Return period cyclonic wind hazard in the Australian region

A review commissioned by the Council of Australian Governments in 2001, Natural Disasters in Australia: reforming mitigation, relief and recovery arrangements, (COAG, 2002) recommended a fundamental shift in focus in relation to natural disasters; beyond relief and recovery towards cost-effective, evidence-based disaster mitigation. The Disaster Mitigation Australia Package (DMAP), administered by the Department of Transport and Regional Services (DoTARS), implements many of the review's recommendations.

Geoscience Australia provides technical advice to DMAP and assists in its implementation. Part of this role involves the development of risk models to estimate the potential physical, economic and social impacts of earthquake, tropical cyclone, landslide, severe wind and tsunami.

The severe wind activity at Geoscience Australia is developing a tropical cyclone hazard model, one of the three components of their infrastructure risk model, in order to rigorously assess the risk around the Australian coastline. The other two models (not discussed) are the exposure and vulnerability models. Therefore, the first stage in assessing the risk of tropical cyclones is to determine the level of hazard. The hazard model is a statistical model of tropical cyclone behaviour, similar to those of Powell et al (2005) and Hall and Jewson (2007). This represents a fundamental difference in methodology to previous estimates of tropical cyclone wind hazard in Australia, outlined in the Australian/New Zealand wind loadings standard (AS/NZS 1170.2, 2002). This assessment of hazard was undertaken by only utilising the observed wind record at a discrete number of "well–exposed" such as airports; no modelled data was employed.

The model utilises distributions of tropical cyclone properties (speed, bearing, intensity) are developed on a grid covering the regions of interest. Tropical cyclones are randomly initiated by sampling from a 2-dimensional probability density function of tropical cyclone origins, and then the subsequent behaviour of the tropical cyclone is determined by sampling from the prepared distributions.

The hazard model is designed to permit as much flexibility to the user as possible. To this end, the wind fields applied to each synthetic track are parametric in nature, so users can test the sensitivity of the hazard levels to the choice of radial wind profile or asymmetry model, for example.

As a preliminary validation exercise of the model, the hazard output is compared to the output from an insurance industry statistical hazard model over northern Australia, and other existing estimates of the wind hazard. We compare the 0.2% AEP wind speed at selected locations and present a spatial estimate of this wind speed, incorporating the effects of shielding, terrain and topography where possible to arrive at a site specific hazard estimate. Some caveats on estimating the appropriate wind hazard across Australia are also highlighted, indicating the difficulties in correctly estimating the hazard levels in a changing climate.

AS/NZS 1170.2, 2002: Structural design actions, Part 2: Wind actions, Australian/New Zealand Standard

COAG; Council of Australian Governments, 2002: Natural disasters in Australia: Reforming mitigation, relief and recovery arrangements, Commonwealth Department of Transport and Regional Services, Canberra

Emanuel, K. A., S. Ravela, E. Vivant, and C. Risi, 2006: A Statistical Deterministic Approach to Hurricane Risk Assessment. Bulletin of the American Meteorological Society, 87 (3), 299-314.

Hall, T. M. and S. Jewson, 2007: Statistical modelling of North Atlantic tropical cyclone tracks. Tellus A, 59 (4), doi:10.1111/j.1600-0870.2007.00240.x, 486-498.

Powell, M., G. Soukup, S. Cocke, S. Gulati, N. Morisseau-Leroy, S. Hamid, N. Dorst, and L. Axe, 2005: State of Florida hurricane loss projection model: Atmospheric science component. Journal of Wind Engineering and Industrial Aerodynamics, 93 (8), 651-674.