Accurate precipitation data are essential for effective water resource management. This study aimed to correct precipitation values from the ERA5_Ag reanalysis dataset using observational data from 20 meteorological stations located in the Tensift basin, Morocco. Five machine learning models were evaluated: MLP, XGBoost, CatBoost, LightGBM, and Random Forest. Model performance was assessed using RMSE, MAE, R², and bias metrics, enabling the selection of the best−performing model to apply the correction. The results showed significant improvements in the accuracy of precipitation estimates, with R² ranging between 0.80 and 0.90 in most stations. The best model was subsequently used to correct and generate raster maps of corrected precipitation over 42 years, providing a spatially detailed tool of great value for water resource management. This study is particularly important in semi−arid regions such as the Tensift basin, where water scarcity demands more accurate and informed decision−making.

Climate change, marked by decreasing rainfall and increasing extreme events, represents a major challenge for water resources, particularly in semi-arid regions. To estimate aquifer recharge, it is essential to assess the fraction of precipitation contributing to groundwater recharge and to implement a water balance model. However, the limited number of rainfall stations has led researchers to rely on satellite and reanalysis rainfall products. The accuracy of these datasets is essential for reliable hydrological modeling. In this study, we evaluated five rainfall products—CHIRPS, ERA5_Ag, CFSR, GPM, and PERSIANN-CDR—by comparing them to ground measurements from gauging stations in the central Haouz region of Marrakech. The evaluation was conducted at three temporal scales: daily, monthly, and annual. Statistical metrics, including RMSE, MAE, NSE, Bias, and Pearson correlation, as well as classification metrics (accuracy, F1 score, recall, precision, and Cohen’s Kappa), and wavelet analysis, were applied to assess the accuracy of the products. The results identified ERA5_Ag and GPM as the most accurate products in capturing rainfall events. Nevertheless, ERA5_Ag showed a high bias. After applying the quantile mapping method to correct the bias, the product exhibited greater accuracy. The corrected datasets from these two products will be used to estimate recharge over the last 30 years, contributing to the development of a hydrogeological model for groundwater dynamics.

Climate change, marked by decreasing rainfall and increasing extreme events, represents a major challenge for water resources, particularly in semi-arid regions. To estimate aquifer recharge, it is essential to assess the fraction of precipitation contributing to groundwater recharge and to implement a water balance model. However, the limited number of rainfall stations has led researchers to rely on satellite and reanalysis rainfall products. The accuracy of these datasets is essential for reliable hydrological modeling. In this study, we evaluated five rainfall products—CHIRPS, ERA5_Ag, CFSR, GPM, and PERSIANN-CDR—by comparing them to ground measurements from gauging stations in the central Haouz region of Marrakech at three temporal scales: daily, monthly, and annual. Statistical and classification measures were applied, as well as wavelet analysis, to improve the accuracy of the evaluation. The results identified ERA5_Ag and GPM as the most accurate products in capturing rainfall events. Nevertheless, ERA5_Ag showed a high bias. After applying the quantile mapping method to correct the bias, the product exhibited greater accuracy. The corrected datasets from these two products will be used to estimate recharge over the last 30 years, contributing to the development of a hydrogeological model for groundwater dynamics.

Located downstream of the Ourika watershed near the commune of Chouiter, the bridge studied is built over the Oued Ourika at the level of the N9 (national road linking Marrakech to Ouarzazat). The flood event of November 2014, reaching a peak flow of 347 m3/s, destabilized the bridge’s overall structure. Hence, the need to rebuild a new bridge, taking into account the magnitude of these extreme events. Hydraulic modeling involves creating mathematical and computer-based representations of hydraulic systems. Its purpose is to simulate water flow behavior and map flood-prone areas. Before initiating any hydraulic simulation, it is imperative to conduct a hydrological study to provide the input data for the hydraulic model. Precipitation records, evaporation rates, soil permeability, historical peak flow data, and site topography constitute, on the one hand, the product of the hydrological study and, on the other, the inputs for the hydraulic model. The hydraulic model is based on Saint-Vant equations to solve the conservation of mass and momentum equations. The aim of this work is to determine the critical water level at the bridge for a 100-year flood in order to optimize the height of the new bridge.