In this study, we carry out an assessment of the impact of urbanization on Butare, southern Rwanda and evaluate its effect on the hydrological process at the Migina catchment. We used data from the meteorological stations of the Migina catchment, land-use maps of the region (obtained in 1974 and 2010) and GIS technology in order to analyse hydrological fluctuations in Migina catchment and the expansion of Butare town. We observed that, between the two hydrological years (1974 and 2010) water runoff increased by over 3.5% with increasing flood risks. Furthermore, evapotranspiration and groundwater level have decreased by about 3% and 0.5% leading to water scarcity. Research on urban planning is necessary in order to improve on water resources management and reduce the effects of urbanization on water resources such as in the Migina catchment.
Water Resources Urban Planning Urbanization Hydrological Cycle Land Cover Migina Catchment Rwanda1. Introduction
Rain water, either returns to the atmosphere through evapotranspiration or infiltrates the soil, runs off on the surface and likely ends up into a river, ocean or lake (Figure 1) [1] - [3] . The relative proportions depend on the nature of the surface, duration of rainfall and intensity [1] [3] [4] .
Water cycle diagram showing different processes involved before and after urbanization. Sourced from UNCE, 2015
Development of an urban area involves covering the ground with artificial surfaces, and it significantly increases the amount of surface runoff in relation to infiltration and evapotranspiration (Figure 2). This therefore increases the total volume of water reaching the river when or after it rains. As a result, this will reduce water that would be stored within a catchment and contribute to water scarcity [5] . Furthermore, the construction of storm sewers and culvert of natural streams and channels which takes place during urbanization results in water being transmitted into the drainage network more rapidly; this increases outflow velocity [6] [7] . Since a larger runoff is discharged within a shorter interval, peak rate of water inevitably increases, giving rise to the danger of sudden flooding of rivers. These subsequent floods damage human and natural resources as well as properties [8] .
When people develop within a watershed, the hydrologic cycle is affected. The increase in impervious or hard surfaces, including rooftops and pavement (roads, driveways, and parking lots) decreases the amount of water that soaks into the ground or infiltrates [2] [9] . This increases the amount of surface runoff (Figure 3). If infiltration is significantly decreased, groundwater levels may decline, affecting stream flows especially during the dry season. Meanwhile, lowered groundwater levels can result in subsequent well failures [1] [10] . While the effects of urbanization on the water cycle can be major, if wise choices are made during the development process, the impacts can be minimized and future water supply can be protected [10] [11] .
Although Rwanda is among the least urbanized countries with an urban population of 28.8% in 2015 [12] , it had a rapid population growth of about 2.7% in 2010-2015 (UN, 2016) and a relatively high urbanization rate of 6.43% in 2015 [12] [13] . On the other hand in Rwanda, observations from existing data showed that over the last 30 years, some parts of Rwanda have experienced unusual irregularities in climate patterns including variability in rainfall frequencies and intensity, persistence of extreme weather conditions such as heavy rainfall in the northern regions and drought in the eastern and southern regions [14] .
The meso-scale Migina catchment (257.57 km2) located in southern Rwanda experiences serious human activities in its northern zone. This is as a result of the rapid urbanization of Butare town as it progressively destroys the natural land cover surrounding the catchment area. The topography and land cover of drainage basins affect phenomena like erosion sediments that are transported from storm runoffs towards rivers. These produce changes in water balances of the concerned drainage basin [15] . Also, as new development projects are continuously being implemented within the area, their growing impact affects the hydrologic cycle in general [3] [10] [16] .
The importance of the hydrological cycle has been in existence for billions of years and continuously renews and refreshes finite water supplies [10] . Therefore, lack of public awareness on implication of their daily activi-
Effect of urbanization on hydrological cycle. Sourced from Carlos, 2007. (a) Pre-urbanization; (b) Urbanization.
Influence of urbanization on different components of the water cycle. Sourced from CWP, 2015
ties on water resources has contributed to its inefficient use and scarcity. This paper looks at urbanization of Butare town and studies its effects on the long term availability of water resources within the catchment. The study strives to espouse scientific evidence that would aid the local population, urban planners and decision makers, such as to promote cleaner and safer urban expansion while ensuring continuous water availability within the Migina catchment.
