Journal of Hydrology: Regional Studies 65 (2026) 103475

Contents lists available at ScienceDirect

Journal of Hydrology: Regional Studies

journal homepage: [Link]/locate/ejrh

Floodplain inundation modeling and mapping using integrated

HEC-HMS and HEC-RAS model on Dirma River Watershed, Lake
Tana Basin, Ethiopia
Fitamlak T. Fikadie a,* , Banteamlak K. Abebe b, Abreham M. Belete a ,
Abebe G. Chekole a , Abrham B. Wassie a , Adugnaw N. Endalew a , Alebel A. Bizuneh a ,
Henok A. Bonger c , Sendeku A. Kebad d, Gebrehiwot G. Kassie a , Aniley T. Tesfaye a
a
Department of Hydraulic and Water Resources Engineering, University of Gondar, Gondar, Ethiopia
b
Department of Hydraulic and Water Resources Engineering, Dilla University, Dilla, Ethiopia
c
Department of Civil Engineering, University of Gondar, Gondar, Ethiopia
d
Department of Physics, University of Gondar, Gondar, Ethiopia

A R T I C L E I N F O A B S T R A C T

Keywords: Study region: Dirma River Watershed, Lake Tana Basin, Ethiopia.
ArcGIS Study focus: Floods are devastating, damaging natural disasters with major environmental, social,
Dirma River watershed and economic consequences. This study focuses on modeling and mapping floodplain inundation
HEC-HMS
in the Dirma River watershed to analyze flood frequency, identify sensitive hydrological pa­
HEC-RAS
Floodplain inundation
rameters, and generate inundation maps for risk management. Hydro-meteorological, geometric,
and spatial datasets were collected from on-site measurements and governmental sources, then
pre-processed and quality tests were controlled. Flood frequency was analyzed using EasyFit
software, rainfall-runoff model was simulated using HEC-HMS, and flood extent was determined
using HEC-RAS model. The model was calibrated (2000–2013) and validated (2014–2020).
New hydrological insights for the region: Model sensitivity analysis indicated that the Soil Conser­
vation Services Curve Number (SCS-CN-CN), Soil Conservation Services Curve Number Initial
Abstraction (SCS-CN-Ia), and SCS Unit Hydrograph-Lag Time are the most influential parameters.
The model demonstrated good performance, with coefficient of determination (R2) values of 0.82
(calibration) and 0.87 (validation), and Nash-Sutcliffe efficiency (NSE) values of 0.73 and 0.78,
respectively. Using simulate peak discharges from HEC-HMS, the HEC-RAS model generated
floodplain inundation maps, showing inundation extents from 9.58 km² (2-year return period) to
11.68 km² (100-year return period). These results highlight increasing flood risks and emphasize
the need for integrated flood management strategies, including structural and non-structural
measures such as levees, afforestation, and land-use planning.

* Corresponding author.
E-mail address: fithiamlak19@[Link] (F.T. Fikadie).

[Link]
Received 31 October 2025; Received in revised form 20 April 2026; Accepted 21 April 2026
Available online 4 May 2026
2214-5818/© 2026 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY license
([Link]
F.T. Fikadie et al. Journal of Hydrology: Regional Studies 65 (2026) 103475

1. Introduction

Flooding is a frequent and destructive natural hazard worldwide, causing substantial socioeconomic and environmental damage,
including loss of life, property, and infrastructure (Acosta-España et al., 2024; Manzoor et al., 2022). Effective flood-warning systems
can assist in taking appropriate action during natural calamities and may save more lives (Mehta et al., 2022). As a widespread natural
disaster, flooding produces significant impacts on floodplains (Devitt et al., 2023). Floods occur when excessive precipitation or runoff
overflows riverbanks, reservoirs, or drainage systems beyond their carrying capacity (Asiedu, 2020; Javadinejad, 2022). These results
were coming from complex hydro-meteorological interactions, including intense rainfall, overtopping of river channels, and elevated
groundwater levels (Sharma et al., 2024; Rawat et al., 2023). Globally, floods have the greatest impact on more people than any other
disaster, affecting hundreds of millions of people annually (Wang et al., 2024; Chan et al., 2023).
Flooding is common in Ethiopia, especially in areas that are dominated by monsoon rainfall, where roughly 80% of the rainfall
happens from June to September (Gemeda et al., 2021). In Ethiopia, there are two main types of flooding: the first one is caused by
heavy rainfall, while the second one arises due to dam releases and intense local rainfall (Belay et al., 2023; Mphefu, 2024). One of the
areas that is heavily affected by flooding is the Dirma River basin, which is part of the Lake Tana sub-basin. Flooding often occurs in the
Dirma River basin during the rainy period from June to September due to overbank flows that are produced by heavy rains upstream in
the highlands (Desalegn and Mulu, 2021). These events damage crops, displace communities, and exacerbate food insecurity (Cherinet
et al., 2022; Bryson et al., 2024).
The identification and accurate mapping of areas that are vulnerable to flooding is fundamental for efficient flood risk management
(Bodoque et al., 2023; Grigg, 2023). The flooding mitigation strategies may be applied and include structural measures such as levees
and embankments or non-structural approaches like hazard mapping, forecasting, and early warning systems (Pandey and Swarnkar,
2025). In developing countries, where financial and technical capacities are limited, non-structural solutions are often more practical
and cost-effective (Ashour et al., 2024; Apollonio et al., 2025). Non-structural flood management relies on flood inundation modeling
to assess the spatial and temporal extent and potential impacts of flooding (Kumar et al., 2023; Navarro, 2024). The HEC-RAS model,
integrated with hydrological inputs from HEC-HMS, is widely applied for simulating river hydraulics and flood dynamics (Ansarifard
et al., 2024; Abd-Elhamid et al., 2025). The model combines topographic, hydrologic, and hydraulic data, such as river geometry,
discharge, roughness coefficients, and terrain characteristics, to analyze flood behaviour under different scenarios (Nagarajan et al.,
2022).

Fig. 1. Location map of the study area.

2
F.T. Fikadie et al. Journal of Hydrology: Regional Studies 65 (2026) 103475

Consequently, this study aims to model floodplain inundation in the Dirma River watershed, Lake Tana Basin, Ethiopia, using an
integrated HEC-HMS and HEC-RAS approach. More precisely, it intends to: (i) select the optimal flood frequency distribution method
for annual peak discharge, (ii) identify the most influential parameters affecting rainfall-runoff simulation in the HEC-HMS model; and
(iii) create a map of the spatial extent of floodplain inundation across the selected return periods using the HEC-RAS model. The
motivation for conducting this research study comes from the research question of whether the integrated hydrologic-hydraulic model
approach is able to properly simulate runoff from a watershed and identify potential areas prone to flooding.
The importance of this research study is in generating flood inundation information for supports land-use planning, disaster
preparedness, and early warning strategies in the Lake Tana Basin or other basins that lack hydrological data. This research is also
important in developing evidence-driven, cost-effective, non-structural flood management systems, which are especially crucial for
countries like Ethiopia facing technical limitations and financial challenges.

