Artificial Intelligence helps to detect fires and predict fire behaviour. We need to have a forward view to predict what a fire will do to accurately predict the spread and intensity of any fire in Australia.


Photo taken from a plane shows bushfires in north-eastern New South Wales – November 9, 2019. Photo Credit: Tom Bannigan / AFP via Getty Images.

The Bushfire Data Quest 2020 has brought together the best scientific minds in bushfire research, machine-learning and earth-observation in an intensive research sprint to:

  • more accurately predict fuel loads across Australia;
  • detect fires earlier; and
  • better predict fire behaviour.

The interdisciplinary approach of Data Quest aims to accelerate research and generate more innovative, outside-the-box solutions to these complex issues.

The challenge given to the multi-disciplinary expert teams was:  Can AI, primed with data from multiple satellites and local sensor networks, detect fires earlier and predict fire behaviour?

Minderoo Foundation’s Fire and Flood Resilience initiative is the challenge partner in the inaugural Bushfire Data Quest 2020, which is based on a model that has been very successful in addressing large and complex issues in the United States and Europe. The event is also supported by the New South Wales Government and a number of academic research partners.  Supporting an open innovation model, the results of the research sprint will be available for further testing and deployment to organisations looking to provide better prediction technology and platforms for emergency services.

“With the support of the Minderoo Foundation, the teams participating in the Bushfire Challenge have made some real strides in remote sensing and AI prediction, which have the potential to find new ways to protect our homes, lives, environment and economy from the destructive impact of bushfires.”

Dr Cormac Purcell, Data Quest lead

Stay informed with news FROM FIRE & FLOOD RESILIENCE.