Water is a vital resource that has been a central concern in Morocco’s politics for several decades. Owing to its increasing scarcity and the pressures exerted by climatic hazards such as recurrent droughts and floods, sustainable management of this resource has become a national priority (Bennouna, 2020). This challenge is even more complex in countries where agriculture plays a key role (World Bank, 2024). Therefore, it is imperative to ensure a stable water supply to support agriculture as well as the industrial and tourism sectors.

To address this growing vulnerability and to better regulate water availability, Morocco adopted an ambitious dam construction policy as early as the late 1960s. The aim was to store water during periods of abundance to cope with future droughts. This strategy has enabled the country to have 152 large dams, offering a storage capacity of 19.9 billion m^³, according to the General Directorate of Hydraulics (2023). However, despite this infrastructure, major challenges remain, notably the silting of reservoirs, which compromises their efficiency.

Water erosion is a pressing global environmental challenge that poses a significant threat to natural resources by degrading soil and inducing dam siltation (Ed-Dakiri et al., 2022). It involves the displacement of soil particles by runoff and is a major driver of reservoir sedimentation (El Mokaddem et al., 2019). This process is exacerbated by increasing rainfall erosivity patterns, especially in vulnerable regions, such as Europe and North Africa, highlighting the role of climate-related factors in accelerating soil degradation (Panagos et al., 2015). Morocco’s climate is characterized by substantial spatiotemporal variability in both yearly and seasonal rainfall, partly because of its extratropical location, making it prone to various large-scale fluctuations. Moreover, the country’s large topography, characterized by the Rif and Atlas Mountains, has a considerable impact on its climatic conditions (Borrelli et al., 2020; Gumus et al., 2023). These climatic characteristics intensify erosion and increase sedimentation in water reservoirs, thereby challenging water management efforts.

Silting of dams reduces their storage capacity by approximately 0.5%-1% per year, or approximately 75 million m^³ annually (Choukri et al., 2019; Schleiss et al., 2016). This progressive loss hampers reservoir operations and threatens essential services, such as drinking water supply, irrigation, power generation, and flood control (Hilgert et al., 2024; Lawmchullova & Rao, 2024; Skariah & Suriyakala, 2021). In addition sediment trapping triggers adverse downstream effects, including intensified erosion, ecosystem disturbances, and altered flow regimes (Kondolf et al., 2014; Maavara et al., 2020). In semi-arid regions, siltation interferes with soil–water interactions and groundwater recharge, especially under declining rainfall conditions (Luo et al., 2023).

This issue is directly linked to the degradation of natural resources, particularly in developing countries. Therefore, the management of watersheds facing climatic hazards is a crucial issue, especially in arid and semi-arid regions, such as Morocco (Maurya et al., 2023; Messaoudi et al., 2025).

In northern Morocco, particularly in the Rif region, water erosion presents a critical risk for soil degradation and dam sedimentation (Ed-Dakiri et al., 2024). The average annual soil loss in this region ranges from 30 to 60 metric tons per hectare, resulting in an annual siltation of water reservoirs of 125 million m^³ and a consequent 0.5% yearly reduction in their storage capacity.

The Ibn Battouta Dam, essential for the potable water supply of Tangier, exemplifies this issue. Its storage capacity has decreased from 43 million m^³ in 1978 to only 30.1 million m^³ by 2020 (Department of Water Research and Planning, 2024). This decline has been attributed to the high sedimentation rate caused by water erosion in the Tleta watershed. Sediment transport in riparian environments is a complex phenomenon influenced by various natural and anthropogenic factors. Flood events, particularly in the Tleta watershed, have accelerated sediment transport and redistribution (Ockelford et al., 2020).

Moreover, dams interrupt the longitudinal transport of sediments, leading to upstream accumulation and alterations in downstream erosion processes (Serra et al., 2022). This dynamic justifies a reassessment of the sediment sources contributing to the silting of the Ibn Battouta Dam, considering not only the upstream inputs but also the morphological changes downstream. Beyond Morocco, similar issues have been observed globally: sediment retention by dams has been shown to increase coastal and downstream erosion (Mosaid et al., 2024). Human infrastructure, such as dams and ports, has modified sediment flows and disrupted morphological processes (Laïbi et al., 2014). Therefore, it is essential to consider downstream erosion and update the existing studies by integrating recent technological advances and new data.

