NSW turns to AI to predict bushfire activity

The NSW government is looking at using machine learning to analyse data from satellites and local sensor networks in a bid to better detect, predict and respond to bushfires.

Deputy Premier John Barilaro on Sunday said the government was “investigating further” how this data might be used to avoid a repeat of the catastrophic Black Summer bushfires.

It follows the Bushfire Data Quest earlier this month, a week-long research sprint that sought to determine whether fusing data from these sources could speed up detection and response times.

The quest looked at three challenges centred around improved fuel assessment, early detection and improving the identification of false positives and machine learning fire behaviour models.

Working in teams of five, machine learning specialists joined bushfire researchers to solve the problems, with support to be provided by a network of faculty.

Barilaro said the government is now working with the teams to understand how satellite data could help emergency services predict and fight bushfires during the bushfire season.

“Predicting the behaviour of bushfires is a hugely difficult problem, made more complicated by a myriad of factors such as fuel load, atmospheric conditions, soil moisture, and availability of water,” Barilaro said.

“Using data from satellites is a great advancement on the tools we have traditionally used with much of the task of planning on-the-ground bushfire response relying on the experience and instincts of firefighters - who are often volunteers.

“We are investigating further how we use the data from multiple satellites and local sensor networks to create algorithms that will help detect fires earlier, predict fire behaviour, and help emergency services respond more effectively to protect homes, people and nature.”

Bushfire Data Quest lead Dr Cormac Purcell welcomed the government’s support for the challenge, which is also supported by Andrew Forrest’s Minderoo Foundation.

“By drawing on a wide range of experience, skills and viewpoints, the teams will naturally create better outcomes,” he said.

“We harness the best practices from the private, academic and non-profit sector to accelerate research for the benefit of all humanity.”