2. Methodology and Data Analysis2.1. Study Area
Migina catchment is located between 2˚32' and 2˚ 48' South, 29˚42' and 29˚48' East, in the Southern province of Rwanda. It is drained by perennial streams with the main flow following a north to south direction. The main stream is located in the eastern part of the catchment with most of the valleys draining from north-west to south-east towards the main stream. The Migina catchment drains into the Akanyaru River, which forms the border between Rwanda and Burundi. Downstream, the Akanyaru River joins the Nyabarongo River to form the Kagera River, which is the largest tributary of Lake Victoria and a precursor of the Nile River, Figure 4.
The Migina catchment area is mountainous with elevation ranging from 1434 m in the south to 2251m around the Huye Mountains in the North West. Land use is dominated by pasture and arable farming such as rice, sorghum, maize, and sweet potato [17] . Butare town, with over 100,000 inhabitants is located in Huye district, stretching along the main axes of Akanyaru, Nyabarongo and Kagera Rivers.
The meteorological data recorded from different meteorological stations in the catchment (Butare airport, Butare mission, Kansi and Rubona) shows that the mean annual temperature is around 24˚C and shows a progressive increase in the area’s temperature, Figure 5 [18] .
2.2. Determination of Two Annual Water Balances
The water balance is an account of the inputs and outputs of water, and it often applies to a watershed or a surface water reservoir. Any difference between the inflow (I) and the outflow (O) over a certain time period represents the water that has entered or left stored (S) within the catchment [19] .
According to [15] , the water balance may also be written as;
Land cover of Migina catchment
Mean annual temperature variation within the study area between 1971 and 2010
The present study considers Inflow as Precipitation (P), as Evapotranspiration (E) + Outlet discharge (Q), while Stock (S) is the quantity of water stored in the catchment. The components of water balance were thus estimated by:
The measured evaporation (Epan) has to be compensated with a pan factor (fpan) to obtain the real open water evaporation [20] .
We used the most popular method for the calculation of the potential evapotranspiration (PET), which computes the crop evapotranspiration according to [16] . Where kc is the crop factor and ETref is the reference evapotranspiration.
The actual evaporation (AET) is then calculated by:
The Rational Formula for estimating peak runoff rate, introduced in the USA in 1889 has become widely used as a tool for drainage design, particularly for sizing water conveyance structures [21] .
where, Q is the peak runoff rate (m3/s), C, the runoff coefficient (unit less, ranging from 0 to 1) while I is the rainfall intensity (mm/hr), and A is the watershed area (ha). The conversion factor K is an estimated value 0.00278 (unit less).
2.3. Determination of Inflow
Rainfall data was collected from 12 meteorological stations located within the catchment area between June 2009 and May 2011 and Thiessen polygons were used to estimate the mean annual rainfall, Table 1.
Similarly, the quantity of water that entered the catchment between 1973 and 1974 was obtained by collecting data from four rain gauges installed at Save, Butare airport, Nyakibanda and Kansi. Thiessen polygons were then used to calculate the mean (average) annual rainfall, Table 2.
Annual rainfall for 2009/2011 using Thiessen polygon method
YEAR
Month
Rango P (mm)
Mubumbano P (mm)
Murama P (mm)
Vumbi P (mm)
Mpare P (mm)
Sovu P (mm)
Save B P (mm)
Muyira P (mm)
Kibilizi P (mm)
Gisunzu P (mm)
Rwasave P (mm)
Kansi A P (mm)
2009
June
6.73
7.57
4.46
5.72
10.94
12.79
4.04
18.68
7.41
21.60
4.00
20.20
July
1.01
0.00
0.00
0.00
0.17
2.52
0.67
0.84
5.55
0.17
1.60
0.00
August
22.55
48.22
50.16
26.93
38.71
4.38
2.78
40.06
25.00
28.61
3.50
55.63
September
50.24
59.58
56.89
44.27
47.63
55.71
28.95
42.25
43.09
48.05
37.70
89.38
October
134.99
94.51
91.99
66.49
78.02
151.82
121.44
55.38
64.47
