Sustain. Water Resour. Manag.

(2016) 2:55–70
DOI 10.1007/s40899-015-0041-4

ORIGINAL ARTICLE

WEAP modeling of surface water resources allocation in Didessa

Sub-Basin, West Ethiopia
Tena Bekele Adgolign1 • G. V. R. Srinivasa Rao1 • Yerramsetty Abbulu1

Received: 8 May 2015 / Accepted: 26 November 2015 / Published online: 14 December 2015
Ó Springer International Publishing Switzerland 2015

Abstract The extensive water resources development of scenarios (2050), there will be a 1.101 billion cubic
plan that Ethiopia has recently embarked on encompasses meters (BCM), i.e., 10.3 %, reduction in the total annual
all major river basins, including the Abbay Basin where flow of Didessa River. Besides, it was found out that some
Didessa Sub-basin is situated. The main objective of the of the schemes in Anger, Dabena and Upper Didessa
study was to assess the impact of water resources devel- watersheds will have unmet demands at the last year of
opment on instream and downstream water availability, scenarios. Nonetheless, it was shown that IFR of 5.46 BCM
and identify intra-sub-basin locations vulnerable to short- will be fully delivered at the outlet of Didessa River; hence,
age of surface water, in Didessa Sub-basin. The Water no sub-basin wide absolute scarcity of water would
Evaluation And Planning (WEAP) Tool was used to model develop.
the surface water resources allocation of the sub-basin,
based on the water demand data extracted mainly from the Keywords Demand site  Didessa sub-basin  IFR 
Abbay Basin integrated development master plan. Existing Streamflow  Supply source  Unmet demand  WEAP
demand sites were collected from irrigation and water
supply sectors of the government. To estimate the crop
water requirement and instream flows requirement (IFR), Background
respectively CROPWAT 8.0 and Indicators of Hydrologic
Alteration (IHA) programs were used. The shortage of Ethiopia has embarked on extensive water resources
recorded data of streamflow was supplemented by the development plan since few years back. Though the
outputs of SWAT hydrological model. Three development development activities encompass all major river basins of
scenarios, viz., current development (2014), medium-term the country, the huge agricultural and hydroelectric power
future development (2015–2030) and long-term future potentials in the Abbay (Upper Blue Nile) Basin have
development (2031–2050), were built. It was demonstrated attracted considerable attention. Hence, there are currently
that the model sufficiently represented surface water flow a number of water resources development projects in the
of the sub-basin. The results indicated that at the last year construction and planning phases in Didessa Sub-basin of
the Abbay Basin.
A water-stressed situation in a watershed does not
& Tena Bekele Adgolign occur instantaneously; rather, it is a phenomenon which
[email protected] develops through time. It has been a common practice to
G. V. R. Srinivasa Rao evaluate water crises after symptoms of water scarcity
[email protected] have begun to manifest themselves. It is, of course, very
Yerramsetty Abbulu useful to make water allocation analysis even after water
[email protected] shortage has manifested itself, as it would lead to
1 seeking a win–win situation among different water users
Department of Civil Engineering, College of Enginering (A),
Andhra University, Visakhapatnam 530003, Andhra Pradesh, (demand sites) in the watershed. However, assessing the
India overall water resources potentials and the existing and

123
56 Sustain. Water Resour. Manag. (2016) 2:55–70

planned demand centers in the basin ahead of time Dataset and data sources for Didessa Sub-basin
would help in limiting developments only to the carrying WEAP model
capacity of the resource, while considering the sustain-
ability issues which need to address the right of the The major data types that are used in this study are
future generation to make their lives from the resource. hydrological time series data, water demand data, water
The main objective of this study is to assess the supply schemes and their corresponding geographical
impact of additional water demand and surface water locations, GIS vector map of Didessa Sub-basin and
resources development in Didessa Sub-basin on down- Didessa river system. Existing and planned projects have
stream and instream water availability, using the Water been included in the model.
Evaluation And Planning (WEAP) system. Besides, McCartney and Girma (2012) witnessed the lack of
vulnerable locations to shortage of water within the information on water resources and the implications of
sub-basin due to water abstraction and use will be different investment options had been a major challenge in
identified. the Abbay Basin. The condition in Didessa Sub-basin is
even worse. Due to the lack of sufficient hydrological data
(shortage of streamflow monitoring stations) at important
Materials and methods points in the sub-basin, it was not possible to use recorded
flow data in this model. Hence, an output of the SWAT
Location of the study area hydrological model of the sub-basin, as suggested by (The
Nature Conservancy 2009), has been used.
Didessa Sub-basin forms the southwestern protruding Table 1 summarizes the dataset used in WEAP model-
portion of Abbay Basin in West Ethiopia. The study area is ing of the study area, and their corresponding sources.
situated between 07°420 4000 –09°580 1700 North latitude and
35°330 1400 –37°070 5200 East longitude. The total area of the Demand sites and supply sources
watershed is about 28,092 km2. The major water resources
development projects and demand sites to be input to the Demand sites
analysis are distributed on the main stem of Didessa River
and along the four major tributaries, namely Anger, The demand sites and their corresponding annual activity
Dabena Wama and Upper Didessa rivers. Figure 1 shows levels have been identified from the Abbay Basin inte-
location of Didessa Sub-basin. grated development master plan project document and

Fig. 1 Location map of

Didessa Sub-basin

123
Sustain. Water Resour. Manag. (2016) 2:55–70 57

Table 1 Types and sources of input data for Didessa Sub-basin WEAP model
Data type Source Scale/period Description

Mean monthly Output of SWAT model Mean of 31 years

streamflow (1982–2012)
Mean monthly Ethiopian meteorological agency Mean of 31 years
evapotranspiration (1982–2012)
Demand sites and Ministry of Water irrigation and energy 1998 Abbay Basin integrated
supply sources Oromia irrigation development authority development master plan
Nekemte and Bedele towns water supply and Existing irrigation development
sewerage authorities schemes
Mean daily water supply

Table 2 Summary of existing

Duration Irrigable area (ha) Total
and planned irrigation demand
sites Small-scale Medium scale Large scale

