Resilience in social-ecological systems: Models and field studies


♪ Music ♪ In this lecture on the foundations and current applications of resilience theory, Dr. Steve Carpenter overviews the major concepts in historical evolution of resilience thinking. He explains that resilience theory came out of an understanding of adaptive cycles, which have a stable frontloop where changes are small or non-critical and a chaotic backloop where an accumulation of change eventually pushes a system to a tipping point and reorganization. He uses an example from a scenario planning project in Northern Wisconsin to highlight the need to identify potential critical transitions and backloops before they occur in order to build capacity in human systems to deal with uncertainty. He also highlights current thinking and writing from younger resilience scholars, which focuses on concepts and methods to better understand complex adaptive systems. These include maintaining heterogeneity and connectivity, broadening participation, and identifying slow variables to help manage the state of the system. He concludes by noting that more work needs to be done in characterizing the backloops of systemic change and highlights the use of scenarios and models to explore these changes before they occur. I was asked to talk about theory, and I had a lot of fun trying to think about how to talk about it. And in the end, I decided to just tell you about my conceptions of social-ecological systems as they evolved over the last 20 or so years. So I’m going to, for me, this started with a project called The Resilience Network, a little over 20 years ago, and I’m going to give you a very quick history of resilience thinking for social-ecological systems. I also have collaborated with economists and economics offers some particular tools for thinking about SES, and I want to talk about those a little bit. There is a component of resilience thinking called the backloop, which I’ll explain momentarily, which I think actually it, thinking about the backloop, I think is really the center of the mysteries of understanding social-ecological systems. I want to give you a current synthesis that I was not involved in by a group called the RAYS, which I think is having a huge influence in resilience thinking right now and is dominating a lot of the work that is going on around the world on resilience, so I want to just talk about that a little bit and then close with a few comments about field testing ideas, which I will continue tomorrow. So first, brief history of resilience. So in the mid-90s, Buzz Holling, who actually developed many of the ideas about resilience in a seminal 1986 paper, organized a group called The Resilience Network, which was an interdisciplinary group of about twenty people that he brought together for a series of meetings to think about concepts of change in social-ecological systems and the photos here show five people from that network and I could have picked any of the twenty, I suppose, but these are the five whose ideas I borrowed for the talk that I’m about to give you, so I wanted to give them some credit, and I’ve learned a lot from many meetings over many years and from my friendships with those five people. So the basic idea of resilience as we conceived it in the mid-90s was that there are two phases of change. There’s a sort of a routine change that we call the foreloop, or the frontloop. The frontloop moves from a growth phase nicknamed “R” for the famous logistic equation of population ecology to a conservation phase, which we nicknamed the “K” phase, again, by allusion to population ecology. And there are many models for routine change as a system develops, as plants colonize a new substrate, as a new business develops, as a new organization is formed. There is also a phase of change that’s very turbulent and complex and not so easy to understand; we call that the backloop. The backloop begins with a collapse of a highly developed system which releases resources in some form, maybe money, but maybe nutrients or carbon or space or whatever. And then eventually that leads to a reorganization that starts a new foreloop, so the contrast between routine change, or the frontloop, and turbulent change, or the backloop, is fundamental, so you put it all together, it’s something called the Adaptive Cycle, and that was the basic unit of change that we were using to organize our thinking in the mid-90s. An example from ecosystems is forest fire, so at the beginning, you have young trees, an open canopy, essentially no fuel. After a few hundred years you have a forest of old trees, a closed canopy, lots of dead trees on the ground, lots of branches, a lot of fuel lying around. Eventually, there may be a fire and the site is barren for a while, but eventually a reorganization begins, and that can come from sprouts of roots from root stalks of plants that were there before, from seed input, seed rain from neighboring sites, any number of sites, so ecological succession is a, kind of a simple example you can think about to remember the adaptive cycle. Of course, ecosystems are also organized hierarchically in space and time. This is a plot of the log of the space of a process versus the log of its turnover time in years. This is called a Stommel diagram after the oceanographer Henry Stommel, who introduced these in the 1930s to think about turbulence and pattern in the