Assessment of landslide susceptibility for Meghalaya (India) using bivariate (frequency ratio and Shannon entropy) and multi-criteria decision analysis (AHP and fuzzy-AHP) models

Abstract

The main goal of the present study is to generate the GIS-based landslide susceptibility map (LSM) of Meghalaya, India. For this purpose, two bivariate statistical (FR and SE) and multi-criteria decision analysis (AHP and Fuzzy-AHP) methods are utilised. The study area is situated over the Shillong Plateau in the North Eastern Region of India and is subjected to numerous landslides due to heavy rainfall and seismic activities. To obtain the LSMs, a landslide inventory of 1330 events and 15 conditioning factors are prepared. The inventory dataset is randomly split into a 70/30 ratio to make training and testing datasets for the study. The results show that the southern escarpment, the southeast region and the hillslopes along the roadsides are highly susceptible to a landslide. The LSMs are validated using the area under the curve (AUC) values and F1-scores, in which AHP (AUCAHP = 0.913; F1-score = 0.884) shows the highest prediction accuracy for Meghalaya. The generated LSMs can help in landslide risk mitigation strategies and sustainable infrastructure development.