Chapter 2 introduced the concept of ecosystem services and outlined an “ecosystem services approach” to damage assessment for events such as the Deepwater Horizon (DWH) oil spill. This chapter addresses resilience and its relationship to ecosystem services.
Ecosystems are subject to natural disturbances such as fires, floods, droughts, and disease outbreaks, as well as human-caused disturbances, including oil spills. Ecosystems are also subject to slowly changing long-term stresses, such as nutrient enrichment and changes in sediment supply. These long-term stresses can affect the ability of the system to respond to a shock such as the DWH oil spill.
A resilience framework focuses attention on shocks (pulse disturbances), long-term stresses (press disturbances), and the response of complex systems to these shocks and stresses (Ives and Carpenter, 2007). Does a system recover slowly or rapidly from a shock? What factors within the system allow for more rapid and more complete recovery? Do attempts to stabilize certain components of systems lead to a reduction in overall system resilience and greater potential for large changes (Gunderson and Holling, 2002)?
Considerations of resilience are especially important in systems that undergo persistent and fundamental shifts in structure and function after disturbances (“regime shifts”). Increasing resilience can reduce the risk that the system will cross critical thresholds and undergo a detrimental regime shift. On the other hand, decreasing resilience can increase the probability of a beneficial regime shift. Understanding of how to increase (or decrease) system resilience places a premium on knowledge of system dynamics, including feedbacks among system components, as well as of uncertainty and variability in dynamic systems (Walker and Salt, 2006). Can factors that increase system resilience be identified and managed to increase (decrease) the resilience of a system to a “desirable” state?
System resilience can play an important role in maintaining conditions that will sustain the provision of ecosystem services that contribute to human well-being. However, a narrow focus on stabilizing complex systems to provide a constant flow of ecosystem services may reduce system resilience and increase vulnerability (Gunderson et al., 1995). An event such as the DWH oil spill may disrupt service provision, but a resilient ecosystem will allow faster recovery so that services will return sooner rather than later or never.
Finding 3.1. Resilience provides a useful conceptual framework for managing complex systems. It focuses attention on system dynamics and how systems are affected by short-term disturbances and long-term stresses.
In the context of resilience of the Gulf of Mexico (GoM) ecosystem, the committee was charged with addressing the following questions from the Statement of Task (see Chapter 1):
In light of the multiple stresses on the Gulf of Mexico ecosystem, what practical approaches can managers take to restore and increase the resiliency of ecosystem services to future events such as the Deepwater Horizon Mississippi Canyon-252 spill? How can the increase in ecosystem resiliency be measured?
Although resilience is an important concept, much like the concept of ecosystem services, providing practical and specific advice to managers to “increase the resiliency of ecosystem services” is a difficult task at present. As noted above, managing resilience requires understanding of the dynamics of complex and highly variable systems, which is often quite limited. Without such understanding, it can be difficult in practical contexts to know how management actions affect resilience. Some researchers have gone so far as to state that resilience is “too vague of a concept to be useful in planning” or ecosystem management (Lindenmayer and Hunter, 2010).
Finding 3.2. Limited data and understanding of complex system dynamics make it difficult to provide specific practical advice to managers on how to restore or increase the resilience of ecosystem services.
The next section explores definitions of resilience, its application in ecosystems and in integrated social-ecological systems, and the relationship between resilience and provision of ecosystem services. The final sections discuss options to manage systems to enhance resilience and approaches to measuring changes in resilience.
RESILIENCE IN SOCIOECONOMIC SYSTEMS AND ITS RELATIONSHIP TO ECOSYSTEM SERVICES
Defining Engineering and Ecological Resilience
The study of resilience emerged primarily in ecology, with initial applications focused on the resilience of ecosystems (Holling, 1973), but resilience is now used more broadly in ecology, economics, engineering, law, natural resource management, psychology, sociology, and other disciplines. The concept of resilience has been applied to ecological systems, social systems, and more recently to integrated social-ecological systems (Berkes and Folke, 1998; Berkes et al., 2003). With this expanded use of resilience has come multiple definitions of resilience. As long as each field of study used a definition that was well defined within the contexts of that field, few problems arose. However, in an interdisciplinary context such as the resilience of ecosys-
tem services in response to a perturbation such as the DWH oil spill, multiple definitions of resilience can come into play, leading to confusion (Gallopin, 2006).
For this report, resilience is defined as the ability of a system subject to disturbance to retain its essential structure, function, and feedbacks (Walker and Salt, 2006) and return to its predisturbance state. Two distinct notions of resilience fit within this general definition, depending on whether the system is characterized by a single equilibrium or multiple equilibria (Holling, 1996). With regard to a dynamic system, “equilibrium” refers to the state where the system will not change unless subjected to a disturbance, also called “steady state.” More generally, one can refer to a “regime” of a dynamic system in which the system follows a given trajectory rather than being static. So, for example, a system may follow a regular cycle unless disturbed. The notion of resilience for a system with a single equilibrium focuses on the speed with which it returns toward equilibrium following a disturbance and is referred to as “engineering resilience” (Holling, 1996). The notion of resilience for a system with multiple equilibria focuses on the magnitude of disturbance the system can absorb without shifting to a new equilibrium (Walker et al., 2004). This form of resilience is referred to as “ecological resilience” (Holling, 1996).
Engineering resilience refers to the speed with which a system returns to equilibrium after a disturbance. For example, how fast does a material return to its original form after a shock that deforms the material? A related notion is resistance, which is the ability of a disturbed system to stay near equilibrium. A rubber band is easily stretched when force is applied (low resistance) but quickly returns to its original shape (high resilience) once the force is removed. A steel rod, on the other hand, is highly resistant to force, but it will bend once the force is sufficiently strong and remain bent even when the force is removed (low resilience). Both the rubber band and the steel rod can retain their original forms after being subject to a disturbance, but only the rubber band returns to its original form after it is deformed by the force of the disturbance. A sufficiently strong disturbance will cause the rubber band to break or the steel rod to bend, which may lead to a new type of equilibrium discussed below under the heading of ecological resilience.
Ecologists have a long history of studying disturbance and recovery. Clements (1936) held that the predictable succession of species followed trajectories toward a specific climax state after a disturbance. Considerable progress was made in the 1950s to provide a solid theoretical basis and predictive capability for this view of resilience as recovery to a climax state. This line of investigation led to seminal works on post-disturbance ecosystem recovery (e.g., Connell and Slater, 1977; Odum, 1969).
