Climate change is becoming a global concern, dictating the water resource availability of a certain region. This study focuses on the impact of climate change on the streamflow of the Melka Wakena reservoir in the Upper Wabi Shebelle sub-basin using a Soil Water Analysis Tool (SWAT). The climate model variables (precipitation and temperature) are obtained from the Coordinated Regional Downscaling Experiment (CORDEX-Africa) and applied under representative concentration pathway (RCP4.5) and (RCP8.5) scenarios. Bias correction was applied to the climate variables before transferring to the hydrological model (SWAT) to simulate discharge. The performance measures R2, NSE, and Percent Bias (PBIAS) for calibration and validation were 0.69, 0.65, and −5.10 and 0.65, 0.61, and −9.81, respectively. Streamflow was simulated for two consecutive periods from 2021 to 2050 and from 2051 to 2080 for both scenarios and compared with the base period from 1986 to 2015 to explore the impact of climate change on inflows to the reservoir. The result obtained showed that the precipitation was predicted to be mildly decreasing, whereas overall annual flow was projected to be insignificantly decreasing under RCP4.5 and RCP8.5 scenarios. The impact of climate change on the water resource of the study area was predicted to be a statistically insignificant reduction.

  • The study deals with the impact of climate change on the water resources.

  • The study highlights the application of ArcSWAT to predict the flow for climate change.

Graphical Abstract

Graphical Abstract
Graphical Abstract

Climate change is defined as the change in climate parameters, such as the change in precipitation trend and air temperature, which indirectly affects the natural ecosystem, human health, irrigation potential, energy use, and availability of water in many areas of the world (Nega 2008). For instance, it has a significant effect on the distribution of freshwater because the distribution of water resources is very sensitive to climate change (Solomon 2007).

Global warming is directly linked to changes in the large-scale hydrological cycle (Anwar & Adem 2016). Linkages between global warming and hydrological cycle components include increasing atmospheric water vapor content (especially potential evaporation); changing precipitation patterns, intensity, duration, and extremes; reduced snow cover and widespread melting of ice; and changes in soil moisture content in the soil and runoff generated (Bates 2008a, 2008b).

The available amount of water is limited, scarce, and not spatially distributed, so a proper management of water resources is essential to satisfy the current demands as well as to maintain sustainability. Water resource planning and management has become a challenge in the 21st century due to the conflicting demands from various stakeholder groups, increasing population, rapid urbanization, and climate change producing shifts in hydrologic cycles (Abdella 2013).

Changes in temperature and precipitation patterns due to global climate change are expected to alter regional hydrological conditions, affecting water resource availability and discharge or flows into rivers (Vandana et al. 2019). The changes in climate parameters such as precipitation, maximum temperature, minimum temperature, potential evapotranspiration, and streamflow have a direct implication on water resource management (Kundzewicz 2009).

Climate change will have a direct impact on the availability and variability of freshwater as the frequency of climatic extremes such as heatwaves, drought, and changes in the rainfall pattern increases in response to global warming (IPCC 2012).

The impact of climate change on water resources is the most crucial research agenda at the global level (Parry et al. 2007a). This change in climate has a significant impact on water resources by disturbing the normal hydrological processes. Future change in overall flow magnitude, variability, and timing of the mean flow event is among the most frequently cited hydrological issues.

The IPCC finding indicates that developing countries such as Ethiopia will be more vulnerable to climate change. Because of the poor flexibility to adjust the economic structure and being largely dependent on sectors that are sensitive to climate change, such as agriculture, forestry, and fishing, the impact of climate change has far-reaching implications in Ethiopia (Worqlul et al. 2018). Hence, assessing climate change impact as part of a comprehensive program is crucial for the country (Zarey 2006).

Modern agriculture largely depends on available water resources, while a large part of the country is categorized as arid and semi-arid regions highly prone to drought and desertification. The search for alternative sources of livelihoods as a result of increasing population pressure ultimately degrades the fragile ecosystem to a greater extent (Melkamu et al. 2017).

River water levels along the Shebelle River in Ethiopia are also currently below normal water levels. Water levels have decreased significantly in the Melka Wakena reservoir, which is located in the upper part of the Shebelle River. The low flows along the Shebelle River both in Somalia and in Ethiopia are due to a hydrological drought or change within the basin, especially in the upper part of the catchment (Hayicho et al. 2019).