Base year (1998) 3478 3478

Current development (2014) 9196 9196
Medium-term future development (2015–2030) 22,241 1144 109,283 132,668
Long-term future development (2031–2050) 41,225 5227 143,418 189,870

Oromia Irrigation Development Authority (OIDA). The

major demand sites in the sub-basin are irrigation, domestic
water and hydroelectric power demand sites. The irrigation
schemes identified in the master plan amount to about
79,180 ha (MWR 1999). According to this document,
small-scale irrigation (SSI) (less than 200 ha) potential in
the sub-basin is more than 51,849 ha. However, the final
and accurate location of the SSI schemes which would be
implemented during the master plan period was not given.
Hence, for the present study, the planned SSI schemes are
located approximately. Table 2 shows the summary of
irrigation development demand sites in the sub-basin,
which have been identified to be implemented during the
master plan period.
Besides, one hydroelectric power generating scheme,
with an installed capacity of 300 MW has been identified
on the master plan.
The demand sites are distributed all over the sub-basin.
Especially, as the areal extent of SSI schemes is small, and Fig. 2 Distribution of demand sites
they are often managed at farm households’ level, it is
common to find them distributed along the majority of the River and its major tributaries is depicted on Table 3
small streams. Figure 2 shows the distribution of demand below.
sites. Currently, there are one reservoir and one pumping
station as supply sources for domestic water supply for two
Supply sources urban areas in the sub-basin. One major dam and reservoir
project, Arjo-Didessa Dam and Reservoir Project, is under
All supply sources of the existing and planned devel- construction, while four other dam and reservoir projects
opment in Didessa Sub-basin are fed from surface water are encompassed in the medium and long-term develop-
resources of the main stem of Didessa River and its ment plans. The salient features of the supply sources are
tributaries. The mean annual outflow from Didessa portrayed on Table 4 below.

123
58 Sustain. Water Resour. Manag. (2016) 2:55–70

Table 3 Mean annual outflow

Sr. no. River Catchment area (km2) Mean annual outflow (BCM*)
of the major tributaries and the
main stem of Didessa River 1 Anger 7405 3.31
2 Dabena 3341 1.22
3 Upper Didessa 5540 1.98
4 Wama 3372 1.51
5 Didessa (main stem) 28,092 10.71
Source: Tena et al. (2015)
BCM* Billion cubic meters

Table 4 Major supply sources

Sr. no. Supply source Location Dam height Storage capacity
(m) (MCM**)
Latitude (°) Longitude (°)

1 Arjo-Didessa reservoir 8.525 36.664 47 1049.9

2 Anger reservoir 9.692 36.74 80 1102
3 Nekemte reservoir 9.431 36.499 30 71.5
4 Negesso reservoir 8.859 36.547 30 104
5 Meka reservoir 9.011 36.482 21 11
6 Bedele water supply 8.436 36.245 – Pumping
7 Lower Didessa reservoir 9.486 35.985 78 5510
Source: MWR (1999)
MCM** Million cubic meters

Annual water use rates A. Schemes located in Wama Watershed (Negesso Irri-
gation, small and medium scale irrigation schemes
I. Irrigation water demand: situated in Wama Watershed), schemes situated in
Upper Didessa Watershed (Arjo-Didessa Irrigation,
Scarcity of weather monitoring stations, and short records
small and medium scale irrigation schemes situated in
in many of the existing stations is a challenge to the use of
Upper Didessa Watershed and the upper reach of the
site specific crop water requirement estimation. Hence, the
main stem of Didessa River), and schemes situated in
existing and planned schemes in the sub-basin were clus-
Dabena Watershed are categorized under Didessa
tered into two categories depending on proximity and the
irrigation water requirement scheme. To estimate the
agro-ecological similarity of the areas. Accordingly, two
crop water requirement for this scheme, the meteoro-
conventional meteorological stations with relatively better
logical parameters of Bedele Meteorological Station is
data availability, Bedele and Anger stations, were inten-
used.
tionally selected, as they are fairly representing the two
B. Schemes located in and around Anger Watershed
agro-ecologic zoning of the existing and planned projects.
(Anger Irrigation, Nekemte Irrigation, SSI schemes
It has been assumed that Bedele station, located at an
situated in Anger Watershed and schemes situated on
altitude of 2011 m above sea level (m.a.s.l), and Anger
the lower reach of the main stem of Didessa River) are
station, located at an altitude of 1350 m.a.s.l, can, respec-
categorized under Anger irrigation water requirement
tively, represent projects located in mid altitude (Woina
scheme.
Dega) and lowlands (Kolla) agro-climatic zones. The
projects clustered under Anger Irrigation water demand The irrigation water demand has been estimated for a
scheme are found in the altitude range of about crop that requires the highest amount of water (sugarcane)
1160–1500 m.a.s.l while those clustered under Bedele as it is going to be grown widely, especially on Arjo-
(Didessa) irrigation water demand scheme are found in Didessa Irrigation Development Project, the main objective
altitude range of about 1500–1970 m.a.s.l. The other point of which is to develop sugarcane on 80,000 hectares of
that was considered as criteria for clustering was proximity land. Sugarcane is considered to estimate the crop water
of the areas to one another (adjacency of the sites). The requirement for the rest of the schemes, too, for two rea-
clusters are: sons. The first reason is that the freedom that small holder

