Seismic vulnerability functions for Australian buildings by using GEM empirical vulnerability assessment guidelines

Australia has a low to moderate seismicity by world standards. However, the seismic risk is significant due to the legacy of older buildings constructed prior to the national implementation of an earthquake building standard in Australia. The 1989 Newcastle and the 2010 Kalgoorlie earthquakes are the most recent Australian earthquakes to cause significant damage to unreinforced masonry (URM) and light timber frame structures and have provided the best opportunities to examine the earthquake vulnerability of these building types. This paper describes the two above-mentioned building types with a differentiation of older legacy buildings constructed prior to 1945 to the relatively newer ones constructed after 1945. Furthermore, the paper presents method to utilise the large damage and loss-related data (14,000 insurance claims in Newcastle and 400 surveyed buildings in Kalgoorlie) collected from these events to develop empirical vulnerability functions. The method adopted here followed the GEM empirical vulnerability assessment guidelines which involve preparing a loss database, selecting an appropriate intensity measure, selecting and applying a suitable statistical approach to develop vulnerability functions and the identification of optimum functions. The adopted method uses a rigorous statistical approach to quantify uncertainty in vulnerability functions and provides an optimum solution based on goodness-of-fit tests. The analysis shows that the URM structures built before 1945 are the most vulnerable to earthquake with post-1945 URM structures being the next most vulnerable. Timber structures appear to be the least vulnerable, with little difference observed in the vulnerability of timber buildings built before or after 1945. Moreover, the older structures (both URM and timber) exhibit more scatter in results reflecting greater variation in building vulnerability and performance during earthquakes. The analysis also highlights the importance of collecting high-quality damage and loss data which is not only a fundamental requirement for developing empirical vulnerability functions, but is also useful in validating analytically derived vulnerability functions. The vulnerability functions developed herein are the first publically available functions for Australian URM and timber structures. They can be used for seismic risk assessment and to focus the development of retrofit strategies to reduce the existing earthquake risk.