A model for reservoir yield under climate change scenarios for the water-stressed City of Bulawayo, Zimbabwe

Tags: water supply, Zimbabwe, precipitation, Filabusi, Esigodini, precipitation pattern, Bulawayo, decade, Zimbabwe National Water Authority, Upper Mzingwane, maximum temperatures, rainfall, catchment areas, municipal water supplies, excessive evaporation, Bulawayo Mean Min, increase, noticeable decrease, warming trend, maximum temperature, storage works, data collection, Matopos Research Station, Bulawayo, Zimbabwe, global climate change, Department of Civil Engineering, Department of Engineering Services, Mount Pleasant, Harare, Zimbabwe, Harare, Zimbabwe, southern Zimbabwe, Zambezi Basin, Monthly runoff data, University of Zimbabwe, water rationing, annual rainfall, coefficient of Variation, Data Department, water rights, Mzingwane
Content: A MODEL FOR RESERVOIR YIELD UNDER climate change SCENARIOS FOR THE WATER STRESSED CITY OF BULAWAYO, ZIMBABWE Bekitemba Moyo1,2, Elisha Madamombe3 and David Love 4,5,6 1 Department of Civil Engineering, University of Zimbabwe, PO MP 167, Mount Pleasant, Harare, Zimbabwe 2 Department of Engineering Services, Beitbridge RDC, PO 32, Beitbridge, Zimbabwe 3 Research and Data Department, Zimbabwe National Water Authority, Harare, Zimbabwe 4 WaterNet, PO Box MP600, Mount Pleasant, Harare, Zimbabwe 5 ICRISAT Bulawayo, Matopos Research Station, PO Box 776 Bulawayo, Zimbabwe 6 Department of Geology, University of Zimbabwe, PO Box MP 167, Mount Pleasant, Harare, Zimbabwe ABSTRACT The City of Bulawayo, the second largest city in Zimbabwe (population 981,000), although located in the Zambezi Basin, sources its water mainly from five reservoirs located in the Upper Mzingwane Subcatchment, Limpopo Basin. The reservoirs have failed to store their expected 4 % yields. The city has frequently implemented water rationing or lesser restrictions. The city is currently under stringent water rationing since the total runoff into the water supply reservoir was low. A study was undertaken to determine if reservoir yields were falling, and possible causes thereof. 1
Examination of 30-year rainfall and runoff records showed declining precipitation and runoff on the five year moving averages. Dam yields, determined by Yield 200 model, have been declining over the same period, from 131.3 Mm3 in 1980 to 67.90 Mm3 in 2005. The phenomenon, which might influence such a trend, is global climate change. Future yields of the reservoirs were estimated by two methods. The yield was projected from the 30-year trend to year 2030 using the best-fit line. This formed one scenario, while two additional projections were determined using two scenarios of runoff and precipitation decline, predicted from IPCC SRES emissions scenarios. Both of these predicted less reduced yields by 2030 (45.04 Mm3 and 41.72 Mm3) than the yield projection from current data (67.90 Mm3), suggesting the possibility that the impact of climate change in southern Zimbabwe may be higher than predicted by global models. Unrestricted water demand for the city is projected to 83.70 Mm3 of raw water by year 2030. This is far above even the most optimistic yield projection. It can therefore be concluded that the City should have determined additional sources of water already. The state-owned Mtshabezi dam and the Nyamandlovu aquifer connection could provide an increased total water supply in the short-term, but longer-term solutions, water demand management or additional sources, or both are required. Key words: Catchment management, climate change, emission scenarios, reservoir yields 2
INTRODUCTION Climate change occurs at a time scale of over one hundred years. It may be artificial: due to increasing concentrations of greenhouse gases in the atmosphere (IPCC, 2000) or natural: caused by periodic changes in distribution of incoming solar radiation resulting from variations in the earth's orbital geometry that is the tilt, precision of equinoxes and eccentricity (Graciano et al., 2004). Greenhouse gas emissions are starting to change our climate, and an increasing body of evidence shows the climate is changing: · The global average surface temperature has increased over the 20th century by about 0.6°C · Temperatures have risen during the past four decades in the lowest 8 kilometres of the atmosphere · Snow cover and ice extent have decreased · Global average sea level rose between 0.1 and 0.2 metres during the 20th century · The effect of anthropogenic greenhouse gases is detected, despite uncertainties in sulphate aerosol forcing and response (Houghton et al., 2001). Climate change is projected to substantially reduce available water (as reflected by projected runoff) in many of the water-scarce areas of the world (Arnell, 2003). Future prediction shows that Southern Africa will get drier and the region will find it difficult to cope with impacts given the present level of preparedness. Any increase of 1.7oC in mean temperature, the region's precipitation is expected to decrease by 5-20% in all major river basins except for Congo basin which is going to decrease by 10% (Table 1). Basing on individual countries, Zimbabwe will 3
have shortfall precipitation of 19%. The potential evaporation estimated for Zimbabwe ranges between 7.5% and 132% representing the varied climate regimes across the country.
