Foundations of Ecological Resilience

By Lance H. Gunderson and C. S. Holling

Ecological resilience provides a theoretical foundation for understanding how complex systems adapt to and recover from localized disturbances like hurricanes, fires, pest outbreaks, and floods, as well as large-scale perturbations such as climate change. Ecologists have developed resilience theory over the past three decades in an effort to explain surprising and nonlinear dynamics of complex adaptive systems. Resilience theory is especially important to environmental scientists for its role in underpinning
adaptive management approaches to ecosystem and resource management. Foundations of Ecological Resilience is a collection of the most important articles on the subject of ecological resilience—those writings that have defined and developed basic concepts in the field and help explain its importance and meaning for scientists and researchers. The book’s three sections cover articles that have shaped or defined the concepts and theories of resilience, including key papers that broke new conceptual ground and contributed novel ideas to the field; examples that demonstrate ecological resilience in a range of ecosystems; and articles that present practical methods for understanding and managing nonlinear ecosystem dynamics. Foundations of Ecological Resilience is an important contribution to our collective understanding of resilience and an invaluable resource for students and scholars in ecology, wildlife ecology, conservation biology, sustainability, environmental science, public policy, and related fields.

• Language: English
• Publisher: Island Press
• Release date: Jul 16, 2012
• ISBN: 9781610911337

Book preview

article.

INTRODUCTION

Why Resilience? Why Now?

LANCE H. GUNDERSON AND CRAIG R. ALLEN

LANCE H. GUNDERSON , Department of Environmental Studies, Emory University, Atlanta

CRAIG R. ALLEN, Nebraska Cooperative Fish & Wildlife Research Unit, University of Nebraska, Lincoln

PERHAPS ECOLOGY, rather than economics, should be called the dismal science because the popular application of the information generated by ecologists is generally bad news. Since the 1960s, prominent ecologists have been referred to by the popular press as the New Jeremiahs, because some have interpreted their ecological research of human impacts on ecosystems as prophecies of doom and gloom. Indeed, many of their successors continue the trend of documenting the increasing size and magnitude of the human footprint on the planet. A recent Millennium Ecosystem Assessment (2005) determined that global and regional ecosystems have been altered by human activity more in the past fifty years than at any time in history, a trend that is likely to continue.

As we begin the twenty-first century, the rate (and spatial scale) of ecological change is accelerating. Across the planet, natural disasters are severely affecting individual and collective societies. Over time, humans adapt and learn how to cope with these events. Yet the abrupt and often unpredictable dynamics associated with these events have led to greater uncertainty in spite of technological and scientific advances. An increasing human population and anthropogenic land use and land cover change have left humankind more vulnerable to these events (Kasperson et al. 1995) and the potential loss of ecosystem goods and services (Vitousek et al. 1997, Millenium Ecosystem Assessment 2005). The root of the word disaster literally means bad star, suggesting an extraplanetary origin of these events but perhaps better describing the large uncertainties associated with them. Such large-scale disasters include outbreaks of disease, such as influenza (Barry 2004), AIDS, and hantavirus; recurring tsunamis (Adger et al. 2005); tropical cyclones, such as Hurricane Katrina in 2005 (Bohannan 2005); and global climate change (Steffen et al. 2004), to name but a few.

Resilience theory (Walker and Salt 2006) was developed by ecologists over the past three decades to explain surprising and nonlinear dynamics of complex adaptive systems (Gunderson and Holling 2002, Walker et al. 2004). Moreover, resilience theory is the basis for adaptive management, which embraces uncertainty of complex resource systems (Holling 1978, Walters 1986, 1997). The following section presents a review of the concepts of resilience as developed by ecological theorists and applied ecologists.

Ecological Resilience

Development of ecological resilience theory began in the 1960s with attempts to mathematically model dynamic ecosystems. Early models (Lewontin 1969, May 1977) focused on the stability of systems—that is, the processes that control or mediate the persistence of ecological structures. Part of the construct of modeling systems was to create boundaries that identified the system. In doing so, processes were categorized as internal or external to the system, with a focus on interactions between the external and internal. Disturbances were viewed as external drivers or perturbations. Human interventions, such as the harvest of natural resources (e.g., fishing, lumbering), were included in the category of disturbances to an ecological system. A major assumption for theory and practice was that ecological systems were stable. The stability assumptions rested on two additional assumptions: (1) that a system would generally persist in form and function (unless, of course, humans perturbed it) and (2) that a system would recover to its former equilibrium state after disturbances. Implicit to this was the notion of global or stable equilibrium, such as population or ecosystem carrying capacity. Although rates of recovery might vary, perturbed systems would eventually recover to a predistur-bance state. When applied to renewable resources, these theories led to policies of maximum sustained yield, or harvest rates that assumed whatever was being harvested would be replaced. Hence, the idea of optimal control of harvests was introduced and persists.

