Abstract

Climate change results in precipitation variation in rivers and other water bodies, resulting in a decline in reservoir inflow and hydropower generation. This study is aimed at the assessment of climate change under different environments for hydropower generation and optimal reservoir operation using updated representative pathways (RCPs) in the Koka reservoir of Ethiopia. The power transformation function and variance scaling techniques of numerous meteorological data were adopted for bias correction. The simulation of flow was performed using HEC-HMS and the generation of hydropower from the reservoir was estimated using MODSIM 8.1 under different climatic scenarios. NSE and R² (90.72 and 0.70) were calculated based on the model performance features for calibration and validation, respectively. There was a remarkable anomaly in the pattern of precipitation and temperature based on the projection of future climate scenarios. This change ultimately affects the power generation with an apprehension of reduction of 0.54% and 0.72% in the near term (2021–2050) and long term (2051–2080) under RCP4.5. Similarly, the mean generation of energy will reduce by 1.04% in the short term and 1.32% in the long-term for RCP8.5. This shows that the reduction is more prevalent in RCP8.5 in comparison to RCP4.5. Simply, it can be concluded that if strict measures are not timely initiated, there will be an acute power shortage. The research findings warn the concerned bodies to take timely measures to reduce recurring sediment deposits and ensure future hydropower output.

1 Introduction

The hydrologic cycle and reservoir water availability are two of the recognized effects of climate change on the use of water resources systems (Abera et al. 2018; Alazard et al. 2015; Alemseged and Rientjes 2015). Reductions in reservoir storage capacity and an increase in stream flow frequency also express changes in the hydrologic system. The nation's economic operations are subsequently impacted by hydrologic hazards and water disasters (Alemu and Dioha 2020).

Electricity generated by hydropower projects accounts for 16% of the world’s renewable sources (Anand et al. 2018). Micro-hydropower plants produce energy from runoff; rivers have the potential to produce 200 gigawatts (GW) of hydroelectricity across the globe (Awulachew et al. 2007). However, global warming-related climatic variability affects the world's precipitation, which articulates hydropower plants at risk (Berhe et al. 2013; Bekele et al. 2019). Water resources have an impact on climate and tend to be a major concern for both industrialized and developing nations (Berhe et al. 2013).

The primary reasons behind a nation’s economic growth are based on the development of their water resources and hydroelectric power production systems (Bruce 1967). Africa has a total hydropower production capacity of around 70 gigawatts (GW), but over 25% of its hydroelectric power facilities are inoperative because of inadequate maintenance (Chaleeraktrakoon and Chinsomboon 2015). The existing hydroelectric capacity in the sub-Saharan Africa (SSA) nations is 27 GW, and 15 GW hydropower plants are now under completion (Chiang et al. 2013), out of which grids supply 45% of the electricity produced, accounting for more than 50% of the total electricity generation (Falchetta et al. 2019; Gentilucci et al. 2021).

Ethiopia is a developing nation with abundant water resources that can produce 45,000 megawatts (MW) of hydropower. In 2017, the nation's ability to produce electricity increased from 850 MW to 4300 MW (Goswami et al. 2015; Guangli et al. 2021). Presently, demand is steadily rising, leading to frequent load shedding (Gunathilake et al. 2020). Furthermore, the government is making whopping efforts to limit the nation's 2030 greenhouse gas (GHG) emissions to 150 Mt CO₂ (Hunt et al. 2020). Conventional and unsustainable methods of exploiting natural resources are most likely to be blamed for the emissions (Huntington 2006).

Ethiopia's hydropower potential and generation are the backbone for the nation's economic development (INDC 2021). One of the most crucial prerequisites is the ability of the reservoir to operate at optimum efficiency (Khan et al. 2006). Climate change is causing changes in temperature, precipitation, and stream flow (Khaniya et al. 2019; Khaniya et al. 2020). These changes have a substantial influence on hydropower schemes and different water-related structures (Goharian and Burian 2018; Khaniya et al. 2018; Kitlasten et al. 2021; Luo et al. 2018). To maintain hydropower production and balance the system's water shortage, reservoirs are crucial (Labadie 2010; Lin and Rutten 2016; Mallick et al. 2021). Hydrological effects on reservoir operation and hydropower generation caused by climate change factors like temperature and precipitation are a matter of concern (Mamman 2018; Mekonnen et al. 2022). In addition to projecting future power generation, Ethiopia has made significant investments in hydroelectric power generation irrespective of the impact of climate change on precipitation and temperature variations on hydropower production (Meng et al. 2021). Due to the lack of structured capability and financial growth, the poorest nations are susceptible to the effects of climate change (Moriasi et al. 2007). Despite various schemes undertaken by the government, frequent climate change has a significant impact on Ethiopia's hydropower output and reservoir operations. Regarding Ethiopia's reservoir performance considering climate change, there is a great deal of uncertainty. This may result in a reduction of 50% in the production capability in the years to come (Mulat et al. 2018). Hence, the installed capacity must increase overall by 5.9% from 169 GW to 179 GW in comparison to the new policy scenario.

Many researchers have used GCMs (Global Climate Models), and RCMs (Regional Climate Models) to study the effects of climatic variation on hydropower output (Norouzi 2020; Otto and Josef 2000). One of the techniques for transferring data from GCMs to smaller scales is dynamic downscaling, which makes use of regional climate models (RCMs) with better resolution (Pilesjo and Al-Juboori 2013; Haile and Rientjes 2015; Pörtner et al. 2022; Qin et al. 2020).

