. 2007. Advanced Unsaturated Soil Mechanics and Engineering, London, Taylor and Francis.
NIXON, P. R. & LAWLESS, G. P. 1960. Detection of deeply penetrating rain water with neutron-scattering moisture meter. Transactions of the American Society of Agricultural Engineers, 3.
NWS. 2010. The Hydrologic Cycle [Online]. National Weather Service, National Oceanic and Atmospheric Administration, U.S. Department of Commerce.: http://www.srh.noaa.gov/jetstream/atmos/hydro.htm. [Accessed 05-10- 2012].
PENMAN, H. L. 1948. Natural Evaporation from Open Water, Bare Soil and Grass. Proceedings of the Royal Society of London - Series A, Mathematical and Physical Sciences, 193, 120-145.
PEREIRA, J. H. F. 1996. Numerical analysis of the mechanical behavior of collapsing earth dams during first reservoir filling. Ph.D, University of Saskatchewan.
PHILP, M. & TAYLOR, M. 2012. Beyond Agriculture - Exploring the application of the Thornthwaite Moisture Index to infrastructure and possibilities for climate change adaptation [Online]. http://www.nccarf.edu.au/settlements-infrastructure/content/beyond-agriculture-exploring-application-thornthwaite-moisture-index-infrastructure-and. [Accessed 13-01- 2014].
PITT, W. G. 1982. Correlation Between the Real Behaviour and the Theoretical Design of Residential Raft Slabs MEng, Swinburne Institute of Technology, Melbourne.
PTI 2004. Design of Post-tensioned Slabs-on-ground, Post Tensioning Institute. RAHARDJO, H. & LEONG, E. Suction Measurements. In: MILLER, G. A., ZAPATA,
C. E., HOUSTON, S. L. & FREDLUND, D. G., eds. Fourth International
Conference on Unsaturated Soils, April 2-6, 2006 Carefree, Arizona, United States. American Society of Civil Engineers, 81-104.
RAJEEV, P., CHAN, D. & KODIKARA, J. 2012. Ground–atmosphere interaction modelling for long-term prediction of soil moisture and temperature. Canadian Geotechnical Journal, 49, 1059-1073.
RAJEEV, P. & KODIKARA, J. 2011. Numerical analysis of an experimental pipe buried in swelling soil. Computers and Geotechnics, 38, 897-904.
RANGANATHAM, B. V. & SATYANARAYANA, B. 1965. A rational method of predicting swelling potential for compacted expansive clays. International Conference on Soil Mechanics and Foundation Engineering. Montrial, Canada.
REN, G. & LI, J. 2010. Monitoring In Situ Soil Moisture Variations of Expansive Clay Using Neutron Probes. Deep Foundations and Geotechnical In Situ Testing.
RICHARDS, B. G. 1965. Measurement of the Free Energy of Soil Moisture by the Psychrometric Technique Using Thermistors. In: AITCHISON, G. D. (ed.) Moisture Equilibria and Moisture Changes in Soils Beneath Covered Areas. Butterworths, Sydney.
RICHARDS, B. G., PETER, P. & EMERSON, W. W. 1983. The effects of vegetation on the swelling and shrinking of soils in Australia. Geotechnique, 33, 127–139.
RUSSAM, K. & COLEMAN, J. D. 1961. The Effect of Climatic Factors on Subgrade Moisture Conditions. Geotechnique, 3, 22-28.
SAMUELS, S. G. & CHENEY, J. E. Long-term heave of a building on clay due to tree removal. Proceedings of Conference on Settlement of Structures, 1975 Cambridge. Pentech Press, London 212-220.
SATTLER, P. J. & FREDLUND, D. G. 1991. Numerical modelling of vertical ground movements in expansive, soils. Canadian Geotechnical Journal, 28, 189-199.
SCHOFIELD, R. K. The pF of the Water in Soil. Transactions, 3 "d International Congress of Soil Science, 1935. 37-48.
SEED, H. B., WOODWARD, R. J. & LUNDGREN, R. 1962. Prediction of Swelling Potential for compacted clays. ASCE. Soil Mechanics and Foundation Division.
SHROFF, A. V. 2003. Soil Mechanics and Geotechnical Engineering, Taylor & Francis.
SKEMPTON, A. W. 1953. The colloidal “Activity” of clays. Proceedings of the 3rd International Conference of Soil Mechanics and Founda-tion Engineering.
