Summary

Coral reef managers are faced with a new decision crisis: deteriorating environmental conditions are reducing the health and functioning of coral reef ecosystems worldwide. Recent episodes of sustained above-average water temperatures have increased the frequency of coral bleaching events and are associated with increased disease outbreaks. Carbon dioxide dissolution in the ocean is lowering the pH of seawater, which is slowly impairing the ability of corals to grow or maintain their skeletons via calcification. These growing threats compound the persistent local stresses coral reefs have experienced for decades from pollution, overfishing, and habitat destruction. A growing body of research on “coral interventions” aims to increase the ability of coral reefs to persist in these rapidly degrading environmental conditions. These new tools are needed because established approaches for managing coral reefs are neither sufficient nor designed to preserve corals in a changing climate. Coral interventions that address the impacts of ocean warming and ocean acidification are part of a three-pronged approach for coral reef management that crucially also includes the mitigation of greenhouse gas emissions and the alleviation of local stressors.

New coral interventions include activities that affect the genetics, reproduction, physiology, ecology, or local environment of corals or coral populations with the goal of enhancing their persistence and resilience in degraded environmental conditions. They build on a growing understanding of how the coral holobiont—the coral, its symbiotic algae, and the rest of its microbiome—responds, acclimatizes, and adapts to stress.

Ultimately, the goal of the interventions is to alter the reef in some way by shifting population structures, altering genes, or changing the composition of symbiont and microbiome communities. These changes may benefit coral reefs, the species that live on them, and the human communities that depend on them. But these changes provide very different benefits across sites and may have unintended consequences that will similarly vary across locations.

An ad hoc committee was convened by the National Academies of Sciences, Engineering, and Medicine to evaluate the potential for these new interventions to increase the persistence and resilience of coral reefs, and to provide a framework for evaluating their risks and benefits. In its first report (NASEM, 2019), the committee reviewed the state of science on potential interventions. The report addresses what is known about the benefits and goals, current feasibility, potential scale, risks, limitations, and infrastructure needs for 23 novel approaches. The committee’s tasks for this second report were to (1) provide a framework for assessing relative risks and benefits of interventions, including in comparison to a baseline or no action; (2) describe a decision pathway that spans the range of actions from new research to future implementation; (3) identify research needs that would refine the intervention strategies and reduce critical uncertainties in the environmental risk assessments; and (4) assess the potential for interventions to meet management objectives for Atlantic/Caribbean coral reefs (the full Statement of Task can be found in Box 1.1). This study was requested and funded by the National Oceanic and Atmospheric Administration, with additional support from the Paul G. Allen Family Foundation.

Best practices for decision making rely on a reef-specific, structured approach that includes input from relevant stakeholders to develop objectives and identify preferred options and decision criteria, as well as detailed modeling of the coral ecosystem and how it affects ecosystem services. Community involvement happens early in the process, and because obtaining this input for any specific location was outside the tasks set for the committee, the committee could not provide specific decision analyses for any local reef system. Instead, the report provides a set of guidelines and principles that incorporate best practices for informing decisions under uncertainty when evaluating new interventions, including an example analysis to illustrate the use of such a framework.

In this report, the committee first reviews the interventions from the first report to assess their practical readiness and context dependencies to illuminate how managers may select interventions most suited to their situation for further assessment in a decision framework (Chapter 2). Next, the committee describes best practices for structured decision making, including evaluating risks and benefits, for taking an intervention

from research to inclusion in a management strategy, and the available tools that have been, or might be, applied to coral reef management (Chapter 3). The committee provides an illustrative model and decision analysis of a coral reef system to exemplify the challenges and insights associated with decision making around coral interventions (Chapter 4). The committee then highlights research areas that would help inform decision making by improving the understanding of the baseline reef system, assessing risks, and managing the beneficial impacts of potential interventions (Chapter 5). Finally, the committee uses the Atlantic/Caribbean region as a case study for how managers may consider their individual context and objectives in an evaluation of possible intervention strategies (Chapter 6).

