Writers : Rodeano Roslee, Tajul Anuar Jamaludin, Mustapa Abd Talip, James Anthony Collin, Budirman Rudding & Ismail Abd Rahim
Date : 11-12 June 2011
Publisher : National Geoscience Conference 2011
Location : The Puteri Pacific Johor Bahru & Persada International Convention Centre, Johor
A very good landslide risk analysis (LRA) model was developed for the area of Kota Kinabalu, Sabah. Using multivariate statistics, the interactions between spatial factors and landslide distribution were tested, and the importance of individual factor to the LRA was defned. On the basis of the statistical results, landslide prediction model were developed using the Analytical Hierarchy Process (AHP) method. The AHP is defned as theory of measurement through pairwise comparisons and relies on the judgements of experts to derive priority scales. It is these scales that measure intangibles in relative terms. The comparisons are made using a scale of absolute judgements that represents, how much more, one element dominates another with respect to a given attribute. The judgements may be inconsistent, and how to measure inconsistency and improve the judgements, when possible to obtain better consistency is a concern of the AHP. The derived priority scales are synthesised by multiplying them by the priority of their parent nodes and adding for all such nodes. LRA, beside risk evaluation and risk assessment, is part of the holistic concept of risk management. Within this study, LRA is considered only, focussing on the risks to life. To calculate landslide risk, the spatial and temporal probability of occurrence of potential damaging events, as well as the distribution of the elements at risk in space and time, considering also changing vulnerabilities, must be determined. Within this study, a new raster-based approach is developed. Thus, all existent vector data are transformed into raster data. The specifc attribute data are attributed to the grid cells, resulting in specifc raster data layers for each input parameter. The calculation of the LRA follows a function of the input parameters hazard, damage potential of the elements at risk and their vulnerability. Within the quantitative LRA the associated uncertainties are estimated semi-quantitatively. Ratings of different spatial factors from the best models calculated with the AHP method were derived. The results showed that the geology (17 % variance), geodynamic features (17 % variance), slope conditions (30 % variance), hydrology/hydrogeology (17 % variance), types of landuse (6 % variance), and engineering characteristics of soils (10 % variance), and rocks (10 % variance) play an important role in landslide susceptibility in general. In terms of LRA, the result indicates that 14% of the area is in Very Low Risk zone, 10% in Low Risk zone, 52% in Medium Risk zone, 22% in High Risk zone and 2 % in Very High Risk zone. In the study area the highest risks throughout all of the analyses (individual risk to life and object risk to life) are caused by landslide, showing that risk heavily varies depending on the process considered. The resultant maps show areas, in which the individual risk to life exceeds the acceptable risk (defned in the aforementioned regulation), so that for these locations risk reduction measures should be developed and implemented. Accuracy assessment of landslide risk map was performed using correlation coeffcient. The correlation coeffcient result is equal 0.88. It can be concluded that the newly developed method works satisfactory and is applicable to further development in Kota Kinabalu, Sabah, and potentially to expand with different environmental settings.