The study of watershed morphology and erosional processes is essential for understanding the impact of climate change on the hydraulic infrastructure (Lu et al., 2021). Modern remote sensing techniques, such as satellite imagery and digital elevation models (DEMs), enable a detailed analysis of morphological changes, particularly erosion and sediment deposition. The integration of geospatial data into GIS software facilitates the assessment of hydrogeomorphological parameters and improves our understanding of erosional processes that contribute to sedimentation (Hema et al., 2022; Rai et al., 2018).

Although several studies have been conducted on the Tleta watershed, they have mainly focused on upstream erosion, which is considered the primary source of sediment. This approach has led to neglecting downstream erosion, which can play a significant role in sediment accumulation. Thus, updating the current knowledge is necessary, considering recent technological advancements and the newly available data.

Our study falls within this framework and aims to reassess the erosional processes downstream of the watershed. Using modern remote sensing tools and multi-source data, we analyzed morphological changes around the Ibn Battouta Dam reservoir. By comparing images taken during different periods, we identified areas particularly affected by shoreline degradation to better understand the full range of erosional processes that influence the watershed.

Understanding these erosional dynamics is essential for addressing the challenges of sustainable water resource management in Morocco. Combating dam sedimentation is a priority to ensure water security and guarantee a long-term water supply. This study assessed the contribution of riverbank erosion to the siltation processes of the Ibn Battouta Dam by integrating shoreline dynamics analysis with multitemporal satellite imagery and bathymetric surveys.

Located in the extreme northwest region of the Rif Mountain ridge, the Tleta watershed covers an area of 177 km^² (2021). As shown in Figure 1, national and regional roads strategically connect Tangier, Tétouan, and Larache. It has an average altitude of 171 m and a slope of up to 16%. This watershed, drained by the Oued Kebir and its tributaries, feeds the Ibn Battouta Dam, whose initial capacity of 43.6 million m^³ dropped to 30.11 million m^³ in 2020 owing to sedimentation caused by erosion. Under a subhumid Mediterranean climate, the region receives an average annual rainfall of 750 mm, concentrated between October and March (Sabir et al., 2019). However, heavy rain frequently causes flooding and accelerates soil erosion. Geologically, the watershed consists of marl and schist formations, which, owing to their fragility, facilitate sediment transport toward the dam. Similar conditions have been observed in other Mediterranean watersheds, where marly and schistous formations contribute significantly to the sediment yield because of their low cohesion and high erodibility (Rodrigo-Comino et al., 2018). The distribution of vegetation is largely influenced by lithology and soil type. On marl substrates and clay-textured soils, the predominant vegetation formations include tall and dense matorral, low and sparse matorral, and very sparse Erme zones. In contrast, sandstone substrates are dominated by matorral formations, primarily resulting from cork-oak forest degradation (Ezzine et al., 2011).

These characteristics render the Tleta Watershed hydrologically dynamic but vulnerable, thus requiring appropriate management to preserve its water resources and mitigate the impact of sedimentation on hydraulic infrastructure.

This study monitored the morphological evolution of the shores of the Ibn Battouta dam reservoir and highlighted the changes that occurred between 1984 and 2024. To this end, the Normalized Difference Water Index (NDWI) was used via the Google Earth Engine platform to enable the accurate mapping of water contours over the decades.

In parallel, a Digital Elevation Model (DEM) developed from contour lines extracted from a topographic map enables the analysis of morphological variations within a detailed geomorphological framework.

The results of this study highlight a significant reduction in the water surface and storage capacity of the Ibn Battouta reservoir, confirming the persistent issue of siltation affecting Moroccan dams. Similar observations were reported by El Boukhari et al. (2020) at the Hachef Dam, which is also located in northern Morocco, where sediment accumulation was found to severely compromise water storage efficiency and operational lifespan. These findings underscore the broader trend of accelerated reservoir silting across the country, largely attributed to soil erosion within upstream catchments and insufficient sediment management strategies.

The analysis of water surfaces extracted using the Normalized Difference Water Index (NDWI) made it possible to assess the effectiveness of satellite imagery in detecting water bodies. By comparing the results obtained from Sentinel-2A and Landsat, we evaluated the accuracy of the extractions in terms of absolute and relative errors expressed in square kilometers.