SELECTING INTERVENTIONS FOR DECISION ANALYSIS

Managers and decision makers are faced with the task of evaluating the benefits and risks of a growing number of interventions, separately and in combination. A number of factors help narrow the field of options. Primarily, readiness or the timeframe for achieving readiness for implementation will determine which interventions are practical on short- or long-term timeframes. For example, testing corals for heat resistance and growing them in nurseries is available to support managed selection, managed breeding, and managed relocation. In contrast, using genetic engineering to increase resilience or marine cloud brightening to reduce light and cool reef surface waters are not yet feasible, and the risks and benefits of these options are not well defined. Interventions available immediately can buy time until more, and potentially more impactful, interventions can become ready for use, or until greenhouse gas emissions are reduced or removed and warming is abated.

The interventions have different risks, benefits, and feasibilities in different regions. Biophysical attributes of a reef or region that influence the choice of intervention include the current degree of reef degradation, disease prevalence, bleaching history and future projections of bleaching events, water quality, herbivory, recruitment, connectivity, spatial extent of the reef, potential for cold shocks, and temperature variability. Moreover, these dependencies are likely to differ when considering where to test versus where to deploy or scale up interventions. Existing management programs and infrastructure will also influence which interventions are more readily deployed. These existing programs and resources will include restoration programs based on sexual reproduction (for genetic and reproductive interventions), technical expertise and research infrastructure availability (for various physiological interventions), propagation programs (for population- and community-level interventions), and

significant engineering infrastructure and scalability (for environmental interventions). The process of identifying appropriate interventions, as well as the locations best suited for their implementation, will likely require coordination across management entities that may be multijurisdictional or multinational in nature. Furthermore, there may be efficiencies that can be found across interventions that share biological or infrastructural resources and can be implemented in combination.

Conclusion: Multiple, interrelated interventions provide a toolbox for increasing coral reef persistence and resilience. This set of options can be tested and deployed based on community goals, ecological objectives for reef management, and the benefits and risks across multijurisdictional or even multinational boundaries. These efforts are likely to evolve over time as interventions become more feasible and as new interventions are developed.

A STRUCTURED, ADAPTIVE APPROACH TO DECISION MAKING

Coral reefs are socioecological systems; humans are responsible for the greatest threats to reef persistence and resilience yet are also among the primary beneficiaries of healthy and functional coral reefs that provide a variety of ecosystem services and benefit streams. A structured decision approach for managing coral reefs links driving forces, human and natural pressures, ecosystem states, measured impacts on ecosystem services, and societal and management responses in a coral reef system. The evaluation of coral interventions is part of a broader decision context that includes managing other stressors (e.g., water quality, overfishing, habitat destruction) to achieve overall coral reef conservation objectives based on community values. Within this framework, an adaptive management approach provides an explicit process for planning, evaluating, implementing, monitoring, and adjusting specific management strategies (e.g., interventions, conventional restoration activities, and their timing) when outcomes are uncertain, based on their measured impact and the overall management goals. The steps of this adaptive process, focused on evaluating coral interventions, are outlined below.

Step 1: Identify the Decision Context

An iterative adaptive management process begins with a planning and problem formulation stage to establish the decision context: identifying long- and short-term goals, objectives, possible biophysical outcomes, and their relationship to evaluation metrics and decision criteria.

Although decision makers have primary responsibility for problem formulation, stakeholder involvement is required to establish shared goals and objectives. Stakeholders may have multiple and sometimes conflicting viewpoints, and also have varying risk tolerances to management alternatives (both unintended and known potential negative consequences). The process of reaching consensus on objectives and criteria will make preferences and risk tolerance explicit and will help guide how some objectives might need to trade off with others given stakeholder and decision-maker values.

