- Email: [email protected]
Evaluation of multi-storage hydropower development in the upper Blue Nile River (Ethiopia): regional perspective
Journal of Hydrology: Regional Studies 16 (2018) 1–14
Contents lists available at ScienceDirect
Journal of Hydrology: Regional Studies journal homepage: www.elsevier.com/locate/ejrh
Evaluation of multi-storage hydropower development in the upper Blue Nile River (Ethiopia): regional perspective
T
⁎
Asegdew G. Mulata, , Semu A. Mogesb, Mamaru A. Mogesa a b
Blue Nile Water Institute (BNWI), Bahir Dar Institute of Technology, Faculty of Civil and Water Resource Engineering, Bahir Dar University, Ethiopia School of Civil and Environmental Engineering, AAiT, Addis Ababa University, Ethiopia
AR TI CLE I NF O
AB S T R A CT
Keywords: Eastern Nile Water resource modeling Blue Nile cascades Renaissance Dam
Study region: Eastern Nile River Basin (Ethiopia, Sudan and Egypt). Study focus: This study aims to understand the future water development perspective in the Eastern Nile region by considering the current water use situation and proposed reservoirs in the upper Blue Nile (Abbay) River basin in Ethiopia using a simulation approach. The study was carried out by using a monthly time step and historical ensemble time series data as representative of possible near future scenarios. Series of existing and proposed cascaded water development projects in the upper Blue Nile were considered in the study. New hydrological insights for the region: The results indicated an overall energy gain in the Eastern Nile region increases by 258%. The upstream country Ethiopia can generate as much as 38200 GWh/year of Energy while the energy production in Sudan increases by 39%. The cascaded developments integrated with existing water resources systems have a performance efficiency of above 92%. This study was an indicative analysis of the potential benefit of upstream Nile development without significantly affecting existing development in the Nile Basin. Further scientific analysis in this direction would help the Nile countries to reach a water use agreement.
1. Introduction There is huge hydro power development potential in the Eastern Nile Basin (Waterbury, 2002; Swain, 2002). The greatest development potential, about 58% of the total in the Nile Basin, is located in Ethiopia; this is because of the great differences in altitude (Ethiopia, 2000). There are a number of water resource development projects in Ethiopia specifically in the Abbay River Basin. The projects are at different stage of development, some are in operational stage, some are under construction and the other are in the studying and design phases. Due to the high variability in annual rainfall, conservation of water and irrigated agriculture has been considered as a way to mitigate the effects of drought. According to the master plan study of Abbay river basin (BCEOM, 1999) the total potential for hydropower generation in the basin is about 13,000 MW. This is many times as much as the existing installed capacity. The Abbay basin master plan in Ethiopia and other regional projects, for instance the Joint Multipurpose Projects (JMP), has identified three hydropower dams upstream of the Grand Ethiopian Renaissance Dam (GERD). An evaluation of the impacts and benefits of the Grand Ethiopian Renaissance Dam (GERD) during the impounding phase on downstream structures, especially on High Aswan Dam (HAD), was conducted by considering different flow scenarios (Mulat and Moges, 2014). The study concluded that under normal and wet flow scenarios of the 6 years filling period, GERD has no significant
⁎
Corresponding author. E-mail addresses: [email protected] (A.G. Mulat), [email protected] (S.A. Moges).
https://doi.org/10.1016/j.ejrh.2018.02.006 Received 30 May 2017; Received in revised form 16 February 2018; Accepted 19 February 2018 Available online 07 March 2018 2214-5818/ © 2018 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/BY-NC-ND/4.0/).
Journal of Hydrology: Regional Studies 16 (2018) 1–14
A.G. Mulat et al.
impact on downstream water uses. The agricultural water requirement from HAD which is a concern in Egypt could not be affected under these scenarios. The reduction in the storage volume of HAD never reached its minimum operation level. While the regional energy increases by about 63% due to early start of energy production at GERD. If the worst scenario of 6 years consecutive drought such as in 1980s (18% of the longer term mean) would occur, or if the flow is reduced by 10% or more from the long term mean annual flow, then the planned 6 years filling of GERD is insufficient to fill the reservoir without affecting downstream water uses. The objective of this study was therefore to provide quantitative analysis of water resources management for the Eastern Nile region by considering the current irrigation water use and proposed reservoirs in the Abbay-Blue Nile River basin. The simulation starts after the end of the GERD filling stages. This analysis was then used for assessing the impacts and benefits of the upper cascades on downstream structures and the basin as a whole during its filling and full operation. In addition, the study tested the utility of hydropower development on the Abbay-Blue Nile River by evaluating different development or dam scenarios. 2. The Eastern Nile There are two major basins within the Nile basin. These are the Eastern Nile which is includes Abbay-Blue Nile, Tekeze (Atbara), Baro Akobo (Sobat), and the Nile Equatorial Lake including mainly Lake Victoria basin and Equatorial Lakes. The Blue Nile is the most important tributary of the Nile. It contributes about 60% of the total annual Nile flow (Sutcliffe and Park, 1999). More than 70% of the flow of the Blue Nile is generated by the four months (June–September) wet season rainfall in Ethiopia. Annual average rainfall over the Blue Nile basin is 400 × 109 m3/yr where 62.5% falls on the Ethiopian plateau (Mageed, 1994). As the Blue Nile drops into the lowlands and into southern Sudan, rainfall decreases and evaporation increases, resulting in a significant net loss. Temperatures also increase in variability, and reach substantially higher levels than at Lake Tana. The Sennar region, located in the southeastern part of Sudan, experiences evaporation rates of 2500 mm/yr and receives 500 mm/yr of rainfall with mean daily temperature of 30 °C (Sutcliffe and Park, 1999). The Blue Nile basin (BNB) is characterized by highly rugged topography and considerable variation in altitude. Total area of the basin is 311,437 km2, of which approximately 63% is in Ethiopia and 37% is in Sudan. The elevation of the basin varies greatly from over 4000 m in the headwaters of some tributaries to 700 m at the foot of the plateau. The highest point in Lake Tana is about 1800 m and the river enters Sudan at an elevation of 490 m at the border of the two countries, i.e., with a gradient of approximately 1.5 m/ km. This gives the Blue Nile its unique feature of huge potential energy opportunity to develop hydropower. 3. Existing and proposed water resources developments The physical characteristics for the High Aswan Dam and Lake Nasser/Lake Nubia were extracted from the Power Toolkit provided by ENTRO. The information contains descriptions of the turbine characteristics including explicit relationships between operating head, turbine releases and power generation. Several water resources development projects in the eastern Nile basin are categorised as under operation (existing), under construction and proposed to be developed. Most of the operational reservoirs are in Sudan and Egypt. The under construction Grand Ethiopian Renaissance Dam (GERD) and the other three cascades under design and study phases are located in Ethiopia (Fig. 1 and Table 1). 3.1. Existing developments 3.1.1. Tana Beles project Lake Tana is a natural reservoir that controls the Tana sub-basin flows. The Lake is fed by the Gilgel Abbay, Megech, Ribb and Gumara rivers; and its surface area ranges from 3000 to 3500 km2 depending on season and the amount of rainfall. The lake level has been regulated since the construction of the control “Chara Chara” weir where the lake discharges into the Blue Nile. This controls the flow to the Blue Nile Falls (Tis Abbay) and hydro-power station, which now the hydropower production is left in favor of the Tana Beles plant. The Tana-Beles hydropower project was completed in May 2010 and the project diverts water directly from Lake Tana and is represented as such in the models. The relationship between the diverted flow and intake elevation for the diversion is defined in the Beles Multipurpose Level 1 Design Report from the Ethiopian Electric Power Corporation (EEPCO, 2006). Water diverted through the Tana-Beles project is returned entirely to the headwaters of the Beles which flows into the Blue Nile upstream of the border between Sudan and Ethiopia. 3.1.2. Roseires Dam Roseires Dam was completed in 1966 with an initial capacity of 3.024 × 109 m3 at level 480 m level. The main objective to supply irrigation demands as first priority, and hydropower generation is the second priority. During its lifetime, the reservoir suffered from serious sedimentation which reduced its storage capacity to less than 2.0 × 109 m3. The deep sluices with sill levels of 435.5m.a.s.l are used to pass the main volume of the flood, and to flush the sediment. Due to the large seasonal fluctuation, relatively small storage volume, and high amount of sediment accumulating in Roseires, the operational criteria is specified to draw down the reservoir starting in mid-January and maintain a minimum elevation until the peak flow has passed in September. Therefore, meeting target elevation criteria is the primary guiding principle of the operation of Roseires Dam as it was indicated in the ENTRO power tool kit. 3.1.3. Sennar Dam The Sennar Dam was built in 1925 on the Blue Nile near the town of Sennar, Sudan. The dam is 3025 m long, with a maximum 2
Journal of Hydrology: Regional Studies 16 (2018) 1–14
A.G. Mulat et al.
Fig. 1. Schematics of Eastern Nile River Basin.
height of 40 m. It provides water for irrigation in the Al Jazirah region. Similar to Rosaries Dam, a primary operational objective of this reservoir is to achieve drawdown and refill elevations on specified dates. In addition, the Gezira Managil diversion takes water directly from the reservoir for agriculture purposes. The minimum diversion elevation of the Gezira Managil diversion is 417.0m.a.s.l and therefore demands can be met when the pool elevation is greater than this level. All water in the reservoir above this elevation is considered available for diversion (ENTRO power tool kit). 3.1.4. Jebel Aluia The Jebel Aulia dam built 40 km upstream of Khartoum in 1937 on White Nile River to store water for later use in Egypt. The rapid silt up of this reservoir and the construction high Aswan dam in Egypt in 1965 stopped its function (Shahin, 1985, 2002). 3.1.5. Khashm El Girba Dam Khashm El Girba Dam just downstream of the confluence of the rivers and the Tekeze Dam on the Tekeze River. The primary 3
Journal of Hydrology: Regional Studies 16 (2018) 1–14
A.G. Mulat et al.
Table 1 Proposed and implemented Hydropower plants in the Eastern Nile (installed and target power in megawatt (MW) and tail water in meter above sea level (m.a.s.l). Hydropower plant
Tana Beles Karadobi Bekoabo Upper Mendia Bekoabo Low GERD Rosaries Sennar Merowe Tekeze Jebil Aluia HAD
Hydro power (MW) Installed capacity
Target Power
460 1600 1940 1700 935 6000 425 15 1000 300 28.5 2100
460 933 1329 802 514 1807 200 15 650 180 15 783
Tail water level (m.a.s.l)
Country (location)
Remarks
1356 890 800 640 603 500 437 406 240 967 300 120
Ethiopia Ethiopia Ethiopia Ethiopia Ethiopia Ethiopia Sudan Sudan Sudan Ethiopia Sudan Egypt
In operation Under study Under study Under study Under study Under construction In operation In operation In operation In operation In operation In operation
consumptive use is a direct diversion from the Khashm El Girba Dam (ENTRO power tool kit). 3.1.6. Tekeze Dam Tekeze Dam is a double-curvature arch dam in the northern Ethiopia on the Tekeze River, a Nile tributary that flows through the deepest canyons into the main Nile downstream of Khartoum. At the time of completion, the 188 m high dam is said to be Africa's largest arch dam. The operation of this reservoir is to primarily meet a target power generation of 112 MW, with a maximum power capacity of 300 MW. A maximum pool elevation of 1140 ma.s.l and a minimum operation level of 1096 ma.s.l was used as the range over which power could be generated. 3.1.7. Merowe Dam The Merowe Dam is recently constructed which is operated to meet the primary objective of hydropower generation. The target power generation is 625 MW with a maximum power capacity of 1250 MW. To accomplish this operation, a rule was used that specifies a turbine release to meet the power generation objective, followed by a flood control rule that spills any water in excess of a specified elevation. A maximum pool elevation of 300 ma.s.l and a minimum operation level of 284.90 ma.s.l was used as the range over which power could be generated (ENTRO power tool kit). 3.1.8. High Aswan Dam (HAD) Aswan High Dam Reservoir extends for 500 km along the Nile River and covers an area of 6000 km2, of which northern two-thirds (known as Lake Nasser) is in Egypt and one-third (called Lake Nubia) in Sudan. The dam, completed in 1968 at a distance of 7 km south of Aswan City, is 2325 m long, 111 m high over the original river bed, and 40 m and 980 m wide, respectively, at its crest and bottom. Nile flow is allowed to pass only through the open-cut channel at the eastern side of the dam, where six tunnel inlets provided with steel gates are constructed for discharge control and water supply to power plants. An escape is also provided at the western side of the dam to permit excess water discharge (ENTRO power tool kit). The total capacity of the reservoir (162 × 109 m3) consists of the dead storage of 31.6 × 109 m3 (147 ma.s.l. of Lake water level), the active storage of 90.7 km3 (147–174 m.a.s.l) and the emergency storage for flood protection of 41 × 109 m3 (175–182 m.as.l). The reservoir is surrounded by rocky desert terrain. To the west is the great Sahara Desert, and the Eastern Desert on the east side extends to the Red Sea. The Aswan High Dam contributed greatly to the economic development of Egypt by supplying irrigation water and about 2000 MW hydroelectricity and protecting the lower reaches of the Nile from flood disasters. 3.2. Proposed developments 3.2.1. Karadobi hydropower project The Karadobi hydropower project is located on the Abbay River (Blue Nile). The proposed dam site is located 1.7 km downstream of its confluence with Guder River at about 135 km (air distance) north west of Addis-Ababa. The project was studied at reconnaissance level in the Abbay River Master Plan Project (BCEOM, 1998). The present pre-feasibility study of the Karadobi MultiPurpose Project indicate a rolled concrete gravity dam of maximum structural height of 260 m and length of about 684 m at the crest, with a corresponding installed capacity of 1600 MW. The reservoir area full level has an area of 445 km2 with a capacity of 40200 MCM. The project is an element in the Nile Basin Initiative regarding development of power trade among the countries in the Nile Basin (Norplan-Norconsult International, 2006). 3.2.2. Bekoabo hydropower project Beko-Abo dam height 285 m, crest length 880 m and reservoir live storage 17.5 × 106 m3 capacity and 2000 MW (12100 GWh/yr) (ENTRO power tool kit). 4
Journal of Hydrology: Regional Studies 16 (2018) 1–14
A.G. Mulat et al.
