@article {bnh-8192, title = {Risk mitigation from prescribed burning in Kangaroo Island and Mount Lofty Ranges - Black Summer final report}, number = {690}, year = {2021}, month = {08/2021}, institution = {Bushfire and Natural Hazards CRC}, address = {MELBOURNE}, abstract = {

According to the Independent Review into South Australia{\textquoteright}s 2019-20 Bushfire Season, conditions were the worst on record with fires resulting in the loss of three human lives, 196 homes, 660 vehicles, 68,000 livestock, $200m of agricultural production. Around 280,000 ha were burnt by the fires, including total or partial burning of several National Parks.

The Bushfire and Natural Hazards CRC commissioned this project as part of a larger set of Black Summer fires research projects aimed at understanding the record-breaking fire season. This project focuses on answering questions about the effectiveness of prescribed burning, also known as hazard reduction burning, in mitigating risk in two areas affected by fires during the season: The Mount Lofty Ranges east of Adelaide, and Kangaroo Island.

The key questions were:

  1. How does risk respond to treatment in Kangaroo Island, an area with little formal quantification of prescribed burning benefits and costs?
  2. What was the risk in the leadup to the 2019-20 fire season in the Mt Lofty Ranges, and how will risk change in the next five years as a result of the implied fuel reduction from the fires, as well as alternative prescribed burning strategies?

These questions were answered using a well developed methodology combining large scale fire behaviour simulations and Bayesian risk quantification. Similar analyses have been carried out for a range of case study landscapes in southern Australia as part of the Hectares to tailor-made solutions CRC project, with results available online via the end-user tool the Prescribed Burning Atlas, and also the NSW Bushfire Risk Management Research Hub{\textquoteright}s projects for the NSW Bushfire Inquiry.

We found a clear relationship between the rate of prescribed burning and area subsequently burnt by wildfire in the Kangaroo Island case study. This translated into reductions in loss of life and property as well. Risk mitigation was more sensitive to edge treatment than landscape treatment, although both reduced risk. Conversely, increasing treatment (particularly at the edge) resulted in higher areas of the landscape exposed to vegetation being burnt below its minimum tolerable fire interval.

In the Mt Lofty Ranges, we found complex patterns of risk are likely in the aftermath of the 2019-20 fires. In the absence of further wildfire events, risk of area burnt is likely to rise substantially by 2025, regardless of prescribed burning rates, with a similar result for vegetation exposed to too frequent fire. However, risk sto life, property and infrastructure are projected to remain similar to current levels.

Our work contributes to the evidence base for prescribed burning planning in South Australia, with future work potentially examining new management values (e.g. smoke health costs, new biodiversity measures) and exploring empirical relationships between prescribed burning and fire-affected area in 2019-20.

}, keywords = {black summer, kangaroo island, mitigation, Mount Lofty Ranges, Prescribed burning, risk}, issn = {690}, author = {Hamish Clarke and Brett Cirulis and Owen Price and Ross Bradstock and Matthias M. Boer and Anthony Rawlins and Trent Penman} } @article {bnh-7343, title = {Effect of weather forecast errors on fire growth model projections}, journal = {International Journal of Wildland Fire}, year = {2020}, month = {08/2020}, abstract = {

Fire management agencies use fire behaviour simulation tools to predict the potential spread of a fire in both risk planning and operationally during wildfires. These models are generally based on underlying empirical or quasi-empirical relations and rarely are uncertainties considered. Little attention has been given to the quality of the input data used during operational fire predictions. We examined the extent to which error in weather forecasts can affect fire simulation results. The study was conducted using data representing the State of Victoria in south-eastern Australia, including grassland and forest conditions. Two fire simulator software packages were used to compare fire growth under observed and forecast weather. We found that error in the weather forecast data significantly altered the predicted size and location of fires. Large errors in wind speed and temperature resulted in an overprediction of fire size, whereas large errors in wind direction resulted in an increased spatial error in the fire{\textquoteright}s location. As the fire weather intensified, fire predictions using forecast weather under predicted fire size, potentially resulting in greater risks to the community. These results highlight the importance of on-ground intelligence during wildfires and the use of ensembles to improve operational fire predictions.