Ecologists have been especially interested in the relationship between species diversity and resilience. Intuitively, it seems that high diversity should be related to greater resilience. Similar to how an asset portfolio is diversified to reduce financial risks, highly diverse ecosystems are more likely to contain species that can respond well to particular kinds of disturbances, thereby allowing the system as a whole to better maintain functions in the face
of disturbances (Tilman et al., 1996, 2005). However, May (1974) showed mathematically that greater diversity could lead to lower system stability. In principle, empirical evidence is largely supportive of the hypothesis that greater species diversity leads to greater system stability (Tilman, 1996; Tilman and Downing, 1994; Tilman et al., 2006). However, some debate remains as to whether high species diversity contributes to system stability by increasing resilience (Ives and Carpenter, 2007; McCann, 2000; Naeem, 2002; Rooney et al., 2006).
During the 1960s ecologists questioned the view that ecological systems tended toward a unique climax state and instead raised the possibility that systems might have multiple potential equilibria (Holling, 1965; Lewontin, 1969). Once disturbed, new conditions can foster a new set of feedbacks and prevent the system from returning to its pre-disturbance equilibrium. For example, plants absorb phosphorus and limit algal growth in shallow lakes with low levels of phosphorus. An increase in phosphorus inputs, however, can lead to algal blooms that reduce light penetration and kill plants, releasing more phosphorus for algae. Algal domination can persist even when phosphorus inputs into the lake decline back to original levels (Carpenter, 2003).
With multiple equilibria, which equilibrium a system will tend to move toward depends upon the set of system conditions. Multiple equilibria generate the potential for a system to cross critical thresholds and flip between equilibria (see Figure 3.1). In the context of multiple equilibria, resilience typically refers to the ability of a system to undergo disturbance without crossing a critical threshold and therefore return to the original equilibrium state (Carpenter et al., 2001; Gunderson and Pritchard, 2002; Holling, 1973, 2001). When a system crosses a critical threshold, it is said to have undergone a “regime shift.” Once a regime shift has occurred, it may be difficult for the system to reverse back to the original equilibrium. Simply removing the
FIGURE 3.1 Illustration of a regime shift. Initially, the system is in equilibrium in regime 1. It will remain in regime 1 unless a sufficiently strong disturbance pushes it over the tipping point and into regime 2. Once the system is in regime 2, it will need an even stronger disturbance to shift it back to regime 1. SOURCE: Biggs et al., 2012a; http://www.stockholmresilience.org/news/researchinsights/regimeshifts.106.d3c937136e935f4864514.html
stresses that lead to a regime shift may be insufficient to return the original state (Scheffer et al., 2001).
Evidence of regime shifts has been found in a several types of ecosystems (e.g., Beisner et al., 2003) including shallow lakes (Scheffer and Carpenter, 2003), grassland grazing systems (Van de Koppel et al., 1997), coral reefs (Hughes et al., 2003), and other marine systems (Nyström et al., 2012; Petraitis et al., 2009). However, some researchers have been skeptical of this evidence. Sousa and Connell (1983, 1985) employed strict criteria to judge whether an observed change represented a single system with multiple steady states and found little convincing evidence for multiple states. Schroder et al. (2005) reviewed experimental studies testing for alternative stable states and found mixed results.
The Mechanics of Ecological Resilience
Ecological systems with strong negative feedbacks are resilient. Negative feedbacks work in opposition to any disturbance and tend to push a system back to equilibrium. A thermostat provides an example of a negative feedback. In economic systems, price responses often function as negative feedbacks. In cases of excess demand, price increases reduce demand and increase supply, thus bringing demand back into balance with supply. That said, the existence of positive feedbacks tends to reinforce disturbances, leading to instability. For example, the introduction of fire-adapted invasive grass species tends to promote fire, which in turn reinforces the competitive advantage of these grasses, leading to a new equilibrium dominated by fire-adapted invasive species (Mack and D’Antonio, 1998).
The concept of negative feedback can be traced to early 20th century work by Russian physiologist P. K. Anokhin (1935). Indeed, physiology has a long history of describing negative feedback systems that operate at the organismal level to maintain homeostasis in body temperature and chemistry. Homeostasis in physiology is the maintenance of a constant internal environment in response to a changing external environment. A great deal of research has been done at the organismal level to understand how feedback mechanisms maintain homeostasis.
Salt marshes, such as those found in the GoM, provide a good example of negative feedbacks that maintain ecosystem conditions in the face of disturbances and long-term stresses. Salt marshes accumulate sediments at rates that can keep the marsh in equilibrium with sea level even with sea level rise (Morris et al., 2002). This type of feedback has been termed ecogeomorphological (Fagherazzi et al., 2004) or ecogeomorphic feedback (Kirwan et al., 2010). The sedimentation of suspended solids carried by tides over the marsh surface increases with the duration of flooding (Friedrichs and Perry, 2001; Krone, 1985) and with the density of standing vegetation (Morris et al., 2002). Flood duration is proportional to the depth of the marsh surface below mean high water, at least until the surface falls below the lowest tide level. In addition to surface deposition, production of organic matter, primarily of roots and rhizomes, contributes to the total accumulation rate (Reed, 1995; Turner et al., 2001). Density and primary productivity of vegetation are related to flood duration and depth (Morris et al., 2002). For the
dominant salt marsh species, smooth cordgrass (Spartina alterniflora), plant growth occurs between mean sea level and mean high water (McKee and Patrick, 1988), with maximal growth near the middle of this range (Morris et al., 2002; Figure 3.2).
Sea level rise, an issue of particular concern in the northern GoM, will decrease relative elevation of coastal marshes. If initial marsh elevation is above the elevation for optimal growth, then sea level rise will increase primary production and sedimentation (Figure 3.2). This feedback maintains the elevation of the marsh, within limits. However, if the rate of sea level rise is too great, then a decrease in elevation of the marsh will decrease primary production and sediment accretion and the marsh surface will fall further behind rising sea level, culminating in the collapse of the marsh and loss of ecosystem services. Consequently, the most stable elevation is one that is higher than optimal for maximum vegetation growth. This illustrates an important general point: aiming to maximize a particular ecosystem service can be counterproductive to other desirable properties, such as resilience, or other ecosystem services, such as storm protection.
