Australia’s deserts are quietly staging a comeback – and the heroes of this story are tiny, vulnerable marsupials learning to outsmart one of the continent’s most destructive predators: the feral cat. But here’s where it gets controversial… instead of trying to wipe out every cat, scientists are asking a daring question: can native animals be “trained” by nature itself to live alongside them?
Deep in the harsh interior of the Australian outback, many of the country’s most distinctive small marsupials – such as bilbies, bandicoots and quolls – vanished from the landscape decades ago, largely because of habitat clearing and the hunting skills of feral cats. These cats, descended from domestic animals brought by European colonisers, are incredibly fast, agile and efficient hunters, and the native mammals had never evolved alongside such a flexible, relentless predator. They simply did not have the instincts or adaptations to recognise this new threat in time.
Yet efforts to completely remove cats from the environment have repeatedly fallen short, despite decades of control programs and significant investment. In response, a team of researchers has launched a bold ecological experiment: instead of relying only on eradication, they are testing whether small marsupials can, over time, learn behaviours and develop traits that allow them to survive in a landscape where cats are still present. In other words, they are trying to turn these animals into “predator-smart” survivors rather than permanent refugees behind fences.
At the centre of this approach is a specially designed “training zone” of about 100 square kilometres inside the much larger Sturt National Park in north‑western New South Wales. Nearby, conservationists have constructed two fully fenced sanctuaries where feral cats are excluded, and within these safer areas they have reintroduced six species of small marsupials, including crest‑tailed mulgaras, bilbies and quolls. These species once lived in the region around a century ago, before cats and foxes decimated them, leaving the desert without many of its small native mammals.
The big challenge is straightforward to state but difficult to solve: how do you help native animals “wise up” to a predator they did not evolve with? The training zone provides a kind of middle ground. It is not completely safe, but it is not as dangerous as the surrounding landscape either. Two sides of the zone are bordered by long dingo fences, which incidentally block some – though not all – cats from entering, helping to reduce predator pressure without removing it entirely.
Inside this learning landscape, managers keep cat numbers low using several control methods. Cats are sometimes removed by shooting, but more sophisticated devices are also deployed that can distinguish between a native animal and a cat before spraying a lethal toxin onto the cat’s fur. When the cat grooms itself, it ingests the poison and is humanely killed, reducing the number of hunters in the area. Thanks to these measures, cat density in the training zone is now only about three cats per square kilometre, while outside the zone there can be ten times as many.
The stakes are high. Some estimates suggest that feral cats kill more than two billion animals across Australia every year, from reptiles and birds to mammals and even insects. Against that backdrop, even a modest reduction in cat numbers – combined with smarter, more cautious native prey – could make a huge difference to whether species cling on or disappear altogether. And this is the part most people miss: the goal is not to create a perfectly safe bubble, but to create conditions where natural selection and learning can actually work.
The Wild Deserts project, which runs this experiment, is led on the ground by ecologist Dr Bec West, who lives with her family in an old homestead within the national park. Since 2024, the project has released dozens of western quolls, hundreds of bilbies and hundreds of golden bandicoots into the training zone, all of which are officially listed as nationally threatened. Soon, burrowing bettongs – another once‑abundant desert mammal – are expected to join them.
The team is realistic about the costs. Some animals will be killed by cats, and West is upfront that predation is a natural process. The crucial question is whether enough individuals survive, learn, breed and pass on useful behaviours or traits to the next generation. Recently, the team trapped and monitored 57 quolls, bilbies and bandicoots inside the training area, with bilbies and quolls having survived at least a year there, and bandicoots making it through three months. Those survival times hint that many individuals are figuring out how to persist in this semi‑dangerous environment.
To track how things are changing over time, West’s group reviews images from around 50 motion‑activated cameras spread throughout the training zone every month. In the past few months, they have started seeing more photos of bilbies and quolls than of cats – a visual sign that native animals are becoming more established. For the scientists, this feels like a symbolic turning point: the marsupials are, quite literally, starting to take back the desert.
So how exactly do these animals adapt? Part of the answer may be behavioural learning. Some marsupials might observe cues in their environment – such as scents, sounds or the sudden disappearance of other animals – and gradually become more cautious. Other shifts may be evolutionary, with individuals that are naturally more wary or faster on their feet more likely to survive and reproduce.
Studies in another desert region in South Australia have already shown that burrowing bettongs become more watchful and vigilant when cats are present, paying closer attention to potential threats around them. In a separate experiment, scientists monitored two groups of bilbies over several years: one group lived in an area where cats were present, and the other group lived without cats. After about five years, the bilbies that had grown up with cat exposure had developed noticeably larger feet and fled from approaching humans much sooner, suggesting they had become more cautious and better prepared to escape.
Those larger feet might give the bilbies a speed or agility advantage when sprinting away from predators across soft desert sands. Just as importantly, their increased tendency to flee early shows a stronger instinct to treat unfamiliar movement as dangerous. Both traits could be favoured over generations, leading to populations that are better equipped to handle the constant threat of predation.
Within the training zone, the reintroduced marsupial species have now spread across the entire area, indicating that they are not just surviving in small pockets but using the landscape more fully. They have bred successfully, and their young have grown up knowing only a world where cats exist but are relatively scarce. The hope is that these “predator‑savvy” animals will eventually be used to establish new populations in other parts of the desert, where conditions are tougher and cat numbers higher.
Beyond their own survival, these small mammals play a vital ecological role. Species like bilbies, bandicoots and bettongs are often called ecosystem engineers because their constant digging turns over the soil, creates small pits that trap water and organic matter, and helps seeds settle and germinate. Over time, this activity boosts native plant growth, improves soil structure and supports a richer, more resilient ecosystem.
Evidence of this engineering power is already visible near the training zone, especially inside the two fully cat‑free fenced areas. West reports that the ground cover there has been transformed as the animals go about their daily digging. Native plants are thriving in what was once sparse, dry country, and the soil surface is now riddled with little hollows where seeds and moisture collect. The soil is so uneven, in fact, that people walking through the site risk twisting an ankle on the many small pits created by the marsupials’ foraging.
Here’s a provocative angle: is it enough – or even ethical – to rely partly on evolution and learning to solve a conservation crisis that humans created by introducing cats in the first place? Some will argue that only aggressive, large‑scale cat eradication is acceptable, while others see this training‑zone approach as a pragmatic way to buy time and give native species a fighting chance. Should conservation focus on restoring a pre‑cat world, or on helping wildlife adapt to the reality that cats are now deeply embedded in Australia’s ecosystems?
What do you think: is it wise to deliberately expose threatened species to a predator that has already driven many of them close to extinction in the hope that they will adapt, or should efforts concentrate solely on making more predator‑free sanctuaries? Do you see this strategy as innovative and necessary, or as an unnecessary risk for animals already on the brink? Share whether you’re for or against this kind of “training the prey” approach – and why.
