Vol.:(0123456789) 1 3 Euro-Mediterranean Journal for Environmental Integration (2020) 5:64 https://doi.org/10.1007/s41207-020-00205-8 ORIGINAL PAPER Comparison of a data‑based model and a soil erosion model coupled with multiple linear regression for the prediction of reservoir sedimentation in a semi‑arid environment Ali EL Bilali 1  · Abdeslam Taleb 1  · Bouchaib EL Idrissi 2  · Youssef Brouziyne 3  · Nouhaila Mazigh 1 Received: 13 March 2020 / Accepted: 21 September 2020 © Springer Nature Switzerland AG 2020 Abstract Reservoir sedimentation is a crucial challenge in planning and managing sustainable surface water resources in arid and semi-arid regions and must be assessed with accuracy. Both data-based models and conceptual models can be valuable tools for predicting reservoir sedimentation. In this study, we used an artifcial neural network (ANN) approach and a modifed Universal Soil Loss Equation coupled with multiple linear regression (MUSLE-MLR) model to predict yearly sedimentation in the Sidi Mohammed Ben Abdellah reservoir, located in a semi-arid region of Morocco. To construct the MUSLE-MLR model, we frst calibrated and validated the MUSLE on 32 storms at four gauging stations upstream of the dam to estimate the sediment yield at these four gauging stations; we then developed the MLR model for combining sediment yield and reservoir sedimentation. The results of this model were then compared with the performance of the ANN model that was trained and validated over the periods 1975–2008 and 2009–2015, respectively. The comparison revealed that the calibrated MUSLE model is fairly useful to predict sediment yield at the watershed level. However, comparison of the two models during the validation process showed that the ANN (R 2 0.91, Nash–Sutclife Efciency [NSE] 0.820) is more accurate and more suitable than the MUSLE-MLR model (R 2 0.819, NSE − 1.592) to predict reservoir sediment in the Sidi Mohammed Ben Abdellah reservoir. The fndings of this study contribute to the armamentarium of potential tools that can be used to predict and manage reservoir sedimentation at the watershed and reservoir levels in a semi-arid context. Keywords Reservoir · Sedimentation · Soil erosion · Conceptual-based model · Artifcial neural network Introduction In arid and semi-arid regions, dam reservoirs are consid- ered to be one of the major water resource infrastructures that play important roles in supplying water for agricultural, domestic and industrial purposes. The efects of climate change in combination with the challenges associated with population growth and economic development have led to increased stress on water resources in these regions (Al- Saidi et al. 2016; Droogers et al. 2012). Reservoir sedimen- tation represents an additional threat to the sustainability of reservoirs in terms of the quantity and quality of col- lected water (Boardman et al. 2009; Palmieri et al. 2001; Zhou and Wu 2008). In general, reservoir sedimentation is assessed using topo-bathymetric data collected in surveys conducted at regular frequency, with the frequency depend- ent on the reservoir’s importance (Bengora et al. 2018; Has- san et al. 2017; Maina et al. 2018). Nowadays, the control and management of land use in watersheds have become a management tool for soil and water conservation using the Best Practices Management (BPM) to reduce soil erosion (Adeogun et al. 2018). Therefore, an accurate prediction of reservoir sedimentation be relevant in terms of integrat- ing sustainable management policies to preserve lands and reservoirs. Communicated by Sudip Chakraborty, Chief Editor. * Ali EL Bilali ali1gpee@gmail.com 1 Hassan II University of Casablanca, Faculty of Sciences and Techniques, Mohammedia, Morocco 2 Institut d’Innovations en Éco-matériaux, Écoproduits et Éco énergies, à base de biomasse, Université du Québec à Trois-Rivières, Trois-Rivières, Canada 3 Mohammed VI Polytechnic University (UM6P), International Water Research Institute, Benguerir, Morocco