The challenge

Predicting bushfire spread

Although a natural occurrence in Australia, bushfires have the ability to be devastating, particularly when they come into contact with homes, infrastructure and people. Improved knowledge of how bushfires spread is critical for emergency management operations, risk prediction and timely issuing of warnings.

Dr Mahesh Prakash led the team behind Spark, our bushfire prediction modelling tool.

This knowledge can also be used to predict the spread of bushfires using computer models. Such predictions can allow better preparation for emergency situations, save lives and make communities safer, stronger and more resilient. However, there are many elements that influence fire behaviour making it difficult to accurately predict fire spread.

Our response

A flexible bushfire simulation framework

We developed 'Spark', an open framework for fire prediction and analysis. It takes our current knowledge of fire behaviour and combines it with state-of-the-art simulation science to produce predictions, statistics and visualisations of bushfire spread.

[Music plays and text appears:  Spark:  A better way to predict the spread of bushfires]

 [Image changes to show a bushfire]

 [Image changes to show Dr Mahesh Prakash, and text appears on screen:  Dr Mahesh Prakash, Group Leader | Computational Modelling and Simulation]

 Dr Mahesh Prakash:  Spark is a bushfire spread modelling capability which has been developed by CSIRO.

 [Image changes to show a computer simulation of a bushfire]

 [Image changes back to Dr Mahesh Prakash]

 It provides emergency management agencies a much better ability to predict fires, and save lives and property.

 [Image changes to show various computer simulations]

Spark is able to provide information around the rate at which the fires spreads, the direction in which it spreads, as well as the locations that the fire actually affects.  Your inputs are essentially your weather data, mainly your wind data, the location where the fire has been started, the data about vegetation, the fuel type, as well as the terrain.

[Image changes back to Dr Mahesh Prakash]

And once you have all these inputs you can run your fire predictions for any given location.

 [Image changes to show computer simulation of a bushfire]

 And it runs better than real time, so you can actually make your decisions around where to actually target the fire fighting efforts.

 [Image changes to show a bushfire]

 [Image changes to show a fireman spraying water on the bushfire from a vehicle]

 [Image changes back to Dr Mahesh Prakash]

 It can also be used by researchers to actually try and understand different types of fire models.

 [Image changes to show various computer simulations of a bushfire]

 [Image changes back to Dr Mahesh Prakash]

Now that Spark’s actually publically available and it’s downloadable from our website, emergency agencies can actually have a firsthand look at how it performs, and Spark’s actually able to be integrated with their existing systems.

[Image changes to show Dr Mahesh Prakash working at a computer]

[Image changes back to Dr Mahesh Prakash]

You also have the ability to incorporate future changes, which might also include climate change effects, into any model that gets developed.

[Image changes to show a bushfire]

[Image changes back to Dr Mahesh Prakash]

People can be better prepared against potential bushfire risks, both from a saving lives perspective, as well as saving property perspective.

[Image changes to show a fireman spraying water on a bushfire from a moving vehicle]

[Image changes back to Dr Mahesh Prakash]

So it’s a very powerful tool for emergency management agencies.

[Music plays and CSIRO logo appears with text: Big ideas start here www.csiro.au]

Spark: A better way to predict the spread of bushfires

Weather conditions such as wind, temperature and humidity contribute significantly to fire behaviour.

Spark can read weather data from meteorological forecasts and use this information directly within fire models. Geographic information, such as land slope, vegetation and un-burnable areas, such as roads and water bodies, also affect the spread of the fire. Spark allows users to easily incorporate such environmental data and to use this information to define a fire spread rate.

The results

More information for better decision making

Predicting the spread of bushfires will have benefits across planning and also emergency management, including:

Spark takes current knowledge of fire behaviour and combines it with state-of-the-art simulation science to predict bushfire spread.

  • infrastructure planning
  • land management and fuel reduction burning
  • fire fighting resource allocation and deployment
  • evacuation route planning
  • reconstruction of historical fire events
  • ecological impacts and fire regime studies
  • suppression strategy analysis.

In February 2021, a partnership between CSIRO’s Data61 and AFAC, the National Council for Fire and Emergency Services, announced the development of a nationally consistent bushfire modelling and prediction capability.

The partnership involves the development of Spark Operational, a cutting-edge bushfire simulation tool based on Data61's Spark fire prediction platform. Fire and emergency service agencies across the nation will now apply state-of-the-art simulation science to produce predictions, statistics and visualisations of bushfire spread, as well as simulating hours of fire spread across a landscape in a matter of seconds.

"Spark Operational will play a significant role in allowing our emergency response teams to effectively plan for and respond to fire emergencies in a variety of landscapes and climates," explained AFAC CEO Stuart Ellis.

"It was identified as the best solution to use to help achieve a nationally consistent system that will take the nation to the next generation of firefighting intelligence, and ensure we are protecting as many lives and assets as possible across multiple scenarios, mitigating the dangers of bushfire."

Read more about Spark Operational at New version of Spark to be used nation-wide to model and predict bushfires .

Do business with us to help your organisation thrive

We partner with small and large companies, government and industry in Australia and around the world.

Contact us now to start doing business

Contact Data61

How can we help you create your data-driven future? Use the form below to send us a message.
Your contact details
0 / 100
0 / 1900
You shouldn't be able to see this field. Please try again and leave the field blank.

For security reasons attachments are not accepted.