The Eurasia Proceedings of Science Technology Engineering and Mathematics

Abstract

Landslide is one of the natural hazards that considers a serious threat to both humans’ lives and properties. Tasmania, Australia is one of those regions where landslides caused considerable damage to people and the State. Landslide damages can be reduced, even stopped, if proper land condition and planning assessment has been done. The main influencing parameters in landslide occurrences are topography, precipitation, and geological formation. Those parameters along with other influencing parameters have been used in landslide susceptibility mapping. In order to have a reliable analysis, a robust method of Decision Tree (DT) has been used to perform susceptibility mapping. According to the hierarchy structure of DT, geology and slope have been selected as the most influential parameters in landslide susceptibility. In order to evaluate the reliability of the outcomes, Area Under the Curve (AUC) has been utilized. Success and prediction rates were 87.64% and 84% respectively. Subsequently, risky features such as buildings, schools, hotels, etc. have been used in overlay analysis in a GIS environment with “very high” and “high” susceptibility classes. The outcome of this research can assist planning parties to secure vulnerable regions and consider those areas in their future decision-making strategies.