Modeling and quantifying surface water resource potential in Awash Bello watershed, Upper Awash River Basin, Ethiopia, using SWAT model

Rising global water demand, combined with climate change, population growth, industrial development, and land use change, highlights the need for quantifying water resources for long-term planning. This study specifically focuses on quantifying surface water resource potential with its seasonal variation in the Awash Bello watershed. The surface water resource potential was quantified using ArcGIS10.4.1, Arc SWAT 2012, and SUFI-2 algorithms in Soil and Water Assessment Tool - Calibration and Uncertainty Program (SWAT-CUP). The meteorological data (1990–2022) and stream flow data (1993–2016) were utilized in the data analysis. The watershed was established to delineate 35 sub-catchments and 236 Hydrological Response Units. Calibration and validation for the SWAT model and sensitivity analysis were also performed. The results of the hydrological model for Awash Bello watershed, covering 4,490 km², estimate approximately 949.43 million cubic meters of annual surface runoff, with a runoff depth of 211.46 mm and total rainfall of 1,074.78 mm. The “Kiremt” season (June, July, August, and September) contributes to 71.21% of annual rainfall, while the “Bega” season (October, November, December, and January) accounts for only 5.58%. In contrast, the “Belg” season, which comprises February, March, April, and May, contributes to 23.21% of the annual rainfall, and it is greater than “Bega.” These results thus highlight the importance of surface water resource management in conserving the runoff and enhancing groundwater recharge toward sustainable water resource management, particularly during periods of dry season and periods of low water balance, ensuring water scarcity is mitigated.

1 Introduction

Globally, many people face severe water shortage and stress, and as development is increasing, water demand across different sectors is rising, further exacerbating the pressure (Sempewo et al., 2023; Ougougdal et al., 2020; Van Vliet et al., 2021).

In Ethiopia, the Awash Basin is an important surface water resource, supporting a lot of ecosystems and livelihoods, yet its sustainability is put in threat because of complex topography, climate variability, and high population density (Tufa, 2021). Muauz et al. (2024) depicted rising water demand because of increasing population as well as the development of socio-economy, like urbanization and intensification of agriculture, placing an escalating pressure on the upper parts of the Awash River Basin (Gizaw et al., 2017). This creates significant service gaps and increases vulnerability to water scarcity and inequalities in various dimensions (Guug et al., 2020; Worku, 2017). Based on past studies, Daniel (2024) noted that the potential of water resources is subject to changes in quality and quantity; hence, there is a need for assessment at the watershed level for sustainable development. This understanding is, therefore, crucial for developing a plan in urbanizing or urbanized areas (Negash et al., 2023) for the proper surface water potential quantification and runoff in order to ensure equitable use of the resource for water security in Ethiopia (Mahtsente et al., 2017; Zeberie, 2019; Beza et al., 2023; Yang et al., 2023). Awash Bello watershed is an important tributary to the large Awash Basin, and its potential water resources are therefore very important but are not being utilized to their full potential. Research gaps exist on current surface water availability and demand, as Singh (2014) asserts that existing studies have largely overlooked the complex dynamics of planning and management of water resources in the area, which are crucial in the improvement of the livelihood of the locals. Escalating water demand because of fast economic growth and population pressure necessitates the need for an effective assessment of resources so as to maximize benefits (Kerim et al., 2016; Gemechu et al., 2024).

Different studies used the Soil and Water Assessment Tool (SWAT) model, which focused on surface water availability and hydrological processes. Gebremichael et al. (2023) elaborate on the projected hydrological change due to climate variation in the Upper Awash River Basin using the QSWAT model. Similarly, Amin and Nuru (2020) assess the SWAT performance in modeling the Mojo river watershed stream flow, while Balcha et al. (2025) compared the NARX and SWAT models for predicting hydrological responses under varying climate scenarios. Despite these insights, wide gaps remain, especially regarding seasonal water availability and spatial surface water runoff distribution, as well as integration with socio-economic factors affecting water resources.

This study aims to address these gaps through several new contributions. It will require enhanced calibration and validation techniques to improve the SWAT model accuracy by integrating recent hydrological and climate data for updated insights and focusing on the availability of potential surface water assessment in the study area. By way of output, the research quantifies surface water potential with its seasonal variation and spatial distribution, addressing gaps in previous studies for improving water resource management strategies for the upper Awash Basin (Tolera et al., 2018). Ultimately, results will emphasize the need for comprehensive strategies to safeguard this vital resource for current and future generations, promoting sustainable usage and effective allocation of water resources to strengthen community resilience.

2 Material and methods

2.1 Study area description

The Awash River Basin covers an area of 113,304 km², from latitudes 7°53′N to 12°N and longitudes 37°57′E and 43°25′E. The river originates in the Ginchi highland at 3,000 m above mean sea level (amsl) and flows for about 1,200 km to Lake Abe, situated 250 m amsl (Tufa, 2021). Within this basin, the Awash Bello Watershed (Figure 1) covers approximately 4,489.96 km², which is located between 8°30′N to 9°00′N and 39°00E′ and 40°00E′ in Oromia. This particular catchment has a different landscape and elevation, ranging between 1,997 and 3,576 m amsl, supporting highly vegetative and agricultural areas with the Melka Kunture River significantly contributing to its flow (Namara et al., 2020). This watershed is vital for local societies, providing water for irrigation, livestock, and domestic consumption, while its subtropical highland climate influences ecological diversity and crop cultivation.

Figure 1

Figure 1. Location map of Awash Bello watershed study site in Awash River Basin, Ethiopia.

2.2 Data collection and sources

The input data used for SWAT integrate spatial, meteorological, and hydrological data to simulate surface water quantities within the watershed (Figure 2). Also, data from EMI, precipitation (PCP), temperature (TMP), solar radiation (SLD), humidity (HMD), wind speed (WND), and MoWIE (stream flow data with elevation, latitude, and longitude) of each station are crucial (Dinku and Kebede, 2022).

Figure 2

Figure 2. Flowchart of the SWAT model process for surface water resource assessment.

2.2.1 Spatial data

The spatial input data necessary for the SWAT model, such as DEM, LULC, soil, and slope maps, were obtained from different sources. A 30 m × 30 m resolution DEM (Figure 3A), obtained from https://earthexplorer.usgs.gov, was used to delineate the catchment into sub-basins, create Hydrological Response Units (HRUs; Dibaba et al., 2020), and define topographic elevation for hydrological modeling. LULC data (Figure 3B) provide crucial information on land cover types, influencing hydrological properties and, when integrated with evaluation and soil data, generating watershed characteristics. Soil map (Figure 3C), extracted from the https://www.fao.org/soils-portal/data-hub/soil-maps-and-databases/harmonized-world-soil-database, improves model accuracy by detailing texture, depth, water-holding capacity, and infiltration rates, essential for estimating surface water potential (Worqlul et al., 2018). Additionally, slope classification, derived from the DEM, categorized land into five FAO-defined classes 0–2% (flat), 2–5% (gently sloping), 5–15% (sloping), 15–30% (strongly sloping), and > 30% (sleep). This classification supports optimal surface water analysis and hydrological modeling (Figure 3D), as the SWAT model allows for flexible slope classification methods (Arnold et al., 2012).