}, keywords = {Bayesian Network, fire prediction, meteorological forecast, sensitivity, simulation}, doi = {https://doi.org/10.1071/WF19199}, url = {https://www.publish.csiro.au/wf/wf19199}, author = {Trent Penman and Dan Ababei and Jane Cawson and Brett Cirulis and Thomas Duff and Swedosh, W and James Hilton} } @article {bnh-7389, title = {From hectares to tailor-made solutions for risk mitigation {\textendash} final project report}, number = {615}, year = {2020}, month = {09/2020}, institution = {Bushfire and Natural Hazards CRC}, address = {Melbourne}, abstract = {

We are pleased to present the 2020 Final Report for the Bushfire and Natural Hazards CRC project, {\textquotedblleft}From hectares to tailor-made solutions for risk mitigation: systems to deliver effective prescribed burning across Australian ecosystems{\textquotedblright}.

Prescribed burning is a central feature of contemporary fire management, not just in Australia but in fire-prone countries around the world. Yet we lack a firm quantitative basis for understanding and comparing its effectiveness at mitigating risk across different regions. This project aims to address these gaps and provide critical support to agency decision makers across southern Australia by undertaking a systematic investigation of the drivers of prescribed burning effectiveness across the region. This project will thus support fire managers in transitioning from hectare targets to a set of tailor-made, risk-based approaches.

The project is divided into two phases: fire behaviour accounting and risk accounting. At the heart of the project is predictive modelling of the effect of prescribed burning on subsequent bushfire (wildfire) behaviour. In this project we combine ignition likelihood modelling, fuel type and arrangement from fire management agencies, weather representing all possible local fire weather conditions from Bureau of Meteorology and fire history including wildfire and variable combinations of edge and landscape treatments, applied to agency or model-derived burn blocks. From thousands of simulations, key outputs such as fire size and intensity are used estimate impacts on key management values: house loss, life loss, length of road damaged, length of powerline damaged and area burnt below minimum tolerable fire interval. We use Bayesian decision networks to estimate risk mitigation, including cost, available through different treatments. Local trajectories of cost for given treatment rates and locations can then be tracked and compared between regions, allowing identification of the most cost-effective prescribed burning strategies, either overall or for a given management value.

The key finding of the project is that the effectiveness of prescribed burning at mitigating area burnt by bushfire and other key values varies considerably across landscapes and values. That is, there is no one-size-fits-all solution to prescribed burning. This has major implications for fire managers, suggesting that tailored prescribed burning solutions are possible, based on the unique risk mitigation profile for any given suite of management values in that region. Further details are included in this report and journal articles listed herein.

While the project now draws to a close, its findings live on in the Prescribed Burning Atlas, a dedicated website for fire managers, researchers and anyone else interested in using our project to support their planning, decision making and communication. The Prescribed Burning Atlas will provide a geographically-based summary of risk for decision makers in an accessible, user friendly format. Our project is unique in placing the design and delivery of this utilisation output at its heart. Active involvement of end-users throughout 2020 and beyond will be crucial in ensuring uptake and translation into outcomes for end-users and the communities they serve. From a research perspective, key outstanding questions include quantification of risk mitigation for new values (smoke effects on human health, biodiversity measures, cultural burning) and expansion of the approach to represent the full richness of planned and unplanned landscape fire over time.

}, keywords = {ecosystems, hectares, mitigation, Prescribed burning, risk, systems, tailor-made solutions}, issn = {615}, author = {Hamish Clarke and Brett Cirulis and Trent Penman and Owen Price and Matthias M. Boer and Ross Bradstock} } @article {bnh-6854, title = {Cost effectiveness of fire management strategies in southern Australia}, journal = {International Journal of Wildland Fire}, volume = {29}, year = {2019}, month = {06/2019}, pages = {427-439}, abstract = {

Fire-management agencies invest significant resources to reduce the impacts of future fires. There has been increasing public scrutiny over how agencies allocate fire-management budgets and, in response, agencies are looking to use quantitative risk-based approaches to make decisions about expenditure in a more transparent manner. Advances in fire-simulation software and computing capacity of fire-agency staff have meant that fire simulators have been increasingly used for quantitative fire-risk analysis. Here we analyse the cost trade-offs of future fire management in the Australian Capital Territory (ACT) and surrounding areas by combining fire simulation with Bayesian Decision Networks. We compare potential future-management approaches considering prescribed burning, suppression and fire exclusion. These data combined costs of treatment and impacts on assets to undertake a quantitative risk analysis. The proposed approach for fuel treatment in ACT and New South Wales (NSW) provided the greatest reduction in risk and the most cost-effective approach to managing fuels in this landscape. Past management decisions have reduced risk in the landscape and the legacy of these treatments will last for at least 3 years. However, an absence of burning will result in an increased risk from fire in this landscape.

