Ingenieur Vol.79 July-Sept 2019 ingenieur 2019 july-sept | Page 32

INGENIEUR The data collection in the study used multiple sources of data to support different stages of the project’s implementation. The data was divided into five main groups namely: geospatial data; expert knowledge on landslide vulnerability and risk estimations; past records of landslides and damage in Malaysia; output from landslide hazard and risk projects in the study area; publications on landslide vulnerability, hazard and risk assessments. The geospatial data was mainly used to evaluate the proposed vulnerability and risk assessment method. The data was used to extract and characterise the critical infrastructures in the study area using various image processing and spatial analysis methods. The output from the landslide hazard and risk project in the study area was used to generate landslide hazard maps. However, the applicability of the landslide hazard information heavily relies on the quality of the hazard map and the requirement of the proposed vulnerability methodology. Expert knowledge was used to support the development of parameters and indicators of landslide vulnerability and risk assessment. Apart from literature reviews, suitable methods, parameters and indicators for landslide vulnerability assessments were obtained from local experts and previous landslide risk assessment projects in Malaysia. well as the amount of data required. For example, the most detailed vulnerability assessment at a local level i.e. of individuals or households usually requires a huge amount of not readily available data. Normally for decision making purposes, vulnerability and risk assessments at regional or national level are adequate. Another related issue is on the down-scaling and up-scaling of assessment results onto different scales that might involve different levels of generalisation and assumptions that significantly affect the results. The quality and quantity of data significantly affects the outcome of the vulnerability and risk assessments. Furthermore, vulnerability and risk assessments are site-specific and take into account the characteristics of the hazard and element-at-risk (Fuchs et al. 2012; Lo et al. 2012). Different areas might require different parameters and approaches for susceptibility and hazard mapping, depending on the natural and anthropogenic causal factors of a specific hazard. Furthermore, it is acknowledged that various types of the same process (e.g. debris flow vs. rock falls for landslide processes, fluvial floods vs. pluvial floods for flood processes) can result in different damage patterns. REVIEW OF LANDSLIDE VULNERABILITY ASSESSMENTS AND MODELS Mazzorana and Fuchs (2010) highlighted issues on the challenges involved in assessing the intensity of the landslide process on individual buildings. There are also gaps in understanding the interaction between the landslide process and the affected element-at-risk which complicate the selection of a suitable intensity parameter in the vulnerability model (Papathoma-Köhle et al. 2015). A thorough review was made of the issues regarding landslide vulnerability assessments and models used to define the risk index of critical infrastructures in urban, urban highland, suburban and rural areas including active tectonic zones in Malaysia. Issues on Scale and Transferability of a Vulnerability and Risk Assessment Approach and Results One of the important issues in vulnerability and risk assessment is the scales of such analyses (Uzielli et al. 2008). The scales account for time or temporal and space scales of the assessments. The scale of the analysis determines the type of approach for vulnerability and risk assessment as 6 30 VOL VOL 79 55 JULY-SEPTEMBER JUNE 2013 2019 Issues on Landslide Intensity Determination for Vulnerability Model Issues on Input Data for Vulnerability and Risk Assessments Detailed vulnerability assessments inevitably require huge amounts of data compiled during previous occurrences of landslides. Remondo et al. (2008) set up an inventory of direct losses due to landslide disasters over a period of nearly 50 years. The inventory was based on field surveys and consultations with both local inhabitants and