Thresholds and regime shifts in Australian freshwater ecosystems PDF Print E-mail

On this page you will find:




Project overview


Australian freshwater ecosystems have undergone considerable change over the past century. Widespread decline in ecological condition, loss of biodiversity and dramatic shifts in ecosystem composition and function have been reported for many habitats and geographic regions. The extent to which this change is irreversible, however, and how much of it reflects the inherent natural variability of these systems, is poorly known. Understanding whether or not species, ecological communities or processes have reached 'tipping points', beyond which changes are irrevocable, consequences dire and restoration unfeasible, clearly has significant implications for the management of our freshwater resources and assets, e.g. application of triage management approaches. Moreover, the ability to predict such ecological thresholds before they are reached will be vital for the conservation and management of these socially, economically and culturally critical ecosystems as we face a future characterised by increasing climate variability and ecological uncertainty. This Working Group will draw on a range of long-term ecological datasets for a variety of freshwater taxa, habitats and ecosystems from across the continent in order to address the following key questions:

1. Can comparable trends be detected amongst the changes in character of freshwater species, habitats and ecosystems that have been observed in Australia during the past century across geographic regions, ecosystem types or species?

2. Do any of the observed changes in the condition of freshwater species, habitats and ecosystems across Australian over the past century represent major shifts in ecological character or can they be considered within the bounds of acceptable variability?

3. Can long-term ecological datasets be interrogated so that past 'tipping points' can be identified and potential future thresholds predicted?

4. What management actions are appropriate in freshwater ecosystems where 'tipping points' have already been reached or where monitoring suggests that a ‘tipping point’ is being approached?


Dry stream bed in the Narran Lakes


Lake Mulwala half full


Photo by Mattinbgn (Own work) [CC-BY-SA-3.0 ( or GFDL (], via Wikimedia Commons


For further enquiries about this group either contact the Principal Investigator, Sam Capon



return to top


Workshop Reports


Workshop 1 Report (23-25 May 2011)

The first workshop for this ACEAS working group was held at the University of Queensland and attended by 13 scientists from around Australia, representing a range of research institutions and management agencies, along with Professor Declan Conway, a visiting fellow from the University of East Anglia. The major objectives of this first workshop were to: i) develop working definitions of key terms (i.e. ecological thresholds, tipping points, ecological regime shits etc.), ii) identify statistical techniques appropriate for use in identifying and predicting change points and state changes from long-term freshwater ecological datasets and iii) determine the availability and suitability of long-term datasets from Australian freshwater ecosystems for such analyses. In addition, this first workshop was structured so as to provide an introduction to the overall aims of the working group and its members and to facilitate the development of a detailed work plan for activities leading up to the second workshop planned for late November this year.

The workshop comprised three days of lively (and sometimes heated) discussions which centred on the controversial topic of whether or not there is any adequately supported evidence for stepped changes in ecological response variables or, rather, if the popularity of this ecological concept has been fuelled more by the use of ambiguous terminology, unreplicable experiments and the appeal of a general ecological theory (think physics envy)! Discussions were punctuated by numerous thought-provoking presentations by working group members including several scene-setting case studies examining long-term trends and decline in particular systems, e.g. Barmah Forest (Keith Ward), the Murray-Darling Basin (Bruce Chessman) and northern Australia (Chris Humphrey), and for specific ecological elements, e.g. waterbirds (Richard Kingsford) and seed and egg banks (Daryl Nielsen). A historical perspective, drawn from the paleolimnological record, was also provided by Peter Gell. Ralph MacNally instructed the group on the latest Bayesian statistical techniques developed specifically for the analysis of change points in ecological data and Matt Collof discussed lessons gleaned from recent projects regarding the collection, collation and analysis of ecological datasets.

