THERMO-HYDRO-MECHANICAL NUMERICAL MODELLING TO ASSESS THE ROLE OF CLIMATIC FACTORS IN THE STABILITY OF ALPINE ROCK SLOPES

This work aims at increasing the knowledge of climate impacts on the stress-strain evolution of rock slopes, through the adoption of a multi-disciplinary approach and the development of multiphysics numerical analyses. Several case studies of past and active rockslide events in the Alpine region were analyzed, for which the effects of different couplings were explored. By means of a methodological process that includes geological and geomechanical characterization of rock slopes, the definition of a geological conceptual model, and the development of numerical stress-strain analysis, this work is intended to be a valid reference for the analysis of rock slope evolution in response to climate inputs. Concerning the hydro-mechanical (HM) couplings, two case histories in which exceptionally heavy rainfall events led to slope failure are discussed. In particular, the Piuro landslide (Bregaglia Valley, Sondrio, Italy), which occurred in 1618, and the 2012 Cimaganda rockslide (San Giacomo Valley, Sondrio, Italy) were back-analyzed with a focus on their triggering mechanisms related to the distribution of pore pressure, resulting from extreme rainfall events. The results confirmed the dominant role of HM couplings in defining slope instability processes. Next, the effects of surface temperature variations on slope stability were analyzed along the Cimaganda rock slope, with particular focus on a large historical rockslide. The combined effect of long- and short-term thermal evolution were numerically explored, overlapping the late-Pleistocene/Holocene warming trend to the seasonal temperature oscillations. It was shown that although slope failure events do not appear to be directly triggered by thermal cycles, the latter over a long period can lead to considerable mechanical slope degradation and strain accumulation. Moreover, a thermal monitoring system placed along the Cimaganda slope provided some preliminary evidences on the impact of current environmental loadings (from daily to seasonal) on the distribution of temperature inside the monitored rock-mass and the resulting rock-joint deformations. Then, to assess the relationship between climate variables and the evolution of an active rockslide, the Ruinon case study (Frodolfo Valley, Sondrio) was analyzed. In particular, after a careful assessment of the factors controlling rockslide evolution via numerical stress-strain analysis and hydrogeological modelling, coupled thermo-poro-mechanical modelling was carried out. Once calibrated, the model allowed to simulate the recently measured velocity history of the rockslide and was then adopted to explore the impact of forthcoming climate changes on rockslide activity. This allowed to gain further insight into the link between climate inputs and rockslide activity, providing the basis for the development of a physics-based warning system. The presented research allowed to: •Correctly simulate the evolution of the analyzed Alpine rock slopes, by considering the mainpredisposing and triggering factors. •Gain deeper understanding on the quantitative relationships between temperature, pore pressuresand strain rates in Alpine rockslides via numerical modeling. •Provide advice on the development and calibration of physics-based numerical models thatinclude climate scenarios in the prediction of future landslide activity. The presented multi-disciplinary approach (geological characterization, conceptual modeling, numerical analysis) has proven to be a valid method for the investigation of rock slopes exhibiting gravitational instability phenomena. It can be deduced that multiphysical approaches should no longer be overlooked in slope stability analyses, as their role in forecasting and predictive applications is of paramount importance.