}, keywords = {Bayesian Network, house loss, life loss, prescribed fire, risk}, doi = { https://doi.org/10.1071/WF18128}, url = {https://www.publish.csiro.au/WF/WF18128}, author = {Trent Penman and Brett Cirulis} } @article {bnh-5648, title = {Developing and testing models of the drivers of anthropogenic and lightning-caused wildfire ignitions in south-eastern Australia}, journal = {Journal of Environmental Management}, volume = {235}, year = {2019}, month = {01/2019}, chapter = {34-41}, abstract = {

Considerable investments are made in managing fire risk to human assets, including a growing use of fire behaviour simulation tools to allocate expenditure. Understanding fire risk requires estimation of the likelihood of ignition, spread of the fire and impact on assets. The ability to estimate and predict risk requires both the development of ignition likelihood models and the evaluation of these models in novel environments. We developed models for natural and anthropogenic ignitions in the south-eastern Australian state of Victoria incorporating variables relating to fire weather, terrain and the built environment. Fire weather conditions had a consistently positive effect on the likelihood of ignition, although they contributed much more to lightning (57\%) and power transmission (55\%) ignitions than the 7 other modelled causes (8{\textendash}32\%). The built environment played an important role in driving anthropogenic ignitions. Housing density was the most important variable in most models and proximity to roads had a consistently positive effect. In contrast, the best model for lightning ignitions included a positive relationship with primary productivity, as represented by annual rainfall. These patterns are broadly consistent with previous ignition modelling studies. The models developed for Victoria were tested in the neighbouring fire prone states of South Australia and Tasmania. The anthropogenic ignition model performed well in South Australia (AUC = 0.969) and Tasmania (AUC = 0.848), whereas the natural ignition model only performed well in South Australia (AUC = 0.972; Tasmania AUC = 0.612). Model performance may have been impaired by much lower lightning ignition rates in South Australia and Tasmania than in Victoria. This study shows that the spatial likelihood of ignition can be reliably predicted based on readily available meteorological and biophysical data. Furthermore, the strong performance of anthropogenic and natural ignition models in novel environments suggests there are some universal drivers of ignition likelihood across south-eastern Australia.

}, keywords = {Bushfire, ignition, Risk modelling, Wildfire}, doi = {https://doi.org/10.1016/j.jenvman.2019.01.055}, url = {https://www.sciencedirect.com/science/article/pii/S0301479719300568}, author = {Hamish Clarke and Rebecca Gibson and Brett Cirulis and Ross Bradstock and Trent Penman} } @article {bnh-5697, title = {From hectares to tailor-made solutions for risk mitigation: annual project report 2018-19}, number = {487}, year = {2019}, month = {07/2019}, institution = {Bushfire and Natural Hazards CRC}, address = {Melbourne}, abstract = {

We are pleased to present the 2018-2019 Annual Report for the Bushfire and Natural Hazards CRC project, {\textquotedblleft}From hectares to tailor-made solutions for risk mitigation: systems to deliver effective prescribed burning across Australian ecosystems{\textquotedblright}. The project aims to provide critical support to agency decision makers across southern Australia by undertaking a systematic investigation of the drivers of prescribed burning effectiveness across the region. This report describes the background, research approach and key milestones since the previous Annual Report in 2017-2018. The report focuses on the research outputs informing the development of the Prescribed Fire Atlas. The project has now delivered a rich, layered dataset capable of addressing nuanced questions about the quantitative risk reduction available through prescribed burning for multiple management values in varying landscapes. The project is now entering its final phase as we complete climate change and cost-effectiveness analyses and launch the Prescribed Burning Atlas in 2019/20. The Prescribed Burning Atlas will provide geographically based summary of risk for decision makers in an accessible, user friendly format. Our project is unique in placing the design and delivery of this utilisation output at the heart of the project. Active involvement of end users throughout 2020 and beyond will be crucial in ensuring uptake and translation into outcomes for end users and the communities they serve.

