A conceptual model to analyse a rainfall-induced slow-moving rock slide

Abstract

Landslide risk assessment and management cannot disregard an accurate analysis of the triggering mechanisms and hydro-mechanical features, that characterise the landslide body in the pre- and post-rupture phases. For these reasons, it should be considered that, in the case of landslides reactivated by precipitation, rainwater infiltration contributes to the increase in pore water pressure along the sliding surfaces. Fluctuations in pore water pressure have a clear effect on the available shear strength and therefore, play a key role in the variation of the cumulative displacements of the landslide body over time. This study focuses on a clay slope where large blocks of fractured rock slowly slide. These blocks are separated from each other by open fractures in which rainwater infiltrates. The rock blocks have different geometric dimensions, with shapes ranging from "towers" to "slabs." The hypothesised scenario is therefore a translational landslide triggered by rainfall. This type of instability is physically interpreted using a new simplified hydro-mechanical model, that allows the mechanical simulation of the landslide body during the propagation phase of movement. The model considers of several crucial aspects of the hypothesized mechanism, such as the effect of rainwater within the fractures, the permeability of the fracture system, the viscosity and shear strength at the sliding surface, and the diffusion of water pressure at the base of the blocks over time. The interaction between two or more adjacent blocks sliding is simplified in the model by the "squeezing" effect of the water present inside the fractures. This effect depends on the variation in fracture aperture and the volume of water stored in the fracture. The model uses a finite difference approach to integrate the velocity over time and to compute the displacement of the different rigid blocks of the landslide body. To evaluate the performance of the model and to highlight the role of the modelling parameters, about 120 simulations belonging to nine different synthetic scenarios were carried out. The scenarios considered both single-block landslide and multi-block systems (i.e., compound landslides). The results obtained show physically reliable simulations, suggesting that the model can be considered as a promising tool for interpreting the pre- and post-failure displacements. Furthermore, the model can be applied to a wide range of hydro-mechanical and climatic scenarios. Future developments of this research include the validation of the model using the monitoring data from a real case study. The case study will have similar characteristics to those of the reference landslide mechanism. The validation will be carried out in a two-step procedure. In the first step, the conceptual model parameters will be calibrated on the monitoring data from a reference period. In the second step, the modelled data will be compared with the measured data from the monitoring period following the reference period.