Table 1. Predicted changes in precipitation, potential evaporation and runoff in major river basins
in Southern Africa, expressed as a 30-year monthly mean for the baseline period (1961-1990)
compared to the period 2010-2039, after Chenje et al. (1996).
River Basin
% Change in
% change in potential % change in
precipitation
evaporation
runoff
Congo
+10
+10 to +18
+10 to +15
Zambezi
- 10 to ­20
+10 to +25
-26 to ­40
Rovuma
-10 to +5
+25
-30 to ­40
Limpopo
-5 to ­15
+5 to+ 20
-25 to ­35
Orange
-5 to +5
+4 to +10
-10 to +10
Runoff is harvested, for use, through damming the rivers or pumping straight from the perennial rivers. Construction of large reservoirs with large storage ratio (full supply capacity/mean annual runoff) aims at carrying over the runoff in high rainfall years for use in below average years in order to assure the supply water for urban, industrial, mining and agricultural purposes (Mitchell, 1987). The function of a storage reservoir is to balance out the variation in both supply and demand and thus to provide the consumer with an assured quantity of water for use as and when required. The variations must be balanced not only within each year, but also over sequences of many years, so that water from year of good rainfall is conserved for use during years of drought (Kabell, 1974)
4
Zimbabwe is in a semi arid region of the world, where water is a vital and scarce commodity. Large dams have been planned on major rivers and tributaries. These storage works have been designed to have carry over capacity to balance out the long sequences of years of varying runoff (Kabell, 1984). Management of such works poses some challenges to the water resources managers. Challenges such as excessive evaporation and siltation are major ones, which affect the reservoir capacity. Such storage works have not been designed to generate the minimum cost of water, but were designed to generate the maximum practical yield from the catchment.
According to the IPCC (1990), vulnerabilities in present water uses (i.e where demand exceeds firm yield) and conflicts among current uses are likely to be exacerbated by global warming in most arid and semi-arid regions. Adaptation is an adjustment in natural or human systems in response to actual or expected climatic stimuli or their effects (Kabat et al., 2003). A variety of adaptations can be considered (Table 2).
Table 2. Recommendations for municipal water supplies (modified from table 4:13, IPCC, 2001)
Supply-side options
Demand-side options
Option
Comments
Option
Comments
Increase reservoir Expensive; potential Incentives to use less Possibly limited
capacity
environmental
(e.g. through pricing) opportunity; needs
impacts
institutional framework
Extra supply: more Potential limited
Legally enforceable Potential political impact:
from rivers and impacts
water use standards usually cost-inefficient
5
groundwater Inter-basin transfers Desalination
(e.g for appliances)
Expensive; potential Increase use of grey Potentially expensive
environmental
water
impacts
Expensive (high
Reduce leakage
Potentially expensive to
energy use)
reduce to very low levels
esp. in old systems
Development of
Possibly too technically
ecological sanitation. advanced for wide
application in large urban
systems
The City of Bulawayo, the second largest city in Zimbabwe, lies at an altitude of approximately 1350 m above the sea level. Its water supply is mainly based on surface water accumulated in five reservoirs located in Upper Mzingwane catchment. The five reservoirs have failed to store their expected 4% yields that they were designed for. It has been noticed that there is some precipitation pattern, which is influencing the yield of the reservoirs. The city is famous for its water rationing and restrictions. For the past eighty years the following were the years when restrictions were imposed: 1938-1943, 1947, 1951, 1953, 1968, 1971-1973, 1983 and 1990 while the rationing were imposed in the following years 1949,1984,1987 and 1991 while it stringent water rationing has been imposed this year as the dams' inflows were poor again.
