<p>In Morocco, groundwater resources constitute the main source of water for irrigation and drinking purposes. The springs-complex of Bittit-Ribaa is a major natural source of water to local villages and Meknes City. The flow of these springs is maintained by Elhajeb-Ifrane aquifer system. In this context, this study presents the development and calibration of a transient groundwater flow model for understanding the dynamics of Elhajeb-Ifrane aquifer, evaluating the uncertainty in model parameters and forecasts of interests to groundwater resources managers. The groundwater flow model was simulated with a finite-difference modular flow model (MODFLOW) code, and represented the hydrogeological dynamics of the region, incorporating elements such as hydraulic conductivity, specific storage, and boundary conditions derived from geological settings. The Drain packages simulate surface water interactions, while groundwater withdrawals are modeled through the Well package. Calibration was achieved using the PEST software, incorporating automated pilot points combined with Tikhonov regularization to refine parameter estimates. Furthermore, an uncertainty analysis was carried out for the estimated parameters and predictions using relative error variance reduction. The results of the calibration showed a good fitting between simulated and observed values. The model’s water balance indicates a decrease in recharge and spring discharge, underscoring the aquifer’s vulnerability to climatic variations and anthropogenic activities. The analysis of relative error variance reduction gives insights on which parameters are optimized more by the calibration dataset, predictive variance reduction and the contribution of parameters to the predictive uncertainty.</p>