123
Sustain. Water Resour. Manag. (2016) 2:55–70 59

farm households have to select the type of crop to be grown increasing recognition of environmental flows as vital in
would pose a challenge on determining the accurate type of maintaining healthy, productive and sustainable river and
crops grown and their corresponding areal extent; and the groundwater systems deserves the attention of all actors.
second is that the trend of organizing individual farmers as That is, the arbitration in the competition for water between
out-growers of sugarcane, that is being promoted in the different users, including the environment, should go
former sugar factories, is expected to be practiced in the beyond being the duty of certain institutions or environ-
vicinity and nearby areas to Arjo-Didessa Sugar Factory as mental activists. As physical water scarcity is said to occur
well. when there is not enough water to meet all demands,
Irrigation water demand was estimated using CROP- including environmental flows, the environment need to be
WAT version 8.0, a programme developed by the Food and allocated its fair share of water (FAO 2012). The allocation
Agricultural Organization of the United Nations. The out- of water for instream flow requirement right before the
puts of CROPWAT 8.0 (monthly irrigation requirements) symptoms of physical water scarcity like severe environ-
for the two areal categories are summarized on Table 5. mental degradation, declining groundwater, and water
allocations that favor some groups over others, manifests,
II. Domestic water demand:
would help minimize the unfavorable effects occurring as a
There are two major urban areas which have been result of shortage of water.
deriving water from the tributaries of Didessa River for Hence, environmental flows is one of the various water
domestic use. Nekemte, the capital city of East Wollega demands that need to be incorporated in water resources
Administrative Zone, is deriving water from a reservoir allocation modeling. Current aquatic ecology and empirical
constructed on Meka Stream, and Bedele Town is deriving observations suggest that a hydrological regime charac-
its water from Dabena River. The World Health Organi- terized by the full or nearly full range of natural variation is
zation (WHO) recommends about 100 l of water per person necessary to sustain the full native biodiversity and integ-
per day to ensure that most basic needs are met and few rity of aquatic ecosystems (Richter et al. 1997).
health concerns arise (WHO 2003). On the other hand the Quantifying the ecological reserve for rivers involves
water supply authorities of the two towns claim that they determining the water quantity and quality requirements
are distributing 4095 and 8400 m3/day for Bedele and that will ensure that they are sustained in a pre-determined
Nekemte, respectively. For this study, per capita water condition. This condition can vary from largely natural to
supply of 100 l per day (36.5 m3/cap/year) is used. largely modified where there is a large loss of natural
habitat, biota and basic ecosystem functioning (Hughes and
III. Instream (environmental) flows requirement:
Hannart 2003). It is, therefore, recognized that while some
Environmental flows is required in order to protect rivers are environmentally important, the requirements for
natural reserves and preserve the ecosystems. The ever socio-economic development in a water scarce country

Table 5 Summary of irrigation

Month Station
requirements for the two
stations Anger Didessa
3
mm m /ha Monthly variation (%) mm m3/ha Monthly variation (%)

January 48.7 487 6.37 96.1 961 19.66

February 73.5 735 9.61 99.1 991 20.28
March 130.8 1308 17.10 80.5 805 16.47
April 175.9 1759 23.00 43.0 430 8.80
May 151.0 1510 19.75 47.6 476 9.74
June 68.5 685 8.96 6.6 66 1.35
July 9.1 91 1.19 0 0 0
August 0.0 0 0 0 0 0
September 2.8 28 0.37 0 0 0
October 0.0 0 0 0 0 0
November 23.3 233 3.05 24.2 242 4.95
December 81.1 811 10.61 91.6 916 18.74
Total 764.7 7647 100.00 488.7 4887 100.00
Crop type: sugarcane

123
60 Sustain. Water Resour. Manag. (2016) 2:55–70

Table 6 Annual streamflow and instream flow requirements at some important reaches
Sr. no. Subbasin/Reach Name Streamflow (BCM) IFR (BCM) Remark

1 Anger-5 0.78 0.47 D/s of Anger reservoir

2 Anger-13 2.48 1.52 D/s of Nekemte reservoir
3 Anger-7 3.31 2.13 Outlet of Anger river
4 Dabena-24 1.22 0.73 Outlet of Dabena river
5 Upper Didessa-23 1.98 0.90 D/s of Arjo-Didessa reservoir; outlet of Upper Didessa river
6 Wama-17 0.36 0.18 D/s of Negesso reservoir
7 Wama-21 1.51 0.97 Outlet of Wama river
8 Didessa-Nr. Arjo 3.90 2.41 Gauging station
9 Didessa-15 6.41 3.33 D/s of Dabena’s confluence with Didessa
10 Didessa-9 6.63 3.57 U/s of Anger’s confluence with Didessa
11 Didessa Main-1 10.71 5.46 Outlet of Didessa main
Source for streamflow: Tena et al. (2015)

suggest that not all rivers can be retained in a near natural catchment has subsided, the river returns to its base- or
state. It is also recognized that resource development low-flow level. These low-flow levels are sustained by
should be environmentally sustainable and therefore all groundwater discharge into the river. The seasonally-
rivers should retain some basic ecological functioning. varying low-flow levels in a river impose a fundamental
According to Smakhtin et al. (2006), the desktop envi- constraint on a river’s aquatic communities because it
ronmental flows assessment (EFA) methods are more determines the amount of aquatic habitat available for most
suitable for initial, reconnaissance-level assessments of of the year. This has a strong influence on the diversity and
environmental flows requirements in unregulated river number of organisms that can live in the river (The Nature
basins and/or river basins where the pressure on water Conservancy 2009).
resources is not yet extreme, but starting to grow. Didessa Thus, the low flows parameter of the IHA’s environ-
Sub-basin falls under this category of rivers; hence, the mental flows component satisfies the environmental water
method is appropriate at this level to make initial estimates demand (instream flow requirements) of rivers. Hence, for
of EFA. Notwithstanding the importance of obtaining a Didessa Sub-basin, the instream flow requirements at the
well refined environmental flows estimates based on eco- downstream of the major reservoirs, at the outlet of the four
logical and livelihood studies, data scarcity and the fact major tributaries and the outlet of Didessa River have been
that hydroecological surveys have rarely been conducted in determined using the IHA. The mean annual outflow and
the sub-basin would compel one to use the desktop model, instream flow requirements for some important reaches are
at least at this initial stage. shown on Table 6.
The Indicators of Hydrologic Alteration (IHA) calcu- Figure 3 shows location of points where IFR are
lates parameters for five different types of environment determined.
flow components (EFCs): low flows, extreme low flows,
high flow pulses, small floods, and large floods. This The WEAP model
delineation of EFCs is based on the realization by research
ecologists that river hydrographs can be divided into a Background
repeating set of hydrographic patterns that are ecologically
relevant. It is the full spectrum of flow conditions repre- The Water Evaluation And Planning version 21 (WEAP21)
sented by these five types of flow events that must be integrated water resource management (IWRM) model
maintained in order to sustain riverine ecological integrity. seamlessly integrates water supplies generated through
Not only is it essential to maintain adequate flows during watershed-scale hydrologic processes with a water man-
low flow periods, but higher flows and floods and also agement model driven by water demands and environ-
extreme low flow conditions also perform important eco- mental requirements and is governed by the natural
logical functions (The Nature Conservancy 2009). watershed and physical network of reservoirs, canals, and
The low flows is the dominant flow condition in most diversions (Yates et al. 2005). WEAP is a PC based surface
rivers. In natural rivers, after a rainfall event or snowmelt and groundwater resource simulation tool, based on water
period has passed and associated surface runoff from the balance accounting principles, which can test alternative

123
Sustain. Water Resour. Manag. (2016) 2:55–70 61

In the present study, WEAP21 is used to evaluate the

implications of existing and planned water resources
development in Didessa Sub-basin, on instream and
downstream water availability.

