A Decision Framework for Interventions to Increase the Persistence and Resilience of Coral Reefs (2019) / Chapter Skim
Currently Skimming:
4 Illustrative Model of Decision Tools
4 Illustrative Model of Decision Tools
Pages 83-114
The Chapter Skim interface presents what we've algorithmically identified as the most significant single chunk of text within every page in the chapter.
Select key terms on the right to highlight them within pages of the chapter.
From page 83... ...
The committee uses a coral reef community model to simulate the effects of two example interventions on a simplified coral reef system: assisted gene flow and atmospheric shading. As part of this analysis, the committee illustrates the use of Bayesian networks to evaluate the impact of the interventions on management objectives in a probabilistic manner, represented by a range of model outcomes resulting from uncertainty around climate projections, intervention efficacy, and intervention risk.
|
From page 84... ...
The committee's model serves as a communication tool to illustrate how the problem of identifying and choosing among restoration and intervention options to build reef resilience under climate change could significantly benefit from the use of quantitative models to inform the decision-making process. Quantitative models and associated nalyses a can open the door for reef managers and policy makers to assess and compare benefits and risks of available intervention options, and thereby make more informed strategy choices under uncertainty.
|
From page 85... ...
The two broad categories of coral morphology represent extremes in the range of possible coral life history strategies and responses to thermal stress in a community (Darling et al., 2012; Loya et al., 2001) and a diversity in response strategies to disturbance among community members (high stress tolerance versus fast regrowth)
|
From page 86... ...
The focal stressor is sea surface warming driven by climate change; for simplicity of this illustration, additional possible global-change-dependent stressors such as ocean acidification (OA) and stronger storms are ignored.
|
From page 87... ...
Blue circles are mortality observations from the 2005 mass bleaching event in the Caribbean (Buddemeier et al., 2011)
|
From page 88... ...
. The external recruitment can provide initial insight into the role of connectivity between locations, but the focus on a single location does not explicitly account for how variability in conditions and climate impacts across locations might affect recruitment, including recruitment declines that might arise from anthropogenic impacts such as pollution and sedimentation as well as coral regional declines (Richmond et al., 2018)
|
From page 89... ...
. Projections of coral reef persistence into the future under climate change without adaptation typically predict coral collapse regardless of future climate scenario (i.e., under committed climate change)
|
From page 90... ...
over a particular historic time window (shaded orange area) , calculate the mean value (solid red line with a dashed orange line highlighting the value calculated FIGURE 4.1.1 Illustration of the approach to calculate thermal stress experienced as degree heating months (which can be converted to degree heating weeks)
|
From page 91... ...
ILLUSTRATIVE MODEL OF DECISION TOOLS 91 within the window, "maximum of the monthly mean" [MMM] climatology)
|
From page 92... ...
scenarios developed by the Intergovernmental Panel on Climate Change (Collins et al., 2013)
|
From page 93... ...
. A rolling window of 80 years was chosen as the default natural adaptation case because this window predicts high-frequency bleaching in severe but not more moderate climate scenarios (Logan et al., 2014)
|
From page 94... ...
The different modeling frameworks described in Table 3.2 (and the "Relevant Mechanistic Modeling Framework" column of Table 4.1) can more mechanistically represent many of these risks and benefits for improved predictive ability.
|
From page 95... ...
) Increase thermal Managed selection, Narrow rolling Genetic model tolerance via managed breeding, genetic window for genetic adaptation manipulation, assisted calculating the gene flow DHW value in hX(τ)
|
From page 96... ...
For the purpose of this modeling study, two example interventions that affect different processes in the system have been chosen for illustration: assisted gene flow (for increased thermal tolerance) and atmospheric shading (for reduced exposure to thermal stress)
|
From page 97... ...
Mechanistic models can account for uncertainty in whether anticipated benefits are realized. Atmospheric shading interventions cool the sea surface over coral reefs locally, regionally, or globally (Gattuso et al., 2018)
|
From page 98... ...
The addition of the transported coral fragments or recruits with assisted gene flow also adds a small amount to coral cover, which is added to the coral recruitment rEX,X (10% increase in value) for each focal coral X
|
From page 99... ...
. The herbivory and macroalgal growth rate values are set at model initialization to represent the local conventional management context, while the parameter changes associated with atmospheric shading or assisted gene flow change in a specified "deployment year" to represent additional interventions implemented in the local context.
|
From page 100... ...
Furthermore, responses to new interventions are in many cases uncertain because the research and development necessary to establish how the environment or the ecosystem responds to intervention has not been carried out. However, because the purpose of this example is to provide an illustration of assessing the scope that new interventions might have in improving coral condition or preventing coral loss, the focus is on relative impacts rather than projecting absolute reef states that depend on precise parameter estimates.
|
From page 101... ...
