Post-failure analysis of landslides using the material point method

Abstract

Slope stability analysis is undoubtedly one of the most complex problems that civil engineers deal with. The evolution of deformation mechanisms of slopes is commonly schematized in four different stages: pre-failure, failure, post-failure and eventual reactivation. Traditional numerical methods, such as the finite element method and the finite difference method, are commonly employed to analyze the slope response in the pre-failure and failure stages under the hypothesis of small deformations. On the other hand, these methods are unsuitable for simulating the post-failure behavior due to the occurrence of large deformations. However, an adequate analysis of this latter stage and a reliable prediction of the landslide kinematics could be particularly useful for minimizing the associated risk or to establish the most suitable mitigation measures for land protection. Among the numerical techniques which have been recently developed to overcome the above-mentioned limitation, the material point method (MPM) is employed in this dissertation to analyze the post-failure stage of two real landslides: the Senise landslide (Basilicata) and the Maierato landslide (Calabria), both in Southern Italy, occurred in 1986 and 2010, respectively. The numerical analyses allow to faithfully simulate the real phenomena. In particular, with referring to the Senise landslide, the numerical analysis provides results that match satisfactorily well the field observations when both the slip surfaces detected by the installed inclinometers are accounted for. Besides, the lowest values of the shear strength parameters obtained from the laboratory tests have to be used. Moreover, an improvement of results is gained accounting for the presence of the existing structures as well. Concerning the Maierato landslide, symbol of the hydrogeological instability in Calabria (Southern Italy), the analysis performed using the material point method allows to successfully simulate the observed phenomenon, despite the complexity of this landslide regarding its size, catastrophic failure and long run-out distance. The obtained results demonstrate that an adequate analysis of the post-failure stage can lead to a better understanding of the complex mechanical mechanisms that characterize some landslides, and therefore help significantly to establish the most effective stabilization measures.