2. Material and methods

2.1. Description of the study area

The Dirma River watershed is located in the Lake Tana sub-basin, which is a constituent of the larger Upper Blue Nile basin. With
regards to geography, the Dirma River watershed is located in the Central Gondar Zone, and it spans over the East Dembia (Koladiba)
and West Dembia (Chuahit) Woreda. These administrative areas are located in the north-western parts of Ethiopia. The watershed
exists within a specific geographic location, which is between 12◦ 16' – 12◦ 21' N latitude and 37◦ 18' – 37◦ 22' E longitude. The wa­
tershed's elevation varies considerably, with a high altitude range spanning from a minimum of 1793 m to a maximum of 2760 m above
the average sea level. This elevation and other details are depicted by the map that has been illustrated in Fig. 1. Climatic data about
the watershed is gathered through a network of meteorological stations situated in Ayikel, Ayimba, Chuahit, Gondar, Gorgora, and
Koladiba, ensuring comprehensive monitoring of the region's weather patterns. Collectively, the Dirma River watershed covers an area
of approximately 354 km2. In terms of precipitation, the watershed experiences a mean annual rainfall of 1225 mm. However, the
annual rainfall is not constant and fluctuates from receiving as little as 850 mm to others experiencing rainfall totals as high as
1600 mm. The monthly maximum temperatures in the area range from 23.3 ◦ C to 32.4 ◦ C, while the monthly minimum temperatures
fall between 11.8 ◦ C and 15.7 ◦ C.

2.2. Topography

The Dirma River watershed’s topography (Fig. 2a), ranging from 1793 to 2760 m above mean sea level, includes plateaus, plains,
escarpments, valleys, and flatlands. Topography is categorized using modified Food and Agriculture Organization (FAO) slope gra­
dients (Balasubramani et al., 2019), as detailed in Table 1 and Fig. 2b, and ranges from flat to very gently sloping to very steep.

2.3. Climate

Ethiopia's climate is divided into five zones based on altitude and temperature: Bereha (hot and hyper-arid, <500 m), Kola (hot and
arid, 500–1500 m), Woina Dega (warm temperate, 1500–2300 m), Dega (temperate highland, 2300–3300 m), and Wurch (cold,

Fig. 2. Elevation (a) and slope (b) map of the Dirma River watershed.

3
F.T. Fikadie et al. Journal of Hydrology: Regional Studies 65 (2026) 103475

Table 1
Slope class of the Dirma River watershed.
Class Gradient Area coverage Weighted Slope class

​ (%) (km2) Percentage (%) Description

1 1–2 34.20 9.65 flat to very gentle slope
2 2–5 74.22 20.94 gently sloping
3 5–10 114.06 32.18 Sloping
4 10–15 60.00 16.93 strongly sloping
5 15–30 62.91 17.75 moderately sloping
6 30–60 8.80 2.48 Steep
7 > 60 0.25 0.07 very steep
Total ​ 354.44 100.00 ​

>3300 m) (Wubaye et al., 2023). The Dirma watershed, situated between 1793 and 2760 m, falls within the Dega and Woina Dega
zones and exhibits considerable climatic variation, with mean monthly rainfall ranging from 0.01 mm at Ayimba station to 12.06 mm
at Gondar station (Fig. 3). Mean monthly minimum temperatures range from 11.8 ◦ C (Gorgora) to 15.7 ◦ C (Gondar). Meanwhile, mean
monthly maximum temperatures range from 23.3◦ C (Gondar) to 32.4◦ C (Koladiba). These climatic conditions are illustrated in
Table S1, Table S2, Figure S1, and Figure S2 based on the collected input data over 23 years (1998–2020).

2.4. Soil type

Soil classification by the Ministry of Water and Energy's five soil types dominates the watershed, as seen in Table S3 and Figure S3.
Eutric Vertisols are the most prevalent (36.74%), followed by Eutric Leptosols (31.69%), Eutric Fluvisols (17.57%), and Chromic
Luvisols (13.99%). Water bodies constitute a negligible fraction (0.003%).

2.5. Land use land cover

Land use land cover (LULC) recognized 14 categories in the watershed. Intensely farmed land is the most dominant, covering
48.5%, followed by moderately farmed land (21.6%) and farm villages (18.2%). The remaining 11.7% includes grassland, shrubland,
and other minor uses, which were described in Table S4 and Figure S4.

2.6. Data collection and analysis

Data collection describes gathering information from various sources to test hypotheses, inform decisions, or gain insights, using
methods like surveys, interviews, observations, and automated digital data collection. Data analysis, on the other hand, focuses on
interpreting collected data to identify patterns, trends, and relationships through statistical analysis, data visualization, and machine
learning. Both primary and secondary datasets were used for the study.
The primary geometric data were collected through field surveys using a total station to obtain cross-sectional coordinates and river
elevation profiles required for HEC-RAS geometry development. The secondary data, including daily Streamflow data for the Drima
River gauging station, were obtained from the Ministry of Water and Energy (MoWE) and used for model calibration and validation
(Table S5). Meteorological data included daily rainfall, minimum and maximum temperature, wind speed, relative humidity, and

Fig. 3. Mean monthly rainfall distribution of the Dirma River watershed station.

4
F.T. Fikadie et al. Journal of Hydrology: Regional Studies 65 (2026) 103475

sunshine hour records for five meteorological stations, namely Gondar Azezo, Gorgora, Koladiba, Ayikel, and Chahit, covering the
period from 1998 to 2020, obtained from the Ethiopian National Meteorological Service Agencies (NMSA). These data were used as
input to estimate peak discharge at 2, 10, 25, 50, and 100-year return periods. The percentage contribution of each metrological station
to the study area was interpolated using the Thiessen polygon method in ArcGIS (Table S6 and Figure S5). Spatial datasets included a
12.5 m resolution DEM obtained from the Alaska Satellite Facility (ASF) by visiting the website [Link], and the data
related to LULC and soil conditions obtained from the MoWE were used for watershed delineation, slope classification, drainage
network analysis, and RAS-Mapper and HEC-RAS model input preparation for floodplain inundation modeling.

2.7. Data pre-processing and checking

Data errors, stemming from faulty recorders or measurement devices, necessitate thorough examination and removal before
detailed analysis. Data consistency was verified, and missing values were imputed accordingly.