FIGURE 3.2 Conceptual model of the effect of relative elevation on salt marsh primary production and sediment accretion. MSL = mean sea level; MHW = mean high water. SOURCE: Based on model in Morris et al. (2002).
Resilience in Social-Ecological Systems
Humans are an integral part of virtually all ecosystems, including the GoM ecosystem. Humans have modified the flow of the Mississippi River, redirected sediment and nutrient loadings into the GoM, loaded pollutants into the water, built cities and towns along the coastline, built canals through coastal marshes, harvested various species of fish and shellfish, and extracted resources such as oil and natural gas. Inclusion of human behavior, and the feedbacks between human behavior and environmental conditions, can mean that the resilience of social-ecological systems differs in important ways from the underlying resilience of ecological systems without consideration of humans (Berkes et al., 2003; Walker et al., 2006). In thinking about the resilience of the GoM system and the ecosystem services that it provides, it is the resilience of the social-ecological system that is of greatest relevance. Knowledge of the resilience of social-ecological systems, however, is even more limited than knowledge of the resilience of ecosystems. Much of the literature on the resilience of social-ecological systems is abstract and general, making it difficult to provide specific, practical suggestions to managers.
Social scientists have begun to incorporate notions of resilience into their analyses of social systems. Perhaps the most relevant social science analysis in the context of ecosystem services is “community resilience.” Community resilience refers to a community’s ability to recover from losses imposed by disturbances, such as oil spills and hurricanes, without significant external assistance (Cutter et al., 2008; Mileti, 1999).
The resilience of human communities is related to the resilience of ecosystems to which communities are linked, but community resilience and ecological resilience are not perfectly correlated. Resilient ecosystems can contribute to social and economic well-being by providing a stable base for a diverse set of industries (fishing, recreation, and tourism) and stable provision of ecosystem services that contribute to quality of life (Adger et al., 2005). Resilient ecosystems can also provide protection against disturbances that would reduce income or quality of life, such as storm surge protection and flood control. However, a community may be resilient despite being linked to irresilient ecosystems. For example, a community that has diversified sources of income may be resilient to disturbances to ecosystems that disrupt specific industries such as commercial fishing. If community members have ready access to other forms of employment or to community support mechanisms that allow for retooling and retraining, then they may not suffer much long-term loss when one particular industry is disrupted (Cutter et al., 2008). In general, a highly diversified economy, including sectors that are buffered from disturbances, will tend to reduce the dependence of the social and economic systems on ecosystem resilience.
Resilience of human communities can also impact the resilience of ecosystems. Human actions resulting in changes in land use, nutrient cycling, hydrology, or pollution levels can reduce ecosystem resilience. For example, increased sediment loading and overharvesting of grazing fish can make coral reefs more susceptible to bleaching and die-offs. When human communities are overly reliant on the provision of a small set of industries, such as sole reliance on fisheries or oil and gas production, they may become locked into certain patterns of behavior that add stress to ecosystems, which can make it difficult to maintain social-ecological system resilience.
Changes in ecosystems and in social systems can be mutually reinforcing (positive feedbacks). In some cases, this can lead to a virtuous cycle where improved community well-being leads to better environmental protection that further promotes community well-being. But the cycle can also go in the other direction. For example, the decline of coral reefs may cause a decline in fisheries or tourism, with consequent decreases in community vitality and increases in social stress. Such stress may lead to reductions in coral reef protection, with further degradation of the reefs. In fishery-dependent communities, environmental shocks that affect the underlying productivity of a fishery may push fishermen to intensify their harvest to maintain their income. But intensified harvesting can further reduce stocks and harm future productivity, requiring even more harvesting pressure in the future. Such downward spirals lead to unsustainable outcomes.
The ability to change human behavior and management in light of changes in underlying conditions (“adaptability”) is an essential component of overall social-ecological resilience (Folke et al., 2010). Highly functioning human systems can learn and adapt their behavior to maintain overall social-ecological system resilience. For example, good management that reduces fishing pressures when fish stocks are depleted and that allows harvests to expand when stocks are abundant can help to stabilize fish populations. In contrast, rigid adherence to constant harvest quotas set for average conditions in the face of a variable environment can lead to fishery collapse (Roughgarden and Smith, 1996). Poverty or lack of properly functioning governance can lead to maladapted behavior and rapid decline of system conditions.
Living successfully in a dynamic and changing world involves adapting to new conditions, and some authors have broadened the concept of social-ecological resilience to allow for adaptation. Folke et al. (2010) define resilience of social-ecological systems as the ability “to continually change and adapt yet remain within critical thresholds. This adaptability is part of resilience.” Adaptability may require that some aspects of the system change in order for the whole system to retain essential system properties. For example, sea level rise may require shifts in the location of coastal marshes as well as human infrastructure to maintain desirable ecosystem services and to avoid future damages from flooding. Some adaptation, especially in the face of large disturbances, may require fundamental transformations (Folke et al., 2004; Walker et al., 2004). Transformations can be traumatic, as when fisheries collapse and fisherman must find a new way to make a living. Planning ahead for the possibilities of transformations can reduce the burden, such as by providing retraining support during periods of adjustment. The institutional and governance capacity of social-ecological systems is a crucial element for having capacity to adapt, transform, and innovate (Folke et al., 2010; Gunderson, 2000; Gunderson and Folke, 2011; Olsson et al., 2004).
Finding 3.3. Resilience in social-ecological systems depends on the feedbacks between ecological and human communities. Because of the feedbacks between human behavior, environmental conditions, ecosystem services, and human well-being, understanding of resilience of social-ecological systems requires integrated analysis that includes both nature and people.
The Relationship Between Resilience and the Provision of Ecosystem Services
Productive ecosystems are necessary for the supply of ecosystem services. Ecosystems that lack resilience are vulnerable to disturbances that can lead to reductions in the supply of ecosystem services. Maintaining a stable supply of ecosystem services, therefore, depends on having resilient ecosystems. The GoM ecosystem is subject to periodic disturbances from hurricanes or oil spills, which can cause temporary declines in service provision. But a resilient ecosystem will allow speedy recovery of ecosystem services provision. After the DWH oil spill, many fisheries in the GoM were closed because of health concerns, and tourism revenue for the GoM coastal region was substantially lower in 2010 than in 2009. However, fisheries were reopened and tourism rebounded quickly. In several locations, the numbers of visitors and tourism revenues were higher in 2011 than in 2009 (AGCCVB, 2012; Mississippi Development Authority/Division of Tourism, 2012).
