2016 travel retrospective

I spent one-third of 2016 outside of Melbourne! Much of my travel was motivated by my work. Though I had lofty goals of blogging on the go, I didn’t progress beyond a single draft post. But once December rolled round I shared a slideshow of travel highlights with the QAECO lab – here’s a few annotated pics.

In June I visited Seattle for ISEC. QAEcologists Pia & Nick also attended, and we even got to catch up with QAECO alumnus Kim Millers! I took training in Bayesian model selection and R-NIMBLE, then attended just about every HMM session I could.
In the evening, we found time for baseball. Iadine Chadès was on hand to guide Pia, Kim & I through the rules.
I flew east in time to observe the 4th of July in Washington DC. I spent a couple of days at Patuxent Wildlife Research Center catching up with adaptive management mavens Sarah Converse & Mike Runge. I made a new friend at Resources for the Future in Becky Epanchin-Niell – we have abundant common research interests and I hope we’ll get to collaborate soon.
From DC, I took a bus and a small leap of faith in visiting another stanger. Sandy Liebhold is based at the US Forest Service in Morgantown WV and a partner investigator on the ARC Discovery Project I’m part of. He’s a forest entomologist, enthusiastic natural historian and generous host. It was great to get to know him and the gypsy moth program better. I was also unexpectedly smitten with the local culinary curiosity, salt rising bread.
Sandy invited me to stick around a further week to attend the IUFRO Workshop on Forest Invasions. It was a diverse mix of invasion and management science, policy and social science from across Europe, Africa, the Americas, Asia & Oceania. I learned a lot!
In September I was a guest at Kyushu University in Fukuoka, Japan. I was hosted by the wonderful Joung Hun Lee (who I first met at the University of Tokyo), and also benefited from an hour with the agile mind of Yoh Iwasa. I gave three presentations around the Japanese Society of Mathematical Biology conference and relished the abundant equations.
From Japan I flew to Perth for the Australasian Weeds Conference, which I wrote a bit about previously. I always appreciate the pragmatism of the work shared at this conference. I also spared a day to visit weed modeller Michael Renton over at UWA.

Travel is one of the great privileges of my job! In 2016 it exposed me to such a diversity of ecosystems, management challenges and research approaches. In 2017 I’m expecting to stay a little closer to home, transferring these insights to my own work.

The last friend I made while travelling – a quokka on Rottnest Island.

Malleefowl workshops, Perth & Mildura

PerthWorkshopMy flight from Tokyo had barely landed before I set off for Perth with the malleefowl adaptive management research team (including Mike Bode, José Lahoz-Monfort, Tim Burnard and Joe Benshemesh). We’re embarking on an ambitious landscape-scale experiment to help understand the role of foxes in malleefowl conservation. While we know that foxes do take malleefowl eggs, juveniles and occasionally even adult birds, is this predation a key threat to malleefowl persistence? And furthermore, is baiting an effective tool for mitigating any impacts that foxes may have on malleefowl?

IMG_8540_cropOur approach is to speak to land managers across Australia who are monitoring malleefowl populations. In Western Australia, they include mining companies, DPaW, an Indigenous Protected Area and NGOs like Bush Heritage Australia and the Australian Wildlife Conservancy.

We’re finding out who is baiting, who isn’t, and what sites are sufficiently similar to be compared. José’s initial analysis of historical malleefowl data suggests suggests that we’d need at least ten site pairs monitored over at least 5 years to pick up a plausible difference in malleefowl breeding between baited and unbaited locations. This workshop setting allowed us to get to know the specific sites, their management contexts and their malleefowl populations better, working out what kind of monitoring is feasible and potential barriers for inclusion.

MilduraWorkshopA week later we moved on to Mildura (this time with Rosanna van Hespen in tow) to meet land managers spanning New South Wales, Victoria and South Australia and representing OEH NSW, Parks NSW, the Western LLS, Parks Victoria, DEWNR SA, Federation University, the Australian Wildlife Conservancy, the VMRG and private properties. There were many more people and sites to get to know. We also had some important conversations about fox monitoring – how can we develop reliable indicators of fox density, activity or predation? Rosanna’s research project will offer important new insights into the capacity of motion-triggered cameras for detecting differences in fox activity across sites.

IMG_8607_colourArmed with all this new information, we can also refine José’s original power analysis to reflect the unique characteristics of each monitored malleefowl site. We’ll be working on ways to analyse data as they come in over the years, and new modules for the National Malleefowl Database to share what we learn with the malleefowl conservation community. We’re grateful that so many of its members are sharing their knowledge with us and entertaining the possibilities offered by this major collaborative project.

