Vulnerability assessment and landslide risk analysis : application to the Loures Municipality, Portugal

• Autor(es): Guillard, Clemence
• Data: 2016
• Identificador Persistente: http://hdl.handle.net/10451/25144
• Origem: Repositório da Universidade de Lisboa
• Assunto(s): Teses de doutoramento - 2016
• Tipo de Documento: Tese de doutoramento
• Idioma: Inglês
• Contribuidor(es): Repositório Científico de Acesso Aberto da ULisboa; Zêzere, José Luís, 1962-

Descrição

The present study aims to develop a method for assessing the landslide vulnerability at the municipal scale which supports the landslide risk analysis. Three susceptibility assessments to deep-seated rotational and translational slides and to shallow slides in the Loures municipality are presented in the first chapter. A bivariate statistical method called Information Value Method was used to cross the 686 inventoried landslides with seven predisposing factors (slope angle and aspect, plan slope curvature, inverse of the topographic wetness index, geology, soil types and land use). The accuracy and the robustness of the models were assessed by success and prediction-rate curves. The 20.3% of the municipality where 70% of the future landslides should occur according to the susceptibility models were selected to be included in the National Ecological Reserve. The second chapter presents a study of the social vulnerability of the 149 civil parishes of the Greater Lisbon. The method used is the Social Vulnerability Index (SoVI) method, which consists in selecting socioeconomics variables, removing the auto-correlated variables and applying a Principal Component Analysis (PCA). The seven principal components resulting from the PCA were interpreted and the SoVI values were classified based on standard deviation. The risk delimitation was effectuated by combining the SoVI map with the susceptibility maps of the main natural hazards which threaten the Greater Lisbon (earthquakes, floods, flash floods, landslides, tsunami, and coastal erosion). The exposition of the population was finally considered by combining the number and location of the residents with the risk zones map. The third chapter shows a physical vulnerability assessment of the Loures municipality buildings for different landslide magnitudes. The average of the vulnerability attributed by a pool of European landslide experts and by a sub-pool of landslide experts who know the study area on the basis of the structural type of the buildings was used. The variability of the answers was assessed by standard deviation calculation. Then, the economic value of the buildings was assessed based on the Portuguese Tax Services approach. In addition, the landslide hazard was calculated by combining the landslides spatiotemporal probability and their frequency-magnitude relationship. Finally, the landslide hazard was combined with the vulnerability and the value of the buildings in order to obtain the landslide risk. In the conclusion section, the social vulnerability and the physical vulnerability of the Loures municipality were combined twice. First, the considered social vulnerability was extracted from the second chapter results; its scale is the civil parish scale. Second, a new social vulnerability assessment was made at the basic geographic entity (BGRI) scale. Finally, the landslide risk was analysed considering the total vulnerability provided by the combination of the physical vulnerability and the new social vulnerability assessment, the landslide susceptibility, the exposition of the population and the economic value of the buildings.