3.2.3. Mendya hydropower project Extensive reservoir impounding is foreseen for the projects which are expected to again feature large dam construction, at sites 220 km from Bekoabo. While Beko abo and Mendia are separate projects the headwaters of the lower, Mendya scheme will extend all the way up to the location of Beko Abo. Mendya dam height 170 m, crest length 1400 m and reservoir live storage 10.3 × 106 m3 M, Capacity 1700 MW and 8220 GWh/yr (ENTRO power tool kit). 3.2.4. Under Construction Grand Ethiopian Renaissance Dam (GERD) The GERD site is the most downstream of the hydropower sites under construction within Ethiopia. The eventual site for the Grand Ethiopian Renaissance Dam was identified by the United States Bureau of Reclamation during a Blue Nile survey conducted between 1956 and 1964 (United States Bureau of Reclamation (USBR), 1964). The Ethiopian Government made the project public on March 31, 2011. The first two generators are expected to become operational after 44 months of construction. It is slated for completion in July 2017. The dam will be a roller-compacted concrete (RCC) gravity-type comprising of two power stations, three spillways and a saddle dam. The dam will be 150 m in height and 1800 m in length (Coyne ET BELLIER and TRACTEBEL Engineering, 2011). 4. Data in the eastern nile 4.1. Stream flow data The source of hydrological data (for the gauged data) is from Ministry of Water of Water, Irrigation and Electricity (MoWIE). There are about 164 hydrometric stations in the Blue Nile river basin in Ethiopia. According to World Meteorological Organization (WMO) recommendations, the number of stations are more than satisfactory, i.e. one station should cover 1800 km2 (BCEOM, 1999). But most of the stations are not working properly, missed data and stopped functioning. As indicated in Mulat and Moges (2014), the monthly input stream flow series is reconstructed from 50 sub-basin flows of Upper Blue Nile (Abbay River). Other stream flows such as that of Tekeze-Setet-Atbara (TSA) and the White Nile at Melakal was collected from Eastern Nile Technical Regional Office (ENTRO) spreadsheet. The later data sets are essential to construct the complete Eastern Nile Model up to HAD in Egypt (Fig. 2). 4.2. Reservoirs data For the standard reservoir, the time series information required includes bottom, crest, spillway, top of dead storage level, and minimum operation pool elevations; minimum and maximum valve releases; precipitation, seepage loss, and evaporation; and flood control and operational rules levels for any water users attached to the reservoir. Simulation starts from their bottom level for upper cascades with sequences of 5 years interval that is intended to offset the impacts of filling the reservoirs on operational reservoirs including GERD and HAD. 4.3. Water users data The most common water use in Eastern Nile River Basin is irrigation water demand. The water system retained for the purpose of this study will assume water requirements for irrigation purposes in Sudan and water requirements associated to High Aswan Dam (HAD). HAD outflows aim to satisfy Egyptian irrigation water demand as priority to energy generation. 55.5 × 109 m3/year are allocated to Egyptian irrigation supply downstream of HAD. This water is also used to generate power within the HAD installed capacity. 18.5 × 109 m3/year is allocated to Sudan for irrigation supply at four different points upstream of HAD in the water system model. As long as HAD water level is equal or greater than the Minimum Operating Level (147 m), water volumes for HAD irrigation are supplied at a monthly time-step according to water demand presented in Table 2. The irrigation water demand in BCM will be changed to m3/s for model input with 10% return flow.
Fig. 2. Seasonal flow distribution of Eastern Nile Basin at selected stations.
5
Journal of Hydrology: Regional Studies 16 (2018) 1–14
A.G. Mulat et al.
Table 2 Water demand input data (m3/sec) used for modeling. Month
Jan
Feb
Mar
Apr
May
June
July
Aug
Sep
Oct
Nov
Dec
HAD Saboloka D/S Sennar U/S Sennar Gizira Dongola Khashim Jubial
983.8 18.85 7.51 67.46 202.34 13.44 45.56 29.97
1296.3 10.75 8.1 66.45 191.27 10.75 51.35 34.55
1628.1 20.37 10.42 44.69 118.13 20.37 51.82 38.74
1662.8 25.08 11.72 34.87 30.78 25.08 55.41 43.26
2137.4 25.92 12.39 35.09 44.49 25.92 46.73 45.03
2800.9 23.26 12.54 77.06 180.89 23.26 45.23 47.85
2835.7 18.85 9.04 97.39 268.64 18.85 42.24 78.52
2442.1 12.85 7.22 93.75 258.55 12.85 39.78 78.66
1736.1 5.34 10.28 132.38 337.76 5.34 27.43 146.06
1388.9 4.63 11.35 148.2 321.77 4.63 16.89 166.39
1284.7 7.17 10.11 141.85 249.6 7.17 19.44 145.5
1219.1 14.81 7.25 82.78 185.58 14.81 40.83 45.75
4.4. Hydropower data All input data which enables for running the simulation are from Ethiopian Electric Power Corporation (EEPCO) Grand Ethiopian Renaissance Dam Project study and ENTRO. For all hydropower plant the engine efficiency is considered to be 95%. Table 1 shows proposed and implemented Hydropower plants in the Eastern Nile (installed and target power in megawatt (MW) and tail water in meter above sea level. 4.5. Evaporation losses Losses/gains were placed at all reservoir locations within the basin to account for reservoirs gains due to precipitation and losses due to Evaporation. These net losses represent the difference between gross precipitation on the reservoir and natural losses due to Evaporation. The input data for net losses from reservoirs are shown in Table 3. 5. River basin modelling and setup The natural river system of the Eastern Nile river basin was schematized and represented with a node-branch structure. A number of nodes and corresponding reaches were established based on the river network configuration (Fig. 2). Branches represent the main river and its main tributaries, and nodes represent major river confluences, reservoirs, and control points for off take. Off take points were selected on the main river and/or tributaries to release water to cover downstream irrigation and hydropower water demands. In the simulation, only two major water demand sectors were considered: irrigation and hydropower. There are eighteen irrigation sites within the basin considered here (Table 2). The major irrigation system in the basin is gravity flow (normally from reservoirs). The schemes were then allocated to the nearby nodes for their water withdrawals, and return flows from established schemes were directed to the immediate downstream nodes. 6. Development scenarios During this study, an investigation has been conducted to assess the future probable changes of hydrological process which may have impacts on the spatial and temporal distribution of water availability in the basin. There are three proposed dams (Karadobi, Bekoabo high/low and Mendya) in the Abbay basin. There are two options for Bekoabo (Bekoabo high and Bekoabo low). It is not possible to combine Bekoabo high and Karadobi because of the minimum available elevation difference between them. The construction and fillings of each cascades is based on five year intervals from the top to downstream, i.e., Karadobi, Bekoabo and Mendya will be constructed and fill in order. To get the best combination of the reservoirs operating simultaneously, the Table 3 Monthly Net evaporation losses (mm/day). Reservoir
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
HAD GERD Khashim Elgibra Jubil Aluia Lake Tana Rosaries Sennar Karadobi Tekeze Merowe Beko abo Mendia
4.41 6.48 5.80 6.10 4.47 5.80 5.99 6.43 4.84 6.35 5.57 6.22
5.95 6.61 6.80 8.11 4.37 6.37 6.16 6.12 5.45 7.66 5.37 6.34
8.61 5.13 8.07 7.50 3.30 7.31 7.55 4.91 5.32 9.52 5.13 4.75
10.64 4.94 8.27 7.40 2.57 7.37 7.37 4.52 5.60 11.03 4.20 4.52
12.25 3.76 8.60 7.40 2.63 6.10 6.30 3.51 5.97 11.77 2.10 3.39
13.51 −0.25 7.93 7.50 −1.17 2.09 2.09 −0.69 3.80 11.33 −2.17 −0.69
13.3 −6.06 4.20 3.80 −5.13 −0.90 −0.93 −7.30 −2.65 10.13 −4.60 −7.07
13.16 −3.93 2.70 3.70 −6.10 −0.84 −0.87 −5.02 −3.55 9.77 −1.53 −4.85
11.55 1.04 5.40 4.50 −1.23 0.65 0.65 0.57 2.90 10.57 0.27 0.57
8.96 4.11 6.10 4.90 3.77 4.03 4.16 3.92 4.45 9.90 2.20 3.79
5.56 5.16 6.27 5.80 4.73 5.23 5.23 4.93 4.70 7.70 3.90 4.93
5.11 6.25 6.10 5.68 4.90 5.58 5.58 6.00 4.60 6.63 4.83 6.00
6
Journal of Hydrology: Regional Studies 16 (2018) 1–14
A.G. Mulat et al.
Table 4 Hydropower dams development scenarios in the Eastern Nile basin with their implementation intervals in years. Starting Year
Before 2020
Scenario Scenario Scenario Scenario Scenario Scenario Scenario Scenario
Currently operating reservoirs in the Eastern Nile Basin
1 2 3 4 5 6 7 8
2020
2025
GERD GERD GERD GERD GERD GERD GERD
Karadobi Karadobi Karadobi Karadobi Bekoabo high Bekoabo high
2030
2035
Mendya Bekoabo low Bekoabo low
Mendya
Mendya
following scenarios have been considered and summarized in Table 4.