}, keywords = {cost-effective, ecosystems, fire management, Prescribed burning}, author = {Hamish Clarke and Owen Price and Matthias M. Boer and Brett Cirulis and Trent Penman and Ross Bradstock} } @conference {bnh-6396, title = {A new decision support tool for prescribed burning risk assessment}, booktitle = {Bushfire and Natural Hazards CRC Research Day AFAC19}, year = {2019}, month = {12/2019}, address = {Melbourne}, abstract = {

In most Australian jurisdictions, the use of prescribed fire is promoted on the basis of its efficacy in mitigation of risk. Despite this, formal attempts to evaluate effects on risk to people, property and environmental values across different jurisdictions are generally lacking. In particular, there is no basis for assessing the generality of attempts to predict risk in response to any particular strategy for use of prescribed fire (e.g. the 5per cent target recommended by the 2009 Victorian Bushfires Royal Commission). General principles therefore need to be developed about how to apply a risk-based approach across widely varying environments, human communities and combinations of key management values.

In this Bushfire and Natural Hazards Cooperative Research Centre project, researchers from the University of Wollongong, Western Sydney University and the University of Melbourne have come together with end users across southern Australia to design a project to systematically investigate how risk to any particular management value will respond to variations in the spatial location and rates of treatment. Project outputs are currently being moulded for utilisation by end users in a dedicated tool, the Prescribed Fire Atlas, which will guide the implementation of {\textquoteleft}tailor-made{\textquoteright} prescribed burning strategies to suit the biophysical, climatic and human context of all bioregions across southern Australia.

}, keywords = {decision support, Fire, perscibed burning, risk management}, url = {https://knowledge.aidr.org.au/resources/australian-journal-of-emergency-management-monograph-series/}, author = {Hamish Clarke and Brett Cirulis and Ross Bradstock and Matthias M. Boer and Trent Penman and Owen Price} } @article {bnh-5615, title = {Quantification of inter-regional differences in risk mitigation from prescribed burning across multiple management values}, journal = {International Journal of Wildland Fire}, year = {2019}, month = {06/2019}, abstract = {

Fire agencies are moving towards planning systems based on risk assessment; however, knowledge of the most effective way to quantify changes in risk to key values by application of prescribed fire is generally lacking. We present a quantification and inter-regional comparison of how risk to management values responds to variations in prescribed burning treatment rate. Fire simulations were run using the PHOENIX RapidFire fire behaviour simulator for two case study landscapes in interface zones in Tasmania and the Australian Capital Territory (ACT), Australia. A Bayesian network approach used these data to explore the influence of treatment and weather on risk from wildfire. Area burnt, length of powerline damaged and length of road damaged responded more strongly to treatment in the ACT than in Tasmania, whereas treatment mitigated house loss and life loss more strongly in Tasmania than the ACT. The effect of prescribed burning treatment rate on area burnt below minimum tolerable fire interval was similar in each case study landscape. Our study shows that the effectiveness of prescribed burning at mitigating area burnt by wildfire and other key values varies considerably across landscapes and values.

}, keywords = {Bushfire, Climate change, trends, Wildfire, wildland fire}, doi = {https://doi.org/10.1071/WF18135}, url = {http://www.publish.csiro.au/WF/WF18135}, author = {Brett Cirulis and Hamish Clarke and Matthias M. Boer and Trent Penman and Owen Price and Ross Bradstock} } @article {bnh-6856, title = {Quantification of inter-regional differences in risk mitigation from prescribed burning across multiple management values}, journal = {International Journal of Wildland Fire}, volume = {29}, year = {2019}, month = {09/2019}, pages = {414-426}, abstract = {

Fire agencies are moving towards planning systems based on risk assessment; however, knowledge of the most effective way to quantify changes in risk to key values by application of prescribed fire is generally lacking. We present a quantification and inter-regional comparison of how risk to management values responds to variations in prescribed burning treatment rate. Fire simulations were run using the PHOENIX RapidFire fire behaviour simulator for two case study landscapes in interface zones in Tasmania and the Australian Capital Territory (ACT), Australia. A Bayesian network approach used these data to explore the influence of treatment and weather on risk from wildfire. Area burnt, length of powerline damaged and length of road damaged responded more strongly to treatment in the ACT than in Tasmania, whereas treatment mitigated house loss and life loss more strongly in Tasmania than the ACT. The effect of prescribed burning treatment rate on area burnt below minimum tolerable fire interval was similar in each case study landscape. Our study shows that the effectiveness of prescribed burning at mitigating area burnt by wildfire and other key values varies considerably across landscapes and values.

