Next Article in Journal
Evapotranspiration and Crop Coefficient of Economically Important Fruit Trees in the Eastern Amazon
Next Article in Special Issue
Multi-Objective Calibration of a Pre-Modern Nile Hydrologic Model Using Recovered Records
Previous Article in Journal
Assessing Nonstationary Hydroclimatic Impacts on Streamflow in the Soan River Basin, Pakistan, Using Mann–Kendall Test and Artificial Neural Network Technique
Previous Article in Special Issue
Surface Runoff Responses to Forest Thinning in Semi-Arid Oak–Pine Micro-Catchments of Northern Mexico
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Sediment Yield Assessment and Erosion Risk Analysis Using the SWAT Model in the Amman–Zarqa Basin, Jordan

by
Motasem R. AlHalaigah
1,
Michel Rahbeh
1,*,
Nisrein H. Alnizami
1,
Mutaz M. Zoubi
2,
Heba F. Al-Jawaldeh
1,3,
Shahed H. Alsoud
1,3,
Yazan A. Alta’any
3,
Qusay Y. Abu-Afifeh
1,3,
Ali Brezat
1,
Rasha Al-Rkebat
4,
Safa E. El-Mahroug
1,
Bassam Al Qarallah
5 and
Ahmad J. Alzubaidi
3,6
1
Department of Land, Water and Environment, The University of Jordan, Amman 11942, Jordan
2
Department of Chemistry, The University of Jordan, Amman 11942, Jordan
3
Department of Civil Engineering, The University of Jordan, Amman 11942, Jordan
4
Department of Environmental Systems Research, National Agricultural Research Center, Baq’a 19381, Jordan
5
Department of Horticulture and Crop Science, The University of Jordan, Amman 11942, Jordan
6
Water Resources Unit, Crawford, Murphy & Tilly, Inc., Davenport, IA 52807, USA
*
Author to whom correspondence should be addressed.
Hydrology 2026, 13(4), 107; https://doi.org/10.3390/hydrology13040107
Submission received: 14 February 2026 / Revised: 7 April 2026 / Accepted: 7 April 2026 / Published: 9 April 2026

Abstract

Sediment accumulation in reservoirs represents a critical challenge for sustainable water resources management in semi-arid regions. In Jordan, accelerated sedimentation threatens the operational capacity of major dams, including the King Talal Dam (KTD), which serves as a key water resource in the Amman–Zarqa Basin (AZB). This study assesses sediment yield and erosion risk at the catchment scale using the Soil and Water Assessment Tool (SWAT) integrated with the Modified Universal Soil Loss Equation (MUSLE). The AZB was subdivided into 31 sub-basins and 586 Hydrological Response Units (HRUs) based on land use, soil characteristics, topography, and slope. The model was calibrated for the period 1993–2002 and validated for 2003–2012 using hydrological and sediment observations from 17 monitoring stations. Long-term simulations covering more than two decades were conducted to quantify spatial and temporal sediment yield patterns across the basin. Results indicate a mean annual sediment yield of 2.79 t ha−1 yr−1, corresponding to approximately 0.59 MCM yr−1 of sediment inflow to the reservoir. These estimates closely agree with bathymetric survey results reported by the Jordan Valley Authority, which indicate sedimentation rates of 2.59 t ha−1 yr−1 (0.55 MCM yr−1). Overall, the model demonstrates strong agreement between observed and simulated sediment loads, confirming its reliability for sediment dynamics assessment. The findings are relevant to Sustainable Development Goals (SDGs) 6 (clean water and sanitation) and 15 (life on land) by informing sustainable watershed and soil erosion management practices.

1. Introduction

Erosion is a geological phenomenon where soil and rock materials are gradually worn away and transported by natural forces such as water and wind. This process causes soil loss and sediment production from drainage areas, even in the absence of direct human disturbance [1]. In arid and semi-arid environments such as Jordan, soil erosion is a key driver of land degradation and downstream sedimentation [2]. Traditional land-management practices, vegetation loss during prolonged dry periods, increasing temperatures, and land-use change can further increase soil susceptibility to erosion by reducing ground cover and soil cohesion and enhancing runoff generation during subsequent high-intensity storms [3]. Consequently, erosion in these settings reflects the combined influence of natural controls (e.g., rainfall intensity and slope) and human pressures (e.g., overgrazing, inadequate conservation practices, and land conversion), often accelerating desertification processes on hillslopes [4].
Jordan is among the most water-stressed countries globally, which makes the protection of existing water resources and infrastructure a national priority. Water-scarcity responses commonly include dams and reservoirs, rainwater harvesting, desalination, and water reuse [5]. However, the long-term sustainability of reservoirs is strongly constrained by sediment inflow. Sediment delivery to reservoirs depends on hydro-climatic forcing (rainfall and discharge), catchment properties (land use, vegetation cover, soils, and topography), and temporal changes in rainfall patterns that can intensify erosion and sediment transport [6]. In this context, basin-scale sediment assessment is essential to support effective watershed management and to protect reservoir storage and operational performance.
Jordan receives only about 200 mm of rainfall annually over most of its territory, while water demand continues to increase due to population growth, industrial development, and improved living conditions [7]. Major urban and industrial centers—including Amman—are largely located in the northern highlands, particularly within the Amman–Zarqa Basin (AZB) [7]. The AZB therefore combines intensive land-use pressures with strong hydro-climatic variability, which increases the likelihood of episodic sediment pulses associated with storm events. Reservoir sedimentation is expected to persist under long-term landscape change, and it can progressively reduce the effective lifespan and benefits of dam infrastructure. The AZB faces compounded challenges related to water scarcity, soil erosion, and sedimentation, and the King Talal Dam (KTD) reservoir is particularly vulnerable because high-flow events and flash floods can mobilize and deliver substantial sediment loads to the reservoir [8].
The KTD, located near Jerash, is a non-homogeneous earth-fill dam approximately 350 m long with a height of 108 m. Constructed in 1978, the reservoir has an 83 MCM capacity (including dead storage) and drains a watershed of about 4300 km2 [9]. As one of Jordan’s major dams supporting irrigation in the Jordan Valley, KTD has experienced persistent sedimentation and storage loss, largely driven by sediment transported from the AZB via surface runoff and channel processes [10]. Compared with other Jordanian dam catchments that have been investigated for sedimentation (e.g., Wadi Al-Arab, Tannur, and Wadi Shu’eib), the AZB stands out due to its larger scale and stronger interaction between urban/industrial expansion and hydrologic extremes, which can complicate erosion dynamics and sediment routing to KTD.
Hydrological models are widely used to support watershed and reservoir management under changing climatic and land-use conditions. Examples include MIKE-SHE, HEC-RAS, GSFLOW, MODFLOW-based frameworks, and the Soil and Water Assessment Tool (SWAT) [11]. SWAT is broadly applied for simulating runoff and sediment dynamics and for evaluating management impacts on water supply, sediment yield, and watershed pollution [11]. It is commonly used at the catchment scale to assess erosion and sediment transport under varying land-use and management scenarios [12]. SWAT represents sediment yield using the integrated Modified Universal Soil Loss Equation (MUSLE), which links event runoff and peak flow to erosion and sediment delivery and enables spatially distributed sediment assessment across hydrologic response units (HRUs) [13,14,15].
Previous applications in Jordan and the region demonstrate SWAT’s usefulness for sediment-yield estimation and identifying erosion-prone sub-basins. For example, SWAT has been used to estimate sediment yield and erosion patterns in the Tannur Dam watershed [16] and to predict sediment yield and reservoir impacts in dam catchments where sedimentation threatens active storage. At Wadi Al-Arab Dam, SWAT-based simulations quantified sediment yield using MUSLE and reported reservoir-scale sedimentation implications over a multi-year period [17]. Modeling has also been applied to Wadi Shu’eib Dam to evaluate long-term hydro-sedimentary behavior and sediment yield at the dam outlet [18]. Beyond Jordan, SWAT-based studies have assessed best management practices and sediment reduction strategies, highlighting the role of parameter sensitivity, land cover, and conservation measures in controlling sediment fluxes [19]. Hybrid approaches integrating SWAT with data-driven models have also been used to improve sediment prediction under limited monitoring conditions in reservoir catchments [20,21]. These studies collectively show that sediment dynamics are strongly conditioned by land management, storm runoff generation, and sediment routing, but they also underscore a recurring challenge in semi-arid basins: limited sediment observations and the need for basin-specific calibration/validation to ensure credible sediment estimates.
Despite these advances, the existing literature provides limited consolidated evidence on sediment sources, spatial erosion hotspots, and long-term sediment delivery to KTD at the full AZB scale, especially in a way that links erosion risk mapping with reservoir sedimentation estimates. While hydrological and sedimentary conditions in the AZB have been previously examined using SWAT [22,23], additional basin-scale synthesis is needed to clarify (i) where erosion risk is concentrated, (ii) how sediment yield varies spatially and temporally across sub-basins, and (iii) how simulated sediment delivery compares with reservoir sedimentation evidence. Addressing these gaps is critical for designing targeted soil and water conservation measures and for prioritizing interventions that extend reservoir service life.
Accordingly, this study quantifies sediment yield and erosion risk within the AZB using SWAT coupled with MUSLE, supported by GIS-based spatial analysis. Also, this study evaluates conservation-practice scenarios to quantify their potential to reduce sediment yield and erosion risk across the basin. The model is calibrated for 1993–2002 and validated for 2003–2012 using available hydrological and sediment observations, and long-term simulations are used to characterize spatial and temporal sediment-yield patterns. The study further supports scenario-based evaluation of sustainable land management (SLM) interventions by adjusting the support practice factor (USLE_P) within defined ranges to represent potential conservation measures.