Holling (1973) introduced the word resilience to describe three aspects of changes that occur in an ecosystem over time. The first was to describe the persistence of relationships within a system and the ability of systems to absorb changes of state variables, driving variables and parameters, and still persist. The second concept recognized the occurrence of alternative and multiple states as opposed to the assumption of a single equilibrium and global stability; hence, resilience was the size of a stability domain or the amount of disturbance a system could take before it shifted into alternative configuration. The third insight was the surprising and discontinuous nature of change, such as the collapse of fish stocks or the sudden outbreak of spruce budworms in forests. These insights altered the way in which theorists perceived ecological systems and how practitioners have attempted to manage these systems.

The community of ecologists has viewed and defined resilience in multiple ways, each based on different assumptions. One interpretation of the idea of resilience is rooted in the etymology of the word, which has been traced to the Latin word resilire, meaning to leap back. Some ecologists (Pimm 1991) define resilience as how fast a variable that has been displaced from equilibrium returns to that equilibrium. This is essentially a return time, or time of recovery, which can be mathematically defined but is based on an assumption of behavior around a single equilibrium. Ludwig et al. (1997) assert that this perspective is applicable to a linear system or to the behavior of a nonlinear system in the vicinity of a local equilibrium that can approximated by a linear function.

Holling (1996) distinguished two types of resilience: engineering and ecological. Engineering resilience is defined as the rate or speed of recovery of a system following a shock. Ecological resilience, on the other hand, assumes multiple states (or regimes) and is defined as the magnitude of a disturbance that triggers a shift between alternative states (Holling 1973, 1996). In this sense, a regime shift occurs when the controlling variables in a system (including feedbacks) result in a qualitatively different set of structures and dynamics of these systems (Walker et al. 2004). Whether or not alternative regimes or states exist in ecological systems has been the subject of debate over the past three decades.

One of the first demonstrations of alternative states and regime shifts came from an international modeling project conducted by the International Institute for Applied Systems Analysis (IIASA) on the outbreaks of spruce budworms in the boreal forests of Canada (Ludwig et al. 1978, Holling 1978, Clark et al. 1979). This project modeled pest outbreaks by using a small number of variables (budworm population, predation effectiveness, and the volume of forest canopy) and the interactions or relationships among these variables. The model depicted two states. One state (no outbreak) is characterized by low numbers of budworm and young, fast-growing trees, and the other state (outbreak) is characterized by high numbers of budworm and old, senescent trees. The shift between regimes occurs when the growing trees result in an increase in canopy volume such that the set of birds that eat budworms can no longer control the pest populations (Holling 1988).

A similar model has been used to explain the outbreak of such human diseases as influenza, hantavirus, or malaria (Janssen and Martens 1997). One state is an outbreak state, when the infection is rapidly spreading among susceptible humans. The other is a dormant or inactive state, when either no or few cases of infection are present. As documented in many cases of regime shifts, the transition among states or regimes is mediated by the interaction between slower and faster components in ecosystems (Holling and Gunderson 2002). In this model, the faster variables include the population or numbers of disease organisms, the slower variables include susceptibility of the hosts and disease vectors, and the slowest variables are the mutation rates of the disease and the size of the human population (May 1977). Some of these factors, such as disease vectors and host densities, may be managed by humans, whereas others, such as mutation rates, may not.

The demonstration that ecosystems exhibited alternative states or regimes gained momentum in the 1980s and 1990s, but not all studies supported the idea. Sousa and Connell (1985) examined a multiple-decade time series data set of marine populations and found no evidence for alternate regimes or states. Other ecologists, working in disturbance-driven ecosystems, found that resilience theory was the only theory that helped explain the complex changes that they were studying. Walker (1981), Westoby et al. (1989), and Dublin et al. (1990) studied semiarid rangelands and found dramatic shifts between grass-dominated and shrub-dominated systems. Those shifts were mediated by interactions among herbivores, fires, and drought cycles. Scheffer (1998) and Scheffer et al. (2001) described two alternative states (clear water with rooted aquatic vegetation and turbid water with phytoplankton) in shallow lake systems. Coral reef systems may shift from a coral-dominated state to a macroalgae-dominated state (Hughes 1994, McClanahan et al. 2002, Hughes et al. 2003, Bellwood et al. 2004). Multiple pathways have been documented for this transition, including overfishing and resulting population declines of key grazing species, increases in nutrients, and shifts in recruitment patterns (Jackson et al. 2001). Estes and Duggins (1995) and Steneck et al. (2004) have shown how near-shore temperate marine systems shift between dominance by kelp or dominance by sea urchins as a function of the density of sea otters and other grazers.

