 |
 |
Response of Suction, Moisture and Temperature of Unsaturated Granitic Residual Soil to Rainfall
Department of Civil Engineering, University Putra, Serdang, Selangor, Malaysia. Faisal Hj. Ali and Affendi Abdullah Department of Civil Engineering, University of Malaya, Kuala Lumpur, Malaysia |
ABSTRACT
Soils located above the groundwater table such as residual soils are generally unsaturated and possess negative pore water pressures (matric suction). However, heavy, continuous rainfall can result in the decreased in matric suction and even increased in pore water pressures to a significant depth, resulting in the instability of the slope. This paper describes a study that has been made to determine the response of suction to rainfall of an unsaturated residual soil slope of weathered granite. Suctions are measured using tensiometers, and the rainfall is recorded by using an automatic logging tipping bucket rain gauge. In addition, the soil moisture and temperature response to rainfall are also monitored. The value of suction is found to vary with depth and soil weathering grades. Suction is observed to decrease with increase in depth below the ground surface. Granitic residual soil of weathering grade IV appears to have the lowest value of suction, indicating the more porous nature of soil of this grade. During the dry spell, the suction values at the shallower depth are observed to be generally larger than the suction values at the deeper depths. Upon rain (wetter period), there is a general lost (reduction) of suction in all cases. However, the suction response to antecedent rainfall is less pronounced as the depth increases. The drop in suction at the deeper depth is only significant over a long wet period, indicating longer time required by the water to infiltrate the slope. With regards to soil moisture, there appears to be an increase in the soil moisture with rain. After the rain has stopped, there is a gradual decrease in the moisture content as the soil recovers its suction. Likewise there is a decrease of soil temperature due to the rain.
Keywords: Moisture, rainfall, suction, temperature, unsaturated residual soil.
INTRODUCTION
The microclimatic conditions in an area are the main factors causing a soil deposit to be unsaturated. Therefore, unsaturated soils or soils with negative pore-water pressures can occur in essentially any geological deposit. An unsaturated soil could be a residual soil, a lacustrine deposit, and soils in arid and semi arid areas with deep ground water table.
Tropical residual soils have some unique characteristics related to their composition and the environment under which they develop. Most distinctive is the microstructure which changes in a gradational manner with depth. Their strength and permeability are likely to be greater than those of temperate zone soils with comparable liquid limits. Most classical concepts related to soil properties and soil behavior have been developed for temperate zone soils, and there has been difficulty in accurately modeling procedures and conditions to which residual soils will be subjected. Engineers appear to be slowly recognizing that residual soils are generally soils with negative in situ pore-water pressures, and that much of the unusual behavior exhibited during laboratory testing is related to a matric suction change in the soil (Fredlund and Rahardjo 1985, 1993). There is the need for reliable engineering design associated with residual soils (Ali & Rahardjo 2004).
When the degree of saturation of a soil is greater than about 85%, saturated soil mechanics principles can be applied. However, when the degree of saturation is less than 85%, it becomes necessary to apply unsaturated soil mechanics principles (Fredlund and Rahardjo 1987). The transfer of theory from saturated soil mechanics to unsaturated soil mechanics and vice versa is possible through the use of stress state variables. Stress state variables define the stress condition in a soil and allow the transfer of theory between saturated and unsaturated soil mechanics. The stress state variables for unsaturated soils are net normal stress (s-ua) and matric suction (ua-uw), where s is the total stress, ua is the pore-air pressure and uw is the pore-water pressure. The stress state in an unsaturated soil can be represented by two independent stress tensors as follow (Fredlund and Morgenstern 1977):
 |
(1) |
 |
(2) |
where sx, sy, sz in Equation 1 are the total normal stresses in the x-, y-, and z-directions, respectively; and txy, tyx, txz, tzx, tzy, tyz are the shear stresses.
Inhabited areas with steep slopes consisting of residual soils are sometimes the sites of catastrophic landslides that claimed many lives (Brand 1984, Shaw-Shong 2004). The soils involved are often residual in genesis and have deep water tables. The surface soils have negative pore water pressures that play a significant role in the stability of the slope. However, heavy, continuous rainfall can result in decreased in matric suction and even increased in pore-water pressures to a significant depth, resulting in the instability of the slope.