The results confirm the reliability of satellite-based remote sensing, particularly the use of the Normalized Difference Water Index (NDWI), for accurately detecting and monitoring surface water dynamics (Gao et al., 2016). Among the analyzed datasets, Sentinel-2A imagery demonstrated superior performance over Landsat, primarily because of its higher spatial resolution (10 m) and enhanced spectral sensitivity. The use of the Multispectral Instrument (MSI) sensor on Sentinel-2A enables improved mapping of inland water bodies, increasing the global water surface mapping resolution from 14.25 m to 10 m (Verpoorter et al., 2014). This enhancement provided more detailed and precise imagery. Additionally, the high-frequency acquisition of high-resolution data makes it possible to create dynamic surface-water maps. These maps can distinguish between permanently and temporarily inundated water bodies. This approach provides a clearer and more up-to-date understanding of the evolution of water surfaces (Toming et al., 2016). These improvements resulted in lower absolute and relative errors in surface extraction, making Sentinel-2A particularly well-suited for precise hydrological monitoring (Fig. 12).

A long-term analysis conducted over a 40-year period (1984-2024) revealed substantial temporal and spatial variability in the reservoir water surface (Fig. 9). According to Figure 7, it is evident that the variation in the shape and size of the polygons is primarily attributable to differences in the hydrometric water levels, which range between 35.85 and 48.02 meters. This significant surface variation between the images was particularly noticeable in the upstream part of the reservoir, suggesting that the main tributaries bring in sediments that contribute to deposition and alter the morphology of the reservoir.

These fluctuations were strongly correlated with hydrometric data, suggesting that changes in water levels are largely driven by seasonal and interannual variations. Although hydrological variability explains most of the observed fluctuations, the results also suggest that geomorphological changes affect the shape and depth of the reservoir. These transformations may reflect cumulative sediment deposition, gradual delta formation, and shifting inflow channels, which are processes commonly reported in Mediterranean reservoirs exposed to intense erosion and irregular rainfall.

The overlap of the polygons shown in Figure 7 sometimes makes it difficult to visualize the planimetric variations of the water surface in the reservoir. This complexity arises from the fact that, during the analyzed period, several hydrometric levels are similar, making it harder to distinguish the changes

One of the most critical drivers of siltation in reservoirs is riverbank erosion, particularly along the tributaries that feed into them. Riverbank erosion is a geological process by which soil and sediment are detached and transported from riverbanks into the watercourse, leading to changes in the morphology of banks and adjacent areas (Saadon et al., 2021). This process can occur at varying speeds, sometimes resulting in landslides or collapses, and is influenced by multiple factors: (1) Fluctuating water levels: Seasonal and interannual variations increased bank saturation and promoted collapse; (2) Wave action: Wind-driven waves generate hydraulic pressure and directly dislodge bank material; (3) Tributary inflows: Concentrated flows at confluence intensify shear stress and accelerate erosion; (4) Bank geometry and material: Steep banks composed of unconsolidated soils erode faster, especially when vegetation cover is sparse (Mohd Rosli et al., 2024).

Spatial asymmetry is evident, as erosion is more pronounced along the southeastern and northeastern banks, likely due to a combination of stronger tributary inflows, steeper slopes, and weaker lithology. Conversely, the northwestern zones showed higher sediment accumulation, indicating the downstream deposition of eroded materials.

The retreat of riverbanks caused by wave action, rainfall-triggered slope failures, and insufficient vegetation cover leads to the direct introduction of eroded soil into the reservoir. These sediments tend to accumulate rapidly at river mouths, reducing water depth and significantly accelerating reservoir infilling. The observed spatial mismatch between the theoretical (simulated) and actual water surface areas highlights zones of concentrated sedimentation that are often aligned with areas prone to persistent erosion. The sediment distribution maps (Fig. 13) further confirm a consistent pattern of shoreline retreat and sediment deposition, indicating an alarming and ongoing trend.

By subtracting the 2013 DEM from the 2020 DEM, a more detailed analysis of the recent evolution of sediment deposits was obtained. This shorter time frame reveals contemporary trends, such as an increase in the sedimentation rate or, conversely, stabilization of deposits. This recent comparison helps to detect rapid or recent changes in the dynamics of sediment accumulation.

The map for the most recent period showed more moderate elevation differences, ranging from -7 m to 10 m. Areas with elevations between 3 m and 10 m indicate recent deposits, although of a lower magnitude compared to the previous period. The deposition zones were more scattered, which may suggest a more homogeneous redistribution of sediments or a reduction in sediment input in some parts of the reservoir.