National Malleefowl Forum, Dubbo

MalleefowlWorkshopEarlier this month the 5th National Malleefowl Forum was held in Dubbo, NSW. Mike Bode, Libby Rumpff, Rosanna van Hespen and I were there to represent the QAEco malleefowl research team. In fact, we made the most of this gathering of experts and enthusiasts by holding a small workshop to develop our malleefowl ecosystem models further. Mike has been translating the network structures formed during the project’s first workshop into literally millions of possible models with varying interaction strengths. In this second iteration we sought to refine this set of models with our experts using their knowledge of the mallee ecosystem.

Mike proposed activities that centred around completing worksheets and evaluating graphs, and I must admit I felt skeptical that we’d effectively capture the interest and knowledge of our participants in this way. I was proven completely wrong – look at ’em all working studiously in that top photo! We made time for talking too, of course, but this original exercise ensured that everybody – not just the loudest or most confident participants – made a major contribution. I think the other great feature of Mike’s approach was that it first invited participants to reject any proposed models that looked implausible, rather than requesting that they compose relationships from scratch. No-one was stuck staring at a blank page, and the rejection exercise built up some familiarity with the kind of model outputs we are interested in. Our experts were in precisely the right frame of mind to create their own graphs and identify any gaps in our work during the afternoon group sessions.

MalleefowlForumThe Forum proper attracted over one hundred attendees from right across the country. Malleefowl are monitored nationally by volunteers, government agencies, mining companies, land holders, researchers and private consultants so the Forum is an important venue for communication across contexts. It’s a place to learn about malleefowl population trends across the country as well as conservation activities (such as captive breeding and feral animal control), technological developments (including LiDAR to detect mallefowl mounds and motion-sensing camera monitoring), Indigeneous knowledge and collaboration, and research findings (Taneal Cope’s genetic research and Jemima Connell’s fire modelling were highlights).

IMG_4251_colourcropNational co-ordinator Tim Burnard and malleefowl ecologist Joe Benshemesh cheekily summed up the Forum with a horrendogram that might have been inspired by our research team’s complicated ecosystem models. It did serve to highlight the diversity of skills, tools and institutions involved in malleefowl conservation across Australia – the Forum committee did well to capture them all into the same room at the same time.

IMG_4290_cropFlying from Dubbo to Melbourne involves a stopover in Sydney, and I snuck in a couple of hours at the Art Gallery of New South Wales on my way home. Here I was delighted to discover Mallee in halflight by Pablo Byass among the finalists for the 2014 Wynne Prize. The artist wrote that he eventually developed a fondness for the mallee through his mother. I have to credit an entire community of committed and knowledgeable mallee and malleefowl enthusiasts for sharing this landscape with me.

NCTC, Shepherdstown WV

IMG_1914_colourMy third travel week was spent at the National Conservation Training Center, the hub of professional development for the US Fish & Wildlife Service. Everything about the forested campus has been designed with sustainability in mind, and they’ve created an utterly immersive environment where you can lodge, eat, meet and think without distraction.

NCTCcampusI attended their Structured Decision Making Workshop, one of an impressive suite of Decision Analysis programs offered regularly at the Center. This involves developing rapid decision-analytic prototypes for solving a real environmental management problem over five days. With Fred Johnson and Christina Romagosa, I co-coached a team of managers and researchers concerned with the spread of Burmese pythons through Florida’s Everglades (and eventually) to Loxahatchee refuge.

PythonTeamThis is a role that demands constant thought and adaption – is everyone getting a say? have we really drilled down to the fundamental objective? what kind of model would best characterise this management problem? what activity can we next set the group to obtain the information we need? Luckily for us coaches, our team committed to the process, hatched some great ideas and made important contributions to our framing of the python invasion problem. Completing a full decision analysis will require extensive data collation, some thoughtful optimisations and simulations, and regular reality checks from our team of experts; Fred, Christina, Mathieu Bonneau and I are likely to stay on hand to see this project through.

IMG_1916_cropOn our last afternoon meeting as a group, team co-ordinator Rebekah Gibble presented each team member with a challenge coin as a token of appreciation from the Loxahatchee refuge staff. It’s a lovely keepsake that I’ll pin somewhere prominent in my office, and perhaps take out on occasion if I’m seeking a free drink.