Some ecosystem services are valuable precisely because they increase the resilience of social-ecological systems. Ecosystem services such as flood mitigation and coastal protection from storm surge reduce the size of disturbances and the destruction that floods and storm surge cause to human communities. Several studies have analyzed the value of coastal marshes, mangroves, and other habitats for protection of coastal communities from storm surge (e.g., Barbier et al., 2008; Costanza et al., 2008; Das and Vincent, 2009). It is often difficult to be precise about how much protection ecosystems are likely to provide given the variability of storms, including wind speed and direction, duration, and arrival of the storm relative to high tides (Koch et al., 2009), but there can be little doubt that their contributions can be significant (see Chapter 5 for additional details).
Some ecosystem services can be replaced by human-engineered services. For example, storm protection can be provided by green infrastructure (coastal marshes) or grey infrastructure (concrete barriers), and water purification for clean drinking water can be provided by forests and wetlands that filter nutrients and pollutants or by water treatment facilities. Loss of ecosystem resilience can compromise ecosystem services and may require replacement of the ecosystem service with a human-engineered service. In some instances, provision of a humanwengineered service may undermine the provision of other services—for example, concrete barriers could protect one section of coastline but intensify erosion in nearby locales, resulting in increased vulnerabilities elsewhere along the coast.
Coastal communities are often highly dependent on coastal resources for their livelihood. The services provided by the ecosystem to the community may be quite diverse, as they are in the GoM, and may include commercial and recreational fishing, beach tourism, wildlife viewing, marine transportation, and other extractive uses. Any event that impacts the ecosystem, such as an oil spill, can have an impact on multiple ecosystem services, with consequent impacts on the human community (Adger, 2000). Highly dependent communities can be severely impacted by disruptions to ecosystems that cause a loss of services.
Finding 3.4. Communities with highly diversified economies and social structures are better placed to withstand disturbances to ecosystems that affect the provision of ecosystem services.
MANAGING FOR RESILIENCE
The health and well-being of coastal communities in the GoM are tied to the continued flow of services provided by coastal and marine ecosystems. A key question for such communities is how to maintain the flow of valuable ecosystem services in the face of long-term social and ecological changes and short-term disturbances such as oil spills and hurricanes. Resilience of the system in the face of ongoing shocks and stresses should be an important objective of resource management.
This section begins with a discussion of ways in which system resilience could potentially be influenced by management actions, both when the specific sources of disturbance can be identified (specific resilience) and when they cannot (general resilience). This section then discusses current management approaches used by federal agencies to manage the resilience of the GoM social-ecological system and then explores metrics that can be used to measure changes in resilience. Finally, the section discusses ecosystem restoration, which is a management strategy of particular interest after a human-caused disturbance such as an oil spill.
Managing for General and Specific Resilience
When managers know the specific types of disturbances they are likely to face, they can often take specific measures to increase the system’s resilience to these disturbances (Adger et al., 2005). Management for specific resilience involves using knowledge of specific types of disturbances to design plans to minimize damages and to aid recovery. For example, a coastal system subject to periodic hurricanes could increase social-ecological resilience by
• increasing the areal extent and health of coastal marshes and mangroves, which can serve as a buffer to absorb wave energy between the open water of the ocean and coastal communities (Barbier et al., 2008; Costanza et al., 2008; Das and Vincent, 2009), although there remain questions about the effectiveness of vegetation to provide protection (Feagin et al., 2010);
• improving communications and early warning systems to provide information about impending danger to people;
• investing in disaster preparedness and planning; and
• changing infrastructure and building design and location to minimize the risk of loss from storm surge or wind damage.
Building system resilience to more swiftly recover from oil spills can be aided by many of these same measures as well as by investments in safety procedures and engineering to reduce the risk of catastrophic accidents, and in emergency response capabilities should an accident occur.
Managers face a more challenging task in creating “general resilience” that allows for greater capacity to respond to many different types of disturbances, some of which will un-
doubtedly be a surprise (Adger et al., 2005; Anderies et al., 2006; Walker and Salt, 2006, 2012). Certain system properties are thought to increase general system resilience in a wide range of circumstances. Several authors have described properties to increase general resilience, which are summarized in five strategy categories (see Table 3-1) and described in more depth below.
TABLE 3.1 Principles of Management for General Resilience of Social-Ecological Systems
|Resilience Strategy||Levin (1999): Eight Commandments for Ecosystem Management||Walker and Salt (2006): Nine Attributes of Resilience||Biggs et al. (2012b): Seven Generic Principles for Enhancing Resilience|
|Maintain diversity, variability, and redundancy||Maintain heterogeneity||Promote and sustain diversity in all forms (biological, landscape, social, and economic)||Maintain diversity and redundancy|
|Preserve redundancy||Embrace and work with ecological variability rather than attempt to control and reduce it|
|Manage feedbacks and slowly changing variables||Tighten feedback loops||Tighten feedbacks
Focus policy on slowly changing variables associated with thresholds
|Manage slow variables and feedback|
|Manage modularity and connectivity||Sustain modularity||Modularity||Manage connectivity|
|Apply adaptive management, learning; anticipate potential future outcomes||Reduce uncertainty
|Emphasize learning, experimentation, and locally developed rules, and embrace change||Encourage learning and experimentation
Foster understanding of social-ecological systems as complex adaptive systems
|Improve governance and increase social capital||Build trust||Promote trust, social networks, and leadership||Broaden participation|
|Do unto others as you would have them do unto you||Include redundancy in governance structures and a mix of institutional types
Include all ecosystem services in development proposals and assessments
|Promote polycentric governance systems|
Maintaining Diversity, Variability, and Redundancy
Sophisticated financial investors know that diversification reduces risk, or to use a more biologically based metaphor, that it isn’t smart to “put all your eggs in one basket.” In socialecological systems, diversification can be accomplished through a variety of ways, including maintaining biodiversity or increasing economic diversification.
