Rapid fire spread occurred unexpectedly near Aberfeldy in eastern Victoria on the evening of 17 January 2013, threatening to impact the catchment of Lake Thomson, Melbourne’s largest water supply. The fire is believed to have begun around 11:30 am, in the first 16 hours it spread 30 km to the southeast and burned 20,000 ha. It was then impacted by a south-westerly wind change and burned a further 25,000 ha in only eight hours. This investigation aims to identify meteorological factors that may have contributed to the fire prediction modelling under-predicting rapid spread, to document situations in which fire prediction models might be vulnerable to unusually large errors. 

Fire agencies will benefit from working closely with the Bureau of Meteorology in their efforts to predict bushfire behaviour. This case study shows how post-incident meteorological and fire behaviour analysis can improve our ability to predict and respond to bushfires. By working together, our collective predictive capability will be improved, which will result in better messaging and warning by fire authorities. The investigation will explore high-resolution numerical weather prediction simulations, using the Australian Community Climate and Earth-System Simulator (ACCESS) and observational data such as radar imagery. The use of radar imagery to assist in analysing fire behaviour, and in particular radar interpretation of smoke plumes, is an area of study for further development.