Transcript FIRST STAKEHOLDER MEETING FOR

Likely Effects Of Climate Change on
Runoff from Limpopo Basin Botswana
By
Prof. B. P. Parida
Dept. of Environmental Science
University of Botswana
Background


Much of Southern Africa below Latitude 20o S
is basically Semi-Arid
This includes BOTSWANA where
Annual Rainfall ~ 250 – 500 mm
Evaporation ~ 2000 mm
Topography ~ Flat and Deep Sandy Soil
(These together limit the effectiveness of Rainfall)
Sources of Water

Rivers and Streams : Average runoff is
very low~0.6 mm for the whole country
> All endogenous rivers originating in Botswana
are ephemeral with an average of 10–70 days of
flow in a year.

Other Sources: Desalination, Water carriers
to tap international rivers – Very Expensive

Groundwater :
Location of Boreholes in Botswana
Sources..(Groundwater..contd)
Wellfield
Jwaneng N.
Extraction Rate Recharge Rate
in M m3
in M m3
10
2
Orapa
5
0.2
Palla
12
2.4
(Source: BNWMP, 1991)
Water Supply..

Clearly Botswana is one amongst the
water stressed countries of the world
(where water use is more than 20% of their renewable water
supply)

Botswana is highly dependent on its
natural resource base : mining,
livestock, wild-life based tourism,
textiles, soda ash and agriculture
In Summary:
Water is the single most important
constraint to the development of
Botswana
The Limpopo Basin



The major source of water in the eastern
Botswana and it is important because of its
strategic value to Botswana, South Africa
Zimbabwe and Mozambique.
Drains from a catchment area of about
80 000 km2 (~1/8 area of Botswana)
Approximately carry about 10 mm of surface
water annually.
The Limpopo Basin


The catchment serves as the main source for
Gaborone & rapid and ongoing expansion of
southeastern Botswana.
In Botswana alone.. 4 dams viz: Gaborone,
Bokaa, Letsibogo and Shashe have been built
with a total capacity of 350 M m3 (against
projected demand of about 325 M m3 by
2020)
Indicative Studies Reveal that:

A rise of 0.02 to 0.05oC per decade in Africa
in the present century (IPCC)
(greatest over the interior of semi-arid margins of
Sahara and Central Southern Africa.)

Limpopo basin is likely to experience between
5 to 15% decrease in rainfall,
5 to 20% increase in Evaporation, and
25 to 35 % decrease in streamflow. (Desanker and
Magadza, 2001).
Main objective
To quantify the effect of climate change
on the flow regime of Limpopo basin –
hence develop flow scenarios.


to find by what percentage the runoff will
either increase or decrease in the coming
years?
what will be the trend in flow over the coming
decades in this century?
Philosophy:


Rainfall ~ Runoff is a complex paradigm as it
represents the catchment’s response which is
a function of land use
So, any adaptation strategy that needs to be
developed to overcome the effects of climate
change, would primarily centre around a well
structured catchment management strategy
along with strategies for water use
management.
PROPOSED METHODOLOGY



Use past climatic and flow information
(monthly rainfall, PE, and flow data over the
past 20-30 years including data on landuse)
to carry out water balance, hence compute
the annual yield.
Use the past data as above to develop an
ANN model which can be used for flow
predictions.
Use the generated climatic scenario to
develop flow scenario for the Limpopo.
CONCEPT OF MULTI-CELL WATER BALANCE :
P = AET + OF + ΔSM + ΔGWS +GWR + Interception
P
I
AET
Infiltration
SM
OF
GWS
Percolation
to
G/w ater
GWR
Impervious
stream
Data Collection:




Rainfall Data: Monthly data at 37
stations (1971 – 1998)
Evaporation Data: Monthly data at 4
stations (1971 – 1998)
Flow Data : Monthly data at 1 station
(1978 – 1998)
Soil Data : Field Capacity of major soil
types in the basin as per the soil atlas.
Procedure for Multi-cell Water
Balance:




Long term average rainfall values and
evaporation values are superimposed on the
soil map and the catchment is then divided
into 75 cells – each cell basically representing
uniform rainfall, evaporation and soil type.
For each cell and for a given year runoff is
computed. (Thornthwaite and Mather, 1957)
For each year total runoff from the catchment
is computed at the outlet by adding up flows
from the 75 cells.
Total runoff / Total rainfall = Runoff coeff.
A TYPICAL WATER BALANCE COMPUTATION TABLE
( Field Capacity = 125 mm )
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
DEC
TOTAL
P
103
33
21
12
32
0
0
12
12
68
122
69
484
PE
139
80
86
92
68
52
62
79
118
141
120
135
1172
P-PE
-36
-47
-65
-80
-36
-52
-62
-67
-106
-73
2
-66
APWL
-102
-149
-214
-294
-330
-382
-444
-511
-617
-690
0
-66
SM
55
37
22
12
9
6
3
2
1
0
125
73
ΔSM
-18
-18
-15
-10
-3
-3
-3
-1
-1
-1
125
-52
AET
121
51
36
22
35
3
3
13
13
69
120
121
DEFICIT
18
29
50
70
33
49
59
66
105
72
0
14
SURPLUS
0
0
0
0
0
0
0
0
0
0
123
62
TARO
31
16
8
4
2
1
0
0
0
0
123
62
RO
15
8
4
2
1
1
0
0
0
0
61
31
DET
16
8
4
2
1
0
0
0
0
0
62
31
565
123
Average Rainfall in Limpopo Catchment system
between 1971 and 1998
Avearge Rainfall (mm)
y = -3.108x + 6556.2
800.00
600.00
400.00
200.00
0.00
1970
1975
1980
1985
Years
1990
1995
2000
Runoff Coefficient for Limpopo Catchment system
between 1971 and 1998
Runoff Coefficients
y = 0.0049x - 9.255
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
1970
1975
1980
1985
Years
1990
1995
2000
CUSUM Analysis of Rainfall Values
in the Limpopo Catchment
CUSUM Rainfall Values
800.00
600.00
400.00
200.00
0.00
1970
-200.00
1975
1980
1985
-400.00
Years
1990
1995
2000
Runoff Coefficients (%)
Rainfall (mm)
Runoff Coefficient in % and Rainfall for the Limpopo
Catchment between 1981 and 1998
700
600
500
400
300
200
100
0
1980
1985
1990
Years
1995
2000
CONCEPTUALISATION OF ARTIFICIAL
NEURAL NETWORK BASED ON A
BIOLOGICAL NEURON
Dendrides
Cell body
Axon
Nucleus
Axons
Nucleus
Dendrites
Sunapses
HOW DOES THE ANN WORK ??
Xo
Wo
W1
X1
Sum
W2
Transfer
X2
Output
path
W3
X3
.
.
.
Processing
element
W4
Xn
Inputs Xn
Weights W n
The basic structural functioning of a neuron
A MULTI-LAYERED FEED FORWARD NEURAL NETWORK
WITH ONE HIDDEN LAYER
input layer
Hidden layer
Output layer
Output array
A multi-layer feedfoward neural network with one hidden lay er
Conclusion
Using scenario of monthly rainfall and
evaporation values from the GCM / RCM
models (AF 07), scenarios on likely
runoff from the Limpopo basin by the
above two techniques will be created.
Co-operation needed:


Monthly rainfall and evaporation data
scenarios for the Limpopo basin in the
next 30 – 50 years will be necessary
(GCM /RCM / Empirical / Other climate
Model).
Formulation of appropriate catchment
management strategies as possible
adaptation strategy.