Statistical Approaches for Landslide Susceptibility Assessment and Prediction

Thisbookfocuses on the spatial distribution of landslide hazards oftheDarjeeling Himalayas. Knowledge driven methods and statistical techniques such as frequency ratio model (FRM), information value model (IVM), logistic regression model (LRM), index overlay model (IOM), certainty factor model (CFM), analytical hierarchy process (AHP), artificial neural network model (ANN), and fuzzy logic have been adopted to identify landslide susceptibility. In addition,a comparison between various statistical models were made using success rate cure (SRC) and it was found that artificial neural network model (ANN), certainty factor model (CFM) and frequency ratio based fuzzy logic approach are the most reliable statistical techniques in the assessment and prediction of landslide susceptibility intheDarjeeling Himalayas. The study identified very high, high, moderate, low and very low landslide susceptibility locations to take site-specific management options as well as to ensure developmental activities intheDarjeeling Himalayas. Particular attention is given to the assessment of various geomorphic, geotectonic and geohydrologic attributes that help to understand the roleofdifferent factors and corresponding classes in landslides, toapplydifferent models,andto monitor and predict landslides. The use of various statistical and physical models to estimate landslide susceptibility is also discussed. The causes, mechanismsand types of landslides and their destructive character are elaborated in the book. Researchers interested in applying statistical tools for hazard zonation purposes will find the book appealing.