}, keywords = {Bushfire, Climate change, trends, Wildfire, wildland fire}, doi = {https://doi.org/10.1071/WF18135}, url = {https://www.publish.csiro.au/WF/WF18135}, author = {Brett Cirulis and Hamish Clarke and Matthias M. Boer and Trent Penman and Owen Price and Ross Bradstock} } @conference {bnh-6549, title = {Quantifying dynamic fire behaviour phenomena using Unmanned Aerial Vehicle technology }, booktitle = {23rd International Congress on Modelling and Simulation}, year = {2019}, month = {12/2019}, address = {Canberra}, abstract = {

Catastrophic wildfires are often a result of dynamic fire behaviours. Although some of these behaviours have been described and investigated, others require further study. Fire coalescence and junction fires are particular cases of merging fire fronts and are common phenomena observed during bushfires. They are important as they can cause rapid escalation of fire behaviour and be dangerous for ground-based emergency personnel. There are a few studies devoted to investigation of merging fires in field conditions. There is a need for high temporal and spatial measurements of fire behaviour in field conditions in order to better understand this phenomenon and evaluate risks during bushfires and prescribed burning. The aim of this study was to test emerging technologies for better quantification of fire behaviour at field scales and evaluate their potential as an operational tool.

Several small and medium scale field experiments were conducted during April 2019 on farmland in Victoria, Australia. Harvested wheat fields were used as experimental plots, as they form relatively homogeneous fuel beds. Fuel height varied from 18 to 40 cm with fuel load and moisture content at 1.1 t/ha and 11.9 \% respectively. Wind speed varied in the range of 1.5-6.5 m/s. An Unmanned Aerial Vehicle (UAV) was used to capture high definition video imagery of fire propagation in synchronisation with sensor data from the on-board Global Positioning System (GPS) and Inertial Measurement Unit (IMU). These sensors enabled the platform/camera orientation and position in space to be aligned with the video footage of fire propagation and to be georeferenced in GIS software.

Twenty-one junction fires and five inward parallel fire fronts (parts of the fire lines approaching each other) were identified during the experiments. The rate of spread (ROS) of merging fire fronts was found to be at least two times higher than for the basic fire fronts (the rate of spread of a linear fire front in the same fuel bed in no-slope conditions) and for junction fires with acute angles (\< 14{\textdegree}) it increased 6 times and more. Inward parallel fire fronts spread much slower, varying between 0.05 and 0.25 m/s. Forty-six percent of junction fires had increase of the ROS at the final stage of the merging process in contrast to Thomas et al. (2017) and Viegas et al. (2012). Also, it was observed that the angle between two oblique fire fronts did not change significantly in time for the initial angles smaller than 34{\textdegree}. It can be assumed that the main fire front influences on the shape and ROS respectively of junction fires and laboratory experiments cannot fully replicate these conditions.

Although the initial experimental conditions were very different in relation to scale, fuel and wind conditions, similar ROS to that shown in numerical simulations by Thomas et al. (2017) were observed in our field experiments. Further investigation is required to explain the similarities as the relationship between fuel load, wind speed and scale is not known. The comparison of corrected values of dimensionless ROS for different angles between fire fronts with laboratory experiments of Viegas et al. (2012) showed reasonable quantitative agreement.

These experiments have shown that the method of using UAV{\textquoteright}s to capture georeferenced video footage can be used reliably to quantify fire behaviour phenomena for research, operation and management purposes.

Catastrophic wildfires are often a result of dynamic fire behaviours. Although some of these behaviours have been described and investigated, others require further study. Fire coalescence and junction fires are particular cases of merging fire fronts and are common phenomena observed during bushfires. They are important as they can cause rapid escalation of fire behaviour and be dangerous for ground-based emergency personnel. There are a few studies devoted to investigation of merging fires in field conditions. There is a need for high temporal and spatial measurements of fire behaviour in field conditions in order to better understand this phenomenon and evaluate risks during bushfires and prescribed burning. The aim of this study was to test emerging technologies for better quantification of fire behaviour at field scales and evaluate their potential as an operational tool.

