Search for a command to run...
The Draa Sfar polymetallic mine, located near Marrakech in Morocco, represents the deepest currently operating underground mine in North Africa, with workings extending beyond depths of −1200 m. At such depths, mining activities are conducted within weak, highly anisotropic foliated black pelites, where recurrent instability mechanisms, most notably rib buckling and crown deterioration, are frequently observed, especially in drifts developed parallel to the foliation planes. In this context, the present study integrates detailed structural field observations with two-dimensional finite-element modelling using RS2 in order to analyse excavation-scale stability within these schistose pelitic rocks. Both numerical simulations and field evidence indicate that increasing depth-related confinement, together with a dominant in situ stress regime, favours stress channelling and localized damage development, while the pronounced transverse weakness of the pelites exerts a primary control on failure kinematics, including schistosity-parallel spalling, asymmetric rib buckling, and shear along inclined foliation intersecting the excavation back. Instability processes are further intensified by excavation geometry and mine layout: angular, square-shaped profiles and foliation-parallel drift orientations generate steeper stress gradients and greater convergence compared to arched sections, while proximity to stopes and adjacent openings enhances mining-induced stress redistribution and associated deformation. Intersection areas emerge as the most critical configurations, where the superposition of stress perturbations and structurally controlled damage mechanisms accelerates wall convergence and roof sagging. Overall, these findings demonstrate that drift stability cannot be adequately evaluated using generic design criteria when excavation geometry, interaction effects, and structural anisotropy exert a dominant influence on mechanical behaviour. Consequently, a fully integrated approach that combines drift geometry optimisation, detailed structural mapping, site-calibrated numerical modelling, and in situ monitoring is required to achieve reliable stability assessment and control.