• Baseline scenario (Scenario 1): the reference scenario or current condition. • GERD only (Scenario 2): the combination of GERD and currently operating reservoirs. • Karadobi and GERD (Scenario 3): the second scenario with Karadobi. • Karadobi, GERD, and Mendya (Scenario 4): this one is the combination of Karadobi, GERD and Mendya with the currently scenario. • Karadobi, GERD and Bekoabo low (Scenario 5): this is scenario 4, but Bekoabo low instead of Mendya. • Karadobi, GERD, Mendya and Bekoabo low (Scenario 6): here four reservoirs are considered with current condition. • Bekoabo high and GERD (Scenario 7): this similar with scenario 3 but Bekoabo high instead of Karadobi. • Bekoabo high, GERD and Mendya (Scenario 8): Mendya with scenario 7 7. Results and discussion 7.1. Inflow to reservoirs The flow regime of the downstream dam is governed by the operations of the upstream hydroelectric power schemes and runoff from the catchments (the incremental flow) between the two dams. Reservoirs in the upstream have effects on the downstream reservoirs inflow. During filling period of new constructed reservoir, the amount of water released will be reduced and more regulated. For instance, there is slight inflow reduction to GERD during the impounding of upper cascades in the Abbay-Blue Nile River. During full operation phase, the reservoirs in the downstream will gain maximum flow in the dry months and the flood will decrease in wet months as a result of regulating effects of upstream cascades. Fig. 3 demonstrates mean monthly inflow to Rosaries dam for different dam scenarios. Releases from Karadobi, Bekoabo and Mendya dam constitute the major inflow into GERD dam, This means that the more the release from upper reservoir the faster the downstream reservoir fill up and excess will be discharged. The operation of GERD and other upstream dams dictate operation pattern in downstream dam, excess releases at GERD will force the reservoir manager at Rosaries dam to release so as to accommodate releases from GERD and thereby causing flooding at the downstream area. Because of its storage capacity, GERD has significant effect on the inflow of the Sudanese reservoirs. Rosaries and Sennar dam receive more regulated flow due to GERD implementation. The monthly reservoirs inflow coefficient of variance is maximum in baseline scenario as compared to the other scenarios. The mean monthly inflow to downstream reservoirs will also be improved, for instance the mean monthly coefficient of variance for Rosaries is 1.2 and 0.2 for baseline and other scenarios respectively. For Sennar it is 1.3 for baseline scenario and 0.16 for the other scenarios. The same is true for HAD inflow, the coefficient of variance of inflow to HAD decreases from 1 (for baseline scenario) to 0.5 (for other scenarios).
Fig. 3. Mean monthly inflow to Rosaries (106 m3/month).
7
Journal of Hydrology: Regional Studies 16 (2018) 1–14
A.G. Mulat et al.
There are changes in other flow statistics. The mean, maximum and minimum monthly flow to downstream reservoirs will be changed due to the implementation of upper cascades. For instance, the maximum inflow to HAD in dry month (May) is 3.4 × 109 m3 for base line scenario but for the scenarios 1–8, it is 4.6 × 109 m3 that is 35% incremental. But for the wet month (August) it is 32.6 × 109 m3 and 19 × 109 m3 for baseline and for scenario 1–8 respectively that is 42% reduction. The inflow to HAD in the dry months (March, April, May and June) of selected dry year periods (from 1982 to 1988) will increase by 133% (from 1272 × 106 m3 to 2967 × 106 m3) due to the implementation of upper cascades. This shows water availability at HAD could be safe even drought occurs. For the selected wettest years (from 1996 to 2001) and wet months (July, August, September and October), flow to Khartoum will decrease by 100% (from 10113 × 106 m3 to 5084 × 106 m3). That is, if flow which cause flood occurs in the highlands of Ethiopia, it will be controlled by the upstream reservoirs and will be released in regulated manner. The impact of the upstream cascades on the mean annual flow to reservoirs in the downstream is insignificant. For instance the difference between maximum (65294 × 106 m3, with current scenario) and the minimum (62438 × 106 m3, for the scenario 4) mean annual inflow to HAD is 4.4%. For Merowe the difference between maximum and minimum mean annual inflow is 3.3%. 7.2. Reservoir water level Filling period for Karadobi is assumed to be 5 years and after that it start operating with full capacity. Bekoabo low will start filling when Karadobi reaches its full level and then Mendya filling will start. The construction and filling of Mendya lags by 5 years of Karadobi or Bekoabo high filling, but for the three reservoirs combination (Karadobi + Bekoabo low + Mendya), the filling will be after 10 years of Karadobi filling. There is no reservoir upstream of Karadobi which affects its operation. There are slight impacts of upstream cascades filling on the GERD pool level. There could be up to 7 m monthly water level reduction of GERD during filling periods of the upper cascades. After filling of all upstream reservoirs, the maximum GERD mean monthly reservoir water level difference, which is between scenario 2 and scenario 4, is 3 m. Fig. 4 show Sudanese reservoir (Rosaries) water levels for different scenarios. As shown in the figures, all Sudanese reservoirs water level will increase due to upper cascades implementation. Their pool level never be below baseline scenario’s level in both filling and operation phases. The maximum pool level for all Sudanese reservoirs is scenario 2. Though it is not below baseline scenario level, there could be up to 11 m water level reduction of Rosaries during the filling of Karadobi, Media and Bekoabo (scenario 4) as compared to scenario 2. Sennar water level will not be affected by reservoirs in Abbay-Blue Nile river basin in both filling and operation phases. Because of its storage capacity there is water level reduction in the dry months for the baseline scenario but this could be mitigated by implementing reservoirs in the upstream. There is slight impact of upper cascades filling on Merowe reservoir as compared to scenario1 and scenario2. There could be up to 5 m water level reduction of Merowe reservoir due to upper cascades filling as compared to scenario 2, but it will never reach below water level of baseline scenario. At the end of impounding of all reservoirs HAD reservoir water level decreases up to 14 m. The maximum reservoir water level reduction will be if all proposed reservoirs in Abbay River are implemented. HAD reservoir water level will have less fluctuation when there are more reservoirs in upstream. Reservoir water level range is 36 m in the simulation period of baseline scenario, but for the other scenarios it is around 31meters. 7.3. Energy production This section discusses the production of energy in the Eastern Nile region by considering scenarios shown in Table 1. Regional and each Eastern Nile country’s’ hydropower production are discussed in the following sub-sections. 7.4. Energy production in abbay-Blue nile river (Ethiopia) The energy production impact of upstream cascade development in the Abbay-Blue Nile River basin is similar with their impacts on the reservoir water level. The simulation results of energy production for the different scenarios are provided in Tables 5 and 6. As
Fig. 4. Mean monthly Rosaries water level during full operation phase.
8
Journal of Hydrology: Regional Studies 16 (2018) 1–14
A.G. Mulat et al.
Table 5 Mean annual Energy generation (GWh/yr) of each reservoir for different scenarios during filling and operation phases of upper reservoirs. Mean Annual Energy (GWh/yr) Scenarios
GERD
Roseires
Sennar
Merowe
Had
Scenario Scenario Scenario Scenario Scenario Scenario Scenario Scenario
16256 15717 15269 15699 15376 15848 15394
2200 3485 3774 3426 3476 3425 3474 3465
131 131 131 131 131 131 131 131
5946 7790 7698 7587 7691 7587 7698 7615
7462 6982 6628 6411 6608 6508 6736 6493
1 2 3 4 5 6 7 8
Mendia
Bekoabo Low/High
6147 3498 3498 7699 7699
4316 5860
Karadobi
EN
Diff (%)
7132 7132 7132 7132
15739 34644 41080 46103 44235 47973 41586 46657
120 161 193 181 205 164 196
Table 6 Country wise and Basin wide (Eastern Nile) energy generation full operation phases of all reservoirs in the Eastern Nile. Countries
Ethiopia GWh/yr
Scenarios Scenario Scenario Scenario Scenario Scenario Scenario Scenario Scenario
1 2 3 4 5 6 7 8
HAD (Egypt) Diff GWh/yr
16256 24876 33264 29594 38200 28133 36525
8620 17008 13338 21944 11877 20269
GWh/yr %
53 105 82 135 73 125
7462 6982 6893 6791 6891 6816 6926 6820
Sudan Diff
GWh/yr
GWh/yr
%
−482 −571 −674 −573 −648 −539 −644
−6 −8 −9 −8 −9 −7 −9
8277 11407 11428 11382 11429 11380 11450 11396
Eastern Nile Diff (%)
GWh/yr
GWh/yr
%
3130 3151 3104 3152 3103 3172 3119
38 38 38 38 37 38 38
15742 34645 43197 51436 47914 56396 46508 54742
Diff GWh/yr
%
18904 27456 35695 32173 40654 30767 39000
120 174 227 204 258 195 248
shown in Fig. 4 and in Table 5, the energy generation at GERD will be slightly affected due to the implementation of upper cascades. Because of their maximum capacity, Karadobi and Bekoabo high have significant impacts on the energy production of GERD during their filling phase. During Karadobi filling period, the energy production of GERD will reach minimum level (9163 GWh/yr) which is 68% of scenario 2 (Fig. 5). Bekoabo high filling will reduce the GERD energy to 10883 GWh/yr which is 2546 GWh/yr less than that of scenario 2. There is slight fluctuation of energy production when only GERD is in operation (scenario 2). For instance, the coefficient of variance of the energy within the simulation period for GERD is 0.4 for scenario 2, while it is 0.35 for the other scenarios. At full development level, maximum GERD energy (16256 GWh/yr) could be generated when there is no cascade upstream of GERD. When there are more reservoirs in operation above GERD, there could be slight energy reduction of GERD (Table 5). The different between the maximum and the minimum is 463 GWH that is 3% reduction from scenario 2. The energy production at Karadobi will not be changed because there are no reservoirs above it. The mean annual energy at Karadobi is 8864 GWh/yr during its full operation phase. There are three scenarios for Mendya that is Mendya with Bekoabo high, Mendya with Karadobi and Mendya with Karadobi and Bekoabo low. Fig. 6 shows energy production of Mendya during filling and operation phases for the combination of Mendya with; Karadobi, Karadobi + Bekoabo low and Bekoabo high. The impounding period is assumed to 5 years and maximum energy production will be after the filling of the reservoir to its full reservoir level (808 m.a.s.l). Fig. 7 shows energy production of Bekoabo low and Bekoabo high during filling and full operation phases respectively. There are
Fig. 5. Annual energy generation of GERD during filling phases of all reservoirs in Abbay river basin (GWh/yr).
9
Journal of Hydrology: Regional Studies 16 (2018) 1–14
A.G. Mulat et al.
Fig. 6. Mendya energy production during impounding and operation phases when combined with different combination of the upper cascades.
Fig. 7. Bekoabo (Bekoabo low and Bekoabo high) energy production during impounding and operation phases.
two options for Bekoabo that is Bekoabo high and Bekoabo low with Karadobi. As the name indicates, there is difference between Bekoabo high and Bekoabo low in storage capacity and normal water level. Therefore, there is big mean annual energy production difference between Bekoabo High and Bekoabo (Table 5). The construction and filling of Bekoabo low is after the implementation of Karadobi. Fig. 8 demonstrates the cumulative (total) energy production in Abbay-Blue Nile River (Ethiopia) after all reservoirs are completed. As shown in Figures and Tables 5 and 6, the overall energy production in Ethiopia will increase significantly due to the construction of the upper cascades. The maximum mean annual energy production (38200 GW/yr) is scenario 6 which combines four reservoirs (GERD, Karadobi, Bekoabo low and Mendya) in Abbay-Blue Nile River basin. The next maximum mean annual energy production (36525 GWyr) is scenario 8 which combines GERD, Bekoabo high and Mendya.
7.5. Energy production in Sudan Energy production in Sudan is from Rosaries, Sennar and Merowe dams. Sudan would also benefit from the upstream infrastructures: reduced spillage makes more water available for hydropower generation. Fig. 9 present the monthly average energy generated by the major hydropower plants (Rosaries) for all development scenarios. There is slight reduction of energy at Roseires during the filling periods of cascades in Abbay-Blue Nile River as compared to scenario 2. But energy generation of Rosaries is always more than that of scenario 1. Even though there is a slight effect of upstream cascades filling on Merowe energy production as compared to the scenario 2, there is significant incremental of energy as compared to scenario 1 For instance there is around 900 GWh more energy production during Bekoabo high filling as compared to scenario 1. Generally there are positive impacts of upper cascade development for energy production in Sudan. Due to their less storage capacity, the energy production in the dry months is less but the upper cascade development can mitigate this problem. Fig. 10 shows the cumulative energy production of reservoirs located in Sudan during filling and full operation phases. Due to the upper cascades, there will be flow regulation and reservoirs in Sudan will get uniform flows and the energy production will increase due to the upper cascades. The minimum energy production in Sudanese reservoirs is in the current situation (when there are no
Fig. 8. Total annual energy generation in upper Abbay River (Ethiopia) during impounding and operation phases of the upper cascades (GWh/yr).
10
Journal of Hydrology: Regional Studies 16 (2018) 1–14
A.G. Mulat et al.
