Integrated disaster risk index model for the Malaysian local assessment

Abstract

Malaysia is considered a high-risk country on a global level due to the increasing number of natural disasters in recent years. Considering the increasing impact of natural disasters, implementing a local disaster risk assessment would improve the understanding and identification of potential disaster risks that could affect social system, the economy, and numerous institutions. Understanding and evaluating integrated disaster risk must consider multi-hazard and multidimensional vulnerability at the local level, particularly in developing nations like Malaysia. This primary gap has never been recorded in earlier research, and the purpose of this work is to close the gap. Therefore, this study developed an integrated disaster risk assessment index (IDRI) model to measure disaster risk within local administrative boundaries in Malaysia. The emphasis of this thesis is to assist decision makers in identifying high-risk areas that are exposed to natural disasters by considering local vulnerability factors. The proposed index model could enhance government disaster risk reduction measures by implementing an (IDRI) model and guiding decision maker on how to properly evaluate and analyse risk for mitigation, preparedness, and planning. The index was developed by expanding on the multi-hazard spatial overlapping and Methods for the Improvement of Vulnerability Assessment in Europe (MOVE) theoretical framework. In this study, the multi-hazard spatial overlapping combined two common hazards in Malaysia which are floods and landslides. This study used a quantitatively structured questionnaire survey to choose relevant IDRI model indicators based on expert opinion. The multidimensional vulnerability index (MDVI) model was developed using a combination of expert opinion and Principal Component Analysis (PCA). The IDRI map was created using Catastrophe Theory and Geographical Information Analysis (GIS) analysis. Based on the expert interviews, the study revealed that multidimensional vulnerability encompasses six dimensions, which in turn comprise 16 subdimensions and 54 indicators. This approach was applied in three urban districts of Selangor, Malaysia: Sepang, Kuala Langat, and Hulu Langat, which are located within the Langat River catchment and consist of 17 subdistricts. The spatial vulnerability assessment was conducted to classify vulnerability and risk in the study areas. The map produced five vulnerability categories (very low, low, medium, high and very high). The findings indicate that of the total vulnerability areas in the study, 7% were in the very high class, 12.6% were in the high class, 25.7% were in the medium class, 34.7% were in the low class and 20% were in the very low class. Overall, 32.9% of the total study area was found to be at risk, with 4.3% in the very high-risk area. Based on the Receiving Operating Characteristics (ROC) validation, the integrated disaster risk index model accuracy was 0.888, suggesting that the proposed model is good for evaluating risk. In comparison with the latest flood events in 2021, the IDRI components were highly correlated with disaster impact. In conclusion, the contribution of this study provides a novel perspective on disaster risk assessment by addressing several types of hazards and multidimensional vulnerability, as compared to the previous study focusing on a single hazard and a physical vulnerability factor. The model produced in this study will help governments at local levels to develop better strategies for disaster risk reduction practices and policies.