The Melka Wakena hydro-electric power plant has been constructed on the Melka Wakena catchment, which is the sub-basin of the upper Wabi Shebelle River. The capacity of the installed power plant is 153 MW, which is not a multi-purpose one or is used only for hydropower generation. Previous studies show that there is a difference between the actual capacity and the installed capacity of the power plant. The plant is not producing the required power, and the studies indicate that it produces at the rate of 12.21% below the design capacity . The reason for this might be the reduction in inflows to the reservoir and water loss due to evaporation. The reservoir inflows have been reduced to about 5% from the designed reservoir inflow and there is also a water loss from the current power plant (Gebresenbet 2010). This entails a critical evaluation of the impact of climate change on the water resource.

Even though much research has been done in continental or global scale studies related to climate change, the magnitude and type of impact at the regional watershed level are not investigated in many parts of the world, including Ethiopia.

Therefore, there is a need to quantify and assess the future scenarios of climate change and its impact on water resources of the country in general and the river basin in particular, because the country's future development activities depend on this resource.

Study area

The case study was conducted in the Melka Wakena watershed, which is located in a headwater of the Upper Wabi Shebelle river basin in the Oromia region of Ethiopia as shown in Figure 1. It is located between 6.5° and 7.5 °N latitude and 39°–39.7 °E longitude. The watershed is one of the major tributaries of the Upper Wabi Shebelle river basin and it covers an area of 4,380 km2. It is part of the seven districts of Adaba, Dodola, Gedebi-Assessa, Kofale, Kore, Kokkosa, and Lemu Bilbilo (Meraro). The elevation in the watershed ranges between 2,243 and 4,178 m above sea level.

Figure 1

Location of the study area.

Figure 1

Location of the study area.

Close modal

The mean monthly rainfall over the station varied from 8.73 mm at the Hunte station in December to 211.69 mm at the Adaba station in August for the period 1985–2015, and the annual rainfall of the study area ranges from 600 to 1,600 mm. The mean monthly minimum temperature over the study area was between 3 and 9 °C, and the mean monthly maximum temperature over the study area was between 15 and 25 °C. However, there is a slight variation in maximum and minimum temperatures from month to month.

Data collection and methods

Regional climate model data

Regional climate model (RCM) simulation from the Coordinated Regional Climate Downscaling Experiment Program (CORDEX) driven by ICHEC–EC–EARTH is obtained from the CORDEX Project under the Africa Domain with a spatial resolution of 50 km (0.44°). The RCM used in this study is Regional Climate Model, version 2.2 (RACMO22T), which is nested in the ICHEC–EC–EARTH Global Circulation Model (GCM) developed at the KNMI center of the Netherlands, with the 1985–2015 base period for simulation.

Observations and climate projections provide abundant evidence that freshwater resources are vulnerable and have the potential to be strongly impacted by climate change, with wide-ranging consequences for societies and ecosystems (Bates 2008a, 2008b). These impacts are mainly due to an increase in temperature, evaporation, sea-level rise, and rainfall variability (Kundzewicz 2004).

Observed data

For an evaluation of climate change and variability, observed data were used to develop temporal climate change scenarios of the current and future projection of precipitation, temperature, evapotranspiration, and streamflow to the watershed. For this study, climate data downloaded from RCMs, and the RACMO22T results were compared against actual weather station data collected from the Ethiopian National Meteorological Services Agency. To obtain good results, climate data having little gaps and missing records are recommended, in order to reduce uncertainty in climate change. Accordingly, for both rainfall and temperature, stations having a smaller number of missing records were used for this study, and these stations included the following: Adaba, Dodola, Hunte, Kofale, and Meraro. For all stations, the missing values were less than 10% for the periods of 1985–2015 and the missing values were filled with mean values.

The study used observed streamflow data to calibrate and validate the SWAT model. The observed streamflow data were collected from the hydrology department of the Ethiopian Ministry of Water, Irrigation, and Energy (MWIE) and the Wabi Shebelle Basin Authority.

For calibration and validation, one hydrological station at the Melka Wakena dam site using SWAT hydrological model parameters was optimized after sensitivity analysis. Finally, the future inflow volume into the Melka Wakena reservoir was generated and used to estimate the reservoir performance under climate change.

The future scenarios of climate for the study area were projected using outputs from the GCMs. GCMs have been considered as credible tools to simulate the response of the global climate system to increasing greenhouse gas concentrations (IPCC-TGCIA 1999). However, climate change projections involve uncertainties due to various assumptions in GCM conceptualizations and scenario developments (IPCC 2014).

This study, therefore, used climate data from the RACMO22T and under the ICHEC–EC–EARTH GCMs to estimate the future climate change of the Melka Wakena watershed.

Finally, after calibrating and validating the SWAT model based on observed streamflow data, it was used for the climate change impact analysis on streamflow. The impact of climate change on the streamflow of the watershed was studied by introducing the percent change in mean monthly rainfall and the difference in mean monthly temperature (Δ°C) into the calibrated and validated SWAT model for two future periods (the 2050s and 2080s) under RCP4.5 and RCP8.5 climate change scenarios.