Figure 3

Figure 3. (A) DEM, (B) land use/cover, (C) soil type, (D) slope category map for Awash Bello.

Land use significantly impacts hydrology in the watershed, with agricultural land (75.86%) dominating and shaping hydrological characteristics through HRU creation with elevation and soil data. Other uses include forest-mixed (7.78%), range-grasses (12.38%), and smaller areas of forest, brush, and residential zones that reduce water regulation, increase erosion, and affect SWAT model accuracy, highlighting the need for conservation and reforestation (Table 1).

Table 1

Table 1. Land use classification in Awash Bello watershed.

A SWAT-compatible lookup table was developed from the soil map for integration into the model. From Table 2, Pellic Vertisols were the most dominant type of soil with 2981.984 km² (66.41%) coverage areas, followed by Eutric Nitisols (9.34%) and Orthic Solonchaks (5.6%), while other soil types have much smaller coverage.

Table 2

Table 2. Major soil types in Awash Bello watershed.

2.2.2 Meteorological data

Daily time series meteorological data were utilized that were obtained from EMI (1 January 1990–31 December 2022) for 14 rain gauges within and nearby stations of Awash Bello watershed (Table 3). Parameters such as PCP, TMP, HMD, SLR, and WND were arranged for use in the SWAT model. For further analysis with weather generator (WGEN), four principal stations were selected: Addis Ababa Bole, Addis Ababa Observatory, Ambo Agriculture, and Bui, based on their data quality. They provide essential meteorological parameters and were used to generate data for incomplete record stations. Particularly, the Holeta station lacks HMD data. Seven stations have PCP and TMP data, while two have only PCP data. To address these missing values, estimations were made using data from nearby stations based on the assumption of having closely located stations in relation to the station with missing records (Hırca and Eryılmaz Türkkan, 2024).

Table 3

Table 3. Name of stations, location, and meteorological parameters for the study area.

Generally, the study has utilized spatial, metrological and hydrological data for the purpose of SWAT model (calibration, validation and simulation), land use and soil classification (See Table 4).

Table 4

Table 4. Major data for SWAT model.

2.2.3 Hydrological data

Monthly average stream flow data were utilized in the calibration and validation process of SWAT model to ensure simulated stream flow aligns with observed data, which is highly significant for accurate performance evaluation in hydrological modeling (Odusanya et al., 2021). Five gauging stations (Addis Alem, Awash Bello, Teji, Ginchi, and Melka Kuntere; Figure 4) were analyzed using daily data covering 24 years (1993–2016). Data from the watershed outlet, Melka Kunture gauge station, play a significant role in the calibration and validation process (Mihret et al., 2025; Tolera et al., 2018), elaborate reliable stream flow data to enhance the model's calibration, which is critical for obtaining valid result in scarcity of data in the Upper Awash Basin, and observe stream flow data gap in the sub-basin, which were handled by estimating hydrological parameters by regionalization and transferring calibrated parameters from gauged watersheds.

Figure 4

Figure 4. Meteorological and hydrological gauging stations.

2.3 Method

2.3.1 Quality of data

2.3.1.1 Filing missing data

Effective water resource management depends on accurate hydrological and meteorological data, necessitating quality assessment before its use in the SWAT model (Rasheed et al., 2024). Missing data, often due to rain gauge malfunctions or relocations, can lead to misleading results (Adilah and Hannani, 2021). In this study, the percentage of missing data in all methodological stations was calculated in detail using the total number of days and data to calculate the total number of missing days (Table 5). The percentage of missing data of rainfall with corresponding meteorological stations of the watershed is summarized (Table 6). Most rainfall-runoff models depend on consistent data reliability, which might be absent for days. Methods normally adopted include normal ratio (NR), multiple linear regression (MLR), and inverse distance weighting (IDW), among others (Caldera et al., 2016; Wangwongchai et al., 2023). Therefore, in this study, the nearby meteorological stations were used to estimate missing rainfall records. Hırca and Eryılmaz Türkkan (2024) and the WGEN were used to complete daily weather data, while the normal ratio method calculates rainfall gaps using each station's normal annual values, and the missing values of each station were calculated (Equation 1).

$\begin{array}{cc}{P}_X=\frac{1}{n}{\displaystyle {\sum }_{i=1}^n}\frac{N_X}{N_i}*P_i& (1)\end{array}$

where P_x is the missing precipitation for any storm at the interpolation station “x,” P_i is the precipitation for the same storm during the same period at the “ith” within a group of index station, N_x represents the normal annual precipitation value for station “x,” and N_i is the normal annual precipitation value for “ith” station.

Table 5

Table 5. Percentage of meteorological missing data calculation for the study area.

Table 6

Table 6. Summary of meteorological missing data percentage of study area.

To achieve the objectives of the study, ArcGIS 10.4.1 was employed for gathering spatial data, and the SWAT 2012 Model for evaluating hydrological conditions and modeling surface water dynamics. Supporting tools like SWAT-Checker, WGEN, and Soil and Water Assessment Tool - Calibration and Uncertainty Program (SWAT-CUP) (SWAT Calibration Uncertainty Program) utilized weather data for visualization, sensitivity analysis, uncertainty assessment, calibration, and validation of SWAT outputs (Arnold et al., 2012). This integrated approach enhances water security and evaluates freshwater availability for sustainable development (Janjić and Tadić, 2023; Regasa and Nones, 2023).

2.3.2 Model set up

SWAT is a tool that assesses management impacts on the water resources and predicts hydrological responses of flow dynamics influenced by land cover and climate changes. It incorporates key components, such as weather, hydrology, soil properties, and land management, through a hierarchical structure of sub-watersheds and HRUs based on homogeneous characteristics (Arnold et al., 2012). This method effectively balances computational efficiency with the accurate representation of the watershed's natural variability (Her et al., 2015).

The SWAT model helps to evaluate the hydrological component of Awash Bello Watershed by utilizing water balance equation water balance equation (Equation 2), simulating the land phase of the hydrological cycle at the HRU level. Also, this approach enables detailed hydrological simulation and visualizes spatial variations in water balance components across the watershed, in which a localized understanding of the hydrological process is crucial and effective for watershed management (Swami and Kulkarni, 2016).

