Modeling transient vegetation effects in slope stability analysis for rainfall-induced shallow landslides: CRITERIA-1D and CRITERIA-3D

Instability of natural slopes is a worldwide problem, with an increasing trend in the occurrence of landslides and the associated aggravation of socioeconomic losses. With regard to shallow landslides, phenomena exacerbation is due to extreme rainfall events, which are becoming more frequent because of the global temperature arise. Several attempts to model landslides’ occurrence over the years have been done, involving probabilistic, deterministic or hybrid approaches. Statistical-based models need to be trained upon past landslide events for detecting new future occurrences. However, many shallow landslides were recently observed to occur as new activation, and thus their triggering conditions are mainly experienced for the first time. Physically-based models have the potentiality of quantifying the different processes involved during the landslide formation. Therefore, they are particularly suitable for addressing consequences of extreme climatic events. When involving transient hydrological phenomena, these models can reach high levels of accuracy, although they are forced to make simplifications in order to be practically applied. The interest in modeling vegetation effects on both unsaturated soil mechanics and hydrology is relatively recent. Yet, only few models consider the transient reinforcement effects exerted by plants on slopes, as the most common approach in slope stability models is to consider them as static components in the analyses. The aim of this thesis is to explore the effectiveness of modeling transient stability effects induced by vegetation growth and evapotranspiration activity in physically-based deterministic models for shallow landslides. The proposed models, CRITERIA-1D and CRITERIA-3D, are agro-hydrological models developed by the Regional Agency for the Environmental Protection and Energy of Emilia-Romagna region (Arpae-SIMC). In both models a simplified slope stability analysis, based on the infinite slope assumption and on the Factor of Safety computation, has been added. Results show that the proposed methodology is effective in reproducing real case studies at different application scales, and that neglecting vegetation effects in transient hydrology computation can lead to unsatisfactory results.