Blue-tongued lizards and sulphur-crested cockatoos are among the native animals that are often smuggled overseas.
While the number of live animals confiscated by the Australian government has tripled since 2017, the full extent of the problem eludes us as authorities often do not know where and how wildlife is being traded. Now we can add a new technology to Australia’s arsenal against this cruel and inhumane industry.
Our study published today shows the potential for new technologies to detect illegal wildlife in baggage or mail. This technology uses artificial intelligence to recognize the shapes of animals as they are scanned at international frontlines such as airports and postal centers.
Exotic species are also smuggled in the land, like snakes, tortoises and fish. This could disrupt Australia’s billion-dollar agricultural industry by introducing pests and diseases, and also threaten fragile native ecosystems.
Shingleback lizards are one of Australia’s most traded animals. Photo: Shutterstock.
An animal welfare issue
The wildlife trade is driven by several factors, including alleged medicinal purposes, animals for ornamental value, or for the illegal pet trade.
It can have fatal consequences, as it usually involves individual animal transport in narrow or cramped environments. This often leads to animals becoming stressed, dehydrated and dying.
Some people have even attempted to smuggle Australian wildlife in packets of chips.
Human traffickers often transport several individuals at once in hopes that an animal will bring them alive.
We do not have the full picture of what animals are being traded, how they are being traded, or when. But examples from confiscated cases in Australia suggest traffickers value Australian reptiles and birds.
For example, the dwarf lizard, a species of blue-tongued lizard, is considered one of Australia’s most traded species.
Just another Sulphur-crested Cockatoo for you? These Australian birds are exotic in the international pet trade and a known victim of the illegal wildlife trade. Photo: dr Vanessa Pirotta.
Aside from being cruel and inhumane, the wildlife trade can also facilitate the introduction of alien species into new environments.
This poses significant biosecurity risks. For example, zoonosis (diseases that spread from a nonhuman animal to a human) affects people who interact with stressed, wild animals. Exotic species can also disrupt natural ecosystems, as we famously saw with the damage done by cane toads in northern Australia.
Unregulated wildlife entering the country may also harbor new diseases or destructive parasites. This could hurt the agribusiness and potentially increase the price of our fruit and veg.
Creating a trafficking image library
Our new research documents a variety of wildlife species scanned using state-of-the-art technology to create “real-time tomography” computer algorithms.
Real-time tomography is an imaging technique that uses a series of X-rays to scan an object (such as a lizard). It then creates a three-dimensional image of the animal, which in turn is used to develop algorithms. For example, mail and luggage can be scanned at the airport, and if wildlife is trapped, the algorithms will alert operators of their presence.
Our study scanned known species of traded Australian animals to create an image reference library. A total of 294 scans of 13 species of lizards, birds and fish were used to develop first wildlife algorithms with a detection rate of 82% and a false alarm rate of just 1.6%.
Wildlife algorithm successfully detects a dwarf lizard. This is a screenshot of the user interface that alerts the operator to a spotted shingle-backed lizard (Tiliqua rugosa) via the green bounding box that identified it as a lizard. Image: Pirotta et al. 2022
This study is the first to document the use of 3D X-ray CT security scanning technology to protect wildlife in the peer-reviewed scientific literature. It is also the first to present results for the detection of reptiles, birds and fish in such scans.
The detection tool is designed to complement existing detection efforts by Australian Border Forces, biosecurity officers and detection dogs, which remain vital in our fight against wildlife crime.
How else can we stop the wildlife trade?
The tools currently used to help detect and limit wildlife trafficking rely primarily on human detection methods.
This includes cybercrime investigations or manual bag searches by Australian border police and biosecurity officers. Biosecurity detector dogs patrolling airports are also useful, as are smartphone reporting apps like the Wildlife Witness app.
Efforts to dismantle illegal trade networks at the source are also crucial. It does this by understanding and reducing consumer demand for wildlife and wildlife products, providing alternative livelihoods for would-be poachers, and enforcing stronger governance and monitoring.
Confiscated animals can be used as evidence to identify traffickers, with previous cases leading to successful prosecutions by environmental investigators. For example, a former rugby league player has been sentenced to four years in prison after being caught trying to smuggle a variety of animals in and out of Australia.
continuation of the fight
All of these measures are helping to combat the wildlife trade, but there is no single solution to predict when and where the events are likely to take place.
Wildlife traders can often adjust their behavior to avoid detection. Therefore, innovative and adaptable solutions, like our new technology, are crucial to support existing detection techniques.
Any attempt to eradicate this horrifying activity is a step in the right direction, and the potential of 3D detection allows us to adapt and evolve the way traffickers can change their behavior.
we want dr Phoebe Meagher of Taronga Conservation Society Australia for her contribution to this study and article.
Vanessa Pirotta, postdoc and wildlife scientist, Macquarie University and Justine O’Brien, Manager of Conservation Science, Taronga Conservation Society Australia, University of Sydney, UNSW Sydney
This article was republished by The Conversation under a Creative Commons license. Read the original article.