132.82
129.30
113.00
November
202.57
116.81
122.70
191.38
250.96
211.41
201.81
112.60
110.16
141.72
168.50
113.50
December
123.55
120.01
131.96
119.34
132.38
134.65
136.42
128.76
63.62
127.74
51.70
111.10
2010
January
199.29
209.81
182.46
335.29
226.39
151.99
190.28
213.60
148.96
219.68
134.40
147.13
February
208.38
260.64
238.93
237.66
183.44
180.52
183.63
280.25
167.98
253.48
288.70
288.20
March
320.90
161.59
159.99
153.00
148.04
158.05
84.92
229.42
134.07
161.43
205.50
194.06
April
390.92
168.23
144.92
226.39
245.67
158.81
123.55
196.17
142.57
181.87
112.60
138.88
May
194.26
259.21
177.45
154.01
234.67
163.32
111.66
214.89
176.31
141.10
172.22
166.38
June
32.08
33.41
53.81
46.79
27.50
23.38
11.92
11.00
7.70
27.50
17.88
37.58
July
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
August
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
September
256.67
77.09
103.03
51.00
74.25
126.50
103.13
66.92
59.13
79.48
106.33
97.81
October
231.00
51.84
114.31
47.30
90.75
107.71
53.17
62.33
45.83
45.93
92.13
52.34
November
286.92
147.45
157.12
108.40
141.17
39.88
115.96
172.33
99.88
120.54
138.42
89.83
December
151.25
87.36
108.63
109.49
154.00
77.46
185.17
70.13
103.22
119.63
107.25
93.59
2011
January
196.17
100.82
101.57
97.96
118.25
98.08
252.08
87.08
108.17
73.61
154.92
293.33
February
239.25
144.67
89.83
74.40
132.92
192.04
150.88
166.83
132.55
125.03
197.54
179.76
March
220.00
130.53
178.20
126.91
132.46
126.96
123.75
326.79
126.50
186.08
145.29
148.59
April
420.75
147.19
120.54
127.25
155.83
109.08
141.17
546.33
333.30
178.75
142.54
136.68
May
285.08
112.60
122.28
102.00
141.17
86.63
129.25
125.58
197.82
124.30
118.25
87.82
An. Rain. 1st Year (mm)
1855.39
1506.19
1361.90
1560.47
1597.01
1385.99
1190.16
1532.90
1089.18
1458.28
1309.72
1437.44
An. Rain. 2nd Year (mm)
2319.17
1032.97
1149.32
891.49
1168.29
987.71
1266.47
1635.33
1214.08
1080.84
1220.54
1217.33
Annual Rainfall (mm)
2087.28
1269.58
1255.61
1225.98
1382.65
1186.85
1228.31
1584.12
1151.63
1269.56
1265.13
1327.39
Thiessen Polygon (Km2)
18.18
33.81
34.35
23.99
6.36
34.72
11.82
11.45
11.27
28.72
17.45
25.45
2.4. Determination of Outflow
Data for pan evaporation (Epan) were obtained from two self-made replications of the US Weather Bureau “Class A” evaporation pans installed at the Rwasave fishpond and Gisunzu meteorological station. The reference potential evapotranspiration was established for data collected from June 2009 to May 2011. In order to obtain the real open water evaporation, the estimated Epan was to be compensated for by a pan factor (fpan), where the pan factor depends on the type of pan used, pan environment and climate. For the “Class A” evaporation pan,
Annual rainfall for 1973/1974 using Thiessen polygon method (mm)
Year
Month
Butare P(mm)
Kansi P(mm)
Nyakibanda P(mm)
Save P(mm)
1973
January
182.40
120.40
190.00
123.50
February
110.50
176.30
67.80
100.10
March
84.70
42.70
61.40
90.50
April
141.00
223.20
189.10
177.60
May
228.60
211.00
252.30
334.80
June
54.50
20.70
0.00
19.50
July
0.00
0.00
0.00
0.00
August
3.60
17.70
0.00
16.20
September
216.30
182.60
166.50
133.80
October
117.50
107.60
146.50
66.90
November
154.00
104.80
109.80
163.90
December
72.20
69.50
104.80
36.50
1974
January
77.50
62.60
71.60
46.90
February
58.90
49.90
100.80
85.00
March
253.80
152.40
167.30
164.90
April
172.00
172.50
183.40
211.40
May
117.20
138.90
119.20
142.70
June
71.90
16.80
30.10
51.50
July
21.20
17.60
29.70
49.40
August
12.90
0.00
0.00
4.20
September
90.30
69.20
114.80
117.10
October
55.00
47.80
45.00
17.70
November
178.40
162.30
104.60
179.90
December
86.40
125.90
91.80
77.00
Annual Rainfall 1973 (mm)
1365.30
1276.50
1288.20
1263.30
Annual Rainfall 1974 (mm)
1195.50
1015.90
1058.30
1147.70
Avg. Annual Rainfall (mm)
1280.40
1146.20
1173.25
1205.50
Thiessen Polygon (Km2)
60.30
105.02
58.77
33.47
fpan varies between 0.35 and 0.85 with an average value of 0.70 [20] . Based on a relative high humidity (>70%), moderate wind speed and green crops on a windward side distance in the order of 10 m [20] , an empirically derived pan factor of 0.55 was used, Table 3.
