This framework is a reference for individuals and agencies involved in bushfire risk assessment in Australia who seek to improve information on bushfire risk from quantitative methods compared to qualitative methods. It is aimed at bushfire researchers and risk managers in fire, planning and related agencies. Computational bushfire risk assessment is in an early stage of development in Australia. It is an opportune time to establish a framework sufficiently broad that it will accommodate pre-existing and new methods to assess bushfire risk while encouraging innovation. Current methods for assessing bushfire risk in Australia use different terminologies and approaches, and application of an overarching framework improves the potential to compare methods and confidence in comparing results between studies.

Extreme events in a changing climate A climate event is 'extreme' when it (or a series of events) occurs with greater intensity, frequency or duration than is normally expected. Every region of the world experiences extreme events from time to time and natural climate variability already produces extreme events in Tasmania. This includes heat waves, cold waves, floods, droughts and storms. Extreme events can have devastating and wide ranging effects on society and the environment, impacting infrastructure, agriculture, utilities, water resources and emergency planning.

This presentation will provide an overview of some of the work currently being undertaken at Geoscience Australia GA) as part of the National Coastal Vulnerability Assessment (NCVA), funded by the Department of Climate Change (DCC). The presentation will summarise the methodology applied, and highlight the issues, including the limitations and data gaps.

The Climate Futures for Tasmania (CFT) research project is the Tasmanian Government's most important source of climate change data at a local scale. The project has created fine-scale (14 kilometre) climate information for Tasmania by downscaling five global climate models (GCM-s) with two IPCC emission scenarios (A2 and B1) to generate climate information from 1961 to 2100. This new dataset is being used to interpret the impact of the changing climate on four main disciplines: General Climate, Water and Catchments, Extreme Events and Agriculture. As part of the extreme events component, Geoscience Australia is conducting severe wind hazard and risk studies in the Tasmanian region under both current and future climate conditions. In this paper we present severe wind hazard maps for Tasmania for current and future climate. The CFT fine scale climate simulations which provide high-resolution spatial detail of the wind speed (hourly maximum time-step mean wind speed used) were used. The methodology is described in an accompanying paper ('Dynamical downscaling of severe wind hazard: Methodology', in these proceedings).

Geoscience Australia has developed a model to assess severe wind hazard for large-scale numerical model-derived grided data. The severe wind modelling approach integrates two models developed at Geoscience Australia: a) A statistical model based on observations which determines return periods (RP) of severe winds using Extreme Value distributions (EVD), and b) A model which extracts mean wind speeds from high resolution numerical models (climate simulations) and generates wind gust from the mean speeds using Monte Carlo simulation (convolution with empirical gust factors) This methodology is particularly suitable for the study of wind hazard over large regions, and is being developed to provide improved spatial information for the Australia/NZ Wind Loading Standard (AS/NZS 1170.2, 2002). The methodology also allows comparison of current and future wind hazard under changing climate conditions. To illustrate the characteristics and capabilities of the methodology, the determination of severe wind hazard for a high-resolution grid encompassing the state of Tasmania (south of the Australian continent) will be presented and discussed, considering both the current and a range of possible future climate conditions (utilising IPCC B1 & A2 emission scenarios).

Overhead transmission lines are a key element of the electrical power system for transferring bulk power from generators to communities. Lattice type transmission towers carrying conductors form the physical backbone of the power transmission system. Transmission tower safety and reliability assessment is necessary to plan for minimisation of the risk of disruption of power supply resulting from in-service tower failure. Lattice type transmission towers are constructed using angle section members and are eccentrically connected. They are regarded as one of the most difficult forms of lattice structures to analyse for dynamic loads. Analysis is difficult due to fabrication errors, inadequate joint details and material properties being hard to quantify as a combination. Proof loading and full-scale tower testing is a traditional form of design validation for lattice type towers [1]. However, loading conditions experienced in severe wind events are dynamic and relatively short term loads and this behavior is confirmed in a limited way through full scale measurements of aero-elastic models in wind tunnels [2].