Flow transfer to the dam site

The streamflow gauging stations are located at a far-off distance from the dam site. Hence, transferring streamflow data to the point of interest in many reservoirs becomes important. To transfer flow data to Melka Wakena near the dam, Equation (1) was used to transfer the flow from the gauged site to the ungauged site. The entire flow of the ungauged site at the Melka Wakena dam site was obtained using the following equation (Nruthya & Srinivas 2015):
(1)
where Aungauged is the drainage area at the site of interest, Agauged is the drainage area of the gauged site, Qungauged is the discharge at the site of interest, and Qgauged is the discharge at the gauged site. n is a value that varies between 0.6 and 1.2. An average value of 0.9 was used for the estimation.

Bias correction method

For this study, the distribution mapping bias correction method was used among different methods of bias correction like linear scaling, which is highly suitable for mean estimation. But the linear scaling method does not adjust the standard deviation and the percentiles, while the distribution mapping method does (Teutschbein 2013).

The outputs of an ensemble of high-resolution RCMs from the CORDEX-Africa under RCP4.5 and RCP8.5 climate scenarios were used after bias correction. The output of the downscaled ICHEC–EC–EARTH (GCM) using nested RACMO22T under RCP4.5 and RCP8.5 scenario conditions for the study area was used. Generally, after bias correction, the future projection is compared to the base period analyzed, average mean monthly projected precipitation, and evapotranspiration, which results in producing some variation with the observed climate data of the Melka Wakena sub-basin.

Model calibration and validation

Rainfall simulation and flow modeling were done using the SWAT model. First, the performance of the model was evaluated by using hydrograph comparison at the Melka Wakena dam site through calibration. Second, using the optimized parameters at the Melka Wakena dam site meant that after selecting the objective function and optimized value of parameters, the SWAT model was validated. The calibration was done by using the observed discharge with simulated discharge using SWAT-CUP by selecting the following three performance measures: R2, NSE, and D. Flow calibration at the Melka Wakena dam site was done by using the hydro-meteorological data of 1988–2004, whereas validation was done by using 2005–2012 data.

Performance evaluation

For an evaluation of the performance of the model, three efficiency measuring techniques, namely, Nash–Sutcliffe Efficiency (NSE), Coefficient of determination (R2), and percent in volume difference (D), were used.

NSE: The efficiency, NSE, proposed by Nash and Sutcliffe (1970) is defined as one minus the sum of the absolute squared differences between the predicted and the observed values normalized by the variance of the observed values during the period under investigation. Melkamu et al. (2017) recommended monthly time steps stating that NSE values between 0.75 and 1 indicate very good values and an NSE value between 0.65 and 0.75 indicates a good value.
(2)
where NSE is the Nash–Sutcliffe Efficiency, Qobs (t) is the observed discharge at time t, Qsim (t) is the simulated discharge at time t, and Qobs is the mean average observed discharge.
Coefficient of determination (R2). The coefficient of determination R2 is defined as the squared value of the coefficient of correlation. It can also be expressed as the squared ratio between the covariance and the multiplied standard deviations of the observed and predicted values. It is calculated as
(3)

This performance measure is widely used in hydrological modeling. With regard to the accepted range of R2, typically values greater than 0.5 are considered acceptable (Santhi et al. 2001).

Percent Bias (PBIAS)

The percent difference over a specified period with the total days calculated from the measured and simulated values of the quantity in each model time step as a percent difference between +5 and −5 indicates that a model performs well, while a difference between +5 and +10% and −5 and −10% indicates a model with reasonable performance.
(4)

The overall research steps followed during the case study are presented in a general framework presented in Figure 2.

Figure 2

Simplified general framework of the methodology.

Figure 2

Simplified general framework of the methodology.

Close modal

Comparison of precipitation under baseline period

The mean monthly precipitation for the baseline period from 1985 to 2015 of observed and bias-corrected data from the selected climate model is shown in Figure 3.

Figure 3

Observed, raw simulated historical, and bias-corrected mean monthly precipitation in the Melka Wakana sub-basin baseline period (1986–2015).

Figure 3

Observed, raw simulated historical, and bias-corrected mean monthly precipitation in the Melka Wakana sub-basin baseline period (1986–2015).

Close modal

The model underestimates monthly precipitation during January, February, March, April, and December, whereas it overestimates during May, June, July, August, September, and October. However, during the month of November, the model simulates almost similar rainfall values with observed data. As it is revealed in Figure 3, the difference between observed and simulated rainfall is large before bias correction is applied. However, after bias correction, the graph of bias-corrected rainfall and observed rainfall values show similar patterns and are close to each other.

