Key points are not available for this paper at this time.
An integrated landslide susceptibility analysis is carried out for the central highland region of the Chaliyar River Basin in Kerala, India using bivariate statistical methods, namely the Frequency Ratio (FR) and Shannon Entropy (SE). The study addresses the complex nature of landslides, influenced by natural as well as anthropogenic factors, with specific focus on assessing the landslide likelihood of the study area. The methodology involves a systematic approach of collecting the inventory data, identifying various landslide causative factors and developing their corresponding thematic maps, spatial analysis of landslide occurrence and causative factors using GIS software and generation of Landslide Susceptibility Model (LSM) employing FR and SE algorithm, followed by model validation. Various causative factors considered for the study include slope angle, slope aspect, slope curvature, elevation, lithology, drainage density, landuse and landcover (LULC), Topographic Wetness Index (TWI) and Normalized Difference Vegetation Index (NDVI). The FR and SE algorithm enable the spatial classification of the study area into four landslide susceptibility categories namely Low, Moderate, High, and Very High. Validation of both the LSMs was carried out using Landslide Density Index (LDI) and Area Under the Curve (AUC) methods. LDI demonstrate a positive fit for both the models, which is indicative of reliability of the susceptibility predictions of the study area. A slightly higher AUC value of SE model is an indication of a high accuracy rate of SE model over FR model. This research brings out a robust methodology for predicting and identifying the landslide risks of the study area.The outcomes of this study will help in developing effective strategies to manage the landslide hazards in geologically vulnerable areas. Keywords: Landslide Susceptibility, Landslide Conditioning Factors, Frequency Ratio, Shannon Entropy, Landslide Density Index(LDI), Area Under the Curve (AUC)
P.R. et al. (Tue,) studied this question.