Thesis Abstract

Abstract
This thesis presents the development of a Geographical Information System (GIS)-based system for landslide susceptibility mapping in a hilly region. Landslides are natural hazards that pose significant risks to human lives, infrastructure, and the environment. The study area selected for this research is characterized by hilly terrain, which is susceptible to landslides due to various factors such as steep slopes, soil characteristics, land use patterns, and rainfall intensity. The primary objective of this research is to develop a GIS-based system that can effectively map and assess landslide susceptibility in the study area. The research methodology employed in this study includes a comprehensive literature review to understand the existing methods and techniques for landslide susceptibility mapping. The data collection process involved acquiring topographic maps, satellite imagery, geological maps, and rainfall data. Various GIS tools and techniques were utilized to analyze and process the collected data, including slope analysis, aspect analysis, land cover classification, and rainfall intensity mapping. Machine learning algorithms, such as logistic regression and random forest, were applied to develop a predictive model for landslide susceptibility mapping. The findings of this research reveal that the developed GIS-based system can effectively identify areas within the hilly region that are prone to landslides. The system incorporates multiple layers of spatial data to generate a comprehensive landslide susceptibility map, which can be used by decision-makers, urban planners, and emergency response teams to implement mitigation measures and reduce the impact of landslides in the study area. The limitations of the study include data availability, scale of analysis, and uncertainties associated with landslide susceptibility modeling. The significance of this research lies in its contribution to enhancing the understanding of landslide susceptibility mapping in hilly regions using GIS technology. The developed system provides a valuable tool for assessing and managing landslide risks, thereby improving disaster preparedness and response strategies. The integration of machine learning algorithms with GIS techniques offers a novel approach to landslide susceptibility mapping, which can be further refined and applied in other regions facing similar challenges. In conclusion, the GIS-based system developed in this research represents a significant step towards enhancing landslide susceptibility mapping in hilly regions. The study highlights the importance of utilizing spatial data and advanced analytical tools to assess natural hazards and mitigate their impact on vulnerable communities. Future research directions include refining the predictive models, incorporating real-time monitoring data, and expanding the application of GIS technology in disaster risk reduction efforts. Keywords GIS, Landslide Susceptibility Mapping, Hilly Region, Machine Learning, Spatial Analysis, Disaster Risk Reduction

Thesis Overview