Projected precipitation, temperature, and evapotranspiration

Precipitation

The future climate data were projected using the selected climate model under RCP4.5 and RCP8.5 scenarios. These future climate data include precipitation, evapotranspiration, and maximum and minimum temperatures. The evaluation was made into two consecutive 30 years of data ranging from 2021 to 2050 and from 2051 to 2080. The projected changes were evaluated for both RCP4.5 and RCP8.5 scenarios and then compared with the baseline period from 1985 to 2015 (Table 1).

The projected monthly rainfall patterns under both scenarios are expected to be lower in all months, except July. In July, the projected rainfall is expected to be higher than the observed rainfall under both scenarios, as shown in Figure 4 (RCP4.5) and Figure 5 (RCP8.5).

Figure 4

Projected mean monthly rainfall for the period from 2021 to 2050 and from 2051 to 2080 under the RCP4.5 climate scenario.

Figure 4

Projected mean monthly rainfall for the period from 2021 to 2050 and from 2051 to 2080 under the RCP4.5 climate scenario.

Close modal
Figure 5

Projected mean monthly rainfall for the period from 2021 to 2050 and from 2051 to 2080 under the RCP8.5 climate scenario.

Figure 5

Projected mean monthly rainfall for the period from 2021 to 2050 and from 2051 to 2080 under the RCP8.5 climate scenario.

Close modal

Overall, the mean annual precipitation over the study area is projected to decrease with mild statistical significance under the RCP 4.5 scenario and with no statistically significant decreasing trend for the RCP 8.5 scenario for the periods of 2021–2080. In these periods, for both RCP4.5 and RCP8.5 scenarios, the rainfall is expected to decrease in all months of the year, except for the month of July.

Generally, the results indicate that for both future periods under both scenarios, mean annual precipitation is expected to decrease over the study area. However, mean monthly variations will be higher than annual variations. The monthly simulated rainfall under RCP8.5 is expected to be larger than under the RCP4.5 scenario in March and April. However, the annual reduction trend is significant for the RCP4.5 scenario.

Considering the seasonal change, the projected change in precipitation in the future period, the precipitation is expected to decrease in the next 60 years for RCP4.5 and RCP8.5 climate scenarios in all seasons.

Overall, mean seasonal precipitation over the study area is projected to decrease by 22.03 and 24.81.0% under RCP4.5 and RCP8.5 scenarios, respectively, for the period of 2021–2050. In this period, for both scenarios, the rainfall is expected to decrease in all four seasons of the year. The maximum percentage change under the two scenarios is projected to decrease by 56.78 and 72.32 in the winter season, respectively (Figures 6 and 7).

Figure 6

Mean annual and seasonal precipitation variation over the Melka Wakena sub-basin under the RCP4.5 climate scenario.

Figure 6

Mean annual and seasonal precipitation variation over the Melka Wakena sub-basin under the RCP4.5 climate scenario.

Close modal
Figure 7

Mean annual and seasonal precipitation variation over the Melka Wakena sub-basin under the RCP8.5 climate scenario.

Figure 7

Mean annual and seasonal precipitation variation over the Melka Wakena sub-basin under the RCP8.5 climate scenario.

Close modal

Similarly, for the period of 2051–2080, the mean seasonal precipitation is projected to decrease by 24.20 and 26.23% under RCP4.5 and RCP8.5 scenarios, respectively. The maximum percentage decrement is projected to decrease in the winter season for both RCP4.5 and RCP8.5 scenarios. The values are 57.92 and 90.98% for RCP4.5 and RCP8.5 scenarios, respectively (Figures 6 and 7).

Temperature

The Melka Wakena sub-basin exhibits an increase in projected minimum and maximum temperatures under RCP4.5 and RCP8.5 climate scenarios (Figures 8 and 9).

Figure 8

Mean annual and seasonal temperature projection in the Melka Wakana sub-basin under minimum temperature change under RCP4.5 (left) and RCP8.5 (right) climate scenarios.

Figure 8

Mean annual and seasonal temperature projection in the Melka Wakana sub-basin under minimum temperature change under RCP4.5 (left) and RCP8.5 (right) climate scenarios.

Close modal
Figure 9

Mean annual and seasonal temperature projection in the Melka Wakana sub-basin under maximum temperature change under the RCP4.5 scenario (left) and under maximum temperature change under the RCP8.5 (right) scenario.

Figure 9

Mean annual and seasonal temperature projection in the Melka Wakana sub-basin under maximum temperature change under the RCP4.5 scenario (left) and under maximum temperature change under the RCP8.5 (right) scenario.

