Landslide Stability Analysis Based on Generalized Information Entropy

Abstract

Landslide is a kind of serious geologic disaster. In the viewpoint of system theory, the landslides may be regarded as a nonlinear open system, and they are ceaselessly exchanging information and energy with their surrounding environment and inside of themselves. The occurrence of landslide is due to the energy the landslides obtain from the environment, and then the states of landslide triggering factors will be changed from disorder to order. Based on information theory, this paper presents a novel landslide stability analysis approach, that is, generalized information entropy approach. First of all, the surveying data time series of landslide triggering factors should be transformed into serial data on probability, and then the generalized information entropy of these landslide triggering factors can be evaluated by these probability serial data. From the change of generalized information entropy, it can be seen that there is a sudden increase of generalized information entropy before landslide occurs, and then generalized information entropy trends to stationary change after landslide occurs.