Step 2: Model Linkages Across Interventions, Biophysical Outcomes, and Objectives

Evaluation of expected intervention risks and benefits requires modeling, as quantitatively as possible, the biophysical (and, as appropriate, social and economic) consequences of implementing different interventions or a set of interventions. Model design and input parameters should be tailored to specific locations at relevant spatial and temporal scales. Biophysical models assume that coral reef systems are naturally highly dynamic and characterized by stochastic variability, and that there is uncertainty in knowledge of these dynamics. Model outputs for reef condition can include attributes such as coral growth, reproductive capacity, coral cover, coral diversity, herbivore biomass, coral disease, macroalgae cover, and other metrics identified by researchers and stakeholders as critical indicators of coral reef health and resilience. To assess the consequences of interventions (or inaction) for society, metrics of coral reef health are mapped directly to agreed-on decision criteria that translate to impacts on ecosystem services that represent the economic, social, and cultural values of stakeholders and decision makers.

Step 3: Analyze Tradeoffs in Criteria Across Alternatives

Reef managers are likely to consider a range of management alternatives, including using one or more interventions in concert with conventional restoration activities as well as taking no action. These combinations, along with uncertainty in knowledge about the reef system and future environment, will yield a range of modeled outcomes across alternatives with tradeoffs in their abilities to meet management objectives and minimize risk. For example, some intervention strategies may support the growth of a small subset of coral species that provides fish habitat but not the solid reef structure that is needed to provide coastal protection from storm waves. If fish habitat and strong reef structure are both key objectives for different stakeholder groups, then tradeoffs

need to be made to reconcile different priorities or value preferences. A number of tools are available for analyzing these tradeoffs to guide a preferred course of action. These include multi-criteria decision analysis, decision trees, system dynamics models, and Bayesian networks. Most importantly, analyzing tradeoffs requires a deliberative approach with stakeholder values at the center.

Step 4: Select Interventions or Combination of Management Activities and Determine Evaluation Metrics

Once decision makers understand the potential performance of the suite of intervention strategies relative to the multiple objectives, and tradeoff analyses have generated an agreed subset of preferred strategies, one or more strategies can be selected for implementation. Measurable evaluation metrics are developed across decision criteria that link to the objectives established in Step 1 and are used to iteratively evaluate whether objectives are being achieved. For example, a decision criterion might be to increase coral cover and diversity to create fish habitat. The associated evaluation metrics might then be abundance and diversity of coral and fish species measured at appropriate time intervals.

Steps 5 and 6: Implement Interventions, and Initiate and Sustain a Monitoring Plan

A targeted monitoring program conducted prior to, during, and after implementation that is based on specific biophysical outcomes is needed to provide the data necessary to quantify the evaluation metrics. Effective and targeted monitoring is critical to assess intervention performance compared to objectives, and to reduce critical uncertainty in models.

Steps 7, 8, and 9: Evaluate, Communicate, and Adapt

Evaluation of monitoring data can identify progress made toward meeting management objectives (including partial success or failure), or reveal the need for more information. The results of the evaluation can be used to communicate progress in meeting objectives to stakeholders and decision makers. The adaptive management framework allows for monitoring data to inform iterative improvements to model design and input parameters to inform strategy adjustment.

Conclusion: Although many tools exist for structured decision making to evaluate interventions as part of a reef management strategy, there is no single generalizable approach and no substitute for working through

a structured decision process with stakeholders in the local context. This effort provides a data- and values-informed basis for selecting and evaluating management options against a set of objectives.

Recommendation: A structured, adaptive management framework that considers all drivers and pressures affecting coral reefs should be developed to evaluate tradeoffs across alternatives and identify when and where new coral intervention(s) will be beneficial or necessary. This framework should include

• Engagement of a broad set of stakeholders to establish objectives and courses of action that reflect community values.
• Development of models tailored to the local environmental and ecological setting, management objectives, and preferred intervention options.
• Targeted monitoring of short- and long-term metrics of reef health and resilience.
• Iterative evaluation and adjustment of management strategies.