Model setup

Modeling assumptions Regardless of the differences in

financial returns expected from each scheme, all demand sites
were designated equal priority in water provision. Hence,
recognizing the environment as a legitimate user of water, it
was also designated equal priority to the other demand sites.
With the exception of Arjo-Didessa Dam which is currently
under construction, and is expected to be commissioned in
2015, all major dams listed in the Abbay Basin integrated
development master plan corresponding to Didessa Sub-basin
are at the planning stage. Hence, operating rule curves are
currently not available for anyone of them, and none is
incorporated in the WEAP model of the sub-basin.
The other important assumption to be noted is that the
annual water use rate for agriculture (irrigation) is esti-
mated for sugarcane, a crop which will be widely grown,
and which has, relatively, higher crop water demand.

Development scenarios
Fig. 3 Location of WEAP IFR points
WIKIPEDIA-The Free Encyclopedia defines Scenario
sets of conditions of both supply and demand. The change analysis as ‘‘a process of analyzing possible future events
in water demand, supply and pollution can be projected by considering alternative possible outcomes (sometimes
over a long-term planning horizon to develop adaptive called ‘‘alternative worlds’’)’’ (WIKIPEDIA 2015).
management strategies. Accordingly, the scenario analysis, which is a main method
WEAP has been used in several water related researches/ of projections, does not try to show one exact picture of the
projects across the world. For example WEAP was used to future. Instead, it presents consciously several alternative
determine the likely impact of a number of possible devel- future developments. Consequently, a scope of possible
opment scenarios in Lake Tana Catchment, Ethiopia, on lake future outcomes is observable.
water levels (Alemayehu et al. 2010). It was also used to In a way more relevant to water resources, scenarios are
evaluate Ethiopia’s policy of large dam construction in the defined as alternative sets of assumptions such as different
Blue Nile River basin by simulating the impact of one operating policies, costs, and factors that affect demand
downscaled midrange climate change scenario (A1B) on the such as demand management strategies, alternative supply
performance of existing and planned hydropower and irri- sources and hydrologic assumptions, with changes in these
gation schemes (McCartney and Girma 2012). data able to grow or decline at varying rates over the
WEAP is widely used in other parts of the world, too. planning horizon of the study. Thus, scenarios are not
For example, Arranz and McCartney (2007) applied created from the minds of the requirements engineers; their
WEAP Model to assess future water demands and resour- definition should be anchored on real situations (Leite et al.
ces in the Olifants Catchment, South Africa. Mounir et al. 2000). The typical characteristic of this (scenario analysis)
(2011) also conducted a study using WEAP model to method is that it can model many real problems where
investigate scenarios of future water resource development decisions are based on uncertain information presented as a
in the Niger River Basin in Niger Republic. Similarly, it set of possible outcomes (i.e., scenarios) (Weng et al.
was used to assess the potential effects of climate change 2010).
and adaptive management on irrigation water supply in the The Abbay River Basin integrated development master
Cache Creek watershed in California (Mehta et al. 2013). plan MWR (1999) has been the basis to build development
Hence, WEAP model has high global popularity in sce- scenarios for the present study. The master plan was for-
nario analyses of water demands and supplies. mulated in 1998, for a time horizon of 50 years. Hence, the

123
62 Sustain. Water Resour. Manag. (2016) 2:55–70

years 1998 and 2050 are, respectively, considered as the schedules as long as they are not preceded by a project
current accounts year and last year of scenarios. whose start up period has passed.
Assuming different growth rates, two scenarios were Three development scenarios, namely reference (current
formulated on the Abbay Basin integrated development development) (1999–2014), medium-term future develop-
master plan, as to when to implement the planned projects: ment (2015–2030) and long-term future development
conservative scenario and accelerated scenario (MWR (2031–2050) scenarios have been proposed.
1999). The development periods were classified as Regarding the developments which were undertaken
1999–2014, 2015–2030 and 2031–2050. However, it was during the reference scenario time horizon, additional
observed that most of the schemes proposed on the master inputs from irrigation development and water supply sec-
plan have not yet been constructed except a few SSI tors have been incorporated. As McCartney and Girma
schemes. Hence, the existing developments are directly (2012) observed, Ethiopia has utilized very little of the
translated into base year and current accounts scenario, and Blue Nile water due to physical, infrastructural, socio-
a minor adjustment has been made on the implementation economic, etc., constraints. This is particularly true when
timing of the planned developments without significant Didessa River is considered. So far, only few SSI schemes
change in the current accounts year, intermediate years and and no hydropower development exist in Didessa Sub-
last year of scenarios. Major projects whose start up years basin. Table 7 portrays the major demand sites and their
have not yet elapsed are maintained as per their initial corresponding annual activity levels in the sub-basin.

