Predicción de susceptibilidad a movimientos de ladera usando Base-SIG y modelos de aprendizaje automático en la ciudad de Loja, Ecuador

Abstract

Abstract:This research focuses on the analysis of soil susceptibility to slope movements in the city of Loja, the methodology focuses on finding the best statistical method that allows us to define the favorableness of the physical environment, analyze variables such as geology, slope and geomorphology in order to predict the occurrence of landslides. In order to find the best and most effective method, three maps of susceptibility to slope movements were made, the logistic regression method, the random forest method and the neural network method using the GIS Base and the landslide inventory of the city of Loja. With the definition of the best method, it is hoped to be able to implement the model and thus be able to counteract the inconveniences of instability of hillsides existing in the city, reduce the threat and support the territorial ordering of the urban area that has been affected for a long time.