Wednesday, March 17, 2010

Out of the frying pan ...

... and into the fire! Last week was the AMCS meeting with everyone by and large on the same page with regards to AM. This week is the Missouri River Natural Resources Conference in Nebraska City, where AM was mentioned by nearly every plenary speaker, although I'm not sure there is a clear and shared understanding of what those words actually mean. One thing is clear - the Missouri River is the focus of one of the largest, if not the largest, ecosystem recovery programs in the country. And there is hope that AM will play a significant role in this effort.

A couple of key take home messages from this morning:
  • John Seeronen pointed out that AM does not excuse an agency from meeting regulatory requirements under other legislation.
  • Erik Blechinger gave a preview of some future guidance from the USACE on the overall goals of ecosystem recovery - it will seek strategies to simultaneously maximize economic, social and environmental outcomes. Hm. I thought it was well known that it is impossible to simultaneously maximize more than one objective.
  • Never, ever, use black text on a dark blue background for a powerpoint. Never. 'nuff said.
  • John Cooper, who has spent alot of time in DC and is a past president of the Association of Fish and Wildlife Agencies said that AM plans must deal with climate change, somehow.
John Seeronen has also prepared an excellent legal history of litigation in the Missouri Basin with respect to reservoir operations - useful background reading.

The afternoon was parallel sessions on BiOp actions and Pallid Sturgeon. I hung out mostly in the sturgeon sessions trying to learn something useful. What I learned is that there is a tremendous amount of information accumulating, what analysis is being done is fragmented across agencies, and none of it is connected to the decision making. And, despite some assertions to the contrary, decisions are being made.

Tuesday, March 9, 2010

Armwaving by Jim Nichols

Jim always says something that changes the way you think, despite the fact that he insists he knows nothing at all. In this case, he pointed out that ARM can deal with non-stationarity induced by climate change in at least two ways:
1) if we're smart enough to develop models of the trajectory (e.g. of rainfall or something else) then add a state variable to make it stationary in the expanded state space
2) if we're not smart enough, optimize over short time horizons with the expectation that you will revisit it frequently, and be careful about the state that you leave the system in.

Option 2 is the new one, that is both computationally manageable and perhaps palatable.

Assisted Migration

Eve Mcdonald-Madden from CSIRO and UQueensland in Australia talked about assisted migration, and in particular how long to spend learning before taking action in the face of climate change.

This work demonstrates the potential pedagogical power of simple models - articulating our beliefs about what is happening to carrying capacity in source and destination areas as a key part of the description.

Another key point Eve brought up is that even when she can construct optimal solutions to high dimensional problems, she can't represent them in a way that any human being can understand. In the social literature on decisionmaking this is known as "bounded rationality" - our limited sensing and cognitive capabilities prevent us from using the best possible solution.

Fred Johnson pointed out that this was a nice example of Ken Williams situation 3 - non-stationary dynamics, even from a simple perspective this model is helpful in framing up the question.

climate change and song birds

"How can I miss you if you won't go away" - old blues song.

Mike Conroy touched on the implications of climate change for song bird conservation - clearly this is in the adaptation to climate change situation, because the cat is out of the bag, frankly.

His key point is that we want to avoid spending large amounts on things that don't work, and that we need to think about many different scales - particularly large scales. He had some nice BBS maps showing how some of the long range migrants are decreasing at the southern end of their ranges, and increasing at the northern end of the range.

Some nice quotes:

"We can deal with epistemic uncertainty ... we predict from different models, do the averaging thing ... this is all in the book."

"Data in the short term cannot discriminate between different hypotheses" when one involves a tipping point.

"We might need to go back to the Walterian model rather than the incremental one we've been preaching for the last 20 years."

Climate change and AM

Ken Williams (USGS) introduced the session on system change and adaptive management with a brief talk on non-stationarity and climate change. He introduced three decision making situations:
Situation 1: stationary system dynamics with epistemic certainty (classic OR)
Situation 2: stationary system dynamics with epistemic uncertainty (AM optimization focuses here)
Situation 3: non-stationary system dynamics with unpredictable process change

This third situation is the new one introduced by climate change.

Key quotes
"Scenarios is a spoon type concept ... it is time to be sharp-edged not spoon-like"

ARM in wikipedia

ARM article in wikipedia - worth a look.

Fred Johnson (USGS) gave a bit of background (10 years of AMCS meetings). Tim O'Meara from the Florida fish and wildlife commission director is a big supporter of this mission, sponsoring the meeting and bringing on USGS to help directly through funding Fred and Bob Dorazio.

Dr. Steve Humphrey, director of School of Natural Resources and the Environment at UF, gave a welcoming address -including a great story of how the Interdisciplinary Ecology degree program got its name. "The first step to wisdom is giving things their right names" - chinese proverb. The word ecosystem meant different things to different people, and restrictive. So a new name was coined - social-ecological systems.

Quote for the day from Dr. Humphrey
"The most forceful, volatile, and manageable part of the social-ecological system is the human component."

Monday, March 8, 2010

Florida!

For the Adaptive Management Conference Series meeting in Gainesville. And, I've had that most marvelous experience of actually meeting someone who is reading the blog! Amazing! People other than my students actually do read this blog.

I'm looking forward to this meeting - its a small group of academics and agency folks who have gotten together annually to 'talk shop' - in this case stochastic dynamic programming and Bayesian updating mixed with waterfowl ecology and alligator harvesting. Its the sort of place where on one side of you someone is talking about heuristics for solving N-P hard combinatorial optimizations and on the other how many Manatees they saw on last week's survey flight. My kind of place, in other words!

Tuesday is a day of hard-core numerical optimization talks, Wednesday is a joint session with SouthEast Partners-in-Flight initiative, Thursday is a hands on workshop on AM for alligator harvest, and Friday is contributed talks including 2 by my crew. And one by me. Stay tuned for updates.

Saturday, March 6, 2010

Do you know Kahneman and Tversky?

If you don't, then (IMHO) you don't have any business asking people what they value and why in the context of making decisions under uncertainty.

Is disciplinary consensus a prerequisite for utility?

At a recent meeting relating to an AM program, I commented on the stark absence of social science expertise in the room. That particular program, which I'll not name to protect the not-so-innocent, isn't alone in this lack. Every single program I've been involved with in the last couple of years falls into the same category. One of the meeting participants, an engineer with experience in other AM programs, mentioned that a previous program had invited several sociologists to a meeting, but that the sociologists couldn't agree with each other about what the best way to proceed was. Much laughter - in which I participated, I confess.

Retrospectively however, I realized that I spend most of my time these days in meetings with ecologists who can't agree with each other. Or hydrologists. So obviously these disciplines aren't useful to AM either. Especially not on rivers. Engineers seem to be able to agree with each other at least, so I guess that's the answer - leave it to the engineers.

In fact, this perspective that disciplinary consensus is necessary for utility cuts straight to the heart of the issue - if there is consensus, then there is no uncertainty, and thus no need for AM. The fact that we think ecologists and hydrologists are necessary, but sociologists are not, suggests that we vastly underestimate the complexity of the socio- part of socio-ecological systems. We, rather arrogantly, assume that there should be consensus on how that works, and thus no need to incorporate that expertise into the process.

Whew. Glad I got that off my chest, its been bugging me for weeks!