2. Methodology

2.1. Study Site Description

The AZB (Figure 1) is in the northwestern region of Jordan, positioned between the latitudes of 31°54′ and 32°24′ N, and the longitudes of 35°42′ and 36°36′ E. The basin releases water at the confluence of the Zarqa and Jordan Rivers, which is situated at an elevation of approximately 350 m. The streamflow of the Zarqa River is regulated by the KTD, the largest dam in Jordan, located southwest of Jerash. The Zarqa River ranks as the third-largest river in the region based on annual flow rate [9,24,25]. Its water resources are primarily allocated for irrigation and industrial applications. Additional inflow sources include small springs and effluents from the Samra Wastewater Treatment Plants (SWWTP). The basin covers a total area of 3900 km2 and comprises a complex mosaic of urban areas, agricultural lands, and both irrigated and rain-fed crops. Combined with the spatial variability of rainfall, this results in substantial heterogeneity across the AZB. Rainfed crops are cultivated in the northern areas, while the rest of the watershed area consists of irrigated agriculture. The agricultural areas within the basin consist of vegetables and fruit trees. In general, the main source of irrigation for farms located adjacent to the Zarqa River, specifically between Khirbet Al-Samra rainfall station and KTD, is diverted directly from the main reach of the river; meanwhile, the farms further away from the river are supplied by groundwater [25]. The soil characteristics of the AZB vary considerably with topography. In the humid western region, soil is generally reddish to brownish in color, consisting of clay and clay loams. While the eastern soils shift toward silty loam to loamy textures with higher carbonate content and colors ranging from yellowish-brown to strong brown. Soil erosion on steeper slopes influences alluvial processes, with sheet erosion particularly prevalent in the eastern parts.

2.2. Input Data Preparation for SWAT Model

The hydrological model utilized in this study, the SWAT (SWAT Model with ArcGIS 10.1 2012 extension), is widely used and adaptable to various watershed conditions and operates on a daily time step. The model is based on the soil water balance equation. It is designed as a tool to simulate watershed hydrology, infiltration, evapotranspiration, and runoff, and improve watershed management practices. The research methodology is illustrated in Figure 2.
The SWAT modeling required a digital elevation model (DEM), land-use/land-cover (LULC) map, soil map, and daily climatic data (precipitation and temperature). Soil erosion was simulated using MUSLE within SWAT, which estimates sediment yield at the HRUs (hydrologic response units) scale based on runoff-driven processes; HRUs represent the fundamental spatial computation units of SWAT and each HRU contains a unique combination of land use, soil, and slope characteristics [27]. A delineation was carried out for the basin, which consists of 31 sub-basins and 586 HRUs. The total number of HRUs was generated according to the recommended land use and the dominant soil type for five slope classes, which were (0–5%, 5–10%, 10–20%, 20–30%, and greater than 30%).
All input datasets were obtained from the Ministry of Water and Irrigation (MWI), Jordan (Figure 3), and were prepared following a structured workflow prior to model setup, including the DEM, LULC map, soil map, daily weather records, and the available hydrological/sediment monitoring records used for calibration/validation. A DEM, with a spatial resolution of approximately 250 m (247.55 × 247.55 m), was used as the primary terrain dataset to delineate the AZB watershed and stream network and to derive topographic parameters (slope classes) required by SWAT. Daily precipitation and temperature data were obtained from MWI station records [28] and were used to drive SWAT climate inputs, including evapotranspiration estimation using the Hargreaves method.
The hydro–meteorological inputs included daily rainfall, temperature, wind speed, and relative humidity records from 17 weather stations distributed across the basin for the period 1992–2012 (Table 1). For streamflow calibration, the Jerash Bridge gauge station (AL0004) provided the principal discharge record and was used as the most reliable hydrological reference within the basin. The total base flow was separated using the method of the straight line. The process of separation is done between the points that are direct surface flood-initiated and the points where the normal base flow resumes [29].
To calculate the actual value of the baseflow, it was determined as a proportion of the SWWTP effluent; the effluent proportion used upstream (for irrigation) was deducted from the proportion of the treatment plant effluent. Monthly stream flow from the SWWTP was obtained as an input to the SWAT model, during the process of collecting climate data (rainfall, temperature, wind speed, dew point, relative humidity, solar radiation). The Hargreaves equation was utilized, which depends on the rainfall and temperature variables. Not all weather data is accessible every day. In these situations, the SWAT fills in the gaps in measured records by estimating missing data using a weather generator called WGNmaker 4.1, which employs monthly statistics. WGNmaker 4.1 produces daily precipitation on its own first. Then, depending on whether there are wet days or not, the maximum temperature, minimum temperature, solar radiation, and relative humidity are calculated. Therefore, it is critical to evaluate how accurate these techniques are at estimating potential evapotranspiration (PET) using weather factors produced by WGNmaker 4.1. This study’s primary goal was to assess the predictive power of WGNmaker 4.1 for missing data needed by the PM technique and how it affected the SWAT forecasts of real evapotranspiration (ET) and streamflow. This was done by referring to the longitude, latitude, and altitude values for each rain gauge station. It should be noted that the only station within the basin that has an extended and documented database is the Jerash Bridge station, which would be the most suitable option in case of flood forecasting, due to its location at the dam’s eastern gate [30].
To improve transparency and reproducibility, Table 2 summarizes the key input datasets used for SWAT model preparation and evaluation in the AZB, including their sources, temporal coverage, resolution, and intended use.

2.3. SWAT Model Parameterization for Sediment Yield Simulation

The quantification of sediments in the AZB requires the evaluation of erosion from agricultural land and routing within the natural streams toward the basin’s outlet. The determination of erosion involves the MUSLE model, which consists of five factors that include peak runoff (R), soil erodibility (K), slope length and steepness (LS), the cover management (C), and support practice (P). The parameterization of these factors needs careful consideration, which involves identifying the prevailing support practices (interventions) and agricultural practices, in addition to the SWAT model parameters related to runoff simulation and routing.
The formula of the MUSLE is:
Sed = 11.8 (Qsurf × qpeak × Ah)0.56 × K × LS × C × P × CFRG
where Sed is the sediment yield on a given day (tons), Qsurf is the surface runoff volume (mm H2O/ha), qpeak is the peak runoff rate (m3/s), Ah is the HRU area in (ha), K is the soil erodibility factor, C is the cover and management factor, P is the support practice factor, LS is the topographic factor, and CFRG is the coarse fragment factor. The coarse fragment factor can be calculated in percentages based on this equation:
C FRG   = e 0.053   r o c k
where rock represents the rock percentage in the first soil. The LS is automatically calculated during HRU delineation from the DEM based on slope and slope length, and C is assigned from the land use/land cover class linked to each HRU. These values are critical for estimating the peak runoff rate ( q peak ), a key component of the MUSLE used in SWAT to simulate sediment yield. Furthermore, the support practice factor (P) was determined based on land management and intervention scenarios defined for each HRU, while the soil erodibility factor (K) was derived from the soil property database associated with each soil type. Together, these parameters ensure a more accurate representation of sediment detachment and transport processes in the model.

2.4. Sediment Data

The annual sediment accumulation in the KTD (Figure 4) was obtained from the Jordan Valley Authority, which provided the sediment data used in this study. The sediment data were collected through bathymetric and topographic surveys, depending on the water level in the reservoir. Bathymetric surveys were carried out during high-water levels using ultrasonic sensors, while topographic surveys were carried out when the dams were dry. The annual inflow values were derived from the available monthly flow records by converting each monthly flow to a monthly volume and then summing the 12 monthly volumes to obtain the annual flow volume (MCM/year) for each hydrological year.
The missing years were estimated with a documented cumulative inflow–sediment scaling. Following the attached engineering analysis, cumulative sediment (ΣSy) was related to cumulative inflow (ΣWf) by a power law [30]:
S y = 0.6042 × W f 0.3694
where Sy is the cumulative sediment yield of the KTD, and Wf is the annual water inflows at the KTD. Missing years were infilled using a documented cumulative inflow–sediment relationship (Equation (3)). The Wf were accumulated to obtain ΣWf, and cumulative sediment ΣSy was then estimated using the power-law scaling of Shammout and Abualhaija [31].
Equation (3) was applied only for the period(s) in which paired annual flow and sediment information required for parameterization were available, and the specific years used to derive and apply the equation parameters are now reported in the manuscript. In Equation (3), the summation symbol denotes accumulation over the full set of time steps within a year (i.e., aggregating the estimated sediment contribution across the defined temporal units used in the equation) to obtain an annual transported sediment estimate for each year of application.
It is important to note that Equation (3) estimates sediment transport/delivery (mass or load transported through the system), whereas the “sedimentation volume” shown in Figure 4 represents reservoir deposition/accumulation (stored sediment volume within the reservoir). These two variables have different physical meanings and are not expected to match one-to-one in magnitude because deposition is influenced by additional processes such as trapping efficiency, settling, resuspension, and operational/hydraulic conditions.
Accordingly, there was a noticeable relationship between higher flow and increased sediment yield, as seen in the peaks of 1992–1993 and 2005–2006 in Figure 4. However, after 1993, sediment yield decreases and stabilizes despite fluctuations in flow. Two notable events are:
  • 1992–1993 “Spike”: A significant flood or extreme weather event may have caused a surge in both water flow and sediment transport.
  • 2005–2006 “Peak in Flow”: Another high-flow event occurred, but sediment yield did not rise proportionally, suggesting potential sediment control measures or changes in land use.
Further changes may be attributed to sediment management practices, especially terraces interventions in the northwestern part of the basin where slopes are generally greater than 10%. Overall sediment “decline”: Over time, sediment yield appears to have decreased, possibly due to improved watershed management, soil conservation techniques, or dam constructions. In addition to the baseline simulation, conservation-practice scenarios were assessed by adjusting the USLE support practice factor (USLE_P) to represent improved soil and water conservation measures.