6
It is imperative that the nature of the rainfall pattern and its variability with time be looked at closely so that the runoff to and storage in the dams is appreciated. There is some realization that climate is changing in the city's reservoir catchments. There are times of inadequate rainfall and times of plenty but it seems the rainfall pattern and magnitude is changing. There is a bigger variation between dry and wet years with poor years having a bigger effect than years of plenty. This study has three objectives: the first is to establish whether there is any significant changes in temperature and precipitation in the city council's reservoirs' catchment areas, the second being determining quantitatively the expected changes in inflows and yields into the city's water supply reservoirs due to expected climate variability and finally, to develop adaptation strategies that will help mitigate the adverse impacts due to climate variability on the water supply reservoirs. Two greenhouse emission scenarios will be analysed, one with high emissions and another with low emission, using IPCC SRES emissions scenarios (IPCC, 2000). Their predicted effects will be superimposed on the city's projected future water demand and the deficit will be determined. Two demand scenarios will be considered for the city; one when water demand management is assumed to be fully implemented according to NORPLAN AS 2001 recommendations and one when the year 2000 consumption levels remain unrestricted and the population increases exponentially at a rate of 3% per annum. Thereafter, adaptation strategies will be proposed for the City. METHODS 7
The evaluation of the effect of climate change on the reservoir yields in Upper Mzingwane catchment was carried out for all four reservoir catchments namely, Mzingwane, Ncema (Lower and Upper), Inyankuni and Lower Insiza (Mayfair) dams. Information required for evaluation was rainfall, temperature, demography, water use, evaporation, stream flow, and upstream and downstream water rights for each reservoir. Temperature, rainfall, evaporation and stream flows were considered for stations within the catchments and nearby catchments to assess consistency and possible gap filling for missing data. The data collected covered thirty years from 1970 to 2000. This period was chosen since the city's raw water supply has been operating continuously without additional source since 1976 when the last dam was connected to the system. The upper limit of year 2000 was chosen as most of processed and updated data available was done up to year 2000. Daily rainfall data for several locations within Upper Mzingwane catchment were obtained from the Meteorological Department. The stations eventually utilized were to meet the following basic criteria: location within the respective dam catchments containing sufficiently long observed records falling within the chosen period, and fair spread on the length of the catchment to give good coverage. The stations selected were Esigodini, Mzinyathini, Mzingwane Dam, Mbalabala, Falcon College and Filabusi Police station. Falcon College results were rejected as they only for 15 year period (between 1975 &1990). The Esigodini data was used to represent the pattern for Ncema and Inyankuni dams, Mzinyathini data was used for Mzingwane dam, and Filabusi data was used for Lower Insiza Dam. Daily rainfall data for all the above stations (1930-2000) were converted into 8
annual data, arranged in such a way so as to coincide them to the Zimbabwean hydrological year, which spans from 1 October to 30 September the following year. The 5-year moving average for mean annual rainfall (M.A.R/F) and the coefficient of Variation (Cv) for each station were computed.
A hydrological gauging station was identified for each dam so as to represent the inflows into the dams (Table 3). The stations satisfied the following criteria: Long enough record of flows The station has reliable data The station is located just upstream of the dam and not affected by the dams' through back at highest dam level.
