A Spatially Explicit Approach for Sensitivity and Uncertainty Analysis of GIS-Multicriteria Landslide Susceptibility Mapping

GIS multicriteria decision analysis (MCDA) methods are increasingly being used in landslide susceptibility mapping for the prediction of future hazards, decision making, as well as hazard mitigation plans. However, the uncertainties associated with MCDA techniques are inevitable and model outcomes are prone to multiple types of uncertainty. In this paper, the spatiality explicitly method is employed to assess the uncertainty associated with two methods of GIS-MCDA namely, Analytical Hierarchical Process (AHP) and Ordered Weighted Averaging (OWA). Our methodology consists of three stages. The first-stage analysis makes use of the AHP matrix weights for GIS-MCDA based landslide susceptibility mapping. This phase is based on a multicriteria evaluation which assesses the susceptibility areas and the landslide hazard potentiality by considering causal and diagnostic criteria. In the second stage the sensitivity and uncertainty analysis of the AHP weights is performed using Monte Carlo Simulation (MCS) and Global Sensitive Analysis (GSA). Finally, validation of results was performed using the existing landslide inventory. This paper carried out a GIS-MCDA uncertainty analysis and demonstrates a solution for the uncertainty modelling. Through the validation exercise with known existing landslides, the AHP clearly performed best. Results of this research demonstrate that further improvement of the accuracy of GIS-based MCDA can be achieved by employing the spatiality explicitly method and accordingly applying MCS and GSA for sensitivity analysis of the AHP weights.