Close modal

Mean seasonal minimum temperature varies from +2.3 to +3.7 °C under the RCP4.5 scenario and from +3.1 to +5.9 °C under the RCP8.5 climate scenario for the two time periods. Similarly, the mean maximum seasonal temperature shows an increasing trend and varies from +0.90 to +2.05 °C under the RCP4.5 scenario and from +1.55 to +3.5 °C under the RCP8.5 climate scenario of future periods. It also shows that there is an increase in daily minimum temperature higher than maximum temperature over the next 60 years under both climate scenarios with no significant trend.

The projected seasonal maximum and minimum temperatures in the Melka Wakena sub-basin, as shown in Figures 8 and 9, indicated a consistent increase during spring (March–April), summer (June–August), autumn (September–November), and winter (December–January) in both scenarios for all periods. Moreover, under CORDEX-Africa climate scenarios, the Melka Wakena sub-basin temperature projection of the RCP8.5 predicted a temperature increase higher than that of the RCP4.5 scenario.

Potential evapotranspiration

Figure 10(a) and (b) shows daily average estimated potential evapotranspiration for both RCP4.5 and RCP8.5 scenarios in the two future time horizons. In this study, evapotranspiration was calculated on the basis of Hargreaves's method, because no complete data on solar radiation, wind speed, and humidity in the study area were available, and, therefore, this method was used for this study area. The projected air temperature is expected to increase in both RCP scenarios. Therefore, there will be increased changes in evapotranspiration following the temperature increases in all months under both RCP4.5 and RCP8.5 scenarios. Under the RCP4.5 scenario, the increment will vary from −0.0229 to 0.0.4054% in 2021–2050, from July to December, and −0.01443 to 0.5590% in 2051–2080. Also, under the RCP8.5 scenario, evapotranspiration changes will vary from −0.03155 to 0.4992% and 0.05–0.5146% in 2021–2050 and in 2051–2080, respectively. Similarly, because of the annual change in potential evapotranspiration under the RCP4.5 scenario, the increments will vary from 0.1324 to 0.2679% in 2021–2050 and 2051–2080, respectively, and also under the RCP8.5 scenario, the increments will vary from 0.1611 to 0.2880%, respectively.

Figure 10

Projected monthly average percentage change in potential evapotranspiration under RCP4.5 and RCP8.5 scenarios for (a) 2021–2050 and (b) 2051–2080 compared with baseline periods (1985–2015).

Figure 10

Projected monthly average percentage change in potential evapotranspiration under RCP4.5 and RCP8.5 scenarios for (a) 2021–2050 and (b) 2051–2080 compared with baseline periods (1985–2015).

Close modal

The changing pattern of potential evapotranspiration under RCP4.5 and RCP8.5 scenarios is expected to increase in both future time horizons due to higher temperature changes in RCP4.5 than in RCP8.5.

Predicted streamflow

Flow calibration at the Melka Wakena dam site

Flow calibration was done by using the observed flow at the Melka Wakena dam site. The calibration and validation were done on a monthly basis: an NSE of 0.65, an R2 of 0.69, and a percentage volume difference (D) of 0.63 were obtained for calibration, and an NSE of 0.61, an R2 of 0.67, and a percentage volume difference (D) of 0.15 were obtained for validation. These values demonstrate that the model has good capability to simulate the observed flow future periods with less uncertainty.

Generally, these results indicate that the model parameters obtained are stable; hence, the model can be used for future streamflow generation or future flow projection.

Streamflow projection

The impact of climate change on streamflow in the two future periods was analyzed after calibration (Figure 11) and validation (Figure 12) of the SWAT hydrological model. In addition, after bias correction of precipitation and temperature, the potential evapotranspiration was calculated by the Hargreaves method, and finally, the simulated discharges for the future periods of 2021–2080 were estimated and compared with the baseline period from 1985 to 2015 to evaluate the change in streamflow. Overall, the change in precipitation for the next 60 years shows almost a statistically insignificant decreasing trend. Accordingly, the streamflow also shows an insignificant decreasing trend in the next 60 years in both scenarios.

Figure 11

Mean monthly observed vs. simulated discharge at the Melka Wakena dam site (calibration).

Figure 11

Mean monthly observed vs. simulated discharge at the Melka Wakena dam site (calibration).

Close modal
Figure 12

Mean monthly observed vs. simulated discharge at the Melka Wakena dam site (validation).

Figure 12

Mean monthly observed vs. simulated discharge at the Melka Wakena dam site (validation).

