Environmental Engineering Reference
In-Depth Information
appear to change with time and were similar for all ceramic
stones tested.
A hysteresis model was proposed by Feng (1999) to mit-
igate the influence of hysteresis associated with the ceramic
block. The model was designed to trace scanning curves
both during the wetting process and during the drying pro-
cess. Calculated and measured scanning curves associated
with changing from the drying curve to the wetting curve
are shown in Fig. 4.61. The model can be used to subse-
quently modify a dataset of measurements obtained during
field monitoring.
The monitoring readings are collected and soil suction
values are first computed through use of the drying calibra-
tion curve. Once the data are compiled in a database, it is
possible to further process the data by examining whether
the past data points have been undergoing a drying or wet-
ting process. This information is used to determine which
calibration curve is best to use for the final calculation of
soil suction at any particular time. In this way it is possi-
ble to increase the accuracy of the measured soil suctions.
Figure 4.62 shows the difference between using the dry-
ing calibration curve or the wetting calibration curve when
interpreting field monitoring suction data. The improved
approach utilizes the drying calibration curve when the soil
is undergoing drying and the wetting calibration curve when
the soil is undergoing a wetting process.
Figure 4.63 shows a comparison between matric suctions
measured using a tensiometer and a GCTS thermal conduc-
tivity suction sensor where hysteresis was taken into con-
sideration. Both sensors were installed in the Indian Head
till and the soil was subjected to a series of drying and
wetting processes. During the drying process, the drying
calibration curve was used to calculate soil suction, and
during the wetting process, the wetting calibration curve
105
100
Measured
Calculated
95
90
85
80
75
70
65
0
1
10
100
1000
Suction, kPa
Figure 4.61
Comparison of applied suctions and predicted suctions when using hysteresis model
proposed by Feng et al. (2002).
46
140
36
120
Temperature
26
100
16
80
6
Suction (drying curve)
Suction with correction for hysteresis
60
-4
40
-14
Suction (wetting curve)
20
-24
-34
0
3/4/2001
12/7/2001
10 days
Time (days)
Figure 4.62
Effect of using both drying and wetting calibration curves to interpret thermal
conductivity soil suction measurements in the field (from Tan et al., 2007).
Search WWH ::
Custom Search