Devices commonly used for measuring matric suction include tensiometers, null-type pressure plate and thermal conductivity sensors. Matric suction can be measured either in a direct or indirect manner. This paper described a study that has been made to determine the response of suction to rainfall of an unsaturated residual soil slope. Suctions are measured using tensiometers, and the rainfall is measured using an automatic logging tipping bucket rain gauge. In addition, the soil moisture and temperature response to rainfall are also recorded.
FIELD SITE
The field site chosen for this study was a cut slope at KM 31 of the Kuala Lumpur- Karak highway, near Kuala Lumpur, Malaysia. The cut slope basically composed of a residual soil that had developed over the more commonly outcropping Permo-Triassic mesozonal granite rock of Peninsular Malaysia (Raj 1985). Figure 1 shows a cross section and instrumentation of the cut slope. The slope generally comprised of weathered material of grade VI to II, based on the commonly use classification of Little (1969). Prior to the cutting of the slope, the water table was located within horizon IID.
Note: S1 to S4: Instrument cluster for measuring soil suction, moisture and temperature at depths of 32 cm, 92 cm, and 24 cm.
Figure 1. Cross – section of cut slope under study
FIELD INSTRUMENTATION
In this study, the following data are needed to be recorded in the field, i.e. soil suction, soil moisture and temperature, as well as record of the rainfall. Figure 2 shows a schematic diagram of the field instrumentation. They basically consist of a tensiometer for measuring soil suction, soil cell for measuring soil moisture and temperature, automatic rain gauge for recording the rainfall, and an automatic data logging unit. Figure 3 shows a typical detail of the instrumentation cluster at each measuring location.
Figure 2. Schematic arrangement of the field instrumentation
Figure 3. Typical detail of the instrumentation cluster at each measuring location
Jet-fill tensiometer which was used for measuring the soil suction consisted of a sealed tube with a porous ceramic cup at one end, and a pressure measuring device and a jet-fill water reservoir, as shown in Figure 4. Note that the tensionmeter can only measure soil suction up to 1 bar or 100 kPa.
Figure 4. Jet-fill tensiometer (Soil Moisture Equipment Corp., USA, www.soilmoisture.com)
The soil cell of type EL 514-054 is used for the measurement of soil moisture and temperature. The cell is made up of two half-cases of soil moisture cell, spot-welded together to ensure uniform spacing between the screens of the electrode sandwich, as shown in Figure 5.
Figure 5. Details of soil moisture (ELE International, UK, www.ele.com)
The antecedent rainfall is recorded using an automatically logged tipping bucket rain gauge. The data is stored in a memory card. The rain gauge is self-contained and facilitates easy data collection. The rainfall recorder utilizes solar cell and a back up battery as source of power.
FIELD RESULTS AND DISCUSSION
Suctions at Different Depths and Weathering Grades
Figure 6 shows the variation suction at different depths for different berms. It can be seen from the figure that the value suction generally decreases with increasing depth. Similar observation was made in studies conducted on residual soils in Hong Kong (Sweeney 1982).
Note that berms 1, 2, 3 & 4 represent approximately granitic residual soils of weathering grades III, IV, V & VI respectively (see Figure 1). Looking at the above figure, it appears that at particular depth, granitic residual soil of weathering grade VI (berm 4) generally has the higher suction. While soil of weathering grade IV (berm 2) appears to have the lower suction, which can be attributed to a relatively higher infiltration rate (permeability) of granitic residual soil of this grade.
Figure 6. Suction at different depths
Suction Response to Rainfall
Plotted also in Figure 6 is the amount of cumulative rain in millimeter. In general, suction at shallow depth shows greater variation than those at the deeper depth. For example suction at 30 cm depth fluctuate from 85 kPa to just about 10 kPa. At deeper depth (92 cm and 124 cm) the variation in suction is only between 65 kPa to 25 kPa.
During the dry spell, the suction values at the shallower depth (30 cm) are generally larger than the suction values recorded at the deeper depths (92 cm and 124 cm). Upon rain (wetter period), there is a general lost (reduction) of suction in all cases.
As shown in Figure 6 above, the soil suction appears to fluctuate considerably and reflects a mirror-image with the rainfall. However, the suction response to antecedent rainfall is less pronounced as the depth increases. The suction at the shallower depth (30 cm) drops to only about 10 kPa, while the minimum recorded values of suction at deeper depths (92 cm and 124 cm) are still twice higher (>20 kPa).