Regime shifts have been observed in hundreds of different ecosystems, including marine, freshwater, and terrestrial (Gunderson and Pritchard 2002, Scheffer et al. 2001, Folke et al. 2004). In all of these systems, the transitions among regimes, and the resilience of the system, can be traced to a small number of variables, including biological and physical controls and recurring larger-scale perturbations (or disturbances). A key insight is that ecological resilience is mediated and lost due to the interaction of variables that operate at distinctive scales of space and time (Holling 1986, Holling and Gunderson 2002).

Resilience and Scale

Complex ecological systems operate across wide ranges of scales of space and time. An example is global influenza outbreaks, as occurred with the Spanish flu in the early part of the twentieth century (Barry 2004). These outbreaks involve structures and dynamics that range from interactions among microscopic organisms to the scale of the planet. Populations of bacteria in host organisms operate on time scales of minutes to weeks, suggested by the normal time course of the disease within a host. Numbers of infected or susceptible humans can literally cover the planet. The many pathways (especially air travel) by which humans move can accelerate rates of spread, so that the disease can flare up in one place and within days be spread around the planet. As argued by Barry (2004), ignorance of the vectors of spread lead to greater concentrations of humans and larger outbreaks.

Another example of a natural disaster that indicates the cross-scale nature of dynamics is from Florida Bay, a shallow, subtropical marine ecosystem located at the southern terminus of the Florida peninsula. For most of the twentieth century, the bay had clear water and a bottom covered by sea grass. Around 1990, the sea grass began dying over most of the bay. The die-off was viewed as an ecological crisis and created great political, social, and economic turmoil. Because much of the bay is in Everglades National Park, the social objectives of conservation in the park were brought into question: Would the grass return? What (if any) management actions led to the die-off? Sport fishing and tourism relied on the clear-water state of the bay. Many wealthy people (including the U.S. president at that time) used the bay for recreation, so the crisis became instantly politicized.

The sea-grass die-off resulted in the system flipping from a clear-water, grass-dominated system to one with muddy water and recurring algae blooms. Living sea grass plants store nutrients, and their root systems stabilize sediments. The loss of sea-grass released nutrients into the water column and allowed sediments to become suspended in the water column by wind-generated currents. But much wrangling and discussion went into trying to understand what caused the shift in ecological regimes.

A number of hypotheses were proposed to explain the system shift (Gunderson 2001). These included hypersalinity resulting from a decrease in freshwater flow and altered water circulation, an increase in nutrient inputs from surrounding urban and agricultural areas, a lack of hurricanes, a loss of herbivores (turtles and manatees), disease, and temperature stress. The most plausible explanation involved a spatially homogenous stand of high sea grass biomass (probably related to a lack of disturbances, such as storms and grazers). The stress caused by high temperatures caused local die-offs as photosynthesis could not produce enough oxygen to keep up with respiratory demand. Because the beds contained high biomass, the die-off spread as dead material in the water column further depressed photosynthesis. Without sea grass, the sediments and nutrients became suspended in the water column, leading to algae blooms and muddy water. Both of these factors inhibited subsequent sea grass establishment. Hence, loss of ecological resilience (the amount of disturbance that the system can absorb without changing state) was related to the slowly changing variable of sea grass biomass. The regime shift (or state change) was related not to a single stressor but to a small set of factors, including sea grass biomass, disturbance regimes, and sediment stability.

The cross-scale nature of natural disasters requires that assessment and monitoring should cover multiple scales, up to the scale of the planet. Local, regional, national, and international agencies and institutions are not sufficient by themselves. Integration and communication are key to management and require novel approaches to dealing with the cross-scale issues. Longstaff and Yang (2008) describe the role of trust in leadership following natural disasters. Scale issues apply to policy, management, and leadership. In the best cases, leadership spans social-political scales; one person can lead for a time, but several are better locally, regionally, and politically (Westley et al. 2002; Folke et al. 2002, 2005).

Adaptive Management

The dynamics of complex adaptive systems following large-scale disasters, such as disease outbreaks, flood and drought cycles, hurricanes, and cyclones, present problems of predictability for management (Holling 1978, Walters 1986, Gunderson 1999, 2003). Yet planning and management require some estimate of future conditions. Certainly, many things are known, especially the broad and the general. In August 2005, it was well known, at least three days prior to landfall, that Hurricane Katrina would strike the Gulf Coast of the United States (with a given probability), but the specific impacts and the exact location of landfall could not be predetermined. In the case of influenza, similar levels of uncertainty abound. Certainly, as described elsewhere, the broad and the general forms of an influenza outbreak can be estimated and imagined, but the particulars, such as where and when a large-scale outbreak will occur, cannot. Our predictive abilities of such systems are limited for many reasons, including the evolutionary, adaptive, and cross-scale nature of complex systems as well as the lack of data to monitor and to test ideas about system dynamics across scales.

