6.4 Decision-making

6.4.1 Fischer et al, 2009

“Integrating resilience thinking and optimisation for conservation” (Fischer et al. 2009)

Key contribution: This article discusses the interface between resilience thinking (a perspective that embraces complexity and nonlinearities) with optimization (explicity acknowledgement of resource constraints on decision making), which are too voluminous literatures that are only beginning to become integrated.

Key notes:

“Resilience thinking” (perspective):

• provides integrated perspective for analysis that emphasizes nonlinearities within a system as well as interdependency of social and ecological systems
• emphasizes interface of gradual changes with punctuated disturbances, feedbacks, alternative stable states, cross-scale relationships & regime shifts

“Resilience” - capacity of a system to absorb disturbance and reorganize under change such that it restores its fundamental function, structure, identity and feedbacks

“Optimisation for conservation” (tool):

• An “outcome-oriented tool” that functions under resource scarcity to promote rational and transparent decision making

Integrating the two:

• Given that one is a perspective and the other is a tool, they may be relevant at different stages of a project or research program and thus may be integrated relatively efficiently
• Resilience thinking may be particularly useful in defining boundaries of a system (spatial and temporal scales as well as key actors or variables in a system)
  • particular importance is involvement of key actors due to focus on social-ecological system
  • thresholds are unknown thus may be seen as working hypotheses that are updated with new information
• Thus, optimal solution is an interative process and may potentially change given new insights

The authors identify three themes worth particular attention:

1. Addressing social issues
2. Addressing uncertainties and interface with decision-making
3. Avoiding undesirable states that constrain reversibility

Social issues

Optimization may be well suited to handle some social issues (e.g., easily quantified finances), but less so for factors that are not easily quantified (human values, cultural practices). Can use optimization as part of social or political framework and analyze for lessons learned to understand these less-easily quantified factors, however.

Uncertainties

Uncertainties are common within social ecological systems and greatly complicate optimization techniques.

“Novel ecosystems” or co-occurrence of two or more species in a single place and time represents one such uncertainty-laden situation. Resilience thinking may help promote more effective science and management around such situations, for example focusing on “drivers” rather than “patterns”.

Avoiding undesirable states

Key challenge is to maintain desirable states and avoid undesirable ones that are difficult to get out of.

Social factors can complicate things that may otherwise appear straightforward to optimize. For example:

• Protection of reserves may appear straightforward at first, but may not respond to climate change or other shifting environmental parameters
• Use of corridors or other connectivity networks thus may be important, but ultimately may not account for social, economic or political factors that impact the efficacy of connectivity plans
• Ultimately need to consider conservation efforts both inside as well as outside the reserves, as well as the key social factors that impact them

6.4.2 Folke et al, 2010

“Resilience thinking: Integrating resilience, adaptability and transformability” (Folke et al. 2010)

Key significance: This study reviews resilience thinking and describes the concepts of “resilience”, “adaptability”, and “transformability” as they relate to it. Prompted largely by the issue in that dynamical systems theory doesn’t necessarily account for inherent dynamism in systems in that they change naturally over time.

Key notes:

“Resilience” as applied to ecological systems original came from Holling, 1973 and was related to capacity of a system to respond to stressors and yet retain underlying function, structure and characteristics.

“Regime shifts”" become “critical transitions” when a system passes to a different stable domain.

One of main limitations of dynamical systems theory is that they do not account for fact that underlying nature of systems may change with time. Thus, understanding of social-ecological systems requires other concepts.

Adaptability and transformability

“Adaptability” - capacity of actors in a system to influence resilience

“Transformability” - defined as capacity to create a fundamentally new structure when ecological, economic or social structures make existing system untenable

Interface between the two and resilience is complex. Strategies that increase adaptability for socially desirable qualities may increase vulnerability of social-ecological systems to undesirable states.

Specified vs. general resilience

“Specified resilience” - resilience in reference to particular aspects or components of a system that may respond to a particular set of sources or shocks; “resilience of what, to what?”

