Workshop Summary

The first meeting of the working group on Optimal and robust monitoring for conservation and natural resource management was held at Freycinet Lodge, Freycinet National Park, Tasmania in August 2007. The overall purpose of the working group was to further explore the role of decision theory in monitoring and adaptive management and to develop some case-studies based on current State and Commonwealth agency research and management needs.

The key workshop objectives were:

1. Address some key questions about monitoring design within a decision theory framework and develop some motivating case studies based on current agency needs.
2. Explore the role of monitoring in dealing with severe uncertainty in decision making.
3. Instigate new, and enrich existing collaborations between agencies, AEDA CERF, and other academy researchers in the area of monitoring and adaptive management.
4. Identify major research priorities arising out of agency monitoring and adaptive management needs.
5. Propose ongoing projects to service agency monitoring needs, improve monitoring practices and contribute to the development of monitoring theory through publication in the scientific literature.

The workshop commenced with a series of short presentations (10-15 min) by each participants. Twelve working group topics were identified from individual presentations and participants formed subgroups around each topic. The outputs from the workshop deliberations will take the form of improved management and monitoring practices through ongoing collaboration between CERF and agency researchers and managers, and published research findings, based on detailed case-studies that will advance monitoring and adaptive management theory and knowledge. Overall the optimal and robust monitoring working group was extremely enjoyable and already productive in terms of the increase in Hub-Agency awareness and the number of collaborative links that have emerged.

Convenor: Dr Brendan Wintle

Background

The focus of the workshop is on optimal/robust monitoring. A central role of monitoring is to inform management aimed at continuous improvement by iteratively updating information about the effectiveness of actions or the magnitude of their impacts (Walters 1986, Possingham 2001). Monitoring is frequently identified as a central element in sustainable resource and environmental management in academic literature (Meffe & Carol 1994, Burgman & Lindenmayer 1998), public policy documents (Commonwealth of Australia 1992, 1996(a,b)) and environmental impact assessments (Environment Protection (Alligator Rivers) Act 1978, Environment Protection and Biodiversity Conservation Act 1999). Despite its prominence in the management vocabulary, surprisingly few documented examples of its successful application exist. There is little evidence that current approaches to monitoring have led to better environmental outcomes, timely remediation of environmental impacts or improvements to management efficiency (Mulder et al. 1999, Wintle et al. 2005). The only certainty is that monitoring is expensive and has a high opportunity cost in light of the finite resources allocated to conservation and environmental management.

    Monitoring has failed to influence management in the way it should, not because it isn’t important, but because managers and scientists have failed to implement a decision framework that explicitly links monitoring results to management decisions (Possingham 2001, Gerber et al. 2005). The questions to be addressed by monitoring have been poorly defined and its ability to change management actions is not clearly stated. Furthermore, monitoring programs are often inadequate for discerning relevant changes (Possingham 2001) and the standard statistical inference methods are often inappropriate for dealing with monitoring data. Sub-optimal monitoring leads to economic losses through inefficient or wasteful expenditure on monitoring, and to serious, yet avoidable environmental impacts resulting from a failure to detect impacts or act on trends in a timely manner.
    Active adaptive management (AAM) is a decision theoretic approach to monitoring and management that enables managers to value learning explicitly within the context of maximizing management objectives (Shea & Possingham 2000; Nichols & Williams 2006). AAM shows great potential for improving the efficiency of monitoring design and management performance. However technical and theoretical problems explored to date are not readily transportable to realistically complex management problems characterized by severe uncertainty. Recent publications on the topic highlight the value of closely integrating management and monitoring, introducing economic efficiency, and utilizing formal decision frameworks (Field et al. 2004, Gerber et al. 2006, Nichols & Williams 2006). A major challenge is to find a practical, general framework that can be applied routinely by managers, and that is sufficiently robust to the severe uncertainty typical of environmental monitoring problems.
    In this working group, we hope to move beyond the commonly asked question: “how many samples do I need to detect a particular change or impact”, to address the more salient question: “how much information do I need to make a good decision?”.