Fig. 9. Mean monthly energy generation at Rosaries during full operation of all reservoirs (GWh/month).
Fig. 10. Annual energy generation in Sudan during full operation of the upper cascades (GWh/yr).
cascades in operation in the upper Blue Nile). But if there is a reservoir in upper Blue Nile, cumulative energy production in Sudan will increase by at least 37% (Table 6). 7.6. Energy generation at HAD Energy production at HAD will decrease due to the upper cascades. The production of hydroelectricity from HAD in Egypt would be reduced on average by 8.43% as compared to baseline scenario. Fig. 11 depicts the energy production of HAD during filling and operation phases of upper cascades. Karadobi or Bekoabo high filling will reduce the HAD mean annual energy to 5039 GWh/yr which is 33% less than the baseline scenario. The overall impacts of the upper cascades filling are insignificant (Table 6). Maximum energy reduction at HAD will be 14% of the baseline scenario. During full operation phases, the maximum energy (7462 GWh/yr) is the baseline situation (Table 6). Maximum energy reduction as compared to the baseline scenario is 9% (Table 6) during the full operation phases. 7.7. Energy production in Eastern Nile At the basin scale, the annual production of hydroelectricity is boosted by at least 34644 GWh/yr amongst which 3130 GWh/yr is from Sudanese reservoir due to the regulation capacity of Ethiopian reservoirs. As shown in Fig. 12, Tables 5 and 6, the energy production in the whole Eastern Nile system will increase significantly when all reservoirs are at their full operation level. During their filling phase there is slight reduction of energy generation as compared to scenario 2 (Fig. 11). The column ‘Diff’ in Tables 5 and 6 indicates the difference in energy production of Eastern Nile between the current situation (without GERD) and other scenarios. During the full operation phase, Eastern Nile energy will increase at least by 120% and it will reach up to 258% incremental. If each reservoir is considered as shown in Table 5, energy productions in Sudan will increase due to the upper cascades but in HAD energy production will slightly decrease because of head reduction. The maximum energy in the Eastern Nile could be 56396 GWh/yr for scenario 6 which combines GERD, Mendya, Bekoabo low, and Karadobi with the baseline scenario.
Fig. 11. Annual energy generation at HAD during filling and operation of the upper cascades (GWh/yr).
11
Journal of Hydrology: Regional Studies 16 (2018) 1–14
A.G. Mulat et al.
Fig. 12. Energy generation of eastern Nile with upper cascades in Abbay River during filling and operation phases of upper reservoirs (GWh/yr).
7.8. Water loss from reservoirs Water loss is due to evaporation from the water surfaces of the reservoirs. As the water level increases there will be more free water surface area and the evaporation loss will be increased. The column ‘Diff’ in Table 7 indicates the difference in water loss from the reservoirs of Eastern Nile between the current situation (scenario 1) and other scenarios. Evaporation loss (mm/day) in the upstream part of the basin is less as compared to downstream. Meanwhile as a dam is constructed upstream, there will be additional free water surface area and more evaporation loss. The loss in Ethiopia will increase from nil up to 2024 × 106 m3 because of the reservoirs. There is more regulated and constant flow of water to the Sudanese reservoirs and the reservoir level is maintained at higher level regularly in an increase of water surface area. Due to that the loss in Sudanese reservoirs increases by at least 35% during full development stages of all upper reservoirs. Water loss at HAD slight decrease as the water level also slightly decreases with addition of reservoirs in the upper basin. The maximum HAD loss is 11673 × 106 m3 is in baseline scenario and the minimum loss is 10190 × 106 m3 is in scenario 2.
7.9. Irrigation water demand deficit Irrigated agriculture in Sudan will benefit from upstream storage in Ethiopia since the Sudanese annual withdrawals are lower in all scenarios. Those allocation decisions illustrate that once water has passed through the Ethiopian hydropower plants, irrigated agriculture starts competing with hydropower generation and irrigation withdrawals become more economically sound. Table 8 illustrates the benefits and impacts of storage in Ethiopia on the irrigated agriculture sector. The regulation capacity of the reservoirs located in Ethiopia would increase irrigation water availability in Sudan. No significant impacts are observed for Egypt. The filling impacts on water demand is insignificant because there are only two events of irrigation water demand deficit at HAD due to the filling of the reservoirs. The irrigation water demand deficit has been considered in the downstream projects, especially in HAD, in all scenarios during full operation phase. The difference between scenarios is the extent of demand deficit within the simulation period. The performance analysis is made by using equation shown below. As shown in Table 8, the water demand deficit is not significant. That is the performance is nearly 100% in irrigation projects in Sudan and more than 92% for HAD irrigation water demand.
Event − based reliability =
Total number of non − failure months Total number of months
Table 7 Mean annual losses (106 m3) from each reservoir for deferent scenarios during filling and operation phases of all reservoirs. Mean Annual Loss (x106 m3) Scenarios
GERD
Rosaries
Sennar
Merowe
HAD
Scenario Scenario Scenario Scenario Scenario Scenario Scenario Scenario
1459 1446 1416 1445 1435 1460 1426
241 419 424 422 424 416 428 428
198 200 210 210 210 210 210 210
1647 2251 2224 2191 2226 2196 2221 2207
11673 10190 11153 10714 11099 10988 11490 10929
1 2 3 4 5 6 7 8
Mendya
Bekoabo Low
363 21 21 174 174
247 360
12
Karadobi
EN
Diff (%)
Eth loss
Sudan loss
245 245 245 245
13759 14519 15702 15561 15670 15758 15983 15734
0 6 14 13 14 15 16 14
1459 1691 2024 1711 1948 1634 1960
2086 2870 2858 2823 2860 2822 2859 2845
Journal of Hydrology: Regional Studies 16 (2018) 1–14
A.G. Mulat et al.
Table 8 Water demand deficit (number of months with water demand deficit, amount of deficit water in million cubic meters (×106 m3) and Nash-Suttcliffe coefficient (NS) for full operation phase. Scenarios
Scenario Scenario Scenario Scenario Scenario Scenario Scenario Scenario
1 2 3 4 5 6 7 8
HAD
Gizira 6
3
No of Month
10 m
25 34 35 36 40 42 34 40
93368 113245 113951 114876 126205 125497 111750 123486
NS
No of Month
106 m3
NS
0.95 0.93 0.93 0.93 0.92 0.92 0.93 0.92
2 1 0 0 0 0 0 0
10.9 72.4 0 0 0 0 0 0
1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00
8. Conclusion and recommendations The aim of this study is to understand the future water development perspective in the Eastern Nile region by considering the current irrigation water use and proposed reservoirs in the upper Blue Nile (Abbay) River basin in Ethiopia using a river basin simulation approach. Simulations was carried out on a monthly time step and using historical ensemble time series data as representative for possible near future scenarios during the implementation of all planned reservoirs in the upper Blue Nile. The scenarios were evaluated on economic benefits for the whole eastern Nile, basin wide, and at the project site. The economic benefits considered are energy production and irrigation water demand satisfaction. From all the scenarios Ethiopia could have maximum energy (38200 GWh/yr) could be achieved when Karadobi, Bekoabo low, Mendia and GERD are in operation simultaneously. The next maximum energy (36525 GWh/yr) could be when Bekoabo high, Mendia and GERD are combined. The maximum Eastern Nile energy (56396 and 54742 GWh/yr) could be achieved in these scenarios. Energy in the eastern Nile will increase at least by 126% for GERD only case and could increase by 258% for Karadobi, Bekoabo low, Mendia and GERD combination. The energy production in Sudan will increase by 39% due to the construction of a single reservoir in the Abbay River. There could be small energy reduction (up to 9%) at HAD due to reduced reservoir water level. There are also slight water loss changes in the whole Eastern Nile due to the implementation of reservoirs. The minimum loss is in the current situation. Development of Abbay River will increase the free water surface area and the evaporation in the basin. The construction of new reservoirs in the Blue Nile results in two competing effects with respect to evaporation. The first effect is increased evaporation resulting from the proposed reservoirs while the other is reduced evaporation from HAD as a result of decreased inflows to the reservoir and reduction of the free water area. The maximum incremental water loss in the Eastern Nile could be 10% as compared to the baseline scenario. The amount of water flowing to the reservoirs will be more regulated and more constant due to the upstream cascades. Reservoirs in Sudan are small in storage capacity as compared to the other reservoirs in Ethiopia and Egypt. Due to that they cannot store water up to their full level in the dry seasons and there are fluctuations in energy production and reservoir water level. But the upper cascades in Abbay River would enable them to get regulated and more water flow in the dry months and to maintain their water level and energy production constant and at higher level. There will be reduction of the risk due to hydrological variability with sequences of dry and wet years. Indeed, at the present time, excessive water may be spilled during wet period at downstream reservoirs because existing reservoirs are already filled and, as a result, water demand failure can occur during a dry period. With GERD, the total storage capacity along the Nile River will significantly increase in the long term. Basin Water Management will be easier to optimize with higher storage capacity and upstream regulation capacities The water demand deficit is insignificant in all scenarios, that is, the performance is more than 92% in all cases. In Gizira irrigation projects there are no months with water demand deficit due to the upper cascades while there are 2 months in the current scenario. As recommendation, cascades in Abbay River basin are not studied at their final stage and their construction period is not specified. So, further evaluations may be necessary after their design and construction time is specified. Further detailed study could also be necessary by considering all proposed and implemented small and medium irrigation and water abstraction projects in Abbay River basin. Research work for the way to improve water use efficiency is crucial. Managing water resources in a way that brings benefits to all and establishing a regional operation plan that integrates a hydrological forecast model and dynamic operation rules are necessary. It is better to recommend managing water resources as a single system by establishing more effective institutional arrangements than those currently existing. It is to be hoped that such arrangements will be devised through the protracted negotiations currently under way.
Conflict of interest None. 13
Journal of Hydrology: Regional Studies 16 (2018) 1–14
A.G. Mulat et al.
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.ejrh.2018.02. 006. References BCEOM, 1998. Abbay River Basin Integrated Development Master Plan Project, Phase 2, Volume III- Water Resources, Part 1, 2- Climatology, Hydrology. Ministry of Water resources, The Federal Democratic Republic of Ethiopia, Addis Ababa. BCEOM, 1999. Abbay River Basin Integrated Master Plan, Main Report. Ministry of Water Resources, Addis Ababa. Coyne ET BELLIER and TRACTEBEL Engineering, 2011. Grand Ethiopian Renaissance Dam Project: Hydraulic and Reservoir Simulation Studies. Ethiopian Electric Power Corporation (EEPCO), 2006. Beles Multipurpose Level 1 Design Report. Ethiopian Electric Power Corporation. Ethiopia, 2000. Country paper of the federal democratic republic of Ethiopia, In: proceedings, comprehensive water resources development of the nile basin: priorities for the new century. In: VIII Nile 2002 Conference. June 26–30, Addis Ababa. pp. 54–56. Mageed, Y.A., 1994. The Nile Basin: lessons from the past. In: Biswas, A.K. (Ed.), International Waters of the Middle East, from Euphrates-tigris to Nile, Water Resources Management Series. Oxford University Press. Mulat, A.G., Moges, S.A., 2014. Assessment of the impact of the Grand Ethiopian Renaissance Dam on the performance of the high Aswan Dam. J. Water Resour. Prot. 6, 583–598. Norplan-Norconsult International, 2006. Karadobi Multipurpose Project Pre-Feasibility Study. Ministry of Water Resources, Federal Democratic Republic of Ethiopia. Shahin, M., 1985. Hydrology of the Nile Basin. Elsevier, Amsterdam, pp. 575. Shahin, M., 2002. Hydrology and Water Resources of Africa, Water Science and Technology Library. Kluwer Academic Publishers, Dordrecht/Boston/, London. Sutcliffe, Park, 1999. The Hydrology of the Nile, IAHS Special Publication No. 5. IAHS Press, Institute of Hydrology, Wallingford, Oxford shire OX10 8BB, UK. Swain, A., 2002. The Nile River Basin initiative: too many cooks, too little broth. SAIS Rev. 22 (2), 293–308. United States Bureau of Reclamation (USBR), 1964. Land and Water Resources of the Blue Nile Basin, ” Main Report. United States Department of Interior Bureau of Reclamation, Washington, DC, USA. Waterbury, J., 2002. The Nile Basin: National Determinants of Collective Action. Yale Univ. Press, New Haven.
