Avaliação da suscetibilidade a deslizamentos em taludes de terraços agrícolas: modelos matemáticos de base física = Evaluation of landslides susceptibility in agricultural terraces: physicaly based mathematical models

• Autor(a) principal: Bateira, Carlos
• Data de Publicação: 2018
• Outros Autores: Costa, António, Fernandes, Joana, Fonseca, Bruno
• Tipo de documento: Livro
• Idioma: por
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: https://hdl.handle.net/10216/116947
• topic: Geografia; Geography

The Douro Valley presents in the eastern region extensive areas of vineyards that belong to the oldest demarcated region dedicated to wine production (RDD). The instability modeling of the agricultural terraces base on the digital elevation models (DEM) and on the physical parameters that characterize the terrain, it is possible to model the spatial variation of the susceptibility to the occurrence of shallow landslides on terraced slopes, using the SHALSTAB (Shallow Landslide Stability Model). It is verified that the DEMs resolution is an essential element of the modeling process, only possible with the acquisition of images at a low altitude, capable to build DEMs of very high resolution. In the case of SHALSTAB it can be seen that the best results become from the combination of higher resolution DEMs for the modeling of the instability component and the lower resolution ones for modeling of the hydrological component. With the contingency tables validation it is possible to identify that the model that combine a DEM of 1m resolution for the contributory area and a DEM of 40cm for the instability modeling of the riser terraces has better performance, presenting a predictive capacity of 97% of the landslides. Among the models with the best predictive capacity, there is also a relation between the true positive index and the false negative index of 2.1. The use of higher resolutions (20cm and 40cm) revealed inefficiency in the prediction of most landslides due to the low efficiency to model the internal flow. This results from the absence of similarity between the internal flow and the topographic surface deeply modified with the construction of the agricultural terraces. This similarity exists between the general topographic configuration of the slope and the internal flow, before the organization in agricultural terraces. This justifies the better performance of the combination of the lower resolution DEMs for the hydrological component and higher resolution DEMs for the instability component.