A project undertaken at The University of Western Australia, and supervised by Dr Nicola Mitchell.
Australia has the worst record of mammal extinction of any continent. Increasingly, in order to prevent further extinctions, we rely on predator-free havens; islands and fenced areas within which vulnerable fauna can persist. While this strategy has successfully secured many species, the predator-free conditions inside havens can cause rapid evolutionary loss of antipredator behaviours. Once these behaviours are lost, havened mammals become easy prey for even their natural predators, and so are unfit for release into unfenced habitats.
With only a small fraction of potentially habitable areas captured in havens, rapid loss of predator awareness may be jeopardising national conservation strategies for Australian mammals. Hence is it imperative that we monitor for, and manage against, loss of antipredator traits.
Figure 1. Natasha processing a juvenile woylie after capture – behaviour during capture and handling can potentially give valuable insights into the behaviour of animals in the face of predators (Image: Natasha Harrison)
To do this, we need the ability to make meaningful measures of behavioural phenotypes in havened mammal species. Genetic data cannot solve this problem because the link between genotype and behavioural phenotype is, at best, very murky. Our capacity to genotype animals has improved by orders of magnitude in a decade, but our capacity to phenotype animals has not. Improved phenotyping is where investment is currently needed.
To quantify the extent to which antipredator traits have been lost will require the development of ground-truthed behavioural assays that are easy to implement, and can be applied to a range of threatened mammal species by different researchers/organisations. In this project, our overarching objective is to develop behavioural assays that can be applied in field conditions on free-ranging animals. We will develop assays that build off monitoring techniques already widely in use (spotlighting; trapping; remote cameras). We will then test if the assays we design are repeatable across individuals, and sensitive enough to detect the behavioural differences that emerge between havened and non-havened populations.
Figure 2. A quenda enters a food station designed to measure latency to approach, behaviour on approach, and giving up densities, all of which are behavioural measures used to understand animal personality and response to predators (Image: Natasha Harrison)
We are working with two mammal species that occur in Western Australia: woylies, and quenda. Woylies are small (~1.2kg) bettongs that function as ecosystem engineers, and while recently down listed from Critically Endangered to Endangered, were prioritised for recovery in Australia’s first Threatened Species Strategy (2015-2020).
In contrast, quenda are not threatened, and are increasingly being translocated into urban and peri-urban habitats to provide ecosystem services (e.g. improve soil turnover, lower habitat flammability) and to restore lost biodiversity. Both species are excellent models because they occur at high density inside several havens, have populations that have been havened for different lengths of time (e.g. 2 to 20 years), have extant populations existing outside of havens, fall within the critical weight range (35 g to 5500 g) defining the continent’s most imperilled mammal taxa; and have close relatives that are also of conservation concern.
Figure 3. A woylie joey (Image: Natasha Harrison)
By developing a library of feasible, ground-truthed assays for these species, we expand the toolkit available for working on critical weight range mammals generally.
This project will:
- • Develop simple, repeatable behavioural assays for these species that measure both passive (neophobia, boldness), and active (vigilance, flight response) antipredator traits;
- • Compare behavioural phenotypes between havened and non-havened populations;
- • Communicate the results to conservation managers nationally.