SMITH, J. B., SCHNEIDER, S. H., OPPENHEIMER, M., YOHE, G. W., HARE, W., MASTRANDREA, M. D., PATWARDHAN, A., BURTON, I., CORFEE-MORLOT, J., MAGADZA, C. H. D., FÜSSEL, H.-M., PITTOCK, A. B., RAHMAN, A., SUAREZ, A. & VAN YPERSELE, J.-P. 2009. Assessing dangerous climate change through an update of the Intergovernmental Panel on Climate Change (IPCC) “reasons for concern”. Proceedings of the National Academy of Sciences, 106, 4133-4137.
SMITH, R. 1993. Estimating soil movements in new Areas. Seminar - Extending the Code beyond Residential Slabs and Footings. The Institution of Engineers, Australia.
SRIDHARAN, A. 1999. Volume change behaviour of expansive soils. International Symposium on Problematic Soils. Sendai, Japan.
STATISTICS, A. B. O. 2012. GEOGRAPHIC DISTRIBUTION OF THE POPULATION [Online]. http://www.abs.gov.au/ausstats/[email protected]/Lookup/by%20Subject/1301.0~2012~
TADANIER, R. & NGUYEN, V. 1984. Index Properties of Expansive Soils in New South Wales. Fifth International Conference on Expansive Soils. Adelaide, South Australia: Institution of Engineers Australia.
THE-AGE. 2011. Owners find homes are cracking under pressure [Online]. http://www.domain.com.au/news/owners-find-homes-are-cracking-under-pressure-20111221-1p5or/. [Accessed 27-02- 2012].
THE-AGE. 2014a. Thousands of suburban home owners facing financial ruin [Online]. http://www.theage.com.au/victoria/thousands-of-suburban-home-owners-facing-financial-ruin-20140607-39q4z.html. [Accessed 10-06- 2014].
THE-AGE. 2014b. When a dream home turns into a nightmare [Online]. http://www.theage.com.au/victoria/when-a-dream-home-turns-into-a-nightmare-20140607-39q50.html. [Accessed 21-07- 2014].
THOMAS, D. E. 1967. Geology of the Melbourne District, Victoria, Mines Department Melbourne, Victoria, Australia.
THORNTHWAITE, C. W. 1948. An Approach Toward a Rational Classification of Climate. Soil Science, 66, 77.
THORNTHWAITE, C. W. 1952. The water balance in arid and semiarid climates. Proceding of International Symposium on Desert research. Jerusalem: Research Council of Israel, Special Publication No. 2.
THORNTHWAITE, C. W. & MATHER, J. R. 1955. The Water Balance. Publications in Climatology. Centerton, New Jersey: Laboratory of Climatology, Drexel Institute of Technology.
THORNTHWAITE, C. W. & MATHER, J. R. 1957. Instructions and Tables for Computing Potential Evapotranspiration and the Water Balance. Publications in Climatology. Centerton, New Jersey: Laboratory of Climatology, Drexel Institute of Technology
TODD, D. K. 1980. Groundwater Hydrology, New York., John Wiley & Sons. TUCKER, B. M. 1955. SOME APPLICATIONS OF CLIMATIC ANALYSIS. In:
CSIRO (ed.) Division of Soils. Adelade: CSIRO. UMS. 2013. HYPROP - Laboratory evaporation method for the determination of pF-
curves and unsaturated conductivity [Online]. http://www.ums-muc.de/en/products/soil_laboratory/hyprop.html. [Accessed 08-29- 2013].
VADOSE 2013. Vadose Zone modeling with VADOSE/W, Calgary, Alberta, Canada, Geo-Slope International Ltd.
VAN GENUCHTEN, M. T. 1980. A Closed-form Equation for Predicting the Hydraulic Conductivity of Unsaturated Soils. Soil Sci. Soc. Am. J., 44, 892-898.
VU, H. Q. & FREDLUND, D. G. 2004. The prediction of one-, two-, and three-dimensional heave in expansive soils. Canadian Geotechnical Journal, 41, 713-737.
WALSH, P. F. 1975. The Design of Residential Slabs on Ground. Division of Building Research, C.S.I.R.O. Melbourne, Australia.
WALSH, P. F. 1978. The Analysis of Stiffened Rafts on Expansive Clays. Division of Building Research, C.S.I.R.O. Melbourne, Australia.
WALSH, P. F. & CAMERON, D. A. 1997. The design of residential slabs and footings, Standards Australia.
WARD, A. L. & WITTMAN, R. S. 2009. Calibration of a neutron hydroprobe for moisture measurements in small diameter steel-cased boreholes. Pacific Northwest National Laboratory.
WASHUSEN, J. A. 1977. The Behaviour of Experimental Rafts Slabs on Expansive Clay Soils in the Melbourne Area. MEng, Swinburne College of Technology, Melbourne.