Although May (1974) showed that increased biodiversity could reduce system stability, recent theory and empirical work have led to the conclusion that increased biodiversity generally enhances ecosystem resilience and stability (Folke et al., 2004; Tilman and Downing, 1996). Biodiversity represents a form of insurance, providing redundancy that can buffer ecosystems against losses of particular species and can reduce the chance that important ecosystem functions will be compromised (McCann, 2000; Naeem, 2002). Greater diversity also increases the probability that there is a species present that is well suited to particular conditions, thereby allowing a high level of function under varying conditions (Chapin et al., 1995; Isbell et al., 2011). Redundancy within functional groups such as primary producers is important (Levin et al., 1998), as is response diversity, which means there is a broad range of species-specific responses to perturbations (Elmqvist et al., 2003; Folke et al., 2002). Greater biodiversity should therefore increase the probability that a system will provide a consistent level of performance over a given time (Naeem and Li, 1997).
The relationship between biodiversity and ecosystem resilience may take different forms, depending on the underlying mechanism (Figure 3.3). There are three possible models for the relationship between biodiversity and ecosystem resilience: (a) the linear model; (b) the keystone model, in which the presence of one or a few keystone species is crucial for ecosystem resilience; and (c) the redundancy model, in which the functions that maintain ecosystem stability are duplicated by numerous species. The available evidence favors the redundancy hypothesis (Cardinale et al., 2011). However, species redundancy decreases over time as species sort into niches (Reich et al., 2012), and certain species may be critical for ecosystem functioning under particular environmental conditions (Isbell et al., 2011).
Although greater diversity tends to lead to greater system resilience, some ecosystems are highly resilient even though they are not particularly diverse, such as coastal wetlands. Plant biodiversity and biomass decrease with increasing salinity moving from tidal freshwater wetlands to salt marshes (Wieski et al., 2010). Salt marshes on the GoM coast are dominated by the smooth cordgrass Spartina alterniflora. Despite low plant biodiversity, salt marshes are extremely resilient, even in the face of the DWH oil spill (Silliman et al., 2012). Salt marshes exemplify extreme environments characterized by hypoxic soils, high salinity, and high sulfide concentrations (soluble sulfide concentrations of 1–5 millimolar in soil are lethal to most plant life). S. alterniflora also tolerates organic toxins (DeLaune et al., 1984; Li et al., 1990; Webb et al., 1985), which partially explains their rapid recovery following oil spills.
FIGURE 3.3 Conceptual models of possible alternative responses of ecosystem resilience to biodiversity. SOURCE: Modified from Naeem (2002).
Enhancing economic diversity is another route to increasing resilience of social-ecological systems. Communities whose economies are heavily dependent on specific natural resources are vulnerable to shocks to those resources. Highly diversified economies, such as biodiversityrich ecosystems, can weather shocks to particular resources or particular industries, much as ecosystems can lose some species and still maintain overall system performance (Briguglio et al., 2009; Rose, 2007). Sectoral diversification is expected to have a positive influence on economic resilience (Simmie and Martin, 2010).
The reliance of its economy on specific natural resources makes the GoM region vulnerable to shocks such as the DWH oil spill. Fishing grounds were closed, and recreational and tourism activities suffered large losses in 2010 after the spill. Some communities along the Gulf Coast are almost totally reliant on the oil and gas industry, fishing, recreation, and tourism. When these industries experience downturns, the economies of these communities suffer. In the case of the DWH oil spill, however, much of the economic harm—though serious—was temporary.
The immediate aftermath of the spill in 2010 was very difficult for many GoM communities (see Box 2.2 in Chapter 2). However, oil and gas production, fishing, and tourism largely rebounded in 2011. Drilling was allowed to resume, fisheries were reopened, and tourists returned.
Managing Feedbacks and Slowly Changing Variables
Strong negative feedbacks increase system resilience by offsetting the effect of a disturbance that pushes a system away from equilibrium. Negative feedbacks exist in ecological systems in which biological organisms perpetuate conditions favorable to their continued existence. In social-ecological systems, actions aimed at maintaining existing conditions can provide negative feedbacks. For example, fisheries management that adjusts harvest quotas based on current population size can help to stabilize fish populations by reducing harvests when stocks are low and increasing harvests when stocks are high (Reed, 1978, 1979). However, maintaining stable stocks can mean highly variable harvests from year to year. But maintaining constant harvests, especially if set to maximize yields in an average year, can lead to fishery collapse (Roughgarden and Smith, 1996). Managing systems to maintain stability for certain outcomes, such as harvests, can reduce stability in other dimensions and can affect overall system resilience (Gunderson and Holling, 2002; Gunderson and Pritchard, 2002).
Long-term changes or trends in environmental or social conditions (“slow variables”) can also have a large influence on resilience. Much as an individual whose general health is declining is more susceptible to disease, ecosystems compromised by ongoing stress may be more vulnerable to rapid change from disturbances (Gunderson and Holling, 2002; Walker and Salt, 2006). For example, levees and channels to reduce flooding along the Mississippi River have changed the supply of sediment to coastal systems and have made parts of Louisiana more susceptible to damage from coastal storm surges (Costanza et al., 2006).
Modularity and Connectivity
Connections among components within a system can allow shocks to propagate through the system so that tightly coupled systems may be more vulnerable to systemwide risks (May et al., 2008). Building in a degree of modularity may be important for preserving overall resilience (Levin, 1999). Placing booms to limit the spread of oil into coastal marshes is one example of modularity in practice. However, marine systems are inherently highly interconnected, and sometimes interconnections provide resilience, such as when local populations are reestablished through recolonization after local extinction events. An objective of an ecosystem services approach is to understand interconnections, but our ability to manage connectivity/ modularity in practical situations is often limited.
Adaptive Management and Learning
If we knew exactly what the future had in store, then it would be easier to plan for it. But as Yogi Berra said, “It’s tough to make predictions, especially about the future.” In adaptive man-
agement, managers try to gain information through experimentation and learning to reduce uncertainty and improve ongoing management (Holling, 1978; Lee, 1993; Walters, 1986). Adaptive management has proven difficult to implement in practice, in part because it involves risk taking that can put managers in a difficult position of justifying failures even when it provides valuable information, as well as because it requires resources for ongoing monitoring and evaluation (Lee, 1999).
