Tuesday, December 20, 2011


My strongest memory of New Zealand’s South Island, unfortunately, will be driving two fisted, white knuckled down narrow streets twisting over insanely steep mountains, and crossing one lane bridges. Did I mention New Zealand is a country where high octane fuel is available at every pump? I finally realized that the “keep left” signs weren’t meant for foreign tourists, but rather Kiwis gleefully taking shortcuts around right hand bends at 140 kph.
Q: Why did the Weka (a native bird that looks a bit like a long legged chicken) cross the road?
A: Unlike the chicken, the Weka didn’t have a good reason to cross the road, but like all good Kiwis he doesn’t feel comfortable unless risking life and limb on the highway.
Back to the Great Plains now, for the Xmas season, turning in final grades, and preparing for next semester’s classes. Have a great holiday season everyone!

Monday, December 19, 2011

Schooling again


It seems pretty clear that the term “Adaptive Management” has gone the way of “Sustainable Development” – it’s popular so everyone wants to do it, and as a result there is a proliferation of interpretations of AM. Rather than waste time arguing over whose interpretation of AM is the right and true path, Jaime McFadden and I tried to identify attributes of different interpretations, and then classify exemplars in the literature into a few, or in fact two, schools of thought. At the time, we had a 3rd category “other”, where we stuck everything that didn’t obviously fit in the other two. By nature I’m a lumper, not a splitter, so it causes me great pain, but I’ve concluded that there needs to be a 3rd school of thought on AM.
Credit for identifying this new school goes to Mike Runge, and the “Redefining Adaptive Management” symposium that we partially sat in on at the ICCB meeting. The key attributes are 1) having measurable objectives, 2) carrying out a management action intended to move the system closer to the objectives, and 3) effectiveness monitoring to determine if the system has in fact moved closer to the objectives, and if not 4) try something else. For the moment, I’m going to dub this the Foundations of Success school, after the consortium of international conservation organizations that put together the FoS umbrella, and the ICCB symposium. In the Artificial Intelligence literature they call this “trial and error learning”. I'll write some more about this later. 

Thursday, December 8, 2011

End of the story

I’m outta here, they are talking about effectiveness monitoring again. Although it is an interesting dataset - snares found per km walked in a forest park in Rwanda. Looks like they need to  control for observation effort - otherwise the ranger posts are attracting snares. 

Big Partnerships


Kim Lutz … or no, some other person talking about Environmental Flow Prescriptions: an adaptive partnership
This is a pilot project partnering with USACE on retiming flows from dams – gee, where have I heard that before.  They want to get both high peaks and subsequent low discharge periods as well. At least one of the projects has peak flows < 1 kcfs, so … although it looks like they include the Missouri River.
On the Savannah River, where to release, how much and when, what change to expect – phrased as a research project. They were surprised by fish response to a managed flow – warm water caused fish to move downstream – an example of learning! Trial and error, not AM.
On the Connecticut River they have management models of the system, a big one for the whole system, and a mini Stella model to interact with managers “hands on”. That’s cool. But no indication that the models make ecological predictions, or that they use them to predict the effects prior to choosing a strategy.
OK, so they've used this on some small rivers so far, not yet on the Missouri. 
Takehome lessons: translate between modelers and ecologists – OK, I’m not always easy to understand, I get it. Iterate stakeholder analysis – also a good idea.
Hmm, still not seeing any AM. 

Redefining Adaptive Management


Welcome to the first LIVE blog direct from the Epsom 3 room at the Skycity convention center in Auckland. This is a special session redefining Adaptive Management organized by Craig Groves and Jensen Montambault from The Nature Conservancy. Definitions are good, so I’m looking forward to adding a new school of thought to my pantheon. I’m sitting here with Mike Runge (USGS), waiting eagerly to hear how our lives will be different. Well, maybe that’s just me.
Mike has just offered a perspective that what non-decision theoretic AM people worry about are the unknown unknowns only – the surprises that are unanticipated.
Here comes Craig Groves.
Survey of AM from conservation measures partnership
Of 7000 projects, 5% of projects do the full cycle, although 2500 have plans.
Why? AM is too complicated, and there is no mandate from senior management.
Overcoming Barriers
Use risk and leverage to guide investments in AM. Invest in projects that are high risk, with potential to generalize to other projects. Need to have the best statistics to be able to say that things are actually happening the way. They have a nice little decision tree that leads to diagnosing when an experimental approach (AM?) would be needed.
Focus AM on addressing questions that managers need to answer – This seems obvious, but it isn’t clear that he means which decision to make. Mike says “Looks like evaluation monitoring”, and I agree. www.conservationgateway.org is the place to go for the details, apparently.
Stop reinventing the wheel – yes! There’s 60 years of literature on decision analysis! This is a good idea, collaborate with other agencies and analyze data across projects within the TNC – but not AM.
Get senior managers to support the idea – yep, hard to disagree with that. Another signal that it is evaluation monitoring in disguise, is that they “peer review” their plans for evaluating effects.
Summary
Not all projects need scientifically rigorous AM, some do not.
Training and tools matter but so does leadership
And we need more success stories.
It’ll be interesting to see how they define success? Getting around the Plan, Do, Check, Adapt cycle? He didn’t define AM :(. 

Tuesday, December 6, 2011

Congressing

I'm fortunate enough to be down in Auckland NZ this week for the International Congress on Conservation Biology. This is the first of the new bi-annual format meetings of the Society for Conservation Biology. So far, I've not been blown away by anything, but its fun to catch up with people. I've run into alot of people from my time in Oz, which is harder to do at conferences in North America. There was a good session yesterday on modelling future responses to climate change, which, for once, included a talk that expressed some skepticism of the utility of static species distribution models for this purpose (John Leathwick, NIWA). Walter Jetz (Yale), gave a remote talk describing his labs work on building models of every bird species on the planet - bold stuff (see www.mappinglife.org). There was some talk of testing predictions from these models, but no discussion of how much accuracy is enough, or of what these models would be used for. Luckily, I also saw Helen Regan's (UC Santa Barbara) talk where she laid a stochastic population model on top of climate affected future habitat distributions. That was sufficient antidote to residual frustration from earlier in the day. Although I worry about using downscaled point predictions of climate in this way - the uncertainty in these predictions is huge.