14
Contents lists available at ScienceDirect
Journal of Hydrology: Regional Studies journal homepage: www.elsevier.com/locate/ejrh
Evaluation of multi-storage hydropower development in the upper Blue Nile River (Ethiopia): regional perspective
T
⁎
Asegdew G. Mulata, , Semu A. Mogesb, Mamaru A. Mogesa a b
Blue Nile Water Institute (BNWI), Bahir Dar Institute of Technology, Faculty of Civil and Water Resource Engineering, Bahir Dar University, Ethiopia School of Civil and Environmental Engineering, AAiT, Addis Ababa University, Ethiopia
AR TI CLE I NF O
AB S T R A CT
Keywords: Eastern Nile Water resource modeling Blue Nile cascades Renaissance Dam
Study region: Eastern Nile River Basin (Ethiopia, Sudan and Egypt). Study focus: This study aims to understand the future water development perspective in the Eastern Nile region by considering the current water use situation and proposed reservoirs in the upper Blue Nile (Abbay) River basin in Ethiopia using a simulation approach. The study was carried out by using a monthly time step and historical ensemble time series data as representative of possible near future scenarios. Series of existing and proposed cascaded water development projects in the upper Blue Nile were considered in the study. New hydrological insights for the region: The results indicated an overall energy gain in the Eastern Nile region increases by 258%. The upstream country Ethiopia can generate as much as 38200 GWh/year of Energy while the energy production in Sudan increases by 39%. The cascaded developments integrated with existing water resources systems have a performance efficiency of above 92%. This study was an indicative analysis of the potential benefit of upstream Nile development without significantly affecting existing development in the Nile Basin. Further scientific analysis in this direction would help the Nile countries to reach a water use agreement.
1. Introduction There is huge hydro power development potential in the Eastern Nile Basin (Waterbury, 2002; Swain, 2002). The greatest development potential, about 58% of the total in the Nile Basin, is located in Ethiopia; this is because of the great differences in altitude (Ethiopia, 2000). There are a number of water resource development projects in Ethiopia specifically in the Abbay River Basin. The projects are at different stage of development, some are in operational stage, some are under construction and the other are in the studying and design phases. Due to the high variability in annual rainfall, conservation of water and irrigated agriculture has been considered as a way to mitigate the effects of drought. According to the master plan study of Abbay river basin (BCEOM, 1999) the total potential for hydropower generation in the basin is about 13,000 MW. This is many times as much as the existing installed capacity. The Abbay basin master plan in Ethiopia and other regional projects, for instance the Joint Multipurpose Projects (JMP), has identified three hydropower dams upstream of the Grand Ethiopian Renaissance Dam (GERD). An evaluation of the impacts and benefits of the Grand Ethiopian Renaissance Dam (GERD) during the impounding phase on downstream structures, especially on High Aswan Dam (HAD), was conducted by considering different flow scenarios (Mulat and Moges, 2014). The study concluded that under normal and wet flow scenarios of the 6 years filling period, GERD has no significant
⁎
Corresponding author. E-mail addresses: [email protected] (A.G. Mulat), [email protected] (S.A. Moges).
https://doi.org/10.1016/j.ejrh.2018.02.006 Received 30 May 2017; Received in revised form 16 February 2018; Accepted 19 February 2018 Available online 07 March 2018 2214-5818/ © 2018 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/BY-NC-ND/4.0/).
Journal of Hydrology: Regional Studies 16 (2018) 1–14
A.G. Mulat et al.
impact on downstream water uses. The agricultural water requirement from HAD which is a concern in Egypt could not be affected under these scenarios. The reduction in the storage volume of HAD never reached its minimum operation level. While the regional energy increases by about 63% due to early start of energy production at GERD. If the worst scenario of 6 years consecutive drought such as in 1980s (18% of the longer term mean) would occur, or if the flow is reduced by 10% or more from the long term mean annual flow, then the planned 6 years filling of GERD is insufficient to fill the reservoir without affecting downstream water uses. The objective of this study was therefore to provide quantitative analysis of water resources management for the Eastern Nile region by considering the current irrigation water use and proposed reservoirs in the Abbay-Blue Nile River basin. The simulation starts after the end of the GERD filling stages. This analysis was then used for assessing the impacts and benefits of the upper cascades on downstream structures and the basin as a whole during its filling and full operation. In addition, the study tested the utility of hydropower development on the Abbay-Blue Nile River by evaluating different development or dam scenarios. 2. The Eastern Nile There are two major basins within the Nile basin. These are the Eastern Nile which is includes Abbay-Blue Nile, Tekeze (Atbara), Baro Akobo (Sobat), and the Nile Equatorial Lake including mainly Lake Victoria basin and Equatorial Lakes. The Blue Nile is the most important tributary of the Nile. It contributes about 60% of the total annual Nile flow (Sutcliffe and Park, 1999). More than 70% of the flow of the Blue Nile is generated by the four months (June–September) wet season rainfall in Ethiopia. Annual average rainfall over the Blue Nile basin is 400 × 109 m3/yr where 62.5% falls on the Ethiopian plateau (Mageed, 1994). As the Blue Nile drops into the lowlands and into southern Sudan, rainfall decreases and evaporation increases, resulting in a significant net loss. Temperatures also increase in variability, and reach substantially higher levels than at Lake Tana. The Sennar region, located in the southeastern part of Sudan, experiences evaporation rates of 2500 mm/yr and receives 500 mm/yr of rainfall with mean daily temperature of 30 °C (Sutcliffe and Park, 1999). The Blue Nile basin (BNB) is characterized by highly rugged topography and considerable variation in altitude. Total area of the basin is 311,437 km2, of which approximately 63% is in Ethiopia and 37% is in Sudan. The elevation of the basin varies greatly from over 4000 m in the headwaters of some tributaries to 700 m at the foot of the plateau. The highest point in Lake Tana is about 1800 m and the river enters Sudan at an elevation of 490 m at the border of the two countries, i.e., with a gradient of approximately 1.5 m/ km. This gives the Blue Nile its unique feature of huge potential energy opportunity to develop hydropower. 3. Existing and proposed water resources developments The physical characteristics for the High Aswan Dam and Lake Nasser/Lake Nubia were extracted from the Power Toolkit provided by ENTRO. The information contains descriptions of the turbine characteristics including explicit relationships between operating head, turbine releases and power generation. Several water resources development projects in the eastern Nile basin are categorised as under operation (existing), under construction and proposed to be developed. Most of the operational reservoirs are in Sudan and Egypt. The under construction Grand Ethiopian Renaissance Dam (GERD) and the other three cascades under design and study phases are located in Ethiopia (Fig. 1 and Table 1). 3.1. Existing developments 3.1.1. Tana Beles project Lake Tana is a natural reservoir that controls the Tana sub-basin flows. The Lake is fed by the Gilgel Abbay, Megech, Ribb and Gumara rivers; and its surface area ranges from 3000 to 3500 km2 depending on season and the amount of rainfall. The lake level has been regulated since the construction of the control “Chara Chara” weir where the lake discharges into the Blue Nile. This controls the flow to the Blue Nile Falls (Tis Abbay) and hydro-power station, which now the hydropower production is left in favor of the Tana Beles plant. The Tana-Beles hydropower project was completed in May 2010 and the project diverts water directly from Lake Tana and is represented as such in the models. The relationship between the diverted flow and intake elevation for the diversion is defined in the Beles Multipurpose Level 1 Design Report from the Ethiopian Electric Power Corporation (EEPCO, 2006). Water diverted through the Tana-Beles project is returned entirely to the headwaters of the Beles which flows into the Blue Nile upstream of the border between Sudan and Ethiopia. 3.1.2. Roseires Dam Roseires Dam was completed in 1966 with an initial capacity of 3.024 × 109 m3 at level 480 m level. The main objective to supply irrigation demands as first priority, and hydropower generation is the second priority. During its lifetime, the reservoir suffered from serious sedimentation which reduced its storage capacity to less than 2.0 × 109 m3. The deep sluices with sill levels of 435.5m.a.s.l are used to pass the main volume of the flood, and to flush the sediment. Due to the large seasonal fluctuation, relatively small storage volume, and high amount of sediment accumulating in Roseires, the operational criteria is specified to draw down the reservoir starting in mid-January and maintain a minimum elevation until the peak flow has passed in September. Therefore, meeting target elevation criteria is the primary guiding principle of the operation of Roseires Dam as it was indicated in the ENTRO power tool kit. 3.1.3. Sennar Dam The Sennar Dam was built in 1925 on the Blue Nile near the town of Sennar, Sudan. The dam is 3025 m long, with a maximum 2
Journal of Hydrology: Regional Studies 16 (2018) 1–14
A.G. Mulat et al.
Fig. 1. Schematics of Eastern Nile River Basin.
height of 40 m. It provides water for irrigation in the Al Jazirah region. Similar to Rosaries Dam, a primary operational objective of this reservoir is to achieve drawdown and refill elevations on specified dates. In addition, the Gezira Managil diversion takes water directly from the reservoir for agriculture purposes. The minimum diversion elevation of the Gezira Managil diversion is 417.0m.a.s.l and therefore demands can be met when the pool elevation is greater than this level. All water in the reservoir above this elevation is considered available for diversion (ENTRO power tool kit). 3.1.4. Jebel Aluia The Jebel Aulia dam built 40 km upstream of Khartoum in 1937 on White Nile River to store water for later use in Egypt. The rapid silt up of this reservoir and the construction high Aswan dam in Egypt in 1965 stopped its function (Shahin, 1985, 2002). 3.1.5. Khashm El Girba Dam Khashm El Girba Dam just downstream of the confluence of the rivers and the Tekeze Dam on the Tekeze River. The primary 3
Journal of Hydrology: Regional Studies 16 (2018) 1–14
A.G. Mulat et al.
Table 1 Proposed and implemented Hydropower plants in the Eastern Nile (installed and target power in megawatt (MW) and tail water in meter above sea level (m.a.s.l). Hydropower plant
Tana Beles Karadobi Bekoabo Upper Mendia Bekoabo Low GERD Rosaries Sennar Merowe Tekeze Jebil Aluia HAD
Hydro power (MW) Installed capacity
Target Power
460 1600 1940 1700 935 6000 425 15 1000 300 28.5 2100
460 933 1329 802 514 1807 200 15 650 180 15 783
Tail water level (m.a.s.l)
Country (location)
Remarks
1356 890 800 640 603 500 437 406 240 967 300 120
Ethiopia Ethiopia Ethiopia Ethiopia Ethiopia Ethiopia Sudan Sudan Sudan Ethiopia Sudan Egypt
In operation Under study Under study Under study Under study Under construction In operation In operation In operation In operation In operation In operation
consumptive use is a direct diversion from the Khashm El Girba Dam (ENTRO power tool kit). 3.1.6. Tekeze Dam Tekeze Dam is a double-curvature arch dam in the northern Ethiopia on the Tekeze River, a Nile tributary that flows through the deepest canyons into the main Nile downstream of Khartoum. At the time of completion, the 188 m high dam is said to be Africa's largest arch dam. The operation of this reservoir is to primarily meet a target power generation of 112 MW, with a maximum power capacity of 300 MW. A maximum pool elevation of 1140 ma.s.l and a minimum operation level of 1096 ma.s.l was used as the range over which power could be generated. 3.1.7. Merowe Dam The Merowe Dam is recently constructed which is operated to meet the primary objective of hydropower generation. The target power generation is 625 MW with a maximum power capacity of 1250 MW. To accomplish this operation, a rule was used that specifies a turbine release to meet the power generation objective, followed by a flood control rule that spills any water in excess of a specified elevation. A maximum pool elevation of 300 ma.s.l and a minimum operation level of 284.90 ma.s.l was used as the range over which power could be generated (ENTRO power tool kit). 3.1.8. High Aswan Dam (HAD) Aswan High Dam Reservoir extends for 500 km along the Nile River and covers an area of 6000 km2, of which northern two-thirds (known as Lake Nasser) is in Egypt and one-third (called Lake Nubia) in Sudan. The dam, completed in 1968 at a distance of 7 km south of Aswan City, is 2325 m long, 111 m high over the original river bed, and 40 m and 980 m wide, respectively, at its crest and bottom. Nile flow is allowed to pass only through the open-cut channel at the eastern side of the dam, where six tunnel inlets provided with steel gates are constructed for discharge control and water supply to power plants. An escape is also provided at the western side of the dam to permit excess water discharge (ENTRO power tool kit). The total capacity of the reservoir (162 × 109 m3) consists of the dead storage of 31.6 × 109 m3 (147 ma.s.l. of Lake water level), the active storage of 90.7 km3 (147–174 m.a.s.l) and the emergency storage for flood protection of 41 × 109 m3 (175–182 m.as.l). The reservoir is surrounded by rocky desert terrain. To the west is the great Sahara Desert, and the Eastern Desert on the east side extends to the Red Sea. The Aswan High Dam contributed greatly to the economic development of Egypt by supplying irrigation water and about 2000 MW hydroelectricity and protecting the lower reaches of the Nile from flood disasters. 3.2. Proposed developments 3.2.1. Karadobi hydropower project The Karadobi hydropower project is located on the Abbay River (Blue Nile). The proposed dam site is located 1.7 km downstream of its confluence with Guder River at about 135 km (air distance) north west of Addis-Ababa. The project was studied at reconnaissance level in the Abbay River Master Plan Project (BCEOM, 1998). The present pre-feasibility study of the Karadobi MultiPurpose Project indicate a rolled concrete gravity dam of maximum structural height of 260 m and length of about 684 m at the crest, with a corresponding installed capacity of 1600 MW. The reservoir area full level has an area of 445 km2 with a capacity of 40200 MCM. The project is an element in the Nile Basin Initiative regarding development of power trade among the countries in the Nile Basin (Norplan-Norconsult International, 2006). 3.2.2. Bekoabo hydropower project Beko-Abo dam height 285 m, crest length 880 m and reservoir live storage 17.5 × 106 m3 capacity and 2000 MW (12100 GWh/yr) (ENTRO power tool kit). 4
Journal of Hydrology: Regional Studies 16 (2018) 1–14
A.G. Mulat et al.
