Extinction risks of frogs under climate change PDF Print E-mail

On this page you will find:




Project overview


The current and future impact of climate change on biodiversity is poorly understood but is likely to result in large numbers of extinctions and substantial changes to ecosystem processes. Frog populations have undergone serious declines in recent decades and a third of species are now listed as threatened worldwide. Eastern Australia has been identified as a global hotspot of frog decline. Frogs are a functionally important group with critical dependence on climatic moisture. They are also exposed to threats including habitat loss, disease, predation and wetland degradation. While frog populations are thought to be highly susceptible to future climatic changes, little attention has been directed toward quantifying the risk of extinction due to climate change and the synergistic effects other threats pose. A common approach for predicting species’ responses to climate change uses bioclimatic envelopes. These models use present-day species-climate relationships to project the potential distribution of species under future climates, but are incomplete because they fail to account for important processes that influence extinction. Species responses to climate change will depend on complex processes involving physiological tolerances, population turnover, species interactions and landscape dynamics. Important advances that overcome the limitations of these models include the development of coupled-species modelling, which integrates mechanistic process models with bioclimatic habitat models, and mechanistic habitat models based on physiological budgets of energy and water. This working group will bring together leading ecological modellers, biologists and managers to apply this powerful new approach to Australian frog species. The working group will:


i) produce reliable projections using the most recent frog population and climate data with state-of-the-art modelling approaches;

ii) identify particular groups of species that are most at risk and the critical processes underlying those risks; and

iii) develop effective climate adaptation strategies that address threats to reduce extinction risks.


Litoria raniformis (Growling grass frog)

Litoria raniformis (Growling grass frog)


Principal Investigator: This e-mail address is being protected from spambots. You need JavaScript enabled to view it


return to top




Products and outcomes





FINAL REPORT available for download


Download PDF


return to top of Products and outcomes




Keith D.A., Mahony M., Hines H., Elith J., Regan T., Baumgartner J.B., Hunter D., Heard G.W., Mitchell N.J., Parriss K.N., Penman T., Scheele B., Simpson C.C., Tingley R., Tracy C.R., West M. and Akçakaya R. (2014) Detecting extinction risk from climate change by IUCN Red List criteria. Conservation Biology 28(3):810-819. DOI: 10.1111/cobi.12234


Penman T.D., Keith D.A., Elith J., Mahony M.J., Tingley R., Baumgartner J.B. and Regan T.J. (2015) Interactive effects of climate change and fire on metapopulation viability of a forest-dependent frog in south-eastern Australia. Biological Conservation 190: 142-153. doi:10.1016/j.biocon.2015.05.020





The group is developing a series of animations of predicted distribution change in Australian frogs. An example of the distribution for Litoria ranifomis (Keferstein, 1867) Common names: Growling Grass Frog, Southern Bell Frog.


Grid colour ramp (to which the legend corresponds): Habitat suitability at a 9975 x 9975 m pixel. This grid is the result of aggregating (summing) each 35 x 35 cell block of 285 x 285 m cells, each of which had a habitat suitability score of between 0 and 1. The cells shown here therefore have a theoretical maximum of 1225 (35 x 35).

Points: These indicate centroids of occupied patches of suitable habitat. The colour of points indicates their mean simulated population size at a given timestep, and ranges from white (abundance = 1 female) to black (abundance = maximum observed across timesteps and populations).

Grey grid pixels indicate the IBRA regions from which the Maxent background sample was drawn




return to top of Products and outcomes


return to top



Workshop Reports




Workshop 1 Report (25–29 July 2011)

The first meeting was held at the ACEAS meeting room at the University of Queensland from July 25th - 29th 2011. Fifteen people attended comprised of frog ecologists and managers, population modellers and biophysical modellers from various research institutions and government agencies around Australia. The overall aim of the working group is to utilize a recently developed innovative modelling technique that integrates mechanistic process models with species distribution models and mechanistic habitat models based on physiological budgets of energy and water to estimate risk of extinction under climate change and to explore and devise appropriate climate change adaptation strategies that minimize risk of extinction.

We spent a large part of the first two days describing and discussing the modelling methods, suitable species to use as case studies, assessing and verifying existing data, deciding on appropriate environmental predictor variables for spatial modelling, and identifying additional spatial layers to be prepared. We also spent a considerable amount of time discussing how to model important life history characteristics and responses to different threats, in particular how the chytrid fungus, a major threat to frog populations, can be modelled most appropriately under climate change. We identified fourteen frog species for modelling that differ in their geographic location, demography, current distribution, and their response to different threats. The remainder of the week was spent building species distribution models and population viability analysis models for each of the fourteen frog species. Participants split into smaller groups comprising of several frog biologists and were responsible for building population models for the frog species they were most familiar with, while the population modellers roved between these groups to assist in the modelling process and software. Spatial modellers concentrated on preparing additional spatial layers and then closely collaborated with frog biologists when developing the species distribution models for each of the species. This working group structure proved to be a very efficient way of building many detailed models in a very short time span, as the frog biologists were integrated into the modelling process directly and could discuss and resolve any issues regarding the species biology relatively quickly. The roving modellers were able to assist in resolving data analysis and modelling issues and together the modellers and frog biologists could determine whether the population dynamics of the species were being captured in the models appropriately.