Workshop proceedings

We followed an NCEAS working group (http://www.nceas.ucsb.edu/) format that we have found to be particular productive in recent years, and that we have adopted in a number of recent domestic working groups. We commenced with a series of short presentations (10-15 min) by each participant. Participants were invited to nominate potential working topics. Working group topics arise from individual presentations. Topics were synthesised at the end of the formal presentations (see Working group topics below). Individuals were asked to nominate themselves to work on particular topics and subgroups were formed around those topics. A series of subgroup meetings were scheduled to allow individual topics to be more fully developed and work on nominated topics to commence in earnest. Subgroups then worked on their allotted topics for the remainder of the week (see agenda for more details of work program). The full group re-convened periodically to allow subgroup presentations on preliminary outcomes and canvassing of opinions and advice from the whole group. Group membership was somewhat fluid with some people choosing to work between several groups. The pattern of subgroup work and group presentations continued until the final day when a full group meeting concluded proceedings and identified ongoing work required to complete sub-group projects.

Working group topics

12 working group topics for identified for detailed investigation throughout the week. Collaboration on each of these topics will be ongoing, resulting in specific management recommendations and, where possible, published outcomes. This section provides a plain English outline of the specific problems that were prioritized and developed throughout the week of the workshop and that will form the basis of ongoing collaborations. This is by no means an exhaustive list of the optimal/robust monitoring research being undertaken by the AEDA CERF and/or its collaborators. While most topics were discussed by all members of the working group, subgroups (listed under each topic title) hold responsibility to progress the problem.  Subgroup coordinators are identified with an asterisk.

1.    How much effort should be invested in monitoring NE Tasmanian devils to see if the species can persist with the disease or recover from the disease?
2.    Combining monitoring methods to identify marine pest outbreak.
3.    Maximising the efficiency of mitigation measures through optimal learning about their effectiveness: DFTD culling trials.
4.    Efficient monitoring of the unknown unknown.
5.    Locating a disease front with imperfect detectors.
6.    How intensively should we monitor state-wide Tasmanian devil population trends to support our management planning?
7.    Optimal quarantine and surveillance
7a    Which is the best strategy for establishing wild insurance populations given risks of disease invasion, costs of eradication, quarantine and surveillance?
8.    Eradication of invasive species to protect island biodiversity
9.    Using species lists to detecting changes in abundance.
10.    Are ‘learning organisations’ the missing ingredient for successful long-term, environmental monitoring?
11.    Endangered species recovery: when to manage and when to monitor
12.    Monitoring and reporting on the performance on investments in vegetation restoration
13.    Robust replication of biological surveys

1. How much effort should be invested in monitoring NE Tasmanian devils to see if the species can persist with the disease or recover from the disease?

Menna Jones (DPIW, Tasmania), Clare Hawkins (DPIW, Tasmania), Tracy Rout* (University of Melbourne), Hugh Possingham (University of Queensland).

DFTD first occurred in Devils in NE Tasmania and has since spread over much of the island.  The devil population in NE Tasmania is now so low spotlighting is an unreliable/useless method for monitoring the population.  However we still know there are devils present through intensive ($300,000 per annum?) monitoring.  Is there any value to continuing this investment in monitoring a population which is now almost gone?  Yes – if the devil population stabilizes at a low level with the disease then this significantly diminishes the urgency for other conservation actions.  Indeed if devils and their disease can coexist at low levels we may abandon more expensive and intrusive conservation efforts and focus on developing a vaccine.  So it looks like that knowledge is valuable, but is it worth $300,000 per year?  How long, and with what intensity should DPIW monitor the declining NE Tasmania population?

2 Combining monitoring methods to identify marine pest outbreak.

Jan Carey* (University of Melbourne), Cindy Hauser (University of Melbourne - ACERA), Mark Antos (Parks Victoria), Yakov Ben-Haim (Technion, Israel), Gerry Maynes (Commonwealth NRM), Eve McDonald-Madden (University of Queensland), Joslin Moore (University of Queensland), Dorian Moro (Chevron), William Probert (University of Queensland), Michael Runge (USGS, Patuxant).

