LANDSLIDE SUSCEPTIBILITY MAPPING USING GIS-BASED FREQUENCY RATIO AND SHANNON ENTROPY, IN PART OF CHAMOLI DISTRICT, UTTARAKHAND, INDIA

Abstract

Landslides are one of the most adverse naturally or artificially occurring hazards that results in a great loss to life and property across the globe. Moreover, people living in the upside of landslide completely lose their access to the basic amenities like: markets, food, schools, hospitals, etc. with a fear of upcoming danger of another landslide event. Therefore, it becomes essential to determine areas which are most likely to get affected by landslides, for ensuring risk mitigation and disaster resilient planning of infrastructure projects by government and private entities involved in construction projects. This study aims at evaluating the effectiveness of two GIS-based statistical approaches namely, Frequency ratio (FR) and Shannon Entropy (SE) for the landslide susceptibility mapping of a region in Chamoli district, in Uttarakhand, India. There are a lot of ongoing and proposed infrastructure projects in the area due to which, there is a surge in tourism in the area. A total of ten landslide factors which directly or indirectly influence the landslide occurrence are considered in this research study namely: elevation, slope, aspect, curvature, lithology, distance to roads, distance to faults/lineaments, distance to river, topographic wetness index, and stream power index. The landslide inventory data (from 2005-2019) has been prepared from Bhukosh Portal by Geological Survey of India in point shapefile format. The obtained dataset for landslide inventory was then randomly divided into training (80%) and testing (20%) datasets. The relationship existing between landslide points and the taken landslide causative factors has been validated through the application of the incorporated models. A total of two landslide susceptibility maps were prepared from the outputs of this research work and were validated through Area Under Curve (AUC) of Receiver Operator Characteristics (ROC) curve. The resultant maps from this research study can be extremely useful for the planners and designers from similar areas for ensuring a safe and disaster resilient infrastructure against a destructive natural calamity like landslide.