Table 7 Scenario building

Sr. Description Scenarios
no.
Base Year (1998) Reference (current Medium-term future Long-term future
development) development development
(1999–2014) (2015–2030) (2031–2050)

A Domestic water demand (MCM/year at the end of scenario period)

A.1 Nekemte town domestic 1.93 3.67 6.97 15.57
water demand
A.2 Bedele town 0.50 0.95 1.81 4.04
domestic water
demand
B Agricultural (irrigation) activity levels (ha)
B.1 EW-SSI 3161 8079 19,903 37,951
B.2 WW-SSI 80 80 80 80
B.3 IAB-SSI 201 401 822 822
B.4 Jima-SSI 36 636 1436 2372
B.5 Arjo-Didessa irrigation – – 80,000 80,000
B.6 Anger irrigation – – 8000 14,450
B.7 Nekemte irrigation – – 6000 11,220
B.8 Negesso irrigation – – 5000 21,315
B.9 Hida irrigation – – 5650 5650
B.10 Wama irrigation – – 325 325
B.11 Lugo irrigation – – 285 285
B.12 Dimtu and Didiga irrigation – – 5167 5167
B.13 Upper Dabena irrigation – – – 1150
B.14 Dabena irrigation – – – 6150
B.15 Urgessa irrigation – – – 2933
C Hydropower activity level (GWH)
C.1 Lower Didessa hydropower – – – 1582
(300 MW installed
capacity)
D Environmental flows H H H H
requirement
Data source: MWR (1999), OIDA (2014)

123
Sustain. Water Resour. Manag. (2016) 2:55–70 63

Schematic of the study area Results and discussion

A schematic, or schematic diagram, is a representation of The WEAP model performs a mass balance of flow
the elements of a system using abstract, graphic symbols sequentially down a river system, making allowances for
rather than realistic pictures. A schematic usually omits all abstractions/water use and inflows. The elements that com-
details that are not relevant to the information the sche- prise the water demand–supply system and their spatial
matic is intended to convey, and may add unrealistic ele- relationship are characterized within the model. The WEAP
ments that aid comprehension (WIKIPEDIA 2015). The model for Didessa River system was setup to simulate the
Schematic View is the starting point for all activities in base year (1998) situation and three subsequent scenarios:
WEAP (SEI 2011). the Reference (current) situation (1999–2014), the Medium-
WEAP allows the display of GIS layers as overlays or term future development (2015–2030), and the long-term
backgrounds on the WEAP Schematic. Hence, Didessa future development (2031–2050) scenarios. The basis for the
Sub-basin vector layer, which includes Didessa River timelines of these scenarios is the Abbay Basin integrated
system, has been added. An image of the WEAP schematic development master plan of 1998.
is shown below (Fig. 4). For each scenario, the main outputs analyzed include the
The existing and planned development in the sub-basin extent of satisfying the water demand of different sectors,
during the three development scenarios have been sum- the degree to which the instream flow requirements is
marized on the following figures (Fig. 5). satisfied, and the spatial and temporal variation in water
The monthly headflows data were obtained from the shortages, if any. Due consideration was rendered to
SWAT Hydrological Model of the Sub-basin (Tena et al. determining the streamflow at the catchment outlets and the
2015) and entered into the headflows environment for each outlet of the main stem of Didessa River so as to determine
stream. Similarly, the data for all demand sites, supply and the cumulative impact of additional upstream water
resources, key assumptions, etc. were entered. abstraction on instream and downstream water availability.

Fig. 4 Screenshot of the schematic of Didessa Sub-basin WEAP model

123
64 Sustain. Water Resour. Manag. (2016) 2:55–70

Fig. 5 Line representation (schematic) of the model for different development scenarios

123
Sustain. Water Resour. Manag. (2016) 2:55–70 65

Water demand, streamflow and instream flow Energy. Weather data from selected weather monitoring
requirements (IFR) stations were used to setup the model.
The model calibration and validation were performed
The total annual water demand under the three develop- using the sequential uncertainty fitting (SUFI-2) algorithm
ment scenarios is depicted on Table 8 below. of SWAT calibration and uncertainty programs (SWAT-
The annual streamflow of the main stem of Didessa CUP). The model predicted monthly discharge with high
River and the four major tributaries as per the WEAP accuracy, with Nash Sutcliff Efficiency (NSE) values of
model of the sub-basin, for the three development scenarios 0.87 and 0.80 for the calibration (1982–1986) and valida-
are depicted below (Fig. 6). tion (1992–1996) periods, respectively; and coefficient of
From the SWAT hydrologic model of the sub-basin, the determination (R2) values of 0.87 and 0.80 for the cali-
mean annual flowout at Didessa-Abbay confluence is 10.71 bration and validation periods, respectively. The calibra-
BCM (Tena et al. 2015). This value is sufficiently com- tion and validation of the model were performed for a
parable with WEAP’s 10.63 BCM for the reference sce- measured flow data at Didessa Near Arjo gauging station,
nario. The SWAT hydrologic modeling of Didessa Sub- which is located at 8°40 N latitude and 36°250 E longitude.
basin involved setting up the model using a DEM of 20 m The following chart (Fig. 7) depicts the temporal and
horizontal resolution, land use/cover and soils maps spatial distribution of streamflow at the end of the long-
obtained from the Ministry of Water, Irrigation and term future development scenario. The streamflow for the

Table 8 Total annual water

Scenario Demand (MCM) % of the mean annual streamflow
demand under the three
development scenarios Reference 74.72 0.70
Medium-term Future 746.22 6.97
Long-term Future 1199.33 11.20

Fig. 6 Annual streamflow Mean annual stremflow of Didessa River and the major tributaries
under each scenario (output of 12.000
Annual streamflow (BCM)

WEAP model) 10.000

8.000
6.000
4.000
2.000
0.000
Anger Dabena Upper Wama Didessa
Scenario Didessa Main
Reference 3.243 1.219 1.977 1.487 10.632
Medium-term Future
3.049 1.218 1.537 1.474 10.016
Development
Long-term Future Development 2.928 1.189 1.533 1.430 9.609
River

Fig. 7 Monthly streamflow at Mean monthly stream flow (BCM)

the end of Long-term future 0.80 2.60
Streamflow for Didessa Main (BCM)

development scenario
2.10
0.60
1.60
Streamflow (BCM)

0.40 1.10

0.60
0.20
0.10

0.00 -0.40
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Month
Anger Dabena Wama Upper Didessa Didessa Main Natural (SWAT output)

123
66 Sustain. Water Resour. Manag. (2016) 2:55–70

Monthly Instream Flow Requirement (MCM) the urban, agricultural and environmental water needs of the

1,154
1500 currently existing demand sites (Bedele Town Water Supply)

1,045
IFR (MCM)

1000 and the plan to be implemented until 2050.

