Rainfall-induced Landslide Hazard Rating System

Abstract

This research develops a Landslide Hazard Rating System for the rainfall-induced landslides in the Chenyulan River basin area in central Taiwan. This system is designed to provide a simplified and quick evaluation of the possibility of landslide occurrence, which can be used for planning and risk management. A systematic procedure to investigate the characterization of rainfall distribution in a regional area is developed in the first part of the thesis. Rainfall data for approximately one decade, 2002 to 2008, from 9 rainfall stations in the study area are included, in which a total of 46 typhoons are selected and categorized into 3 typhoon paths: the Northeastern, Northwestern, and Western. The rainfall distribution affected by typhoon paths in a region is thereby determined. The second part of the thesis is the Landslide Hazard Rating System, which integrates different hazard factors: bedrock geology, aspect, and slope gradients. This analysis is based on the specific characterization of the study area, which consists of the relative topographic relief (aspect and slope gradients) and variable bedrock geology. The method of normalized difference is used for examining the relationship of the topographic features to landslide occurrence. Although this study is conducted in a specific area, this landslide hazard rating system can be applied to other locations. Finally, a concept of a rainfall-induced landslide analytical system is proposed to combine the rainfall distribution analysis and the landslide hazard rating system. This analytical system is intended to include and address the relationship of rainfall and landslide occurrence by combining characterizations of rainfall, topography, and landslide potential. Additionally, this study recommends that, in future work, theoretical models of rainfall distribution and laboratory tests of soil and rock samples be included. Together, these will constitute a basis for the prediction of landslide occurrence. The ultimate goal of future work should be the development of a system for assessing and forecasting rainfall-induced landslide risks, which can become the foundation for a comprehensive risk management system for use in planning.