From her bush block in the Blue Mountains, Bianca Nogrady considers how researchers will tackle the ‘new normal’ for severe bushfire weather.
On February 7, 2009, several bushfires in Victoria burned through 450,000 hectares over two days, destroying more than 2000 homes, killing 173 people plus an estimated one million wild and domestic animals.
Between September 2019 and March 2020, hundreds of bushfires burned through almost 19 million hectares across Australia, destroying more than 2500 homes, killing more than one billion birds, animals and insects, and driving many species to the brink of extinction. But the death toll was lower, at 33 people.
Every one of those lives lost was a tragedy that devastated families, friends, colleagues and communities. But given the unprecedented scale, ferocity and duration of this season’s fires, the fact that comparatively few lives were lost suggests that since Black Saturday, we have made profound changes to the way we predict, understand and respond to bushfires.
Many of those changes have come about because of the collaborative research undertaken by the Bushfire & Natural Hazards Cooperative Research Centre.
“We strongly believe the work of the CRC has been instrumental in reducing the death toll out of these fires,” says Dr Richard Thornton, Chief Executive Officer of the Bushfire & Natural Hazards CRC (BNH CRC). The relatively low death toll from the 2019–2020 bushfire season is astonishing to many people working with bushfires, especially given that there were several days during the season when conditions were as catastrophic as those experienced during the 2009 Black Saturday fires.
Getting better, earlier warnings
One of the areas of change is communication: how, where, why and to whom emergency warnings are delivered, and the content and wording of those warnings. Communication has been an active area of research from the early days of the original Bushfire CRC, which was established in 2003 and morphed into a focal program of the current BNH CRC from 2013.
Associate Professor Amisha Mehta, a risk and crisis communications expert at the Queensland University of Technology (QUT) Business School, has been working with the BNH CRC on risk communication during bushfires and other natural hazards.
She says the wording of risk and emergency warnings is a delicate balancing act between trust and encouraging individual responsibility.
“In our co-designed messages, we have enhanced trust but at the expense of reducing people’s perception of personal responsibility,” says Mehta. “So we learn from that and talk about ways to maintain trust and enhance or build people’s ability or confidence in being able to take the actions needed.”
Extensive research on how people respond to emergency warnings and risk messages has led to some bushfire and emergency agencies changing the wording of their message headings. The three tiers of ‘Advice’, ‘Watch and Act’ and ‘Emergency Warning’ are classifications firmly embedded into the way emergency management organisations work since the Black Saturday bushfires.
However, those headings don’t necessarily reflect the way individuals think in a bushfire situation. Instead, some agencies are switching to what Mehta describes as more ‘community-minded’ language. “Instead of ‘emergency warning’, the lead is ‘leave now’, or ‘leave immediately’, or ‘shelter in place’, so it’s the behaviour that is captured in the heading,” she says.
After those headings comes more detailed information about the location and type of hazard, timing and other aspects.
For example, the Queensland Fire and Emergency Services have amended their headings to ‘Advice: monitor conditions’, ‘Watch and Act: conditions are changing’, and ‘Emergency: you are in danger’.
Mehta and her colleagues’ research also showed that adding a reason for the warning helped the message get across. “If you add a rationale, such as ‘leaving now is your safest option’, even though it makes the message longer, it enhances residents’ self-efficacy, so it makes them feel more confident in taking the behaviour.”
When to leave and who to tell
Another active area of CRC-led research is when and to whom those messages should be sent.
Bushfire predictions have taken a quantum leap forward in recent years with the development of a variety of computerised modelling systems that can predict bushfire risk from as far out as one year ahead of a season to an hour-by-hour update on where a bushfire is likely to spread or its embers land.
Two products helping fire agencies calculate risk are Phoenix and SABRE Fire. Ben Twomey is a fire behaviour analyst and executive manager of advanced capability in the Queensland Fire and Emergency Services, and he works with both these fire modelling systems. “Part of the reason for the comparatively low number of deaths this season is the prediction capability and our ability to get people out of the way of fires that we know are going to be catastrophically bad,” he says.
Phoenix is a fire simulator program developed by the Bushfire CRC and the University of Melbourne. It characterises fire spread across the landscape based on forecast weather — temperature, humidity, wind speed and direction — fuel maps, topography of the landscape, wind modification by the landscape, vegetation, the fire history of an area and other factors such as roads, fire breaks and rivers.
When that data is put into the system, along with weather inputs to account for factors like the effect of pyrocumulonimbus, the output is a map forecasting where that fire is likely to spread to in the next hours and days.
But as anyone who has ever tried to plan a picnic knows, the weather forecast is a probabilistic prediction; it offers a likelihood — not a certainty — of the weather being a certain way. To account for that element of uncertainty, there’s SABRE Fire.
“We don’t know those inputs with a great deal of certainty and we’d rather be broadly right than precisely wrong,” says Twomey.
SABRE Fire calculates a range of scenarios by varying some of the inputs at random, like a much higher wind speed or much lower humidity, “so when we talk about worst case scenarios, theoretically it’s already built in”.
When fire controls the weather
There are some wildcards in fire behaviour modelling and pyroconvective fire behaviour is one of those. This refers to the enormous weather systems that can develop above intense fires — so-called ‘fire thunderstorms’ — and drastically alter weather and fire behaviour. “It’s like putting a chimney on top of the fire and getting a nice big draw around the edge – it goes berserk,” says Twomey.
Research by the BNH CRC has worked out the minimum heat required for these pyrocumulonimbus systems to develop in various environments, which should help better forecast these events and allow for their incorporation into fire behaviour modelling.
This research will be part of the next generation of fire spread models, which includes weather modelling conducted by the Bureau of Meteorology, as well as in more real-time models being developed with CSIRO and capable of being used in control centres just like Phoenix.
Even with all the best predicting, forecasting and messaging, people and properties are still being caught in bushfires. Understanding how and why that happens is the question Dr Josh Whittaker from the University of Wollongong and the BNH CRC is trying to answer.
After fires have passed through, Whittaker and colleagues visit those communities and interview residents about their awareness of bushfire risk, what planning and preparations they had done before the fire, what information and warnings they received and how they responded.
“Those findings are fed back into fire services to help them better communicate with communities and better prepare them for bushfire in the future,” says Whittaker.
Having worked on many fires since the devastating Black Saturday bushfires, Whittaker says there is always a wide range of preparedness levels in the community.
“There always seem to be people who have done little or nothing to prepare and therefore have difficult experiences in fires,” he says. “Through to the other end of the spectrum, where people are very well prepared and are either able to safely leave, as is their plan, or remain to defend their farms or houses.”
Understanding those decisions is critical to helping shape messaging and awareness campaigns about bushfire preparedness. Whittaker says one interesting finding is how many people in bushfire-exposed areas are unaware they are even at risk.
One change he has seen over time is greater focus and understanding of the ‘leave early’ message during bushfires.
There’s also greater awareness of the high level of planning and preparation that is required by anyone thinking of staying to defend their property. And it seems the most important message is being heard: don’t leave anything to the last minute.
There is a mountain of research to be done after this bushfire season, which will help researchers understand what contributed to the severity of this season and how well predictions, preparedness and response systems worked. But there is also the question of how Australia’s experience of bushfires will change in the future.
There is also population growth and how that changes the relationship between humans and the bush. There is the question of settlement strategies and decisions by local and state governments, and what role the insurance industry will play in that.
There are new technologies that will present opportunities and challenges, when it comes to preparedness and response to bushfires. And finally, there is perhaps the greatest challenge of all: climate change.