ABSTRACT
This study was conducted to develop the soil water characteristic curve for residual soils in landslides
prone areas of Sri Lanka. Soil water characteristic curve is the relationship between matric suction
and volumetric water content, which is used to understand the behavior of the unsaturated soil.
Reduction in the matric suction, due to the water infiltration in rainy seasons caused the land slides.
Three soil samples collected from Erapola and Dickdeniya in Ratnapura District were tested in this
study. To determine the Soil water characteristic curve, an apparatus was developed based on the axis
translation principle in the Department of civil engineering laboratory, OUSL. Tests were done on
compacted soil samples, after achieving the possible saturation. 500kPa Zero air entry porous disk
was used for the test. The index properties of soil such as sieve analysis, hydrometer test, Atterberg
limit test and Specific gravity test were performed according to BS 1377 part 02 1990.
Results shows that the plasticity index between 2.13-12.32 and specific gravity between 2.7-2.8 and
the soils were classify as silty sand (SM). The result data of the Soil water characteristic curve
experiment were best fitted with Fredlund and Xing (1994) equation and the fitting parameters were
found. The results show that the air entry values of samples were 170kPa, 120kPa and 280kPa and the
residual water contents were 0.280, 0.170 and 0.185 respectively. As for the fitting parameters,
parameter a was between 50-98, parameter m was between 0.19-0.4 and parameter n was between
0.5-1.3. Using the Feng and Fredlund (1999) equation wetting curve was best fitted for two samples
and the hysteresis nature of soil water characteristic curve was showed.
4.3 DISCUSSION
From the soil classification result it can be concluded that, each of the four subdivided samples show
the similar mechanical properties. These have minimum variation and the soil types were similar
SWCC test has a higher accuracy due to the higher standard of the test and does not show a significant
variation with the soil type. Soil types that have similar properties show similar SWCC. Hence it was
acceptable to conduct the SWCC test on one sample obtained from the four subdivided specimens.
During this study the ceramic disk with an air-entry value of 500kPa was used and Table 4.12
(Padilla, 2005) recommends this selection for Silty Sands (SM).
Table 4.12 Selection of ceramic disk (Padilla 2005)
Type of soil Air entry value of ceramic disk (kPa)
Clean sand 100
Silty sand, Clayey sand 300
Sandy silt, Sandy clay 500
Clay 1500
As mentioned in the Introduction (Chapter 1) even the soil types were same, SWCC fitting
parameters, air entry value and residual water content have slightly different values. Even the soil type
were same plastic limit, liquid limit and plasticity index have slightly different properties. Sample 01-
1 has the lowest plastic index (2.13) while sample 03-3 shows the highest index value (12.32). But
the plastic limit of the sample 01-1 has highest plastic limit value and sample 02-1 (34.35) and sample
03-1 (33.90) have similar plastic index.
As described in Chapter 02 (Literature review) SWCC is hysteresis. To show this hysteresis nature,
sample 02-1 and 03-1 had been tested. The wetting curve was best fitted using Feng and Fredlund
equation. In figure 9 and figure 10 show not only the drying part of the curve but also the wetting part
of the curves too.
When this investigated results compared with the literature it can be understand that the fitting
parameters were different with the shape of the curve. As described in Chapter 2 in Fredlund and
Xing (1994) model with a large a value curve shift towards the higher soil suction region. The
investigated results also show the same results with the parameter a. Parameter n related to the pore
size distribution index. Since the investigated n values were low, the uniformity of the pore size in the
soil was also low. Since the investigated samples were compacted pore sizes can be less uniform.
Small values of m results the moderate slope in the lower suction range and steeper slope in the high
suction range.
Figure 4.9 shows two different curves with their parameter values. The first soil layer of the slope
consisted of a silty sand with unit weight of 20 kN/m3, effective cohesion of 9 kPa and air entry value
was 15kPa, the second soil layer consisted of silty sand with unit weight of 20 kN/m3, effective
cohesion of 0 kPa and air entry value was 5kPa.
Figure 4.9 SWCC for two silty sands (Rahardjo, 2010)
The main results of this study were the air entry values of SWCCs. This air entry values change with
the amount of fine particles in the soil samples. As shows in Figure 1.5, it is also related to the
plasticity index of the soil. Higher plasticity index value will shift the curve towards the higher
suction region. If a soil contains high percentage of fine particles, the air-entry value of the SWCC
can be high. Air entry value also changes with the amount of compaction and initial water content.
