ABSTRACT
There are few studies on shallow foundations placed close to the crest of slopes. Many of those studies are based on small scale physical models and include recommendations for the reinforcement layout (usually normalised to the width of the footing, B). Such studies have limitations, particularly regarding the unrealistically low stress levels applied. To address such limitations, this paper presents a numerical analysis of a full-scale footing close to the crest of a sandy slope reinforced with geosynthetics, using the finite element method. The parametric analysis focuses on the effect of the depth of the upper reinforcement layer (u) and number of reinforcement layers (n). The response of numerical models reinforced with one layer of geosynthetic was compared to that of the unreinforced model analysed under the same conditions. The installation of a reinforcement at 0.3B allowed a significant increase in the bearing capacity of the shallow foundation. For the same prescribed displacement, the failure mechanism of the reinforced model involved a smaller mass of soil and exhibited more localised shear strains than the unreinforced model. In contrast to what has been reported in the literature for reduced scale models, the optimal depth for one layer of reinforcement was 0.3B (and not 0.5B). This value agrees with recent studies performed in centrifuge and full-scale models. For the optimal depth of reinforcement, adding a second reinforcement layer (equally spaced from the first), led to the maximum bearing capacity improvement: 70% more of the unreinforced model and 38% more of the model with one layer of reinforcement.
