Drones Used to Monitor Koalas

Drones have been aiding in environment conservation projects for some time now. In another significant contribution to conservation of wildlife, researchers at Queensland University (QUT) of Technology have developed an innovative method for detecting koala populations using drones and infrared imaging that is more reliable and less invasive than traditional animal population monitoring techniques.

In the study led by Dr Grant Hamilton, from QUT’s School of Earth, Environmental and Biological Sciences, co-authored with PhD student Evangeline Corcoran and Dr Simon Denman from QUT, and John Hanger and Bree Wilson from Endeavour Veterinary Ecology.

Dr Grant Hamilton is looking at the invasive pest myrtle rust and its spread across the Australian bush and into Brisbane. Photo:QUT Marketing and Communication/Erika Fish.

In their study, published in Nature journal Scientific Reports, the QUT researchers detail the technique that has great potential to improve management of koala populations and other threatened species as well as being used to detect invasive species.

While detection systems for other animal population by drones is fairly simple for example looking for seals on a beach or animals on the savannah.

“A seal on a beach is a very different thing to a koala in a tree,” Dr Hamilton said. Dr Hamilton said the researchers were able to correlate the detection of koalas from the air using ground surveys of tracked radio-collared koalas in Petrie, Queensland.

The system uses infrared imaging to detect the heat signals of the koala despite the canopy coverage of the eucalyptus trees. Researchers carried out aerial sweeps of the area at early morning during colder months, when difference between the body heat of the koalas and the background was likely to be greatest. The drone covered an area in a “lawnmower” pattern, going up and down over a specific area. After the flight, the data was put through an algorithm that was designed to identify the heat signatures of the koalas while ignoring other objects and animals like cars and kangaroos in the area.

For these initial tests, the accuracy of the system was checked by comparing the inferred koala locations with ground truth measurements provided by GPS units on some animals and radio tags on others. The system detected about 86 percent of the koalas in a given area, considerably better than an “expert koala spotter,” who rates about a 70.

Drone detection is also quicker and cheaper than using human spotters for the same area. Dr Hamilton added that high accuracy rate of the drone detection did not render other means of determining koala population, such as by human spotters or dogs redundant. “There are places that people can’t go and there are places that drones can’t go,” Hamilton said adding, “what we do know now is that this is a really powerful tool within the tool box.

Following the success of the study, Dr Hamilton said the researchers were looking to expand the area where they had studied koala populations, and would use the drone system in other parts of Brisbane, south-east Queensland and northern New South Wales. With plans to adapt the algorithm the technology could be used to detect invasive species such as deer.