3.2.3. Mendya hydropower project Extensive reservoir impounding is foreseen for the projects which are expected to again feature large dam construction, at sites 220 km from Bekoabo. While Beko abo and Mendia are separate projects the headwaters of the lower, Mendya scheme will extend all the way up to the location of Beko Abo. Mendya dam height 170 m, crest length 1400 m and reservoir live storage 10.3 × 106 m3 M, Capacity 1700 MW and 8220 GWh/yr (ENTRO power tool kit). 3.2.4. Under Construction Grand Ethiopian Renaissance Dam (GERD) The GERD site is the most downstream of the hydropower sites under construction within Ethiopia. The eventual site for the Grand Ethiopian Renaissance Dam was identified by the United States Bureau of Reclamation during a Blue Nile survey conducted between 1956 and 1964 (United States Bureau of Reclamation (USBR), 1964). The Ethiopian Government made the project public on March 31, 2011. The first two generators are expected to become operational after 44 months of construction. It is slated for completion in July 2017. The dam will be a roller-compacted concrete (RCC) gravity-type comprising of two power stations, three spillways and a saddle dam. The dam will be 150 m in height and 1800 m in length (Coyne ET BELLIER and TRACTEBEL Engineering, 2011). 4. Data in the eastern nile 4.1. Stream flow data The source of hydrological data (for the gauged data) is from Ministry of Water of Water, Irrigation and Electricity (MoWIE). There are about 164 hydrometric stations in the Blue Nile river basin in Ethiopia. According to World Meteorological Organization (WMO) recommendations, the number of stations are more than satisfactory, i.e. one station should cover 1800 km2 (BCEOM, 1999). But most of the stations are not working properly, missed data and stopped functioning. As indicated in Mulat and Moges (2014), the monthly input stream flow series is reconstructed from 50 sub-basin flows of Upper Blue Nile (Abbay River). Other stream flows such as that of Tekeze-Setet-Atbara (TSA) and the White Nile at Melakal was collected from Eastern Nile Technical Regional Office (ENTRO) spreadsheet. The later data sets are essential to construct the complete Eastern Nile Model up to HAD in Egypt (Fig. 2). 4.2. Reservoirs data For the standard reservoir, the time series information required includes bottom, crest, spillway, top of dead storage level, and minimum operation pool elevations; minimum and maximum valve releases; precipitation, seepage loss, and evaporation; and flood control and operational rules levels for any water users attached to the reservoir. Simulation starts from their bottom level for upper cascades with sequences of 5 years interval that is intended to offset the impacts of filling the reservoirs on operational reservoirs including GERD and HAD. 4.3. Water users data The most common water use in Eastern Nile River Basin is irrigation water demand. The water system retained for the purpose of this study will assume water requirements for irrigation purposes in Sudan and water requirements associated to High Aswan Dam (HAD). HAD outflows aim to satisfy Egyptian irrigation water demand as priority to energy generation. 55.5 × 109 m3/year are allocated to Egyptian irrigation supply downstream of HAD. This water is also used to generate power within the HAD installed capacity. 18.5 × 109 m3/year is allocated to Sudan for irrigation supply at four different points upstream of HAD in the water system model. As long as HAD water level is equal or greater than the Minimum Operating Level (147 m), water volumes for HAD irrigation are supplied at a monthly time-step according to water demand presented in Table 2. The irrigation water demand in BCM will be changed to m3/s for model input with 10% return flow.
Fig. 2. Seasonal flow distribution of Eastern Nile Basin at selected stations.
5
Journal of Hydrology: Regional Studies 16 (2018) 1–14
A.G. Mulat et al.
Table 2 Water demand input data (m3/sec) used for modeling. Month
Jan
Feb
Mar
Apr
May
June
July
Aug
Sep
Oct
Nov
Dec
HAD Saboloka D/S Sennar U/S Sennar Gizira Dongola Khashim Jubial
983.8 18.85 7.51 67.46 202.34 13.44 45.56 29.97
1296.3 10.75 8.1 66.45 191.27 10.75 51.35 34.55
1628.1 20.37 10.42 44.69 118.13 20.37 51.82 38.74
1662.8 25.08 11.72 34.87 30.78 25.08 55.41 43.26
2137.4 25.92 12.39 35.09 44.49 25.92 46.73 45.03
2800.9 23.26 12.54 77.06 180.89 23.26 45.23 47.85
2835.7 18.85 9.04 97.39 268.64 18.85 42.24 78.52
2442.1 12.85 7.22 93.75 258.55 12.85 39.78 78.66
1736.1 5.34 10.28 132.38 337.76 5.34 27.43 146.06
1388.9 4.63 11.35 148.2 321.77 4.63 16.89 166.39
1284.7 7.17 10.11 141.85 249.6 7.17 19.44 145.5
1219.1 14.81 7.25 82.78 185.58 14.81 40.83 45.75
4.4. Hydropower data All input data which enables for running the simulation are from Ethiopian Electric Power Corporation (EEPCO) Grand Ethiopian Renaissance Dam Project study and ENTRO. For all hydropower plant the engine efficiency is considered to be 95%. Table 1 shows proposed and implemented Hydropower plants in the Eastern Nile (installed and target power in megawatt (MW) and tail water in meter above sea level. 4.5. Evaporation losses Losses/gains were placed at all reservoir locations within the basin to account for reservoirs gains due to precipitation and losses due to Evaporation. These net losses represent the difference between gross precipitation on the reservoir and natural losses due to Evaporation. The input data for net losses from reservoirs are shown in Table 3. 5. River basin modelling and setup The natural river system of the Eastern Nile river basin was schematized and represented with a node-branch structure. A number of nodes and corresponding reaches were established based on the river network configuration (Fig. 2). Branches represent the main river and its main tributaries, and nodes represent major river confluences, reservoirs, and control points for off take. Off take points were selected on the main river and/or tributaries to release water to cover downstream irrigation and hydropower water demands. In the simulation, only two major water demand sectors were considered: irrigation and hydropower. There are eighteen irrigation sites within the basin considered here (Table 2). The major irrigation system in the basin is gravity flow (normally from reservoirs). The schemes were then allocated to the nearby nodes for their water withdrawals, and return flows from established schemes were directed to the immediate downstream nodes. 6. Development scenarios During this study, an investigation has been conducted to assess the future probable changes of hydrological process which may have impacts on the spatial and temporal distribution of water availability in the basin. There are three proposed dams (Karadobi, Bekoabo high/low and Mendya) in the Abbay basin. There are two options for Bekoabo (Bekoabo high and Bekoabo low). It is not possible to combine Bekoabo high and Karadobi because of the minimum available elevation difference between them. The construction and fillings of each cascades is based on five year intervals from the top to downstream, i.e., Karadobi, Bekoabo and Mendya will be constructed and fill in order. To get the best combination of the reservoirs operating simultaneously, the Table 3 Monthly Net evaporation losses (mm/day). Reservoir
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
HAD GERD Khashim Elgibra Jubil Aluia Lake Tana Rosaries Sennar Karadobi Tekeze Merowe Beko abo Mendia
4.41 6.48 5.80 6.10 4.47 5.80 5.99 6.43 4.84 6.35 5.57 6.22
5.95 6.61 6.80 8.11 4.37 6.37 6.16 6.12 5.45 7.66 5.37 6.34
8.61 5.13 8.07 7.50 3.30 7.31 7.55 4.91 5.32 9.52 5.13 4.75
10.64 4.94 8.27 7.40 2.57 7.37 7.37 4.52 5.60 11.03 4.20 4.52
12.25 3.76 8.60 7.40 2.63 6.10 6.30 3.51 5.97 11.77 2.10 3.39
13.51 −0.25 7.93 7.50 −1.17 2.09 2.09 −0.69 3.80 11.33 −2.17 −0.69
13.3 −6.06 4.20 3.80 −5.13 −0.90 −0.93 −7.30 −2.65 10.13 −4.60 −7.07
13.16 −3.93 2.70 3.70 −6.10 −0.84 −0.87 −5.02 −3.55 9.77 −1.53 −4.85
11.55 1.04 5.40 4.50 −1.23 0.65 0.65 0.57 2.90 10.57 0.27 0.57
8.96 4.11 6.10 4.90 3.77 4.03 4.16 3.92 4.45 9.90 2.20 3.79
5.56 5.16 6.27 5.80 4.73 5.23 5.23 4.93 4.70 7.70 3.90 4.93
5.11 6.25 6.10 5.68 4.90 5.58 5.58 6.00 4.60 6.63 4.83 6.00
6
Journal of Hydrology: Regional Studies 16 (2018) 1–14
A.G. Mulat et al.
Table 4 Hydropower dams development scenarios in the Eastern Nile basin with their implementation intervals in years. Starting Year
Before 2020
Scenario Scenario Scenario Scenario Scenario Scenario Scenario Scenario
Currently operating reservoirs in the Eastern Nile Basin
1 2 3 4 5 6 7 8
2020
2025
GERD GERD GERD GERD GERD GERD GERD
Karadobi Karadobi Karadobi Karadobi Bekoabo high Bekoabo high
2030
2035
Mendya Bekoabo low Bekoabo low
Mendya
Mendya
following scenarios have been considered and summarized in Table 4.