The Awash basin is one of the biggest tributaries in the Koka watershed and is distinguished by a high degree of climatic change and water needs (Ranzani et al. 2018). The Koka reservoir and hydropower generation are further stressed by the fluctuating climate and the need for water for agriculture, industry, and cattle. The present study specifically highlights the risk of reduction in power production due to surface runoff instability. Water that is in the reservoir for a long period of time can be safely utilized, but there are some factors which affect the reservoir capacity like poor watershed management practice in and around the watershed, environmental impact assessment, erosion and flooding, and climate change. The impact of climate change on the water resources of the Upper Awash watershed is well articulated (Norouzi 2020; Otto and Josef 2000) but the effect of hydropower generation in the Koka reservoir has not been studied yet. Due to this, further investigation on the impacts of climate change on Koka hydropower generation has been initiated using MODSIM model and specifically this study focuses on 5 climate change impacts in the watershed, Hydropower generation was enunciated with RCP 4.5 and 8.5 scenarios and adopting reservoir operation guide curve. Therefore, it is crucial to quantify the expected consequences of climate change on hydropower generation.

The reservoir, which is expounded by the Koka dam, is primarily used for hydropower generation and downstream irrigation expansion purposes, but the hydro climatic extreme has an influence on the amount of inflow to the reservoir due to climate, poor watershed management, and flooding. Hence, this study addressed the effect of climate and river discharge variation on hydropower generation to predict the reservoir inflow and other variables under the future climate projection to optimize the hydropower production. A HEC-HMS hydrological model is used for river simulation, and MODSIM is used for hydropower generation under climate change.

2 Materials and methods

2.1 Study area

The study of climate change is important for the Koka reservoir because the reservoir is affected by serious environmental issues related to climate change. The reservoir, situated in the Awash River basin of Ethiopia, is located at 8°26′ 0″N Latitude and 39°2′0″E Longitude (Figure 1). The areal coverage is around 180 square kilometers. The reservoir can hold a maximum of 1850 million cubic metres (MCM), out of which 1650 MCM, is utilized for storage. The operation level is 1590.7 m at the highest point and 1580.7 m at its lowest point, respectively. The power plant is equipped with firm capacities of 43.2 megawatts (MW) and 34.5 MW, respectively, at working heads of 40 m and 32 m.

Figure 1 Upper Awash River basin: Sub basins, Koka dam, hydrological and metrological stations, and other detailed watershed information.

2.2 Data collection

Time series data

The daily meteorological data for the nine stations and the gauged hydrological data for the main dam site stations were acquired from the National Meteorological Agency and Ministry of Water, Irrigation, and Energy of Ethiopia, respectively (Table 1).

Table 1 Description of meteorological stations found in and around the watershed and near the Koka reservoir (Fanta et al. 2023).

S. No.	Name	Lat (Deg)	Long (Deg)	Elevation (m)
1	Addis Abeba	9.01891	38.7475	2355
2	Bulbula	7.72	38.6525	1606
3	Boneya	8.48	38.39	2879
4	Ginchi	9.01667	38.1333	2492
5	Hombole	8.368167	38.78	1860
6	Koka dam	8.471	39.157	1661
7	Sebeta	8.93	38.63	2213
8	Sendafa	9.152167	39.0215	2482
9	Tulubolo	8.658	38.211	2310

Spatial data

A 12.5 m * 12.5 m Digital Elevation Model (DEM) was taken from SRTM (Shuttle Radar Topography Mission) using the website https://asf.alaska.edu/ in October 2023, and this spatial data ID was used to delineate the catchment using ArcGIS 10.7.

Climate data

To create the representative concentration pathways (RCP4.5 and RCP8.5) conditions for the years 1951–2100, CORDEX–Africa program was used. In the fifth assessment report (Shrestha et al. 2014), the IPCC used the new RCP scenarios to represent the emission paths of RCP2.6 (low emission), RCP4.5 (middle emission), RCP6 (moderate emission), and RCP8.5 (high emission scenario) (IPCC 2014). The RCP4.5 and RCP8.5 scenarios were employed in this study to predict meteorological data with respect to the reference period (1971–2000) for both the short-term (2021–2050) and long-term (2051–2080) periods.

2.3 Data analysis

Bias correction of RCP data

Since these techniques are deemed more effective considering the validity of recorded temperature and rainfall, they were used for temperature bias correction and precipitation bias correction, respectively (Saied et al. 2013; Schaeffer et al. 2012). The variance and average of normally distributed temperature data were adjusted using the VARI method and the advantage of using this method is discussed in detail by Saied et al. (2013).

RCP data and accuracy assessment of rainfall simulations

The systematic error in rainfall quantity (Bias), root mean square error (RMSE), and coefficient of variation were used in this work to assess the precision of the rainfall simulation using RCP data (Scola et al. 2014).

Trend analysis

The effects of climate change on meteorological data in the Koka watershed have been identified using a trend analysis. There are a lot of methods used for trend analysis tests such as the metrological data/ VARI method, Mann-Kendall test, Penman equation, Hargreaves method, and Penman-Monteith method. Among those methods, the Mann-Kendall test is a non-parametric test that doesn't need the data to have a normal distribution and the principles and applications of each method (Shrestha et al. 2014) and has been used for hydro meteorological data trend analysis. The non-parametric Mann-Kendall test is frequently used to detect monotonic trends in a series of environmental data and is widely used to test increasing or decreasing trends in climate data or hydrological data. The Penman-Monteith method is a refinement of the original Penman equation, incorporating additional parameters to improve the estimation of evapotranspiration. The Penman-Monteith method is considered the most accurate and widely used method for estimating reference evapotranspiration, as it considers a broader range of meteorological factors.

A detailed study was conducted to find the patterns in the climatic data for five stations spread over the Koka watershed. The Mann-Kendall trend analysis was then calculated for more approximations (Sohoulande Djebou and Singh 2016).