AN ILLUSTRATIVE DECISION MODEL

The committee provides a simple model and Bayesian network analysis to illustrate the principles of a decision analysis for a simplified reef system and the potential questions faced and insights gained from the approach. Though the results of the analysis are not reef specific, the illustration provides a concrete example of the construction of a decision support framework to analyze the risks and benefits of example interventions, evaluate the likelihood of achieving intervention goals under different scenarios, and communicate potential outcomes. The example focuses on a subset of interventions through which a range of benefits and risks of intervention can be illustrated: assisted gene flow and atmospheric shading (i.e., marine cloud brightening). In practice, a more extensive set of interventions and their combinations would be explored in a localized context. The potential for these example interventions to support reef persistence under climate change was also analyzed in combination with the management of local stressors to demonstrate the relative importance of continuing these practices along with adopting new strategies.

Biophysical Model

The committee constructed a biophysical coral community model to capture the basic ecological dynamics of the reef system (see Figure S.1). The committee used a simple model appropriate for qualitative,

comparative interpretation; any model constructed for a decision-making process would use a more realistic and locally tailored modeling framework appropriate for more quantitatively precise predictions. As a proxy for a range of coral responses and community states, the model predicts the proportion of area covered by macroalgae and two types of coral: slow-growing coral (such as a foliose or massive coral) and fast-growing coral (such as a branching coral). Dynamics of growth, baseline mortality, and competition are modeled as continuous processes; bleaching and recruitment are modeled as discrete events. The model builds on the approach presented in Mumby et al. (2007), which has been adapted by others.

The proportion of mortality due to bleaching is modeled as a function of predicted degree heating weeks (DHWs) based on global climate model

projections. The committee includes estimates of how natural adaptation by corals, whether through acclimatization and/or genetic adaptation, might keep up with predicted increases in thermal stress associated with climate change. The committee represents the potential for natural adaptation using variable windows for rolling climatologies as presented in Logan et al. (2014), in which adaptation or acclimatization occurs as the corals’ threshold for experiencing thermal stress (DHW) changes based on recent thermal history. Accounting for natural adaptive processes enables better identification of situations where interventions will be needed and also allows for a process to represent the benefit of interventions that accelerate adaptation.

Modeling Benefits and Risks of Interventions

Model evaluation of interventions requires representation of both risks and benefits in the model parameters. Table S.1 identifies how the risks and benefits of each intervention may be modeled using the committee’s example framework. However, different modeling frameworks will offer more detailed mechanistic representation across risks and benefits for necessary predictive power. For example, using a rolling climatology window calculation of DHWs implicitly accounts for adaptation or acclimatization processes, but explicitly accounting for genetic dynamics could allow a model to predict the expected evolutionary outcome from interventions that affect genetic composition rather than assuming a particular adaptation rate.

For the example interventions, assisted gene flow occurs as an accelerated adaptation in thermal tolerance with the risks of increased mortality and decreased growth (due to outbreeding depression, demographic tradeoffs with thermal tolerance, and disease introduction), and shading occurs as reduced DHW exposure with a risk of slowed adaptation in thermal tolerance and occasional shading failure. In addition, the committee also considered two other scenarios that can be applied alone or in combination with these interventions: (1) no intervention (a baseline of no action), and (2) applying conventional management strategies by varying herbivory rate (v; influenced by management of overfishing) and macroalgal growth rate (γ; influenced by control of nutrient runoff). The committee also explored a range of values for initial coral cover (5% versus 30%), climate scenarios based on the Intergovernmental Panel on Climate Change’s (IPCC’s) representative concentration pathways (RCP2.6 versus 8.5), and intervention start dates (2025 versus 2035), with decadal target dates for assessing the coral cover outcome (2020-2060).

TABLE S.1 Anticipated Benefits and Risks of Interventions and the Biophysical Parameters for Modeling Using the Committee’s Example Model

NOTE: DHW = degree heating week; OA = ocean acidification.

