Chapter 8 - GIS-based landslide susceptibility zonation mapping using fuzzy gamma operator model in part of Trans-Asian Highway (Mao-Kangpokpi), Manipur, India

Abstract

NH-39 is the only lifeline of the state Manipur. After the Indian Government adopted look east policy, the trade and road connectivity was improved. The main NH-39 turns into Asian Highway-1 (AH-1) (Imphal-Mao) and Asian Highway-2 (AH-2) (Imphal-Moreh). This highway connects all the South-Asian countries. AH-1 is highly landslide prone due to their complex geological, geomorphological, and tectonic settings along with growing anthropogenic activities, which leads to frequent landslides occurrences in this study region; hence, the road connectivity is frequently blocked. In the present study, the section of Mao to Kangpokpi from AH-1 was chosen for the landslide susceptibility zonation (LSZ) mapping using fuzzy gamma operator (FGO) model through integrated remote sensing and geographical information systems (GIS) techniques. The purpose of the study is to generate and validate landslide susceptibility zones. The landslide distribution map was prepared through fieldwork and visual interpretation of high-resolution satellite images. The landslide distribution data were divided into training (to build the model) and test data (to validate the model). There are 15 landslide influencing factors such as slope gradient, slope aspect, slope curvature, elevation, topographic wetness index (TWI), stream power index (SPI), drainage density, drainage proximity, geology, soil, lineament density, lineament proximity, land use land cover, rainfall distribution, and normalized difference vegetation indices (NDVI), which were considered to build a relationship with training landslide inventory datasets using FGO model. The final LSZ map was classified into different susceptible zones, viz., very high, high, moderate, low, and very low. The LSZ map was validated using the area under curve (AUC) method to derive success (using training data) and the prediction rate (using test data) curve and AUC value. The AUC value of success rate was 0.7935, and in the prediction rate, it was 0.7003.