Potential evapotranspiration

Evapotranspiration (ETo) for the Koka watershed was calculated for the baseline and extended areas using the Penman and Hargreaves methods (Tessema et al. 2020). The same ETo estimation method was also applied in this study. For future cases, the adjustment factor between the Penman and Hargreaves approaches was applied. The correction appropriately aligned with the methodology is used for reservoir modeling, model validation, and calibration (Sýs et al. 2021). To calculate probable evaporation in the future, a regression equation was created. 

Reservoir evaporation

Reservoir evaporation was not determined directly; instead, it was incorporated to implicitly estimate using a variety of techniques (Taka et al. 2020). In this work, the monthly evapotranspiration rate of the Koka reservoir was estimated using the Penman-Monteith method. The Addis Ababa Synoptic Station, located next to the basin, provided numerous meteorological data. This data was utilized for the estimation of the monthly precipitation from the Koka reservoir.

Description of HEC-HMS 4.2.1

The HEC-HMS model utilized raw data from the Geospatial Hydrologic Modeling Extension (HEC-GeoHMS), and Arc Hydro tools, including river flow, rainfall, evapotranspiration, and various watershed characteristics (slope, soil, and digital elevation models). The geospatial hydrology tool HEC-GeoHMS, which helps to define streams, was utilized in the GIS 10.7 environment. ArcGIS extensions, called Arc Hydro tools, helped to define streams, process terrain data, and draw the boundaries of the required watershed.

HEC-HMS model performance

The model's performance was assessed using the Nash-Sutcliffe coefficient (NSE), whose values ranged from zero to one. The model's performance is considered excellent when the NSE is one, and poor when it is zero. Very good performance is defined as values between 0.6 and 1.0. The strength of the simulated and observed flow was calculated using the coefficient of determination (R²) (Tessema et al. 2020). The volume errors between the simulated and observed values were quantified using relative volume error or RVE. There is no difference between simulated and observed runoff when the RVE value is zero, indicating optimal performance (Tiruneh and Worku 2018) (Figure 2).

Figure 2 Overview of GIS, HEC-GeoHMS (Kitlasten et al. 2021).

MODSIM 8.1 model

The MODSIM model is a decision support system for river basin management that was used for reservoir operation, river basin management, and estimating the production of hydroelectric power (Yang et al. 2010). A robust, interactive graphical user interface for building, locating, and connecting river basin network components was included in the MODSIM model (Visweswararao and Viswanadh 2019). A MODSIM model is used for river basin management decision support systems and for river basin management and reservoir operation, and to determine hydroelectric power generation. A MODSIM model includes a powerful, interactive graphical user interface for creating, locating, and connecting river basin network components (Yang et al. 2010). In this study, the input data used in the MODSIM model were stream flow, water demand, environmental release, net evaporation, maximum reservoir capacity, minimum and initial reservoir capacity, reservoir area and elevation, power plant capacity, load factor, tail water discharge, efficiency, and reservoir node properties.

3 Results and discussion

3.1 Performance evaluation of RCP rainfall data

Table 2 shows the basin's mean annual observed rainfall data from the reference period 1971–2000 as 950 mm, while the RCP 4.5 rainfall value is 1213 mm. The root mean square (RMSE) was 33 mm/year in comparison to the observed value, indicating that the rainfall from RCP data was overestimated by 26%. 

Table 2 Summary of accuracy evaluation for rainfall with RCP data.

	Mean annual rainfall (mm)	Bias (%)	CV (%)	RMSE (mm/year)
Observed	950	-	9.2	-
RCP 4.5	1213	23	7.3	33

3.2 Climate data analysis

Figure 3a illustrates the maximum rainfall in the years 1993 and 1998, the minimum occurred in 1989 and 2003, as shown in Figure 3a. With a minimum value of 714.82 mm/year, the rainfall value has significantly reduced. The year 2002 saw the least amount of rainfall in the Koka watershed. There was a noticeable upward trend in the temperature (Figures 3c and d). In the watershed, the mean annual maximum and minimum temperatures have risen at the rate of 0.42°C and 0.38°C per decade, respectively. Between 1999 and 2005, there was an increase in potential evapotranspiration (PET) (Figure 3b).

Figure 3 Historical climatic data trend test; (a) rainfall, (b) evapotranspiration, (c) maximum temperature, and (d) minimum temperature data.

3.3 Bias correction

The outcome of the bias correction revealed that from April to September, a significant error was noted between the climatic and the observed data (Figure 4). RCP 8.5 rainfall data for the months (not mentioned), contains a minor inaccuracy (Mirani et al. 2022). As such, in this study, RCP 8.5 has a minor inaccuracy and only focuses on RCP 4.5 to magnify the bias correction for minimum and maximum temperature, precipitation.

Figure 4 Bias correction of average monthly precipitation.

For every month, there are a few underestimations in the RCP 4.5 average highest temperature and the actual temperature measurements (Figure 5) (Mirani et al. 2022).

Figure 5 Bias correction for maximum temperature.

The actual lowest temperature and the RCP average minimum temperature are shown in Figure 6 (Mirani et al. 2022). Despite possible exceptions from January to May, the RCP mean lowest temperature had been a little undervalued in most of the parts (Figure 6).

Figure 6 Bias correction of average monthly minimum temperature.

3.4 Precipitation, temperature, and future scenarios

In Figure 7, the mean precipitation per month in the Koka watershed over the baseline (1971–2000) and subsequent years (2021–2050 and 2051–2080) are shown within the RCP4.5 and RCP8.5 scenarios. The RCP4.5 scenario produced a mean monthly precipitation decreasing/increasing by -8.1% and 14% in 2021–2050 and 2051–2080, respectively. The RCP8.5 scenario showed an average monthly precipitation reduction of 15.2% and 16.1% in 2021–2050 and 2051–2080, respectively. An arithmetic mean of statistical data to draw the plot and the variation of mean monthly precipitation varied within the year from June to September, and was incorporated since the watershed gets excess rainfall during this season.

