Landslide Analysis Approaches in Tropical Environment Region for Disaster Risk Reduction

Abstract

Landslides are natural disasters that befall practically in almost every country worldwide. Since the behaviour of the Earth is varied and the influencing factors that induce landslides are not constant, there seems to be no precise technique for assessing and forecasting the occurrence of landslides. This study selected the Cameron Highlands district, situated in Pahang, Malaysia, which accentuates reviewing numerous methods by the preceding local researcher to analyse and assess landslide incidence. A country like Malaysia is highly vulnerable to landslides due to its geographical features of high and lowlands, relatively intense precipitation, and locality in the distribution of tropical rain forests typified by dense vegetation, hot and humid temperatures throughout the year. In comprehending the landslide, most prior researchers employed numerous approaches and methods, where three qualitative methods (acceptable accuracy), two semi-quantitative methods (78% to 86% accuracy) and five quantitative methods (86% to 98% accuracy) were identified. These methods appraise multiple parameters and employ various techniques for factor research and understanding, where each method has its own set of benefits and shortcomings. The diversity of the landslide scale requires specific research in determining landslide mapping, whether by inventory, susceptibility, hazard, or risk. An application of the programme and software platform can forecast the accuracy of landslide occurrence modelling for future landslide mitigation planning. Based on the review findings, GIS and remote sensing play a crucial part in translating spatial data for more accessible analysis in furthering the research, as supported by field survey results. Each method comprising various techniques indicates that overall accuracy is applicable for the landslide analysis approaches.