2.7.1. Missing data estimation

The missing data were estimated using the arithmetic mean method, the normal ratio method, and the inverse distance method.
The arithmetic mean method was adopted for the percent of missed less than 10%, while the inverse distance and normal ratio methods
were used for the percent of missed greater than 10%. In this study, the arithmetic mean was adopted based on the above expression of
the missing data estimation techniques using the formula as in (Eq. 1).
p1 + p2 + p3 + . . . pn
Px = (1)
n

Where: - PX is the precipitation for the station with missed records, P1, P2, P3, Pn are corresponding precipitations at the index station,
and n is the number of index stations.

2.7.2. Data quality control tests

[Link]. Consistency test. Double mass curve analysis (Fikadie et al., 2022) confirmed the consistency of rainfall data across all sta­
tions. As a sample, the Gondar Azezo rainfall stations demonstrated a data consistency test (Fig. 4).

[Link]. Homogeneity test. The areal precipitation homogeneity test verified the uniformity of rainfall (Franco and Tesfaye, 2020),
Fig. 5.

[Link]. Outlier test. Outliers can indicate data quality issues, and clearly erroneous outliers should be removed or treated as missing
data. However, genuine extreme values, while potentially skewing statistics, contain valuable climatological information. Outlier
detection was conducted using skewness and threshold methods. Skewness (Cs) was computed as in Eq. (2):

Fig. 4. Double mass curve of contributed stations.

5
F.T. Fikadie et al. Journal of Hydrology: Regional Studies 65 (2026) 103475

Fig. 5. Rainfall station homogeneity test.

∑
N (Y − Ym )3
Cs = (2)
(N − 1)(N − 2)δn 3

Where: Cs is the skewness of the sample data, N is the number of sampling data, and Y is the logarithm of the sampling data, Ym is the
mean logarithm of sampling data, δn is the mean standard error of the sampling data and Xm is the mean of sampling data. By esti­
mating the skewness of the data, check whether the data is higher, lower, or both lower and higher outliers for a sampling data of
Gondar Azezo stations (Table 2).

a) Test for a higher outlier

YH = Ym + Kn ∗ σy

Y= 10YH

In the above formula, the value of Y is greater than the corresponding maximum value of the given data; the data is acceptable
and adequate; otherwise not adequate.
b) Test for lower outlier
YL = Ym − Kn ∗ σy

Y = 10YL

Whereas the value of Y is less than the corresponding minimum value of the given data, the data is acceptable and adequate; otherwise
not adequate.
Coefficient of skewness (Cs) is 0.25, falling within − 0.4–0.4, indicating the need to test both lower and upper outliers. The
calculated bounds were Ylower = 1.43 mm and Yhigher = 1.83 mm, while the observed range (1.62–1.77 mm) confirmed the adequacy
of the dataset.

2.8. Regionalization methods in ungauged catchments for flow prediction

Regionalization techniques include transferring calibrated data from a gauged catchment to an ungauged one with similar char­
acteristics, even with limited Streamflow records (Guo et al., 2021; Singh and Devi, 2022). This method calibrates, analyzes, and

Table 2
Outlier checking standards.
Case 1 If Skewness (Cs) < -0.4 Check for lower outliers
Case 2 If Skewness (Cs) > +0.4 Check for higher outliers
Case 3 If Skewness (Cs) b/n − 0.4 < CS < +0.4 Check for both outliers

6
F.T. Fikadie et al. Journal of Hydrology: Regional Studies 65 (2026) 103475

validates rainfall-runoff models and watershed characteristics using the gauged catchment before transferring them to the ungauged
catchment (Singh and Devi, 2022).
Several regionalization approaches, including spatial proximity, arithmetic averaging, physical similarity, regression analysis, and
hydrologic similarity (Qi et al., 2022). Although study sites were initially gauged up to 2010 G.C., data from 2011 to 2020 were used
via a regionalization approach to obtain more current data and improve findings, to strengthen model reliability and ensure temporal
continuity (Guo et al., 2021). For this study, due to the research catchments' similar drainage area, channel length, elevation, slope,
temperature, precipitation, vegetation cover, and soil types, the physical similarity technique was applied (Cupak, 2020). The
regionalization approach for this study estimates ungauged discharge based on the physical similarity of catchments (Barbhuiya et al.,
2023; Narbondo et al., 2020) using the formula as expressed in (Eq. 3).
{ }
Aung Pung
Qung = ∗ Qg ∗ (3)
Ag Pg

Where: - Qung is discharge at un-gauged site (m3s− 1), Aung is drainage area of un-gauged site (km2), Pung is areal rainfall at the un-
gauged site (mm), Qg is discharge at gauged site (m3s− 1), Ag is drainage area at gauged site (km2), and Pg is areal rainfall at the
gauged site (mm).

2.9. Identifying fitted flood probability distribution methods

To determine the best flood probability distribution, several common methods were tested for their goodness-of-fit to the
Streamflow data (Tegegne et al., 2020). The models considered encompassed Normal, Log-Normal, Log-Pearson Type III, Generalized
Extreme Value, and Gumbel distributions (Montes-Pajuelo et al., 2024).
Using EasyFit software, the statistical distributions were applied to a sample of the Streamflow data (Mengistu et al., 2024; Al-Salih,
2025). Chi-square, Anderson-Darling, and Kolmogorov-Smirnov tests were used to assess the performance of each distribution's
goodness of fit (Ghayyadah, 2022). The significance level for all tests was α = 0.05 (Alam et al., 2018). The flood probability best-fit
distribution agreed most closely with the observed data, exhibiting the lowest statistical value across all tests (Tegegne et al., 2020).

2.10. HEC-HMS model setup

Simulation of rainfall-runof process in the Dirma River watershed was conducted via the Hydrologic Modeling System (HEC-HMS)
modeling technique that followed the six steps described below.
Basin delineation and characterization: In GIS, river reaches, sub-basins, and basin boundaries were delineated through the use of
the digital elevation model (DEM). Basin physical properties, including its area, slope, and stream length, were obtained.
Meteorological model: The creation of the meteorological model entailed rainfall allocation to each sub-basin through the Thiessen
polygon method using precipitation data recorded in the weather stations.
Loss method: For the calculation of the amount of water losses resulting from the infiltration process during rainfall events, the
Initial and Constant Rate Loss Method, based on initial loss (mm) and constant infiltration rate in mm/hr was applied.
Transform method: The transform method entailed the conversion of excess rainfall into a direct runoff hydrograph based on the
sub-basin lag time and the SCS unit hydrograph method.
River routing: The Muskingum-Cunge method correctly routes the flow hydrograph by simulating the movement of water in
different sections of the river.
Models of calibration and validation: Calibration of the HEC-HMS model entailed adjusting the model parameter values using the
flow data of the Dirma River, and validation was done using the other set of data observed.

