It has long been known that failure of soils is controlled by intrinsically weak mineral phases (such as clays) and transient pore pressure fluctuations. Both factors may be involved in triggering of slope failure; however, it is unclear to what extent each of them contributes. To shed light on this problem, we compare geotechnical shear tests on dry mineral standards (clays, quartz) and numerical ‘shear box’ experiments using the Discrete Element Method (DEM) on dry particle assemblages. The role of fluid is additionally monitored by adding water in the analogue tests (humid to fully saturated conditions). A series of geotechnical shear tests (up to 40MPa normal stress) and numerical ‘shear box’ models in a fluid-free environment indicate that sediment composition (namely the presence of clay minerals) is a major factor in shear strength or frictional stability of granular materials. Because µ is further affected by fluid saturation, an effect that is maximized in swelling clays, added humidity or aqueous fluid cause µ to decrease by a factor of 4 in the analogue tests. However, our data suggest that mineralogical control alone may be sufficient to trigger slope instability, for instance owing to the concentration of clay minerals in zones of weakness. Fluids may enhance the effect by lowering effective stresses.
Comparative numerical and analogue shear box experiments and their implications for the mechanics along the failure plane of landslides