Table 3. Hydrological stations used for data collection
Station Name
Hydrological River Zone
Location
Catchment Year Area (km2) opened
B30
BUZ4
Mzingwane Mzingwane dam 448
1958
U/S
B11
BNC
Ncema
Ncema dam U/S 640
1943
B60&B61 BIK
Inyankuni Inyankuni dam 365
1962
U/S
B89 & BIN2
Insiza
Insiza dam U/S 982
1973
B88
9
Monthly runoff data was collected from Zimbabwe National Water Authority (ZINWA) data and research branch. The annual mean runoff was plotted against the hydrological years for each gauge station. Gaps (missing data) were filled by best fitting line from linear regression from nearby stations. Student t-test was used for testing the trend on the data series. The five-year moving average was calculated for Mean Annual Rainfall (M.A.R) and Coefficient of variation (Cv) of runoff. Both daily minimum and maximum temperatures were collected from the Meteorology Department and converted into annual time steps, from 1 October to the following 30 September. Pan evaporation data was obtained for each dam. The Summer and winter water commitments for all the dams were obtained from the water rights data bank of Zimbabwe National Water Authority. The individual water rights were reviewed noting the point of abstraction relative to the dam, the history, the quantity of the right and the owner of the property served by the water right. Each right was split into the winter and summer commitment. The rights were summed for each dam according to their location (upstream or downstream). The fractional area covering the water-righted areas were also obtained by plan metering the water right map. These parameters are the inputs into the yield model as well. Data on raw water abstraction history, consumption pattern and dams operations was collected from the City of Bulawayo. A site visit to all the dams was undertaken to gather information on the operation, maintenance, conveyance capacities and their water level states during the study 10
period. This data mainly was used for demand versus availability comparison and future demand forecasting. Participatory appraisal workshops were held in the dam catchments. The objective of the workshops was to obtain oral appraisal of the participants' perception about the climate change and variability from the 1920s to present. Each workshop consisted of seven old citizens of the area including the local leadership. Overall, the age range for the participants was fifty-eight years and eighty-three years. The parameters which were discussed were variability in rainfall, temperature, stream flows, reduction in volume of wetlands, land use changes, situations before and after dam constructions, extreme events like flooding, cold winters, drought and hot summers as well as the general water availability for domestic and livestock in their areas. The reservoir yield calculations were carried out using Yield 200 model developed by Mitchell (1989). The model uses an empirically fitted Weibull distribution to establish the probabilities of annual inflows corresponding to the annual yield or draw off from a reservoir. The model was written in QWBASIC form computer programme.(Mitchell, 1989) The monthly inflows are obtained from the gauge station upstream of the dam. This model gives a correct answer within the accuracy of the data supplied, and treats the runoff as a stochastic variable. It is applicable to larger dams (storage ratio equal to 0.5 M.A.R and above) where there is carry over of stored water from year to year. Inputs into the model were: dam catchment (km2), dam upstream mean annual runoff (mm) 11
According to Arnell (2003), the predicted changes in runoff by year 2050 compared to 1990 greenhouse emission levels in Zimbabwe under model Had CM3 scenario A1 are 30% reduction for high greenhouse gas emission and a reduction of 20% for low emission scenario using Other model under scenario A2 .The model predicts that precipitation will fall by 30% under high emission scenario while it will fall by 15 % under low emission scenario. These predictions were done using Had CM3 and CSIRO under scenarios A2 and MKII respectively (Arnell, 2003). These two predictions up to 2030 formed the data for the calculation of predicted mean annual runoff and rainfall, coefficients of variation of both runoff and rainfall. These were then input into the yield model with other input variables assumed constant. These two sets of yield results formed two scenario results while the third scenario was the projection of the current trend. The water demand projection was determined using two scenarios as well. One scenario was unrestricted water demand. In this scenario it is assumed the city supplies the water at a level of demand such the water is supplied as the residents wish, with same levels of losses and without any restrictions. The consumption growth rate has been assumed to be directly proportional growth rate of 3% (from City Council master plan) for this scenario. The other scenario is when the city fully implements the water demand management strategy with losses, estimated to be reduced to 10 000m3 per day. RESULTS AND DISCUSSION Temperature 12
The maximum temperature trend (Fig. 1) shows a warming trend at a rate of 0.410C per decade while minimum shows a warming trend of 0.340C. This implies that winter seasons are becoming warmer and the rate of increase is higher than that of maximum temperatures.
30.0
Temperature (oC)
25.0
Mean Max. Bulawayo
Mean Min. Bulawayo
20.0
Linear (Mean Max. Bulawayo)
Linear (Mean Min. Bulawayo)
15.0
10.0
y = 0.0412x - 55.405 y = 0.0342x - 54.975
5.0
0.0 1970
1975
1980
1985
1990
1995
2000
Fig. 1. Mean Minimum and maximum temperature variation for Bulawayo (1970-2000).