“General resilience” - resilience in reference to all sorts of shocks; coping with uncertainty in all ways

Important to consider both, as one often thinks in terms of specified resilience and thus may not be considering important or unexpected shocks.

Tranformability and resilience at multiple scales

Transformability may occur slowly as a result of underlying factors, or may be deliberately induced by social actions.

Oftentimes, large-scale transformations of a system are too complex or expensive to perform, but sequential actions can allow for feedbacks that move a system towards another, as well as learning processes as decisions are made.

Case studies of social ecological systems suggest transformations occur through three phases:

1. Being prepared for or even preparing the social ecological system for change
2. Navigating transition by making use of crisis as a window of opportunity for change
3. Building resilience of new social-ecological regime

Questions for consideration:

1. Are there deeper, slower variables in social systems, such as identity, core values, and worldviews that constrain adaptability?
2. What are the features of agency, actor groups, social learning, networks, organizations, institutions, governance structures, incentives, political and power relations or ethics that enhance or undermine social ecological resilience?
3. How can we assess social ecological thresholds and regime shifts and what governance challenges do they imply?

6.4.3 Polasky et al, 2011

“Decision-making under great uncertainty: Environmental management in an era of global change” (Polasky et al. 2011)

Key contribution: This article discusses the difficulty of environmental management under the pervasive uncertainties that exist due to global change. The authors review several strategies for informed decision-making that may help to alleviate challenges due to uncertainty.

Key notes:

The authors note that it is important to consider the potential for learning from decisions in addition to the potential impacts that may happen. Decisions that are made should be made such that they will inform iterative management as addtional understanding is achieved.

In the authors view, ideal decision-making approaches should clarify the effect that alternative decisions have on probable desirability of outcomes in terms of stated objectives.

Decision theory:

“Decision theory” requires information about probabilities of various outcomes under alternative management options, as well as the desirability of those outcomes.

In standard decision theory:

1. Uncertainty is represented by assuming many different states of the system, as well as the various probabilities that each of those outcomes will occur
2. Outcomes are joint products of action and state, which produces a conditional probability of outcomes given the action
3. The outcomes are expressed in common metrics, and general best “decision” is the one that maximizes the “utility” of an action

Advantage of standard decision theory is that it provides a clear and repeatable framework for generating a course of action that is based on empirical data.

Disadvantage is that the necessary data (i.e., all alternative states) is unlikely to exist.

Incorporating learning into the decision-making process may be one means of mitigating the lack of data: “forward-looking decision makers should take account of how current decisions might influence future conditions, future decisions and the probable impacts of decisions on current and future well-being.”

Idea that “decision-making” should be treated as an experiment.

Alternative methods to use in place of or beside decision theory:

Thresholds approach - focus attention on critical boundaries that may lead to persistence of a new alternative state (often times with major consequences)

Common example is the 2 degrees celsius threshold thought to be catastrophic GHG emissions.

Thresholds are rarely ever precisely known, and thus each action incorporates a certain degree of risk with it. Decision-making ultimately needs to decide on what levels of risk are acceptable.

In this context – those in power are ultimately deciding that the risks currently seen by climate change are “fine”.

Scenario planning - thinking creatively and systematically about complex futures.

Scenario planning does not incoporate probabilities, but rather models and explores potential alternative outcomes supported by data and simulations. Scenario planning can be helpful in conceptualizing the future.

The key weakness of scenario planning is it is difficult to assess the likelihood of any given scenario occurring.

Resilience thinking - focuses on critical thresholds for system performance, as well as the capacities to both adapt and shift to a new system should a former one become untenable.

Employs methods that use multiple perspectives to better-understand and identify uncertainty (similar to scenario planning).

Resilience thinking requires an iterative approach, as assumptions and actions are re-evaluated with increasing availability of information (i.e., “adaptation”).

Decision making for global change

Standard decision making theories will fail as likelihoods of alternative states are likely to be too subjective and will induce disagreement such that decisions are ultimately not taken. The alternative methods proposed in the study may be one route by which decision-making can be informed, without the necessity of assigning probabilities to particular outcomes or actions.