The final, actual or adjusted crop factors (Kadj) was calculated as shown in Table 4.
The dominant crops and their corresponding characteristics were described in [7] and the adjusted evaporation coefficient (Kadj) is therefore equal to 0.74.
The Actual Evapotranspiration (Equation (2.5)) becomes:
Reference evapotranspiration for 2009-2011 (mm)
Year
Month
Rwasave E (mm)
Gisunzu E (mm)
Avg. E (mm)
ET (mm)
Column (1)
Column (2)
(3) = (1)*(2)/2
0.55*(3)
2009
June
140.00
125.60
132.80
66.40
July
218.60
145.17
181.88
90.94
August
145.50
113.61
129.56
64.78
September
140.70
154.05
147.38
73.69
October
151.30
141.80
146.55
73.28
November
149.50
138.68
144.09
72.05
December
118.70
133.75
126.23
63.11
2010
January
134.40
167.74
151.07
75.53
February
100.90
159.07
129.98
64.99
March
103.60
129.43
116.51
58.26
April
124.90
133.87
129.38
64.69
May
116.35
113.17
114.76
57.38
June
135.58
130.50
133.04
66.52
July
198.00
147.00
172.50
86.25
August
194.00
170.00
182.00
91.00
September
179.33
154.48
166.90
83.45
October
170.13
153.93
162.03
81.01
November
147.42
138.54
142.98
71.49
December
151.25
129.63
140.44
70.22
2011
January
154.92
116.61
135.76
67.88
February
183.54
127.03
155.29
77.64
March
181.29
141.17
161.23
80.61
April
98.04
157.75
127.89
63.95
May
91.25
87.32
89.28
44.64
An. Et 1st Year (mm)
1644.45
1655.94
1650.19
825.10
An. Et 2nd Year (mm)
1884.74
1653.94
1769.34
884.67
Annual ET (mm)
1764.60
1654.94
1709.77
854.88
Calculation of adjusted evaporation coefficient 2009-2011
To estimate the final crop factors to be used for the calculation of the Actual Evapotranspiration (1973-1974), the land cover was assessed for the entire catchment during 1973-1974 (Table 5).
The adjusted evaporation coefficient (Kadj) is therefore equal to 0.75.
The Actual Evapotranspiration (AET) becomes:
a. Discharge in 2009-2010
To investigate surface water discharge in Migina catchment, five gauging stations were installed in 2009 Table 6.
Before the newly installed gauging stations in 2009, six old gauging stations were installed in 1993, but were destroyed during the Rwandan crisis of 1994. Therefore, Runoff coefficient was used to deduce the amount of discharge that flowed out of Migina catchment between 1973 and 1974. The rational method (Equation (2.6)) was used to know the runoff coefficient for the catchment.
Calculation of adjusted evaporation coefficient for 1973-1974
For urban areas, the runoff coefficient is C1 = 1.
The C2 (runoff coefficient) for the remaining land cover was therefore obtained as shown below;
The discharge
3. Results and Discussions3.1. Water Balance of Migina Catchment between 2009 and 2011
The water balance (Equation (2.1)) is given by:
Inflows in this study are related to Precipitation (P).
Outflow relates to Evapotranspiration (E) + Outlet discharge (Q).
Stock (S) is the quantity of water stored in the catchment.
Therefore, the inflow in Migina catchment or the annual rainfall for June 2009-May 2011 is 1325.34 mm.
The outflow for Migina catchment is represented by evapotranspiration and outlet discharge where the annual reference evapotranspiration is 854.88 mm while the annual actual evapotranspiration for Migina catchment is equal to 597.80 mm. Also, the outlet discharge (the average) is equal to 1.13 m3/s.
In estimating the water balance of Migina catchment for June 2009-May 2011, the following equation was used.