2.10.1. Sensitivity analysis, calibration, and validation of the HEC-HMS model

Sensitivity analysis was carried out through the one-at-a-time parameter variation method by changing one model parameter while
other parameters were kept constant (Abbenante et al., 2024). The influence of the shape of the hydrograph, peak discharge, runoff
volume, and objective function values was examined to identify the most influential parameters. Based on this procedure, SCS-CN,
initial abstraction, and lag time were found to be the most sensitive parameters in the Upper Blue Nile Basin and Lake Tana
sub-basin (Bolt, 2025). Yimer (2020) Sensitivity analysis identified Soil Conservation Service Unit Hydrograph (SCS-UH), starting
discharge, recession constant, lag time, Soil Conservation Service Curve Number (SCS-CN), and initial abstraction as the HEC-HMS
model parameters with the greatest impact on simulation results. Through modifying each parameter and assessing its impact on
the objective function(s), it is possible to ascertain which parameters have the greatest impact on model simulation (Haile et al., 2016).
Streamflow data for the period of 1998–2020 were used to calibrate and validate the model after a 2-year warming-up period. Model
parameters were adjusted during a 14-year calibration period (2000–2013). The parameters developed through the calibration process
were then evaluated independently using 7 years of validation periods (2014–2020) to achieve high NSE and R2 for the whole study
period (Ayalew and Bharti, 2022).

2.10.2. Evaluation of model performance

The HEC-HMS model's performance was evaluated to ensure accuracy and dependability (Belayneh et al., 2020). Both objective
and subjective evaluations were performed by comparing simulated outputs with observed data (Ruman et al., 2020), using the

7
F.T. Fikadie et al. Journal of Hydrology: Regional Studies 65 (2026) 103475

Nash-Sutcliff Efficiency (NSE) and coefficient of determination (R2) as key statistical metrics. The R2 measures the predictability of
observed data variation from simulated data, ranging from 0 (no correlation) to 1 (perfect match) (Nasiri et al., 2020).
The NSE assesses the fit between simulated and observed data relative to a 1:1 line, representing the difference between residual
and measured data variance. An NSE of 1 indicates a perfect match, 0 means the model is no better than using the mean of observed
data, and a negative value signifies that the observed data mean is a better predictor than the model (Waseem et al., 2017). Eqs. 4 and 5
illustrate the estimation formula of the coefficient of determination and the Nash-Sutcliffe Efficiency.

{ Σ (Qsi − Qsm ) ∗ (Qoi − Qom ) }2

R2 = (4)
Σ(Qsi − Qsm )2 ∗ (Qoi − Qom )2

Σ(Qoi − Qsi )2
NSE = 1 − (5)
Σ(Qoi − Qom )2

Where: - Here, m and s represent measured and simulated values, respectively, and Q is a variable.
The model adopted minimal suggested values for R2 > 0.6 and NSE > 0.5, as per Kasuni (2017) and Shame Usman (2022), ensuring
adequate calibration and validation. Table 3 displays the performance rating of the R2 and NSE statistics.
In this study, model performance was primarily evaluated using NSE and R², which are usually practical in hydrological modeling
and provide robust measures of fit and predictive strength. Although Root Mean Square Error (RMSE) and Percent bias (PBIAS) were
not included in this study analysis, these indicators can provide additional insight into model residuals and bias, and are recommended
for future studies to further strengthen model performance evaluation.

2.11. Structure of HEC-RAS model

The Hydrologic Engineering Center–River Analysis System (HEC-RAS) was employed to perform one-dimensional (1D) hydraulic
simulation and floodplain inundation mapping for the Dirma River (Mehta et al., 2022). The model was driven by design flood
hydrographs generated from the calibrated HEC-HMS model (Mehta et al., 2023). The HEC-RAS setup consisted of geometric data
preparation, assignment of hydraulic parameters, input of design flows, hydraulic simulation, and generation of inundation maps
(Mehta et al., 2022).
The geometric representation of the river system was developed using a 12.5 m resolution Digital Elevation Model (DEM) in RAS-
Mapper. The river centerline, bank lines, flow paths, and cross-sections were extracted from the terrain data, and field-surveyed cross-
sectional information obtained using a total station was incorporated to improve geometric accuracy. This combined dataset provided
the basis for representing the channel and overbank topography within the 1D HEC-RAS framework (Mehta et al., 2022).
Subsequently, the hydraulic parameters were assigned in the model. The Manning roughness coefficient values for the channel and
overbank zones were determined according to land use and land cover information and observations from the site conditions (Mehta
et al., 2022). These roughness values were used to calculate resistance to flow in the channel and overbank zones (Soliman et al.,
2022).
The hydrological inputs to the hydraulic model included the simulated peak flows of 2, 10, 25, 50, and 100-year return periods,
which were obtained from the calibrated HEC-HMS model. The outputs generated by this model were used to determine the design
discharge of the river at the cross-sections, which were further used to run the hydraulic simulations in HEC-RAS. The hydraulic model
was run using the 1D flow regime to obtain information on the hydraulic parameters along the channel and floodplain (Mehta et al.,
2022).
Once the simulation process was complete, the results were extracted and processed, including water surface elevation, flood
depth, and flow velocity in RAS-Mapper (Mehta et al., 2022). These results were integrated with the DEM to produce flood inundation
maps for the selected return periods. The resulting maps were used to delineate the spatial extent of flooding and to support flood
hazard assessment in the Dirma River watershed.

2.11.1. Mapping of floodplains inundation

The floodplain inundation maps were developed using RAS-Mapper through the outputs produced by the 1D HEC-RAS model
(Mehta et al., 2022). First, the DEM was developed in the projected raster form and served as the basis of terrain for developing the
hydraulic model (Fig. 6a). The terrain map is critical for obtaining geometric data for the simulation process, such as the center line of
the river, banks, flow path, and cross-sections (Greer, 2019).
Digitization of the centerline of the river, the banklines, and the direction of water flow was done in the upstream-to-downstream
direction along the natural path of the Dirma River. Next, cross-section cut lines were established across the river and the floodplain

Table 3
Overall performance evaluation of suggested statistics.
Performance evaluation

Statistics Excellent Very good Good Fair Satisfactory

R2 0.9–1 0.75–0.9 0.65–0.75 0.5–0.65 0–0.5

NSE 0.9–1 0.75–0.9 0.65–0.75 0.5–0.65 0–0.5

8
F.T. Fikadie et al. Journal of Hydrology: Regional Studies 65 (2026) 103475

Fig. 6. (a) Terrain TIN and (b) River geometry cross-section of the Dirma River.