$\begin{array}{cc}S{W}_t=S{W}_0+{\displaystyle \sum _{i=1}^t}(R_{day}-Q_{surf}-Ea-W_{seep}-Q_{gw})& (2)\end{array}$

where, SW_t is the final water contained (mm), SW₀ is the initial soil water contained on day i (mm), t: is the time (days), R_day is the amount of precipitation on day i (mm), Q_surf is the amount of surface runoff on day I (mm), Ea is the amount of evapotranspiration on day i (mm), W_seep is the amount of percolation on day i (mm), and Q_gw is the amount of return flow on day i (mm).

In this study, for the estimation of surface runoff using SWAT modeling, soil conservation service (SCS) and curve number (CN) method with daily precipitation data and the SCS curve number equation (Equation 3) were used. Also, this method, combined with daily maximum and minimum TMP, SLR, HMD, and WND speed, estimates potential evapotranspiration (PET) using the Penman-Monteith method (Abeysiriwardana et al., 2022).

$\begin{array}{cc}{Q}_{surf}=\frac{{(R_{day}-0.2S)}^2}{R_{day}+0.8S}& (3)\end{array}$

where Q_surf is the accumulated runoff or rainfall excess (mm), R_day is the depth of daily precipitation (mm), and S is the retention parameter in (mm), which can be represented as $S=25.4(\frac{100}{CN}-10)$, where CN is the curve number for the day and 25.4 is used to convert the S value to mm (Singh and Saravanan, 2020).

The SWAT model was setup for the stimulation of flow by delineating the watershed into interconnected sub-watersheds. A projected DEM of the Awash Bello watershed was extracted using GIS and imported into ArcSWAT for the analysis. The DEM was also processed to create the study area stream network and the outlet of the sub-basin. HRUs were developed using land use, soil, and slope map, with data input in ASCII (America Standard Code for Information Interchange) format for compatibility and ease of editing. After organizing meteorological data, the SWAT model was executed, allowing for accurate modeling of hydrological processes and informed decision-making in watershed management and environmental planning.

The delineation process defines streams, identifies outlet points, and calculates sub-basin parameters based on the resolution of the DEM, with sub-basin size determined by a threshold area for stream formation. Thus, in the study, DEM with 30 m × 30 m resolution and ArcGIS10.4.1 were used. Setting 9,000 hectares as the threshold area, the watershed was delineated into 35 sub-basins covering 4,489.96 km², and the outlet point was identified (Figure 5).

Figure 5

Figure 5. Delineated sub-basin of Awash Bello watershed map.

In this study, using Arc SWAT version 10.4.1, the HRU tool integrates land use and soil maps, which showed overlaps of 97.65% and 97.49%, with the delineated watershed of Awash Bello, respectively. SWAT database parameter was aligned with setting thresholds of 10% slope, 10% soil, and 15% land use. Therefore, areas below these thresholds were excluded, resulting in 236 HRUs in the Awash Bello watershed.

The meteorological data were simulated for 33 years (January 1990–December 2022), including a warm-up period for 3 years to stabilize the model, while missing values in all parameters except precipitation were set to −99 and excluded from WGEN statistical calculations. Daily weather data for 33 years from the principal stations were used to establish parameters for WGEN. All the station's data with their location and elevation were exported to WGEN, which computed overall statistics and created .txt files for each weather parameter compatible with SWAT. These WGEN statistics were then integrated into the SWAT database by importing an Excel file generated by WGEN, after which SWAT created the necessary database table. Finally, the climatic data were imported during the Arc SWAT modeling process at the input table writing stage, following the HRU analysis.

The model simulates physical systems to predict real-world performance (Janjić and Tadić, 2023), requiring realistic, process-based input parameters. Meteorological data for 33 years were sourced from the EMI, followed by a 3-year warm-up period (1990–1992) to stabilize initial values and ensure reliable results. Stream flow data of the watershed were simulated, calibrated, and validated using time series from the Melka Kunture gauge station. Results were evaluated with the SWAT-Checker software for model integrity, and predictive accuracy was enhanced through sensitivity analysis, calibration, and validation against independent datasets (Gizaw et al., 2022; Sane et al., 2020), supporting effective watershed management.

2.3.3 Sensitivity analysis and model validation

Sensitive parameters identification is the most important for the SWAT framework that enhances calibration efficiency by removing non-sensitive ones, thereby reducing calibration time and aiding in assessment at the watershed level (Duguma, 2018; Gassman et al., 2014; Guo and Su, 2019). Sequential uncertainty fitting (SUFI-2) algorithm and SWAT-CUP were conducted by evaluating monthly stream flow data of 24 years from the Melka Kunture gauge station (1993–2016) using t-stat and p-value metrics to identify sensitive parameters and improve calibration efficiency while minimizing uncertainty (Arnold et al., 2012). Therefore, the utilized parameters were from research near the study area and default SWAT results (Khalid et al., 2016).

Calibration in hydrological modeling is essential as it enhances the model's reliability and predictive capability by accurately reflecting environmental dynamics through comparisons with real-world data (Dibaba et al., 2020). Analysis of stream flow utilized the SWAT-CUP SUFI-2 algorithm, applying statistical test regression correlation coefficient (R²) and Nash–Sutcliffe model efficiency (NSE) to evaluate the model's predictions against observed data (Nash and Sutcliffe, 1970). In this study, stream flow data were divided into two-thirds for calibration (1 January 1993–31 December 2008) and one-third for validation, ensuring a strong assessment of performance.

The model validation process is the next step after calibration is completed, which includes evaluating the accuracy of models using simulations with observed data, validating their ability to produce reliable results. In this study, one-third of the total collected stream flow data from 1 January 2009 to 31 December 2016 was employed for evaluation.

Evaluating model performance by plotting simulated and observed data is important to assess the alignment and identify over- or under-prediction (Moriasi et al., 2015). Main indicators recommended by SUFI-2, such as R², NSE, percent bias (PBIAS), and root mean square (RMS), assess models' accuracy and reliability. Then the performance was evaluated depending on the criteria for the statistical test at the watershed level (Table 7).

Table 7

Table 7. Performance evaluation criteria of statistical test at watershed level (Moriasi et al., 2015).