2.5. Model Calibration and Validation for Streamflow

Calibration and validation were performed by dividing the observed data into two periods: 1993–2002 for calibration (excluding the first three years from 1990), and 2003–2012 for validation. The SWAT-CUP 2019 (version 5.2) was employed to calibrate key hydrological variables such as runoff and sediment, and to validate the model performance on an independent dataset. The Parasol method within the SWAT-CUP optimized hydrological parameters by conducting sensitivity and uncertainty analyses, thereby enhancing the reliability of simulation results [32].
The calibration process required multiple SWAT simulations, with the resulting database serving as input for the SWAT-CUP to refine parameter estimation and complete both calibration and validation. The SWAT-CUP provided a robust framework for assessing sediment transport and improving prediction accuracy [33]. The software further allowed evaluation of the performance of sensitive parameters for simulating both streamflow and sediment load, as summarized in Table 3.
Parasol sensitivity analysis in SWAT-CUP identified the most sensitive parameters based on high absolute t-stat values and p-values close to zero [34]. The analysis used 144 streamflow observations and 10 sediment observations, with ~700 simulations (maximum trials). Streamflow was calibrated at the monthly scale (Jerash Bridge data), while sediment was calibrated annually because the available sediment observations are survey-based reservoir sedimentation volumes. Uncertainty was evaluated using the 95% prediction uncertainty (95PPU) framework (p-factor and r-factor), and relative errors were checked for extreme sediment years. Accordingly, the five most sensitive parameters for flow were CN2, ALPHA_BF, GW_DELAY, GWQMN, and ESCO [34], while the five most sensitive parameters for sediment were USLE_K, USLE_P (Slope (10–20)), USLE_P (Slope (20–30)), USLE_P (Slope (30–9999)), and RAINHHMX.
The SWAT-CUP calibration of the SWAT project resulted in a CN2 adjustment value of −0.6, indicating a relative decrease in the curve number and, thus, reduced potential for surface runoff generation throughout the AZB. This result shows the basin’s semi-arid to arid climate, characterized by a wet season from November to March, and an extended dry season, during which low antecedent soil moisture improves infiltration and decreases runoff. Even in wetter months, the heterogeneous land cover and permeable soils support spatially variable but primarily reduced runoff, which requires such a reduction adjustment in the CN2 parameter to accurately determine the basin’s hydrological reality.
The Nash–Sutcliffe efficiency (NSE) is extensively employed in hydrology to evaluate the model’s predictive capability compared to the observed data [34]. It has a range of (−∞ to 1.0). Where NSE = 1 indicates a perfect fit between the observed and simulated values. On the other hand, NSE = 0, suggests that the model predictions are as accurate as finding the mean of the observed data. Lastly, if NSE is less than 0, that means the model performs worse than using the mean observed value, which emphasizes the poor predictive skill.
NSE = 1 i = 1 n O i P i 2 i = 1 n Ō O i 2
where O i is the observed value, P i is the simulated or predicted value, and Ō is the mean of the observed values.
In the calibration part, the used parameters inserted into the SWAT-CUP were: SCS Curve number “CN2”, available water capacity “SOL_AWC”, USLE soil erodibility “USLE_K”, soil evaporation compensation factor, average slope length, Manning’s coefficient for the main channels, effective hydraulic conductivity, threshold depth of water in the shallow aquifer, and surface runoff lag coefficient).
To maintain the calibration criterion, the priority was to minimize the error between the simulated and observed mean annual flow for the total calibration period at the Jerash Bridge gauging station (AL0004). All the data cover the base flow, and the flood records are available, within the New Jerash Road Bridge, daily (m3/s) according to the MWI.
As a condition for the successful performance, including calibration and validation of the monthly streamflow data related to the AZB, the NSE coefficient should be greater than 0.5. This is also true for the error between the simulated and the measured mean annual sediment yield at the KTD.
The formula of coefficient of determination (R2) is given below:
R 2 = 1 S S E S S T
where SSE is the sum of the squared errors (residuals) between the predictions and the observations, and SST is the total sum of squares (variability in the observed data). A higher R2 (closer to 1) always indicates a better model fit.
The percent bias (PBIAS) expresses the average tendency of the simulated data to be greater or smaller than their observed counterparts. The PBIAS is commonly utilized in hydrology to evaluate water balance errors, flow-out simulations, and other environmental model outputs. It gives clear percentage-based evidence of model bias. The formula of the PBIAS:
PBIAS   =   100   ×   i = 1 n ( O i P i ) i = 1 n ( O i )
where n is the number of data points. An optimal PBIAS value is 0, indicating perfect alignment between observed and simulated data. Positive values indicate a model underestimation bias (simulated less than observed values), while negative PBIAS indicated a model overestimation bias (simulated values are greater than observed values).
In addition to NSE, R2, and PBIAS, model performance was further evaluated using Kling–Gupta Efficiency (KGE), root mean square error (RMSE), and mean absolute error (MAE), computed as follows:
K G E = 1 r l 1 2 + α 1 2 + β 1 2
where r l is the linear correlation coefficient between observed and simulated values, α = σ s i m / σ o b s is the variability ratio, and β = μ s i m / μ o b s is the bias ratio; σ and μ denote the standard deviation and mean, respectively.
R M S E = 1 n i = 1 n P i O i 2
M A E = 1 n i = 1 n P i O i
KGE was used as an integrated measure combining correlation, bias, and variability components (optimal value = 1), while RMSE and MAE quantify the magnitude of errors (optimal value = 0).

2.6. Time-Series Diagnostics (RAPS, ITA and Pettitt)

To strengthen the reliability of interpretations drawn from the observed and modeled time series, additional time-series diagnostics were conducted using (i) the rescaled adjusted partial sums (RAPS) approach, (ii) the innovative trend analysis (ITA) method and (iii) Pettitt test. RAPS was used to visually diagnose potential irregularities and fluctuations, including possible sub-period shifts and departures from stable behavior that may not be evident from the raw series alone [35]. The ITA was applied as a complementary, assumption-light approach to evaluate trend behavior by comparing ordered halves of the series and enabling identification of monotonic and non-monotonic tendencies without requiring distributional assumptions [36]. The Pettitt test was applied as a non-parametric change-point test to assess time-series homogeneity by detecting whether a statistically significant shift (breakpoint) occurs in the median of the series and identifying the most probable change year [36]. These diagnostics were used to examine potential irregularities, homogeneity-related behavior, and time-dependent fluctuations in the analyzed series and to support the robustness of the sediment-reduction recommendations derived from the SWAT outputs.
The formulas of those tests are as follows:
R A P S k = t = 1 k Y t μ σ  
S I T A = 1 m i = 1 m y i x i
K = max 1 t n 1 i = 1 t j = t + 1 n s g n   x i x j
where Y t is the value at time t , k is the position in the time series, S I T A is the ITA trend indicator (slope index), m is half of n, xi and yi are the sorted values of the first second halves, K is the Pettitt test statistic, sgn is the sign function and t is a candidate change-point position in the series.

3. Results and Discussion

3.1. Sensitivity Analysis and Uncertainty Assessment

The sensitivity analysis indicates that streamflow simulation in the AZB is governed mainly by runoff generation, groundwater dynamics, evapotranspiration partitioning, and channel routing. The strongest control is exerted by the SCS Curve Number (r__CN2.mgt = −0.6), where −0.6 to 0.2 represents the SWAT-CUP calibration range for the relative CN2 adjustment, highlighting the dominant role of surface runoff generation in shaping the outlet hydrograph. The negative adjustment of CN2 suggests that reducing runoff potential was necessary to better match the observed response under the prevailing basin conditions.
Groundwater-related parameters were also highly influential, reflecting the importance of delayed subsurface contributions to streamflow. The calibrated baseflow alpha factor (v__ALPHA_BF.gw = 0.68727; range 0–1) indicates a relatively responsive groundwater recession behavior, while the groundwater delay (v__GW_DELAY.gw = 263.18 days; range 30–450) reflects a long lag between recharge and groundwater contribution to the stream. In addition, the shallow aquifer threshold for return flow (v__GWQMN.gw = 4.8927; range 0–10) suggests that baseflow occurs after a moderate level of aquifer storage is achieved, supporting a threshold-type groundwater contribution.
Evapotranspiration controls were represented by ESCO and EPCO, both of which ranked among the sensitive parameters. The soil evaporation compensation factor (v__ESCO.hru = 0.8; range 0.8–1) and plant uptake compensation factor (v__EPCO.hru = 0.76254; range 0–1) indicate that soil–plant water partitioning substantially affects simulated water balance and low-flow behavior, which is expected in a semi-arid basin where evaporative losses are significant.
Channel routing parameters further influenced simulated discharge timing and attenuation. The calibrated Manning’s n for the main channel (v__CH_N2.rte = 0.070344; range 0–0.3) indicates sensitivity of flow routing to channel resistance. The effective hydraulic conductivity of the main channel alluvium (v__CH_K2.rte = 5; range 5–130) was calibrated at the lower bound, implying limited transmission losses were required to preserve discharge at the outlet. Bank storage dynamics were also relevant, with the baseflow alpha factor for bank storage calibrated at the upper bound (v__APHA_BNK.rte = 1; range 0–1), indicating a rapid exchange response within the adopted routing conceptualization.
Soil hydraulic properties contributed to sensitivity through their control on infiltration and percolation. The available water capacity adjustment (r__SOL_AWC().sol = 0.4) and the saturated hydraulic conductivity adjustment (r__SOL_K().sol = 0.36472) confirm that soil moisture storage and conductivity needed tuning to reproduce seasonal flow behavior. Finally, snow-related parameters (v__SFTMP.bsn = 1.8228; v__SMTMP.bsn = 13.4; v__SMFMX.bsn = 7.3781; v__SMFMN.bsn = 13.306) were sensitive, indicating that cold-season processes can influence runoff generation in the basin during years with snowfall and subsequent melt.
Regarding the sediment parameters, the sensitivity analysis indicates that simulated sediment yield is mainly controlled by hillslope erodibility and support practices (MUSLE/USLE-related factors), storm intensity characteristics, and channel sediment routing. The USLE soil erodibility factor (r__USLE_K().sol = 0.46532 within −0.5–0.7) ranked among the most influential parameters, confirming that soil susceptibility to detachment is a primary driver of sediment generation in the basin. The USLE support practice factor (v__USLE_P.mgt) was also highly sensitive across slope classes, with calibrated values decreasing as slope increases (0.77638 for 10–20%, 0.71922 for 20–30%, and 0.62451 for >30%). This pattern indicates that erosion-control effectiveness is strongly slope-dependent and that steeper areas contribute disproportionately to sediment yield unless conservation measures are improved.
Rainfall intensity forcing further affected sediment response through the maximum 0.5-h rainfall parameter (r__RAINHHMX().wgn = 1.5 within −0.5–1.5), highlighting the importance of short-duration, high-intensity storms in mobilizing sediment under semi-arid conditions. Channel processes were represented by sediment re-entrainment and routing parameters. The linear re-entrainment parameter (v__SPCON.bsn = 0.0001; range 0.0001–0.01) calibrated at the lower bound suggests limited re-entrainment under the adopted setup, while the exponent term (v__SPEXP.bsn = 0.92879; range 0.5–1.5) indicates a moderate nonlinearity in re-entrainment behavior. The peak rate adjustment factor for sediment routing (v__PRF_BSN.bsn = 1.0424; range 0–2) further confirms sensitivity to peak-flow conditions, consistent with sediment transport being strongly driven by high-flow periods. Collectively, these sensitivities indicate that sediment reduction in the AZB is best targeted through improving soil conservation practices in steep HRUs, reducing hillslope erodibility impacts, and limiting sediment connectivity and re-entrainment during intense storm events.
Table 4 summarizes the sensitive streamflow and sediment parameters identified by SWAT-CUP and reports their calibrated values and corresponding calibration ranges.