Because the behavior of unsaturated soil is strongly related to its pore size and pore geometrical
distribution (Miao 2001). Pore structure of undisturbed soil has certain directivity and even some
connected channels exist in undisturbed soil because soil particles are arranged according to certain
direction and order during the formation process of undisturbed soil. For compacted soil sample,
pores in soil are basically uniform distribution and non-directional. Its pore gradation is bad gradation.
So air enters difficultly and soil has high air entry value (Yu Li, 2009). Higher initial water content
has a higher air entry value. Also the shape of soil particle, organics coating, the entrapped air, and
some adsorptive forces acting on the surface of soil particles can be affected the SWCC on fine
grained soils. In this research the investigated samples were compacted samples and achieved possible
saturation.
Silty sand that had a clay fraction (<0.002 mm) less than 8% with a low plasticity index (PI = 2) had
been used in the curve that in Figure 1.6 (Vanapalli 1996). In this SWCC the air entry value was 14
kpa.
In this research the sample 02-1 and sample 03-1 had around 40% fine particles while sample 01-1
had around 30% fine particles. And the plasticity index value in sample 01-1 was 2.13, in sample 02-1
was 6.08 and in sample 03-1 was 14.16.
When comparing the sample 01-1 results with the literature (Figure 1.6) that even the plastic index
was around 2, the fine percentage difference is 22% and the air entry value is also higher in sample
01-1 ( difference is 156)
In this study the apparatus that used to measure the SWCC had a limitation. It can only measure the
air pressure higher than 10kPa. Hence the SWCC graph was plotted as matric suction value starting
from 1kPa. The lower suction range of the curve cannot be plotted. According to the literature if a
sample has dual porosity, having this lower range is important. Soils which have dual porosity have
two air entry values. Figure 4.10 shows a curve (soil 1) that has dual porosity. The first air entry value
(i.e.0.05 kPa) might be associated with macro porosity while the second air entry value (i.e. about
1,000 kPa) might be associated with micro porosity of the soil. It can be clearly identified that the first
air entry value was occurred in the lower range of the suction. Hence it is difficult to say that the
investigated air entry values were higher due to the dual porosity or not.
Figure 4.10 Example of dual porosity (Hernandez 2011)
Since the lower range of the curve was missing, the air entry value can be shifted towards higher
suction range than the actual value. As shown in Figure 4.11 and Figure 4.12 measurement of air
entry value can be changed with the range of the curve.
According to the above explanation it can be concluded that this air entry values can be acceptable.
1.000
0.800
0.600
0.400
0.200
0.000
1 10 100 1000 10000 100000 1000000
Figure 4.11 SWCC without lower suction range
1.000
0.800
0.600
0.400
0.200
0.000
0.001 0.010 0.100 1.000 10.000 100.000 1000.00010000.000100000.000
1000000.000
Figure 4.11 SWCC with lower suction range
CHAPTER 5: CONCLUSIONS
In this study three soil water characteristic curves were developed. The experimental data obtained
from SWCC test apparatus were best fitted with the Fredlund and Xing (1994) equation and the fitting
parameters were then collected.
The main conclusions that can be drawn from this study are.
According to the classifications all the investigated soils were silty sand (SM)
The air entry value and residual water contents are in the range of 120kN/m2 to 250kN/m2 and
0.17-0.28 respectively. Since the investigated samples were compacted and have a higher
percentage of fine particles (around 40%), these results are acceptable.
The fitting parameters such as a, m and n are in the range of 0.24 – 0.299, 1.7 – 4.8 and 0.142
– 0.440 respectively.
Using this investigated soil water characteristic curves and its values, relationships that
related to unsaturated soil mechanics (permeability function (Fredlund, 1994), shear strength
(Fredlund, 1995) and volume–mass change (Fredlund, 1999)) can be obtained. These data
helping to understand the unsaturated soil behavior.
CHAPTER 6: RECOMMENDATIONS
This study was done using disturbed soil samples. It can be recommended to use undisturbed
soil samples to develop the SWCC and compare the results with the results of this study to
find out the difference between two conditions.
Generally when SWCC tests were done, effect of the consolidation pressure was neglected.
This study also neglected the effect of the consolidation pressure when SWCC tests were
done. Consolidation pressure affects the Air entry values, hysteresis and the shape of the
SWCC. It can be recommended to find the effect of the consolidation pressure on SWCC.
The apparatus used to this SWCC test measured the air pressure starting 10kPa. Hence the
SWCC graph plotted starting matric suction value as 1kPa. But some soils have air entry
values in the range of 0.01kPa-1kPa. It helps to measure an accurate air entry value.
Measurements in the lower range also help to identify the dual porosity of the soils. So it can
be recommended to improve the SWCC test apparatus to measure the lower air pressures also.