Bayesian Network Analysis

Simulation of all intervention strategy options over a range of conditions and time, along with environmental stochasticity in DHWs, results in a distribution of possible outcomes for total coral cover, the model’s focal evaluation metric. The committee demonstrates the use of Bayesian network analysis to convert the range of outputs from the dynamic model to a network of conditional probabilities to inform decision making. That is, it shows how one condition (e.g., achieving coral cover of at least 20%) depends on the probability of other conditions being met and which conditions matter the most (e.g., the implementation of an intervention versus strong climate mitigation). Figure S.2 illustrates the probabilities of achieving at least 20% coral cover across the various management strategies and contrasting climate scenarios. Example conclusions that decision makers might draw from an analysis of this type include (1) intervention risks outweigh benefits under strong climate mitigation (RCP2.6), but benefits outweigh risks for business-as-usual emissions (RCP8.5), and (2) under business-as-usual climate change, intervention success relies on effective local management. These conclusions will vary with context-dependent model development, parameterization, and ground truthing.

Conclusion: A successful modeling framework requires substantial effort in tailoring the model structure and parameters to the decision context, risks and benefits of the interventions under consideration,

local environmental conditions, and reef ecosystem dynamics. As demonstrated by the committee’s illustrative effort, the utility and payoff of this approach is the ability to identify

• The conditions necessary for new and potentially risky interventions to outperform the no-action alternative under different future climate scenarios.
• The interventions expected to be most effective at achieving management objectives.
• Potential synergistic and antagonistic interactions across multiple interventions, including management of local stressors.
• The key dynamics and parameters resolving empirically in order to improve the capacity to predict intervention efficacy and risks.

Further application of such a modeling framework includes identifying indicators for context- or condition-dependent decisions, monitoring, and adaptive management. The insight provided by a quantitative

model enables decision makers and reef stakeholders to compare the benefits and risks of different intervention options with more clarity and transparency than that provided by qualitative or conceptual approaches or by expert opinion only. The benefits of a quantitative model are greatest when local ecosystem and evolutionary dynamics are known, and when primary sources of uncertainty are considered.

ADVANCEMENTS THROUGH RESEARCH

Despite the rapid pace of research on coral biology and conservation that is occurring on a global scale, there are many gaps and unresolved issues that need to be addressed in the short and long term. Priority research would improve understanding of the risks and benefits associated with a potential intervention and reduce critical uncertainty to better inform decision making and modeling. These priorities include improved ways to identify, measure, and monitor fitness parameters of corals; greater understanding of factors that contribute to stress tolerance and associated tradeoffs for corals; and measuring the impact of interventions on demographic processes within reef ecosystems.

Research on Fundamental Coral Reef Biology

Effective intervention approaches for reefs require an improved understanding of which factors underpin coral health and how these lead to reef resilience at scale. Though these topics are inherently broad, there are a number of research areas that can be prioritized. These areas include the following:

• Identifying the cellular mechanisms of bleaching and how these pathways are influenced by recent thermal history, host genetics, symbiont type, and microbiome.
• Identifying underlying causes of coral diseases and developing biomarkers of coral health, heat susceptibility, and disease diagnosis, as well as ecosystem health.
• Determining functional roles of, and tradeoffs among, members of coral reef communities at multiple ecological scales from coral-associated symbionts and microbiomes up to the composition of coral species in reef communities.
• Identifying population structure, determining evidence for local adaptation, and defining relevant management units for population recovery.

• Developing methods to improve recruitment and survivorship of corals that are released, planted, relocated, or settled on reefs at the reef scale.
• Developing extensive, freely available databases on coral hosts, symbionts, and microbiomes to support studies on population structure, genotype–phenotype relationships, population structure, and community dynamics.
• Identifying species-specific threshold responses of corals to changes in temperature, light incidence, and ocean pH, as well as reef-scale threshold responses to disturbance and environmental change.