Summer reading

PlanningLast week I was fortunate to attend a workshop in Queensland with colleagues from CSIRO and UQ, hosted by Iadine Chadès. It was a small and focused group skewed heavily towards the most quantitative environmental management research. Together we reviewed and classified adaptive management optimisation algorithms, from the foundational fisheries work of Carl J. Walters et al. in the 1970s and 80s, through the waterfowl hunting papers of the 1990s and 2000s to the recent contributions of our own research labs and the advances made in computer science that could determine our future directions.

Literature tableIt was a luxury to spend three straight days poring over the literature (not to mention the 5pm head-clearing beach visits). Most of us had several “Aha!” moments as we took the time to really interrogate the models and philosophies in articles we might have already read and cited several times before. Though they were severely restricted in computational capacity, early adaptive management researchers had some incredibly valuable insights into the nature and value of experimental management strategies; insights that we’ve sometimes forgotten or overlooked in subsequent work.

I’d encourage other researchers (and especially students!) to dig deep into the literature of your field – you’re bound to find some gems, and may save yourself from rehashing a problem already solved.

Adaptive management hall of fame:

Walters C.J. (1975) Optimal harvest strategies for salmon in relation to environmental variability and uncertain production parameters. Journal of the Fisheries Research Board of Canada 21: 1777-1784.

Walters C.J. & Hilborn R. (1978) Ecological optimization and adaptive management. Annual Review of Ecology and Systematics 9: 157-188.

Smith A.D.M & Walters C.J. (1981) Adaptive management of stock-recruitment systems. Canadian Journal of Fisheries and Aquatic Sciences 38: 690-703.

Nichols J.D., Johnson F.A. & Williams B.K. (1995) Managing North American waterfowl in the face of uncertainty. Annual Review of Ecology and Systematics 26: 177-199.

Johnson F.A., Moore C.T., Kendall W.L., Dubovsy, J.A., Caithamer D.F., Kelley, J.R. & Williams B.K. (1997) Uncertainty and the management of mallard harvests. The Journal of Wildlife Management 61: 202-216.

Parma A.M. & the NCEAS Working Group on Population Management. (1998) What can adaptive management do for our fish, forests, food, and biodiversity? Integrative Biology 1: 16-26.

Johnson F.A., Kendall W.L. & Dubovsky J.A. (2002) Conditions and limitations on learning in the adaptive management of mallard harvests. Wildlife Society Bulletin 30:176-185.

Runge M.C., Converse S.J. & Lyons J.E. (2011). Which uncertainty? Using expert elicitation and expected value of information to design an adaptive program. Biological Conservation 144: 1214-1223.

Shameless nepotism:

McCarthy M.A. & Possingham H.P. (2007) Active adaptive management for conservation. Conservation Biology 21: 956-963.

Hauser C.E. & Possingham H.P. (2008) Experimental or precautionary? Adaptive management over a range of time horizons. Journal of Applied Ecology 45: 72-81.

Moore A.L., Hauser C.E. & McCarthy M.A. (2008) How we value the future affects our desire to learn. Ecological Applications 18: 1061-1069.

Rout T.M., Hauser C.E. & Possingham H.P. (2009) Optimal adaptive management for the translocation of a threatened species. Ecological Applications 19: 515-526.

McDonald-Madden E., Probert W.J.M., Hauser C.E., Runge M.C., Possingham H.P., Jones M.E., Moore, J.L., Rout T.M., Vesk P.A. & Wintle B.A. (2010) Active adaptive conservation of threatened species in the face of uncertainty. Ecological Applications 20(5): 1476-1489.

Moore A.L. & McCarthy M.A. (2010) On valuing information in adaptive-management models. Conservation Biology 24: 984-993.

Probert W.J.M., Hauser C.E., McDonald-Madden E., Runge M.C., Baxter P.W.J. & Possingham H.P. (2011) Managing and learning with multiple models: objectives and optimization algorithms. Biological Conservation 144: 1237-1245.

Kangaroo-vegetation workshop, University of Melbourne

Workshop introduction
In May, José and I put our facilitation training into practice by hosting our own workshop at uni. The workshop fits into our ‘kangaroo project’, a collaboration with Parks Victoria to develop an adaptive management program for kangaroo control at Wyperfeld National Park. Kangaroo control is undertaken with the purpose of reducing grazing pressure and enhancing the regeneration of pine-buloke woodland. Our task is to build a model and help frame future management plans such that kangaroo management is explicitly linked to vegetation objectives and condition.