Figure 7 Average monthly precipitation of future scenarios for the Koka watershed.

Figure 8 describes the maximum temperature in the RCP4.5 and RCP8.5 climate scenarios for the baseline period (1971–2000) and the future periods (2021–2050 and 2051–2080). For RCP4.5 (2021–2050 and 2051–2080), the monthly average maximum temperature increases by 0.2°C and 0.5°C, respectively, and for RCP8.5 (2021–2050 and 2051–2080) by 0.7°C and 1°C, respectively.

Figure 8 Average monthly maximum temperatures for future scenarios.

The lowest recorded temperature for the baseline period (1971–2000) and subsequent periods (2021–2050 and 2051–2080) in the climate change scenarios of RCP4.5 and RCP8.5 are illustrated in Figure 9. For the RCP4.5 scenario, the mean monthly lowest temperature increased by 0.5°C and 0.8°C for the periods from 2021–2050 and 2051–2080, respectively. In addition, during the coming years (2021–2050) and (2051–2080), the average monthly minimum temperatures increased by 0.4°C and 0.5°C, respectively under RCP8.5.

Figure 9 Average monthly minimum temperatures for future scenarios.

3.5 Climate change impact in the Koka watershed

In the RCP4.5 scenario, the precipitation varied by -8.1% (2021–2050) and 14% (2051–2080). Precipitation indicates a decreasing pattern according to the RCP8.5 scenario, leading to 15.2% (2021–2050) and 16.1% (2051–2080). The impacts of the changing climate on the Koka watershed were investigated regarding the RCP4.5 and RCP8.5 scenarios, as well as the period of observation (1971–2000) and subsequent periods (2021–2050 and 2051–2080) (Figure 10).

Figure 10 Change in average monthly rainfall over the periods 2021–2050 and 2051–2080.

For the RCP4.5 scenario, the highest possible temperature was raised by 0.2% (2021–2050) and 0.5% (2051–2080), respectively (Figure 11). The RCP8.5 scenario's average monthly temperature indicates an increase with changes of 0.62% (2021–2050) and 0.95% (2051–2080).

Figure 11 Change in average monthly maximum temperature.

According to the RCP4.5 scenario, in the future, the mean monthly minimum temperature over the Koka watershed increased by 0.3% (2021–2050) and 0.45% (2051–2080) (Figure 12). The RCP8.5 scenario's average monthly minimum temperature change indicates a rise in temperature.

Figure 12 Change in average monthly minimum temperature.

3.6 Evapotranspiration

The mean monthly evapotranspiration rising in subsequent conditions is shown in Figure 13 for the two different scenarios (RCP4.5 and RCP8.5) (Mirani et al. 2022). According to Figure 13, the mean monthly evapotranspiration increased for RCP4.5 by 2.5% (2021–2050) and 4.4% (2051–2080), and for RCP8.5 by 4.5% (2021–2050) and 7.2% (2051–2080).

Figure 13 Changes in average monthly evapotranspiration.

3.7 Evaporation

The Koka watershed's monthly rate of evaporation was significantly increased in the RCP4.5 and RCP8.5 scenarios during the short (2021–2050) and long (2051–2080) periods (Figure 14).

Figure 14 Changes in average monthly evaporation from Koka reservoir.

3.8 HEC-HMS model sensitivity analysis, calibration, and validation

For further model calibration, more sensitive parameters of the model might be identified through a sensitivity analysis. Researchers investigated the impact of ±30% variation in the model parameter values (Yang et al. 2010). Each of the model parameter's values can be changed by up to ±30% to test the sensitive parameters. The sensitive variables were identified following modification of the model parameters by ±30% of the peak volume. The most sensitive parameters were the dimensionless weight (X), storage coefficient (SC), and constant rate (CR) (Figure 15). Less sensitive parameters included the recessional constant (RC), initial loss (IL), time of concentration (TC), initial discharge (ID), ratio-to-peak (RP), and travel time (K).

Figure 15 Volume change percentages with variation in percentage of parameters, the HEC-HMS model was calibrated (1992 to 2003) and validated (2004 to 2009) using streamflow data.

Stream flow calibration is a continuous procedure that examines parameters and evaluates simulation and observed data in modeling to identify the most suitable parameter. For the Koka reservoir, the areal precipitation, areal evapotranspiration, and observed stream flow were calibrated by both automated and manual methods. The calibration result (Figure 16) demonstrated that the peak is accurately captured for most of the year, and the model and the observed and simulated stream flow are in good agreement. Nonetheless, there is a slight undervaluation of the simulated flow in the years 1993, 1997, and 2000. Following model calibration, the relative volume error (RVE) was 0.063%, the coefficient of determination (R²) was 0.67, and the Nash-Sutcliffe coefficient (NSE) was 0.82. Additionally, the model achieved an NSE of 0.72, an R² of 0.7, and an RVE of -4.7% for validation. Figure 16 shows how well the years 2004 and 2006 captured the peak flow of the stream, even though there was an overestimation in 2009 and an underestimation in 2005, 2003, and 2008.

Figure 16 Observed and simulated stream flow at the Koka dam.