Adaptive management is an approach to natural resource management that was developed from theories of resilience (Holling 1978). Adaptive management acknowledges the deep uncertainties of resource management and attempts to winnow those uncertainties over time by using management actions as experiments to test policy (Walters 1986). Management must confront various sources of complexity in systems, including the ecological, economic, social, political, and organizational components of these systems (Holling and Gunderson 2002, Westley 2002), as well as the interactions among system components. The difficulties in managing the interacting aspects of social-ecological systems have led some to call them wicked problems (Rittel and Webber 1973, Ludwig et al. 2001). Developers of adaptive management approaches (Holling 1978, Walters 1986) acknowledged the complex multidimensionality of natural resource issues but focused on analytic approaches primarily in the ecological and economic domains from a systems perspective. Lee (1993) was the first to separate these issues into scientific (primarily ecological) and social (political) components. Adaptive management attempts to bring together disciplinary approaches for analysis and assessment and then integrate those ideas with policy and governance in the social arenas in a framework some describe as adaptive governance (Folke et al. 2005).

Adaptive governance is an emergent framework for managing complex environmental issues. Dietz et al. (2003) used the phrase to describe the social and human context for applying adaptive management. Folke et al. (2005) describe this form of governance as necessary for the management of complex ecosystems, particularly when change is abrupt, disorganizing, or turbulent. Brunner et al. (2006) provide a rich set of examples to illustrate the emergence of adaptive governance as a method for solving problems created by top-down control of decision making and attempts at implementing myopic scientific and technical solutions that are bereft of political considerations. They describe adaptive governance as operating in situations where the science is contextual, knowledge is incomplete, multiple ways of knowing and understanding are present, policy is implemented in modest steps, and unintended consequences and decision making are both top-down (although fragmented) and bottom-up. As such, adaptive governance is meant to integrate science, policy, and decision making in systems that assume and manage for change rather than against change (Gunderson et al. 1995).

Foundations of Resilience

The purpose of this volume is to synthesize the key scientific papers that led to our current understanding of resilience. In these papers lies the corpus of thought and understanding of resilience theory and thinking (Walker and Salt 2006). Our focus is specifically on ecological resilience. The concept of ecological resilience is often applied to social-ecological systems, but the foundations are in ecology. This book is organized both chronologically and categorically, focusing on theory, examples, and models.

The articles in this volume were chosen using a combination of factors. The first criteria for selection was articles that have shaped or defined the concepts and theories of resilience. We chose to include key papers that broke new conceptual ground and contributed a novel idea, or expansion of ideas, to our collective understanding of resilience, including the first formal descriptions of resilience theory from the early 1970s. The second criteria was to include examples that demonstrated ecological resilience in a range of ecosystems—that is, we included a small set of exemplar ecosystems where ecological resilience has been observed and reported. The third criteria for selection was a set of articles that provided the practical and methodological advances in understanding resilience. In essence, these factors became the groupings by which the volume is organized: concepts, examples, and models. The articles selected for each of these groups are described in the following paragraphs.

The first section of the collection includes the papers that are most often cited as the origins of the descriptions of ecological resilience and development of the fundamental concepts. These include the article that started the notion: C. S. Holling’s 1973 article in the Annual Review of Ecology and Systematics. Holling expanded on these ideas in 1986, introducing the concept of an adaptive cycle—a general heuristic of ecological dynamics over time, which is a prototheory of nonlinear dynamics in complex systems of people and ecosystems. Although Holling introduced the idea (and importance) of scale in 1986, the idea was developed first by Folke and colleagues in 1996, along with the entwined arguments of the relationship between diversity and resilience. Holling wrote a short chapter in 1996 to contrast and highlight at least two different ways that ecologists use and define resilience. To round out the section on theory and concepts, we include the annual review article by Folke and colleagues (2004), which is an outstanding summary of the knowledge of ecological resilience that had accumulated over three decades.

The second section of this volume includes classic papers from ecology that reported on regime shifts in very different ecosystems. All were the first publication for the specific ecosystem type that used the idea of alternative regimes to describe the resource dynamics that they observed. The first is by Terry Hughes (1994), who describes phase shifts in coral reefs of the Caribbean. The second is by James Estes and Donald Duggins (1995), who describe trophic-regulated regime shifts in the northern Pacific involving sea otters, sea urchins, and kelp forests. The third, by Craig Allen, Beth Forys, and C. S. Holling (1999), examines regime shifts in animal communities in transforming landscapes.

The final section of the book presents methodological and practical implications of resilience, especially how human intervention erodes or enhances resilience. We include the article on resource science by Holling and Chambers (1973) because it introduces the notion that ecological resilience is most often revealed in managed resource systems, or systems that have a large degree of human intervention. That article also introduced the use of models to help understand the complex dynamics exhibited by such resource systems. We also include two articles that each describe different aspects of forest pest management. Don Ludwig and colleagues (1978) provide elegant quantitative and qualitative analyses of outbreak dynamics. Clark, Jones, and Holling (1979) describe the practical implications of resilience for policy development and implementation.