Our marine reserves are under attack from exotic plants and animals. These invaders can endanger our unique native species. Early detection of any new invasion is vital to limit its impact. There are many different ways to detect a new invader. We construct a decision model to help the park manager choose the best combination of detection methods when many aspects of this problem are uncertain. The model describes the effectiveness of the each of the methods for each species, the chance that each possible pest species will be transported to the reserve and the impact if the pest was not detected and takes hold. This model identifies the amount of funding needed for reliable and timely detection of an invasion.

3. Maximising the efficiency of mitigation measures through optimal learning about their effectiveness: DFTD culling trials.

Eve McDonald-Madden (University of Queensland)*, Micheal Runge (USGS, Patuxant), Cindy Hauser (University of Melbourne), Tracey Rout (University of Melbourne)*, Joslin Moore (University of Melbourne),William Probert (University of Queensland), Menna Jones (University of Tasmania and DPWI), Peter Vesk (University of Melbourne), Brendan Wintle (University of Melbourne)

Environments are under pressure from many threats, including habitat loss, climate change, invasive species and diseases. Environmental managers work to reduce these threats. What makes this difficult is the urgency of the problems we face, and our limited understanding of how the environment works. As a result, when we make decisions in the face of this doubt, there is a chance we may not implement the best action to manage our natural resources. However, every action we take provides the opportunity to learn about the ecosystem, if we monitor the outcome of our actions appropriately.  Since some actions provide more information than others, there is sometimes a tension between doing what we think is best for the environment, and doing what will allow us to learn the fastest. How do we balance these two goals, so that in the long-term, our actions protect and enhance the natural world?  The approach to such a problem is called ‘active adaptive management’.   

One threat that has surfaced in the last 5 years has been the fatal disease faced by one of our iconic Australian species, the Tasmanian devil. Rapid decline in the number of Tasmanian devils has required an urgent response. But such urgency leaves little time to learn the best thing to do in the face of this completely novel disease. We are building a framework to evaluate methods for disease suppression, taking into account both the need to protect the population in the short term, but also learn quickly so we can control the disease and conserve the species in the long term.

4. Efficient monitoring of the unknown unknowns

Brendan Wintle* (University of Melbourne), Mike Runge (USGS, Patuxant), Sarah Bekessy (RMIT University), Adrian Manning (ANU CRES), Mark Antos (Parks Victoria), Clare Hawkins (DPIW Tas), Yohay Carmel (Technion, Israel), Yakov Ben-Haim (Technion, Israel).

Strong arguments have been posited that investment in monitoring should always focus on targeted (management-driven) approaches in order to spend the scarce conservation dollar efficiently.  It is argued that management focussed monitoring avoids wasting effort on collecting data that is unlikely to inform specific management objectives.  Yet, as in homeland security, background surveillance can detect important ‘unknown unknowns’ that could otherwise lead to catastrophic outcomes.  The recent outbreak of the Tasmanian devil facial tumour disease is an example of such a catastrophic event in the conservation world.  We build a model that allows us to explore allocation of investment to background surveillance versus focussed monitoring of the environment given different frequencies and costs of unanticipated threats.  We identify under what circumstances it may be rational and defendable to invest in less focussed “survellience” monitoring (sensu Nichols & Williams 2006). We find that when the probability of detecting unknown threats is included, together with the cost of their impact, an optimal allocation of effort could include background surveillance in some instances.  The framework highlights the necessity for conservation managers to think strategically about the costs and benefits in monitoring allocation decisions.

5. Locating a disease front with imperfect detectors.

Michael Bode* (University of Melbourne), Brendan Wintle (University of Melbourne), Clare Hawkins (DPIW Tas), Menna Jones (DPIW Tas).