3.2. Calibration and Validation

In the hydrological modeling process, the flow-out calibration is a critical step to ensure that the simulated stream flow aligns closely with the observed conditions [36,37]. Two steps were carried out by the SWAT-CUP software. The data spans 20 years (1993–2012). In the case of the flow out, each reading was a single month, while the sediment data set had an annual trend. The SWAT-CUP 95PPU uncertainty evaluation for the monthly streamflow series (n = 180) yielded a p-factor of 0.144 and an r-factor of 0.202, where the p-factor is the fraction of observed data points bracketed by the 95PPU band, and the r-factor is the average thickness of the 95PPU band relative to the standard deviation of the observations. The low p-factor (0.144) indicates that only a small portion of observations is captured within the 95PPU envelope, while the low r-factor (0.202) indicates a relatively narrow uncertainty band. Together, these results suggest a narrow 95PPU envelope that does not bracket much of the observed variability, which is consistent with limited uncertainty coverage under the current parameter ranges and available data constraints. Relative errors were also evaluated for extreme years in the annual sedimentation record (e.g., 1991/1992) by comparing observed and simulated annual sedimentation values, supporting interpretation of model behavior under high-magnitude conditions despite the limited number of annual sediment observations.
Figure 5a,b shows the flow-out during calibration and validation periods.
The calibration period for this study (a) was from 1993 to 2002. The NSE had a value of 0.58, which indicated a good satisfactory performance between the observed and simulated streamflow. On the other hand, the validation period (b) covered the years 2003 to 2012. The NSE had a value of 0.52, indicating a moderate agreement between the model’s predictions and the actual observations, as reported by Tan et al. [37], Xing et al. [38] and Zhang et al. [39].
The NSE and R2 provide complementary but different information on model performance. NSE evaluates how well the simulated series reproduces observations relative to the observed mean, whereas R2 measures the strength of the linear association between simulated and observed values. Therefore, R2 alone does not indicate bias or volumetric accuracy and should be interpreted alongside NSE and other performance metrics [40]. In the current study, the NSE was used, but it was more focused on the R2 and the PBIAS, mainly when calibrating and validating the data. The R2 measures the proportion of variance in the dependent variable interpreted by the independent variables in the linear regression model, while for the standard PBIAS thresholds. Thus, it quantifies how well the model predictions would match the real-world observations, as claimed by Mehedi et al. [41] and Bojer et al. [42].
The complementary performance metrics show clear improvement from calibration to validation. During calibration, KGE was −0.837 with RMSE = 0.964 m3/s and MAE = 0.402 m3/s, indicating limited overall agreement between observed and simulated monthly flows. In contrast, validation achieved KGE = 0.725 with lower errors (RMSE = 0.720 m3/s; MAE = 0.321 m3/s), reflecting substantially improved consistency and reduced deviation magnitude in the simulated streamflow series.
The sediment yield for the AZB was simulated through the phases of calibration (1991–2002) and validation (2003–2012). In general, the observed and simulated sediment loads exhibit an agreement, with noticeable variations during the peak events. These indications aligned with Shaikh et al. [43]. Figure 6 shows the sediment yield (ton/ha) through calibration and validation periods.
In the calibration period (1991–2002), the model actively captured the overall sediment trends. Meanwhile, inconsistencies were observed, especially during the extreme sediment events. The same trends were observed by Sahoo et al. [44]. These variations could be obtained due to some uncertainties in the input data, land use changes, or sediment transport.
In the validation period (2003–2012), the model remained reasonably satisfied. However, the sediment peaks were sometimes underestimated, indicating potential limitations in demonstrating the extreme sediment transport dynamics. The main suggestion describes the model as suitable for long-term estimation but requires further improvements in parameterization and data accuracy to enhance the event’s prediction reliability [45,46]. Figure 7 shows an overall increasing trend in cumulative sediment over time (years). There is a sharp increase around (1990–1993), indicating a period of rapid sediment accumulation. After that, the increase continues but at a more gradual rate.
The basin-average sediment yield of 2.79 t/ha/year can be considered moderate for semi-arid to Mediterranean watersheds. For example, a SWAT application in the semi-arid Tata basin (SE Morocco) reported an average sediment yield of ~2.3 t/ha/year (range 0–11 t/ha/year), which is comparable in magnitude [47]. In contrast, lower long-term average sediment yields have been reported for some semi-arid Mediterranean watersheds with ephemeral streams (e.g., ~1.13 t/ha/year at watershed scale) [48]. Higher sediment yields are also common in more erosion-prone Mediterranean settings; for instance, the Carapelle watershed (southern Italy) reported an average annual sediment load of ~6.8 t/ha/year [49], and event-to-year variability can be large, with measured specific sediment loads ranging from ~0.89 t/ha/year (driest year) to ~7.45 t/ha/year (wettest year) in a Mediterranean basin study [50]. Therefore, although 2.79 t/ha/year is not exceptionally high relative to reported semi-arid/Med basins, it still implies meaningful sediment pressure when aggregated over a large drainage area and when considering that sediment export is typically concentrated in hotspot sub-basins and high-intensity storm periods.

3.3. Time-Series Trend and Change-Point Analysis Top of Form

For sediment, the RAPS curve (Figure 8) showed its highest excursion during the 1991/1992 season, consistent with an extreme event, followed by a sustained downward trajectory that indicates a progressive reorganization of the cumulative departures around the long-term mean. The sediment ITA plot (Figure 9) placed all points below the 1:1 line, indicating that later values were systematically lower than earlier values, and suggesting that the observed reduction in sediment yield is not solely an artifact of the extreme season. Consistent with these findings, the Pettitt test (Figure 10) detected a statistically significant change point in the sediment series in 1996–1997 (K = 324, p = 0.00001), indicating that the series is not homogeneous over the full observation period. The agreement between RAPS, ITA, and Pettitt results supports the interpretation of a sediment regime shift toward lower sediment conditions after the mid-1990s.
RAPS is a standardized diagnostic tool and does not have universal acceptance thresholds; therefore, interpretation is based on the curve pattern (persistent drift and pronounced excursions) rather than a specific RAPSk value. In current results, sediment RAPS shows a clear excursion in 1991/1992 followed by a sustained downward trajectory, indicating a shift toward lower sediment conditions, while flow RAPS shows alternating segments without a comparable persistent drift.
For streamflow, the RAPS curve (Figure 11) exhibited alternating declining, stable, and increasing segments, indicating multi-phase variability rather than a single persistent trend. The flow ITA plot (Figure 12) showed points distributed mainly around the 1:1 line, suggesting no strong monotonic trend, although a slight upward tendency was apparent for low and medium flows. One high-flow point fell below the 1:1 line, implying that the extreme-flow class did not follow the same tendency as the rest of the series. The Pettitt test (Figure 13) identified a statistically significant change point in flow in 1990–1991 (K = 192, p = 0.02848), while the primary sediment break occurred later (1996–1997). This temporal offset indicates that hydrological reorganization preceded the main sediment regime shift.
Overall, the combined evidence indicates that the sediment series experienced irregular and non-stationary behavior, including a persistent decline that cannot be directly explained by the streamflow series. The absence of a strong monotonic trend in flow, together with the later sediment breakpoint and systematic downward shift in sediment ITA, supports the interpretation that the reduction in sediment yield is more closely linked to changes in sediment supply conditions than to long-term hydrological change. This pattern is consistent with the effect of erosion-control and land-management interventions that reduce sediment availability and connectivity, thereby decreasing sediment delivery even when runoff variability remains present.