We hope that this collection is a contribution to our collective understanding of resilience. We have likely overlooked some key articles and perhaps included some that are repetitive. In the end, those reading this volume will be the judge of the collection’s adequacy. In either case, we invite the reader to read or reread what we consider to be the foundational articles to that collective understanding.

Literature Cited

Adger, W. N., T. P. Hughes, C. Folke, S. R. Carpenter, and J. Rockstrӧm. 2005. Social-ecological resilience to coastal disasters. Science 309:1036–39.

Barry, J. 2004. The great influenza: The epic story of the deadliest plague in history. New York: Viking Adult.

Bellwood, D. R., T. P. Hughes, C. Folke, and M. Nystrom. 2004. Confronting the coral reef crisis. Nature 429:827–33.

Bohannon, J. 2005. Disasters: Searching for lessons from a bad year. Science 310:1883.

Brunner, R. D., T. D. Steelman, L. Coe-Juell, C. M. Cromley, C.M. Edwards, and D. W. Tucker. 2006. Adaptive governance: Integrating science policy and decision making. New York: Columbia University Press.

Clark, W. C., D. D. Jones, and C. S. Holling. 1979. Lessons for ecological policy design: A case study of ecosystem management. Ecological Modeling 7:1–52.

Dietz T., E. Ostrom, and P. C. Stern. 2003. The struggle to govern the commons. Science 302:1902–12.

Dublin, H. T., A. R. E. Sinclair, and J. McGlade. 1990. Elephants and fire as causes of multiple stable states in the Serengeti-Mara woodlands. Journal of Animal Ecology 59:1147–64.

Estes, J. A., and D.O. Duggins. 1995. Sea otters and kelp forests in Alaska: Generality and variation in a community ecological paradigm. Ecological Monographs 65: 75–100.

Folke, C., S. Carpenter, T. Elmqvist, L. Gunderson, C. S. Holling, and B. Walker. 2002. Resilience and sustainable development: Building adaptive capacity in a world of transformations. Ambio 31:437–40.

Folke, C., S. Carpenter, B. Walker, M. Scheffer, T. Elmqvist, L. Gunderson, and C. S. Holling. 2004. Regime shifts, resilience and biodiversity in ecosystem management. Annual Review of Ecology and Systematics 35:557–81.

Folke, C., T. Hahn, P. Olsson, and J. Norberg. 2005. Adaptive governance of social-ecological systems. Annual Review of Environment and Resources 30:441–73.

Gunderson, L. H. 1999. Resilience, flexibility and adaptive management. Conservation Ecology 3 (1): 7. http://www.consecol.org/vol3/iss1/art7/.

———. 2001. Managing surprising ecosystems in southern Florida. Ecological Economics 37:371–78.

———. 2003. Adaptive dancing. In Navigating social-ecological systems: Building resilience for complexity and change, ed. Fikret Berkes, Johan Colding, and Carl Folke, 33–53. Cambridge: Cambridge University Press.

Gunderson, L. H., and C. S. Holling. 2002. Panarchy: Understanding transformations in human and natural systems. Washington, DC: Island Press.

Gunderson, L. H., C. S. Holling, and S. S. Light. 1995. Barriers and bridges to the renewal of ecosystems and institutions. New York: Columbia University Press.

Gunderson, L. H., and L. Pritchard. 2002. Resilience and the behavior of large scale systems. Washington, DC: Island Press.

Holling, C.S. 1973. Resilience and stability of ecological systems. Annual Review of Ecology and Systematics 4:1–23.

———. 1978. Adaptive environmental assessment and management. Caldwell, NJ: Blackburn.

———. 1986. The resilience of terrestrial ecosystems: Local surprise and global change. In Sustainable development of the biosphere, ed. W. C. Clark and R. E. Munn, 292–317. Cambridge: Cambridge University Press.

———. 1988. Temperate forest insect outbreaks, tropical deforestation and migratory birds. Memoirs Entomological Society of Canada 146:22–32.

———. 1996. Engineering resilience vs. ecological resilience. In Engineering within ecological constraints, ed. P. C. Schulze, 31–43. Washington, DC: National Academy Press.

Holling, C. S., and L. H. Gunderson. 2002. Resilience and adaptive cycles. In Panarchy: Understanding transformations in human and ecological systems, ed. L. H. Gunderson and C. S. Holling, 25–62. Washington, DC: Island Press.

Hughes, T. P. 1994. Catastrophes, phase shifts, and large-scale degradation of a Caribbean coral reef. Science 265:1547–51.

Hughes, T. P., A. H. Baird, D. R. Bellwood, M. Card, S. R. Connolly, C. Folke, R. Grosberg, et al. 2003. Climate change, human impacts, and the resilience of coral reefs. Science 301:929–33.