818
651

672
500

245
102
57

448
40 Change in streamflow and water consumption
56
0

174
The existing and planned development will cause some
Month
changes in river flow of the tributaries and the main stem of
Didessa River. Table 10 summarizes streamflow for the
Fig. 8 Monthly IFR for Didessa River at Didessa-Abbay confluence current development and the two future development
(2050) scenarios.
Table 10 portrays the mean annual streamflow across
main stem of Didessa River according to SWAT Hydro- scenarios. It shows that at the end of the long-term future
logic Model (‘natural’) has been added for comparison. In development scenario, the annual flow of Didessa River
WEAP, natural flow is meant to indicate the river flow will decrease to 9.609 BCM. This shows that the water
without any development. Besides, the outputs of the consumed by demand sites will be 1.101 BCM (10.3 % of
SWAT model for the major tributaries and main stem of the mean annual flow). Table 10 shows the water con-
Didessa River at its outlet have been considered as natural sumption details for the main stem and the four major
flow, as there is no flow monitoring station towards the tributaries of Didessa River.
downstream ends of the rivers to get recoded flow data and
make analyses of water consumption and instream flow Spatial and temporal occurrence of water shortage/
requirement. The word natural has been put in single Unmet demands under each scenario
quotation mark (‘natural’) to make a distinction that the
flow is not a recorded natural flow. Spatial (Watershed-wise) occurrence of unmet demand
The streamflow data for the tributaries is also reasonably
comparable to those obtained from the SWAT hydrologic The spatial occurrence of unmet demands across the
model. watersheds is shown bellow (Table 11).
The monthly Instream Flow Requirement for the main As shown above, there is unmet demand in all water-
stem of Didessa River at the end of the long-term future sheds particularly for the medium-term and long-term
development scenario is depicted below (Fig. 8). The future development scenarios, although there is variation in
chart shows that the Instream Flow Requirement varies in a the extent of shortage. In this regard, Wama Watershed is
similar trend with the monthly streamflow. in a better condition.
Table 9 shows streamflow, IFR, and IFR delivered at some
important reaches of the main stem of Didessa River and its Spatial (Demand site-wise) occurrence of unmet demand
tributaries. It can be observed from the table that except
Dabena River, the IFR has been satisfied in the mentioned Demand sites with unmet water demand and their corre-
reaches. It seems that the flow of Dabena River cannot satisfy sponding unmet demands are portrayed on Fig. 9 below.

Table 9 Annual streamflow and instream flow requirements for some important reaches
Sr. no. Subbasin/reach name Streamflow (BCM) IFR (BCM) IFR delivered (BCM) Remarks

1 Anger-5 0.78 0.47 0.47

2 Anger-13 2.48 1.52 1.52
3 Anger-7 3.31 2.13 2.13 Outlet of Anger R.
4 Dabena-24 1.22 0.724 0.703 Unmet IFR = 0.021
5 Upper Didessa-23 1.98 0.90 0.90 Outlet of Wama R.
6 Wama-17 0.36 0.177 0.177
7 Wama-21 1.51 0.975 0.975 Outlet of U/Didessa R.
8 Didessa- Nr. Arjo 3.90 2.405 2.405
9 Didessa-15 6.41 3.568 3.568
10 Didessa-9 6.63 3.326 3.326
11 Didessa Main-1 10.71 5.463 5.463 Outlet of Didessa R.

123
Sustain. Water Resour. Manag. (2016) 2:55–70 67

Table 10 Changes in river flow under each scenario for the major tributaries and the main stem of Didessa River
Sr. River ‘Natural’ (output of SWAT Annual stream flow under each scenario (BCM)
no. model)
Current Medium-term future Long-term future
development development development

a. Mean annual streamflow across scenarios

1 Anger 3.31 3.243 3.049 2.928
2 Dabena 1.22 1.219 1.218 1.189
3 Wama 1.51 1.487 1.474 1.430
4 Upper 1.98 1.977 1.537 1.533
Didessa
5 Didessa 10.71 10.632 10.016 9.609
Main
Sr. River ‘Natural’ (output of SWAT Annual water consumed under each scenario (BCM)
no. model)
Current % Medium-term future % Long-term future %
development development development

b. Change (decrease) in annual flow across scenarios

1 Anger 3.31 0.07 2.0 0.26 7.9 0.38 11.5
2 Dabena 1.22 0.001 0.1 0.002 0.2 0.031 2.5
3 Wama 1.51 0.023 1.5 0.036 2.4 0.080 5.30
4 Upper 1.98 0.003 0.2 0.443 22.4 0.447 22.6
Didessa
5 Didessa 10.71 0.078 0.7 0.694 6.5 1.101 10.3
Main

The base for estimation of the consumed is taken as the ‘natural’ flow (SWAT’s output)

Table 11 Spatial occurrence of unmet demands across watersheds

Sr. no. Watershed/reach name Scenarios Volume of water (MCM)
Demand Supply delivered Unmet demand

1 Anger Reference 55.54 50.29 5.25

Medium-term future 254.58 235.20 19.38
Long-term future 450.77 410.18 40.59
2 Dabena Reference 0.95 0.93 0.02
Medium-term future 1.81 1.81 0.00
Long-term future 39.71 34.13 5.58
3 Wama Reference 8.12 7.76 0.36
Medium-term future 47.92 47.92 0.00
Long-term future 165.69 165.69 0.00
4 Upper Didessa Reference 3.11 2.53 0.58
Medium-term future 425.49 410.45 15.04
Long-term future 430.06 414.84 15.22
5 Didessa Main (some reaches) Reference 7.00 6.56 0.44
Medium-term future 16.42 16.25 0.17
Long-term future 113.10 112.93 0.17
Total Reference 74.72 68.07 6.65
Medium-term future 746.22 711.624 34.59
Long-term future 1199.33 1137.77 61.56