3.4. Sediment Hotspots and Implementation of the SLM

SWAT outputs indicate that for 1993–2012 the mean sediment yield delivered to the KTD outlet was 2.79 t/ha, corresponding to approximately 0.59 MCM/year of sediment deposition in the reservoir (based on a contributing area of 3600 km2). Sediment generation is seasonally concentrated, with peak contributions occurring during the winter months (January, February, and December), consistent with the basin’s rainfall–runoff regime. The highest modeled annual sediment yield occurred in 2012 (7.63 t/ha), whereas the lowest was observed in 2006 (2.17 t/ha). Based on the modeled long-term mean, the cumulative sediment deposited since dam commissioning in 1978 is estimated at ~13.65 MCM, equivalent to ~18% of the reservoir’s designed storage capacity, indicating a substantial long-term reduction in effective storage and operational performance.
Spatially, sediment generation is highly heterogeneous and concentrated at both the HRU and sub-basin scales, with a limited fraction of units acting as erosion hotspots that disproportionately contribute to outlet sedimentation (Figure 14). At the sub-basin scale (31 sub-basins), weighted sediment yield ranges from 0 to 16.02 t/ha (mean 2.07 t/ha). Sub-basins 5 and 30 represent the dominant erosion hotspots, with weighted sediment yields of 16.02 and 15.26 t/ha, respectively, and among the highest long-term weighted precipitation totals (~269.77 mm in sub-basin 5 and ~309.90 mm in sub-basin 30). A second group of elevated-yield sub-basins includes 1, 2, 23, 25, and 31 (3.95–6.64 t/ha). In contrast, several sub-basins exhibit negligible long-term average sediment yields despite receiving moderate rainfall (sub-basins 3, 4, 6, 7, 11, 12, 13, 14, 19, 20, 24, 27, 28, and 29), indicating that outlet sedimentation is driven by a limited set of high-yield sub-basins and their internal HRU-level hotspots where rainfall is more effectively converted into runoff and erosion.
Weighted precipitation varies between 59.09 and 309.90 mm (mean 137.74 mm; standard deviation 73.51 mm), whereas sediment yield varies much more strongly across sub-basins. At the sub-basin scale, weighted precipitation and weighted sediment yield show a strong positive relationship (R2 = 0.626; Figure 4), indicating that long-term precipitation differences explain approximately 63% of the spatial variance in sub-basin sediment yield, while the remaining variability is attributable to differences in land use, soil properties, and slope distribution.
Based on slope and rainfall conditions, contour plowing was recommended for areas with rainfall below 250 mm/year and slope greater than 10%, while stonewall terraces were suggested for areas with rainfall below 250 mm/year and slope less than 10%. Data sources included the Ministry of Agriculture and relevant literature on land management in Jordan [51,52,53,54,55,56,57].
The implementation of the SLM was carried out by setting the parameter of the support practice (USLE_P) to a different range of the maximum and minimum values, varied between 0 and 1 (no-till = 0.25, stonewall terracing = 0.50, contour plowing = 0.70) (p-values in Table 4). The HRUs (586 HRU within the basin) were divided into four slope categories: 0–5%, 5–10%, 10–20%, and 20–30%, respectively.
In this study, three main SLMs were recommended to reduce the number of sediments deposited into the KTD. The suggested SLMs [58]:
  • Contour plowing in the areas of slope < 10%
  • No-till in the areas of slope < 10%
  • Stone wall terracing in the areas of slope > 10%
Figure 15 presents the simulated annual sediment load at the outlet of KTD under baseline and SLM scenarios during 1991–2002. The baseline scenario (no intervention) exhibits pronounced interannual variability, with peak sediment loads exceeding 3 million tons per year in the early 1990s. In contrast, all SLM scenarios substantially reduce sediment export throughout the simulation period. Among the tested practices, the superior performance of the combined stone-terrace and contour-plowing scenario is consistent with previous evidence showing that contour-oriented and structural soil and water conservation measures reduce sediment export by slowing runoff and trapping mobilized particles. In a hilly Mediterranean farming context, contour-based practices coupled with undisturbed or minimally disturbed inter-row areas were reported to enhance soil structural stability and reduce erosion and runoff processes, supporting the effectiveness of contour-aligned management for long-term sediment control [59]. Similarly, a systematic review of sustainable land and water management technologies emphasizes that contour-aligned structures (including terraces/bunds) function as slope-control barriers that retain water and capture sediment while reducing surface runoff velocity, and it also notes the soil-health and runoff-related benefits associated with no-till/strip-till systems [60]. At the watershed scale, SWAT-based best management practices (BMP) assessments further indicate that structural measures such as bench terracing generally achieve larger sediment-yield reductions than agricultural practices alone, and that combined BMP scenarios can deliver the greatest overall sediment-load mitigation [57]. International experience shows that reducing sediment impacts requires combining catchment-scale erosion control with in-reservoir sediment management where feasible. In arid and semi-arid regions, vegetation restoration and land-management measures can substantially improve soil structure and stability, increase surface protection, and reduce sediment connectivity, thereby lowering sediment delivery to channels and reservoirs [61]. Beyond watershed interventions, several large reservoirs worldwide apply operational and engineering approaches to manage deposited and incoming sediments, including sediment flushing, sluicing during high flows, and the management of density/turbidity currents to route sediment through or away from the reservoir when hydraulic conditions allow. Numerical modeling demonstrates that turbidity currents can be predicted and strategically managed through outlet operation and reservoir regulation to reduce deposition in critical zones and improve sediment passage efficiency [62]. These experiences highlight that effective sediment-risk reduction is most robust when upstream conservation measures are paired with reservoir-scale operational strategies, guided by monitoring and modeling, to address both sediment supply and sediment trapping processes.

4. Conclusions

This study evaluated the spatial and temporal dynamics of sediment yield and soil erosion in the Amman–Zarqa Basin (AZB), Jordan, using the SWAT model integrated with MUSLE and supported by GIS-based spatial analysis. Long-term simulations (1990–2012) revealed marked interannual variability in sediment transport, reflecting the influence of rainfall intensity and runoff generation processes. Heavy snowfall events in the early 1990s, particularly in 1992, affected hydrological responses in the northern AZB and contributed to increased streamflow toward the King Talal Dam (KTD) through snowmelt-driven runoff. The overall positive relationships among rainfall, streamflow, and sediment indicators are consistent with the expected basin behavior and support the use of the modeling framework for catchment-scale assessment and relative sediment-risk interpretation.
Sensitivity and spatial analyses indicate that sediment generation is concentrated in specific areas of the basin, with agricultural lands emerging as major contributors, reflecting the combined effects of soil erodibility, slope, and land management. Scenario evaluation of SLM practices suggests that contour plowing provides the largest sediment reduction, followed by stonewall terracing and no-till farming, highlighting the potential effectiveness of targeted practices for mitigating erosion and reducing sediment delivery to KTD under semi-arid conditions.
This study also has limitations that should be considered when interpreting the results. Sediment evaluation relied on a limited annual sedimentation dataset, whereas streamflow calibration was performed at a monthly scale, which constrains event-scale assessment of extreme sediment pulses. In addition, climate data gaps required gap-filling, which introduces uncertainty. Therefore, the results are most suitable for interpreting long-term trends, spatial patterns, and relative scenario performance rather than precise event-by-event sediment prediction.
Future research should prioritize higher temporal resolution sediment monitoring and event-based evaluation to better characterize extreme sediment transport processes, particularly under hydroclimatic variability. In addition, complementary approaches such as fully coupled hydrodynamic modeling based on 2D shallow water equations integrated with sediment transport modules could be applied to assess within-channel processes, sediment routing, and reservoir-adjacent dynamics at finer spatial and temporal scales. Integrating remote sensing products and expanded scenario-based land management simulations would further strengthen decision-support capabilities for sustainable watershed planning and reservoir protection. In this context, the outcomes of this study are relevant to Sustainable Development Goal (SDG) 6 (clean water and sanitation) by supporting sustainable water resource management and reservoir protection, and to SDG 15 (life on land) by promoting soil conservation and erosion control strategies that contribute to the long-term stability of terrestrial ecosystems.

Author Contributions

Conceptualization, M.R.A., M.R. and M.M.Z.; methodology, M.R.A., M.R., M.M.Z., Y.A.A. and A.J.A.; software, M.R.A., M.R., M.M.Z., Y.A.A., Q.Y.A.-A. and A.J.A.; validation, M.R.A., M.R., M.M.Z. and Q.Y.A.-A.; formal analysis, M.R.A., M.R., M.M.Z., H.F.A.-J. and Q.Y.A.-A.; investigation, M.R.A., M.M.Z. and H.F.A.-J.; resources, M.R.A., N.H.A., M.M.Z., S.H.A., Q.Y.A.-A., A.B., R.A.-R. and S.E.E.-M.; data curation, M.R.A., N.H.A., M.M.Z., S.H.A., A.B. and R.A.-R.; writing—original draft preparation, M.R.A., M.R., M.M.Z., Y.A.A. and Q.Y.A.-A.; writing—review and editing, M.R., M.M.Z., H.F.A.-J., Q.Y.A.-A., B.A.Q. and A.J.A.; visualization, M.R.A., N.H.A., H.F.A.-J., A.B., R.A.-R. and S.E.E.-M.; supervision, M.R., M.M.Z. and Q.Y.A.-A.; project administration, M.R. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Data Availability Statement

Data will be made available upon request.

Acknowledgments

The authors would like to express their sincere gratitude to the University of Jordan, Amman, Jordan, for providing the resources and support necessary for this research. During the preparation of this work the authors used ChatGPT5.2 to improve language. After using this tool, the authors reviewed and edited the content as needed and took full responsibility for the content of the publication.

Conflicts of Interest

Author Ahmad J. Alzubaidi was employed by the company Crawford, Murphy & Tilly, USA. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Abbreviations

The following abbreviations are used in this manuscript:
α Variability ratio
ALPHA_BF Base Flow Alpha Factor
ARSAgricultural Research Service
ASA ArcGIS Spatial Analysis
AWC Available Water Capacity
AZBAmman-Zarqa Basin
β Bias ratio
BMPBest Management Practices
C Cover management factor
CFRG Coarse fragment factor
CNCurve Number
CN2 Curve Number (Moisture Condition II)
DEMDigital Elevation Model
ESCOSoil Evaporation Compensation Factor
ETEvapotranspiration
GSFLOWGroundwater and Surface-water FLOW model
GW_DELAY Groundwater Delay
GWQMNThreshold Depth of Water in the Shallow Aquifer for Return Flow
HEC-RASHydrologic Engineering Center–River Analysis System
HRUsHydrologic Response Units
JWWTPJerash Wastewater Treatment Plant
kPosition in the time series
KPettitt test statistic
KTDKing Talal Dam
LSSlope length and steepness
μ Mean
MCMMillion Cubic Meters
MIKE-SHE Modeling System for Hydrology and Environment
MODFLOWModular Three-Dimensional Finite-Difference Ground-Water Flow Model
MUSLEModified Universal Soil Loss Equation
MWIMinistry of Water and Irrigation
NDVI Normalized Difference Vegetation Index
NSENash–Sutcliffe Efficiency
OiObserved value of variable i
PBIASPercent Bias
PETPotential Evapotranspiration
PiSimulated value of variable i
r l Linear correlation coefficient
RPeak runoff
R2Coefficient of determination
RAINHHMXMaximum Half-Hour Rainfall
σ Standard deviation
sgnSign function
S I T A ITA Trend Indicator (Slope Index)
SLMSustainable Land Management
SSE Sum of Squared Errors residuals
SWATSoil and Water Assessment Tool
SWAT-CUPSWAT Calibration and Uncertainty Programs
SWWTPSamra Wastewater Treatment Plant
tCandidate Change-Point position
USLEUniversal Soil Loss Equation
USLE_KSoil erodibility factor
USLE_PSupport practice factor
Y t Value at time
WGENWeather Generator