WAYLLACE, A. 2008. Volume change and swelling pressure of expansive clay in the crystalline swelling regime. PhD, University of Missouri.
WIJEYESEKERA, D. C., O'CONNOR, K. & SALMON, D. E. 2001. Design and performance of a compacted clay barrier through a landfill. Engineering Geology, 60, 295-305.
WILSON, G. W. 1990. Soil evaporative fluxes for geotechnical engineering problems. Ph.D, University of Saskatchewan.
WRAY, W. K. 1978. Development of a Design Procedure for Residential and Light Commercial Slabs-on-ground Constructed Over Expansive Soils. PhD, Texas A&M University.
WRAY, W. K. 1997. Using Soil Suction to Estimate Differential Soil Shrink or Heave. Proc. Unsaturated Soil Engineering Practice. Geotechnical Special Publication No. 68: ASCE.
WRAY, W. K. Mass transfer in Unsaturated Soils: A Review of Theory and Practices. Proceedings of the 2nd International Conference on Unsaturated Soils, 1998 Beijing, China. 99-155.
WRAY, W. K. 2005. Three-Dimensional Model for Moisture and Volume Changes Prediction in Expansive Soils. Journal of Geotechnical and Geoenvironmental Engineering, 131, 311-324.
YOSHIDA, R. T., FREDLUND, D. G. & HAMILTON, J. J. 1983. The prediction of total heave of a slab-on-grade floor on Regina clay. Canadian Geotechnical Journal, 20, 69-81.
ZASLAVSKY, D. & ROGOWSKI, A. S. 1969. Hydrologic and Morphologic Implications of Anisotropy and Infiltration in Soil Profile Development. Soil Science Society of America Journal, 33, 594-599.
A_1
10. APPENDICES
A: HYPROP MEASUREMENTS OF BRAYBROOK SOIL
Depth 0 – 0.3 meters
HYPROP TEST RESULTS SHEET
SITE: Braybrook DEPTH: 0-0.3 m
SAMPLE DATE : 1/05/2014 TEST DATE : 23/09/2014
INITIAL VOLUME OF SAMPLE : 249.0 cm3 OVEN DRY WEIGHT OF SAMPLE : 371.5 g
D: SATURATED HYDRAULIC CONDUCTIVITY MEASUREMENTS OF BRAYBROOK SOIL
Depth 0.0 – 0.4 meters
Type of Material Undisturbed clay soil Depth (m) 0.0-0.4 Location Braybrook Date Sampled 24/07/2014 Tested by Aruna Karunarathne Date Tested 30/10/2014
Material Specification Saturation Duration 24 hrs
Curing Duration 24 hrs
Sample Diameter (mm) 50.00
Reading Duration 48 hrs Water Temperature 20 (°C) Sample Area (cm2) 19.63
Epoxy Curing 24 hrs Flow direction Downward Sample height (cm) 10.60 Water density at test temperature (gr/cm3) 0.9982071
Type of Material Undisturbed clay soil Depth (m) 0.5-1.0 Location Braybrook Date Sampled 24/07/2014 Tested by Aruna Karunarathne Date Tested 10/10/2014
Material Specification Saturation Duration 24 hrs
Curing Duration 24 hrs
Sample Diameter (mm) 50.00
Reading Duration 48 hrs Water Temperature 20 (°C) Sample Area (cm2) 19.63
Epoxy Curing 24 hrs Flow direction Downward Sample height (cm) 9.50 Water density at test temperature (gr/cm3) 0.9982071
Type of Material Undisturbed clay soil Depth (m) 1.0-1.4 Location Braybrook Date Sampled 24/07/2014 Tested by Aruna Karunarathne Date Tested 23/09/2014
Material Specification Saturation Duration 24 hrs
Curing Duration 24 hrs
Sample Diameter (mm) 50.00
Reading Duration 48 hrs Water Temperature 20 (°C) Sample Area (cm2) 19.63
Epoxy Curing 24 hrs Flow direction Downward Sample height (cm) 8.90 Water density at test temperature (gr/cm3) 0.9982071
Type of Material Undisturbed clay soil Depth (m) 1.5-1.8 Location Braybrook Date Sampled 24/07/2014 Tested by Aruna Karunarathne Date Tested 18/11/2014
Material Specification Saturation Duration 24 hrs
Curing Duration 24 hrs
Sample Diameter (mm) 50.00
Reading Duration 48 hrs Water Temperature 20 (°C) Sample Area (cm2) 19.63
Epoxy Curing 24 hrs Flow direction Downward Sample height (cm) 11.2 Water density at test temperature (gr/cm3) 0.9982071