Vulnerability analysis of buildings in areas affected by slow-moving landslides and subsidence phenomena

Slow-moving landslides and subsidence phenomena yearly induce huge damages both direct (on structures and/or infrastructures with them interacting) and indirect (corresponding to the associated economic losses). For this reason, studies aimed at analyzing and predicting the aforementioned damages are of great interest for Scientific Community and Authorities in charge of identifying the most suitable strategies for the land-use planning and management of urban areas affected by slowmoving landslides and subsidence phenomena. However, carrying out the activities related to the pursuit of those goals is not straightforward since it usually requires high costs due to the great amount of data to be collected for setting up reliable forecasting models as well as the development of proper procedures that take into account i) the identification and quantification of the exposed elements; ii) the definition and estimation of an intensity parameter; iii) the prediction of the damage severity level (generally associated with the attainment of a certain limit state). In this PhD Thesis some original procedures are proposed. In particular, on the basis of empirical and numerical methods, fragility and vulnerability curves are generated in order to predict the damage to buildings in subsidence- and slow-moving landslide-affected areas. The proposed empirical procedures, based on the joint use of DInSAR data (provided from the processing of images acquired by Synthetic Aperture Radar via Differential Interferometric techniques) and information on damages suffered by buildings (recorded and classified during in situ surveys), were tested on case studies in The Netherlands, affected by subsidence phenomena, and in Calabria Region (southern Italy) for slow-moving landslide-affected areas. The procedure based on the adoption of a numerical method was applied on a structural model representative of a single building. With reference to subsidence phenomena, the analyses were carried out for a densely urbanized municipality following a multi-scale approach. In particular, at medium scale, the subsiding areas that are most prone to ground surface settlements along with their spatial distribution and rates, were preliminarily detected. The above ground surface settlements (here considered as subsidence intensity parameter) combined with the results of an extensive damage survey on masonry buildings, allowed first retrieving, at large-scale (on building aggregates) and at detailed scale (on single buildings), the relationships between cause (settlements/differential settlements) and effect (damage severity level); then, empirical fragility curves were generated for structurally independent single buildings. These latter were validated via their comparison with fragility curves generated, with reference to two others densely urbanized municipalities, for buildings with similar structural typology (masonry) and foundation type (shallow or deep). Finally, fragility and vulnerability curves for masonry buildings were generated by using the entire database of damages. As for slow-moving landslides, the analyses were carried out at large scale. In particular, the joint use of DInSAR and damage surveys data allowed analyzing the consequences induced on the buildings (either of masonry or reinforced concrete) with shallow foundations by retrieving the causeeffect relationships and generating empirical fragility and vulnerability curves. Finally, the numerical analyses carried out on a structural model representative of a single masonry building, allowed to go in-depth in the different aspects contributing to the onset and development of building damages as well as to quantify the uncertainties inherent to the addressed issue. The obtained results highlight the huge potential of the fragility and vulnerability curves generated according to the proposed procedures that, once further calibrated/validated and jointly used with a continuous monitoring of the intensity parameter via conventional (e.g., inclinometers, GPS, topographic leveling) and/or innovative (e.g., SAR images processed via DInSAR techniques) systems, can be valuably used as tools for the analysis and prediction of the damage to buildings for land-use planning and urban management purposes in subsidence- and slow-moving landslide-affected areas. [edited by author]