Since Devil Facial Tumour Disease (DFTD) emerged in the early 1990s, the Tasmanian devil population is estimated to have declined by more than 50%, with a decline of 90% in the area where this infectious cancer was first reported.  One of the ways that The Tasmanian State Government's Save The Tasmanian Devil program is addressing this threat is to establish insurance populations, in captivity and, perhaps in future, on islands.  It is important to avoid including infected animals in these insurance populations, but difficult to ensure this. The addition of an infected devil into a captive population would risk wasting husbandry costs prior to the emergence of DFTD signs. The addition of an infected devil into an island population would jeopardise the entire costly action of establishing such a population.  However, if devils are only collected from a limited area furthest from the known edge of DFTD distribution, genetic variation in the insurance population is limited. The latent period, between time of infection and time when DFTD signs (tumours) appear, is unknown, so that one cannot be sure that a tumour-free devil is DFTD-free.  Furthermore, at the edge of the disease front, an extremely low proportion of devils will be infected so that a trapping trip in this region will be very unlikely to catch a devil with DFTD signs. Funds have been allocated to improve knowledge of the disease front position, but the decision of where to focus intensive checks for DFTD presence is complex.

6. How intensively should we monitor state-wide Tasmanian devil population trends to support our management planning?

Tracy Rout* (University of Melbourne), Hugh Possingham (University of Queensland), Menna Jones (DPIW Tas), Clare Hawkins (DPIW Tas), Adrian Manning (ANU CRES), Sarah Bekessy (RMIT), Brendan Wintle (University of Melbourne).

The Tasmanian State Government's Save The Tasmanian Devil Program was established to address the threat of Devil Facial Tumour Disease (DFTD). Since the emergence of this infectious cancer in the early 1990s, the Tasmanian devil population is estimated to have declined by more than 50%, with a decline of 90% in the area where DFTD was first reported.

Devil population trends have a significant effect on resource allocation and management choices of this Program. Full population recovery would indicate that the government should entirely abandon the Program. If population extinction occurred, the Program would wind down the costly insurance population projects and reintroduce these populations to what would now be a disease free environment. Changes in rates of declines would also affect allocation of resources into relatively long- or short-term acting management options. Choosing the wrong management option, such as reintroduction into an area where DFTD is still present, could be very costly. The cost of an appropriate level of monitoring to distinguish between these trends will be balanced by the benefit of avoiding such an error.

7. Optimal Quarantine and Surveillance

Joslin Moore* (University of Melbourne), Tracy Rout (University of Melbourne), Chris Wilcox (CSIRO Marine), Dorian Moro (Chevron), Menna Jones (DPIW Tas), Clare Hawkins (DPIW Tas), Jan Carey (University of Melbourne), Yakhov Ben-Haim (Technion Israel), Hugh Possingham (University of Queensland)

How much money should an organisation spend on stopping rats from invading Barrow Island?  How much effort should we put into checking that DFTD has not entered one of our disease free populations?  How much money should we spend looking for an invasive sea star in Hansen Bay, or should we enforce recreational boat cleaning? Invasive species, like rats, cats and weeds, threaten many species on Australia’s offshore islands.  Some of these threatened species occur nowhere else in the world.  Some islands do not have these invasive species, yet.  On other islands we have successfully removed one or more of the invaders.  For these pest free islands how much effort should we put into quarantine, reducing the risk of a species invading?  Alternatively we could spend money on surveillance, looking for the pest on the island with the view of eradicating it before it gets out of control.  We have solved the general problem of how much to spend on quarantine and surveillance and we illustrate the solution using a case study.

7a) Which is the best strategy for establishing wild insurance populations given risks of disease invasion, costs of eradication, quarantine and surveillance?

Menna Jones (DPIW Tas), Joslin Moore* (University of Melbourne), Tracy Rout (University of Melbourne), Chris Wilcox (CSIRO Marine), Yakov Ben-Haim (Technion Israel), Hugh Possingham (University of Queensland).