The R² value from 0 to 1, with a value close to 1, shows strong performance, and R² greater than 0.5 is generally acceptable. Evaluation employs statistical and graphical methods to validate the model's effectiveness. The R² obtained from the equation (Equation 4):

$\begin{array}{ll}{R}^2=\frac{{\displaystyle {\sum }_{i=1}^n{(Q_{Obs,i}-{\overline{Q}}_{Obs,i})}^{*}(Q_{Obs,i}-{\overline{Q}}_{sim,i})}}{\sqrt{{\displaystyle {\sum }_{i=1}^n{(Q_{obs,i}-{\overline{Q}}_{Obs})}^2}\ast {\displaystyle {\sum }_{i=1}^n{(Q_{sim,i}-{\overline{Q}}_{sim,})}^2}}}& (4)\end{array}$

where,

• Q_{Obs, i} and Q_{Sim, i} are the observed and simulated stream flow data of i

• Q_obsand ${\bar{Q}}_{Sim}$ refers to the mean of the observed and simulated stream flow

The NSE coefficient evaluates the hydrological model prediction that shows the level of fitness of simulated and observed data. NSE values range from 1 to –∞, where 1 indicates best fit and 0 or less indicates a poor match. Then, in the study, the NSE was computed using Equation 5:

$\begin{array}{cc}NSE=1-\frac{{\displaystyle {\sum }_{i=1}^n}{(Q_{Obs,i}-Q_{Sim.i})}^2}{{\displaystyle {\sum }_{i=1}^n}{(Q_{Obs,i}-{\bar{Q}}_{Obs})}^2}& (5)\end{array}$

Moreover, the model's predictive capacity is evaluated depending on p-factor and r-factor, ranging from 0 to 100% (0–1), which shows the observed proportion in the 95% estimates the range of uncertainty, with a value above 70% considered acceptable modeling. The r-factor measures the depth of 95 PPU envelope, ideally around 1.

PBIAS evaluates the tendency of modeled values to deviate from observed, and then positive values indicate underestimation and negative values indicate overestimation, with an ideal PBIAS, which is closer to 0, reflecting accurate model predictions (Koltsida and Kallioras, 2022). The PBIS in the study were obtained from the Equation 6:

$\begin{array}{cc}PBIAS=\left[\frac{{\displaystyle {\sum }_{i=1}^n}(Q_{OBS,i}-Q_{Sim,i})}{{\displaystyle {\sum }_{i=1}^n}{Q}_{OBS,i}}\right]*100& (6)\end{array}$

3 Result and discussion

3.1 Sensitive analysis of model parameters

Process-based identification of SWAT input parameters within realistic uncertainty limits is important for model calibration, sensitivity analysis was conducted to identify parameters affecting main flow, and careful parameter selection is emphasized for effective watershed management (Tefera et al., 2023; Tesema, 2021; Abbaspour et al., 2017; Khalid et al., 2016). In this study, raking of sensitive parameters was conducted according to the sensitivity analysis by a calibrated value using a significance level of 0.05, and then the analysis revealed the depth from the soil surface to the bottom of the layer (SOL_Z), SCS runoff curve number (CN2), deep aquifer percolation friction (RCHRG_DP), threshold depth of water in the shallow aquifer required for return flow to occur (GWQMN), and available water capacity of the soil layer (SOL_AWC) exhibited the highest t-test values, as they are the most influential parameters. Therefore, 15 parameters with relatively high sensitivity (Table 8) and P-value below 5% (Figure 6) were selected in the study area for calibration, aligning with finding from similar studies (Arnold et al., 2012).

Table 8

Table 8. Sensitive parameters with calibration values for Awash Bello watershed.

Figure 6

Figure 6. Sensitive parameters of stream flow from global sensitivity analysis (AAT).

The qualifier (R_) refers to a relative change in parameter where the default values are multiplied by 1 plus a factor in the parameter range, while (V_) refers to the substitution of the default parameter by a calibrated value. The extensions .mgt, .gw, .sol, .rte, and .bsn indicate the SWAT parameter family.

3.2 Model calibration and validation

Calibration and validation of the model were conducted at the Melka Kunture outlet gauge station using average monthly stream flow data (January 1993–December 2016), data from 1993 to 2008 for calibration, and data from 2009 to 2016 for validation. The result was employed in the SUFI-2 algorithm to adjust model parameters, minimizing differences between observed and simulated data.

Although the SWAT model internally operates over a daily time step using daily meteorological inputs, calibrating the model with monthly average stream flow is a widely accepted and practical approach, especially when daily discharge data contain significant uncertainties. Gaps are a common feature in daily stream flow observations, in addition to measurement errors, due to human limitations, and are therefore less reliable for comprehensive calibration. Monthly averages tend to smooth out most of these data inconsistencies, making the calibration process emphasize broader hydrological trends rather than short-term fluctuations. Moreover, many hydrologic models, such as SWAT, tend to provide stronger performance metrics at the monthly scale (e.g., NSE and R²) even when daily results appear unsatisfactory. This makes the monthly calibration particularly useful for studies where capturing the long-term variability is more important than replicating individual peak flow events.

Calibration at a monthly scale is also useful when the objective is to assess long-term water balance components and overall watershed behavior. Monthly calibration produces more stable estimates of runoff, evapotranspiration, groundwater recharge, and seasonal water availability, which are crucial in planning and managing water resources (Gobezie et al., 2023). Additionally, it reduces the demand for computing, especially for long simulation periods like multi-decadal datasets. Due to these advantages, monthly calibration is generally done by a researcher focusing on long-term hydrological trends, like those in the Melka Kunture basin station, to ensure that the model performs well in representing the general stream flow regime. This approach helps to overcome the limitations of using daily data and the inherent challenges of the model in capturing rapid daily flow dynamics, thus realizing a more reliable overall representation of hydrological processes of the watershed.

The calibration achieved acceptable performance metrics, such as R², NSE, and PBIAS, indicating that the simulated flows are reasonably close to the data. Figures 7, 8 display the results of calibrated flow, and Table 9 summarizes the model's performance.

Figure 7

Figure 7. Average monthly time step during the calibration (1993–2008) and validation period (2009–2016). Surface flow results at Melka Kuntre gauging station.

Figure 8

Figure 8. (A) Scatter plot of Observed and Simulated flow of the Calibration period and (B) Scatter plot of Observed and Simulated flow of the Validation period of Stream Flow Hydrograph at Melka Kunture.

Table 9

Table 9. SWAT model performance summary.

The model shows a strong correlation, where R² = 0.8 and NSE = 0.8 during calibration. When compared to previous studies in the region, Amin and Nuru (2020) observed almost similar performance values of R² = 0.8, NSE = 0.75 in the upper Awash River Basin Mojo, which strengthens the reliability of our findings and underscores their implications for effective water resource management in the study area. On the other hand, Gobezie et al. (2023) observed a lower calibration performance value of R² = 0.64, NSE = 0.61 for a similar SWAT model and applied it to the Awash River Basin Borkena watershed, attributing this to limited meteorological data and small-scale hydraulic structures affecting flow data (Beza et al., 2023). However, the results met the criteria of R² > 0.6 and NSE > 0.5, recommended by Arnold et al. (2012).

Validation results from SWAT-CUP showed NSE = 0.63, R² = 0.66, and the performance assessment of the model indicates a good correlation with the observed data (Figures 7, 8). The model's overall performance during validation is summarized in Table 9.