Site-Specific Research and Assessment

Development of appropriate ecological models and identification of relevant management objectives and goals requires site-specific information. The committee highlighted the assessments that would inform robust decision-making processes. These items include the following:

• Identifying local stressors that influence population recovery and determining whether stressors are likely to influence the success of interventions.
• Developing appropriate metrics and recovery goals that assess the effects of the intervention on ongoing tolerance, health, fitness, and recruitment within the target management unit as well as on connected reefs.
• Evaluating whether population recovery at a specific site can be achieved through translocation or managed breeding, and if so, which intervention is most appropriate.
• Identifying host, symbiont, and microbial populations at restoration sites to ensure that treatments or manipulations aimed at improving coral physiological performance can achieve recovery goals.
• Assessing in a site-specific manner the benefits, risks, and chances of success for implementing environmental interventions.
• Identifying the most appropriate site-specific, synergistic management and intervention strategies that together provide a greater chance of success and reduced risks than the sum of the impacts of each intervention alone.

Research to Improve Specific Interventions

Research is needed to stage interventions from laboratory experiments to full-scale management strategies. Additionally, research can help inform the safety, efficacy, and cost-efficiency of interventions. Priority research to support these goals includes the following:

• Developing protocols for control of pathogens (biosecurity and quarantine).
• Developing effective approaches to modify symbiotic algal and/or microbiome populations.
• Developing effective approaches to determine whether corals that are released, planted, relocated, or settled on reefs contribute to recovery goals while reducing risk to ongoing adaptation and ecological processes.
• Developing and testing genome-editing methods in a wide variety of ecologically important coral species.
• Developing methods of delivery for nutrients, probiotics, antibiotics, phage therapy, and antioxidants at reef scales.
• Assessing feasibility, potential benefits, costs, limitations, and risks associated with environmental interventions.

Research to Inform Risk Assessments and Modeling

The adaptive management cycle requires monitoring and evaluating the results of a management action based on an established monitoring program in order to iteratively gain knowledge and improve information to support decision making. Thus, ongoing improvements to structured decision making require:

• Using targeted monitoring to evaluate performance, improve benefits, and minimize or manage risks.
• Using iterative model design to reduce uncertainties and improve model predictions to increase confidence in the decision support framework.

TROPICAL WESTERN ATLANTIC AND CARIBBEAN CASE STUDY

The committee was tasked with assessing coral intervention strategies and their ability to meet objectives for sustaining coral reefs in the Caribbean and tropical western Atlantic. Coral reefs in this region show

widely variable conditions, but many areas have experienced uniquely devastating losses in recent history. These losses have been due to a wide variety of factors including widespread local stressors, disease outbreaks, hurricanes, and bleaching events.

Ecological decline of Caribbean reefs appeared to accelerate in the late 1970s. Basinwide, major disease epizootics have been responsible for some of the greatest changes, with the loss of both the structurally dominant elkhorn and staghorn corals (Acropora palmata and A. cervicornis), followed by the loss of a keystone herbivore, the black-spined sea urchin (Diadema antillarum). Overfishing has compounded the impacts of the loss of Diadema, and the invasive lionfish, which has also spread rapidly through the region, has aggravated the impacts of overfishing by preying on juvenile fish. Poor land-use practices have also contributed to coral mortality by increasing sedimentation, nutrients, and turbidity. Recently, a virulent new disease, stony coral tissue loss disease, has emerged in Florida and parts of the central and northeastern Caribbean, threatening coral diversity.

In addition to the significant impact of local stressors in the Caribbean, climate change impacts are increasing, and as elsewhere, there are limits to the effectiveness of local management in this context. Coral bleaching events have been observed widely across the Caribbean. Warming also affects corals indirectly by increasing intensities of tropical storms, and by triggering more devastating episodes of coral disease.

