The project aims to improve the capabilities to predict the occurrence of landslide crisis episodes. The research activities focus on the changes in the stability conditions of natural slopes when subjected to wet/dry cycling. Soil properties can undergo strong modifications as a consequence of cyclic variations in the degree of saturation, which can lead to a reduction of the available shearing resistance in the long term. The project involves three main research activities. The first phase deals with study of the evolution of natural soil fabrics as a consequence of imposed weathering processes; modern technologies will be applied to analyse those changes at a microscopic level. This analysis is expected to provide a better understanding of the involved physical processes. During the second phase, the effects of cyclic wetting and drying episodes on natural soil samples will be quantified at a macroscale by an extensive laboratory testing programme. The experimental programme involves the use of advanced testing facilities, available at the host institute, which are necessary to control the pore water pressure of the soil in unsaturated states. A new apparatus will also be developed to measure the shearing resistance of the soil after imposing cyclic changes in the degree of saturation using controlled suction. In the third phase, the results from the experimental work will be used to validate a constitutive model for unsaturated soils. This model will take into account the effects of the changes in the degree of saturation on the mechanical response. Finally, the model will be used in a finite element framework to simulate and forecast the behaviours in selected case studies.