The model achieved strong calibration performance, with NSE = 0.80, R² = 0.80, and PBIAS = −11.1 (Table 9), indicating that the simulated streamflow aligned well with the observed data during the calibration period. This high performance is partly because the calibration relied on historical streamflow datasets that were more consistent with the catchment conditions represented by the 2016 land cover map. However, the validation results showed a moderate decline, with NSE = 0.63, R² = 0.66, and PBIAS = −12.1. Although the land cover dataset remained the same, the streamflow data used for validation reflected different hydrological conditions that were less compatible. As a result, the model's predictive performance decreased slightly during validation. In general, the model performs well under historical conditions but shows reduced accuracy when applied to periods whose hydrological characteristics diverge from the conditions implied by the 2016 land cover.

The performance of the model was evaluated using statistical and graphical methods (Figure 8 and Table 9), comparing simulated and observed monthly stream flow at the outlet. Statistical analysis during calibration, R² values of 0.80, and validation R² values show strong performance, high NSE values, and minimal PBIAS (Table 9). However, consistently higher observed flows compared to simulated values raise concerns about unaccounted climate variability and land use changes, indicating a need for further investigation. Overall, the model showed effective capability and met criteria for good performance, confirming its suitability for analyzing the Awash Bello watershed.

3.3 Water balance and surface runoff estimation

The model outputs (Table 10) provide critical insights for water resources management. In this study, a stream flow to precipitation ratio is 0.32, which indicates significant water losses through evaporation and infiltration. It essentially underlines the need for integrated water management (Fors, 2022). The base flow to total flow ratio of 0.38 emphasizes the role of groundwater in sustaining stream flow. The surface runoff to total flow ratio of 0.62 raises concern about erosion and soil degradation, especially in areas with poor vegetation or heavy rainfall. Low percolation (0.12) and low deep recharge (0.01) indicates very negligible contribution to deep aquifer, highlighting the importance of surface water conservation (Akoko et al., 2021), Finally, the evapotranspiration to precipitation ratio of 0.63 signifies notable water loss, with studies indicating that only effective vegetation can help in enhancing water availability on a sustainable basis (Wanniarachchi and Sarukkalige, 2022).

Table 10

Table 10. Water balance elements and their simulated value in the study area.

After simulating and calibrating with stream flow data (1993–2008) of Awash Bello and validating using data from 2009 to 2016 by checking R², NSE, and p-test values, the results show surface water runoff depth of 211.46 mm, total rainfall of 1,074.78 mm, and lateral soil flow of 53.21 mm annually (Table 11). Then the model generates 949.43 MCM of surface runoff annually from the total watershed area of 4,489.96 km², with a maximum of 323.05 MCM and a minimum of 0.000009 MCM of surface runoff in August and November, respectively.

Table 11

Table 11. Watershed characteristics: average monthly surface water balance value per month.

The “Kiremt” rainy season (June–September) has the highest surface water balance, with a total surface runoff of 161.67 mm depth, rainfall of 765.35 mm, and lateral soil flow of 38.29 mm, for 71.21% of annual rainfall. On the other hand, the “Bega” season (October–January) shows a low surface water balance, contributing only 5.58% of the annual rainfall, characterized by surface runoff of 15.5 mm, rainfall of 59.94 mm, and lateral soil flow of 4.85 mm. The “Belg” season (February–May) shows a moderate balance, which is comparatively greater than “Bega,” with a surface runoff of 34.29 mm, rainfall of 249.49 mm, and lateral soil flow of 10.07 mm, contributing 23.21% of annual rainfall (Table 11).

Spatial distribution of surface runoff provides valuable insights for water resource management and conservation (Jia et al., 2024). Depending on the distribution of average annual surface runoff across 35 sub-basins in the catchment (Figure 9 and Table 12) that are influenced by factors like topography, land use, and soil properties, very low runoff (0–50 m³/s) in the watershed, spanning 19.67% of the area, are attributed to high infiltration or flat terrain, aligning with studies on infiltration dominated regions analyze water infiltration into soil and its impact on surface runoff (Dananto et al., 2022; Kang et al., 2024).

Figure 9

Figure 9. Surface runoff special distribution in Awash Bello watershed map.

Table 12

Table 12. Average annual surface runoff special distribution with its area coverage.

Similarly, low runoff category (50–150 m³/s), 53.42% of the area, and moderate runoff (150–400 m³/s), 19.8%, are linked to high runoff. This agrees with Mitiku et al. (2024), who highlighted the impact of slope gradients on runoff intensity. High runoff (>400 m³/s), limited to 7.11%, is associated with erosion and flood-prone areas. Al-Ghobari et al. (2020) emphasize the need for erosion control in such areas. This runoff distribution, therefore, becomes crucial for risk management and erosion mitigation, which identifies high zones needing interventions, such as vegetative buffers and rainwater harvesting. Most sub-basins fall into the low runoff category, which indicates that sustainable land use practices and hydrological planning are very important. Studies indicate that runoff variability reflects local conditions of topography and soil properties. Alshammari et al. (2023) make the map essential for hydrological studies and water resource management decisions.

4 Conclusion

This study successfully utilized the SWAT hydrological model with essential inputs, such as soil characteristics, land use data, topographic slope, and 33 years of meteorological data, to quantify the surface water resource potential with a view to sustainable resource planning. The model accuracy was ensured by sensitivity analysis, calibration, and validation against 24 years of stream flow data. The results prove that the SWAT model represents the surface water balance effectively in the Awash Bello watershed, demonstrating reliable performance through accurate simulations of evapo-transpiration (ET) and soil water content (SWC) that were validated against the gauged stream flow data. Annually, this model predicts a surface water runoff depth of 211.46 mm, which is equivalent to 949.43 million cubic meters from a watershed area of 4,489.96 km², highlighting the potential of this model as a significant water resource.

Seasonal variation in water contribution is also significant; the “Kiremit” season (June to September) contributes 71.21% of annual rainfall with a runoff depth of 161.67 mm, whereas the “Bega” season (October to January) contributes only 5.58% with 15.5 mm of runoff. The “Belg” season (February to May) contributes moderately with a runoff depth of 34.29 mm and 23.21% rainfall. With increasing temperatures, the potential evapotranspiration (PET) increases, thereby affecting the overall ET. This emphasizes the importance of seasonal rainfall and temperature in managing surface water resources of the Awash Bello watershed, contributing valuable insight for sustainable water resource planning and management that is critically needed for addressing future water demand and challenging water use in the study area.