Close modal

However, the mean monthly percentage change (increment) may vary from 9.0 to 63.0 under the RCP4.5 scenario and 47.0 to 63.0 under RCP8.5 for the period of 2021–2050. The maximum percentage increment is expected in November under both scenarios where the mean monthly flow will be 59 and 68% for RCP4.5 and RCP8.5 scenarios, respectively. There is also a reduction in the percentage of flow for the period of 2021–2050. The percentage reduction is projected to vary from −1.0 to −158.0 and from −1.0 to −14.0 for the two scenarios, respectively. The maximum reduction in percentage is expected in February when the mean monthly flow will be −158.0 under the RCP4.5 scenario and −14.0 in.May for RCP 8.5. The minimum reduction will be happening in January for RCP4.5 and in April for RCP8.5 as shown in Figure 13(a).

Figure 13

Projected mean monthly percentage change in simulated discharge under RCP4.5 and RCP8.5 scenarios for 2021–2050 compared with the baseline period (1986–2015).

Figure 13

Projected mean monthly percentage change in simulated discharge under RCP4.5 and RCP8.5 scenarios for 2021–2050 compared with the baseline period (1986–2015).

Close modal

For the period of 2051–2080 (Figure 14), the mean monthly percentage change is expected to vary from −17.0 to +59.0 and −1.0 to +68.0 under both scenarios, respectively. There will be a large increment in November under both scenarios where the mean monthly flow will be +59.0 and +68.0%, respectively. The reduction in mean monthly discharge in percentage may vary from 17.0 to 55.0 under RCP4.5 and from 1.0 to 89.0 under RCP8.5. A percentage reduction in mean monthly flow will also be seen in February under both scenarios. The reductions are 55.0 and 89.0%. The minimum reduction is in January for RC4.5 and in March for RCP8.5 as shown in Figure 10(b). Desalew & Bhat (2021) also reported such similar inconsistent changes in the climate of the Rib watershed, northwestern Ethiopia. A decrease in the runoff in all periods and the RCP 4.5 and RCP8.5 scenarios was reported by Farzin & Anaraki (2021). Anaraki et al. (2021) also reported a decrease in discharge in both scenarios and periods.

Table 1

Geographic and basic information about the climate stations in the Melka Wakena sub-basin

SRStationLatitude (N)Longitude (E)Elevation (m.a.s.l.)Period (year)
Adaba 7.006 39.39 3,420 1986–2015 
Dodola 7.01 39.05 3,000 1986–2015 
Hunte 7.018 39.43 2,380 1986–2015 
Kofale 7.077 39.78 2,620 1986–2015 
Meraro 7.40 39.24 2,940 1986–2015 
SRStationLatitude (N)Longitude (E)Elevation (m.a.s.l.)Period (year)
Adaba 7.006 39.39 3,420 1986–2015 
Dodola 7.01 39.05 3,000 1986–2015 
Hunte 7.018 39.43 2,380 1986–2015 
Kofale 7.077 39.78 2,620 1986–2015 
Meraro 7.40 39.24 2,940 1986–2015 
Figure 14

Projected mean monthly percentage change in simulated discharge under RCP4.5 and RCP8.5 scenarios for 2051–2080 compared with the baseline period (1986–2015).

Figure 14

Projected mean monthly percentage change in simulated discharge under RCP4.5 and RCP8.5 scenarios for 2051–2080 compared with the baseline period (1986–2015).

Close modal

Rainfall and discharge trend test

Trends were tested in terms of annual time series using the non-parametric Mann–Kendall trend test at a significance level of 5%. According to the Mann–Kendall trend test analysis, the positive and negative values of the S statistic indicate an upward and downward trend. The significance is rejected when the absolute value of Z is less than the tabulated Z = 1.96, and the significance should be accepted (significant) when the absolute value of Z is greater than 1.96 at a significance level of 5% (Table 2). The time series of mean annual simulated discharge and rainfall of observed for the study area are shown in Figure 15.

Table 2

Statistical parameters for trend analysis and their results

VariablesYearsSZSens-SlopeTrends
Rainfall (observed) 1986–2015 −39 −0.68 −1.133 Insignificant 
Discharge (observed) 36 0.69 −12.55 Insignificant 
Rainfall (RCP4.5) 2021–2080 −115 −2.03 −3.729 Negative 
Rainfall (RCP8.5) −77 −1.43 −1.697 Insignificant 
Discharge (RCP4.5) −64 −1.24 −0.799 Insignificant 
Discharge (RCP8.5) −94 −1.84 −0.799 Insignificant 
VariablesYearsSZSens-SlopeTrends
Rainfall (observed) 1986–2015 −39 −0.68 −1.133 Insignificant 
Discharge (observed) 36 0.69 −12.55 Insignificant 
Rainfall (RCP4.5) 2021–2080 −115 −2.03 −3.729 Negative 
Rainfall (RCP8.5) −77 −1.43 −1.697 Insignificant 
Discharge (RCP4.5) −64 −1.24 −0.799 Insignificant 
Discharge (RCP8.5) −94 −1.84 −0.799 Insignificant 
Figure 15

Mean annual simulated discharge for (a) the RCP4.5 scenario and (b) the RCP8.5 scenario and (c) observed rainfall.