Jackson, J. B. C., M. X. Kirby, W. H. Berger, K. A. Bjorndal, L. W. Botsford, L. W. Bourque, R. H. Bradbury, et. al. 2001. Historical overfishing and the recent collapse of coastal ecosystems. Science 293:629–38.

Janssen, M. A., and W. J. M. Martens. 1997. Modeling malaria as a complex adaptive system. Artificial Life 3:213–36.

Kasperson, J. X., R. E. Kasperson, and B. L. Turner II. 1995. Regions at risk: Comparisons of threatened environments. Tokyo: United Nations University Press.

Lee, K. 1993. Compass and gyroscope: Integrating science and politics for the environment. Washington, DC: Island Press.

Lewontin, R. C. 1969. The meaning of stability. Brookhaven Symposium in Biology 22:13–24.

Longstaff, P. H., and S. Yang. 2008. Communication management and trust: Their role in building resilience to surprises such as natural disasters, pandemic flu, and terrorism. Ecology and Society 13 (1): 3. http://www.ecologyandsociety.org/vol13/iss1/art3/.

Ludwig, D., D. D. Jones, and C. S. Holling. 1978. Qualitative analysis of insect outbreak systems: The spruce budworm and forest. Journal of Animal Ecology 47:315–32.

Ludwig, D., B. Walker, and C. S. Holling. 1997. Sustainability, stability, and resilience. Conservation Ecology 1 (1): 7. http://www.consecol.org/vol1/iss1/art7/.

Ludwig, D., M. Mangel, and B. Haddad. 2001. Ecology, conservation, and public policy. Annual Review of Ecology and Systematics 32:481–517.

May, R. M. 1977. Thresholds and breakpoints in ecosystems with a multiplicity of stable states. Nature 269:471–77.

McClanahan, T., N. Polunin, and T. Done. 2002. Resilience of coral reefs. In Resilience and behavior of large-scale systems, ed. L. H. Gunderson and L. Pritchard Jr., 111–64. Washington, DC: Island Press.

Millennium Ecosystem Assessment. 2005. Ecosystems and human well-being: Current state and trends findings of the condition and trends working group. Washington, DC: Island Press.

Pimm, S.L. 1991. The balance of nature? Chicago: University of Chicago Press.

Rittel, H., and M. Webber. 1973. Dilemmas in a general theory of planning. Policy Sciences 4:155–69.

Scheffer, M. 1998. The ecology of shallow lakes. London: Chapman and Hall.

Scheffer, M., S. R. Carpenter, J. Foley, C. Folke, and B. Walker. 2001. Catastrophic shifts in ecosystems. Nature 413:591–696.

Sousa, W. P., and J. H. Connell. 1985. Further comments on the evidence for multiple stable points in natural communities. American Naturalist 125:612–15.

Steffen, W., A. Sanderson, J. Jӓger, P. D Tyson, B. Moore III, P. A. Matson, K. Richardson, F. Oldfield, H. J. Schellnhuber, B. L. Turner II, and R. J. Wasson. 2004. Global change and the earth system: A planet under pressure. Heidelberg: Springer-Verlag.

Steneck, R. S., J. Vavrinec, and A. Leland. 2004. Accelerating trophic level dysfunction in kelp forest ecosystems of the western North Atlantic. Ecosystems 7:323–31.

Vitousek, P. M., H. A. Mooney, J. Lubchenco, and J. M. Melillo. 1997. Human domination of earth’s ecosystems. Science 277: 494–99.

Walker, B., C. S. Holling, S. R. Carpenter, and A. Kinzig. 2004. Resilience, adaptability and transformability in social–ecological systems. Ecology and Society 9 (2): 5. http://www.ecologyandsociety.org/vol9/iss2/art5/.

Walker, B., and D. Salt. 2006. Resilience thinking: Sustaining ecosystems and people in a changing world. Washington, DC: Island Press.

Walker, B. H. 1981. Is succession a viable concept in African savanna ecosystems? In Forest succession: Concepts and application, ed. D. C. West, H. H. Shugart, and D. B. Botkin, 431–47. New York: Springer-Verlag.

Walters, C. J. 1986. Adaptive management of renewable resources. New York: McGraw-Hill.

———. 1997. Challenges in adaptive management of riparian and coastal ecosystems. Conservation Ecology 1 (1). http://www.consecol.org/vol1/iss2/art1.

Westley, F. 2002. The devil in the dynamics: Adaptive management on the front lines. In Panarchy: Understanding transformations in human and natural systems, ed. L. H. Gunderson and C. S. Holling, 333–60. Washington, DC: Island Press.

Westley, F., S. R. Carpenter, W. A. Brock, C. S. Holling, and L. H. Gunderson. 2002. Why systems of people and nature are not just social and ecological systems. In Panarchy: Understanding transformations in human and natural systems, ed. L. H. Gunderson and C. S. Holling, 103–19. Washington, DC: Island Press.

