The impact of wildfires on the Australian outback and wildlife was devasting. The 2019-2020 season was particularly bad, causing unprecedented harm and affecting over 3 billion animals. Of course, it takes time for the environment to recover from such an event, and conservationists are monitoring the recovery via AI technology and a network of cameras.
Seven million photos from around 1,100 sensor-activated cameras in eight bushfire-affected areas across Australia were collected and analysed. The project is a collaboration between the World Wildlife Fund (WWF), Conservation International, and local land managers.
The Google AI technology, Wildlife Insights, was trained on 4 million images of over 150 Australian animals, which allowed researchers to track their recovery with over 90 per cent accuracy. The AI then tags all the animal species, collates and interprets the data, and creates maps and charts to show trends. It’s a massive labour-saving device for researchers.
The images have shown the return of vulnerable species of animals, such as wombats, echidnas, dingos and koalas, to previously burnt-out areas of land.
The dunnarts of Kangaroo Island lost around 90% of their habitat in 2020. However, according to the project, they are showing a steady return to the area. Dunnarts are mouse-like marsupials native to Australia.
Similarly, koalas are returning, although are seen on the cameras to be mostly ground-dwelling rather than living in their usual treetop homes. This could make them more vulnerable to predators.
AI wildlife tracking cameras have been piloted in other parts of the world to assist in conservation projects. In Gabon, AI cameras have been used to effectively monitor and track elephants that are in danger of poaching. The cameras are able to monitor for suspicious activity and alert security guards.
[Via One Green Planet]
FIND THIS INTERESTING? SHARE IT WITH YOUR FRIENDS!