PHOENIX RapidFire

The PHOENIX RapidFire fire simulator places Australia as a world-leader in wildfire tools and analysis. It is used by land and fire managers to support fire management and land-use planning and to support decision-making during bushfires.

The major benefit of PHOENIX RapidFire is that the speed, accuracy and user-friendliness of its information allows quick, informed decisions to be made that save lives and properties. When combined with clear and timely warning messages, communities at risk can are able to informed actions while under the threat of a major fire.

By using PHOENIX RapidFire, fire and land management agencies can reduce natural hazards impacts on property and infrastructure both by ensuring appropriate measures are taken before the fire season begins and making more informed decisions about what areas are at risk when a fire has taken hold.

PHOENIX RapidFire is a model that simulates bushfires. It integrates fuel, terrain, weather conditions and suppression to simulate a fire’s development and progression in the landscape.

The model is mechanistic, continuous, dynamic and empirically based. It simulates fire characteristics such as fire spread, flame height, intensity, size and ember density. It can also simulate some of the effects of suppression efforts and the impact of fire on various values and assets.

At a minimum, PHOENIX RapidFire requires fuel data as an input. Inputs including terrain, weather, suppression, fire history and assets are required for realistic simulations. These inputs and outputs are described in the figure below.

PHOENIX RapidFire model: inputs and outputs

THE PHOENIX MODEL: INPUTS AND OUTPUTS

The development of PHOENIX RapidFire was driven by the need to realistically describe bushfires across the Australian landscape.

In the late 1990s researchers at the University of Melbourne began to review the elements contributing to bushfire risk and the current state of knowledge. With the establishment of the Bushfire Cooperative Research Centre in 2004, funding was made available to continue this research. The first stage of this work was defining the Fire Management Business Model, which showed how 54 factors (or elements) of bushfire risk management interacted to reduce bushfire risk for a given level of resources allocated to each element.

Having established this model, it was necessary to characterise and quantify the effect of different bushfire management strategies on reducing the level of bushfire risk. The researchers concluded that the best way to characterise fires across the landscape was to use a fire simulator. This would allow analysis to be spatially and temporally explicit and would also be objective and repeatable. Two international fire simulators were considered, but they were not easily adaptable to Australian conditions.

In the absence of an existing bushfire simulator, in 2005 Professor Kevin Tolhurst and programmer Derek Chong at the University of Melbourne were funded by the Bushfire CRC to build and merge two models: PHOENIX, which described what a bushfire is like at any point in the landscape, and RapidFire, which analysed how a fire interacts with important assets, such as houses, powerlines, catchments and biodiversity.

Initially, PHOENIX RapidFire could only be used as a fire characterisation simulator. Once that was adequately established, the researchers added more functionality to assess the relative level of bushfire risk. This allowed the model to be used for planning of prescribed burns, rather than just response to wildfires. 

The use of PHOENIX RapidFire has resulted in a vast number of benefits for Australian communities, with many examples drawn from recent bushfires. Queensland emergency services have credited PHOENIX RapidFire for enabling a quick responses to rapidly evolving fire threats that occurred across the entire state during the spring and summer of 2018.

On 28 November 2018, a small fire started outside Gracemere, Queensland, a town of 8000 people located just south of Rockhampton. Conditions that day were unlike any seen in the region since the 1960s, with an intense dry heat. The fire danger rating index reached Catastrophic: the first time such a level had been reached in Queensland.

Queensland Fire and Emergency Services’ Inspector Andrew Sturgess said at the time that the PHOENIX RapidFire modelling allowed emergency responders to realise that the small fire west of Gracemere had the potential to turn into a blaze that could threaten the town.

The swift evacuation of the community with clear, direct warning messages has been hailed as an example of success of both the simulator and the CRC research into timely and effective warnings.

Inspector Sturgess noted the fire was “predicted to impact the township of Gracemere, and that's what happened”. He also said the software predicted the path and intensity of the fire which then allowed appropriate directions to fire crews on the ground.

QFES Deputy Commissioner Doug Smith, who led the emergency response during the 2018 fires and is a current board member of the CRC, noted that PHOENIX RapidFire had recognised the potential of the small fire to grow beyond expectations. “The heatwave that came was unusual. We didn’t have experience with it. PHOENIX RapidFire made it possible for us to apply the learnings from Victorian bushfires that often occur in these conditions and apply that to a new situation unfolding in Queensland. Firefighters then fought the fire in a way that was very unfamiliar to Queenslanders. That is the true value of the model”. 

At the end of the Bushfire CRC in 2013, the IP for PHOENIX RapidFire was transferred to three joint parties: the University of Melbourne; the Victorian Department of Environment, Land, Water and Planning; and AFAC. The model’s use is now licensed out commercially by Fire Prediction Services, which is owned by the three IP owners.

Phoenix RapidFire is an example of how investment and research made in the days of the Bushfire CRC is continuing to benefit the Australian community. The Bushfire and Natural Hazards CRC has developed updates and conducted research for inputs into the model, increasing its accuracy across different scenarios and landscapes. Current research projects that are informing the model include:

▪ fire coalescence research

▪ wind reduction in tree canopies

▪ research into soil moisture

▪ fire event spotting.