Generally, the findings of the study provide essential information and inputs to all water resource management-related sectors, which can also be used by the different stakeholders in watershed-level planning, for policymakers in formulating long-term policies promoting sustainability and availability of water resources in the region. This should be approached in a manner that considers the local contexts and traditional practices to enhance community acceptance and effectiveness.

Therefore, future research shall expand the contribution of the current study's findings regarding surface water potential in the Awash Bello watershed by assessing the availability of groundwater and incorporating climate change projections into the SWAT model. This approach will help to evaluate the influences of climate change on both surface and groundwater availability, supporting the improvement of adaptive supply enhancement and demand management strategies. Furthermore, the integration of local knowledge in the management of water will enhance the effectiveness, cultural relevance, and resilience of such measures at the level of the local community.

Data availability statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Author contributions

RT: Project administration, Funding acquisition, Writing – review & editing, Visualization, Methodology, Formal analysis, Writing – original draft, Supervision, Validation, Conceptualization, Data curation, Resources, Software, Investigation. HW: Validation, Conceptualization, Methodology, Supervision, Project administration, Software, Resources, Formal analysis, Visualization, Investigation, Writing – review & editing, Writing – original draft, Data curation, Funding acquisition. AM: Project administration, Formal analysis, Resources, Visualization, Data curation, Validation, Methodology, Investigation, Software, Writing – review & editing, Funding acquisition, Supervision, Conceptualization, Writing – original draft.

Funding

The author(s) declared that financial support was not received for this work and/or its publication.

Acknowledgments

The authors would like to acknowledge Addis Ababa University for providing funds for this study.

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

Abbaspour, K. C., Vaghefi, S. A., and Srinivasan, R. (2017). A guideline for successful calibration and uncertainty analysis for soil and water assessment: a review of papers from the 2016 international SWAT conference. Water 10:6. doi: 10.3390/w10010006

Crossref Full Text | Google Scholar

Abeysiriwardana, H. D., Nitin, M., and Rathnayake, U. (2022). A comparative study of potential evapotranspiration estimation by three methods with FAO Penman–Monteith method across Sri Lanka. Hydrology 9:206. doi: 10.3390/hydrology9110206

Crossref Full Text | Google Scholar

Adilah, A. N., and Hannani, H. (2021). “Comparison of methods to estimate missing rainfall data for short term period at UMP Gambang,” in IOP Conference Series: Earth and Environmental Science, Vol. 682, No. 1 (IOP Publishing), 012027. Available online at: http://dx.doi.org/10.1088/1755-1315/682/1/012027 (Accessed November 21, 2024).

Google Scholar

Akoko, G., Le, T. H., Gomi, T., and Kato, T. (2021). A review of SWAT model application in Africa. Water 13:1313. doi: 10.3390/w13091313

Crossref Full Text | Google Scholar

Al-Ghobari, H., Dewidar, A., and Alataway, A. (2020). Estimation of surface water runoff for a semi-arid area using RS and GIS-based SCS-CN method. Water 12:1924. doi: 10.3390/w12071924

Crossref Full Text | Google Scholar

Alshammari, E., Rahman, A. A., Rainis, R., Seri, N. A., and Fuzi, N. F. A. (2023). The impacts of land use changes in urban hydrology, runoff and flooding: a review. Curr. Urban Stud. 11, 120–141. doi: 10.4236/cus.2023.111007

Crossref Full Text | Google Scholar

Amin, A., and Nuru, N. (2020). Evaluation of the performance of SWAT model to simulate stream flow of Mojo river watershed: in the upper Awash River basin, in Ethiopia. Hydrology 8:7. doi: 10.11648/j.hyd.20200801.12

Crossref Full Text | Google Scholar

Arnold, J. G., Moriasi, D. N., Gassman, P. W., Abbaspour, K. C., White, M. J., Srinivasan, R., et al. (2012). SWAT: model use, calibration, and validation. Trans. ASABE 55, 1491–1508. doi: 10.13031/2013.42256

Crossref Full Text | Google Scholar

Balcha, Y. A., Kaveh, K., Alamirew, T., and Malcherek, A. (2025). Comparability of NARX model to SWAT model in simulating future water resources scenarios using CMIP6 climate model outputs over UASB, Ethiopia. Stochastic Environ. Res. Risk Assess. 1–21. doi: 10.21203/rs.3.rs-4131271/v1

Crossref Full Text | Google Scholar

Beza, M., Hailu, H., and Teferi, G. (2023). Modeling and assessing surface water potential using combined SWAT model and spatial proximity regionalization technique for ungauged subwatershed of Jewuha watershed, Awash basin, Ethiopia. Adv. Civil Eng. 2023:9972801. doi: 10.1155/2023/9972801

Crossref Full Text | Google Scholar

Caldera, H. P. G. M., Piyathisse, V. R. P. C., and Nandalal, K. D. W. (2016). A comparison of methods of estimating missing daily rainfall data. Eng. J. Inst. Eng. Sri Lanka 49, 1–8. doi: 10.4038/engineer.v49i4.7232

Crossref Full Text | Google Scholar

Dananto, M., Aga, A. O., Yohannes, P., and Shura, L. (2022). Assessing the water-resources potential and soil erosion hotspot areas for sustainable land management in the Gidabo watershed, Rift Valley Lake basin of Ethiopia. Sustainability 14:5262. doi: 10.3390/su14095262

Crossref Full Text | Google Scholar

Daniel, H. (2024). Modeling surface water potential using the SWAT model combined with principal component analysis in the ungauged Gelana watershed, Ethiopia. J. Water Clim. Change 15, 703–732. doi: 10.2166/wcc.2023.548

Crossref Full Text | Google Scholar

Dibaba, W. T., Demissie, T. A., and Miegel, K. (2020). Watershed hydrological response to combined land use/land cover and climate change in highland Ethiopia: Finchaa catchment. Water 12:1801. doi: 10.3390/w12061801

Crossref Full Text | Google Scholar

Dinku, M. B., and Kebede, H. H. (2022). Assessment of irrigation potential in Jewuha Watershed, Middle Awash River Basin, Ethiopia. Int. J. Oceanogr. Aquacult. 8, 110. Available online at: https://api.semanticscholar.org/CorpusID:246889857

Google Scholar

Duguma, T. A. (2020). Application of SWAT (Soil and Water Assessment Tool) to the Abay River Basin of Ethiopia: the case of Didessa Sub Basin. Int. J. Environ. Stud. 9, 1–13. doi: 10.20944/preprints201805.0147.v1

Crossref Full Text | Google Scholar

Fors, A. (2022). Improved understanding of water balance in the Malwathu Oya River basin using SWAT and remote sensing (Master's thesis). Uppsala University.