Another small group comprised the biophysical modellers who concentrated on developing biophysical models for a couple of frog species in order to compare correlative spatial distribution models with mechanistic niche models. This group also worked towards developing spatial layers of changes in fecundity rates based on biophysical tolerances to climatic conditions across a landscape and under climate change. These layers can be linked to the population viability analysis models to incorporate a direct mechanistic relationship with climate. By the end of the first workshop we developed 14 species distribution models and 14 population viability analysis models that are at various levels of completion. Biophysical models for two species are also in development. By the next workshop these models will be ready for coupling and simulation under various climate change scenarios and for different management scenarios.

click to enlarge
Left to right: Matt West, Jane Elith, Tracey Regan, Michael Kearney, Nicki Mitchell, Dave Hunter, Trent Penman, John Baumgartner, Ben Scheele, Michael Mahony, Harry Hines, Geoff Heard, Chris Tracy and David Keith. (absent: Kirsten Parris, John Clarke, Brendan Wintle, Chris Simpson)

return to top


Workshop 2 Report (17 – 21 September 2012)

Our second workshop was held at the Women’s College at UQ from September 17th-21st.  Fortunately most of the working group members were able to attend and we had the addition of Reid Tingley, a post doc from the University of Melbourne.

We hit the ground running and after a brief update on progress and approach we leapt straight into the modelling (excuse the frog pun!). We broke into small groups, including one that concentrated on refining the species distribution models (SDM), others that focused on finalising the population viability analysis models (PVA) for individual species, and a small group concentrated on developing a generic biophysical model that could be used for all frog species. This group also focused on searching the published and grey literature for appropriate data to parameterise the biophysical model for the case study species.

A large amount of data processing and manipulation was required so that the spatial models would be in an appropriate format for integration. But thanks to the prowess and several late nights, one of our R modellers, John Baumgartner, developed a script to automate and streamline the process. This made the integration process highly efficient. By the middle of the week we had ironed out all the remaining technical issues and began integrating the SDM models with the PVA models under different climate change scenarios. Several scenarios to investigate the influence of climate change and fire on the persistence of the giant burrowing frog were simulated during the workshop and results discussed among the group. The workshop was intense but highly productive. By the end of the week we had completed SDM models and PVA models for thirteen the frog species and integrated several of them. The generic biophysical model requires some further refining but is anticipated to be ready to parameterise for different frog species shortly.

During the week we spent several hours discussing how best to incorporate the chytrid fungus, a major threat to frog populations, into our climate change models. We assessed an existing SDM model for chytrid to determine its appropriateness as a threat layer. After identifying several data gaps in the model, it was decided to try and fill the gaps and update the model using the spatial layers we have been utilizing, as well as any new locality data available. This required personally contacting frog specialists around Australia for any further data on chytrid locations. This work is on-going.

We identified several outputs and possibilities for publication of work emerging from this workshop. Examples of some of the outputs currently in progress are:


- Draft manuscript titled: “How soon can the IUCN Red List criteria of Threatened Species detect risks posed by anthropogenic climate change?”
- Conference presentation by Trent Penman at the Australian Society of Herpetologists annual conference: “Climate change, fire and one giant frog”
- Draft manuscript on the role of connectivity in the extinction risk of frogs under climate change.
- Synthesis paper focusing on whether particular life history traits make some frog species more susceptible to extinction under climate change than others.


click to enlarge
Extinction risks of frogs under climate change Workshop 2 (September 17–21, 2012)
Attendees left to right: Tracey Regan (University of Melbourne), John Baumgartner (University of Melbourne),David Hunter (NSW Office of Environment and Heritage), Matt West (University of Melbourne),  Michael Kearney (University of Melbourne), Harry Hines (Queensland Department of Environment and Resource Management), Geoff Heard (University of Melbourne), Reid Tingley (University of Melbourne), Mike Mahony (University of Newcastle), Chris Tracy (Charles Darwin University / University of Melbourne), Trent Penman NSW Department of Primary Industries/The University of Wollongong), Nicola Mitchell (University of Western Australia), Chris Simpson (NSW Office of Environment and Heritage), Jane Elith (University of Melbourne), David Keith (University of NSW)


return to top

Last Updated on Wednesday, 15 July 2015 00:18