The comparison between the 2020 and 2013 DEMs highlights the significant accumulation of sediments in several areas of the reservoir. Deposits are particularly concentrated in the northwestern region, where a marked increase in elevation (7-10 m) indicates intense sedimentation.

Bank erosion is especially pronounced in the southeastern and northeastern zones of the reservoir, where a notable decrease in elevation indicates significant erosion during this period.

The observed water level change of 5.13 m at the Ibn Battouta reservoir between November 9, 2015, and July 14, 2024, corresponds to a shoreline displacement of approximately 2.57 m, calculated from a slope angle of 63.43°. This displacement, which exceeds the detection limits of satellite imagery and DEM resolution, confirms the occurrence of measurable geomorphological changes. These findings align with the patterns of sediment deposition and riverbank erosion identified in the DEM comparisons and satellite-derived water surfaces, demonstrating that the reservoir underwent significant morphological evolution over the analyzed period.

Such measurable changes highlight the value of high-resolution, multitemporal monitoring tools, such as GEE, which can provide detailed datasets to support effective reservoir management strategies, particularly for agencies like the Loukkos Basin Water Agency. In practice, these datasets can guide management decisions in several key areas: (1) The early detection of erosion hotspots enables targeted interventions, such as riverbank reinforcement or revegetation; (2) The quantification of sediment volumes aids in predicting reservoir infilling rates, helping schedule dredging or sediment management programs; and (3) Scenario analysis: Simulating water level fluctuations can forecast areas at risk of erosion and optimize reservoir operations to minimize sediment mobilization.

By integrating satellite-derived NDWI data with hydrometric and geomorphological information, water authorities can adopt evidence-based strategies to maintain the reservoir capacity, reduce siltation rates, and improve long-term water resource sustainability.

Despite the valuable insights provided by this study, several limitations should be acknowledged. First, satellite imagery (Landsat and Sentinel) is subject to temporal and spatial resolution constraints and cloud cover, which may affect the accuracy of the water surface extraction and shoreline delineation. Second, bathymetric data were only available for 2013 and 2020, preventing a continuous temporal analysis of sedimentation trends and limiting the ability to capture seasonal or annual variations. Third, minor uncertainties may arise from image preprocessing, DEM interpolation, and geometric correction steps. Future studies should integrate more frequent bathymetric surveys, higher-resolution imagery, and field validation to improve the accuracy, robustness, and reliability of sediment estimations. Additionally, integrating NDWI-based analysis with machine learning shoreline extraction or LiDAR-derived topography could further improve spatial accuracy and detection sensitivity.

Based on these findings and considering the above limitations, we recommend the following actions to support sustainable sediment management and preserve the capacity of the Ibn Battouta reservoir. First, continuous satellite-based monitoring should be implemented to track water surface and sediment dynamics. Second, the structural reinforcement of vulnerable riverbanks should be prioritized to reduce erosion and sediment input into the reservoir. Third, the ecological restoration of degraded upstream areas, including vegetation rehabilitation, will help stabilize soils and limit further sediment transport. These interventions are necessary to maintain reservoir storage, ensure operational capacity, and promote long-term sustainability of water resource management in the region.

In summary, acknowledging these limitations and implementing the proposed recommendations will strengthen sediment management strategies and ensure the sustainable functioning of the Ibn Battouta reservoir and similar Mediterranean watersheds.

This study monitored the morphological evolution of the Ibn Battouta reservoir and assessed sedimentation trends from 1984 to 2024. The actual water-surface contours were extracted from satellite data, and the theoretical surfaces were determined using a reference Digital Elevation Model (DEM). The overlay of these datasets highlighted significant changes over time, revealing progressive reservoir shrinkage and significant sediment accumulation. These changes are mainly attributed to sediment contributions from tributaries and shoreline degradation, which play a crucial role in the gradual silting of the dam. The results of this analysis underscore the major impact of erosion and sedimentation processes on the functional capacity of reservoirs. Shoreline degradation is a significant source of sediment that substantially contributes to siltation. These findings emphasize the need for sustainable and proactive management of reservoir shores. Measures such as shoreline stabilization, reforestation, and watershed management should be considered to limit the sediment input and preserve the long-term functionality of the Ibn Battouta dam.