Of particular importance to managing complex systems is having some ability to predict regime shifts prior to their occurrence. There are some signals of the imminent onset of a regime shift (Scheffer et al., 2009, 2012). Whether these signals can be received in time, and management revised to forestall a regime shift, is doubtful (Biggs et al., 2009). The potential for a harmful regime shift in the future should influence current management. In most cases, this potential shift should cause management to take precautionary actions to reduce the probability of its occurrence (Polasky et al., 2011). However, in some cases the impact on management of a potential shift could work in the opposite direction. For example, the threat of collapse of a resource stock may cause resource managers to use the resource more aggressively (“use it or lose it”; Reed, 1984, 1988).
Improving Governance and Increasing Social Capital
The ability of a social-ecological system to recover from shocks such as a hurricane or an oil spill is improved by having highly functional institutions and a high degree of social cohesion. Disasters that inflict large-scale and widespread damage put ecosystems and human communities under great stress. Help from outside the stricken area, such as from the federal government, is often essential to provision of relief in the immediate aftermath of a disaster. But long-term recovery depends in large part on the resources and resourcefulness of the local community (Picou and Martin, 2006). Good governance also aids in assessing potential threats and long-term stresses and in bringing about changes in management or behavior to address them (Folke et al., 2010). Trust in institutions is a key variable in attaining cooperation to make changes, especially if the change requires at least some short-term sacrifice. Lack of good governance or social capital can lead to social, economic, or ecological decline and a downward spiral in environmental conditions and human well-being (Folke et al., 2010).
• Although systems are complex, it may be possible to manage them in ways that increase system resilience. Management itself can enhance system resilience if it creates responses to current conditions in ways that lessen the impacts of disturbances. The effectiveness of management can be enhanced by improving understanding of system dynamics, reducing uncertainty, and developing better early warning signals of approaches to thresholds.
• Management for specific resilience uses knowledge of specific types of disturbances to design plans to minimize damages and promote recovery.
• Management for general resilience is needed when the type of disturbance is unknown or when many types of disturbances are possible. Conserving biodiversity or increasing
economic diversity can make social-ecological systems more resilient to a range of different of disturbances, as can the other management approaches summarized in Table 3.1. The emphasis on properties to promote general resilience is important given the inability to predict the type, timing, scale, and interaction of future disturbances.
FEDERAL GOVERNANCE OPTIONS AND POLITICAL LIMITATIONS
Although resilience management can result in many benefits, translation of resilience theory into improved agency decision making for ecosystem service management will be difficult. Because resilience management represents a significant change from current management priorities or decision-making processes, congressional authorization in the form of new or amended laws (Ruhl, 2012) may be necessary to make it part of the process. A fundamental principle of federal administrative law, rooted in Article I of the U.S. Constitution, is that agencies cannot act in ways that exceed the statutory authority Congress has given to them (Doremus, 2001). Whether resilience management is consistent with current resource management laws depends upon current statutory authority and the details of resilience management. For example, an agency committed to incorporating resilience management would need to
• craft a firm definition of resilience for the system in question;
• identify a means of measuring the current state of resilience;
• build a model capable of predicting to some degree of certainty the resilience effects of specific decisions (or to design management experiments that could lead to the development of predictive capacity); and
• explain both the process and the results to the public and to the judges likely to review the agency’s decisions (Allen et al., 2011).
In many ways, the challenges that agencies face in resilience management are similar to those in adaptive management. As J. B. Ruhl wrote: “For adaptive management to flourish in administrative agencies, legislatures must empower them to do it, interest groups must let them do it, and the courts must resist the temptation to second-guess when they do in fact do it” (Ruhl, 2012).
Some federal agencies, notably the U.S. Forest Service, the U.S. Fish and Wildlife Service, the Bureau of Reclamation, and the National Oceanic and Atmospheric Administration, have begun to recognize the importance of resilience management and to incorporate it into their broader policy objectives (Benson and Garmestani, 2011). President Obama’s 2010 Executive Order 13,5471 seeks to “improve the resiliency of ocean, coastal, and Great Lakes ecosystems, communities, and economies.” It remains to be seen how agencies will translate these general policy statements into on-the-ground decision making.
Implementing resilience management could, in theory, be easier in the marine environ-
1 Exec. Order No. 13,547, 75 Fed. Reg. 43023 (July 22, 2010).
ment than in terrestrial environments. Unlike terrestrial systems, the oceans are entirely public and are not interspersed with private holdings. The federal government thus has more freedom both to set and adjust regulations. In federal waters outside of state jurisdiction, federal agencies could manage all resource use holistically to attain broad systemic goals. Finally, the laws pertaining to resource use in the oceans have traditionally been designed to function under changing environmental conditions and substantial uncertainty. The Magnuson-Stevens Fishery Conservation and Management Act (MSA) requires regular input of scientific information and annual consideration of management measures, which enable decision makers to respond to environmentally driven stock fluctuations (Carlarne and Eagle, 2012).
Still, resilience management in the GoM ecosystem faces challenges. The United States lacks a legal tradition of managing complex systems as multiple, interacting systems. Congress has traditionally written laws that address resources individually; thus, the MSA governs the use of fisheries, the Outer Continental Shelf Lands Act regulates oil and gas and mineral use, and so on (Eagle, 2006). The resource-by-resource approach is less amenable to resilience management than is a place-based approach. That approach, in which a single agency has jurisdiction over most resource use within a single place (e.g., a national park), permits an agency to manage an integrated system. In addition, the GoM ecosystem is affected by actions taken on land that influence nutrient and sediment inputs and are controlled by agricultural and other policies (Rabalais et al., 2002, 2007).
Finally, although there are no private property interests in federal Gulf waters, there are private interest groups that would likely oppose significant changes to the current system of resource management. Such groups might fear that the “new” regulatory regime would result in significant reductions in their income or in their ability to shape management decisions. Because these groups have the incentives and resources to oppose change at all levels, a management shift would need to be accompanied by an explanation of why its benefits would outweigh the real or perceived costs (Ruhl, 2012).
Finding 3.6. Although the maintenance of resilient ecosystems has many benefits, including provision of a steady flow of ecosystem services, political, legislative, and institutional barriers may impede implementation of resilience management.
Option: Implement a Portfolio Approach to Management
How might governments manage publicly owned resources in a way that minimizes disruption to the flow of ecosystem services during and after a disaster? One approach is to move away from the current legal approach to managing marine ecosystems and toward a set of laws that aim to manage distinct, defined areas of the sea for narrower, specific objectives.