123
68 Sustain. Water Resour. Manag. (2016) 2:55–70

Fig. 9 Annual unmet demand 3

18,000 Annual unmet Demand ('000 m )
per demand sites at the end of

Unmet demand ('000 m3)

each scenario period 16,000
14,000
12,000
10,000
8,000
6,000
4,000
2,000
0

Demand sites with unmet demand

Reference (Ref) Medium-term Future Development (MFD) Long-term Future Development (LFD)

The output of WEAP shows that during the Reference of the annual flow). The demand sites with unmet demand
Scenario period, the water demand of some demand sites were are SSI demand sites in East Wollega, West Wollega and
not satisfied. The major demand sites with unmet demand are Jima Zones, and medium and large scale irrigation demand
SSI demand sites in East Wollega Administrative Zone. Of sites in Dabena and Upper Didessa Watersheds. Bedele
course, there is no sub-basin wide absolute scarcity of water Town Demand Site, which derives its water from Dabena
during this scenario. As indicated in the Abbay Basin Inte- River, is also included among sites with unmet demand.
grated Development Master Plan document, the exact location
of all SSI schemes is not known; therefore, the shortage might Temporal occurrence of unmet demands
have occurred due to the approximate positioning of the
schemes. The second possible cause of the shortage is that for Table 12 shows temporal (monthly) occurrence of unmet
all irrigation demand sites, the crop water requirement was demand under the three development scenarios.
calculated for the highest crop water demand, which might not The above table shows that at the end of the long-term
have been the case in practice. The total annual unmet demand future development scenario (2050), demand will be fully
was about 6.65 MCM. met during the months of June–October only. This is typ-
During the medium-term future development Scenario, ically the main rainy season in the study area. During the
most of the demands are met. Some SSI demand sites and remaining seasons, there are varying amount of unmet
one large scale irrigation demand site have unmet demand. demand.
The annual unmet demand during the medium-term future
development scenario is 34.59 MCM. The total supply Water shortage hotspots
delivered is 711.624 MCM (6.645 % of the total annual
streamflow), against a demand of 746.22 MCM (6.97 % of Similar to the characteristics of rainfall distribution, there
the total Didessa annual flow). is variability in the spatial and temporal distribution of
At the end of the long-term future development sce- streamflow across the sub-basin. The water shortage hot-
nario, the annual unmet demand in Didessa Sub-basin spots are identified depending on the extent to which the
would reach about 61.56 MCM. The annual supply deliv- surface water resource in the corresponding catchment is
ered would be about 1137.77 MCM (10.62 % of the mean capable of meeting the demand of demand sites and the
annual flow), against a demand of 1199.33 MCM (11.20 % instream flow requirements.

Table 12 Temporal occurrence of unmet demand across watersheds (MCM)

Scenario Month Sum
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Long-term future development 16.21 14.75 10.48 7.46 3.92 0 0 0 0 0 0.10 8.65 61.56
Medium-term future development 10.90 9.37 4.09 1.49 0 0 0 0 0 0 0.10 8.65 34.59
Reference 1.87 2.02 0.09 0.07 0 0 0 0 0 0 0 2.60 6.65

123
Sustain. Water Resour. Manag. (2016) 2:55–70 69

With respect to watersheds, the water demand in Anger program developed by the Food and Agricultural Organi-
Watershed is the most unmet. All the demands in Dabena zation of the United Nations. To be on the safest side, the
Watershed, including Bedele Town Demand Site, are highest crop water demand was considered. At the end of
unmet at the end of the long-term future development the master plan period, the total irrigation annual activity
scenario. The shortage of water to meet the water demand level will reach 189,780 ha, and the corresponding annual
of Bedele Town water supply is very critical if alternative water requirement will be 1.162 BCM. The supply
water source is not available. The annual water demand of requirement for the two urban areas, Nekemte and Bedele
Bedele water supply demand site is unmet by about towns, will be 19.612 MCM per year.
2,62,000 m3. Similarly, the environmental flows (instream flow
Similarly, environmental water needs are not met in requirement) was estimated using the indicators of hydro-
Dabena Watershed. To fully meet the annual Instream logic alteration (IHA), a program developed by The nature
Flow Requirement, a deficit of 21 MCM of water is conservancy. At the outlet of Didessa River, the total
observed. Temporally, shortage of water is observed during annual IFR of about 5.46 BCM is fully met at the end of the
the months of November–March. long-term future development scenario (end of the master
In the other watersheds too, there are patch pockets plan implementation period).
where demands are not fully met. The demand of most SSI The analysis has shown that, at the end of the master
demand sites in East Wollega, West Wollega, I/A/Bora and plan implementation period (2050), the reduction in the
Jima zones (Anger, Wama and Upper Didessa Water- mean annual flow of Didessa River due to the develop-
sheds), and that of Hida Irrigation (located in Upper ment will be about 10.3 % of the total. Similarly, the
Didessa Watershed) are unmet during the medium term reduction in the annual flow of the major tributaries will
future development scenario (for some of them) and the be 11.5, 2.5, 5.3 and 22.6 %, respectively, for Anger,
long-term future development scenario (for most of them). Dabena, Wama and Upper Didessa rivers. Watershed
The instream flow requirements are met in all watersheds wise, Dabena is the only one with unmet instream flow
except Dabena. Regarding the SSI demand sites, the requirement. Despite the non-existence of universal
shortage of water might have occurred due to the possible shortage of water in Didessa Sub-basin, there are water
reason of locating some of the sites approximately. shortage hotspots in all watersheds. The temporal occur-
rence of these shortages and the unmet IFR are during the
months of November–March. This is the typical dry
Summary and conclusions season in the area, and it is the season with the highest
irrigation water requirement.
Summary
Conclusions
The main objective of the study was to assess the possible
impacts of upstream surface water abstraction in Didessa As part and parcel of the endeavors to assess the sustain-
Sub-basin on downstream water availability and environ- ability and environmental integrity of the existing and
mental flows. The water evaluation and planning (WEAP) planned water resources development in Didessa Sub-
Tool was used to evaluate the impacts. The three major basin, the water evaluation and planning (WEAP) model
categories of demand sites encompassed in the analysis has shown a clue on what would possibly happen after the
were the irrigation demand sites, domestic (urban) water planned water resources development is fully implemented.
demand sites and hydroelectric power demand site. The It has been shown that at the end of the long-term future
annual activity levels for the demand sites were collected development scenario period (2050), some watersheds in
primarily from the Abbay Basin integrated development Didessa Sub-basin will fall under a water shortage
master plan of 1998. Besides, recent data on irrigation situation.
development and water supply have been collected from Overall, the impact of the existing and planned water
the irrigation development and water supply sectors of the resources development on delivery of instream flow
government. requirements and downstream water availability is mini-
During the model setup, all demand sites, including the mal. By the end of 2050, the reduction in the total annual
environmental flows, were assigned the same priority. The flow of Didessa River due to the existing and planned water
basic reason behind is that the environment is legitimate resources development would be about 1.101 BCM
user of water and, therefore, should be allocated its fair (10.3 % of the total annual flow). A watershed which
share of the resource. should be rendered special focus is Dabena Watershed,
The agricultural (irrigation) water demand (crop water where the water demand of all demand sites, including the
requirement) was estimated using CROPWAT 8.0, a IFR, would not be met at the end of the long-term future

