Estimating population size is crucial for managing populations of threatened species. In the Top End of northern Australia, populations of northern quolls (Dasyurus hallucatus), already affected by livestock grazing, inappropriate burning regimes and predation, have collapsed following the spread of the toxic cane toad (Rhinella marina). Cane toads are currently invading the Kimberley, where they pose a threat to quoll populations. To manage these populations, we need reliable methods for detecting and estimating quoll abundance. We deployed camera traps with lures containing tuna, peanut butter or no bait and found that baited cameras performed better than the unbaited control. Cameras with a tuna lure detected more individuals than cameras baited with peanut butter or no bait. Cameras with a tuna lure yielded more photographs per quoll than those baited with peanut butter or no bait. We identified individual quolls from unique spot patterns and found multiple photographs improved the accuracy of identification. We also found that population estimates for the sample area derived from camera trapping were consistent with those from live trapping using mark–recapture techniques.

Sharing landscapes with humans is an increasingly fraught challenge for wildlife across the globe. While some species benefit from humans by exploiting novel opportunities (e.g., provision of resources or removal of competitors or predators), many wildlife experience harmful effects, either directly through persecution or indirectly through loss of habitat. Consequently, some species have been shown to be attracted to human presence while others avoid us. For any given population of a single species, though, the question of whether they can recognise and change their response to human presence depending on the type of human actions (i.e., either positive or negative) has received little attention to date. In this study, we chose to examine the behavioural plasticity within a single population of eastern grey kangaroos (Macropus giganteus) to both positive and negative human activity. Within a relatively small and contiguous landscape, we identified areas where kangaroos experience a combination of either low and high frequencies of benign and harmful human disturbances. From six sampling sessions over five months, we found that density and group sizes were higher where humans acted benignly towards them, and that these groups had higher representations of sub-adults and juveniles than where humans had harmful intentions. Importantly, we found that the vital antipredator strategy of increasing group size with distance from cover was not detectable at sites with low and high levels of harm. Our findings suggest that these kangaroos are recognising and adjusting their behavioural response to humans at fine spatial scales, a plasticity trait that may be key to the survival of these species in human dominated landscapes.

Globally, wilderness is being converted for rural and agricultural land use. In countryside landscapes, many habitat structures remain intact, providing suitable habitat for wildlife species that can accurately assess novel risks and develop tolerance to benign disturbances. Associative learning that promotes avoidance and also facilitates desensitization to benign disturbance is key to persisting in these landscapes. Conversely, learning to distinguish and avoid negative interactions with humans, like hunting, is vital. To determine if eastern gray kangaroos are capable of learning from previous interactions with humans, we tested the flight responses of wild kangaroos which have previously experienced either low or high frequencies of harmful and benign encounters with humans. We found that eastern gray kangaroos rapidly habituated to benign disturbance as there was no significant difference in assessment distance between groups that previously experienced low or high frequencies of disturbance. The threat of harmful disturbances was not as quickly learnt, as groups that experienced low frequencies of harmful disturbance delayed flight longer than those experiencing frequent harm. We found that the influence of environmental and group parameters on a kangaroo's decision to flee depended on the intent and frequency of previous interactions with humans. Our study indicates that kangaroos are learning from previous encounters with humans, correctly assessing novel risks which may be contributing to their persistence in countryside landscapes.