Property protection from extreme bushfire events under the influence of climate change

Abstract

Natural disasters give rise to loss of life, property (including homes, industry and livelihood) and environmental values and may be increasing with the impacts of climate change. Bushfires are a natural part of the Australian landscape and the ecology of the range of biota found within the various landscapes. They pose significant risks to people and property and require increasing demands for management in the face of these risks. Bushfires (also known as wildland fires) can be highly complex both spatially and temporally within the landscape. Attempts to better explain such events has given rise to a range of fire behaviour models to quantify fire characteristics such as rate of spread, fire line intensity, flame heights and spotting distances. However, there is a need to develop clear criteria when applying these models in land use planning and construction practice for bushfire protection. In Australia, a number of empirical models have been developed to quantify bushfire behaviour. These models have limitations, both in their application and in their capacity to draw upon data with which to utilise them. Two such models are used in the current study, being the McArthur Forest Fire Danger Meter (Mark 5) and the more recent Dry Eucalypt Forest Fire Model, and both have been used to develop design bushfire(dimensions and characteristics of a bushfire in a regional setting) conditions for the state of New South Wales (NSW). These models use different input parameters, as well as different intermediate parameters to describe fire behaviour. In addition, the study utilises and extends the forest fire danger index (FFDI) andKeetch-Byram Drought Index (KBDI) data to all 21 NSW fire weather districts. It also provides a new database for daily fuel moisture content (FMC). By using case studies that show 'validation' of methodological approaches, it can be confirmed that suitable extreme value assessment statistical techniques can be applied to the outputs of the identified models for the purposes of determining design bushfires. The study also seeks to give greater understanding of the frequency and shifts in the seasonality of fire weather, and changes in bushfire severity as consequences of climate change. A technique of generalised extreme value analysis based on moving data window to detect the impact of climate change on recurrence values of various indices has been developed. The evaluation of trends in fire weather through various metrics for FFDI, FMC and KBDI have revealed that a number of districts in NSW exhibit pronounced shifts at the extreme arising from climate change. However, the role of the El Nino Southern Oscillation does not appear to play a major role in these shifts over the long term. The current investigations have provided significant improvements on previous investigations such as improved datasets providing wider representation of all the NSW fire weather districts and covering a longer period of time; the use of new metrics, including the use of the GEV assessment through a moving period approach; the metrics being applied to fire weather parameters other than FFDI; and, trends in fire weather parameters being considered in conjunction with other global factors. The methodology and the technique developed in the current study have the potential to be utilised in many parts of the world for the development of design conditions and to study the impact of the climate change on the local fire weather conditions.

Date of Award	2017
Original language	English