• Baseline scenario (Scenario 1): the reference scenario or current condition. • GERD only (Scenario 2): the combination of GERD and currently operating reservoirs. • Karadobi and GERD (Scenario 3): the second scenario with Karadobi. • Karadobi, GERD, and Mendya (Scenario 4): this one is the combination of Karadobi, GERD and Mendya with the currently scenario. • Karadobi, GERD and Bekoabo low (Scenario 5): this is scenario 4, but Bekoabo low instead of Mendya. • Karadobi, GERD, Mendya and Bekoabo low (Scenario 6): here four reservoirs are considered with current condition. • Bekoabo high and GERD (Scenario 7): this similar with scenario 3 but Bekoabo high instead of Karadobi. • Bekoabo high, GERD and Mendya (Scenario 8): Mendya with scenario 7 7. Results and discussion 7.1. Inflow to reservoirs The flow regime of the downstream dam is governed by the operations of the upstream hydroelectric power schemes and runoff from the catchments (the incremental flow) between the two dams. Reservoirs in the upstream have effects on the downstream reservoirs inflow. During filling period of new constructed reservoir, the amount of water released will be reduced and more regulated. For instance, there is slight inflow reduction to GERD during the impounding of upper cascades in the Abbay-Blue Nile River. During full operation phase, the reservoirs in the downstream will gain maximum flow in the dry months and the flood will decrease in wet months as a result of regulating effects of upstream cascades. Fig. 3 demonstrates mean monthly inflow to Rosaries dam for different dam scenarios. Releases from Karadobi, Bekoabo and Mendya dam constitute the major inflow into GERD dam, This means that the more the release from upper reservoir the faster the downstream reservoir fill up and excess will be discharged. The operation of GERD and other upstream dams dictate operation pattern in downstream dam, excess releases at GERD will force the reservoir manager at Rosaries dam to release so as to accommodate releases from GERD and thereby causing flooding at the downstream area. Because of its storage capacity, GERD has significant effect on the inflow of the Sudanese reservoirs. Rosaries and Sennar dam receive more regulated flow due to GERD implementation. The monthly reservoirs inflow coefficient of variance is maximum in baseline scenario as compared to the other scenarios. The mean monthly inflow to downstream reservoirs will also be improved, for instance the mean monthly coefficient of variance for Rosaries is 1.2 and 0.2 for baseline and other scenarios respectively. For Sennar it is 1.3 for baseline scenario and 0.16 for the other scenarios. The same is true for HAD inflow, the coefficient of variance of inflow to HAD decreases from 1 (for baseline scenario) to 0.5 (for other scenarios).
Fig. 3. Mean monthly inflow to Rosaries (106 m3/month).
7
Journal of Hydrology: Regional Studies 16 (2018) 1–14
A.G. Mulat et al.
There are changes in other flow statistics. The mean, maximum and minimum monthly flow to downstream reservoirs will be changed due to the implementation of upper cascades. For instance, the maximum inflow to HAD in dry month (May) is 3.4 × 109 m3 for base line scenario but for the scenarios 1–8, it is 4.6 × 109 m3 that is 35% incremental. But for the wet month (August) it is 32.6 × 109 m3 and 19 × 109 m3 for baseline and for scenario 1–8 respectively that is 42% reduction. The inflow to HAD in the dry months (March, April, May and June) of selected dry year periods (from 1982 to 1988) will increase by 133% (from 1272 × 106 m3 to 2967 × 106 m3) due to the implementation of upper cascades. This shows water availability at HAD could be safe even drought occurs. For the selected wettest years (from 1996 to 2001) and wet months (July, August, September and October), flow to Khartoum will decrease by 100% (from 10113 × 106 m3 to 5084 × 106 m3). That is, if flow which cause flood occurs in the highlands of Ethiopia, it will be controlled by the upstream reservoirs and will be released in regulated manner. The impact of the upstream cascades on the mean annual flow to reservoirs in the downstream is insignificant. For instance the difference between maximum (65294 × 106 m3, with current scenario) and the minimum (62438 × 106 m3, for the scenario 4) mean annual inflow to HAD is 4.4%. For Merowe the difference between maximum and minimum mean annual inflow is 3.3%. 7.2. Reservoir water level Filling period for Karadobi is assumed to be 5 years and after that it start operating with full capacity. Bekoabo low will start filling when Karadobi reaches its full level and then Mendya filling will start. The construction and filling of Mendya lags by 5 years of Karadobi or Bekoabo high filling, but for the three reservoirs combination (Karadobi + Bekoabo low + Mendya), the filling will be after 10 years of Karadobi filling. There is no reservoir upstream of Karadobi which affects its operation. There are slight impacts of upstream cascades filling on the GERD pool level. There could be up to 7 m monthly water level reduction of GERD during filling periods of the upper cascades. After filling of all upstream reservoirs, the maximum GERD mean monthly reservoir water level difference, which is between scenario 2 and scenario 4, is 3 m. Fig. 4 show Sudanese reservoir (Rosaries) water levels for different scenarios. As shown in the figures, all Sudanese reservoirs water level will increase due to upper cascades implementation. Their pool level never be below baseline scenario’s level in both filling and operation phases. The maximum pool level for all Sudanese reservoirs is scenario 2. Though it is not below baseline scenario level, there could be up to 11 m water level reduction of Rosaries during the filling of Karadobi, Media and Bekoabo (scenario 4) as compared to scenario 2. Sennar water level will not be affected by reservoirs in Abbay-Blue Nile river basin in both filling and operation phases. Because of its storage capacity there is water level reduction in the dry months for the baseline scenario but this could be mitigated by implementing reservoirs in the upstream. There is slight impact of upper cascades filling on Merowe reservoir as compared to scenario1 and scenario2. There could be up to 5 m water level reduction of Merowe reservoir due to upper cascades filling as compared to scenario 2, but it will never reach below water level of baseline scenario. At the end of impounding of all reservoirs HAD reservoir water level decreases up to 14 m. The maximum reservoir water level reduction will be if all proposed reservoirs in Abbay River are implemented. HAD reservoir water level will have less fluctuation when there are more reservoirs in upstream. Reservoir water level range is 36 m in the simulation period of baseline scenario, but for the other scenarios it is around 31meters. 7.3. Energy production This section discusses the production of energy in the Eastern Nile region by considering scenarios shown in Table 1. Regional and each Eastern Nile country’s’ hydropower production are discussed in the following sub-sections. 7.4. Energy production in abbay-Blue nile river (Ethiopia) The energy production impact of upstream cascade development in the Abbay-Blue Nile River basin is similar with their impacts on the reservoir water level. The simulation results of energy production for the different scenarios are provided in Tables 5 and 6. As
Fig. 4. Mean monthly Rosaries water level during full operation phase.
8
Journal of Hydrology: Regional Studies 16 (2018) 1–14
A.G. Mulat et al.
Table 5 Mean annual Energy generation (GWh/yr) of each reservoir for different scenarios during filling and operation phases of upper reservoirs. Mean Annual Energy (GWh/yr) Scenarios
GERD
Roseires
Sennar
Merowe
Had
Scenario Scenario Scenario Scenario Scenario Scenario Scenario Scenario
16256 15717 15269 15699 15376 15848 15394
2200 3485 3774 3426 3476 3425 3474 3465
131 131 131 131 131 131 131 131
5946 7790 7698 7587 7691 7587 7698 7615
7462 6982 6628 6411 6608 6508 6736 6493
1 2 3 4 5 6 7 8
Mendia
Bekoabo Low/High
6147 3498 3498 7699 7699
4316 5860
Karadobi
EN
Diff (%)
7132 7132 7132 7132
15739 34644 41080 46103 44235 47973 41586 46657
120 161 193 181 205 164 196
Table 6 Country wise and Basin wide (Eastern Nile) energy generation full operation phases of all reservoirs in the Eastern Nile. Countries
Ethiopia GWh/yr
Scenarios Scenario Scenario Scenario Scenario Scenario Scenario Scenario Scenario
1 2 3 4 5 6 7 8
HAD (Egypt) Diff GWh/yr
16256 24876 33264 29594 38200 28133 36525
8620 17008 13338 21944 11877 20269
GWh/yr %
53 105 82 135 73 125
7462 6982 6893 6791 6891 6816 6926 6820
Sudan Diff
GWh/yr
GWh/yr
%
−482 −571 −674 −573 −648 −539 −644
−6 −8 −9 −8 −9 −7 −9
8277 11407 11428 11382 11429 11380 11450 11396
Eastern Nile Diff (%)
GWh/yr
GWh/yr
%
3130 3151 3104 3152 3103 3172 3119
38 38 38 38 37 38 38
15742 34645 43197 51436 47914 56396 46508 54742
Diff GWh/yr
%
18904 27456 35695 32173 40654 30767 39000
120 174 227 204 258 195 248
shown in Fig. 4 and in Table 5, the energy generation at GERD will be slightly affected due to the implementation of upper cascades. Because of their maximum capacity, Karadobi and Bekoabo high have significant impacts on the energy production of GERD during their filling phase. During Karadobi filling period, the energy production of GERD will reach minimum level (9163 GWh/yr) which is 68% of scenario 2 (Fig. 5). Bekoabo high filling will reduce the GERD energy to 10883 GWh/yr which is 2546 GWh/yr less than that of scenario 2. There is slight fluctuation of energy production when only GERD is in operation (scenario 2). For instance, the coefficient of variance of the energy within the simulation period for GERD is 0.4 for scenario 2, while it is 0.35 for the other scenarios. At full development level, maximum GERD energy (16256 GWh/yr) could be generated when there is no cascade upstream of GERD. When there are more reservoirs in operation above GERD, there could be slight energy reduction of GERD (Table 5). The different between the maximum and the minimum is 463 GWH that is 3% reduction from scenario 2. The energy production at Karadobi will not be changed because there are no reservoirs above it. The mean annual energy at Karadobi is 8864 GWh/yr during its full operation phase. There are three scenarios for Mendya that is Mendya with Bekoabo high, Mendya with Karadobi and Mendya with Karadobi and Bekoabo low. Fig. 6 shows energy production of Mendya during filling and operation phases for the combination of Mendya with; Karadobi, Karadobi + Bekoabo low and Bekoabo high. The impounding period is assumed to 5 years and maximum energy production will be after the filling of the reservoir to its full reservoir level (808 m.a.s.l). Fig. 7 shows energy production of Bekoabo low and Bekoabo high during filling and full operation phases respectively. There are
Fig. 5. Annual energy generation of GERD during filling phases of all reservoirs in Abbay river basin (GWh/yr).
9
Journal of Hydrology: Regional Studies 16 (2018) 1–14
A.G. Mulat et al.
Fig. 6. Mendya energy production during impounding and operation phases when combined with different combination of the upper cascades.
Fig. 7. Bekoabo (Bekoabo low and Bekoabo high) energy production during impounding and operation phases.
two options for Bekoabo that is Bekoabo high and Bekoabo low with Karadobi. As the name indicates, there is difference between Bekoabo high and Bekoabo low in storage capacity and normal water level. Therefore, there is big mean annual energy production difference between Bekoabo High and Bekoabo (Table 5). The construction and filling of Bekoabo low is after the implementation of Karadobi. Fig. 8 demonstrates the cumulative (total) energy production in Abbay-Blue Nile River (Ethiopia) after all reservoirs are completed. As shown in Figures and Tables 5 and 6, the overall energy production in Ethiopia will increase significantly due to the construction of the upper cascades. The maximum mean annual energy production (38200 GW/yr) is scenario 6 which combines four reservoirs (GERD, Karadobi, Bekoabo low and Mendya) in Abbay-Blue Nile River basin. The next maximum mean annual energy production (36525 GWyr) is scenario 8 which combines GERD, Bekoabo high and Mendya.
7.5. Energy production in Sudan Energy production in Sudan is from Rosaries, Sennar and Merowe dams. Sudan would also benefit from the upstream infrastructures: reduced spillage makes more water available for hydropower generation. Fig. 9 present the monthly average energy generated by the major hydropower plants (Rosaries) for all development scenarios. There is slight reduction of energy at Roseires during the filling periods of cascades in Abbay-Blue Nile River as compared to scenario 2. But energy generation of Rosaries is always more than that of scenario 1. Even though there is a slight effect of upstream cascades filling on Merowe energy production as compared to the scenario 2, there is significant incremental of energy as compared to scenario 1 For instance there is around 900 GWh more energy production during Bekoabo high filling as compared to scenario 1. Generally there are positive impacts of upper cascade development for energy production in Sudan. Due to their less storage capacity, the energy production in the dry months is less but the upper cascade development can mitigate this problem. Fig. 10 shows the cumulative energy production of reservoirs located in Sudan during filling and full operation phases. Due to the upper cascades, there will be flow regulation and reservoirs in Sudan will get uniform flows and the energy production will increase due to the upper cascades. The minimum energy production in Sudanese reservoirs is in the current situation (when there are no
Fig. 8. Total annual energy generation in upper Abbay River (Ethiopia) during impounding and operation phases of the upper cascades (GWh/yr).
10
Journal of Hydrology: Regional Studies 16 (2018) 1–14
A.G. Mulat et al.
Fig. 9. Mean monthly energy generation at Rosaries during full operation of all reservoirs (GWh/month).