Google Scholar

Gassman, P. W., Sadeghi, A. M., and Srinivasan, R. (2014). Applications of the SWAT model special section: overview and insights. J. Environ. Qual. 43, 1–8. doi: 10.2134/jeq2013.11.0466

PubMed Abstract | Crossref Full Text | Google Scholar

Gebremichael, H. B., Raba, G. A., Beketie, K. T., Feyisa, G. L., and Anose, F. A. (2023). Projection of hydrological responses to changing future climate of Upper Awash Basin using QSWAT model. Environ. Syst. Res. 12:25. doi: 10.1186/s40068-023-00305-8

Crossref Full Text | Google Scholar

Gemechu, Y. T., Goshime, D. W., Bushira, K. M., and Asnake, A. B. (2024). Modeling of surface water allocation under current and future climate change in Keleta Catchment, Awash River Basin, Ethiopia. Sustain. Water Resources Manage. 10:176. doi: 10.1007/s40899-024-01156-6

Crossref Full Text | Google Scholar

Gizaw, M. S., Biftu, G. F., Gan, T. Y., Moges, S. A., and Koivusalo, H. (2017). Potential impact of climate change on streamflow of major Ethiopian rivers. Clim. Change 143, 371–383. doi: 10.1007/s10584-017-2021-1

Crossref Full Text | Google Scholar

Gizaw, N. K., Daniel, W., Nichola, K., Miguel, V. O., Eva, C. P. D. P., and Martin, K. (2022). Evaluating SWAT model for streamflow estimation in the semi-arid Okavango-Omatako catchment, Namibia. Afr. J. Environ. Sci. Technol. 16, 385–403. doi: 10.5897/AJEST2022.3155

Crossref Full Text | Google Scholar

Gobezie, W. J., Teferi, E., Dile, Y. T., Bayabil, H. K., Ayele, G. T., and Ebrahim, G. Y. (2023). Modeling surface water–groundwater interactions: evidence from borkena catchment, Awash River Basin, Ethiopia. Hydrology 10:42. doi: 10.3390/hydrology10020042

Crossref Full Text | Google Scholar

Guo, J., and Su, X. (2019). Parameter sensitivity analysis of SWAT model for streamflow simulation with multisource precipitation datasets. Hydrol. Res. 50, 861–877. doi: 10.2166/nh.2019.083

Crossref Full Text | Google Scholar

Guug, S. S., Abdul-Ganiyu, S., and Kasei, R. A. (2020). Application of SWAT hydrological model for assessing water availability at the Sherigu catchment of Ghana and Southern Burkina Faso. HydroResearch 3, 124–133. doi: 10.1016/j.hydres.2020.10.002

Crossref Full Text | Google Scholar

Her, Y., Frankenberger, J., Chaubey, I., and Srinivasan, R. (2015). Threshold effects in HRU definition ofthe soil and water assessment tool. Trans. ASABE 58, 367–378. doi: 10.13031/trans.58.10805

Crossref Full Text | Google Scholar

Hırca, T., and Eryılmaz Türkkan, G. (2024). Assessment of different methods for estimation of missing rainfall data. Water Resources Manage. 38, 5945–5972. doi: 10.1007/s11269-024-03936-3

Crossref Full Text | Google Scholar

Janjić, J., and Tadić, L. (2023). Fields of application of SWAT hydrological model—a review. Earth 4, 331–344. doi: 10.3390/earth4020018

Crossref Full Text | Google Scholar

Jia, Y., Jin, J., Wang, Y., Guo, X., Du, E., and Wang, G. (2024). Evaluating the spatiotemporal distributions of water conservation in the Yiluo River Basin under a changing environment. Water 16:2320. doi: 10.3390/w16162320

Crossref Full Text | Google Scholar

Kang, H., Dou, W., Chen, L., Han, L., Sui, X., and Ding, Z. (2024). A surface water resource asset accounting method based on multi-source remote sensing data. Front. Environ. Sci. 12:1473419. doi: 10.3389/fenvs.2024.1473419

Crossref Full Text | Google Scholar

Kerim, T., Abebe, A., and Hussen, B. (2016). Study of water allocation for existing and future demands under changing climate condition: case of Upper Awash Sub River Basin. J. Environ. Earth Sci. 6, 18–19.

Google Scholar

Khalid, K., Ali, M. F., Abd Rahman, N. F., Mispan, M. R., Haron, S. H., Othman, Z., et al. (2016). Sensitivity analysis in watershed model using SUFI-2 algorithm. Procedia Eng. 162, 441–447. doi: 10.1016/j.proeng.2016.11.086

Crossref Full Text | Google Scholar

Koltsida, E., and Kallioras, A. (2022). Multi-variable SWAT model calibration using satellite-based evapotranspiration data and streamflow. Hydrology 9:112. doi: 10.3390/hydrology9070112

Crossref Full Text | Google Scholar

Mahtsente, T., Assefa, M. M., and Dereje, H. (2017). Rainfall-runoff relation and runoff estimation for Holetta River, Awash subbasin, Ethiopia using SWAT model. Int. J. Water Resourc. Environ. Eng. 9, 102–112. doi: 10.5897/ijwree2015.0601

Crossref Full Text | Google Scholar

Mihret, T. T., Zemale, F. A., Worqlul, A. W., Ayalew, A. D., and Fohrer, N. (2025). Unlocking watershed mysteries: Innovative regionalization of hydrological model parameters in data-scarce regions. J. Hydrol. Regional Stud. 57:102163. doi: 10.1016/j.ejrh.2024.102163

Crossref Full Text | Google Scholar

Mitiku, A., Deribew, K. T., Moisa, M. B., and Worku, K. (2024). Spatial evaluation of surface water irrigation potential areas to improve rural crop productivity in the Gomma district, southwestern Ethiopia. Cogent Food Agric. 10:2328424. doi: 10.1080/23311932.2024.2328424

Crossref Full Text | Google Scholar

Moriasi, D. N., Gitau, M. W., Pai, N., and Daggupati, P. (2015). Hydrologic and water quality models: performance measures and evaluation criteria. Trans. ASABE 58, 1763–1785. doi: 10.13031/trans.58.10715

Crossref Full Text | Google Scholar

Muauz, A., Berehanu, B., and Bedru, H. (2024). Spatial analysis of water budget components in the Upper Awash River Sub-basin, Ethiopia using SWAT Model. J. Earth Environ. Waste Manage. 2, 1–23.