Several small and medium scale field experiments were conducted during April 2019 on farmland in Victoria, Australia. Harvested wheat fields were used as experimental plots, as they form relatively homogeneous fuel beds. Fuel height varied from 18 to 40 cm with fuel load and moisture content at 1.1 t/ha and 11.9 \% respectively. Wind speed varied in the range of 1.5-6.5 m/s. An Unmanned Aerial Vehicle (UAV) was used to capture high definition video imagery of fire propagation in synchronisation with sensor data from the on-board Global Positioning System (GPS) and Inertial Measurement Unit (IMU). These sensors enabled the platform/camera orientation and position in space to be aligned with the video footage of fire propagation and to be georeferenced in GIS software.

Twenty-one junction fires and five inward parallel fire fronts (parts of the fire lines approaching each other) were identified during the experiments. The rate of spread (ROS) of merging fire fronts was found to be at least two times higher than for the basic fire fronts (the rate of spread of a linear fire front in the same fuel bed in no-slope conditions) and for junction fires with acute angles (\< 14{\textdegree}) it increased 6 times and more. Inward parallel fire fronts spread much slower, varying between 0.05 and 0.25 m/s. Forty-six percent of junction fires had increase of the ROS at the final stage of the merging process in contrast to Thomas et al. (2017) and Viegas et al. (2012). Also, it was observed that the angle between two oblique fire fronts did not change significantly in time for the initial angles smaller than 34{\textdegree}. It can be assumed that the main fire front influences on the shape and ROS respectively of junction fires and laboratory experiments cannot fully replicate these conditions.

Although the initial experimental conditions were very different in relation to scale, fuel and wind conditions, similar ROS to that shown in numerical simulations by Thomas et al. (2017) were observed in our field experiments. Further investigation is required to explain the similarities as the relationship between fuel load, wind speed and scale is not known. The comparison of corrected values of dimensionless ROS for different angles between fire fronts with laboratory experiments of Viegas et al. (2012) showed reasonable quantitative agreement.

These experiments have shown that the method of using UAV{\textquoteright}s to capture georeferenced video footage can be used reliably to quantify fire behaviour phenomena for research, operation and management purposes.

}, keywords = {fire front propagation, merging fire fronts, Remote measurements, UAS}, url = {https://www.researchgate.net/publication/338412609_Quantifying_dynamic_fire_behaviour_phenomena_using_Unmanned_Aerial_Vehicle_technology}, author = {Alex Filkov and Brett Cirulis and Trent Penman} } @article {bnh-5035, title = {From hectares to tailor-made solutions for risk mitigation: annual project report 2017-18}, year = {2018}, month = {10/2018}, institution = {Bushfire and Natural Hazards CRC}, abstract = {

This report describes the background, research approach and key milestones since the 2016-2017 Annual Report. The report focuses on the analysis of fire spread simulations for two case study landscapes: the ACT and Tasmania. The analysis demonstrates that it is possible to investigate prescribed burning effectiveness at risk mitigation across a range of treatment levels, management values and landscapes, by placing them on a common baseline of risk.

}, issn = {421}, author = {Hamish Clarke and Owen Price and Matthias M. Boer and Brett Cirulis and Trent Penman and Ross Bradstock} } @conference {bnh-4774, title = {A systematic exploration of the potential for bushfire risk mitigation with prescribed burning}, booktitle = {AFAC18}, year = {2018}, month = {09/2018}, publisher = {Bushfire and Natural Hazards CRC}, organization = {Bushfire and Natural Hazards CRC}, address = {Perth}, abstract = {

Fire regimes vary widely across Australian ecosystems as a function of climate, fuel, terrain and ignition variations. Fundamentally such variation will not only shape the way that prescribed burning can reduce risk to human and environmental assets but also the scope for effective treatment. While many agencies are moving toward planning systems based on risk assessment, knowledge of the best way to use prescribed fire to reduce risk to key values is generally lacking. The BNHCRC Project, {\textquotedblleft}From hectares to tailor-made solutions for prescribed burning{\textquotedblright}, combines simulation and empirical approaches to improve our understanding of how risk to any particular management value will respond to variations in the spatial location and rates of treatment. Here, we present the modelling framework and key results for two landscapes, Tasmania and the Australian Capital Territory. We run a large number of simulations using the PHOENIX RapidFire model, investigating the interaction between fuel treatment and location under various weather scenarios. Key outputs for risk assessment include area burnt, house loss, life loss, roads and powerlines damaged, environmental cost and economic cost. Across both case study landscapes, greater levels of prescribed burning tend to result in reduced wildfire impacts on all risks. However, there is considerable variation in the rate of reduction in risk, including the amount of treatment required to achieve key targets. Further, the particular combination of weather factors underpinning given fire weather conditions (e.g. temperature vs wind driven) can substantially impact the overall level of risk, as well as the response to prescribed burning.\ 

