Hydromechanical Rock Mass Fatigue in Deep-Seated Landslides Accompanying Seasonal Variations in Pore Pressures

Abstract

The episodic movement of deep-seated landslides is often governed by the presence of high pore pressures and reduced effective stresses along active shear surfaces. Pore pressures are subject to cyclic fluctuation under seasonal variations of groundwater recharge, resulting in an intermittent movement characterized by acceleration–deceleration phases. However, it is not always clear why certain acceleration phases reach alarming levels without a clear trigger (i.e., in the absence of an exceptional pore pressure event). This paper presents a conceptual framework linking hydromechanical cycling, progressive failure and fatigue to investigate and explain the episodic behavior of deep-seated landslides using the Campo Vallemaggia landslide in Switzerland as a case study. A combination of monitoring data and advanced numerical modeling is used. The principal processes forcing the slope into a critical disequilibrium state are analyzed as a function of rock mass damage and fatigue. Modeling results suggest that during periods of slope acceleration, the rock slope experiences localized fatigue and gradual weakening through slip along pre-existing natural fractures and yield of critically stressed intact rock bridges. At certain intervals, pockets of critically weakened rock may produce a period of enhanced slope movement in response to a small pore pressure increase similar to those routinely experienced each year. Accordingly, the distribution and connectivity of pre-existing permeable planes of weakness play a central role. These structures are often related to the rock mass’s tectonic history or initiate (and dilate) in response to stress changes that disturb the entire slope, such as glacial unloading or seismic loading via large earthquakes. The latter is discussed in detail in a companion paper to this (Gischig et al., Rock Mech Rock Eng, 2015). The results and framework presented further demonstrate that episodic movement and progressive failure of deep-seated landslides cannot be analyzed by means of classical limit equilibrium tools but require advanced numerical models. When calibrated against slope monitoring data, the improved understanding of episodic slope movements can lead to more reliable early warning forecasting and improved landslide hazard management.