Rainfall
The rainfall trends (Fig. 2) show a decrease in precipitation for all stations. Compared to their respective averages, Esigodini precipitation decreases at a rate of 2.8% per decade while Filabusi decreases at a rate of 2.5% per decade
Stream flow (runoff)
13
There is some noticeable decrease in the amplitude of the runoff peaks into the dams oveer the 30 year period (Fig. 3). After the 1979/80 hydrological year, there was a jump in the pattern and the number of no flow recorded also increased. After that year there was an increase in the duration of number of drought years. As the inflows into a reservoir directly affect its yield, there should be a corresponding jump in decrease of the reservoir yields. The sub catchment also experienced some increase in no flows in its streams. A peak in temperatures coincides with peaks in precipitation. For drought years, there is maximum peak and vice versa during years of plenty precipitation. This is illustrated by taking drought years of 91/92 and 82/83 where both minimum and maximum temperatures had peaks while year 2000 had a drop in both trends. This means that there is a direct relationship between the precipitation pattern and the temperature pattern. The ambient temperature are affected by the ambient humidity, therefore the higher the humidity the cooler it becomes. The runoff peaks also correspond to the precipitation peaks. Generally on all graphs low rainfall in year 91/92 coincides with low flows in the streams. Year 1978/79 was a year of plenty rainfall in all dam catchments, and also there were high recordings of the runoff. This shows that there is a relationship between rainfall and stream flows. 14
Annual Ranfall (mm)
1400 1200 1000 800 600 400 200 0 1931 1934 1937 1940 1943 1946 1949 1952 1955 1958 1961 1964 1967 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 Year Ending September
Bulawayo Esigodini Filabusi Linear (Bulawayo) Linear (Esigodini) Linear (Filabusi)
Fig. 2 Precipitation variation (1930 ­2000) for Bulawayo, Esigodini and Filabusi
15
Fig. 3. Stream flow recorded (1970-200) for the rivers flowing into the dams Reservoir yields and demand Reservoir yield is declining, and predicted to decline further. The rate of decline predicted by the IPCC SRES models is faster than the current trend (Fig. 4). Predicted future yields and the water demands intersect from 2009 onwards. After these intersection points, the demand will be higher than the combined yield and therefore a deficit will start (Table 4). 16
Volume of yield or demand (Mm3/year)
140.00 120.00 100.00 80.00 60.00 40.00 20.00 0.00
Current trend Future high emissions Future low emissions Unrestricted demand WDM- demand Power (Current trend)
2029/30 2024/25 2019/20 2014/15 2009/10 2004/5 1999/2000 1994/95 1989/90 1984/85 1979/80
Hydrological years Fig. 4. Recorded dam yields and water demand (1979-2000), and predicted dam yields and demand (2005-2030).
Table 4. Deficit levels by 2030
Climate
Unrestricted demand scenario
Scenarios
Deficit year
Deficit by 2030 (Mm3/year)
High emissions 2009/10
41.98
Low emissions 2010/11
38.30
Current trend
2014/15
35.37
Water demand management scenario
Deficit year
Deficit by 2030 (Mm3/year)
17/18
27.78
14/15
24.46
20/21
21.17
CONCLUSIONS AND RECOMMENDATIONS
From the results it can be observed that climate is changing in the study area. The river flows are decreasing due to decrease in precipitation at a rate of 2.8% per decade for Esigodini and 2.5% 17
per decade for Filabusi compared to the long-term average of 70 years for Esigodini and 80 years for Filabusi. Temperatures are warming up at a rate of 1.5% per decade and 2.2% per decade for maximum and minimum temperatures respectively. This accelerates evaporation from the open surface reservoirs. Rivers show a strong seasonal variability. Flows are often ephemeral and at times limited to few weeks. There is some relationship between the precipitation variation and runoff variation. In wet season of high rainfall there catchments experience some high inflows. The demand is increasing yet the dam yields show a down ward trend. The city was recommended to implement some water demand management, which was going to reduce the consumption per capita from an average of 150 litres per capita per to 80 litres per capita per day. Using a population growth rate of 3% per annum