Westoby, M., B. H. Walker, and I. Noyr-Meir. 1989. Opportunistic management for rangelands not at equilibrium. Journal of Rangeland Management 42:266–74.

PART ONE

Concepts and Theory

Commentary on Part One Articles

CRAIG R. ALLEN, LANCE H. GUNDERSON, AND C. S. HOLLING

WHILE SOME IN COMMON PARLANCE use the word theory to mean something that is speculative or opinion, we use it in a scientific sense to indicate an explanation based on observation and reasoning, which is consistent with its meaning for over four hundred years (Harper 2001). As such, resilience and panarchy are an integral part of ecological theory because of their application in understanding and explaining commonalities in patterns of change in complex systems. Those systems can be ecosystems or human or social systems, or combinations thereof (Gunderson and Holling 2002, Walker et al. 2004). This section presents six papers that collectively describe much of the theory of ecological resilience. They include the first article to use the word resilience to describe a new way of conceptualizing ecosystem dynamics.

There are two views of human interactions with, and management of, the world. In one, efforts are focused on maintaining a degree of constancy by reducing natural variability. In the other, focus is on maintaining the consistency of relationships among various parts of the ecological system in question. The former comes from traditions of physics, engineering, and other similar quantitative sciences, while the latter focuses on qualitative properties of systems. The first focuses on equilibrium states; the latter, on the persistence of function and structure. Both approaches are useful, under the right circumstances.

Systems have a capacity to absorb disturbances, but this capacity has limits and bounds, and when these limits are exceeded the system may rapidly transform. Holling was the first to recognize the significance of thresholds in ecological systems, and the importance of avoiding them. Holling (1973) outlines how the response of systems can exhibit threshold behavior. Changes in either driving or state variables may cause collapses. Often, the system provides no warning, and collapse follows an unexpected, but often inevitable, event. More recently, increasing variability in some variables has been suggested as an indicator of impending collapse of a system as it approaches the limits of its resilience (Carpenter and Brock 2006, Wardwell and Allen 2009).

Knowledge of the form of critical population processes (particularly predation) suggests that ecological systems have more than one stable state (multiple stable states, multiple equilibria, and meta-stable states are all words used to describe such behavior). Variables move between some of those states, and that variability both results in and is caused by diversity in space, time, and species. Abrupt jumps in variables are the rule, not the exception.

Resilience is described here as the property that allows the fundamental functions of an ecosystem to persist in the face of extremes of disturbance. It can be measured by the size of the viable stability domains. Stability, in contrast to resilience, is used in a narrow sense of elasticity. It is the property that resists departure from equilibrium and that maximizes the speed of return to the equilibrium following small disturbances. Resilience focuses on the role of positive feedbacks, of behavior far from steady states and with internally generated variability. Stability, in the narrow sense above, deals with negative feedback, of behavior near steady states, and with constancy. Different views and definitions are still being used to distinguish between resilience and stability. The view expressed above is generally used by those ecologists who develop theory empirically, who often use simulation models, and who conduct their science integrated with policy and ecological management. They are typically trained within a biological tradition.

Those who define resilience differently use a measure of elasticity or return time, a definition that is the opposite of that above. It is a definition that implicitly assumes there is only one equilibrium state. Scientists holding this view tend to be more deductive in their formation of theory or are influenced by an engineering and applied mathematical tradition. They tend to apply theory to practice rather than to develop theory empirically as part of practice.

One empirically based critique of multi-stable states has been influential (Sousa and Connell 1985). But this critique is inadequate because it relies on only existing published time series data and ignores any kind of analysis of causation. It is a phenomenological investigation, not causal. As a result, behavior is seen as being determined or explainable by only one variable, the time horizon for the variable is too short, and multiscale interactions between variables of different speeds are ignored. This reflects a common limitation of many population studies.

Holling (1973) documents stability domains using empirical evidence from numerous studies. Stability is defined as the return of a system to an equilibrium state following disturbance, and resilience is defined as a measure of a system’s persistence and its ability to absorb change and disturbance but still maintain the same relationships among population or state variables. A system can be highly unstable but very resilient. In fact, a key insight of this paper is that instability may create highly resilient systems (e.g., grassland persistence is reliant on frequently occurring fires). Managing for stability, as humans so often do, has the unexpected outcome of reducing a system’s resilience (Holling and Meffe 1996, Allen and Holling 2008). An equilibrium-focused view is attractive to humans, who often focus on optimizing single elements of systems, but it fails to capture the behavior of complex systems.