surrounding it at appropriate intervals. The objective of establishing such cross-sections was to capture variations in the geometry of
the river as well as the elevation differences of the floodplain on either side of the river. It was ensured that the cross-sections were
drawn such that they intersect the centerline of the river without overlapping one another (Fig. 6b).
Once the geometric data had been prepared, hydraulic parameters and flow conditions were considered. The peak discharges for
the return periods of 2, 10, 25, 50, and 100-years, obtained from the calibrated and validated HEC-HMS model, were provided as input
in HEC-RAS. For the main channel, the manning roughness coefficient is set at 0.05, while for the right overbank and the left overbank,
it is taken as 0.04 and 0.045, respectively, which reflects the hydraulics of the channel and its flood plain. Normal depth was used as the
boundary condition, and the simulation was conducted with steady-state flows for a 1D hydraulics framework (Mehta et al., 2022).
The simulated water surface elevation was then combined with the terrain model in RAS-Mapper to delineate the extent of flood
inundation for each return period. In addition to flood extent, hydraulic outputs such as flow depth and velocity were also derived from
the model results. These outputs formed the basis for the final flood inundation maps, which were used to identify flood-prone areas
and to evaluate the spatial pattern of flooding along the Dirma River floodplain (Naghibi et al., 2017).

3. Results and discussion

3.1. Best fit probability distribution

Five probability distributions were fitted to annual watershed discharge data using EasyFit, and their goodness-of-fit was assessed

Table 4
Comparison of goodness of fit statistics.
No. Distribution Kolmogorov Smirnov Anderson Darling Chi-square

Statistic Rank Statistic Rank Statistic Rank

1 Gamma 0.09829 1 0.426 2 0.604 3

2 Gen. Extreme Value 0.15188 2 0.378 1 0.564 2
3 Gumbel Max 0.26926 4 0.463 4 0.492 1
4 Lognormal 0.19549 3 0.438 3 1.078 5
5 Normal 0.27587 5 0.484 5 0.896 4

9
F.T. Fikadie et al. Journal of Hydrology: Regional Studies 65 (2026) 103475

using Chi-Square, Anderson-Darling, and Kolmogorov-Smirnov tests. The Anderson-Darling and Kolmogorov-Smirnov tests indicated
that Gamma, Generalized Extreme Value, and Lognormal distributions provided a good fit, while the Chi-Square test identified the
Gumbel Max distribution as the best fit.
A comprehensive comparison of the three tests, the Kolmogorov-Smirnov test was identified as the most reliable for the discharge
data based on EasyFit analysis, a conclusion supported by prior research, such as Franco and Tesfaye (2020) in the Gilgel Abay
watershed. Statistical values and ranks for each distribution are presented in Table 4.

3.2. HEC-HMS model sensitivity analysis

Sensitivity analysis of the HEC-HMS model for the Dirma River Watershed indicated that the Soil Conservation Services Curve
Number Factor (SCS-CN-CN factor), Soil Conservation Services Curve Number Initial Abstraction (SCS-CN-Ia), SCS Unit Hydrograph-
Lag Time, Recession Constant, Initial Discharge, and Sub-reaches Parameters significantly influenced rainfall-runoff simulations.
Curve number (CN) is the key parameter for this watershed; it indicates the factor that determines the proportion of the surface runoff
and infiltration. Any slight changes in CN lead to considerable variations in runoff volume, evapotranspiration rate, and groundwater
recharge since CN depends on land use types, soils, and antecedent moisture conditions. In turn, the considerable significance of initial
abstraction and lag time highlights the role of proper representation of the hydrograph constructions (Table 5).
The outcomes of the study coincide with hydrological studies and investigations carried out previously in the Lake Tana Basin. Prior
research by Franco and Tesfaye (2020) on CN and lag time identified the parameters that have the strongest influence on the outcome
in the Gilgel Abay watershed; these results were replicated in the Gumara watershed by Belay et al. (2025) and Chekole et al. (2024),
who also confirmed the critical role of CN and lag time. The consistent significance of CN-related parameters across these studies
highlights their general importance for accurate runoff modeling with HEC-HMS, although parameter sensitivity ranking may vary
between watersheds.

3.3. Calibration and validation of the HEC-HMS model

The performance of the HEC-HMS model during calibration (2000–2013) and validation (2014–2020) periods was satisfactory. The
performance parameters during the calibration period were found to be R2 = 0.82 and NSE = 0.73, whereas during the validation
period, better performance parameters were obtained as R2 = 0.87 and NSE = 0.78. According to the criteria for model performance by
Kasuni (2017) and Shame Usman (2022), the values of NSE and R2 of the HEC-HMS model are greater than 0.65 and 0.7, respectively.
Hence, the HEC-HMS model performed well, and the rainfall-runoff simulation is considered reliable with good to very good
performance.
Even when there are some instances where the simulation is either an overestimate or an underestimate of the flow, the agreement
between the simulations and observations is generally very good, which is a typical trait of hydrological models that can be explained
by inadequate data inputs and natural variability (Figs. 7 –10).
These findings are consistent with earlier studies on the application of HEC-HMS models within the Lake Tana and Upper Blue Nile
sub-basins. Similar NSE and R2 values of 0.5–0.89 have been documented by Franco and Tesfaye (2020) and Belay et al. (2025) for
both calibration and validation purposes, thus indicating the efficacy of the model within this region.

3.4. Floodplain inundation mapping

The mapping of floodplain inundation in the Dirma River watershed has been undertaken through the combined use of ArcGIS,
HEC-HMS, HEC-RAS, and RAS-Mapper tools and models. From the analysis, the flooding of the different return period scenarios has
been mapped, thereby illustrating the link between flood levels and inundation depth and coverage.
From the analysis, the results indicate that there is an existing relationship between larger return periods and both larger flooded
land areas and higher flood depths (Table 6). Specifically, the area inundated increases from 9.58 km² (Fig. 11a) with a 2-year return
period to 11.68 km² (Fig. 11b) for a 100-year return period. In addition, there is an increase in flood depth from 9.64 m to 10.14 m. The

Table 5
HEC-HMS model optimized parameters.
Component Parameter Unit Initial Optimized Rank

Subbasin-1 SCS-CN-CN ​ 72.32 54 1

Subbasin-1 SCS-CN-Ia mm 18 125 2
Subbasin-1 SCS-CN-UH - Lag Time min 17.23 17.23 3
Subbasin-2 SCS-CN-CN ​ 80.2 58.2 4
Subbasin-2 SCS-CN-Ia mm 13.4 129 5
Subbasin-2 SCS-CN-UH - Lag Time min 16.3 16.3 6
Subbasin-3 SCS-CN-CN ​ 72.8 61.6 7
Subbasin-3 SCS-CN-Ia mm 17.5 16.4 8
Subbasin-3 SCS-CN-UH - Lag Time min 21.4 21.4 9
Reach-1 Muskingum – K hr 143.2 143.2 10
Reach-1 Muskingum – X ​ 0.015 0.015 11

10
F.T. Fikadie et al. Journal of Hydrology: Regional Studies 65 (2026) 103475

Fig. 7. Observed and simulated Streamflow with rainfall during calibration.

Fig. 8. Observed and simulated Streamflow with rainfall during validation.

change in the inundation area between the 2-year and the 100-year flood is significantevent but not extremely large, which indicates
the geomorphic characteristics of the Dirma floodplain. The downstream area is commonly flat and sensitive to an increase in water
surface elevation; accordingly, even a relatively small increase in stage can spread floodwater laterally across low-lying areas.
Nevertheless, the general areal extension remains moderate because some parts of the floodplain are topographically constrained and
because additional discharge mainly increases water depth in already inundated zones rather than opening entirely new flooded areas.
This designates that the watershed is highly sensitive to stage variation even under comparatively small increments in peak flow.
It has been proven that due to the relatively low gradient of the floodplain, it is very sensitive to floods. A slight change in the height
of the water surface (an increase from 1994.0 to 1994.3 m) between the two periods has significantly increased the flooded zone. As
can be seen from the figures depicting the areas of flooding for two periods, the 2-year and the 100-year return period (Figs. 12 and 13).
From the analysis above, it is apparent that extensive farming and ineffective land use can exacerbate the risk of flooding. This is
consistent with other research conducted in different Lake Tana watersheds, such as Gilgel Abay by Franco and Tesfaye (2020).

11
F.T. Fikadie et al. Journal of Hydrology: Regional Studies 65 (2026) 103475

Fig. 9. Observed versus simulated scatter plot of Streamflow during calibration.

Fig. 10. Observed versus simulated scatter plot of Streamflow during validation.

Table 6
floodplain inundation model.
Return period, yr 2 10 25 50 100
3
Peak flow, m /s 175 210 236 260 286
Max. flood depth, m 9.64 9.87 9.96 10.05 10.14
Floodplain area, km2 9.58 10.41 10.88 11.27 11.68

Therefore, floodplain mapping is essential for sustainable land use and flood risk management within the Dirma River watershed. The
outcomes of the simulated HEC-RAS model, which include the water surface elevation, depth, and velocity of all flow profiles, are
shown in Table 6, Table S7, and Figure S6.

12
F.T. Fikadie et al. Journal of Hydrology: Regional Studies 65 (2026) 103475

Fig. 11. Floodplain inundation map at 2- yr (a) and 100-yr (b) return period depth.

Fig. 12. Water surface elevation at 8912 river station at 2-yr return period floodplain.

4. Conclusions

These findings were effectively used to apply the HEC-HMS and HEC-RAS models to analyze rainfall-runoff dynamics, obtain
optimal flood frequency distributions, evaluate model performance, and delineate the flooded area in the Dirma River watershed.
However, the "best fit" results from the fitting of the probability distribution of the floods varied slightly with the statistical tests used.
The "best fit" results include the Gamma distribution for the Kolmogorov-Smirnov test, the Generalized Extreme Value distribution for
the Anderson-Darling test, and the Gumbel distribution for the Chi-Squared test, respectively. HEC-HMS sensitivity analysis results
indicate that the most important factors affecting runoff are SCS-CN and SCS Unit Hydrograph-Lag Time. Calibration and validation of

13
F.T. Fikadie et al. Journal of Hydrology: Regional Studies 65 (2026) 103475

Fig. 13. Water surface elevation at 8912 river station at 100-yr return period floodplain.

the HEC-HMS model proved a good-to-very-good accuracy level, with R2 ranging between 0.82 and 0.87 and NSE ranging between
0.73 and 0.78. From the analysis using HEC-RAS, it can be seen that the magnitude of the flooded area becomes greater as the return
period increases. As a result, the flooded area increases from 9.58 km² during the 2-year return period to 11.68 km² for the 100-year
return period, with the depth of the flood water increasing from 9.64 m to 10.14 m. This study clearly shows that the Dirma river basin
is highly vulnerable due to its flat landscape and poor land use conditions.
The following recommendations have been made based on the above results in order to minimize risks of floods and improve
watershed management: The flood maps show that the lowest elevation downstream of the Dirma River is the most susceptible region,
which should be prioritized in efforts towards mitigating risks of flooding. Such regions can be prioritized for structural defense,
improved drainage systems, flood-resistant land use planning, and implementation of early warning systems within communities.
Improved watershed management practices, such as afforestation, soil conservation, and sustainable agricultural techniques, are
implemented to reduce runoff and enhance infiltration. Future studies incorporate high-resolution climate and land use scenarios to
assess future flood risks under changing conditions. Findings are linked with basin-wide water management frameworks for the Lake
Tana sub-basin to support evidence-based decision-making.

CRediT authorship contribution statement

Conceptualization: Fitamlak Teka, Abreham Melaku, and Abebe Gobezie, Banteamlak Kase, project supervision: Banteamlak Kase,
Abreham Melaku, Abebe Gobezie, Abrham Bayeh, Adugnaw Nega, Alebel Abuhay, Henok Akmel, Sendeku Asnakew, Gebrehiwot Getu,
and Aniley Tenaw, Original and revised data acquisition, data processing, writing and revised document, methodology: Fitamlak Teka,
Banteamlak Kase, Abreham Melaku, Abebe Gobezie, Abrham Bayeh, Adugnaw Nega, and Alebel Abuhay, Software, visualization, and
validation: Fitamlak Teka, Banteamlak Kase, and Abebe Gobezie, Writing, editing, reviewing, and improvement of the final version of
the manuscript: all the members of the authors

Funding

This work was supported by the University of Gondar (UoG), Ethiopia, under grant number (ምር/ቴክ/ሽግ/ም/ፕ/05/170/2017).

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to
influence the work reported in this paper.

Acknowledgments

The authors would like to thank the University of Gondar (UoG) for supporting funding for the research project. All the organi­
zations which provided relevant data were highly appreciated, the Ministry of Water and Energy of Ethiopia (MoWE) provides
Streamflow data, the National Meteorology Service Agency of Ethiopia (NMSA) provides metrology and climate data, the Amhara
Design and Supervision Works Enterprise (ADSWE) provides soil and land use data, and Alaska Satellite Facility (ASF) provides a free
12.5-meter by 12.5-meter digital elevation model data.

14
F.T. Fikadie et al. Journal of Hydrology: Regional Studies 65 (2026) 103475

Appendix A. Supporting information

Supplementary data associated with this article can be found in the online version at doi:10.1016/[Link].2026.103475.

Data availability

Data will be made available on request.

References

Abbenante, S.E., Althoff, I., Valdes-Abellan, J., 2024. Comparative analysis of lumped and semi-distributed hydrological models in humid Mediterranean
environments. Hydrol. Sci. J. 69, 2481–2496.
Abd-Elhamid, H.F., Franzetti, L.A., Zeleňáková, M., Kaya, Y.Z., 2025. Development of integrated hydrologic and hydrodynamic models for flood modeling in Radiša
catchment at Western Slovakia using HEC-HMS and HEC-RAS. Nat. Hazards 1–27.
Acosta-España, J.D., Romero-Alvarez, D., Luna, C., Rodriguez-Morales, A.J., 2024. Infectious disease outbreaks in the wake of natural flood disasters: global patterns
and local implications. Le. Infez. Med. 32, 451.
Alam, M.A., Farnham, C., Emura, K., 2018. Best-fit probability models for maximum monthly rainfall in Bangladesh using Gaussian mixture distributions. Geosciences
8, 138.
Al-Salih, S.W.M.A., 2025. Water Quality assessment parameters using frequency analysis and environmental techniques at the Euphrates River Basin in Al-Nasiriyah,
Iraq. Eur. J. Ecol. Biol. Agric. 2, 115–124.
Ansarifard, S., Eyvazi, M., Kalantari, M., Mohseni, B., Ghorbanifard, M., Moghaddam, H.J., Nouri, M., 2024. Simulation of floods under the influence of effective
factors in hydraulic and hydrological models using HEC-RAS and MIKE 21. Discov. Water 4, 92.
Apollonio, C., Iemmolo, G., Di Modugno, M., Apollonio, M., Petroselli, A., Recanatesi, F., Giannetta, D., 2025. A multi-parameter approach to support sustainable
hydraulic risk analysis for the protection of transportation infrastructure: the case study of the Gargano Railways (Southern Italy). Sustainability 17, 4151.
Ashour, M.A., Mahmoud, H.M., Abu-Zaid, T.S., 2024. GIS-based methodology for culvert location evaluation on railways: a case study of the Qena–Safaga Track in
Egypt. AQUA—Water Infrastructure. Ecosyst. Soc. 73, 980–998.
Asiedu, J.B., 2020. Reviewing the argument on floods in urban areas: a look at the causes. Theor. Empir. Res. Urban Manag. 15, 24–41.
Ayalew, L.T., Bharti, R., 2022. Modeling sediment yield of rib watershed, northwest Ethiopia. ISH J. Hydraul. Eng. 28, 491–502.
Balasubramani, K., Gomathi, M., Bhaskaran, G., Kumaraswamy, K., 2019. GIS-based spatial multi-criteria approach for characterization and prioritization of micro-
watersheds: a case study of semi-arid watershed, South India. Appl. Geomat. 11, 289–307.
Barbhuiya, S., Raghuvanshi, A.S., Tiwari, H.L., 2023. Assessment of streamflow in ungauged basin by using physical similarity approach. Arab. J. Geosci. 16, 672.
Belay, B., Ambaw, G., Amha, Y., Tesfaye, A., Workneh, S., Terefe, T., Nigussie, A., Sinshaw, Y., Yigrem, S., Kebede, G., 2023. Climate-Smart. Agric. Train. Guide.
Belay, H., Melesse, A.M., Tegegne, G., Tamiru, H., 2025. Identifying flood source areas and analyzing high-flow extremes under changing land use, land cover, and
climate in the Gumara Watershed, Upper Blue Nile Basin, Ethiopia. Climate 13, 7.
Belayneh, A., Sintayehu, G., Gedam, K., Muluken, T., 2020. Evaluation of satellite precipitation products using HEC-HMS model. Model. Earth Syst. Environ. 6,
2015–2032.
Bodoque, J.M., Esteban-Muñoz, Á., Ballesteros-Cánovas, J.A., 2023. Overlooking probabilistic mapping renders urban flood risk management inequitable. Commun.
Earth Environ. 4, 279.
Bolt, N., 2025. The applicability of large-scale Rain-on-Grid hydrological modelling in South Africa. Stellenbosch University, Stellenbosch.
Bryson, J.M., Patterson, K., Cunsolo, A., Berrang-Ford, L., Lwasa, S., Namanya, D.B., Twesigomwe, S., Kesande, C., Ford, J.D., Team, I.H.A.T.C.C.R., 2024. When you
have stress because you don’t have food: Climate, food security, and mental health during pregnancy among Bakiga and Indigenous Batwa women in rural
Uganda. PLOS Clim. 3, e0000399.
Chan, F., Wang, Z., Chen, J., Lu, X., Nafea, T., Montz, B., Adekola, O., Pezzoli, A., Griffiths, J., Peng, Y., 2023. Selected global flood preparation and response lessons:
implications for more resilient Chinese Cities. Nat. Hazards 118, 1767–1796.
Chekole, A.G., Belete, M.A., Fikadie, F.T., Wubneh, M.A., 2024. Evaluate the performance of HEC-HMS and SWAT models in simulating the streamflow in the Gumara
watershed, Ethiopia. Sustain. Water Resour. Manag. 10, 26.
Cherinet, A., Tadesse, C., Abebe, T., 2022. Drought and flood extreme events and management strategies in Ethiopia. J. Geogr. Nat. Disast. 12, 2167-0587.
Cupak, A., 2020. Regionalization methods for low flow estimation in ungauged catchments-a review. Acta Sci. Pol. Form. Circumiectus 19, 21–35.
Desalegn, H., Mulu, A., 2021. Mapping flood inundation areas using GIS and HEC-RAS model at Fetam River, Upper Abbay Basin, Ethiopia. Sci. Afr. 12, e00834.
Devitt, L., Neal, J., Coxon, G., Savage, J., Wagener, T., 2023. Flood hazard potential reveals global floodplain settlement patterns. Nat. Commun. 14, 2801.
Fikadie, F.T., Wubneh, M.A., Kifelew, M.S., Sinshaw, B.G., 2022. Assessment of surface irrigation potential and crop water requirement: The case of megech
watershed, upper Blue Nile Basin, Ethiopia. Sustain. Water Resour. Manag. 8, 92.
Franco, E., Tesfaye, F., 2020. Floodplain inundation modeling and mapping for Gilgel Abay River: Nile Basin, Ethiopia. Int. J. Eng. Res. Technol. 10 (1).
Gemeda, D.O., Feyssa, D.H., Garedew, W., 2021. Meteorological data trend analysis and local community perception towards climate change: a case study of Jimma
city, Southwestern Ethiopia. Environ. Dev. Sustain. 23.
Ghayyadah, A. 2022. Power Comparison of Some Goodness of Fit Tests.
Greer, A.D., 2019. Autonomous engineering: a multi-scale GIS-based approach to green infrastructure design. The University of Alabama.
Grigg, N.S., 2023. Comprehensive flood risk assessment: state of the practice. Hydrology 10, 46.
Guo, Y., Zhang, Y., Zhang, L., Wang, Z., 2021. Regionalization of hydrological modeling for predicting streamflow in ungauged catchments: a comprehensive review.
Wiley Interdiscip. Rev. Water 8, e1487.
Haile, A.T., Tefera, F.T., Rientjes, T., 2016. Flood forecasting in Niger-Benue basin using satellite and quantitative precipitation forecast data. Int. J. Appl. earth Obs.
Geoinf. 52, 475–484.
Javadinejad, S., 2022. Causes and consequences of floods: flash floods, urban floods, river floods and coastal floods. Resour. Environ. Inf. Eng. 4, 173–183.
Kasuni, S.M. 2017. Impacts of expansion of agriculture and land use change on flow regime of Thiba river, Kenya.
Kumar, V., Sharma, K.V., Caloiero, T., Mehta, D.J., Singh, K., 2023. Comprehensive overview of flood modeling approaches: a review of recent advances. Hydrology
10, 141.
Manzoor, Z., Ehsan, M., Khan, M.B., Manzoor, A., Akhter, M.M., Sohail, M.T., Hussain, A., Shafi, A., Abu-Alam, T., Abioui, M., 2022. Floods and flood management
and its socio-economic impact on Pakistan: a review of the empirical literature. Front. Environ. Sci. 10, 1021862.
Mehta, D., Dhabuwala, J., Yadav, S.M., Kumar, V., Azamathulla, H.M., 2023. Improving flood forecasting in Narmada river basin using hierarchical clustering and
hydrological modelling. Results Eng. 20, 101571.
Mehta, D.J., Eslamian, S., Prajapati, K., 2022. Flood modelling for a data-scare semi-arid region using 1-D hydrodynamic model: a case study of Navsari Region.
Model. Earth Syst. Environ. 8, 2675–2685.
Mengistu, T.D., Chung, I.-M., Talpur, Z., Alemayehu, E., 2024. Quantifying regional low flows under data scarce conditions. Heliyon 10.