}, author = {Brett Cirulis and Hamish Clarke and Ross Bradstock and Matthias M. Boer and Trent Penman and Owen Price} } @article {bnh-4203, title = {From hectares to tailor-made solutions for risk mitigation - systems to deliver effective prescribed burning across Australian ecosystems: annual project report 2016-17}, number = {310}, year = {2017}, month = {09/2017}, institution = {Bushfire and Natural Hazards CRC}, address = {Melbourne}, abstract = {

We are pleased to present the 2016-2017 Annual Report for the Bushfire and Natural Hazards CRC project, From hectares to tailor-made solutions for risk mitigation: systems to deliver effective prescribed burning across Australian ecosystems. By undertaking a systematic investigation of the drivers of prescribed burning effectiveness across southern Australia, the project will provide critical support to agency decision makers across the region.

This report describes the project goals, methods and activities since the 2015-2016 Annual Report. The report focuses on initial results from a key project component: fire spread simulations in a range of case study landscapes throughout southern Australia. Results are presented in detail for case studies in East Central Victoria, Adelaide, the ACT and Hobart. This initial suite of simulations provides good evidence that there is substantial variation between case study landscapes in the response of key risks to different rates of prescribed burning. Further simulations in combination with other project methodologies will provide further evidence about this critical issue.

The report also details progress on data acquisition, empirical analyses and stakeholder engagement, which form the other key activities of the project to date. Finally, we are graetful to Naomi Stephens and Felipe Aires, our End User representatives from the NSW Office of Environment and Heritage, for providing an insight into the project{\textquoteright}s progress from a stakeholder perspective.

}, issn = {310}, author = {Hamish Clarke and Owen Price and Matthias M. Boer and Brett Cirulis and Trent Penman and Ross Bradstock} } @article {BF-4295, title = {Sensitivity Analysis of PHOENIX RapidFire}, year = {2013}, month = {05/2013}, abstract = {An analysis of the sensitivity of the outputs of PHOENIX Rapidfire (PHOENIX) to a range of inputs and simulation parameters was undertaken. This was done using two separate methods; assessment of model response in an artificially generated idealised landscape and assessment using case-studies of real fires. The ideal landscape was used to evaluate model sensitivity in response to temperature, relative humidity, wind speed, fuel type and wind direction relative to slope. The model was evaluated under two sets of weather conditions, mild (representing moderate fire spread potential) and extreme (representing high fire spread potential). Each scenario was evaluated for each of two fuel types, forest and grass. Sensitivity was evaluated in terms of the gross area burnt when the input of interest was systematically changed while all other inputs were held constant. For all evaluations except relative wind direction, model sensitivities were compared to an equivalent area burnt using the corresponding McArthur Forest or grassland fire danger meter (assuming an elliptical fire shape). The combination of wind direction and slope resulted in simulated fires that were not elliptical, so comparisons with shapes generated with the fire danger meters were not valid. PHOENIX predictions differed from those generated using point estimates for some circumstances; however without further investigation it is unclear on what is causing these differences. Differences in predictive performance are not necessarily representative of model error, as there are a number of differing assumptions between the systems used. However, specific situations have been flagged for follow up work. Two case study areas were used for PHOENIX model sensitivity evaluation; Wangary and Kilmore. Case study fires were simulated using observations from the day that the fires occurred with one input systematically varied. Three inputs were evaluated using the case studies; simulation resolution, start time (simulating fire ignition to occur earlier and later than observed) and start location (varying the ignition location in space). Sensitivity was evaluated by considering the change in the Area Difference Index (ADI, an index of the ratio between incorrectly predicted burnt area and the correctly predicted burnt area) from the baseline scenario (simulation resolution of 180m, ignition location and time as observed. Predictive performance varied wide with changing inputs. In general as the difference in input value to the {\textquoteleft}best estimate{\textquoteright} increased, predictive performance degraded.}, author = {Chong, Derek and Tolhurst, K.G. and Thomas Duff and Brett Cirulis} }