Engineers Australia

Landslide inventory, susceptibility, frequency and hazard zoning in the Wollongong and wider Sydney Basin Area

Phil Flentje, David Stirling and Robin Chowdhury


The University of Wollongong Landslide Research Team has been working on the development of GIS-based Landslide Inventory, Susceptibility and Hazard Zoning projects for over 15 years. To undertake the zoning work we use knowledge-based methods including Data Mining techniques which are facilitated within a GIS framework. This work is ongoing, and as with this paper, there are two main aims; firstly for those smaller subregions of Sydney where considerable data have been obtained and the landslide inventory development is comprehensive, increasingly more reliable modelling, analysis and synthesis is being done, and secondly, for the entire 31,000km2 geological extent of the Sydney Basin region where the available data are relatively small scale and the process of developing the landslide inventory is in the early stages, preliminary studies which are described as ‘proof of concept’ have been completed and are reported herein. The most advanced sub-region is a large portion of the Illawarra Escarpment within the Wollongong Local Government Area (LGA). Another advanced sub-region is the Picton area within the Wollondilly LGA. All the while, input data is being refined and improved in particular with the advent of Airborne Laser Scan derived DEM’s and the ongoing development and populating of Landslide Inventories. In tandem with refined input data, computing capabilities are also rapidly evolving and this is enabling ever growing terrain modelling capacity. With higher resolution input data for the Sydney Basin project, including a more rigorous Landslide Inventory which is already well under development, higher resolution geology information and possibly even a better or more recent DEM, the regional yet large scale GIS-based Susceptibility modelling outcomes are likely to be suitable for use at Local Government Planning levels. Furthermore, susceptibility modelling at a national scale to identify preliminary or ‘first pass’ binary type (i.e. in/out) areas for further assessment is also achievable in the very near future.