Figure 15

Mean annual simulated discharge for (a) the RCP4.5 scenario and (b) the RCP8.5 scenario and (c) observed rainfall.

Close modal

These trend tests of time series discharge and rainfall data show the existence of statistically insignificant climate change impacts on the study area, even though some reduction trends are visible. Habte et al. (2021) found that there were no consistent and significant temporal trends of annual and seasonal rainfall in southwest Ethiopia. A rainfall reduction for both scenarios was reported by Farrokh et al. (2021) as well. Desalew & Bhat (2021) found similar results as most of the detected trends were statistically insignificant at 5 and 10% levels of significance in their study of the Rib catchment in northwestern Ethiopia.

The results of the study can be concluded as follows:

  • Projected precipitation and temperature from ensemble CORDEX-Africa RCP scenarios have systematic errors (biases) which may lead to biased simulated streamflow which cannot be corrected by calibration of the hydrological model. Therefore, bias correction improved the quality of the climate model and the distribution mapping method corrected the biases and improved precipitation and streamflow simulations of the study area.

  • The seasonal Tmax and Tmin are consistently projected to mildly increase under both emission scenarios. Percentage increments of maximum and minimum air temperature will be higher in RCP8.5 than in RCP4.5, confirming that RCP8.5 is a higher greenhouse gas–emission scenario with a higher degree of global warming.

  • Mean monthly precipitation is projected to decrease by 33.56 and 40.69.0% for 2021–2050 under RCP4.5 and RCP8.5 scenarios, respectively, over the study area. Similarly, for the period of 2051–2080, the mean monthly precipitation is projected to decrease by 34.41 and 40.46% under both scenarios, respectively. Generally, as the result indicates, for both future periods under both scenarios, mean annual precipitation is expected to mildly decrease over the study area. However, mean monthly variations will be much higher than annual variations.

  • The precipitation in the watershed is predicted to mildly decrease, and consistent with this reduction, the streamflow in the next 60 years shows a decreasing trend, but it will not be a statistically significant trend.

The authors thank the Ethiopian Ministry of Water, Irrigation, and Energy (MWIE), Wabi Shebelle Basin Authority, and Ethiopian National Meteorological Services Agency (ENMSA) for the provision of the relevant data.

All relevant data are included in the paper or its Supplementary Information.

