Led by the University of South Australia (UniSA) and supported by the SmartSat CRC, this pioneering artificial intelligence (AI) technology aims to detect bushfires from space 500 times faster than traditional on-ground methods which have previously lacked the capability to analyse complex images of Earth in real time.
South Australia's first cube satellite, known as Kanyini, uses advanced technology incorporating AI-enabled software into tiny cube satellites that weigh just a few kilograms. It can help can detect smoke, the earliest visible sign of fire, before it grows into a full-blown disaster. This promises swift response times to bushfire outbreaks, potentially protecting human lives, wildlife, property and reducing environmental damage.
Once the solution is operational by 2025, it would be possible to reliably detect wildfires in under an hour, compared to current satellites that take between six to eight hours. In a test involving a past fire event in the Coorong of South Australia, this AI approach detected smoke and relayed the data in less than 14 minutes.
“This will not only prove invaluable in the event of bushfires but also serve as an early warning system for other natural disasters,” Dr. Stefan Peters, lead researcher at UniSA, said in a press release.
According to a WWF-Australia report, between 400-700 million tonnes of carbon dioxide were released during the 2019-2020 wildfire season. To put this in perspective, Australia's annual emissions for the 12 months leading up to June 2019 were approximately 500 million tonnes.
The fires affected approximately 143 million mammals, 2.46 billion reptiles, 181 million birds, and 51 million frogs, with more than 60,000 koalas impacted. Spotting wildfires early will help reduce the amount of carbon dioxide released into the atmosphere, potentially benefiting the ozone layer and saving Australia's invaluable wildlife.
A video explaining the research is also available here.
Planet Ark does not take responsibility for the accuracy of the original information and encourages readers to check the references before using this information for their own purposes.