Overall, the workshop was considered a success by those who attended and, despite some lingering differences of opinion, working group members are still talking to each other and even looking forward to reconvening in November. Prior to the second workshop, the major activities to be undertaken by the group will include: 1. a meta-analysis of published evidence of stepped change in freshwater ecosystems, 2. development of a metadata catalogue of long-term datasets for Australian freshwater ecosystems, and 3. statistical analyses of numerous long-term datasets held by working group members using Bayesian change point analysis techniques. The results of these activities will form the basis for discussion at the second workshop. Any queries about this working group’s activities or offers of relevant long-term datasets for analysis or inclusion in the metadata catalogue can be directed toward the group’s leader, Samantha Capon, by email to This e-mail address is being protected from spambots. You need JavaScript enabled to view it .


Capon Group


return to top


Workshop 2 Report (29 November - 1 December 2011)

The 'Glass half full' Working Group held its second workshop from November 29 to December 1 2011 at Linnaeus near Lennox Head. This workshop was attended by 12 participants, including Dr Jim Thomson of Monash University who was invited to join the group in recognition of his considerable expertise in Bayesian change point analyses. The workshop focused on four major activities:

1) A survey of metadata of long-term datasets pertaining to Australian freshwater ecosystems;
2) A review of approaches to long-term monitoring of Australian freshwater ecosystems and recommendations for future monitoring;
3) Preliminary findings of change point analyses on long-term datasets from the Murray River; and
4) A critical review of evidence for stepped change in freshwater ecosystems.


The first day commenced with a review of the working group’s aims and objectives, which had shifted considerably as a result of the first workshop, and the progress made on planned activities since. Professor Jenny Davis then provided a presentation on long-term patterns in the ecology of wetlands of both Western and central Australia with a particular focus on the applicability of the alternative stable states conceptual framework. Results of the metadata survey of long-term freshwater datasets were then discussed. Since responses to this survey were clearly limited and patchy, it was determined that a better tactic for assessing trends in long-term monitoring of Australian freshwater ecosystems would be to take a State-based approach that capitalised on the group’s diverse experiences. This formed the basis for the second activity (above) which is to be prepared as a manuscript for publication under the leadership of Nick Bond and Ralph MacNally.

On day two, results from preliminary change point analyses of long-term datasets made available by the Murray-Darling Basin Authority were presented by Jim Thomson and discussed by the group. The findings reflected the importance of consistent monitoring methods but also suggested some marked changes in response to driving variables flagged for future exploration. Further datasets, e.g. vegetation data from Barmah, were also identified for similar analyses.

The final activity undertaken during this workshop was to commence analysis of the database compiled by group members of published evidence for stepped change in freshwater ecosystems. Group members presented on the papers they had reviewed and a sub-set of papers was selected for critical appraisal. This preliminary inspection revealed very few papers with empirical evidence of stepped change in freshwater ecosystems and even fewer that met the criteria determined by the group for qualifying as a ‘regime shift’, particularly that of continued stability of a new state following removal of the triggering stressor. A process was determined for further population and analysis of this database under the leadership of Jasmyn Lynch, Jenny Davis and Sam Capon.

At the close of this workshop, it is envisaged that the working group's tangible outputs are likely to include: i) a review for publication of trends in long-term monitoring of Australian freshwater ecosystems with recommendations, ii) a manuscript describing results of change-point analyses for various Murray River datasets, and iii) a meta-analysis of published evidence for stepped change in freshwater ecosystems globally. The third workshop, to have a management focus, will seek to provide a management and policy interpretation of the working group’s deliberations.

Queries about this working group’s activities or offers of relevant long-term datasets for analysis or inclusion in the metadata catalogue can be directed toward the group’s leader, Samantha Capon, by email to This e-mail address is being protected from spambots. You need JavaScript enabled to view it .

click to enlarge
Back row L-R: Bruce Chessman, Nick Bond, Jasmyn Lynch, Daryl Neilsen, Keith Ward, Jim Thomson, Ralph Mac Nally, Max Finlayson. Front row L-R: David Hohnberg (standing), Peter Gell, Jenny Davis and Sam Capon.


return to top

Last Updated on Sunday, 08 February 2015 16:10