3.9 Reservoir inflow

A variety of time perspectives, such as baseline, short-term, and long-term, were forecasted using reservoir inflow (stream flow data) (Mirani et al. 2022). Compared to the observed flow, the mean stream flow will vary by -12.93% (2021–2050) and -17.01% (2051–2080) under the RCP4.5 scenario. The average monthly stream flow is expected to decrease by -20.51% in the case of the RCP8.5 scenario, and -22.87% during 2051–2080, indicating a downward trend (Figure 17). The gauging station in the upper Awash sub basin is presented in Figure 1 and marked in green and the monitoring station for inflow to reservoir is near to the Koka dam reservoir with one inflow site which is directly contributing to the reservoir with a monthly average inflow base for the existing inflow sites (Figure 1).

Figure 17 Koka reservoir inflow discharge at different time horizons.

3.10 Hydropower generation with observed data

Regarding the generation of hydropower, the mean production of power and energy demonstrated substantial variations compared to the observed stream flow data recorded. The reservoir simulation revealed that the average energy generated during the observed period (1987–2005) was 401.23 MWh, while the maximum energy generated was 425.36 MWh (Figure 18).

Figure 18 Average hydropower versus energy for base period scenario.

3.11 Hydropower generation with RCPs scenario

The reservoir's simulation for the RCP4.5 and RCP8.5 scenarios focused on the energy variations between the observed period (1987–2005) and future periods (2021–2050 and 2051–2080). The short-term (2021–2050) energy generated in the RCP4.5 scenario is expected to be 381.85 MWh, with an average of 328.24 MWh. With an average energy of 320.57 MWh, the energy generated in the long-term (2051–2080) scenario will be 376.23 MWh. This suggests that, in the short and long terms, respectively, the average energy generation will decline by 0.62% and 0.78% due to the change in climate under and existence of uncertainties for RCP 4.5 and RCP 8.5. Hydropower generation has been achieved by converting the kinetic energy of water into the energy of electricity by the generator. In hydropower plant generation, water is collected upstream and directed by canals downstream. The difference between upstream and downstream is called the head, and the energy is calculated in kWh on an hourly basis with a function of discharge and head to keep other parameters constant. According to the results of RCP8.5 scenario, the maximum amount of energy produced in short-term scenarios is anticipated to be 375.48 MWh, with an average energy of 334.95 MWh. Long-term energy generation for RCP8.5 is expected to reach a maximum of 378.35 MWh and an average of 318.25 MWh. The average energy generated may drop by 1.14% and 1.42% over the short and long terms, respectively. The sample of the MODSIM output of daily energy and hydropower generation production is displayed in Figure 19.

Figure 19 Daily hydropower generation and energy.

3.12 Reservoir operation guide curve

Three periods of reservoir rule curves were developed for the Koka reservoir based on the reservoir guide curves of Mulat et al. (2018). To meet the projects' target demand, these were the observed period (1987–2005), short-term (2021–2050), and long-term (2051–2080) for the RCP4.5 and RCP8.5 scenarios. The reservoir rule guide curves for the Koka reservoir are displayed in Figure 20, and the results indicate a decreasing trend for future scenarios when compared to the observed period. The outcome indicates that the Koka watershed develops a significant change in climate, which affects reservoir inflow and operation and eventually reduces the amount of hydropower and energy produced in the future. Thus, to mitigate the effects of climate change, implementation of watershed management techniques and controlling the future changes in hydropower output, water resources planners and government authorities must use the research findings.

Reservoir rule curves are graphical representations that show the desired or target storage levels in a reservoir over the course of a year. Reservoir curve rules were determined by analyzing historical inflow data and determining monthly or seasonal water demand for hydropower after the established storage target, followed by developing the rule curve based on operational constraints such as minimum and maximum storage levels, reservoir release requirements, and downstream flow considerations. The resulting reservoir rule curve provides a guideline for reservoir operators to manage the storage and releases from the reservoir throughout the year, helping to ensure efficient and sustainable water resource management (Saied et al. 2013).

Figure 20 Reservoir operation guide curve.

3.13 Discussion

The impact of climate change studies on hydropower generation by coupling climate models (RCM/GCM) with hydrological models involves a range of uncertainties (Yang et al. 2010). Selection of a GCM/RCM model and plausible scenarios are the greatest sources of uncertainty for climate change analysis, in addition to hydrological model simulation (Mulat et al. 2018). Despite these limitations, in this work every effort was made to minimize the uncertainties for climate prediction and hydrological simulation to understand the potential impact of climate change on future hydropower generation (Dagbegnon et al. 2016), such as best climate model (RCM) selection, most plausible climate scenarios (RCP4.5 and 8.5), and selection and bias correction climate model output. This study coupled a single climate model (RCM) under RCP4.5 and 8.5 emission scenarios with a hydrological model (HEC-HMS) to investigate future streamflow changes in the Upper Awash sub-basin. However, the results of this work are consistent with similar studies (Khaniya et al. 2018; Luo et al. 2018; Kitlasten et al. 2021).

Overall, projected temperature and precipitation are expected to increase in the study area. The increment in minimum temperature would be higher compared to maximum temperature, and such changes are common globally (Labadie 2010; Lin and Rutten 2016; Mallick et al. 2021). The temperature change is more severe under RCP8.5 than RCP4.5, confirming that RCP8.5 is the highest carbon emission scenario, and this creates an active hydrological cycle which leads to heavy rainfall. The application of bias correction improved the projected temperature and precipitation accuracy (Qin et al. 2020; Pörtner et al. 2022). On average, annual projected streamflow is expected to decrease by 14.97% under RCP4.5 and 21.69% under RCP8.5 (Berhe et al. 2013; Alazard et al. 2015; Alemseged and Rientjes 2015; Bekele et al. 2019).