as stated in the city's master plan the water demand is also growing at this rate. Apparently the city council uses more water from Ncema dams and Mzingwane because of the installed abstraction capacities in these dams. Inyankuni follows the list then lastly Insiza. The city council needs to install pumps at Insiza to fully utilize its allocated yield from the dam. Currently, it is using gravity to lower Ncema where it pumped to the city. In addition to water demand management, the city is recommended to implement some or all of the following developments: increase the reservoir capacity by connecting Mtshabezi dam (yield = 9.8 Mm3/year), extract more water from ground water from Nyamandlovu acquifer (yield = 4.4 Mm3/year), Umguza aquifer (7.3 Mm3/year), abstraction from Zambezi basin (161 Mm3/year) and increased grey water recycling. 18
ACKNOWLEDGEMENTS This paper contains research results from a M.Sc. project by B. Moyo at the University of Zimbabwe, and is a contribution to WaterNet Challenge Program Project 17 "Integrated Water Resource Management for Improved Rural Livelihoods: Managing risk, mitigating drought and improving water productivity in the water scarce Limpopo Basin", funded through the Consultative Group on International Agricultural Research's Challenge Program on Water and Food. The research is also supported by a Research Grant from RELMA-in-ICRAF and a scholarship awarded to B. Moyo by the WREM Trust. The opinions and results presented in this paper are those of the authors and do not necessarily represent the donors or participating institutions. The cooperation of the Bulawayo City Council and the Zimbabwe National Water Authority (Research and Data Department and Catchment Manager: Mzingwane) has been essential and is gratefully acknowledged. REFERENCE: Allen R.B., Basson, M.S., Pegram, A. and Van Rooyen, B. (Eds.). (1994). Probabilistic management of water resources and hydropower system. Department of Water Affairs and Forestry, South Africa. Arnell, N. (2003). Effects of IPCC SRES emissions scenarios on river runoff: a global perspective. Hydrology and Earth System Sciences 7: 619-641. Chenje, M. and Johnson, P. (Eds.). (1996). Water in southern Africa. SADC/IUCN/SARDC, Harare, Zimbabwe. 19
Graciano, P., Matondo, J. and Msibi, K. (2004). Evaluation of the impact of climate change on hydrology and water resources in Swaziland. Physics and Chemistry of the Earth 29: 1181-1191. Houghton, J.T., Ding, Y., Griggs, M., Noguer, M., van der Linden, P.J. and Xiaosu, D. (Eds.) (2001). Climate Change 2001: The Scientific Basis. Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change (IPCC). Cambridge University Press, Cambridge. IPCC. (1990). Potential Impacts of Climate Change: the IPCC Scientific Assessment. Report of Working Group III to the Third Assessment Report of the Intergovernmental Panel on Climate Change. Mateorological Office, Bracknell, United Kingdom. IPCC. (2000). Special Report on Emissions Scenarios. Intergovernmental Panel on Climate Change and Cambridge University Press, Cambridge. IPCC. (2001). Climate Change 2001: Impacts, Adaptation and Vulnerability, Section, Hydrology and Water Resources. Report of working group II to the Third Assessment Report of the Intergovernmental Panel on Climate Change. Kluwer Academic Publishers, Dordrect, Netherlands. Kabat P., Schulze R.E., Hellmuth M.E., Veraart J.A. (eds.). (2003). Coping with impacts of climate variability and climate change in water management: A scoping paper. DWCReport DWCSSO-01, International Secretariat of the Dialogue on Water and Climate, Wageningen, Netherlands. Kabell, T.C. (1974). The philosophy of reservoir yield and operational system fro the maximum utilization of water resources. Department of Water Development, Harare, Zimbabwe. 20
Mitchell, T.B. (1989). Notes on Programme Yield 98. Internal circular, Ministry of Rural Resources and Water Development. Harare, Zimbabwe. 21

File: a-model-for-reservoir-yield-under-climate-change-scenarios-for.pdf
Title: A model for reservoir yield under climate change scenarios for the water stressed City of Bulawayo, Zimbabwe
Author: Suitor Mujax
Published: Wed Jul 18 10:04:11 2007
Pages: 21
File size: 0.22 Mb


Exposure, 128 pages, 0.53 Mb

Old Testament Survey, 3 pages, 0.11 Mb

Education for What, 33 pages, 0.45 Mb
Copyright © 2018 doc.uments.com