Investigating the co-extinction of Australian insects on threatened plants

A project undertaken at The University of Melbourne, supervised by Peter Vesk, and implemented by Melinda Moir

Background

"The loss of biodiversity is the only truly irreversible global environmental change today” (Dirzo & Raven 2003. Global state of biodiversity and loss. Annual Review of Environment & Resources 28: 137-167). The least known mechanism of extinction is co-extinction: when a dependent species goes extinct because the host that they depend on at some stage in their lifecycle is extinguished (e.g., plant-dwelling insects with their host-plants). About 22,100 (7.4%) of the world’s 300,000 plant species are threatened and many native plants have at least one species-specific herbivorous insect species. Thus, the potential for co-extinction of many host-specific insects is high. However, species co-extinction remains a poorly studied area but one with critical implications to the global biodiversity crisis. A recent overseas modelling study suggested losses to global biodiversity through co-extinction may be high, but only when large numbers of hosts go extinct. However, estimates were constrained because there are no datasets with insects sampled from endangered plants. Thus, the accuracy of the estimates remains unknown, particularly in the Southern Hemisphere. This project will provide the most accurate estimate of co-extinction rates to date by sampling endangered plants and identify management strategies for sustaining viable populations of endangered insects, thereby conserving Australia’s biodiversity.

Practical Objectives

Broadly, this study will sample the insect fauna from plants with a range of threat status in two regions (New South Wales and global biodiversity hotspot, the southwest of Western Australia). We will use the samples to estimate the number of host-specific insects associated with the plant species. These data will allow refined estimates of extinction risk using refinements to published models. Then, we will build meta-population models to estimate the plant population size necessary to maintain a viable population of dependent insect species. This work will give us the tools we need to model rates of insect extinction across broad ranges of geography and taxonomy and to identify factors influencing an invertebrate species’ ‘extinction proneness’. The modelling will enable us to predict the number of plant-dwelling insects potentially co-threatened in Australia, and to determine the likely non-linear effects of plant population size on them. This innovative project will thus permit land managers charged with conserving Australia’s biodiversity to gauge the magnitude of the co-extinction problem, how it may vary with habitat loss, climate change and other threats, and to evaluate the conservation benefits for fauna of in situ plant conservation strategies.

Specifically, we will:

  1. Collect a dependent-host data set of plant-dwelling insects from plants across a range of threat status, phylogeny and localities in Australia;
  2. Estimate the number of plant-dwelling insect species within an assemblage that are host-specific to each plant species considered;
  3. Use these data to model extinction risk with an existing model, a refined model that considers extinction of some insects before hosts become extinct, and population viability analysis; and
  4. Provide land managers with recommendations on how to incorporate the conservation of co-threatened insects into management strategies for plant species which may be, or may become, threatened.
Project Team

Dr Melinda Moir (Post-doctoral Fellow – University of Melbourne)
Dr Peter Vesk (Supervisor – University of Melbourne)
Assoc. Prof. Lesley Hughes (Collaborator – Macquarie University)
Dr David Keith (Collaborator – NSW Department of Environment & Conservation)
Prof. Mark Burgman (Advisor – University of Melbourne)
Dr Mick McCarthy (Advisor – University of Melbourne)

For more information please contact Melinda Moir (mmoir@unimelb.edu.au).

Outcomes from the Project as at 30 June 2010
  • We have collected a dependent-host data set of plant-dwelling insects from plants across a range of threat status, phylogeny and localities in Australia;
  • We have identified and described key insect species that are potentially threatened with co-extinction and are notifying the relevant agencies;
  • We have designed new host breadth models to estimate the insect species most likely to be restricted to a threatened plant species and therefore at high risk of co-extinction;
  • We are currently estimating the number of plant-dwelling insect species within an assemblage that are host-specific to each threatened plant species considered. These data will be used to model extinction risk with an existing probabilistic model, a refined model that considers extinction of some insects before hosts become extinct, and population viability analysis; and
  • We have constructed a decision protocol for identifying co-threatened taxa and will provide land managers with recommendations on how to incorporate the conservation of co-threatened insects into management strategies for threatened plants.

 

Figure 1. Banksia coccinea (red banksia) in foreground and in the background the highly endangered B.Brownii (feather-leafed banksia)

Figure 2. Herbivore damage on the endangered plant Gastrolobium leutofolium.