Fig. 10. Annual energy generation in Sudan during full operation of the upper cascades (GWh/yr).
cascades in operation in the upper Blue Nile). But if there is a reservoir in upper Blue Nile, cumulative energy production in Sudan will increase by at least 37% (Table 6). 7.6. Energy generation at HAD Energy production at HAD will decrease due to the upper cascades. The production of hydroelectricity from HAD in Egypt would be reduced on average by 8.43% as compared to baseline scenario. Fig. 11 depicts the energy production of HAD during filling and operation phases of upper cascades. Karadobi or Bekoabo high filling will reduce the HAD mean annual energy to 5039 GWh/yr which is 33% less than the baseline scenario. The overall impacts of the upper cascades filling are insignificant (Table 6). Maximum energy reduction at HAD will be 14% of the baseline scenario. During full operation phases, the maximum energy (7462 GWh/yr) is the baseline situation (Table 6). Maximum energy reduction as compared to the baseline scenario is 9% (Table 6) during the full operation phases. 7.7. Energy production in Eastern Nile At the basin scale, the annual production of hydroelectricity is boosted by at least 34644 GWh/yr amongst which 3130 GWh/yr is from Sudanese reservoir due to the regulation capacity of Ethiopian reservoirs. As shown in Fig. 12, Tables 5 and 6, the energy production in the whole Eastern Nile system will increase significantly when all reservoirs are at their full operation level. During their filling phase there is slight reduction of energy generation as compared to scenario 2 (Fig. 11). The column ‘Diff’ in Tables 5 and 6 indicates the difference in energy production of Eastern Nile between the current situation (without GERD) and other scenarios. During the full operation phase, Eastern Nile energy will increase at least by 120% and it will reach up to 258% incremental. If each reservoir is considered as shown in Table 5, energy productions in Sudan will increase due to the upper cascades but in HAD energy production will slightly decrease because of head reduction. The maximum energy in the Eastern Nile could be 56396 GWh/yr for scenario 6 which combines GERD, Mendya, Bekoabo low, and Karadobi with the baseline scenario.
Fig. 11. Annual energy generation at HAD during filling and operation of the upper cascades (GWh/yr).
11
Journal of Hydrology: Regional Studies 16 (2018) 1–14
A.G. Mulat et al.
Fig. 12. Energy generation of eastern Nile with upper cascades in Abbay River during filling and operation phases of upper reservoirs (GWh/yr).
7.8. Water loss from reservoirs Water loss is due to evaporation from the water surfaces of the reservoirs. As the water level increases there will be more free water surface area and the evaporation loss will be increased. The column ‘Diff’ in Table 7 indicates the difference in water loss from the reservoirs of Eastern Nile between the current situation (scenario 1) and other scenarios. Evaporation loss (mm/day) in the upstream part of the basin is less as compared to downstream. Meanwhile as a dam is constructed upstream, there will be additional free water surface area and more evaporation loss. The loss in Ethiopia will increase from nil up to 2024 × 106 m3 because of the reservoirs. There is more regulated and constant flow of water to the Sudanese reservoirs and the reservoir level is maintained at higher level regularly in an increase of water surface area. Due to that the loss in Sudanese reservoirs increases by at least 35% during full development stages of all upper reservoirs. Water loss at HAD slight decrease as the water level also slightly decreases with addition of reservoirs in the upper basin. The maximum HAD loss is 11673 × 106 m3 is in baseline scenario and the minimum loss is 10190 × 106 m3 is in scenario 2.
7.9. Irrigation water demand deficit Irrigated agriculture in Sudan will benefit from upstream storage in Ethiopia since the Sudanese annual withdrawals are lower in all scenarios. Those allocation decisions illustrate that once water has passed through the Ethiopian hydropower plants, irrigated agriculture starts competing with hydropower generation and irrigation withdrawals become more economically sound. Table 8 illustrates the benefits and impacts of storage in Ethiopia on the irrigated agriculture sector. The regulation capacity of the reservoirs located in Ethiopia would increase irrigation water availability in Sudan. No significant impacts are observed for Egypt. The filling impacts on water demand is insignificant because there are only two events of irrigation water demand deficit at HAD due to the filling of the reservoirs. The irrigation water demand deficit has been considered in the downstream projects, especially in HAD, in all scenarios during full operation phase. The difference between scenarios is the extent of demand deficit within the simulation period. The performance analysis is made by using equation shown below. As shown in Table 8, the water demand deficit is not significant. That is the performance is nearly 100% in irrigation projects in Sudan and more than 92% for HAD irrigation water demand.
Event − based reliability =
Total number of non − failure months Total number of months
Table 7 Mean annual losses (106 m3) from each reservoir for deferent scenarios during filling and operation phases of all reservoirs. Mean Annual Loss (x106 m3) Scenarios
GERD
Rosaries
Sennar
Merowe
HAD
Scenario Scenario Scenario Scenario Scenario Scenario Scenario Scenario
1459 1446 1416 1445 1435 1460 1426
241 419 424 422 424 416 428 428
198 200 210 210 210 210 210 210
1647 2251 2224 2191 2226 2196 2221 2207
11673 10190 11153 10714 11099 10988 11490 10929
1 2 3 4 5 6 7 8
Mendya
Bekoabo Low
363 21 21 174 174
247 360
12
Karadobi
EN
Diff (%)
Eth loss
Sudan loss
245 245 245 245
13759 14519 15702 15561 15670 15758 15983 15734
0 6 14 13 14 15 16 14
1459 1691 2024 1711 1948 1634 1960
2086 2870 2858 2823 2860 2822 2859 2845
Journal of Hydrology: Regional Studies 16 (2018) 1–14
A.G. Mulat et al.
Table 8 Water demand deficit (number of months with water demand deficit, amount of deficit water in million cubic meters (×106 m3) and Nash-Suttcliffe coefficient (NS) for full operation phase. Scenarios
Scenario Scenario Scenario Scenario Scenario Scenario Scenario Scenario
1 2 3 4 5 6 7 8
HAD
Gizira 6
3
No of Month
10 m
25 34 35 36 40 42 34 40
93368 113245 113951 114876 126205 125497 111750 123486
NS
No of Month
106 m3
NS
0.95 0.93 0.93 0.93 0.92 0.92 0.93 0.92
2 1 0 0 0 0 0 0
10.9 72.4 0 0 0 0 0 0
1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00
8. Conclusion and recommendations The aim of this study is to understand the future water development perspective in the Eastern Nile region by considering the current irrigation water use and proposed reservoirs in the upper Blue Nile (Abbay) River basin in Ethiopia using a river basin simulation approach. Simulations was carried out on a monthly time step and using historical ensemble time series data as representative for possible near future scenarios during the implementation of all planned reservoirs in the upper Blue Nile. The scenarios were evaluated on economic benefits for the whole eastern Nile, basin wide, and at the project site. The economic benefits considered are energy production and irrigation water demand satisfaction. From all the scenarios Ethiopia could have maximum energy (38200 GWh/yr) could be achieved when Karadobi, Bekoabo low, Mendia and GERD are in operation simultaneously. The next maximum energy (36525 GWh/yr) could be when Bekoabo high, Mendia and GERD are combined. The maximum Eastern Nile energy (56396 and 54742 GWh/yr) could be achieved in these scenarios. Energy in the eastern Nile will increase at least by 126% for GERD only case and could increase by 258% for Karadobi, Bekoabo low, Mendia and GERD combination. The energy production in Sudan will increase by 39% due to the construction of a single reservoir in the Abbay River. There could be small energy reduction (up to 9%) at HAD due to reduced reservoir water level. There are also slight water loss changes in the whole Eastern Nile due to the implementation of reservoirs. The minimum loss is in the current situation. Development of Abbay River will increase the free water surface area and the evaporation in the basin. The construction of new reservoirs in the Blue Nile results in two competing effects with respect to evaporation. The first effect is increased evaporation resulting from the proposed reservoirs while the other is reduced evaporation from HAD as a result of decreased inflows to the reservoir and reduction of the free water area. The maximum incremental water loss in the Eastern Nile could be 10% as compared to the baseline scenario. The amount of water flowing to the reservoirs will be more regulated and more constant due to the upstream cascades. Reservoirs in Sudan are small in storage capacity as compared to the other reservoirs in Ethiopia and Egypt. Due to that they cannot store water up to their full level in the dry seasons and there are fluctuations in energy production and reservoir water level. But the upper cascades in Abbay River would enable them to get regulated and more water flow in the dry months and to maintain their water level and energy production constant and at higher level. There will be reduction of the risk due to hydrological variability with sequences of dry and wet years. Indeed, at the present time, excessive water may be spilled during wet period at downstream reservoirs because existing reservoirs are already filled and, as a result, water demand failure can occur during a dry period. With GERD, the total storage capacity along the Nile River will significantly increase in the long term. Basin Water Management will be easier to optimize with higher storage capacity and upstream regulation capacities The water demand deficit is insignificant in all scenarios, that is, the performance is more than 92% in all cases. In Gizira irrigation projects there are no months with water demand deficit due to the upper cascades while there are 2 months in the current scenario. As recommendation, cascades in Abbay River basin are not studied at their final stage and their construction period is not specified. So, further evaluations may be necessary after their design and construction time is specified. Further detailed study could also be necessary by considering all proposed and implemented small and medium irrigation and water abstraction projects in Abbay River basin. Research work for the way to improve water use efficiency is crucial. Managing water resources in a way that brings benefits to all and establishing a regional operation plan that integrates a hydrological forecast model and dynamic operation rules are necessary. It is better to recommend managing water resources as a single system by establishing more effective institutional arrangements than those currently existing. It is to be hoped that such arrangements will be devised through the protracted negotiations currently under way.
Conflict of interest None. 13
Journal of Hydrology: Regional Studies 16 (2018) 1–14
A.G. Mulat et al.
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.ejrh.2018.02. 006. References BCEOM, 1998. Abbay River Basin Integrated Development Master Plan Project, Phase 2, Volume III- Water Resources, Part 1, 2- Climatology, Hydrology. Ministry of Water resources, The Federal Democratic Republic of Ethiopia, Addis Ababa. BCEOM, 1999. Abbay River Basin Integrated Master Plan, Main Report. Ministry of Water Resources, Addis Ababa. Coyne ET BELLIER and TRACTEBEL Engineering, 2011. Grand Ethiopian Renaissance Dam Project: Hydraulic and Reservoir Simulation Studies. Ethiopian Electric Power Corporation (EEPCO), 2006. Beles Multipurpose Level 1 Design Report. Ethiopian Electric Power Corporation. Ethiopia, 2000. Country paper of the federal democratic republic of Ethiopia, In: proceedings, comprehensive water resources development of the nile basin: priorities for the new century. In: VIII Nile 2002 Conference. June 26–30, Addis Ababa. pp. 54–56. Mageed, Y.A., 1994. The Nile Basin: lessons from the past. In: Biswas, A.K. (Ed.), International Waters of the Middle East, from Euphrates-tigris to Nile, Water Resources Management Series. Oxford University Press. Mulat, A.G., Moges, S.A., 2014. Assessment of the impact of the Grand Ethiopian Renaissance Dam on the performance of the high Aswan Dam. J. Water Resour. Prot. 6, 583–598. Norplan-Norconsult International, 2006. Karadobi Multipurpose Project Pre-Feasibility Study. Ministry of Water Resources, Federal Democratic Republic of Ethiopia. Shahin, M., 1985. Hydrology of the Nile Basin. Elsevier, Amsterdam, pp. 575. Shahin, M., 2002. Hydrology and Water Resources of Africa, Water Science and Technology Library. Kluwer Academic Publishers, Dordrecht/Boston/, London. Sutcliffe, Park, 1999. The Hydrology of the Nile, IAHS Special Publication No. 5. IAHS Press, Institute of Hydrology, Wallingford, Oxford shire OX10 8BB, UK. Swain, A., 2002. The Nile River Basin initiative: too many cooks, too little broth. SAIS Rev. 22 (2), 293–308. United States Bureau of Reclamation (USBR), 1964. Land and Water Resources of the Blue Nile Basin, ” Main Report. United States Department of Interior Bureau of Reclamation, Washington, DC, USA. Waterbury, J., 2002. The Nile Basin: National Determinants of Collective Action. Yale Univ. Press, New Haven.
14