The highest priority response to the possible extinction risk for wild Tasmanian devils is to establish “insurance populations” in places that can be isolated from the disease. It is essential that some of these populations are wild-living to retain natural behaviours and ecological adaptations. It is also essential that some of these wild-living populations are in Tasmania so that the parasitic, pathogenic and commensal fauna, some of which is endemic, is also protected. Captive animals evolve adaptations to captivity and parasitic faunas are substantially altered.

There are four types of wild-living insurance populations that could be established. All have different socio-political issues that affect likelihood of their establishment, biosecurity risks of disease invasion, quarantine costs (and methods), surveillance costs, and costs of eradication. Costs and benefits to the local ecosystem of the introduction or maintenance of devil populations also differ. The four types are:
•    Offshore islands (currently all free from devils)
•    Large (100km2) fenced reserves on mainland Australia
•    Large peninsulas in Tasmania (site isolation maintained by water barriers enhanced by fencing and devil-resistant road barriers)
•    Currently disease-free
•    Currently diseased; disease suppression used to eradicate disease

What is the optimal allocation of effort to these types of wild insurance populations? Parameters can be estimated as an extension of the disease suppression trial parameters.

8. Eradication of invasive species to protect island biodiversity

Chris Wilcox* (CSIRO), Joslin Moore (University of Melbourne), Michael Bode (University of Melbourne).

Non-native mammals are responsible for most vertebrate extinctions over the past six centuries, the overwhelming majority occurring on islands.  Three-quarter of seabirds listed by the IUCN are threatened by non-native species, compared to 49% by fisheries, the next most common threat.  Increasingly, eradication of invasive species, particularly invasive mammals, has been demonstrated to be feasible and to have large impacts on the persistence of native species.  However, moving beyond an opportunistic island-by-island approach to addressing this threat to a national or ocean basin scale approach necessitates developing a prioritization system which can determine the most effective locations for actions amongst the myriad of islands, threats, and endemic species.  Decision theory provides a wealth of tools for addressing complex decision-making, such as prioritizing investments in biodiversity conservation.  These methods have only recently seen application in ecological problems, although there is some history of use in the context of reserve selection problems.  In this project we will develop a prioritization system which explicitly considers socio-political constraints, in addition to direct economic costs and biodiversity benefits of eradications, for finding the greatest biodiversity return on financial investments in conservation.

9. Using species lists to detecting changes in abundance.

 
Judit Szabo* (University of Queensland), Chris Wilcox (CSIRO Marine), Peter Vesk (University of Melbourne), Peter Baxter (University of Queensland), Yohay Carmel (Technion, Israel), Oscar Ventar (University of Queensland), Yakov Ben-Haim (Technion, Israel), Hugh Possingham (University of Queensland).

Simple lists of species are probably the most abundant form of biodiversity information available. Naturalists have long recorded the plants, birds and other animals they have sighted or collected and potentially these data may tell us about changes in species abundance and distribution. However, lists have significant shortcomings as data. At the least a list contains two pieces of information, the record of the species and the list length—how many species were recorded. The list length may reflect the collection effort—important information for qualifying the absence of a particular species from a list. Better lists provide information on location and survey dates. Franklin provided an example of potential use of such lists for detecting declines in seed eating birds of northern Australia. We are exploring the potential for wider use of his approach. We are exploring three lines of enquiry: (1) the robustness and sensitivity of Franklin’s method for data of variable quality; (2) analysis of the sampling design required to detect future declines in Atlas surveys; (3) a calibration of community bird data records using high quality data to provide a basis for broad scale monitoring. This research will enable us to use historical data to monitor long-term trends of species, and to better use community monitoring effort to monitor future changes.

10. Are ‘learning organisations’ the missing ingredient for successful long-term, environmental monitoring?

Adrian Manning (ANU CRES), Mark Antos (Parks Victoria), Sarah Bekessy* (RMIT), Gerry Maynes (Commonwealth NRM), Brendan Wintle (University of Melbourne).