Implications for Selecting and Modeling Interventions

Assessing the condition of Caribbean and tropical western Atlantic reefs helps clarify the attributes most relevant to selecting interventions, and influences analyses aimed at deciding which interventions to test and deploy. These attributes include generally poor reef conditions, intrinsic vulnerability, high interconnectedness, low diversity of coral and algal symbionts, high environmental variability across the region, and persistent and destructive disease outbreaks. The social attributes include a relatively widespread and growing network of coral restoration practitioners located in a small (compared to the Indo-West Pacific) but politically complex region.

The committee identifies a range of opportunities for including interventions across management strategies and activities in the region. The following strategies seem to be the most promising for the tropical western Atlantic/Caribbean region based on the regional context dependencies and technical readiness across interventions. They represent options for more detailed evaluation using a model-based decision framework,

including consideration of local management objectives and acceptable courses of action based on stakeholder input. They include

• Identifying heat-tolerant or disease-resistant coral genotypes (and evaluating potential tradeoffs between these traits) among the Caribbean standing stock to provide opportunities for assisted gene flow, managed breeding, and genetic interventions.
• Leveraging existing coral restoration activities and infrastructure (involving the nursery propagation and outplanting of asexually derived clones) to establish a comprehensive region-wide program to boost larval recruitment and survivorship.
• Exploiting sexual restoration activities to test algal symbiont manipulations.
• Expanding coral cryopreservation across the region to provide opportunities for managed breeding and assisted gene flow.
• Testing short-distance managed relocation (i.e., assisted gene flow) of corals across local thermal gradients, where disease incidence is not a limiting factor.
• Leveraging restoration activities to test pre-exposure methods to increase stress tolerance of outplanted corals.
• Assessing feasibility of environmental interventions to reduce heat stress at both local and subregional scales.
• Testing interventions, such as antibiotics, phage therapy, and microbiome manipulations, to halt the spread of emerging diseases, improve coral condition, and increase the success and/or feasibility of other interventions.
• Combining interventions where possible to increase resource efficiencies.
• Testing the efficacy of interventions under a range of different conditions by exploiting variability in the degree of degradation across the region.
• Developing regional and multinational coordination and agreements to meet the scale of the challenge.
• Soliciting and incorporating stakeholder input on interventions to gauge and maximize acceptability/social license.

Conclusion: Coral reef managers in the tropical western Atlantic/Caribbean region have a variety of interventions available to them depending on the localized management context and the specific objectives of stakeholders and decision makers. Available actions include leveraging existing restoration and propagation infrastructure, increasing sexual reproduction and genetic diversity of corals (managed breeding, gamete and larval capture and seeding, coral cryopreservation),

capitalizing on thermally tolerant species and genotypes (managed selection, algal symbiont manipulation), accelerating reef connectivity to boost thermal tolerance when disease is not a factor (managed relocation), reducing disease spread (antibiotics, phage therapy, microbiome manipulation), and/or reducing exposure to stress (environmental interventions). The complex disease geography in the Caribbean requires particular care to ensure that interventions do not facilitate the spread or severity of ongoing disease outbreaks. These rapidly developing new interventions do not replace the need for direct management of local stressors.

Recommendation: The ongoing management and restoration efforts in the Caribbean provide a strong foundation on which to implement newly emerging interventions designed to increase the resilience of individual corals and coral populations. The modeling and decision-making tools outlined in this report should be used to inform more detailed assessments to evaluate which approaches might be appropriate for specific settings, including their interactions with more traditional management approaches. Maintaining genetic diversity in the face of multiple climate-driven stresses (e.g., bleaching and disease) is particularly important. Monitoring corals to maintain genetic diversity and identify resistant phenotypes should be simplified and standardized for research, ex situ propagation, and in situ restoration. Research programs to model and field test the risks, benefits, and efficacy of interventions in this multinational and highly interconnected region should be coordinated to maximize resources, co-learning opportunities, and the ability to achieve management objectives regionally.