By the early 1990s, many authors in the ecological literature (O’Neill et al. 1986, Pimm 1984, Tilman and Downing 1994) had applied the word resilience as the speed or time of return of an ecological system to an equilibrium following a disturbance. This was part of a multifaceted definition outlined by Holling in 1973, but it is a narrow definition and ignores the presence of alternative states. In response, Holling (1996) explicitly contrasts and compares two primary definitions of resilience, which he describes as engineering resilience and ecological resilience. Ecological systems differ from engineering ones in that change is not continuous but, rather, discontinuous; ecological change is characterized by surprising events (such as hurricanes, fires, or pest outbreaks) that open windows of opportunities for establishing new combinations of species and ecological processes (Allen and Holling 2008). Ecological attributes are also distributed discontinuously in space, across scales. Ecosystems don’t have single equilibria; rather, they have multiple equilibrium and are often far from equilibrium and are on dynamic trajectories. Like the location of electrons about an atom, an ecological system has changed by the time it can be measured, and optimal approaches to ecosystem management are prone to failure. Additionally, management actions and policies focused on constant yields and the reduction of variability reduce the resilience of a system. Because these systems are moving targets, management needs to be flexible, adaptive, and experimental and must recognize the multiple critical scales characterizing a given system.

Engineering resilience focuses on equilibrium states and is measured simply as the return time following disturbance. This definition, often used by population biologists, is analogous to the intrinsic rate of increase of a species (r). Engineering resilience focuses on stability, in the sense of elasticity. It is the ability of a system to resist departure from an equilibrium following disturbances and to return to the same equilibrium when sufficiently perturbed.

Ecological resilience focuses on conditions far from equilibrium, when abrupt shifts between multiple stable states are possible. Here, the measurement of resilience is the magnitude or amount of disturbance that can be absorbed without undergoing the shift to an alternative stable state characterized by changes in controlling variables and processes and their dominant scales. More recently, stable states have been characterized by their process regimes, and the term regime shift has been used (Scheffer et al. 2001). Ecological resilience can be measured by the size of the stability domains.

The two differing definitions of resilience lead to grossly different strategies for managing systems and responding to surprise. Managing for stability is suggested by engineering resilience; however, this often has demonstrable negative consequences in the long run. Productivity or yield is often increased over short time periods due to management efficiency and optimization but suffers in the long run as ecological surprises exceed the diminished resilience of the system. This happens because managing for reduced variability in one or a small number of variables alters competitive interactions and the buildup of capital (such as fuel for fires) and leads to the loss of important processes and functions. The reduction in variability means that key structuring variables and processes are lost or greatly diminished (e.g., pest outbreaks or fires).

The changes that occur when the resilience of a system is exceeded can lead to an undesirable, but highly resilient, system state. Reversing the system can be very difficult because undesirable systems can be extremely resilient and the regime shifts may exhibithysteresis. Concomitant with the reduction in resilience when management attempts to reduce variability is an increasingly rigid management bureaucracy and ever more dependent—and vulnerable—human societies. An increasing reliance by humans on systems where management has reduced variability often means that ecological shifts have ever greater and negative implications for human economies and societies.

Holling (1996) also begins to formulate a model of the relationship among ecological diversity, resilience, and scale (formally conceptualized in Peterson et al. 1998). He notes that resilient systems have multiple controls that are most efficient on different scales, and that the distribution of diversity within and across scales is what matters.

Holling (1986) is part of a groundbreaking volume that was one of the first works to build and synthesize understanding around themes of sustainability and global environmental change (Clark and Munn 1986). This was years before global climate change was a widespread research topic or undertaken by large international research bodies. As part of this work, Holling (1986) applied the concept of resilience to help understand how a wide range of ecosystems would respond to broad-scale environmental (climatic) change. Ecological systems exhibit a diverse array of responses to global changes—a characteristic that is inherent in their resilience. Nonadaptive systems with little flexibility in response to perturbation and disturbance would be in a constant state of flux and disarray. This paper provides an early recognition of discontinuous and nonlinear response in ecological systems and represents an early attempt to link ecological and social systems across scales. Holling (1986) recognized that positive feedbacks are responsible for maintaining the systems on which humanity relies—for example, feedbacks between the atmosphere and vegetation.

In many ways, this paper was an early warning of the very real possibility that the resilience of global systems could be exceeded, resulting in very sudden and effectively irreversible regime shifts. Because many of the anticipated changes are global, rather than local, in nature, adaptation to changes caused by the human footprint will need to occur not only within individuals but within institutions and social systems as well. Twenty years later, approaches linking social-economic-ecological systems are commonplace and viewed as the frontier in global change and resilience research (Walker and Salt 2006).

A theme of the Holling (1986) paper is recognizing the inevitability of surprises—unexpected outcomes with causes and responses very different from those anticipated, or results or behaviors that are induced by human actions but which are very different than expected. Such thoughts were given a wide public airing when Donald Rumsfeld, secretary of defense under U.S. president George W. Bush, stated the following at a Department of Defense news briefing on February 12, 2002: "Because as we know, there are known knowns; there are things we know we know. We also know there are known unknowns; that is to say we know there are some things we do not know. But there are also unknown unknowns—the ones we don’t know we don’t