Avaliação a suscetibilidade de deslizamento de terra na bacia hidrográfica do Rio Trombetas via lógica fuzzy

Abstract

The landslides are defined as rapid mass movements, which present a well-defined rupture plane causing considerable social and economic damages worldwide each year. The phenomena of mass instabilisation are conditioned by many factors, such as climate, lithology and rock structures, pedology, morphology, anthropic action and others. Prevention actions such as the effective warning system and the establishment of areas susceptible to these processes are necessary because they can minimise the losses and damages caused by these disasters. In this sense, the objective of this research was to elaborate a map of the susceptibility of the landslide to the Trombetas River basin through fuzzy logic, since the area of study is of great economic interest, mainly to the exploration mineral. We used a Geographic Information System (GIS) environment software and the MATLAB in Mamdani type Fuzzy Inference System (FIS) to develop a qualitative model for predicting susceptibility and critical thresholds of landslide warning scenarios. Seven (7) input parameters were included in the model, characterising the topographic, pedological and environmental conditions. The results showed that in the Trombetas River basin there are areas classified as very low (14.11%), low (47.23%), moderate (35.08%), high (0.20%) and very high (3.39%) susceptibility of the landslide.