4 Conclusion

The assessment of the production of hydropower and reservoir operation according to the effects of climate change can play a significant role as studied from the baseline and subsequent years generated electricity. The water availability of the Koka reservoir is likely to be severely affected by climate change. The result of climate forecasting demonstrates that the RCP data can be observed when the short- and long-term climate scenarios of RCP4.5 and RCP8.5 are considered. During the stream flow calibration period, the HEC-HMS model performed well, with an NSE of 0.72, R² of 0.73, and RVE of -0.063%. The model performed well during the validation phase, with NSE, R², and RVE at 0.74, 0.75, and -4.7%, respectively. The reservoir simulation model (MODSIM) was used to analyze the reservoir operation for both the base and future periods inferred as per expectation. From the simulated results of release and hydropower generation, it is possible to estimate the reservoir operation rule curve and hydropower production for three distinct climate periods: baseline (1987–2005), short-term (2021–2050), and long-term (2051–2080). RCP4.5's quantified maximum energy will be 376.213 MWh for the short-term period, and 370.513 MWh for the long-term period, respectively. Furthermore, the maximum energy of 368.605 MWh and 363.492 MWh, respectively, will probably be generated for short and long periods in the case of RCP8.5. This suggests that energy generation from the Koka reservoir is erratic, and in the future, energy production may reduce continuously. The findings are crucial to watershed management organizations and hydroelectric power authorities. It is also strongly advised that other scientific researchers use a similar methodology for similar reservoirs.