Decision-making around global change may be thought of as two phases with continuous feedbacks:

1. Scoping the problem as broadly as possible to expand imaginable states and outcomes
   • “Scenario planning” and “resilience thinking” can be effective here
2. Actually making decisions given current understanding
   • “Threshold approach” given that science often more robust in identifying vulnerabilities than predicting future

6.4.4 Johnson et al, 2017

“Knowing when (not) to attempt ecological restoration” (Johnson et al. 2017)

Key significance: This opinion piece examines the critical questions of:

1. How do we identify when community change represents a phase shift to a different ecological dynamic that warrants considering restoration?
2. How do we identify discontinuous phase shift and hysteresis that may render futile attempts at restoration?

Key notes: Ecosystems are dynamic and thus it is critical to understand and recognize change when considering or assessing the health or state of an ecosystem. This largely will have impacts on whether or not restoration attempts should be undergone.

The authors lay out two key considerations that must be undertaken prior to restoration:

1. Must interpret ecological change; typical fluctuations represent a properly functioning ecosystem that does not warrant restoration, whereas other less typical fluctuations may indicate loss of resilience or regime shift
2. Ecological change may involve hysteresis and thus consideration as to whether or not restoration will be effective must be taken

Characteristic length scale - spatial scale at which deterministic trends in ecological dynamic are most sharply in focus

CLS can be particularly useful for interpreting change in an ecological system.

Estimating CLS is based on the idea in that dynamic of an entire community can be approximated by dynamic of a single species given that the species is part of some larger network in the community.

In the case in which the attractor (mathematically, the set of numerical values) of the system does not change, can understand fluctuations in the single species to be representative of the community; however, if the attractor changes then the CLS will change.

“Phase space” - is space in which all possible states of the system are represented, with each possible state corresponding to a unique point in the phase space

Exemplifying species invariance

The study shows, for the first time, that in real ecological systems the selection of particular species does not matter and is representative of broader trends of that network within a community.

Although the exact methods are a bit unfamiliar to me, they use predictions of a time series compared with the actual time series to quantify error, and use an inflection point at which the error “levels off” as the CLS.

They investigate this via a marine fouling community, and show that 9 of 10 species are indicative of that network community, whereas the last has different ecology and is understood to be part of a different network.

The significance here is that the findings adhere to theory (that selection of species does not matter), as well as the fact that CLS can be estimated using short time series of data

Hysteresis example

In regards to hysteresis and the point of whether or not to undertake restoration, they provide an example of sea urchins, kelp beds, and lobsters in eastern Tasmania.

In urchin barrens, freezes on lobster harvesting or planting of kelp are unlikely to restore the system, thus restoration action is not being undertaken. Rather, resources are being put towards conservation of existing kelp beds through regulation of lobster harvesting in these spatially explicit areas.

The management actions are a result of understanding the dynamics of key species and functional groups rather than comprehensive understanding of every component of the system.

References

Fischer, Joern, Garry D. Peterson, Toby A. Gardner, Line J. Gordon, Ioan Fazey, Thomas Elmqvist, Adam Felton, Carl Folke, and Stephen Dovers. 2009. “Integrating resilience thinking and optimisation for conservation.” Trends in Ecology and Evolution 24 (10): 549–54. doi:10.1016/j.tree.2009.03.020.

Folke, Carl, Stephen R Carpenter, Brian Walker, Marten Scheffer, Terry Chapin, and Johan Rockstrom. 2010. “Resilience thinking: Integrating resilience, adaptability and transformability.” Ecology and Society 15 (4).

Polasky, Stephen, Stephen R Carpenter, Carl Folke, and Bonnie Keeler. 2011. “Decision-Making Under Great Uncertainty: Environmental Management in an Era of Global Change.” Trends in Ecology & Evolution 26 (8): 398–404. doi:10.1016/j.tree.2011.04.007.

Johnson, Craig R, Rebecca H Chabot, Martin P Marzloff, and Simon Wotherspoon. 2017. “Knowing When (Not) to Attempt Ecological Restoration.” Restoration Ecology 25 (1): 140–47. doi:10.1111/rec.12413.