1160
Proceedings of the 18
th
International Conference on Soil Mechanics and Geotechnical Engineering, Paris 2013
The magnitudes of the predicted changes in temperature and
rainfall for 2030 and 2070 are expected to cause a decrease in
soil moisture content from that experienced at present, thus
leading to higher soil shrinkage on expansive soil sites.
In order to quantify the effect of climate change predictions
on the magnitude of soil moisture changes, it is necessary to first
establish the relationship between climate and seasonal rainfall
and temperature. This is undertaken in this paper using the
Thornthwaite Moisture Index.
3 THORNTHWAITE MOISTURE INDEX
Thornthwaite (1948) proposed an empirical method for
estimating potential evapotranspiration as a climatic factor.
Thornthwaite’s method enables potential evapotranspiration
(
PE
) to be estimated using only monthly average temperature
data (
t
) for a particular location, and a simply derived adjustment
factor, which is applied to correct for latitude and month length
by Equation (1)
PE
a
I
t
10 16
x Latitude correction factor
(1)
where
49239 .0
017921 .0
0000771
.0
000000675
.0
2
3
I
I
I
a
In Equation (1), the heat index (
I
) is given by Equation (2)
12
1
i
i
I
, where
514 .1
5
t
i
(2)
By enabling estimation of monthly potential evapo-
transpiration and balancing this with monthly rainfall,
Thornthwaite (1948) made it possible to estimate seasonal
moisture deficiencies and surpluses.
A moisture deficit is defined as the amount by which the net
monthly potential evapotranspiration exceeds monthly rainfall
during a period of zero soil moisture storage. Similarly, a
moisture surplus is the amount by which the net monthly rainfall
exceeds potential evapotranspiration when soil moisture storage
is at capacity (defined as being 100 mm water); this surplus is
assumed to drain as runoff. Soil moisture recharge occurs during
months when the net moisture balance is positive and soil water
storage is below capacity at the beginning of the month, while
soil moisture depletion occurs when the net moisture balance is
negative, but the soil water storage at the beginning of the month
is non-zero. During the spring, soil moisture is rapidly utilized
by vegetation, particularly annual grasses, which quickly mature,
then die off once the soil moisture has been depleted.
The Thornthwaite Moisture Index (TMI) is calculated from
the derived moisture deficiency and surplus by Equation (3).
TMI
=
PE
Deficit
Surplus
)
(60 )
( 100
(3)
Determination of the TMI based on present day monthly
temperature and rainfall averages for Perth is outlined in
Table A1 of Appendix A. It is shown that for Perth at the present
time, the TMI = +16.
The TMI is used to categorize the climate of a particular
location. A higher positive value of TMI corresponds to a wetter
climate, while a greater negative number is associated with a
more arid climate. Therefore climate change predictions of
higher temperatures and lower rainfall will result in a smaller
TMI than at present for a particular locality.
4 PREDICTED CHANGES IN TMI
The author (Mitchell 2011,2012) used the seasonal temperature
rise and general rainfall decline predictions (Table 1) for
Adelaide, Melbourne, Sydney and Perth for 2030 and 2070 and
applied them to the current climate averages to yield new
average TMI estimates for each city in 2030 and 2070. An
example for Perth for 2070 is given in Table A2 of Appendix A,
and the results for all four cities are shown in Table 2.
Table 2: TMI Predictions for 2030 and 2070
TMI
City
Present
2030
2070
Adelaide
-19
-26
-34
Melbourne
-8
-14
-24
Perth
16
6
-14
Sydney
43
27
3
It is seen from Table 2, the predicted TMI for each city
examined decreases with time, indicating increasing aridity.
From a geotechnical perspective, predictions of a lower TMI
and hence an increase in the severity of desiccation of the soil
profile could potentially lead to an increase in the magnitude of
expansive soil movements for regions of Adelaide, Melbourne,
Sydney and Perth characterised by highly expansive soils. This
is quantified in the next section.
5 RELATIONSHIPS BETWEEN EXPANSIVE SOIL
MOVEMENTS AND TMI
By the Australian standard AS2870-2011, the characteristic
surface movement (
y
s
) of a site is determined from the design
soil suction changes and the soil instability index (
I
pt
) over the
depth of design soil suction change (
h
=
H
s
) by Equation (4) and
Figure 1 (the surface soil suction change is Δu
s
= 1.2 pF by
AS2870 - 2011).
s
H
h
pt
s
hu I
y
0
(4)
Figure 1. Design soil suction extremes for a “normal” site by AS2870-
2011
Equation (4) and Figure 1 are for “normal” sites such as sites
that are not affected by trees. When considering tree effects, one
recommended method of AS2870-2011, considers an additional
soil suction as shown in Figure 2. This leads to an additional soil
movement due to trees (
y
t max
) by Equation (5).
t
H
h
pt
t
s
hu I
y y
0
max
(5)
The value of
y
t
max
is corrected for the distance of the tree
from the footing of a building to give the actual tree effect
y
t
.