The direct management of marine life populations in federal waters of the GoM currently occurs pursuant to two laws: the MSA and the Marine Mammal Protection Act (MMPA). Although the efficacy of these laws in achieving their stated objectives is debatable, it is clear that each law uses a single, science-based concept aimed at achieving a single goal with
respect to the GoM marine species under its auspices. The MSA attempts to achieve “sustainable fisheries” using the science-based concept of “optimum yield,” while the MMPA attempts to maintain an “optimum sustainable population level” using the science-based concept of “potential biological removals.”
Because of the substantial scientific uncertainty involved in estimating necessary numerical targets, such as optimum yield and potential biological removals, it might be possible to increase the resilience of marine ecosystems by diversifying management strategies and objectives on a geographical or stock-by-stock basis. So, Congress could, for example, modify the MSA and the MMPA so that managers would be required to apply a range of strategies over a set of geographically defined management areas or over a set of distinct fish or mammal populations. Some areas or species could be managed using more conservative versions of the traditional optimum yield or potential biological removal tools, and some areas or species could be managed using entirely different approaches and with different ends in mind. So, for example, one could imagine managing a particular area for “ecosystem integrity” rather than “sustainable fisheries.”
The theory underlying this diversification approach is based on well-accepted principles that economists use to guide investors in the business world. The idea is that, in a realm characterized by substantial uncertainty, such as the stock market or marine ecosystems, a “portfolio approach” to investing can help to achieve an optimal combination of risk and return. As Gordon Munro observed, fishery (or ecosystem) management is similar to other forms of capital management:
Economists view capture fishery resources, as they do all natural resources, as a form of natural capital, assets that are capable of yielding a stream of economic returns (broadly defined) through time. Since fishery resources are capable of growth (like forests, but unlike minerals), these resources—natural capital—can be managed on a sustained basis, essentially by skimming off the growth through harvesting. This also means that the resources can provide economic benefits to society indefinitely. It means further that one can, within limits, engage in positive investment in the natural capital by harvesting less than the growth. (Munro, 2008, p. 14)
Designing and implementing a portfolio-based law for managing marine ecosystems such as the GoM is a challenging, but not impossible, task. Conceptually, such laws offer one of the most promising approaches to enhancing system resilience and thereby reducing the impacts of future human-engineered or natural disasters on the people who rely on those systems.
Finding 3.7. One option for facilitating resilience management is employment of a portfolio approach to designing the system of laws regulating use of the Gulf of Mexico. Under a portfolio approach, different management goals and strategies would be applied to discrete geographic ocean and coastal areas. Such an approach, as compared with an aspatial management strategy that tries to ensure that all similar systems provide equal amounts of all services, provides a buffer against uncertainties, including future large-scale disturbances. Implementation of this portfolio approach would require congressional action because a set of new statutes would be needed.
Ecosystem Restoration and Resilience
Following an oil spill or other disturbance to an ecosystem, a key management objective is ecosystem restoration. Ecosystem restoration after a disturbance and ecosystem resilience are closely linked. Systems with low resilience may recover slowly or switch to a different regime and fail to return to original conditions. In the case of the DWH oil spill, of particular interest is how well restoration efforts will work to recover ecosystems to pre-spill conditions, both in terms of restoring the provision of ecosystem services and increasing resilience to further disturbances and ongoing stresses.
Much has been written about the restoration of the Mississippi River Delta (Boesch et al., 1994, 2006; CPRA, 2012; Day et al., 2007; USACE, 2009). Restoration efforts in the delta are taking place in the midst of long-term changes as well as periodic disturbances. Coastal wetlands in the delta have failed to maintain elevation relative to mean sea level because of an inadequate supply of sediment. Prior to European colonization and the engineering of the Mississippi River, sediment-laden water routinely spilled over the river’s natural levees during flood events and nourished the wetlands. This process has been disrupted by armoring levees to prevent flooding, damming tributaries in the river basin, dredging canals, and other alterations that reduce sediment delivery to the delta and alter its hydrology. Since the 1930s, the delta has lost more than 4,800 km2, an area nearly the size of Delaware (Barras et al., 2003; Dunbar et al., 1992) and could lose an additional 1,600 km2 by the year 2050 (Louisiana Coastal Wetlands Conservation and Restoration Task Force and Wetlands Conservation and Restoration Authority, 1998). The U.S. Army Corps of Engineers (USACE) identified 2,460 km2 of marsh or open water in five planning units within the lower delta requiring some 5.3 billion cubic yards (4 km3) of sediment for restoration (USACE, 2009). A report by the National Research Council (NRC, 2009) noted, “If wetlands cannot be maintained, this implies that decision makers and citizens ultimately will have to make hard choices about where restoration can take place and where it cannot.”
Much less is known about restoration efforts in other types of ecosystems. It may be difficult or impossible to fully restore some ecosystems after disturbances. Changes in landscape structure, loss of native species or invasion by exotics, changes in species dominance hierarchies, trophic interactions, and biogeochemical processes may prevent returns to pre-disturbance conditions. Moreover, restoration efforts can be altered by feedbacks among biotic and physical processes (Ehrenfeld and Toth, 1997; Suding et al., 2004). Restoration in the absence of knowledge of these feedbacks may launch an ecosystem down an unpredictable trajectory.
Restoration of ecosystem services does not necessarily follow from restoration of the ecosystem structure or the return of a habitat to a former state. Zedler (2000) concluded that numerous variables, including landscape setting, habitat type, hydrological regime, soil properties, topography, nutrient supplies, disturbance regimes, invasive species, seed banks, and declining biodiversity, can constrain the restoration process in wetlands. Zedler also concluded, “although many outcomes can be explained post hoc, we have little ability to predict the path that sites will follow when restored in alternative ways, and no insurance that specific targets will be met” (Zedler, 2000). However, this predictive capability is precisely what we seek in an ecosystem services approach.