ILLUSTRATIVE MODEL OF DECISION TOOLS 101 TABLE 4.2 Summary of Symbols, Functions, and Default Parameter Values Used in the Model Symbol Unit Interpretation Range Source CF Prop Area covered by fast-growing corals 0-1 -- CS Prop Area covered by slow-growing corals 0-1 -- M Prop Area covered by macroalgae 0-1 -- P Prop Proportion habitat available 1 -- gF yr–1 Growth rate of fast-growing corals 0.5 Anthony et al., 2011; Fung et al., 2011 gS yr–1 Growth rate of slow-growing corals 0.3 Fung et al., 2011 γ yr–1 Growth rate of macroalgae 0.8 Fung et al., 2011 aF yr–1 Rate of macroalgae overgrowing fast- 0.05 Fung et al., 2011 growing corals aS yr–1 Rate of macroalgae overgrowing slow- 0.07 Fung et al., 2011 growing corals α yr–1 Rate of fast-growing coral overgrowing 0.03 Tanner et al., slow-growing corals 1994 mF yr–1 Base rate mortality for fast-growing 0.1 Madin et al., corals 2014 mS yr–1 Base rate mortality for slow-growing 0.05 Madin et al., corals 2014 v yr–1 Baseline grazing rate on macroalgae 0.4 Fung et al., 2010 rEX,F Prop External recruitment of fast-growing 0.001 Fung et al., 2010 corals rEX,S Prop External recruitment of slow-growing 0.001 Fung et al., 2010 corals rEX,M Prop External recruitment of macroalgae 0.001 This study rIN,F nd Internal recruitment of fast-growing 0.005 Fung et al., 2010 corals rN,S nd Internal recruitment of slow-growing 0.005 Fung et al., 2010 corals rIN,M nd Internal recruitment of macroalgae 0.005 This study p1,F nd Shape parameter 1 for fast-growing 0.02 This study corals p2,F nd Shape parameter 2 for fast-growing 0.28 This study corals continued
|
From page 102... ...
is replicated for a set of conditions, specifically early versus late deployment years, low TABLE 4.3 Summary of Conditions and Interventions at Increasing Levels of Intensity Used in Model Simulations as Part of the Strategy Design Levels Conditions Climate change scenario RCP2.6, RCP8.5 Start state (total coral cover) 5%, 30% Deployment year 2025, 2035 Interventions Algal growth (stimulated by nutrient load)
|
From page 103... ...
Figure 4.5 illustrates a selection of example trajectories from the model with varying levels of RCP scenarios, conventional management, and assisted gene flow. Under RCP2.6, low grazing and high nutrients (driving algal growth)
|
From page 104... ...
, and low versus high rates of assisted gene flow (AGF)
|
From page 105... ...
BAYESIAN NETWORK ANALYSES The large number of model outputs (coral cover over time based on 192 combinations of conditions and interventions) produced by the model means that general conclusions cannot be drawn easily from inspecting the many trajectories of outcomes for corals and macroalgae.
|
From page 106... ...
The following examples explore how selecting fixed coral cover as a simple objective, under selected timeframe and climate scenario conditions, influences the relative importance of other parameters, such as the need for intervention or improved management of local stressors. Example 1: What Will Be Required to Sustain Coral Cover Under Substantial Greenhouse Gas Mitigation (RCP2.6)
|
From page 107... ...
Example 3: What Will Be Required to Sustain Corals on the Long Time Horizon Under Business-as-Usual Greenhouse Gas Emissions? To model the specific conservation goal of sustaining high coral cover in the long term under the business-as-usual climate change scenario, the
|
From page 108... ...
108 A DECISION FRAMEWORK FOR CORAL INTERVENTIONS FIGURE 4.6 High-level summary result of the Bayesian network, integrating and synthesizing all outputs of the dynamic coral reef model. Results show distributions of conditional probabilities of global and local conditions and interventions needed to achieve the objective of achieving coral cover greater than 20%.
|
From page 109... ...
Importantly, similar to the general RCP2.6 scenario (see Example 1) , there is twice the chance of sustaining high coral cover under severe climate change by 2060 under a high compared to low rate of algal grazing, a component of conventional management.
|
From page 110... ...
Option C represents a strategy that deploys a high rate of assisted gene flow in combination with best practices conventional management. This is achieved by selecting the 40-year rolling climatology (AGF High)
|
From page 111... ...
This outcome will inevitably vary with parameterization of the amount by which shading dampens and assisted gene flow accelerates thermal tolerance adaptation, and the likelihood of shading failure. Therefore, the illustration highlights the need to explore and resolve this potential antagonistic interaction with a more mechanistic model, such as a genetic model in which the evolutionary outcome emerges from the thermal stress rather than assuming an adaptive rate.
|
From page 112... ...
First, conventional management on its own would very likely not be enough to sustain coral condition. S econd, while new intervention strategies have the scope to build critical coral resilience under business-as-usual climate change, their success requires sustained or intensified conventional management.
|
From page 113... ...
In addition to varying model inputs, different types of model outputs and analyses can inform decisions as they might depend on stakeholder and manager risk tolerance. For example, a model output and analysis focused on minimizing the likelihood of a risk, such as minimizing the probability of particularly low reef state, might be more relevant to a risk-averse manager.
|
From page 114... ...
Conclusion: A successful modeling framework requires substantial effort in tailoring model structure and parameters to the decision context, risks and benefits of the interventions under consideration, and local environmental conditions and reef ecosystem dynamics. As demonstrated by the committee's illustrative effort, the utility and payoff of this approach are the ability to identify • The conditions necessary for new and potentially risky inter ventions to outperform the no-action alternative under differ ent future climate scenarios.
|
Key Terms
This material may be derived from roughly machine-read images, and so is provided only to facilitate research.
More
information on Chapter Skim is available.