References

  1. Bekele, S.; Abate, B. Estimation of sediment yield using SWAT model: A case of Soke River watershed, Ethiopia. Int. J. Eng. Res. 2021, 9, 695. [Google Scholar]
  2. Alkharabsheh, M.M.; Alexandridis, T.; Bilas, G.; Misopolinos, N.; Silleos, N.G. Impact of land cover change on soil erosion hazard in northern Jordan using remote sensing and GIS. Procedia Environ. Sci. 2013, 19, 912–921. [Google Scholar] [CrossRef]
  3. Abuhamoor, D.; AlMahasneh, L.; Tawaha, A. Spatial-based analysis for risk erosion hazard in Jordan. Sains Tanah J. Soil Sci. Agroclimatol. 2025, 22, 157. [Google Scholar] [CrossRef]
  4. Al-Sababhah, N.M.A.; Al Maqablah, M.M. Integrated evaluation of soil erosion-prone areas based on the GIS technique and analytic hierarchy process on hillside slopes, northwest Jordan. Jordan J. Earth Environ. Sci. 2023, 14, 158–174. [Google Scholar]
  5. El-Khateeb, M.; Abu-Dalo, M.; Salih, K.A.; Ayadi, H.; Choura, M. Environmental assessment of the surface sediment in the artificial lake at Jordan University of Science and Technology as a route to sustainable aquaculture. Arab. J. Geosci. 2023, 16, 552. [Google Scholar] [CrossRef]
  6. Shammout, M.W.; Shatanawi, K.; Al-Bakri, J.; Abualhaija, M.M. Impact of land use/cover changes on the flow of the Zarqa River in Jordan. J. Ecol. Eng. 2021, 22, 40–50. [Google Scholar] [CrossRef]
  7. Abdulla, F.; Eshtawi, T. Climate change effect on sediment yield at King Talal Dam (Jordan). Civil Environ. Res. 2015, 7, 13–27. [Google Scholar]
  8. El-Mahroug, S.E.; Suleiman, A.A.; Zoubi, M.M.; Al-Omari, S.; Abu-Afifeh, Q.Y.; Al-Jawaldeh, H.F.; Alta’any, Y.A.; Al-Nawaiseh, T.; Obeidat, N.; Alsoud, S.H.; et al. Predictive modeling of climate-driven crop yield variability using DSSAT towards sustainable agriculture. AgriEngineering 2025, 7, 156. [Google Scholar] [CrossRef]
  9. Abu-Afifeh, Q.; Rahbeh, M.; Al-Afeshat, A.; Al-Omari, S.; Qutishat, T.A.; Brezat, A.; Alkayed, A. Dam sustainability’s interdependency with climate change and dam failure drivers. Sustainability 2023, 15, 16239. [Google Scholar] [CrossRef]
  10. Al-Taani, A.; El Radaideh, N.; Al Kateeb, W.; Howari, F.; Nazzal, Y.; Iqbal, J.; Al Bsoul, A. Sustainable sediment management in Jordanian dams: Feasibility, economic viability, and agricultural reuse potential. J. Water Land Dev. 2025, 64, 211–220. [Google Scholar] [CrossRef]
  11. Sabale, R.S.; Londhe, S.; Jose, M.K. Reservoir sedimentation analysis using SWAT model. In Lecture Notes in Civil Engineering; Springer: Singapore, 2023; pp. 167–180. [Google Scholar] [CrossRef]
  12. Leta, M.K.; Waseem, M.; Rehman, K.; Tränckner, J. Sediment yield estimation and evaluating best management practices in Nashe watershed, Blue Nile Basin, Ethiopia. Environ. Monit. Assess. 2023, 195, 716. [Google Scholar] [CrossRef] [PubMed]
  13. Alam, S.A.; Maiti, R. Estimating sediment yield, water discharge and surface runoff using SWAT model. In Sediment Source to Sink: Deciphering Sediment Connectivity to Large Dams in Damodar River Basin; Springer: Cham, Switzerland, 2025; pp. 33–57. [Google Scholar]
  14. Zalaki-Badil, N.; Eslamian, S.; Sayyad, G.-A.; Hosseini, S.-E.; Asadilour, M.; Ostad-Ali-Askari, K.; Singh, V.; Dehghan, S. Using SWAT model to determine runoff and sediment yield in Maroon-Dam catchment. Int. J. Res. Stud. Agric. Sci. 2017, 3, 31–41. [Google Scholar] [CrossRef]
  15. Chiang, L.-C.; Lu, C.-M. Improvement of sediment simulation in landslide areas using SWAT model with various landslide triggering factors and multiple thresholds. Catena 2024, 240, 107948. [Google Scholar] [CrossRef]
  16. Ijam, A.; Al-Bdoor, A. Sediment yield assessment for Tannur Dam reservoir in Jordan. Civil Environ. Res. 2016, 8, 79–87. [Google Scholar]
  17. Hadadin, N.A.; Al-Adwan, S.R. Estimation of sediment yield using hydrological assessment tool model: A case of Wadi Al-Arab Dam, northern Jordan. Arab. J. Geosci. 2020, 13, 380. [Google Scholar] [CrossRef]
  18. Al-Weshah, R.; Smadi, K.; Ta’any, R. Sediment transport mitigation and management in Wadi Shu’eib Dam. Jordan J. Earth Environ. Sci. 2025, 16, 227–236. [Google Scholar]
  19. Babaei, H.; Nazari-Sharabian, M.; Karakouzian, M.; Ahmad, S. Identification of critical source areas (CSAs) and evaluation of best management practices (BMPs) in controlling eutrophication in the Dez River Basin. Environments 2019, 6, 20. [Google Scholar] [CrossRef]
  20. Kassem, A.A.; Aness, W.T.; Mohammed, M.G. Sediment estimation of the Mosul Dam Lake using SWAT and dual ANN models. Tikrit J. Eng. Sci. 2025, 32, 1853. [Google Scholar] [CrossRef]
  21. Aness, W.T.; Hadadin, N.A.; Zoubi, M.M.; Abu-Afifeh, Q.Y.; Al-Afeshat, A.F.; Al-Jawaldeh, H.F.; Qutishat, T.A.; Alhalaigah, M.R.; Alnizami, N.H.; Aljarrah, A.; et al. Runoff and Sediment Dynamics Modeling for Erosion Risk Reduction and Sustainable Watershed Management. Glob. J. Environ. Sci. Manag. 2026, 12, 813–836. [Google Scholar] [CrossRef]
  22. Rahbeh, M.; Alnizami, N.H.; Zoubi, M.M.; Abu-Afifeh, Q.A.; Qutishat, T.A.; AlHalaigah, M.R.; Al-Jawaldeh, H.F.; Al-Omari, S.; Al Qarallah, B. Comparative Reliability Analysis of CHIRPS and Gauge-Based Precipitation Measurements over the Zarqa River Basin. Jordan J. Earth Environ. Sci. 2026, 17, 83–96. Available online: https://www.researchgate.net/publication/402103200_Comparative_Reliability_Analysis_of_CHIRPS_and_Gauge-_based_Precipitation_Measurements_over_the_Zarqa_River_Basin (accessed on 6 April 2026).
  23. Ghumaid, A.A. Study of changes in hydrological conditions for the Amman–Zarqa Basin using incremental climate change scenarios, SWAT model, and geospatial techniques. An-Najah Univ. J. Res. B (Humanit.) 2023, 37, 943–980. [Google Scholar] [CrossRef]
  24. Al-Afeshat, A.; Zoubi, M.M.; Abu-Afifeh, Q.Y.; Al-Jawaldeh, H.; Qutishat, T.A.; Masoud, A.M.N.; Rahbeh, M. Interrelation of dam sustainability with local communities and water quality. Glob. J. Environ. Sci. Manag. 2025, 11, 157–176. [Google Scholar] [CrossRef]
  25. Rahbeh, M.; Srinivasan, R.; Mohtar, R.H. Numerical and conceptual evaluation of preferential flow in Zarqa River Basin, Jordan. Ecohydrol. Hydrobiol. 2019, 19, 224–237. [Google Scholar] [CrossRef]
  26. GIS Geography. What Is Remote Sensing? The Definitive Guide to Earth Observation. 2021. Available online: https://gisgeography.com/remote-sensing-earth-observation-guide/ (accessed on 23 December 2025).
  27. Arnold, J.G.; Srinivasan, R.; Muttiah, R.S.; Williams, J.R. Large area hydrologic modeling and assessment part I: Model development. J. Am. Water Resour. Assoc. 1998, 34, 73–89. [Google Scholar] [CrossRef]
  28. Hroub, H.A.; Rahbeh, M.; Zoubi, M.M.; Abu-Afifeh, Q.Y.; Al-Jawaldeh, H.; Obeidat, N. Projection of future temperature variations in river basins under climate change scenarios using general circulation models. Glob. J. Environ. Sci. Manag. 2025, 11, 403–426. [Google Scholar] [CrossRef]
  29. Gonzales, A.L.; Nonner, J.; Heijkers, J.; Uhlenbrook, S. Comparison of different base flow separation methods in a lowland catchment. Hydrol. Earth Syst. Sci. 2009, 13, 2055–2068. [Google Scholar] [CrossRef]
  30. Al-Qaisi, M. Climate Change Effects on Water Resources in Amman–Zarqa Basin—Jordan; Ministry of Water and Irrigation (MWI): Amman, Jordan, 2010.
  31. Shammout, M.W.; Abualhaija, M.M. An analysis of long-term yearly water-flow trend and its impact on sediment yield in King Talal Dam. Int. J. Eng. Res. Technol. 2019, 12, 3041–3049. [Google Scholar]
  32. Arnold, J.G.; Moriasi, D.N.; Gassman, P.W.; Abbaspour, K.C.; White, M.J.; Srinivasan, R.; Santhi, C.; Harmel, R.D.; Van Griensven, A.; Van Liew, M.W.; et al. SWAT: Model use, calibration, and validation. Trans. ASABE 2012, 55, 1491–1508. [Google Scholar] [CrossRef]
  33. Abbaspour, K.C. SWAT-CUP: SWAT Calibration and Uncertainty Programs—A User Manual; Swiss Federal Institute of Aquatic Science and Technology (Eawag): Dübendorf, Switzerland, 2015. [Google Scholar]
  34. Andrade, C. The p value and statistical significance: Misunderstandings, explanations, challenges, and alternatives. Indian J. Psychol. Med. 2019, 41, 210–215. [Google Scholar] [CrossRef]
  35. Garbrecht, J.; Fernandez, G.P. Visualization of Trends and Fluctuations in Climatic Records. J. Am. Water Resour. Assoc. 1994, 30, 297–306. [Google Scholar] [CrossRef]
  36. Şen, Z. Innovative Trend Analysis Methodology. J. Hydrol. Eng. 2012, 17, 1042–1046. [Google Scholar] [CrossRef]
  37. Tan, L.; Zan, C.; Liu, T.; Xiong, J.; Zhang, A. Evaluating multi-source precipitation data for streamflow simulation using the SWAT model in the alpine Manas River Basin, Northwest China. Sci. Rep. 2025, 15, 43770. [Google Scholar] [CrossRef]
  38. Xing, K.; Yang, P.; Liu, S.; Zhao, Q. A coupled SWAT-LSTM approach for climate-driven runoff dynamics in a snow- and ice-fed arid basin. Sustainability 2025, 17, 10235. [Google Scholar] [CrossRef]
  39. Zhang, Z.; Xiao, Y.; Chen, R.; Long, K.; Deng, H.; Zheng, Z.; Liao, J. Comparative assessment of machine learning models for daily streamflow prediction in a subtropical monsoon watershed. Sci. Rep. 2026, 16, 7341. [Google Scholar] [CrossRef]
  40. Mitova, M.; Rakotoarimanana, Z.; Kercheva, M.; Ishidaira, H. SWAT model calibration, validation and parameter sensitivity analysis using SWAT-CUP SUFI-2 for the Rusenski Lom River watershed, Bulgaria. In Environmental Protection and Disaster Risks (EnviroRisks 2024); Springer: Cham, Switzerland, 2024; Volume 883, pp. 381–390. [Google Scholar]
  41. Mehedi, M.; Döll, P.; Hosseini-Moghari, S.-M.; Papa, F.; Güntner, A. The benefits and trade-offs of multi-variable calibration of the WaterGAP global hydrological model (WGHM) in the Ganges and Brahmaputra basins. Hydrol. Earth Syst. Sci. 2025, 29, 567–596. [Google Scholar] [CrossRef]
  42. Bojer, A.K.; Abshare, M.W.; Mesfin, F.; Al-Quraishi, A.M.F. Assessing climate and land use impacts on surface water yield using remote sensing and machine learning. Sci. Rep. 2025, 15, 18477. [Google Scholar] [CrossRef] [PubMed]
  43. Shaikh, M.; Lodha, P.; Eslamian, S. A framework for climate change impact assessment and mitigation of hydrological parameters of the watershed through trend analysis and hydrological modeling. In Handbook of Eurasian Forecasting; Eslamian, S., Bintul Huda, M., Eslamian, F., Eds.; Nova Science Publishers: New York, NY, USA, 2022; pp. 79–110. [Google Scholar]
  44. Sahoo, S.; Das, V.K.; Debnath, K. Control of sandbar development on morphological evolution and channel dynamics of the Mahanadi Delta. Int. J. River Basin Manag. 2026, 24, 1–23. [Google Scholar] [CrossRef]
  45. Al-Nouti, A.F.; Fu, M.; Bokde, N.D. Reservoir operation-based machine learning models: Comprehensive review of limitations, research gaps, and future directions. Knowl.-Based Eng. Sci. 2024, 5, 75–139. [Google Scholar] [CrossRef]
  46. Wang, K.; Bertoli, G.; Cheng, S.; Schröter, K.; Caporali, E.; Piggott, M.D.; Wang, Y.; Arcucci, R. AI-empowered latent four-dimensional variational data assimilation for river discharge forecasting. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 2025, 18, 24676–24689. [Google Scholar] [CrossRef]
  47. Echogdali, F.Z.; Boutaleb, S.; Taia, S.; Ouchchen, M.; Id-Belqas, M.; Kpan, R.B.; Abioui, M.; Aswathi, J.; Sajinkumar, K.S. Assessment of Soil Erosion Risk in a Semi-Arid Climate Watershed Using SWAT Model: Case of Tata Basin, South-East of Morocco. Appl. Water Sci. 2022, 12, 164. [Google Scholar] [CrossRef]
  48. Pulighe, G.; Bonati, G.; Colangeli, M.; Traverso, L.; Lupia, F.; Altobelli, F.; Dalla Marta, A.; Napoli, M. Predicting Streamflow and Nutrient Loadings in a Semi-Arid Mediterranean Watershed with Ephemeral Streams Using the SWAT Model. Agronomy 2019, 10, 2. [Google Scholar] [CrossRef]
  49. Ricci, G.F.; De Girolamo, A.M.; Abdelwahab, O.M.M.; Gentile, F. Identifying Sediment Source Areas in a Mediterranean Watershed Using the SWAT Model. Land Degrad. Dev. 2018, 29, 1233–1248. [Google Scholar] [CrossRef]