There is no existing roadmap for restoring all ecosystem services and functions. Currently, the best we can hope to accomplish are a few practical goals. A target for a coastal wetland might be restoration to an elevation that is more resilient and that raises productivity and its value as wildlife habitat, knowing that it may not be practical or even possible to restore its original species composition. Conceptually, an ecosystem might cross a threshold in transitioning from one ecological state to another, such that restoration to a previous state is impeded by biotic and abiotic barriers (Figure 3.4). Examples of biotic thresholds could be the invasion of an exotic species (e.g., the common reed, Phragmites spp.) that excludes less competitive species, or a disease (e.g., chestnut blight) that eliminates a dominant member of the biological community. Examples of abiotic thresholds include saltwater intrusion into a freshwater wetland as a consequence of rising sea level. Whether recovery of ecosystem services or restoration of the ecosystem is the goal, the endpoints must be defined in terms of practical metrics
FIGURE 3.4 Summary of the state (boxes) and transition (arrows) approach to ecosystem degradation and restoration. Hypothesized thresholds, indicated by vertical bars, may prevent recovery from a more degraded state to a less degraded state. SOURCE: Redrawn from Hobbs and Cramer (2008).
that can be monitored and a plan of action that is grounded in a thorough knowledge of the ecosystem.
Finding 3.8. Restoration aimed at returning a system to pre-disturbance conditions is closely related to resilience. Systems that cross critical thresholds after a disturbance can be difficult to restore, and restoration in the absence of a detailed understanding of system dynamics may launch an ecosystem down an unpredictable trajectory. Whether recovery of ecosystem services or restoration of the ecosystem is the goal, the endpoints must be defined in terms of practical metrics that can be monitored.
The definitions of engineering and ecological resilience provide a basis for designing measures of resilience. Engineering resilience is the speed of recovery or the decay rate of the perturbation. Engineering resilience can be quantified post hoc as the time required to return to a pre-disturbance state after an experimental or natural disturbance. In natural systems with sufficiently rapid recovery, such as the sessile biota of the rocky intertidal, monitoring of recolonization can be done after disturbance (Conway-Cranos, 2012). One could also attempt to measure the speed of recovery of social-ecological systems after disturbances. For example, how fast or complete is recovery within a community or an industry after a shock such as a hurricane or an oil spill?
Engineering resilience can also be modeled with various degrees of mathematical sophistication (DeAngelis, 1980; Neubert and Caswell, 1997; Pimm and Lawton, 1980). With a model of system dynamics, an estimate of engineering resilience may be simulated for various severity levels of disturbances.
Although straightforward in theory, measurement of engineering resilience presents at least three practical challenges. First, there is the question of what particular measures or metrics will be used. Some system components or ecosystem services may recover more quickly than others. For example, species diversity may well return to a previous state at a different rate than does primary productivity, or not at all. Hence, quantifying resilience may be heavily influenced by the choice of what is measured. Second, measurements of the speed of return require accurate baselines of conditions prior to the disturbance. The issue of baselines was discussed in the Interim Report (NRC, 2011) and is discussed further in Chapter 5. Third, disturbances that differ in type or scale are likely to result in quantitatively or possibly qualitatively different degrees of responses relevant to resilience (Carpenter et al., 2001). Systems may be highly resilient to some kinds of shocks but not others.
Measurement of ecological resilience is a more complex matter. Social-ecological systems are often too complex to enable researchers to accurately model potential regime shifts or to have much confidence in simulation results. However, attempts to capture critical indicators of ecological resilience have been made. Borrowing the concept of “critical slowing down” from physics, an indicator of ecological resilience might be the rate of recovery from small perturba-
tions when the recovery rate decreases as a regime shift is approached (Scheffer et al., 2012; van Nes and Scheffer, 2007). Changes in other system properties, such as variance, spatial correlation, autocorrelation, and skewness, may also prove useful in detecting approach to critical thresholds (Scheffer et al., 2012). But this field is quite new and does not yet have a body of well-established results to guide management or policy.
Other research has explored the concept of functional diversity (Allen et al., 2005), response diversity (Petchey and Gaston, 2009), and population densities in stochastic systems (Ives, 1995). These approaches require species-level ecological information about function and response. For some groups of organisms, such as birds, this information may be inferred from size and morphology (Cumming and Child, 2009).
When a system is well understood, it may be more straightforward to define appropriate metrics for ecosystem resilience. With respect to tidal wetlands, for example, it seems prudent to adopt practical metrics that are inclusive of numerous ecosystem services and overall ecosystem function. Practical metrics with these characteristics would be (1) change in total wetland area by plant community type and (2) relative elevation (relative to mean sea level). Similarly, critical components of the ecosystem service under study that can be used to measure resilience should be identified. The ability to do this links directly to the understanding of the ecological production functions associated with the service and links directly back to the understanding of ecosystem dynamics.
Finding 3.9. The measurement of resilience poses a number of conceptual and practical challenges. Measures of speed of recovery to pre-disturbance conditions (engineering resilience) depend on having good baseline data and may vary depending on what ecosystem service or system component is measured and what type and severity of disturbance is considered. Measuring how likely a system is to cross a critical threshold and undergo a regime shift (ecological resilience) is a topic that is at the frontier of science. Consensus on practical measures that can be used to predict the location of critical thresholds or the probabilities of regime shifts does not yet exist. However, for specific systems, it may be possible to define a set of metrics that measure key conditions or processes linked to system dynamics that can predict the resilience of the system and the return of provision of ecosystem services.
Resilient ecosystems can lead to the resilience of specific ecosystem services. However, ecosystems typically generate multiple ecosystem services. Changes in ecosystem structure and function will generally not affect all ecosystem services in the same manner. Therefore, different services could have different outcomes after a disturbance. As noted by Carpenter et al. (2001), evaluation of a system’s resilience requires identification of the initial state of the system (regime) and the disturbances that can impact that system. If the system is viewed through the lens of ecosystem services, then there is a need to identify the most important services and how they may be affected by potential alternate states of the system. For example, coastal marsh habitat of the GoM provides several significant services, two of which are storm surge protection and fishing opportunities, both recreational and commercial. The quality of these services is directly linked to the structural condition of the marsh. Fragmented marsh,
characterized as having a high ratio of edge to area, provides an environment for high-quality recreational fishing and also habitat for commercially important species, but it is less effective for storm protection. In contrast, continuous or connected marsh provides higher quality storm surge protection, but it is not as good for recreational and commercial fishing (Minello et al., 1994; Peterson and Turner, 1994; Zimmerman et al., 2002). The quality of each of these services is strongly connected to two alternate states of marsh habitat. Focusing on the resilience of a particular state of a habitat (e.g., fragmented versus continuous marsh) can also lead to resilient ecosystem services that are connected to a specific state of the habitat. As a society we would prefer having more of all ecosystem services, but given that only one condition can occur at one time, there will be tradeoffs between services.
This page intentionally left blank.