From the results obtained:
The water balance of Migina catchment during June 2009 to May 2011 is calculated and found to be:
And can be simplified as:
The change in stock within Migina catchment during June 2009 to May 2011 is estimated to be 152 millions of m3.
3.2. Water Balance of Migina Catchment between 1973 and 1974
The inflow in Migina catchment (annual rainfall for 1973-1974) is 1191.5 mm. The outflow such as the annual reference evapotranspiration is equal to 768.55 mm. Therefore, the annual actual evapotranspiration for Migina catchment is equal to 576.41 mm. Meanwhile, the average discharge for 1973-1974 stands at 0.65 m3/s.
The water balance of Migina catchment for 1973-1974 is obtained as follows;
From the results obtained:
The water balance of Migina catchment during 1973-1974 is calculated and found to be:
And can be simplified as:
The change in stock within Migina catchment during 1973-1973 is estimated to be 138 million of m3.
3.3. Impact of Urbanization on the Water Balance of Migina Catchment
a) During 2009 and 2011
Evapotranspiration represents 45% of the total inflow, while runoff represents 10.5% of the total inflow and change in storage represents 44.5% of the total inflow.
b) During 1973-1974
Evapotranspiration in 1973/1974 is 48% of the total inflow, runoff is 7% of the total inflow and change in storage is 45% of the total inflow. These findings tally and are in close range as that given in [22] . These studies revealed that, during the last 30 years the runoff has increased by 3.5%. This may be attributed to the infiltration of water into the soil [23] [24] .
This may also lead to increase pollution due to surface water run-off which may also spread water borne diseases as a result of little or no water self-purification. This increase in runoff also reduces water storage hence leads to water scarcity within the surrounding area [25] - [27] .
A decrease of 3% in evapotranspiration maybe due to the conversion of green spaces for urbanization purposes especially housing. This reduces surface albe do hence a drop in the contribution of evapotranspiration to rainfall. A progressive reduction in rainfall is gradually created with consequences for agriculture due to recurrent droughts [24] [26] . The change in storage has slightly decreased by 0.5% because increase runoff causes out- flow of catchment water that could have been stored as groundwater [26] [27] .
These changes in the hydrological cycle could be attributed to the new residential plots of Butare town whose population increased from a mere 8400 in 1970 to over 100,000 inhabitants in 2009. This growth in population has been detrimental to the natural landscape cover hence fluctuations in the hydrological cycle.
4. Conclusions
The water balance of Migina catchment was estimated over a 30-year interval with more rainfall experienced during 1973/1974 in Migina catchment than in 2009-2011. These observations were also true for surface water runoff which was considerably lower in 1973/1974 and higher in 2009-2011. These changes are all due to an increase in demand for basic and social amenities as a result of an increase in population of Butare town.
The consequences of urbanization on the hydrological cycle could be curtailed by structural adjustment programmes including urban greening and provision of detention of runoff water in urban drainage systems.
Cite this paper
Olivier Clement Gatwaza,Xin Cao,Mukete Beckline, (2016) Impact of Urbanization on the Hydrological Cycle of Migina Catchment, Rwanda. Open Access Library Journal,03,1-12. doi: 10.4236/oalib.1102830