Abdella
K. M.
2013
The Effect of Climate Change on Water Resources Potential of Omo Gibe Basin, Ethiopia. A Dissertation of doctoral degree, Engineering Capacity Building Program (ECBP), University der Bundeswehr München Germany. (B. Bates, C. Kundzewicz, Z. W. Wu, P. Palutikof, eds). Climate Change and water Technical Paper of the Intergovernmental panel on Climate Change Geneva, Switherland IPCC Secretariat, p210
.
Anaraki
M. V.
,
Farzin
S.
,
Mousavi
S.-F.
&
Karami
H.
2021
Uncertainty analysis of climate change impacts on flood frequency by using hybrid machine learning methods
.
WaterResources Management
35
(
1
),
199
223
.
Anwar
A.
&
Adem
S. A.
2016
Climate Change Impact on Stream Flow in the Upper Gilgel Abay Catchment, Blue Nile Basin, Ethiopia
.
Bates
B. K. Z.
2008a
Climate Change and Water
.
Intergovernmental Panel on Climate Change (IPCC)
,
Geneva, Switzerland
.
Bates
B. C.
2008b
Climate Change and Water. Technical Paper of the Intergovernemental Panel on Climate Change
.
IPPC Secretariat
,
Geneva
,
Switzerland
.
Desalew
M. M.
&
Gangadhara Bhat
H.
2021
Climate change and its implications for rainfed agriculture in Ethiopia
.
journal of Water and Climate Change
12
(
4
),
1229
1244
.
Farrokh
A.
,
Farzin
S.
&
Mousav
S.-F.
2021
Meteorological drought analysis in response to climate change conditions, based on combined four-dimensional vine copulas and data mining (VC-DM)
.
Journal of Hydrology
603
(
Part C
).
https://doi.org/10.1016/j.jhydrol.2021.127135.
Farzin
S.
&
Anaraki
M. V.
2021
Modeling and predicting suspended sediment load under climate change conditions: a new hybridization strategy
.
Journal of Water and Climate Change
12
(
6
),
2422
2443
.
doi: 10.2166/wcc.2021.317
.
Gebresenbet
T. G.
2010
Modeling hydropower plant system to improve its reservoir operation
.
International Journal of Water Resources and Environmental Engineering
2
(
4
),
87
94
.
Habte
A.
,
Mamo
G.
,
Worku
W.
,
Ayalew
D.
&
Gayler
S.
2021
Spatial variability and temporal trends of climate change in Southwest Ethiopia: association with farmers’ perception and their adaptation strategies
.
Advances in Meteorology
2021
,
13
.
Article ID 3863530
.
https://doi.org/10.1155/2021/3863530
.
Hayicho
H.
,
Alemu
M.
&
Kadir
H.
2019
Assessment of land-use and land cover change effect on Melka Wakena Hydropower Dam in Melka Wakena Catchment of Sub-Upper Wabe-Shebelle Watershed, South Eastern Ethiopia
.
Agricultural Sciences
10
(
6
),
819
840
.
doi: 10.4236/as.2019.106063
.
IPCC
2012
Intergovernmental Panel on Climate Change, Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation: Special Report of the Intergovernmental Panel on Climate Change
.
Cambridge University Press
,
New York, NY
,
USA
.
IPCC
2014
Climate change 2014: synthesis report
. In:
Contribution of Working Groups I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change
(
Core Writing Team
,
Pachauri
R. K.
&
Meyer
L. A.
, eds).
World Meteorological Organization
,
Geneva
,
Switzerland
, p.
151
.
IPCC-TGCIA
1999
Guidelines on the use of scenario data for climate impact and adaptation assessment. Version 1. Prepared by Charter TR, Hulme M, Lal M (eds) Intergovernmental panel on climate change, task group on scenarios for climate impact assessment. pp. 69
.
Kundzewicz
Z. W.
2004
Change detection in hydrological records a review of the methodology
.
Hydrological Science Journal
49
(
1
),
7
19
.
Kundzewicz
Z. W.
2009
The implications of projected climate change for freshwater resources and their management
.
Hydrological Sciences Journal
.
https://doi.org/10.1623/hysj.53.1.3
.
Melkamu
T.
,
Negash
W.
&
Abdella
K.
2017
Assessment of the impacts of climate on gibe-III reservoir using reliability, resilience and vurnalbility indices
.
International Journal of Scientific & Engineering Research
8
(
10
),
1606
1628
.
Nega Gudeta Demmissie
T. A.
2008
Predicting the impact of climate change on Kulfo River flow
.
Hydrology
.
doi: 10.11648/j.hyd.20180603.11
.
Parry
M. L.
,
Canziani
O. F.
,
Palutikof
J.
,
van der Linden
P.
&
Hanson
C. P.
2007a
Impacts, Adaptation and Vulnerability: Working Group II Contribution to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change
.
Cambridge University Press
,
Cambridge
,
UK
.
Santhi
C.
,
Arnold
G.
,
Williams
R.
,
Dugus
A.
,
Sirinivasan
R.
&
Hauck
M.
2001
Validation of the SWAT model on a large river basin with point and nopoint sources
.
Journal of the American Water Resources Association
37
(
5
),
1169
1188
.
Solomon
S.
2007
Climate Change 2007-the Physical Science Basis: Working Group I Contribution to the Fourth Assessment Report of the IPCC
.
Cambridge University Press
,
Cambridge
.
Teutschbein
C. a.
2013
Is bias correction of regional climate model (RCM) simulations possible for non-stationary conditions?
Hydrology and Earth System
,
5061
5077
.
doi:10.5194/hess-17-5061-2013
.
Vandana
K.
,
Islam
A.
,
Sarth
P.
,
Sikka
A. K.
&
Kapil
H.
2019
Assessment of potential impact of climate change on streamflow: a case study of the Brahmani River basin, India
.
Journal of Water and Climate Change
.
https://doi.org/10.2166/wcc.2018.129
.
Worqlul
A. W.
,
Dile
Y. T.
,
Ayana
E. K.
,
Jeong
J.
&
Adem
A. A.
2018
Impact of climate change on streamflow hydrology in headwater catchments of the Upper Blue Nile Basin, Ethiopia
.
Water
10
(
120
),
1
18
.
Zarey
L.
2006
Climate Change Impact on Lake Ziway Watershed's Water Availability, Ethiopia
.
MSc Thesis
,
Cologne
,
Germany
.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence (CC BY 4.0), which permits copying, adaptation and redistribution, provided the original work is properly cited (http://creativecommons.org/licenses/by/4.0/).