ocean. And since then, they’ve become fairly popular in ecology and it’s easy in ecological textbooks to find lots of diagrams that show that systems are organized hierarchically in space and time. For example, the green one is a forest, the smallest scale of needles, tree crowns, a patch of trees, a stand of trees, which is a larger unit, a forest itself, and a whole landscape composed of a heterogeneous mosaic of forest. The social scientists that we worked with in the mid-90s very quickly pointed out that similar sorts of space-time hierarchies could be identified in social systems, so scaling in terms of number of people and the persistence time of an entity is potentially useful for thinking about social systems. So this led to an elaboration of the adaptive cycle called the panarchy, so you can imagine multiple adaptive cycles at different spatial extents and turnover times, and they can be connected in various ways. Two of the kinds of connections that are particularly important are the revolt and remember connections. Anyone who has raised children will be very familiar with these, where the smaller scale entity at critical moments of collapse is actually able to induce catastrophic change in the larger scale entity through a spread, a contagious spread of the stresses. And on the other hand, during the reorganization phase of the smaller scale entity, elements of structure that are present at a larger scale may be remembered and influence that reorganization. Going back to the forest, the seed rain on a site may, in fact, be produced from a much larger spatial scale, and essentially, that’s an aspect of remembrance. So these were the basic ideas that we worked with and looking back on it, I think our biggest success with these ideas was the work that Frances Westley did on change in organizations and Frances organized a series of workshops around the world with environmental NGOs, working with those NGOs to understand change in their environments and change in their organizations using ideas of the adaptive cycle and panarchy and I was fortunate enough to be one of the teachers in a couple of those workshops, and here’s Frances leading a workshop in Uruguay last year. And one of the things we learned through Frances in this work with organizations, is what it feels like to be in an organization going through the different phases. So in the growth phase, people have converged on an approach or a process, a product, a question, an intellectual agenda in a science. The kinds of people that are involved are people who are implementers, organizers, team-builders, engineers. It’s an exciting time, a time of flow, high energy. Learning is rapid. There’s a sense of progress. As you move into the conservation phase, the system is near its peak production. In a science, this is the feeling like, “Wow, I’m having a hard time thinking of interesting questions in this area. I think I’ll go work on something else now.” Progress seems to be incremental. Engineers and managers do well. People who enjoy engineering and management do well in this phase. Innovators might be a little bit bored and looking around for something else to do. There is an experience of satisfaction, pride of accomplishment, anxiety about stresses on the system and the system losing momentum, so it’s kind of conflicted feelings in that sense. Eventually, that transitions to a release, because there will be a discontinuation or a breakdown of key processes in a business. A competitor might come in with a better product. In government, there may be a shift in the electorate, a shift in the mood of the electorate, and maybe very different kinds of politicians are beginning to get the votes. In science, people may vote with their feet and they may just decide, “Well, this old research agenda is something I’m not interested in anymore. I’m going to shift my intellectual energy and support into something else.” People who thrive on crisis are very excited and engaged by this phase, and people who miss the old way will mourn the loss. So it’s a time of anxiety. Changing relationships can be a lot of confusion in the organization. During the reorganization phase there’s a recognition that there is a need for innovation and a need for doing something else. There is a, it’s a time of meandering, loss of focus, experiments that may have few measurable outcomes for some time, but also it’s a time of deepening mysteries, and those can evoke excitement in science. People who love to play with uncertainty are happy here, entrepreneurs, innovators, and, I think, researchers actually, I think for many researchers, at least for me that’s, I’m happiest in that phase of reorganization. Lots of false starts, frustration. Occasionally, you figure something out and eventually it leads to a new growth phase. At about the same time Frances was running those workshops and I was occasionally participating, I developed a collaboration with an economist and a mathematician, William Brock, who is on your right, is a mathematical economist at UW Madison. John, Don Ludwig, on the left, is a mathematician from University of British Columbia and at the time we did the work, notions of alternate states, which are pretty prominent in ecology, had not really penetrated economics, and we felt that traditional economic benefit cost analyses were really missing something by not considering alternate states. So we wrote a series