Comparación entre Mora-Vahrson y Lógica difusa en la susceptibilidad a movimientos de tierra en el cantón Morona – Amazonía ecuatoriana

Abstract

Landslides have historically caused the greatest damage on the planet, accounting for approximately 17% of all deaths caused by natural hazards worldwide. Unfortunately, due to the effects of climate change, human and economic losses are not expected to decrease. There are several methodologies for mapping landslide susceptibility, of which, at the regional level in Latin America, the Mora-Vahrson and fuzzy logic models stand out. Ecuador is a tropical country that has suffered several natural disasters, the last one was a landslide in Alausí-Chimborazo where 65 people died; however, there are still areas that have been little studied, such as the Amazon region, which is presented as a case study in the Morona canton. The main objective was to generate a landslide susceptibility model using fuzzy logic and Mora-Vahrson, and to determine the method that best fits their geographical reality. The resulting models were classified into five risk levels, where 80.73% of the landslides recorded in high and very high susceptibility zones were correctly represented with fuzzy logic, while 33.03% of accuracy was obtained with Mora-Vahrson. The comparison between models determined that the parishes of Macas, Alshi, Zúñac, San Isidro presented greater susceptibility, but with the fuzzy probabilistic modeling it was possible to identify in greater detail the level of risk to facilities present in the territory. This advantage is due to the scale of cartographic information of the variables that are used in each method, and that affect the accuracy of the results, so for risk management purposes it is advisable to use methodologies such as fuzzy logic in works at scales ≥1:25000, in places similar to those analyzed in this study, such as the Amazonian territories of Ecuador