NOTESReferencesMunyaneza, O., Mukubwa, A., Maskey, S., Wenninger, J. and Uhlenbrook, S. (2013) Assessment of Surface Water Resources Availability Using Catchment Modeling and the Results of Tracer Studies in the Meso-Scale Migina Catchment, Rwanda. Hydrology and Earth System Sciences Discussions, 10, 15375-15408. http://dx.doi.org/10.5194/hessd-10-15375-2013Adrian, B. and Pede, Z. (2012) Design in Nature: How the Constructal Law Governs Evolution in Biology, Physics, Technology, and Social Organization. Random House, New York.UNCE (2015a) University of Nevada Cooperative Extension. Effects of Urbanization on the Water Cycle. http://www.unce.unr.edu/publications/files/nr/2004/FS0443.pdfMukete, B. and Sun, Y. (2014) Assessing the Effectiveness of Urban Nature Reserves on Biodiversity Conservation. Applied Ecology and Environmental Sciences, 2, 130-134. http://dx.doi.org/10.12691/aees-2-6-1Nahayo, D., Wali, U. and Anyemedu, F. (2010) Irrigation Practices and Water Conservation Opportunities in Migina Marshlands. International Journal of Ecology and Development, 16, 100-112.Mugabe, T., Chitata, T., Kashaigili, J., Chagonda, I. and Nyagumbo, I. (2011) Modelling the Effect of Rainfall Variability, Land Use Change and Increased Reservoir Abstraction on Surface Water Resources in Semi-Arid Southern Zimbabwe. Physics and Chemistry of the Earth, Parts A/B/C, 36, 1025-1032. http://dx.doi.org/10.1016/j.pce.2011.07.058Munyaneza, O., Wali, U., Ufiteyezu, F. and Uhlenbrook, S. (2011) A Simple Method to Predict River Flows in an Agricultural Catchment in Rwanda. Nile Water Science and Engineering Journal, 4, 24-36.Ufiteyezu, F. (2010) Prediction of Migina River Discharge in Contribution to Hydrometeorology in Rwanda. National University of Rwanda, Butare, Rwanda.Hategekimana, D. (2007) Water Balance of Lake Muhazi. MSc Thesis, National University of Rwanda, Butare, Rwanda.UNCE (2015b) University of Nevada Cooperative Extension. Urbanization and the Water Cycle. http://www.unce.unr.edu/publications/files/ho/2005/av0512.pdfMukete, B. and Monono, S. (2014) Assessing the Impact of Consumer Behaviour on Food Security in South West Cameroon. Journal of Food Security, 2, 87-91.Central Intelligence Agency (CIA) (2015) The World Factbook. https://www.cia.gov/library/publications/the-world-factbook/fields/2212.htmlUnited Nations Organization (UN) (2015) UN Data: A World of Information.http://data.un.org/CountryProfile.aspx?crName=RWANDAREMA (2009) Economic Impact of Climate Change in Rwanda. REMA, Kigali, Rwanda. Savenije, H. and De Laat, P. (1992) Textbook of Hydrology. Delft, The Netherlands.Munyaneza, O. (2008) Space-Time Patterns of Hydrological Processes and Water Resources in Rwanda, with Special Focus on the Meso-Scale Migina Catchment. PhD Research Proposal, UNESCO-IHE Institute for Water Education, Delft, The Netherlands.Ying, H., Gerdo, A., Charle, W., Julia, C., Adriaan, G., Matthew, S. and Alexander, R. (2011) The Infrastructure Research Initiative at SWA. Germany.Iraguha, G. (2006) Etude de la dynamique des paysages par télédétection: Cas de la ville de Butare. Mémoire, National University of Rwanda, Butare.Van den Berg, W. and Bolt, H. (2010) Catchment Analysis in the Migina Marshlands, Southern Rwanda. MSc Thesis, Vrije University Amsterdam, Amsterdam.Musoni, J., Wali, U. and Munyaneza, O. (2009) Runoff Coefficient Classification on Nyabugogo Catchment. 10th International Waternet/Warfsa/GWP-SA Symposium, Entebbe, 28-30 October 2009.Carlos, E. and Tucci, M. (2007) Urban Flood Management. Hydraulic Research Institute, Federal University of Rio Grande, Rio Grande.Chibulu, B. (2007) Effect of Rainfall Variability on Crop Yield under Semi-Arid Conditions at Sub-Catchment Level. MSc Thesis, University of Zimbabwe, Harare.Mukete, B. (2014) Overpopulation and Biodiversity Conservation in Sub-Saharan Africa. In Discussions on Forest Conservation Issues of Sub-Saharan Africa Series 1 Volume 2. https://www.academia.edu/7759415/Overpopulation_and_Biodiversity_Conservation_in_Sub-Saharan_AfricaNicholson, S. (2001) Climatic and Environmental Change in Africa during the Last Two Centuries. Climate Research, 17, 123-144. http://dx.doi.org/10.3354/cr017123Mearns, R. and Norton, A. (2010) Social Dimensions of Climate Change: Equity and Vulnerability in a Warming World. New Frontiers of Social Policy, World Bank Publications No. 52097.Fatemeh, A. (2011) A Methodology for Risk-Based Optimization of Urban Drainage System. MSc Thesis, Vrije University Amsterdam, Amsterdam.Tsheko R. ,et al. (2003)Rainfall Reliability, Drought and Flood Vulnerability in Botswana Water SA 29, 389-392.Fischer, B. and Savenije, H. (2009) A Spatial and Temporal Rainfall Analysis of the Pangani Basin and Makanya Catchment, Tanzania. Geographycal Research Abstracts, 11, 79-86.