of papers on the economic implications of alternate state. We used eutrophic lakes as our lab rat for those modeling studies because I had a lot of data on eutrophic lakes available for calibrating models, so the basic idea is lakes have clear and turbid water. States, phosphorous inputs are typically the driver that is most likely to shift them to the turbid state. Phosphorous washes in from the landscape. It comes from excessive use of fertilizer or excessive production of manure in the Dairy State that I come from. The phosphorous ends up in the lakes, builds up in the sediment, eventually can tip the lake into an alternate state where there are high algae concentrations all the time, toxic blooms of cyanobacteria and other kinds of problems, so we were studying the economics of that rapid transition. And quickly summarizing three very long papers, the, you know, the general pattern is that optimal policies, economically optimal policies in a benefit cost sense, often add just enough pollutant to barely avoid crossing the threshold. And eventually, something random happens and you cross the threshold, so a lesson of the models is that mistakes are almost inevitable due to uncertainties and random events in the environment, and backloops are, therefore, going to happen, even in a system that you are trying to control with the best available information. We then embedded that analysis in a series of much more complex models of social-ecological systems working with a very skilled computer programmer named Paul Hanson at the Center for Limnology. And in all of these models, there were agents which are essentially individuals, say, individual farmers or recreators or landowners or whatever, making decisions based on rational beliefs and the information that’s available to them in the model, so there are hundreds of these computer people running around in silico. And the non-linear ecosystem dynamics are responding to human action and are measured imperfectly by the agents and the regulators. And all of these models produce some sort of looping structure like you see here with, you know, times of collapse as you see here in the back of the figure, and then these meandering periods of growth that you see in the front of the figure and if you squint, those look a little bit like the adaptive cycle. As a result of this work, we really began to center our thinking on the backloop. It seemed like we had pretty good models for normal development and growth. Ecologists and social scientists had been thinking about that for a long time and we had a lot of information to go on. It was this, the dynamics of collapse and the dynamics of recovery and reorganization following collapse that we had a poor understanding of actually both in ecosystems and in social systems. So our initial thinking about, and these are the kinds of things we would talk about in workshops in the early 2000s, were things like protect the wisdom, that you’re, the experience that you’re going to need to make wise choices. In other words, basically protect the memory that you need to reorganize. Experiment, but not in a dangerous and potentially catastrophic way, so try to find safe options for experiment, scales or modes of experimentation where if it fails it’s okay, but if it works, maybe there’s a big payoff. Build capacity for adaptation and talk a lot about complexity. Expand and communicate an understanding of change, try to get a lot of people and diverse people thinking about the complex problem that the system faces. About that time, I got interested in an approach called scenarios, which I’ll talk about more tomorrow. Scenarios are just one of many ways of organizing thinking in the backloop. Scenarios have many definitions; I’m using a very particular definition. I mean a set of plausible stories. It can’t be just one. You need multiple stories about how the future of a social-ecological system might unfold from existing patterns, new factors, and human volition, alternative choices that people may make. And Paul Raskin is one of the pioneers of scenario thinking. He did marvelous work at Stockholm Environment Institute in the mid-90s on scenarios, and he wrote an overview paper of the state of the art of scenarios in ecosystems in 2005, from which I’m taking that definition. And in my talk tomorrow especially, I’ll give you more concrete examples of scenarios. Scenarios, one advantage of scenarios is they bring everybody onboard, so unlike complex technological models which only engineers and scientists can think about, anybody, ordinary people, narrative writers, journalists, artists, anybody can think about scenarios and I came across this quote from the novelist John Barth in Sunday’s New York Times. He was talking about plot, and he said, “Plot is the gradual perturbation of an unstable homeostatic system and its catastrophic restoration to a new and complexified equilibrium.” That is the most concise statement of the backloop I have ever found. And he probably never knew what a backloop was, but he knew a lot about backloops and he organized his novels around them, so it’s easy to write stories about backloops, and that’s one reason that scenarios are a good way to go. Our first Scenario Project was in the Northern Highland of Wisconsin. The Northern Highland is the seeded territory, it’s the part of Wisconsin that was, that is managed, co-managed by native people’s