Stochastic finite-fault ground-motion prediction equations (GMPEs) are developed for the stable continental region of southeastern Australia (SEA). The models are applicable for horizontal-component ground motions for earthquakes 4.0 <= MW <= 7.5 and distances less than 400 km. The models are calibrated with updated source and attenuation parameters derived from SEA ground-motion data. Careful analysis of well-constrained earthquake stress parameters indicates a dependence on hypocentral depth. It is speculated that this is the effect of an increasing crustal stress profile with depth. However, rather than a continuous increase, the change in stress parameter appears to indicate a discrete step near 10 km depth. Average stress parameters for SEA earthquakes shallower and deeper than 10 km are estimated to be 23 MPa and 50 MPa, respectively. These stress parameters are consequently input into the stochastic ground-motion simulations for the development of two discrete GMPEs for shallow and deep events. The GMPEs developed estimate response spectral accelerations comparable to the Atkinson and Boore (BSSA, 2006) GMPE for eastern North America (ENA) at short rupture distances (less than approximately 100 km). However, owing to higher attenuation observed in the SEA crust (Allen and Atkinson, BSSA, 2007), the SEA GMPEs estimate lower ground-motions than ENA models at larger distances. The response spectral models are validated against moderate-magnitude 4.0 <= MW <= 5.3 earthquakes from eastern Australia. Overall the SEA GMPEs show low median residuals across the full range of period and distance. In contrast, Eastern North American models tend to overestimate response spectra at larger distances. Because of these differences, the present analysis justifies the need to develop Australian-specific GMPEs where ground-motion hazard from a distant seismic source may become important.

The Australian Bureau of Meteorology (BoM) have been recording peak gust wind speed observations in the Australian region for over 70 years. The current wind loading code and the performance of our infrastructure is based primarily on the Dines anemometer interpretation of the peak gust wind speed. Australian building codes through the Australia/New Zealand Wind Actions Standard [1] as well as the wind engineering community in general rely to a significant extent on these peak gust wind speed observations. In the mid-1980's the Australian Bureau of Meteorology (BoM) commenced a program to replace the aging pressure tube Dines anemometer with the Synchrotac and Almos cup anemometers. Only six Dines anemometers remain in operation, mainly as backup or for high-speed measurement. During the anemometer replacement procedure, many localities had more than one type of anemometer operating, recording extreme events. The passage of Cyclone Vance through Exmouth in 1999 saw Dines and Almos anemometers, separated by 25 metres, recording peak gusts of 144 and 122 knots respectively [2]. A weak cyclone that passed through Townsville in April 2000 recorded a peak gust of 70 knots on the Dines and 59 knots on the Almos anemometer [3]. These systematic differences raise concerns about the consistency and utility of the peak gust wind speed database.

In Global ShakeMap (GSM) applications where access to real-time ground-motion data - which constrains the shaking - is often limited, we must rapidly estimate the shaking distribution in the earthquake source region using solely predictive techniques. Current ShakeMap practice is to first calculate instrumental ground motions using a Ground-Motion Prediction Equation (GMPE). These instrumental ground motions are subsequently converted to macroseismic intensities, which are employed to evaluate human exposure to potentially fatal levels of ground-shaking in PAGER (Prompt Assessment of Global Earthquakes for Response). Here, we use the combined dataset of global instrumental and macroseismic intensity ground motion data gathered for the Atlas of ShakeMaps (Allen et al., this meeting) for evaluating the GSM approach. Several commonly used GMPEs are evaluated for active tectonic crust, subduction zones, and stable continental regions. Using our preferred instrumental GMPE, we subsequently evaluate peak motion to intensity conversion equations. Finally, we evaluate several intensity prediction equations against the ShakeMap Atlas dataset. This review has led us to recommend several fundamental changes to current GSM practice, particularly in the prediction of active crustal ground motions. We also recommend that macroseismic intensities should be predicted using conversion equations that consider earthquake magnitude and distance to rupture, in addition to peak ground motions. Though not exhaustive, this review provides a comprehensive analysis of GMPEs and macroseismic intensity prediction techniques in different tectonic regimes against a large dataset of global ground motion data. The primary purpose of this study is to evaluate these techniques with a view of improving current practices in rapid ground motion prediction for the GSM and PAGER systems.

In addition to the devastating 1989 Newcastle earthquake, at least four other earthquakes of magnitude 5 or greater have occurred in the surrounding Hunter region since European settlement in 1804. Some of these earthquakes caused damage in areas that, at the time, were sparsely populated. Similar events, were they to occur today in populated areas, would certainly cause significant damage. The frequency with which these events have occurred in the Hunter region suggests that earthquakes pose a genuine threat to the communities there. This study presents the most comprehensive and advanced earthquake risk assessment undertaken for any Australian city to date. It has focused on the economic losses caused by damage to buildings from earthquake ground shaking, and not on the impacts from other, secondary hazards such as soil liquefaction and surface faulting. The study has adopted a probabilistic approach that makes allowances for the variability that is inherent in natural processes as well as the uncertainty in our knowledge. The results from this project will assist decision-makers involved in local and state government, policy development, the insurance industry, engineers, architects, and the building and finance industries to manage potential damage and loss of life from earthquakes in Newcastle and Lake Macquarie. The results also have implications for the earthquake risk facing larger Australian cities such as Sydney, Melbourne and Adelaide. This is due to a number of factors, including similarities between the earthquake hazard in Newcastle and Lake Macquarie and other parts of Australia, and similarities between the urban environments, particularly the composition of the building stock.