References

1. Abera, F.F., D.H. Asfaw, A.N. Engida, and A.M. Melesse. 2018. “Optimal operation of hydropower reservoirs under climate change: The case of Tekeze reservoir, Eastern Nile.” Water (Switzerland) 10 (3). https://doi.org/10.3390/w10030273
2. Alazard, M., C. Leduc, Y. Travi, G. Boulet, and A. Ben Salem. 2015. “Estimating evaporation in semi-arid areas facing data scarcity: Example of the El Haouareb dam (Merguellil watershed, Central Tunisia).” Journal of Hydrology: Regional Studies 3, 265–284. https://doi.org/10.1016/j.ejrh.2014.11.007
3. Alemseged, T.H., and T. Rientjes. 2015. “Evaluation of regional climate model simulations of rainfall over the Upper Blue Nile basin.” Atmospheric Research 161–162, 57–64.
4. Alemu, Z.A., and M.O. Dioha. 2020. “Climate change and trend analysis of temperature: the case of Addis Ababa, Ethiopia,” Environmental Systems Research 9 (1). https://doi.org/10.1186/s40068-020-00190-5
5. Anand, J., A.K. Gosain, and R. Khosa. 2018. “Optimisation of multipurpose reservoir operation by coupling soil and water assessment tool (SWAT) and genetic algorithm for optimal operating policy (Case Study: Ganga River Basin).” Sustainability (Switzerland) 10 (5). https://doi.org/10.3390/su10051660
6. Awulachew, S.B, A.D. Yilma, M. Loulseged, W. Loiskandl, M. Ayana, and T. Alamirew. 2007. “Water Resources and Irrigation Development in Ethiopia.” International Water Management Institute Working Paper 123.
7. Bekele, D., T. Alamirew, and A. Kebede. 2019. “Modeling Climate Change Impact on the Hydrology of Keleta Watershed in the Awash River Basin, Ethiopia.” Environmental Modeling and Assessment 24, 95–107. https://doi.org/10.1007/s10666-018-9619-1
8. Berhe, F.T., A.M. Melesse, D. Hailu, and Y. Sileshi. 2013. “MODSIM-based water allocation modeling of Awash River Basin, Ethiopia.” Catena 109, 118–128. https://doi.org/10.1016/j.catena.2013.04.007
9. Bruce, P. 1967. “Meteorology and Water Resources.” In Water Resources of Canada, edited by Claude Dolman, University of Toronto Press, Toronto, Canada, pp. 111–115. https://doi.org/10.3138/9781442653863-016
10. Chaleeraktrakoon, C., and Y. Chinsomboon. 2015 “Dynamic rule curves for flood control of a multipurpose dam.” Journal of Hydro-environment Research 9 (1): 133–144. https://doi.org/10.1016/j.jher.2014.11.002
11. Chiang, J-L., H-C. Yang, Y-R. Chen, and M-H. Lee. 2013. “Potential impact of climate change on hydropower generation in Southern Taiwan.” Energy Procedia 40, 34–37. https://doi.org/10.1016/j.egypro.2013.08.005
12. Dagbegnon C., S. Djebou, and V.P. Singh. 2016. “Impact of climate change on the hydrologic cycle and implications for society.” Environment and Social Psychology 1 (1): 36–49. http://dx.doi.org/10.18063/ESP.2016.01.002
13. Falchetta, G.D., D.E.H.J. Gernaat, J. Hunt, and S. Sterl. 2019. “Hydropower dependency and climate change in sub–Saharan Africa: A nexus framework and evidence-based review.” Journal of Cleaner Production 231, 1300–1417. https://doi.org/10.1016/j.jclepro.2019.05.263  
14. Fanta, S.S., M.B. Yesuf, and T.A. Demissie. 2023. “Investigation of climate change impact on the optimal operation of Koka Reservoir, Upper Awash Watershed, Ethiopia.” Heliyon 9 (5). https://doi.org/10.1016/j.heliyon.2023.e16287
15. Gentilucci, M., M. Bufalini, M. Materazzi, M. Barbieri, D. Aringoli, P. Faravollini, and G. Pambianchi. 2021. "Calculation of potential evapotranspiration and calibration of the Hargreaves equation using geostatistical methods over last 10 years in central Italy.” Geosciences (Switzerland) 11 (8). https://doi.org/10.3390/geosciences11080348
16. Goharian, E., and S.J. Burian. 2018. “Developing an integrated framework to build a decision support tool for urban water management.” Journal of Hydroinformatics 20 (3): 708–727. https://doi.org/10.2166/hydro.2018.088
17. Goswami, B.B., R.P.M. Krishna, P. Mukhopadhyay, M. Khairoutdinov, and B.N. Goswami. 2015. “Simulation of the Indian Summer Monsoon in the Super parameterized Climate Forecast System Version 2: Preliminary Results.” Journal of Climate 28 (22): 8988–9012. https://doi.org/10.1175/JCLI-D-14-00607.1
18. Guangli, F., A. Sarabandi, and M. Yaghoobzadeh. 2021. “Evaluating the climate change effects on temperature, precipitation and evapotranspiration in eastern Iran using CMPI5.” Water Supply 21 (8): 4316–4327. https://doi.org/10.2166/ws.2021.179
19. Gunathilake, M.B., P. Panditharathne, A.S. Gunathilake, and N.D. Warakagoda. 2020 “Application of a HEC-HMS model on event-based simulations in a tropical watershed.” Journal of Engineering and Applied Science Research 47 (4): 349–360.
20. Haile, T.H., and T. Rientjes. 2015. "Evaluation of regional climate model simulations of rainfall over the Upper Blue Nile basin." Atmospheric Research 161–162, 57–64. https://doi.org/10.1016/j.atmosres.2015.03.013
21. Hunt, J.D., G. Falchetta, B. Zakeri, A. Nascimento, P.S. Schnieder, N.A.B. Weber, A.L.A. Mesquita, et al. 2020. “Hydropower impact on the river flow of a humid regional climate.” Climatic Change 163 (1): 379–393. https://doi.org/10.1007/s10584-020-02828-w
22. Huntington, T.G. 2006. “Evidence for intensification of the global water cycle: Review and synthesis.” Journal of Hydrology 319 (1–4): 83–95. http://dx.doi.org/10.1016/j.jhydrol.2005.07.003
23. Intergovernmental Panel on Climate Change (IPCC). 2014. “Africa.” In: Climate Change 2014 – Impacts, Adaptation and Vulnerability: Part B: Regional Aspects: Working Group II Contribution to the IPCC Fifth Assessment Report. Cambridge University Press, 20141199-1266. https://doi.org/10.1017/CBO9781107415386.002
24. International Institute for Environment and Development (INDC). 2021. “Intended Nationally Determined Contribution of the Federal Democratic Republic of Ethiopia,” pp. 1–13, Available online: https://www4.unfccc.int/sites/submissions/INDC/Published%20Documents/Ethiopia/1/INDC-Ethiopia-100615.pdf  (accessed on 21 October 2023).
25. Khan, M.S., P. Coulibaly, and Y. Dibike. 2006. “Uncertainty analysis of statistical downscaling methods.” Journal of Hydrology 319 (1–4): 357–382. https://doi.org/10.1016/j.jhydrol.2005.06.035
26. Khaniya, B., I. Jayanayaka, P. Jayasanka, and U. Rathnayake. 2019. "Rainfall Trend Analysis in Uma Oya Basin, Sri Lanka, and Future Water Scarcity Problems in Perspective of Climate Variability." Advances in Meteorology 3636158. https://doi.org/10.1155/2019/3636158
27. Khaniya, B., Ch. Karunanayake, M.B. Gunathilake, and U. Rathnayake. 2020. "Projection of Future Hydropower Generation in Samanalawewa Power Plant, Sri Lanka.” Mathematical Problems in Engineering 8862067. https://doi.org/10.1155/2020/8862067