In recent decades there has been increasing interest across many domains in the concept of ‘learning organisations’ as a way of harnessing individual learning for the benefit of the organisation at large. A ‘learning organisation’ is one that ‘continually expands its capacity to build its own future’. Two key purposes of becoming a learning organisation are to maintain flexibility in the context of rapid change, and in order to enhance the capacity to innovate. Key features of learning organisations include: open systems (ie. integrated) thinking; improving individual capabilities; team learning; updating mental models (ie. overcoming personal assumptions) and a cohesive organisational vision.  

The key characteristics of learning organisations are directly relevant to the conservation paradigm of active adaptive management and long-term, robust environmental monitoring. Active adaptive management promotes the use of management as a source of learning, which in turn can inform subsequent actions.  Learning from long-term monitoring has many strategic advantages, including detecting important ecological trends or emerging threats before they become too serious and providing evidence that management is effective.  Despite the obvious benefits and the inherent simplicity of the concept, examples of effective adoption of AAM are elusive.  The gap between rhetoric and reality is partly due to institutional barriers that prevent learning and the difficulty of developing management objectives that are clear, widely agreed and based on appropriate logic. Consequently, the creation of learning organisations could potentially help overcome these barriers by facilitating the initiation of long-term, robust monitoring programmes, and by allowing organisations to respond rapidly to information derived from monitoring. Further, in combination with the adoption of strategies which permit the development of clear, explicit objectives, this approach could represent the missing ingredient needed to create successful long-term, robust environmental monitoring programmes.  

In this paper we define a learning organisation in the context of monitoring, describe the development of a coherent vision for active adaptive management and outline the key characteristics of an organisation that will be able to support such a vision.  We explore the applicability of this model to the 54 natural resource management regional bodies that have recently been established across Australia with the task of establishing monitoring and evaluation strategies.

11. Endangered species recovery: when to manage and when to monitor

Oscar Ventar*(University of Queensland), Hugh Possingham (University of Queensland).

Endangered species recovery plans usually identify the main threat facing the species. However there is always some uncertainty whether the threat is actually endangering the species. A manager is then faced with the decision of whether to implement threat management actions based on this prior belief in the threat, or instead to invest in what may be an expensive monitoring scheme to gather more data before deciding how to proceed. Here we formulate a problem that clarifies when to act and when to monitor with the goal of minimizing the net expected cost.

12. Monitoring and reporting on the performance on investments in vegetation restoration

Brendan Wintle* (University of Melbourne), Ted Lefroy (University of Tasmania), David Duncan (ARI, Victoria), Gerry Maynes (Commonwealth NRM), Peter Vesk (University of Melbourne), Hugh Possingham (University of Queensland), David Lindenmayer (ANU CRES), David Freudenberger (Greening Australia).

The Australian Commonwealth and State governments make substantial investments in landscape restoration through CMA funding and other initiatives. Ongoing investment will depend largely on perceived performance of current investments in achieving restoration goals. Assessing the performance of restoration investments is difficult because there are usually multiple objectives of restoration and those objectives are often poorly defined, or intangible and difficult to measure. Perceived benefits of vegetation restoration include improved social and economic well being and ecological integrity, all of which are notoriously difficult to measure. This project aims to utilize decision theory to identify a coherent approach to measuring vegetation restoration success and to assist investors evaluate the relative performance of their investments.

13. Robust replication of biological surveys

Kirsten Parris (University of Melbourne), Michael McCarthy (University of Melbourne), Jan Carey (University of Melbourne)

Decisions about the level of survey effort required for biological surveys require decisions the number of sites and the number of replicates per site. For a given budget of resources and time, allocating more replicates to each site reduces the number of sites that can be surveyed. This trade-off in the survey design has been optimised under the assumption that some of the parameters that are to be estimated during the survey are known. However, these parameters are usually highly uncertain. We examine how to design surveys that are robust to uncertainty in the parameter values.