Small group workOur workshop gathered experts from the Mallee Parks, PV’s city office, academia and private practice to discuss how vegetation could be monitored and evaluated for both short- and long-term purposes. In the short term, we want the capacity to trigger actions that effectively protect regeneration. In the long term, we want to track progress in restoring the woodland and have a clear picture of the role of kangaroo management in that restoration. Our participants tackled these issues with gusto, bringing forth their many years’ experience in observing, researching and/or managing this system.

Workshop participantsJosé and I called on our training to plan a purposeful workshop and arrange activities that we hoped would engage and focus our participants. Some of these plans were highly successful, others were overlooked in the energy of the moment and a few fizzled. When our framing of an issue didn’t capture their perspective, our participants helped with a redefinition. We witnessed some terrific moments of self-organisation, with individuals reflecting and summarising what they’d heard from others, once or twice even pulling the group through a ‘groan zone‘ with little intervention from us facilitators.

That said, it was mighty challenging to draw on the facilitation techniques I’d learned while on my feet, in front of an (occasionally inattentive) audience. It helped enormously that I embarked on this workshop in partnership with José, so that leading and speaking could be alternated with reflecting and strategising. Several PV and university colleagues helped to form links and re-establish focus, while local students Hannah and Stefano assisted with note-taking and odd jobs (including the above photography!). Preparation and a post mortem with the facilitation discussion group helped us recognise challenges, develop strategies and identify paths for future improvement. This support network ensured that the first workshop José and I facilitated was robust in the face of our inexperience, and we’ll gain a little more foresight and confidence each time we take on this role.

Facilitators in training

Facilitation themes
An energetic, packed and deceptively well-organised schedule

I was fortunate to attend an EDG-funded Advanced Facilitation Skills workshop in Brisbane a month ago – there’s already a general report of this workshop on qaeco.com and I think Anna Renwick might have more to contribute in an upcoming issue of Decision Point. It’s given me a lot to think about.

An increasing number of QAEcologists have become accidental (and occasionally intentional) facilitators. If my path is at all representative, it seems as if we’ve burst from our scientific training with its structure and solutions, ready to solve the world’s environmental problems. But there’s rarely the precise kind of data we want to hand, or useful information is stored primarily in the heads of various ‘experts’. Politics and preferences come into play. Pure science isn’t even half the problem.

Even when we our scientific skill can contribute to the solution, we can’t expect to be seen and heard unless we’re willing to look and listen.  Understanding the context helps compose a feasible solution, building trust can open access to data and other resources, developing relationships and networks can connect us to those with power over the problem. It all takes interpersonal skills, not something us scientists (least of all a maths graduate like me!) are known for.

EDG researchers hone their skills at the Advanced Facilitation Skills Training
EDG researchers hone their skills at the Advanced Facilitation Skills training
(left to right: Tracy Rout, Kelly Hunt de Bie, Leonie Seabrook, José Lahoz-Monfort, Carissa Klein, Sana Bau)

It’s reassuring to know, then, that theory exists around some of these very issues. Adopting a technique such as active listening could enhance informal conversations, small meetings and large workshops alike. Reflecting initially on the four Ps (purpose-participants-process-product) would probably help focus and direct almost any gathering. Moreover, our trainer Mary Maher believes that all personality types are equipped to take on the role of facilitator!

That doesn’t mean that facilitation is insensitive to our personality type. Each of us can and should develop our own style, recognising our strengths and weaknesses. For example, many of have have an instinctive, nonconstructive response to stress. Facilitators will frequently be exposed to stress, conflict and low energy – in fact, they’re often necessary precursors to open discussion and creative solutions. Thus facilitators need to be ready to override that instinct, allow the difficult moments to happen and help their group move on to a more constructive, common understanding.

I learned a bit about my own style while participating in a facilitation role-play. Observers praised my encouraging demeanor and active listening. Yet, even in a role-play where I had no stake in the solution, I held my own perception of the problem structure and allowed it to interfere with an even-handed record of the discussion. When a role-playing participant voiced an idea that didn’t fit my implicit structure, I avoided it.  After this was pointed out to me, I remembered a moment in a real workshop that I’d done the same thing and saw it from this new perspective. ‘Performing’ in front of others was nerve-wracking but it was incredibly effective in revealing one of my persistent weak spots, allowing me to reflect on it and (hopefully) address it.

Better networked, ...
Trainer Mary Maher leaves us better informed, better networked, more motivated and more self-directing

We engaged in a broader discussion of impartiality in facilitation. Most of us find ourselves playing facilitator on projects where we are also an expert in some sense, perhaps as an ecologist or modeller. Often we don’t see ourselves as advocates pushing a specific solution but we might still have less obvious stakes in the direction a discussion takes (e.g. would that interpretation fit easily into my mathematical model?).