28. Khaniya, B., H.B. Priyantha, N. Baduge, H.M. Azamathulla, and U. Rathnayake. 2018. “Impact of climate variability on hydropower generation: a case study from Sri Lanka.” ISH Journal of Hydraulic Engineering 26, 1–9. https://doi.org/10.1080/09715010.2018.1485516
29. Kitlasten, W., E.D. Morway, R.G. Niswonger, M. Gardner, J.T. White, E. Triana, and D. Selkowitz. 2021. “Integrated hydrology and operations modeling to evaluate climate change impacts in an agricultural valley irrigated with snowmelt runoff.” Water Resources Research 57 (6): e2020WR027924. https://doi.org/10.1029/2020WR027924
30. Labadie, J. 2010. MODSIM 8.1: River Basin Management Decision Support System User Manual and Documentation. Department of Civil and Environmental Engineering Colorado State University, Ft. Collins, Colorado, USA.
31. Lin, N.M., and M. Rutten. 2016. “Optimal Operation of a Network of Multi-purpose Reservoir, A Review.” Procedia Engineering 154, 1376–1384. https://doi.org/10.1016/j.proeng.2016.07.504
32. Luo, M., T. Liu, F. Meng, Y. Duan, A. Frankl, A. Bao, and P. De Maeyer. 2018. “Comparing Bias Correction Methods Used in Downscaling Precipitation and Temperature from Regional Climate Models: A Case Study from the Kaidu River Basin in Western China.” Water 10, 8. https://doi.org/10.3390/w10081046
33. Mallick, J., A.R.M.T. Islam, B. Ghose, H.M. Touhidul Islam, Y. Rana, Z. Hu., et al. 2021. “Spatiotemporal trends of temperature extremes in Bangladesh under changing climate using multi-statistical techniques.” Theoretical and Applied Climatology 147, 307–324. https://doi.org/10.1007/s00704-021-03828-1 
34. Mamman, M.J. 2018. “Recalibration of Kainji Reservoir Operating Rule for Optimal Operation.” Annals of Reviews & Research 1 (1): 555555. https://doi.org/10.19080/arr.2018.01.555555
35. Mekonnen, T. W., S.T. Teferi, F.S. Kebede, G. Anandarajah. 2022. “Assessment of Impacts of Climate Change on Hydropower-Dominated Power System: The Case of Ethiopia.” Applied Sciences 12, 1954. https://doi.org/10.3390/app12041954
36. Meng, Y., J. Liu, Z. Wang, G. Mao, K. Wang, and H. Yang. 2021. “Undermined co-benefits of hydropower and irrigation under climate change.” Resources, Conservation and Recycling 167, 105375. https://doi.org/10.1016/j.resconrec.2020.105375
37. Mirani, K.B., M.A. Ayele, T.K. Lohani, and T.Y. Ukumo. 2022. “Evaluation of hydropower generation and reservoir operation under climate change from Kesem Reservoir, Ethiopia.” Advances in Meteorology 3336257. https://doi.org/10.1155/2022/3336257
38. Moriasi, D.N., J.G. Arnold, M.W. Van Liew, R.L. Bingner, R.D. Harmel, and T.L. Veith. 2007. “Model evaluation guidelines for systematic quantification of accuracy in watershed simulations.” Transactions of the ASABE 50 (3): 885–900. https://doi.org/10.13031/2013.23153
39. Mulat, A.G., S.A. Moges, and M.A. Moges. 2018. “Evaluation of multi-storage hydropower development in the upper Blue Nile River (Ethiopia): regional perspective.” Journal of Hydrology: Regional Studies 16, 1–14. https://doi.org/10.1016/j.ejrh.2018.02.006
40. Norouzi, N. 2020. “Climate change impacts on the water flow to the reservoir of the Dez Dam basin.” Water Cycle 1, 113–120. https://doi.org/10.1016/j.watcyc.2020.08.001
41. Otto, C.S., and D. Josef. 2000. “World reference base for soil resources. An introduction with special reference to soils of tropical forest ecosystems.” Working paper and preprint, 95/14, ISSS-ISRIC-FAO.
42. Pilesjo, P., and S.S. Al-Juboori. 2013. “Modelling the effects of climate change on hydroelectric power in Dokan, Iraq.” International Journal of Energy and Power Engineering 5 (2): 7. https://doi.org/10.11648/j.ijepe.s.2016050201.12
43. Pörtner, H.O., D.C. Roberts, E.S. Poloczanska. 2022. “Climate Change Impacts, Adaptation, and Vulnerability.” Contribution of Working Group II to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change (IPCC). Cambridge University Press, In Press, Summary for Policymakers, United Kingdom.
44. Qin, P., H. Xu, M. Liu. 2020. “Climate change impacts on three Gorges reservoir impoundment and hydropower generation.” Journal of Hydrology 580, 123922.
45. Ranzani, A., M. Bonato, E.R. Patro, L. Gaudard, and C. De Michele. 2018. “Hydropower future: Between climate change, renewable deployment, carbon and fuel prices.” Water (Switzerland) 10 (9): 1–17. https://doi.org/10.3390/w10091197
46. Saied, S., H. Magsi, F. Marri, and H. Rejeb. 2013. “Under what management practices and mechanisms water resource can be considered as a public good for sustainable development in the developing countries.” Journal of Water Sustainability 3 (1): 1–16. 
47. Schaeffer, R., A. Szklo, and P. de Lucena. 2012. “Energy sector vulnerability to climate change, A review.” Energy 38 (1): 1–12. https://doi.org./10.1016/j.energy.2011.11.056
48. Scola, L.A., R.H.C. Takahashi, and S.A.A.G. Cerqueira. 2014. “Multipurpose Water Reservoir Management: An Evolutionary Multiobjective Optimization Approach.” Mathematical Problems in Engineering, 628259. https://doi.org/10.1155/2014/638259  
49. Shrestha, S., M. Khatiwada, M.S. Babel, and K. Parajuli. 2014. “Impact of climate change on river flow and hydropower production in Kulekhani hydropower project of Nepal.” Environmental Processes 1 (3): 231–250. https://doi.org/10.1007/s40710-014-0020-z
50. Sohoulande Djebou, D.C., and V.P. Singh. 2016. "Entropy-based index for spatiotemporal analysis of streamflow, precipitation, and land-cover." Journal of Hydrologic Engineering 21 (11): 05016024.
51. Sýs, V., P. Fošumpaur, and T. Kašpar. 2021. “The Impact of Climate Change on the Reliability of Water Resources.” Climate 9 (11): 153. https://doi.org/10.3390/cli9110153
52. Taka, G.N., T.T. Huong, I.H. Shah, and H.S. Park. 2020. “Determinants of energy-based CO₂ emissions in Ethiopia: A decomposition analysis from 1990 to 2017.” Sustainability 12, 4175.
53. Tessema, N., A. Kebede, and D. Yadeta. 2020. “Modeling land use dynamics in the Kesem sub-basin, Awash River basin, Ethiopia.” Cogent Environmental Science 6, 1. https://doi.org/10.1080/23311843.2020.1782006
54. Tiruneh, M. A., and G.B. Worku. 2018. “Identification of system vulnerabilities in the Ethiopian electric power system.” Global Energy Interconnection 1, 358–365.
55. Visweswararao, M., and G.K. Viswanadh. 2019. “Operation of an Upstream Reservoir Using Rule Curve to Improve the Performance of the Downstream Reservoirs.” International Journal of Civil Engineering and Technology (IJCIET) 10 (7): 14–21. http://iaeme.com/Home/issue/IJCIET?Volume=10&Issue=7
56. Yang, W., J. Andréasson, and L.P. Graham. 2010. “Distribution-based scaling to improve usability of regional climate model projections for hydrological climate change impacts studies.” Hydrology Research 41, 211–229. https://doi.org/10.2166/nh.2010.004