Google Scholar

Namara, W. G., Damise, T. A., and Tufa, F. G. (2020). Rainfall runoff modeling using HEC-HMS: the case of Awash Bello sub-catchment, upper Awash basin, Ethiopia. Int. J. Environ. 9, 68–86. doi: 10.3126/ije.v9i1.27588

Crossref Full Text | Google Scholar

Nash, J. E., and Sutcliffe, J. V. (1970). River flow forecasting through conceptual models part I—a discussion of principles. J. Hydrol. 10, 282–290 doi: 10.1016/0022-1694(70)90255-6

Crossref Full Text | Google Scholar

Negash, E. D., Asfaw, W., Walsh, C. L., Mengistie, G. K., and Haile, A. T. (2023). Effects of land use land cover change on streamflow of Akaki catchment, Addis Ababa, Ethiopia. Sustain. Water Resources Manage. 9:78. doi: 10.1007/s40899-023-00831-4

Crossref Full Text | Google Scholar

Odusanya, A. E., Schulz, K., Biao, E. I., Degan, B. A., and Mehdi-Schulz, B. (2021). Evaluating the performance of streamflow simulated by an eco-hydrological model calibrated and validated with global land surface actual evapotranspiration from remote sensing at a catchment scale in West Africa. J. Hydrol. Regional Stud. 37:100893. doi: 10.1016/j.ejrh.2021.100893

Crossref Full Text | Google Scholar

Ougougdal, A.yt H., Yacoubi Khebiza, M., Messouli, M., and Lachir, A. (2020). Assessment of future water demand and supply under IPCC climate change and socio-economic scenarios, using a combination of models in Ourika watershed, High Atlas, Morocco. Water 12:1751. doi: 10.3390/w12061751

Crossref Full Text | Google Scholar

Rasheed, N. J., Al-Khafaji, M. S., Alwan, I. A., Al-Suwaiyan, M. S., Doost, Z. H., and Yaseen, Z. M. (2024). Survey on the resolution and accuracy of input data validity for SWAT-based hydrological models. Heliyon 10, 1–23. doi: 10.1016/j.heliyon.2024.e38348

PubMed Abstract | Crossref Full Text | Google Scholar

Regasa, M. S., and Nones, M. (2023). SWAT model-based quantification of the impact of land use land cover change on sediment yield in the Fincha watershed, Ethiopia. Front. Environ. Sci. 11:1146346. doi: 10.3389/fenvs.2023.1146346

Crossref Full Text | Google Scholar

Sane, M. L., Sambou, S., Leye, I., Ndione, D. M., Diatta, S., Ndiaye, I., et al. (2020). Calibration and validation of the SWAT model on the watershed of Bafing River, main upstream tributary of Senegal River: checking for the influence of the period of study. Open J. Modern Hydrol. 10, 81–104. doi: 10.4236/ojmh.2020.104006

Crossref Full Text | Google Scholar

Sempewo, J. I., Twite, D., Nyenje, P., and Mugume, S. N. (2023). Comparison of SWAT and HEC-HMS model performance in simulating catchment runoff. J. Appl. Water Eng. Res. 11, 481–495. doi: 10.1080/23249676.2022.2156401

Crossref Full Text | Google Scholar

Singh, A. (2014). Groundwater resources management through the applications of simulation modeling: a review. Sci. Total Environ. 499, 414–423. doi: 10.1016/j.scitotenv.2014.05.048

PubMed Abstract | Crossref Full Text | Google Scholar

Singh, L., and Saravanan, S. (2020). Simulation of monthly streamflow using the SWAT model of the Ib River watershed, India. HydroResearch 3, 95–105. doi: 10.1016/j.hydres.2020.09.001

Crossref Full Text | Google Scholar

Swami, V. A., and Kulkarni, S. S. (2016). Simulation of runoff and sediment yield for a Kaneri watershed using SWAT model. J. Geosci. Environ. Protect. 4, 1–15. doi: 10.4236/gep.2016.41001

Crossref Full Text | Google Scholar

Tefera, G. W., Dile, Y. T., Srinivasan, R., Baker, T., and Ray, R. L. (2023). Hydrological modeling and scenario analysis for water supply and water demand assessment of Addis Ababa city, Ethiopia. J. Hydrol. Regional Stud. 46:101341. doi: 10.1016/j.ejrh.2023.101341

Crossref Full Text | Google Scholar

Tesema, T. A. (2021). Impact of identical digital elevation model resolution and sources on morphometric parameters of Tena watershed, Ethiopia. Heliyon 7, 1–9. doi: 10.1016/j.heliyon.2021.e08345

PubMed Abstract | Crossref Full Text | Google Scholar

Tolera, M. B., Chung, I. M., and Chang, S. W. (2018). Evaluation of the climate forecast system reanalysis weather data for watershed modeling in Upper Awash basin, Ethiopia. Water 10:725. doi: 10.3390/w10060725

Crossref Full Text | Google Scholar

Tufa, K. N. (2021). Review on status, opportunities and challenges of irrigation practices in Awash River Basin, Ethiopia. Agrotechnology 10:207.

Google Scholar

Van Vliet, M. T., Jones, E. R., Flörke, M., Franssen, W. H., Hanasaki, N., Wada, Y., et al. (2021). Global water scarcity including surface water quality and expansions of clean water technologies. Environ. Res. Lett. 16:024020. doi: 10.1088/1748-9326/abbfc3

Crossref Full Text | Google Scholar

Wangwongchai, A., Waqas, M., Dechpichai, P., Hlaing, P. T., Ahmad, S., and Humphries, U. W. (2023). Imputation of missing daily rainfall data; a comparison between artificial intelligence and statistical techniques. MethodsX 11:102459. doi: 10.1016/j.mex.2023.102459

PubMed Abstract | Crossref Full Text | Google Scholar

Wanniarachchi, S., and Sarukkalige, R. (2022). A review on evapotranspiration estimation in agricultural water management: Past, present, and future. Hydrology 9:123. doi: 10.3390/hydrology9070123

Crossref Full Text | Google Scholar

Worku, H. (2017). Rethinking urban water management in Addis Ababa in the face of climate change: An urgent need to transform from traditional to sustainable system. Environ. Qual. Manage. 27, 103–119. doi: 10.1002/tqem.21512

Crossref Full Text | Google Scholar

Worqlul, A. W., Ayana, E. K., Yen, H., Jeong, J., MacAlister, C., Taylor, R., et al. (2018). Evaluating hydrologic responses to soil characteristics using SWAT model in a paired-watersheds in the Upper Blue Nile Basin. Catena 163, 332–341. doi: 10.1016/j.catena.2017.12.040

Crossref Full Text | Google Scholar

Yang, Y., Cai, S., Wang, H., Wang, P., and Li, W. (2023). Evolution of hydrological conditions and driving factors analysis of the Yongding river in a changing environment: a case study of the Xiangshuipu section. Agronomy 13:2289. doi: 10.3390/agronomy13092289

Crossref Full Text | Google Scholar

Zeberie, W. (2019). Modelling of rainfall-runoff relationship in Big-Akaki watershed, upper Awash Basin, Ethiopia. World News Nat. Sci. 27, 108–120. EISSN 2543–5426.

Google Scholar