  50. Abdelwahab, O.M.M.; Ricci, G.F.; De Girolamo, A.M.; Gentile, F. Modelling Soil Erosion in a Mediterranean Watershed: Comparison between SWAT and AnnAGNPS Models. Environ. Res. 2018, 166, 363–376. [Google Scholar] [CrossRef]
  51. Ziadat, F.M.; Zdruli, P.; Christiansen, S.; Caon, L.; Abdel Monem, M.; Fetsi, T. An overview of land degradation and sustainable land management in the Near East and North Africa. Sustain. Agric. Res. 2021, 11, 11–24. [Google Scholar] [CrossRef]
  52. Al-Bakri, J.T.; D’Urso, G.; Calera, A.; Abdalhaq, E.; Altarawneh, M.; Margane, A. Remote sensing for agricultural water management in Jordan. Remote Sens. 2023, 15, 235. [Google Scholar] [CrossRef]
  53. Jawarneh, R.N.; Abulibdeh, A.; Hashem, N.; Awawdeh, M.; Al-Awadhi, T.; Abdullah, M.M.; El-Kenawy, A.M. Assessing and predicting land cover dynamics for environmental sustainability in Jordan’s arid ecosystems using the CA–Markov model. Remote Sens. Appl. Soc. Environ. 2024, 35, 101262. [Google Scholar] [CrossRef]
  54. Al-Sababhah, N. Land suitability and capability analysis for sustainable allocation of agricultural crops and natural plants, northwest Jordan. J. Geovis. Spat. Anal. 2023, 8, 1. [Google Scholar] [CrossRef]
  55. Zubi, R.A.; Gill, T.E.; Lee, J.A.; Ardon-Dryer, K.; Kandakji, T. Spatiotemporal analysis of sand and dust emission point sources detected from satellite imagery in Syria, Jordan, and Iraq. Sci. Total Environ. 2024, 949, 175157. [Google Scholar] [CrossRef]
  56. Assaf, M.N.; Abdelal, Q.; Hussein, N.M.; Halaweh, G.; Alzubaidi, A.J. Water quality monitoring and management: Integration of machine learning algorithms and Sentinel-2 images for the estimation of chlorophyll-a. Model. Earth Syst. Environ. 2025, 11, 348. [Google Scholar] [CrossRef]
  57. Darwish, H.H.; Talozi, S. Assessment of water footprint in the Jordanian industrial sector as a means for sustainable water resources management. Nafta-Gaz 2024, 80, 450–468. [Google Scholar] [CrossRef]
  58. Debie, E. Effectiveness of sustainable land management initiatives in the highlands of Ethiopia. Reg. Environ. Change 2024, 24, 106. [Google Scholar] [CrossRef]
  59. Masri, Z.; Turkelboom, F.; Huang, C.-H.; Schumacher, T.E.; Govindan, V. Restoring soil and ecosystem functions in hilly olive orchards in northwestern Syria by adopting contour tillage and vegetation strips in a Mediterranean environment. Soil Syst. 2025, 10, 1. [Google Scholar] [CrossRef]
  60. Nguru, W.; Ouedraogo, I.; Muriithi, C.; Karanja, S.; Kinyua, M.; Nduah, A. A systematic review of land- and water-management technologies for resilient agriculture in the Sahel: Insights from climate analogues in sub-Saharan Africa. Sustainability 2026, 18, 787. [Google Scholar] [CrossRef]
  61. Zhou, W.; Li, C.; Wang, S.; Ren, Z.; Stringer, L.C. Effects of Vegetation Restoration on Soil Properties and Vegetation Attributes in the Arid and Semi-Arid Regions of China. J. Environ. Manag. 2023, 343, 118186. [Google Scholar] [CrossRef]
  62. Sharma, N.; Yousuf, A.; Kaushal, A. Evaluating the efficacy of best management practices in mitigating sediment yield and nutrient losses in the Lower Sutlej River Basin, India. Water Air Soil Pollut. 2024, 235, 703. [Google Scholar] [CrossRef]
Figure 1. Location and spatial configuration of the AZB, Jordan, including the regional context map and the detailed basin map showing the KTD, drainage network, sub-basin boundaries, and meteorological station locations [26].
Figure 1. Location and spatial configuration of the AZB, Jordan, including the regional context map and the detailed basin map showing the KTD, drainage network, sub-basin boundaries, and meteorological station locations [26].
Hydrology 13 00107 g001
Figure 2. The overall framework of the methodology.
Figure 2. The overall framework of the methodology.
Hydrology 13 00107 g002
Figure 3. Spatial input datasets used for SWAT model setup in the AZB: (a) DEM (m), (b) derived slope classes (%), (c) soil map, and (d) LULC map. All layers were obtained from the Ministry of Water and Irrigation (MWI), Jordan, and were prepared to support watershed delineation, HRU definition, and erosion–sediment simulations.
Figure 3. Spatial input datasets used for SWAT model setup in the AZB: (a) DEM (m), (b) derived slope classes (%), (c) soil map, and (d) LULC map. All layers were obtained from the Ministry of Water and Irrigation (MWI), Jordan, and were prepared to support watershed delineation, HRU definition, and erosion–sediment simulations.
Hydrology 13 00107 g003
Figure 4. Comparison of annual flow (MCM) and annual sediment yield (MCM) over the years (1981 to 2017) regarding the outlet at the KTD.
Figure 4. Comparison of annual flow (MCM) and annual sediment yield (MCM) over the years (1981 to 2017) regarding the outlet at the KTD.
Hydrology 13 00107 g004
Figure 5. Comparison between observed and simulated monthly streamflow at the basin outlet during (a) calibration (1993–2002) and (b) validation (2003–2012) periods. Model performance is evaluated using the PBIAS, NSE, R2, KGE, RMSE and MAE.
Figure 5. Comparison between observed and simulated monthly streamflow at the basin outlet during (a) calibration (1993–2002) and (b) validation (2003–2012) periods. Model performance is evaluated using the PBIAS, NSE, R2, KGE, RMSE and MAE.
Hydrology 13 00107 g005
Figure 6. Comparison of observed and SWAT-simulated annual sediment yield (t ha−1) in the AZB for the calibration (1991–2002) and validation (2003–2012) periods, separated these periods with a green line.
Figure 6. Comparison of observed and SWAT-simulated annual sediment yield (t ha−1) in the AZB for the calibration (1991–2002) and validation (2003–2012) periods, separated these periods with a green line.
Hydrology 13 00107 g006
Figure 7. Temporal evolution of cumulative sediment yield in the Amman–Zarqa Basin over the study period (1991–2012).
Figure 7. Temporal evolution of cumulative sediment yield in the Amman–Zarqa Basin over the study period (1991–2012).
Hydrology 13 00107 g007
Figure 8. The RAPS of the sediment time series.
Figure 8. The RAPS of the sediment time series.
Hydrology 13 00107 g008
Figure 9. The ITA for sediment series.
Figure 9. The ITA for sediment series.
Hydrology 13 00107 g009
Figure 10. The Pettitt test of the sediment series.
Figure 10. The Pettitt test of the sediment series.
Hydrology 13 00107 g010
Figure 11. The RAPS of the time series.
Figure 11. The RAPS of the time series.
Hydrology 13 00107 g011
Figure 12. The ITA for the flow series.
Figure 12. The ITA for the flow series.
Hydrology 13 00107 g012
Figure 13. The Pettitt test of stream flow series.
Figure 13. The Pettitt test of stream flow series.
Hydrology 13 00107 g013
Figure 14. The AZB sediment hotspots at the sub-basin scale.
Figure 14. The AZB sediment hotspots at the sub-basin scale.
Hydrology 13 00107 g014
Figure 15. Comparison of simulated annual sediment load at the KTD outlet under baseline (no intervention) and three SLM scenarios (stone terraces; stone terraces + no-till; stone terraces + contour plowing) during 1991–2002.
Figure 15. Comparison of simulated annual sediment load at the KTD outlet under baseline (no intervention) and three SLM scenarios (stone terraces; stone terraces + no-till; stone terraces + contour plowing) during 1991–2002.
Hydrology 13 00107 g015
Table 1. The meteorological and streamflow stations used in the study.
Table 1. The meteorological and streamflow stations used in the study.
No.Meteorological Station TypeIDStation NameLongitudeLatitudePercentage of Missing Data *
1RainfallAD0018Ibbin35.8136932.3609-
2RainfallAL0003Bal’ama36.0876432.23574-
3RainfallAL0005Kitta35.8417732.27509-
4RainfallAL0012Sukhna36.0654232.12949-
5RainfallAL0015Zarqa36.0891632.0644-
6RainfallAL0016Ruseifa36.0410832.01693-
7RainfallAL0017Sweilh35.8400232.02261-
8RainfallAL0018Jubeiha35.8664932.02428-
9RainfallAL0026Burma35.7830532.22127-
10RainfallAN0002Wadi EsSir35.8183731.95057-
11RainfallAN0003Na’ur35.8284331.87387-
12Rainfall/ClimateAL0019Amman Airport35.987831.974897%
13Rainfall/ClimateAL0035King Hussein Nursery (Baq’a)35.8457132.080291%
14Rainfall/ClimateAL0053King Talal Dam35.8316632.1939938%
15Rainfall/ClimateAL0057Natural Resources Authority35.8469631.9558486%
16Rainfall/ClimateAL0066Khirebit Es Samra36.1456532.1514613%
17Rainfall/Stream flowAL0004Jarash35.8948832.27932-
* Missing climatic data was filled by WGNmaker 4.1.
Table 2. Summary of input datasets used for SWAT model preparation and evaluation in the AZB.
Table 2. Summary of input datasets used for SWAT model preparation and evaluation in the AZB.
DatasetQuantityQuality Control
DEM~250 m (247.55 × 247.55 m cell size)Projected to UTM Zone 36N (WGS84); slope reclassified into five classes (0–5%, 5–10%, 10–20%, 20–30%, and >30%)
LULCTo be confirmedReclassified to the SWAT land-use classes
Soil map34 coded soil unitsSoil classes linked to the SWAT lookup tables and harmonized with GIS layers
Rainfall station inventory34 rainfall stationsStation IDs, names, coordinates, and annual average precipitation compiled
Selected climate stations used in SWAT17 weather stationsMissing climatic variables were supplemented using WGNmaker 4.1 where required
Hydrological reference station1 gauge stationDaily base flow, flood flow, and total discharge were checked for continuity
Monthly flow seriesMonthly time seriesOne missing monthly value was identified; the series requires explicit clarification in the text as modeled or observed
Weather generator supportWGNmaker 4.1 Tool-basedUsed to supplement missing required climatic inputs; not a replacement for observed records
Table 3. Sensitive stream flow and sediment parameters and their ranges.
Table 3. Sensitive stream flow and sediment parameters and their ranges.
Target VariableParameter Name *Description of ParametersRange Value
Stream flow parametersr__CN2.mgtSCS runoff curve number−0.6–0.2
v__ALPHA_BF.gwBase flow alpha factor0–1
v__GW_DELAY.gwGroundwater delay30–450
v__GWQMN.gwThe threshold depth of water in the shallow aquifer required for return flow to occur0–10
v__ESCO.hruSoil evaporation compensation factor0.8–1
v__EPCO.hruPlant uptake compensation factor0–1
v__CH_N2.rteManning’s “n” value for the main channel0–0.3
v__CH_K2.rteEffective hydraulic conductivity in the main channel alluvium5–130
v__APHA_BNK.rteThe baseflow alpha factor for bank storage0–1
r__SOL_AWC().solAvailable water capacity of the soil layer−0.4–0.4
r__SOL_K().solSaturated hydraulic conductivity−0.8–0.8
v__SFTMP.bsnSnowfall temperature−5–10
v__SMTMP.bsnSnowmelt base temperature−20–20
v__SMFMX.bsnMaximum melt rate for snow during the year (occurs on the summer solstice)0–20
v__SMFMN.bsnMinimum melt rate for snow during the year (occurs on the winter solstice)0–20
Sediment parametersr__USLE_K().solUSLE soil erodibility factor−0.5–0.7
v__USLE_P.mgtUSLE support practice factor (Slope (10–20))0–1
v__USLE_P.mgtUSLE support practice factor (Slope (20–30))0–1
v__USLE_P.mgtUSLE support practice factor (Slope (30–9999))0–1
r__RAINHHMX().wgnMaximum 0.5-h rainfall in the entire period of record for the month−0.5–1.5
v__SPCON.bsnLinear re-entrainment parameter for channel sediment routing0.0001–0.01
v__SPEXP.bsnExponent parameter for calculating sediment re-entrained in channel sediment routing0.5–1.5
v__PRF_BSN.bsnPeak rate adjustment factor for sediment routing in the main channel0–2
* v = absolute value, r = relative change.
Table 4. Sensitive stream flow and sediment calibrated parameters.
Table 4. Sensitive stream flow and sediment calibrated parameters.
Stream Flow Calibrated ParametersSediment Calibrated Parameters
Parameter NameCalibrated ValueParameter NameCalibrated Value
r__CN2.mgt−0.6r__USLE_K().sol0.46532
v__ALPHA_BF.gw0.68727v__USLE_P.mgt0.77638
v__GW_DELAY.gw263.18v__USLE_P.mgt0.71922
v__GWQMN.gw4.8927v__USLE_P.mgt0.62451
v__ESCO.hru0.8r__RAINHHMX().wgn1.5
v__EPCO.hru0.76254v__SPCON.bsn0.0001
v__CH_N2.rte0.070344v__SPEXP.bsn0.92879
v__CH_K2.rte5.0v__PRF_BSN.bsn1.0424
v__APHA_BNK.rte1
r__SOL_AWC().sol0.4
r__SOL_K().sol0.36472
v__SFTMP.bsn1.8228
v__SMTMP.bsn13.4
v__SMFMX.bsn7.3781
v__SMFMN.bsn13.306
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

