In 2020, bushfires burnt over 87,000 hectares of the World Heritage-listed island K’Gari (Fraser Island). In this conference paper, we examine the utility of ensemble modelling approaches to investigate probable fire pathways and risk using big data fire weather analysis and modern fire simulation technology across the island. Traditional bushfire-prone area mapping relies on static fire weather scenarios and simplified behaviour. This paper presents the potential utility of wildfire behaviour ensemble modelling for bushfire-prone area mapping, fire mitigation design, Bushfire Attack Level determination and risk assessment.
The method adopted includes an ensemble modelling approach aggregating fire behaviour from thousands of fire simulations under relevant fire weather originating from eight cardinal directions identified using big-data analysis techniques. We compare the wildfire model to the 2020 fire scar and traditional mapping methods. The aggregated fire behaviour ensemble provided a high spatial correlation with the 2020 burn scar. We then demonstrate the platform’s utility to test the efficacy of various mitigation strategies, including surgical burning near assets for their local protection vs. a broad-scale prescribed burning strategy to reduce fuels and risk.