References

Burgman MA., Lindenmayer DB. 1998. Conservation biology for the Australian environment. Surrey Beatty, Chipping Norton.
Commonwealth of Australia. 1992. National Forest Policy Statement. A new focus for Australia's forests. Canberra.
_____. 1996a. Australia: State of the Environment. Department of Environment, Sport and Territories, Canberra.
_____. 1996b. National strategy for the conservation of Australia's biodiversity. Canberra
Dorazio RM., Johnson FA. 2003. Bayesian inference and decision theory-A framework for decision making in natural resource management. Ecological Applications 13:556-563.
Field SA, Tyre AJ, Jonzen N, Rhodes J, Possingham HP 2004. Minimizing the cost of environmental management decisions by optimizing statistical thresholds. Ecology Letters 7:669-675.
Gerber LR, Beger M, McCarthy MA, Possingham HP 2005 Optimal monitoring of marine reserves Ecology Letters 8:829-837.
Meffe GK, Carroll CR. 1994. Principles of conservation biology. Sinauer, Sunderland, MA.
Mulder S, et al. 1999. Marbled Murrelet effectiveness monitoring plan for the Northwest Forests. PNWGTR-439. U.S. Forest Service.
Possingham HP. 2001. The business of biodiversity: Applying decision theory principles to nature conservation. TELA Report, Melbourne. Weblink: www.acfonline.org.au/uploads/business_of_biodiversity.pdf
Shea K., Possingham HP. 2000. Journal of Applied Ecology 37:77-86.
Walters CJ. 1986. Adaptive management of renewable resources. MacMillan, New York, New York, USA.
Wintle BA, SA Bekessy et al. 2005. The utility of dynamic landscape metapopulation models for sustainable forest management Conservation Biology 19:1930-1943.

Workshop Participants

   
Adrian Manning (ANU CRES) email:    adrian.manning@anu.edu.au
Cindy Hauser (UniMelb) email:  chauser@unimelb.edu.au
Clare Hawkins (DPIW Tas) email:  clare.hawkins@bigpond.com
David Lindenmayer (ANU CRES) email:  davidl@cres.anu.edu.au
Dorian Moro (Chevron) email:  DMMV@chevron.com
Eve  McDonald-Madden (UQld) email:  e.mcdonaldmadden@uq.edu.au
Gerry Maynes (NRM Team) email:  Gerry.maynes@nrm.gov.au
Hugh Possingham (UQld0 email:  h.possingham@uq.edu.au
Janet Carey (UniMelb) email:  j.carey@botany.unimelb.edu.au
Joslin Moore (UniMelb) email:  joslinm@unimelb.edu.au
Judit Szabo (UQld) email: j.szabo@uq.edu.au
Kirsten Parris (UniMelb) email:  k.parris@unimelb.edu.au
Mark Antos (Parks Victoria) email:  mantos@parks.vic.gov.au
Michael Bode (UniMelb) email: mbode.web@gmail.com
Mick  McCarthy (UniMelb) email: mamcca@unimelb.edu.au
Mike Runge (USGS, USA) email:  mrunge@usgs.gov
Oscar Venter (Uqld) email: oventer@uq.edu.au
Peter Baxter (Uqld) email:  p.baxter@uq.edu.au
Peter Vesk (UniMelb) email: pvesk@unimelb.edu.au
Sarah Bekessy (RMIT) email: sarah.bekessy@rmit.edu.au
Tracy Rout (UniMelb) email: t.rout@pgrad.unimelb.edu.au
William Probert (Uqld) email: willprobert@gmail.com
Yakov Ben-Haim (Technion, Israel) email:  yakov@techunix.technion.ac.il
Yohay Carmel (Technion, Israel) email: yohay@tx.technion.ac.il
Brendan Wintle (UniMelb) email:  brendanw@unimelb.edu.au
Menna Jones (DPIW Tas) email:  Menna.Jones@dpiwe.tas.gov.au
Chris Wilcox (CSIRO) email: chris.wilcox@csiro.au
David Salt  (ANU CRES) email:  dsalt@cres10.anu.edu.au
Nick Beeton (Utas) email: beetonn@utas.edu.au
Ted Lefroy  (Utas) email: Ted.Lefroy@utas.edu.au
Joanne Potts  (UAberdeen, UK) email:  joanne@mcs.st-and.ac.uk