AlHalaigah, M.R.; Rahbeh, M.; Alnizami, N.H.; Zoubi, M.M.; Al-Jawaldeh, H.F.; Alsoud, S.H.; Alta’any, Y.A.; Abu-Afifeh, Q.Y.; Brezat, A.; Al-Rkebat, R.; et al. Sediment Yield Assessment and Erosion Risk Analysis Using the SWAT Model in the Amman–Zarqa Basin, Jordan. Hydrology 2026, 13, 107. https://doi.org/10.3390/hydrology13040107

AMA Style

AlHalaigah MR, Rahbeh M, Alnizami NH, Zoubi MM, Al-Jawaldeh HF, Alsoud SH, Alta’any YA, Abu-Afifeh QY, Brezat A, Al-Rkebat R, et al. Sediment Yield Assessment and Erosion Risk Analysis Using the SWAT Model in the Amman–Zarqa Basin, Jordan. Hydrology. 2026; 13(4):107. https://doi.org/10.3390/hydrology13040107

Chicago/Turabian Style

AlHalaigah, Motasem R., Michel Rahbeh, Nisrein H. Alnizami, Mutaz M. Zoubi, Heba F. Al-Jawaldeh, Shahed H. Alsoud, Yazan A. Alta’any, Qusay Y. Abu-Afifeh, Ali Brezat, Rasha Al-Rkebat, and et al. 2026. "Sediment Yield Assessment and Erosion Risk Analysis Using the SWAT Model in the Amman–Zarqa Basin, Jordan" Hydrology 13, no. 4: 107. https://doi.org/10.3390/hydrology13040107

APA Style

AlHalaigah, M. R., Rahbeh, M., Alnizami, N. H., Zoubi, M. M., Al-Jawaldeh, H. F., Alsoud, S. H., Alta’any, Y. A., Abu-Afifeh, Q. Y., Brezat, A., Al-Rkebat, R., El-Mahroug, S. E., Al Qarallah, B., & Alzubaidi, A. J. (2026). Sediment Yield Assessment and Erosion Risk Analysis Using the SWAT Model in the Amman–Zarqa Basin, Jordan. Hydrology, 13(4), 107. https://doi.org/10.3390/hydrology13040107

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop