Bushfires have disastrous effects on Australian communities almost every year. But now a new AI app being developed by University of Sunshine Coast researchers could help Australians living in bushfire prone areas to better predict the likelihood of fires, hopefully also limiting their devastating effects.
The app development project dubbed, NOBURN (National Bush Fire Resilience Network), will be funded by a $500,000 Australian Citizen Science Grant that will be spent on its design and implementation.
Once developed, it’s hoped the app will be used by people living or hiking in bushfire prone areas to collect data in the form of photos and forest fuel samples. This data will then be matched with satellite imagery and used by experts at the Australian Institute for Machine Learning at the University of Adelaide, to train AI systems to predict the probability, severity and burn areas of future bushfires.
Bushfires are common in Australia’s warmer months, with residents in danger areas frequently at risk from devastating infernos. In fact, just between 2019-2020 fire season, over 2,000 homes were destroyed, and millions of dollars lost in industries such as the agricultural and tourism industries. However, bushfires can be annoyingly hard to predict by conventional means.
“A range of factors determine the intensity and speed of fires, including fuel load, moisture, ignition source and wind. However, the methods currently available to predict fire events and risk are complex and not easily accessible for residents in bushfire-prone areas,” says Sunshine Coast University NOBURN project lead, Professor Mark Brown.
Once trained with data, the app will be used when a bushfire ignites to better determine the direction, extent, and severity of the boundary fire, which will allow fire crews to more strategically intervene in bushfire control by passing on key information to fire authorities, forest professionals, landowners and residents living in fire prone areas.
“While naturally occurring bushfires cannot be avoided, there is an opportunity with this project to predict the likelihood and implement strategies to minimise their impact on the environment, property and life,” says Brown.