15
F.T. Fikadie et al. Journal of Hydrology: Regional Studies 65 (2026) 103475

Montes-Pajuelo, R., Rodríguez-Pérez, Á.M., López, R., Rodríguez, C.A., 2024. Analysis of probability distributions for modelling extreme rainfall events and detecting
climate change: Insights from mathematical and statistical methods. Mathematics 12, 1093.
Mphefu, T., 2024. Flood analysis and management for the Mngeni river catchment in South Africa. University of Johannesburg, South Africa.
Nagarajan, K., Narwade, R., Pathak, H., Panhalkar, S., Kulkarni, V.S., Hingmire, A.P., 2022. Review paper for floodplain mapping with applications of HEC-HMS,
HEC-RAS, and ArcGIS softwares—a remote sensing and GIS approach. Int. Res. J. Eng. Technol. 9, 812–826.
Naghibi, S.A., Moghaddam, D.D., Kalantar, B., Pradhan, B., Kisi, O., 2017. A comparative assessment of GIS-based data mining models and a novel ensemble model in
groundwater well potential mapping. J. Hydrol. 548, 471–483.
Narbondo, S., Gorgoglione, A., Crisci, M., Chreties, C., 2020. Enhancing physical similarity approach to predict runoff in ungauged watersheds in sub-tropical regions.
Water 12, 528.
Nasiri, S., Ansari, H., Ziaei, A.N., 2020. Simulation of water balance equation components using SWAT model in Samalqan Watershed (Iran). Arab. J. Geosci. 13, 421.
Navarro, F.A.R., 2024. Framework for flood risk management in urban basins under non-stationary scenarios based on hydrological-hydrodynamic simulations. Univ.
De. Sã o Paulo.
Pandey, S., Swarnkar, S., 2025. Overview of flood concepts: present perspective and future direction. Blue Sky, Blue Water: Strategies for Protecting Air and Water
Quality in the 21st Century. Springer.
Qi, W.-Y., Chen, J., Li, L., Xu, C.-Y., Li, J., Xiang, Y., Zhang, S., 2022. Regionalization of catchment hydrological model parameters for global water resources
simulations. Hydrol. Res. 53, 441–466.
Rawat, M., Jain, S.K., Ahmed, R., Lohani, A.K., 2023. Glacial lake outburst flood risk assessment using remote sensing and hydrodynamic modeling: a case study of
Satluj basin, Western Himalayas, India. Environ. Sci. Pollut. Res. 30, 41591–41608.
Ruman, S., Krpec, P., Rusnok, P., Black, A., Trizna, M., Ball, T., 2020. Impact of missing precipitation values on hydrological model output: a case study from the
Eddleston Water catchment, Scotland. Acta Geophys. 68, 565–576.
Shame Usman, N., 2022. Wabe Shebelle river basin, Ethiopia. Haramaya university, Haramaya. Assess. LAND Suitabil. Irrig. POTENTIAL JERJERTU WATERSHED.
Sharma, K.V., Jadhav, P., Kumar, V., Dongare, P., Munoth, P., Singh, K., 2024. Flood detection in the Upper Krishna Basin through integrated geospatial analysis:
leveraging decision frameworks and statistical measures. J. Hydroinform. 26, 2389–2415.
Singh, N.M., Devi, T.T., 2022. Regionalization methods in ungauged catchments for flow prediction: review and its recent developments. Arab. J. Geosci. 15, 1019.
Soliman, M., Morsy, M.M., Radwan, H.G., 2022. Assessment of implementing land use/land cover lulc 2020-esri global maps in 2d flood modeling application. Water
14, 3963.
Tegegne, G., Melesse, A.M., Asfaw, D.H., Worqlul, A.W., 2020. Flood frequency analyses over different basin scales in the Blue Nile River basin, Ethiopia. Hydrology 7,
44.
Wang, J., Li, K., Hao, L., Xu, C., Liu, J., Qu, Z., Yan, X., Sajjad, M.M., Sun, Y., 2024. Disaster mapping and assessment of Pakistan’s 2022 mega-flood based on multi-
source data-driven approach. Nat. Hazards 120, 3447–3466.
Waseem, M., Mani, N., Andiego, G., Usman, M., 2017. A review of criteria of fit for hydrological models. Int. Res. J. Eng. Technol. (IRJET) 4, 1765–1772.
Wubaye, G.B., Gashaw, T., Worqlul, A.W., Dile, Y.T., Taye, M.T., Haileslassie, A., Zaitchik, B., Birhan, D.A., Adgo, E., Mohammed, J.A., 2023. Trends in rainfall and
temperature extremes in Ethiopia: station and agro-ecological zone levels of analysis. Atmosphere 14, 483.
Yimer, N. 2020. Condition Assessment and Documentation of Ayteyef Palace Dessie, Ethiopia.

16