Slope stability modelling of shallow landslides at a regional scale

Abstract Rainfall-induced shallow landslides can rapidly evolve into debris flows, characterized by high velocities and destructive power, posing a significant threat to many communities. These mass movements often occur as clustered events affecting wide areas, making their prediction a challenging task. This study aims to identify a suitable distributed slope stability model to predict such events by assessing the performance of two physically-based distributed models: HIRESSS (HIgh REsolution Slope Stability Simulator) and SCOOPS3D, in predicting shallow landslides in an Alpine region of Italy (Valle d'Aosta). The models simulated two rainfall events in May 2013 and October 2018. The landslides triggered during these events were used to validate and compare the results of slope stability analyses generated by the two models. To perform a significant comparative study, three saturation scenarios (Pre-event, Peak-event, and Post-event conditions), defined by HIRESSS during the modelling process, were provided to SCOOPS3D to carry out the slope stability analyses. HIRESSS incorporates a transient hydrological model to dynamically track soil saturation changes, while SCOOPS3D does not inherently simulate hydrological processes. The results showed that HIRESSS was particularly effective in modelling the primary triggering mechanism for shallow landslides induced by increasing soil saturation in the context of the study area. This can be attributed to the fact that shallow soils and high relief energy, which define the geohydrological and geomorphological context of the area, align better with HIRESSS's infinite slope approach. This alignment enables HIRESSS to more accurately capture rapid hydrological responses and pore pressure dynamics, which are crucial for predicting shallow landslide triggering within this specific geomorphological setting. The findings emphasize the necessity of a comprehensive evaluation of these factors when selecting slope stability models, providing valuable insights for land-use planning, risk assessment, and the development of early warning systems.