123
70 Sustain. Water Resour. Manag. (2016) 2:55–70

development scenario. If the water resources development References

plans in this watershed need to be implemented without
any major detrimental effect on the environment, it might Alemayehu T, McCartney M, Kebede S (2010) The water resource
implications of planned development in the Lake Tana catch-
be necessary to make arrangements to provide storage
ment. Ethiopia Ecohydrol Hydrol 10(2–4):211–222
facilities that would minimize the ill-effects of the devel- Arranz R, McCartney M (2007) Application of the water evaluation
opment endeavors. and planning (WEAP) model to assess future water demands and
Regarding the irrigation development demand sites resources in the olifants catchment, South Africa. IWMI
Working Paper 116. Colombo, Sri Lanka: Internation Water
whose demands are not met, the watersheds where they are
Management Institute
situated are, generally, not in a water shortage situation. FAO (2012) Coping with water scarcity-An action framework for
Therefore, the problem might be resolved by locating the agriculture and food security. Food and Agricultural Organiza-
schemes based on proper feasibility analyses. tion of The United Nations, Rome
Hughes DA, Hannart P (2003) A desktop model used to provide an
Moreover, this study demonstrates that the problems
initial estimate of the ecological instream flow requirements of
associated with lack of information on water resources, rivers in South Africa. J Hydrol 270:167–181
especially recorded hydrological data, can be tackled by Leite JC, Hadad GD, Doorn JH, Kaplan GN (2000) A scenario
supplementing the existing data from few conventional construction process. Requirements Eng 5:38–61
McCartney MP, Girma MM (2012) Evaluating the downstream
gauging stations with the outputs of hydrological models
implications of planned water resource development in the
like SWAT, and using in water allocation models. Ethiopian portion of the Blue Nile River. Water Int
This analysis was made based on data and information 37(4):362–379. doi:10.1080/02508060.2012.706384
gathered from the master plan project document, which is Mehta VK, Haden VR, Joyce BA, Purkey DR, Jackson LE (2013)
Irrigation demand and supply, given projections of climate and
crude by its nature, as it was prepared for a large river
land-use change, in Yolo County, California. Agric Water
basin, where there is the possibility of overlooking things Manag 117:70–82
that were happening at sub-basin or small watershed scale. Mounir ZM, Ma CM, Amadou I (2011) Application of water
Hence, some of the parameters that were input in the evaluation and planning (WEAP): a model to assess future water
demands in the Niger River (In Niger Republic). Modern Appl
WEAP model were fixed based on assumptions. Nonethe-
Sci 5(1):38–49
less, the results of this study are indicative of what may MWR (1999) Abbay River Basin master plan project—phase 3—
happen and what need to be done; and the data generated main report. Ministry of Water Resources, Addis Ababa
here might be a springboard for further study on issues Richter BD, Baumgartner JV, Wigington R (1997) How much water
does a river need? Freshw Biol 37:231–249
related to sustainable water resources planning and devel-
SEI (2011) WEAP water evaluation and planning system user guide.
opment in the sub-basin. Stockholm Environ Institute, Somerville
The study was conducted on the basis of the Abbay Smakhtin VU, Shilpakar RL, Hughes DA (2006) Hydrology-based
Basin integrated development master plan, where ground- assessment of environmental flows: an example from Nepal.
Hydrol Sci 2:207–222
water was not considered as an alternative supply source, in
Tena BA, Srinivasa Rao GVR, Abbulu Y (2015) Assessment of
order to evaluate the impact of the existing and planned spatio-temporal occurrence of water resources in Didessa Sub-
development on instream and downstream water avail- basin, West Ethiopia. IJCSEIERD 5:105–120
ability. It was considered that the assumptions on which the The Nature Conservancy (2009) Indicators of hydrologic alteration
version 7.1 User’s Manual. User’s Manual. The Nature
master plan study was based will prevail, as this is one
Conservancy
possible scenario. Therefore, it is recommended that fur- Weng S, Huang G, Li Y (2010) An integrated scenario-based multi-
ther study addressing potential climate and land use/cover criteria decision support system for water resources management
change scenarios, and groundwater resources as alternative and planning—a case study in the Haihe River Basin. Expert
Syst Appl 37:8242–8254
supply sources be conducted in the future. Besides, it is
WHO (2003) Right to water. Retrieved from Office of the United
recommended that site specific crop water demand be Nations High Commissioner for Human Rights: http://www2.
estimated based on surveys to perform more refined water ohchr.org/english/issues/water/docs/Right_to_Water.pdf
resources allocation modeling. WIKIPEDIA (2015) Scenario_analysis. Retrieved from WIKIPEDIA
The Free Encyclopedia: http://en.wikipedia.org/wiki/Scenario_
analysis
Acknowledgments The authors are grateful to the editor of the
Yates D, Sieber J, Purkey D, Lee (2005) WEAP21—a demand-,
journal and the anonymous reviewer/s for their relevant and valuable
priority-, and preference-driven water planning model-part 1:
comments on earlier version of the manuscript due to which the
model characterisitics. Int Water Res Assoc 30(4):487–500
contents of this paper have been enriched.

123