I have a hunch that the facilitators here at QAEco will revisit this issue often. And now we have a forum for such debates, as we’ve just set up a facilitation discussion group to continue sharing what we practise and learn. I was delighted to attract more than a dozen people to our first meeting last Thursday, where StefanoJosé and I outlined our purposes and plans for upcoming workshops. At our next meeting (at 3:30pm on May 2 in G27 – all welcome!), Jos will debrief on a workshop she recently led. I’m hoping we might use other sessions to explore the theory of facilitation further, too.

Facilitation is an exciting and intimidating new role to take on, and at QAEco I think we’re better resourced and more motivated than ever to master it.

Malleefowl workshop, University of Melbourne

Brendan elicits objectives
Brendan Wintle helps workshop participants extricate their fundamental objectives from their means objectives

After an illuminating weekend observing malleefowl monitoring practices, the University research team reciprocated by inviting around twenty mallee and malleefowl experts to one of our study sites, G26 Botany North. While the surrounds were not quite as picturesque, we did our best to compensate with comfy accommodation and abundant catering.

Our purpose was to develop a picture of malleefowl persistence with help from the people who know them and their environment best. We aimed to identify:

  • Objectives: What do we want for malleefowl? How will we know if we’re succeeding or failing?
  • Threats and Drivers: What processes influence our ability to achieve the objectives we have set?
  • Actions: What could be done to address the threats and drivers that negatively impact malleefowl?

We ran sessions of structured brainstorming in small groups and developed influence diagrams that connected our actions to threats and drivers, and our threats and drivers to our malleefowl objectives. Dynamics are complicated, with different processes operating at different temporal and spatial scales, and our diagrams often looked like spidery messes even as we made progress in ordering our thoughts and theories.

An influence diagram
An influence diagram in its inevitable ‘horrendogram’ phase

Since we’re embarking on an adaptive management project, we aim to embrace uncertainty. We encouraged our experts to disagree and to cast doubt; this will allow us to characterise our uncertainty, carry it through our modelling processes, and develop strategies that are robust to what we don’t know. Our experts filled us in on what data exist where, and for what interactions there is little or conflicting information.

Workshop group
Still stoically smiling after almost two days of modelling

Eliciting and ordering information from experts is hard work for everyone involved. It’s an art (sensing the mood of the room, maintaining morale and focus) and a science (obtaining relevant information in a meaningful format with as few biases as possible), and something that I’m very new to. A number of QAEco and ACERA researchers are expert expert-wranglers and I’m hoping to learn a lot from them as the malleefowl and kangaroo management projects progress.

Luckily for us, this cohort of experts had as much enthusiasm and stamina for the project as they did knowledge and data. The University research team is enormously grateful for their time, and we’re hoping to continue involving them in our work.

Adaptive management workshop, Brisbane

Terry Walshe deftly crafts a performance metric combining the best and worst ecological indices via summations, logarithms and even a cubed root.

In the last week of September Eve McDonald-Madden and Brendan Wintle brought together a couple dozen EDG researchers and friends to discuss adaptive management. The theory presented by Carl Walters and C.S. Holling in the late 1970s is eminently appealing to scientists and many decision-makers. How can we best manage (eco-)systems in the face of uncertainty? What’s our capacity to learn and reduce uncertainty as we manage? Are the risks of experimenting with our system outweighed by the long-term benefits of learning? (For a primer on adaptive management, you can check out these two articles that I wrote for Decision Point back in 2008.)

The “space and time” group ponder the advantages of spatial replication for experimental adaptive management.

We split into several groups, rearranging ourselves every half-day or so, knocking heads over different issues. Some of us focused on new analytical techniques for specific scenarios, others worked on simple and accessible representations of adaptive management principles, the malleefowl team developed strategies for an upcoming workshop, and on Friday morning we all reflected on the philosophy of adaptive management and why we so often fail to implement it.

I’m an avowed puzzle-solver, always attracted to a mathematically-framed problem, but I got a lot out of the broader-picture discussions that picked apart the assumptions underlying the adaptive management paradigm. There are clearly some (even many or most!) circumstances where it’s not the most suitable approach for environmental management, and we do it a disservice if we try to force every management issue into this framework. On the other hand, the process of framing a problem using adaptive management structures can make many of the conflicts, uncertainties and trade-offs at hand explicit. This can (I hope!) lead to better informed and more transparent decision-making even without fancy models and fully fledged adaptive management.

I’m sure there’s a Nature paper in here somewhere.