treaty rights and by the state agency; it’s largely recreational, second growth forest, thousands of lakes, one of the highest concentrations of lakes in the world and the economy is largely recreation and forestry, and the economy is chronically in trouble and it’s an area with, that has a tremendous amount of conflict over resource management. And so we organized a series of workshops up there to, you know, to think about the backloop that the region seemed to chronically be in. And our approach at that time was, really brought in modeling in the early phases and so we developed a very complex model of the dynamics of the Northern Highland, which had forest management, it had lake shore zoning and lake shore management, it had fisheries in it, it had hunting, it had several dimensions of the economy, and it had a rudimentary sort of social interaction in it. And the, and it was set up as a game, so that it could be managed by setting various levers in the game that ran the computer program. So you could get a group of people around this computer program, say a realtor, a member of the Chamber of Commerce, a fishing guide, somebody who owned a resort, a scientist, what, you know, whatever you could find, and they would manage the thing together. And by observing that and working with them, we saw how people thought their way through computer-generated backloops and the model had certain fragilities in it, so you could enter a pretty catastrophic backloop in ten or fifteen minutes if you weren’t super careful in the way you managed the thing. And you could play a full cycle in half an hour or so. So in a two-hour workshop, you could go through many cycles of learning, and the individuals around the table could talk about what was going on with the system, so essentially, it was a way to do adaptive management with a computer program, and it’s a cheap experiment because you haven’t wrecked anything, you know, the worst that’s going to happen is you have to reboot the computer. So here’s an example of, for each color is a different cycle of a game that was actually played by a group of people in the Northern Highland and we collected a bunch of these. I just want to very quickly mention there’s a book calledResilience Thinkingand a later book calledResilience Practiceby Brian Walker and David Salt that are full of case studies of applied resilience thinking in different regions around the world. The Northern Highland of Wisconsin is one of the case studies in theResilience Thinkingbook. About the same time we were doing that Wisconsin project, I got dragged away into co-chairing the Scenarios Project for the Millennium Ecosystem Assessment, which essentially took all my time for about 5 years and is a huge exercise in scenario thinking, which I’m not going to talk about today, but I’d be happy to talk about it offline, but it is another model of that kind of approach, in this case a global one. I wanted to spend a few minutes talking about this new synthesis by the RAYS, the Resilience Alliance Young Scholars, because I think it might be useful to you in your thinking, so this project developed as work by about a dozen young people who went through their graduate-student and postdoc-hood working on this book as a side project. The leaders are Oonsie Biggs, who’s one of my former grad students who now runs a complex systems institute in South Africa; Maja Schlüter, who was a post-doc with Simon Levin, in the middle, who is an environmental economist at the Beijer Institute of Ecological Economics now; and Michael Schoon on the bottom, who’s a political scientist at Arizona State, and they develop seven, I think it’s seven, principles for building resilience in social-ecological systems, and I’m going to go quickly through those. I should have mentioned there’s, you can read the whole book, which is very easy reading, or if you need the Reader’s Digest condensed version, there is a paper in ARES that you can read pretty quickly. So the first point is maintain heterogeneity, and this is a, you know, fairly familiar argument in ecology related to the role of diversity and I believe it’s also fairly common in the social sciences as well. Heterogeneity is guaranteed to improve resilience under certain mathematical conditions in non-linear systems, according to a recent book by Scott Page. Manage connectivity, and you can’t just say make connectivity high or make connectivity low because epidemics spread through connectivity and fire is spread through connectivity, but also good things like beneficial species and good ideas spread through connectivity, so there, the correct level of connectivity or appropriate level is continually changing and it’s different for different things, so you just have to be thinking about it. Slow variables, the, many environmental problems are actually the result of ignoring slow change that then leads to catastrophic, unexpected effects and there are slow variables in societies as well because slow variables are not on the surface and not always apparent, it’s important, and they don’t change very much, so they look boring, it’s important, that’s what slow means, it’s important to recognize they’re there and bring them to the surface when you are thinking about how a system works. Foster complex adaptive systems thinking, that’s exactly what this immersion program is doing and what we’ll be doing the next few days. Encourage learning