A sample plot for the dry and wet period at various depths for berm 4 is shown in Figure 7 and Figure 8 respectively.
Note that the suction at the shallow depth showed a faster recovery (steeper increases in suction) compared with the deeper depths (92 cm and 124 cm) during the dry spell as shown in Figure 7.
Figure 7. Suction plot for a dry period (at berm 4)
The drop in suction over the wet period at the shallower depth (30 cm) shows a mirror image response to the cumulative rain plot as shown in Figure 8. However for case of the deeper depth, the drop in suction is only significant over a long wet period, indicating longer time required by the water to infiltrate the slope.
Figure 8. Suction plot for wet period (at berm 4)
Effect of Rainfall Intensity
Figure 9 shows the effect of rainfall intensity on the loss of suction. As shown, higher rainfall intensity would results in larger drop in suction, particularly at the shallower (30 cm) depth. However, it is observed that drop in suction is much slower at the deeper (90 cm and 124 cm) depths.
High intensity rainfall, medium long duration (40 mm in 2 hrs - 20 mm/hr)
High intensity rainfall, short duration (20mm in 0.5 hr - 40 mm/hr)
Low intensity rainfall, medium long duration (15 mm in 1.5 hr - 10 mm/hr)
Figure 9. Suction response to rainfall intensity (at berm 4)
Suction Recovery Prediction
A preliminary analysis is undertaken to predict the resulting suction value (Sr) for different period (T) after cessation of rain. In establishing the relationship between Sr and T, the lowest observed values suction after storm, and the recovery values of suction over a period of no rain are used. The suction recovery equations obtained for berm 4 at approximately 32 cm, 92 cm and 124 cm depth respectively are as follows:
| Sr = 0.183 + 0.97 S1 + 0.134 T; at depth 30 cm | (3) |
| Sr = 0.230 + 0.91 S1 + 0.079 T; at depth 92 cm | (4) |
| Sr = 0.423 + 0.98 S1 + 0.053 T; at depth 124 cm | (5) |
where
Sr = resulting suction in kPa
S1 = lowest suction attained during storm in kPa
T = period after rain in hours
Note that above equations are only empirical and they do not allow for positive pore water pressures to be established. The equations could be used only as a rough guide.
Field Soil Moisture Response to Rainfall
The response of field moisture contents to rainfall is in general not that encouraging. But nevertheless as shown in Figure 10, there appears to be an increase in the soil moisture with rain, which is to be expected. After the rain has stopped, there is a gradual decrease in the moisture content as the soil recovers its suction.
Figure 10. Field soil moisture content (at berm 1)
Field Soil Moisture Suction Relationship
Figure 11 shows a plot of field soil moisture with suction. As shown, there is a considerable scatter in the data. In any case it can be observed that the moisture content decreases with the increase in suction, which is to be expected.
Figure 11. Field soil moisture - suction relationship (at berm 1, depth 124 cm)
Soil Temperature Response
Figure 12 shows a typical ground temperature plot. As shown the fluctuation of ground temperature is from 28oC to 31oC. Although the plot seems to show a considerable scatter but the general trend is that there is a decrease of temperature due to the rainfall.
Figure 12. Temperature - time plot (at depth 30 cm)
CONCLUSIONS
From the results of this study, the following conclusions can be drawn.
The value of suction varies with depth and soil weathering grades. Suction is observed to decrease with increase in depth below the ground surface. Granitic residual soil of weathering grade IV appears to have the lowest value of suction, indicating the more porous nature of residual soil of this grade.
During the dry spell, the suction values at the shallower depth are observed to be generally larger than the suction values at the deeper depths.
Upon rain (wetter period), there is a general lost (reduction) of suction in all cases. However, the suction response to antecedent rainfall is less pronounced as the depth increases. The drop in suction at the deeper depth is only significant over a long wet period, indicating longer time required by the water to infiltrate the slope.
The higher rainfall intensity would results in larger drop in suction, particularly at the shallow depth.
After the cessation of the rain, suction at the shallow depth showed a faster recovery compared with the deeper depth.
There appears to be an increase in the soil moisture with rain. After the rain has stopped, there is a gradual decrease in the moisture content as the soil recovers its suction. Likewise there is a decrease of temperature due to the rain.
REFERENCES
 | |
| © 2005 ejge | |