it’s, you know, the thing is, just about the time you think you understand how an ecosystem works, it doesn’t work that way anymore. And the same may be true of social systems, we’ll hear from the experts shortly, but because knowledge is always partial and incomplete and the system is always changing, you’ve always got to be probing to learn, and that’s always got to be on the agenda. You’re never done learning. Broaden participation. This is related to heterogeneity, but this is an explicitly social point that you need to build relationships and trust among key actors in the system in order for the system to function effectively and that involves broadening participation and doing the best you can there. And the final one is promoting polycentric government, governance, which I’ve actually read that chapter twice, and I don’t understand it, and so I’m going to hope that somebody else talks about that and I think somebody might, but it’s in the book anyway. If you want to read about it, go ahead. Field-testing ideas about resilience, I’m moving toward a wrap-up here. Field-testing ideas about resilience, I have, about 20 years ago, I published a paper saying, look, there are four ways to learn about ecosystems, and there’re theory and models, long-term observation, experiments, and comparative studies of ecosystems. And in fact, our most robust concepts about ecosystems are the ones that are supported by all four legs of the table, so the really strong knowledge in ecosystem science is supported by those four methods. And as a conjecture, I would suggest that the same is true for social-ecological systems. And there’s certainly a lot of theory and models for social-ecological systems. There are lots of longitudinal observations, observations of the social-ecological system for a long period of time. There are a lot of comparative studies, comparisons of different societies and the way they have interacted with their environments. Experiments are not quite the same as what an ecosystem scientist means by experiment. By, when I think about a lake experiment, I think about going in and ripping out a trophic level or going in and stocking a new species or changing the chemical composition of the water, and you can do that, but you can’t go into a society and just remove a socio-economic group, for example – I mean, you just – You can, but it’s not pretty. It’s not pretty, right. I mean, but nonetheless, I think you can learn a lot from perturbations that will occur in social-ecological systems that are just going to happen if you’re watching and you have the right controls in place when a perturbation occurs. I’m, this is one of my favorite quotes from one of the papers that influenced me the most as a graduate student, “To find out what happens to a system when you interfere with it, you have to interfere with it.” It’s not enough to just sit there and watch it. And so tomorrow, I’m going to talk with you about a complex systems interference project that we’re running now in the Yahara Watershed around Madison, Wisconsin, where we are developing an assessment of ecosystem services that’s actually designed to change the way people think about how the watershed works and I’m going to say more about that tomorrow, so I’ll just point out that it has a lot of boxes and arrow diagrams, and we did a lot of complicated stuff that I’ll tell you about tomorrow. And now, I’m going to give you a quick summary. So I’ve said a lot and I’ve covered 20 years of my learning and 20 years of research by a lot of people at a very superficial level, but I’d like to draw out a few points that I’d like to make sure you take home. One is routine expansion versus turbulent change is a key distinction. That’s the basic idea of the adaptive cycle and, in my view, the most important idea of the adaptive cycle. Both of these occur at the same time and at different scales and that gives managers and scientists opportunities. It gives managers opportunities to intervene and scientists opportunities to understand by paying attention to what’s going on at different scales. The backloop is the part of the adaptive cycle that we understand the least, and it’s highly influential and I think that is an area that needs a lot of research attention. We found a number of ways to accelerate thinking about backloops and long-term change. Scenarios are one of those that I mentioned. Computer games are another, because you can do experiments very quickly on computer games and I didn’t talk very much about other kinds of games, but there’s a big effort going on right now through the Stockholm Resilience Centre to develop role-playing games and interactive games for thinking about backloops very rapidly in groups of people, and I mentioned models. Engagement of researchers with the people who live in the social-ecological system is essential. I can’t think of any other way to do it, because they know more than you do about the system and you’ve just got to work with them. And currently, it seems like the RAYS seven principles in that book and that paper are providing a framework for organizing social-ecological research that a lot of people are getting excited about. And even though I was not involved in that book, I wanted to mention it to you as something for your consideration. Thanks very much. ♪ Music ♪

Leave a Reply

Your email address will not be published. Required fields are marked *